UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Sensitive and specific chromatographic methods for the pharmacokinetic evaluation of carboplatin in young… Burns, Robbin Bruce 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-565173.pdf [ 8.21MB ]
Metadata
JSON: 831-1.0089694.json
JSON-LD: 831-1.0089694-ld.json
RDF/XML (Pretty): 831-1.0089694-rdf.xml
RDF/JSON: 831-1.0089694-rdf.json
Turtle: 831-1.0089694-turtle.txt
N-Triples: 831-1.0089694-rdf-ntriples.txt
Original Record: 831-1.0089694-source.json
Full Text
831-1.0089694-fulltext.txt
Citation
831-1.0089694.ris

Full Text

SENSITIVE AND SPECIFIC CHROMATOGRAPHIC PHARMACOKINETIC  EVALUATION  METHODS  OF CARBOPLATIN  FOR  IN YOUNG  by ROBBIN B R U C E B U R N S B . S c . University of British C o l u m b i a , 1991 B . S c . (Pharm.) University of British C o l u m b i a , 1994  A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES Faculty of P h a r m a c e u t i c a l S c i e n c e s Division of P h a r m a c e u t i c a l Chemistry  W e accept this thesis a s conforming to the required standard  T H E UNIVERSITY O F BRITISH C O L U M B I A October 2000 ® R o b b i n Burns, 2 0 0 0  THE  PATIENTS  In presenting this degree  thesis  in partial  fulfilment  of the requirements  at the University of British Columbia, I agree that the Library shall make it  freely available for reference and study. I further copying  for an advanced  agree that permission for extensive  of this thesis for scholarly purposes may be granted  department  or  by  his or  her  representatives.  It  is  by the head of my  understood  that  copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department  of  P"^W^m<Au  The University of British Columbia Vancouver, Canada  D  DE-6  a  t  e  (2/88)  OCVoe.ES- U , Z O O O  O ^ v s r ^  ABSTRACT C a r b o p l a t i n , c/s-diamminedichloro-1,1-cyciobutanedicarboxylatoplatinum II, is a s e c o n d g e n e r a t i o n platinum antineoplastic agent that cross-links D N A .  D o s i n g of  carboplatin in adults is frequently b a s e d o n formulas that relate renal elimination to AUC.  P h a r m a c o k i n e t i c studies evaluating t h e s e d o s i n g s c h e m e s are c o m p l i c a t e d by  the a s s a y m e t h o d s utilized.  In particular, substitution  of platinum for  carboplatin  m e a s u r e m e n t s in p h a r m a c o k i n e t i c studies h a s o c c u r r e d d u e to the poor sensitivity of existing H P L C a s s a y s for the parent drug.  T h e objectives of this study w e r e (1) to  d e v e l o p a n d validate a sensitive a n d specific H P L C a s s a y method for determination of carboplatin in p l a s m a ultrafiltrate, a n d (2) to c o m p a r e carboplatin A U C e s t i m a t e s d e t e r m i n e d using free (non protein-bound) carboplatin v e r s u s e s t i m a t e s using free platinum. A s s a y m e t h o d s for quantitation of carboplatin w e r e d e v e l o p e d utilizing H P L C with direct U V detection a n d with U V detection following post-column ( P C ) derivatization. B o t h m e t h o d s y i e l d e d quantitation limits that w e r e approximately 10-fold m o r e sensitive than previous H P L C - U V m e t h o d s .  F o r the H P L C - P C m e t h o d , s o d i u m bisulfite w a s  u s e d to derivatize carboplatin to products p o s s e s s i n g e n h a n c e d U V a b s o r b a n c e at 2 9 0 nm.  T h e greater selectivity of this analytical w a v e l e n g t h r e m o v e d the n e e d for s a m p l e  extraction, resulting in a s i m p l e r a n d m o r e rapid a s s a y p r o c e d u r e . MethCBDCA  (bis(methylamine)cyclobutanedicarboxylatoplatinum  II)  and  M e t h M a l (bis(methylamine)malonatoplatinum II) w e r e s y n t h e s i z e d a s internal s t a n d a r d s for the H P L C - U V a n d H P L C - P C m e t h o d s , respectively, the c h e m i c a l structures being confirmed by H P L C - M S under positive electrospray ionization conditions.  Both H P L C -  U V a n d H P L C - P C m e t h o d s w e r e validated for carboplatin concentrations from 0.05 to 4 0 |ag/mL.  V a l i d a t e d a s s a y p a r a m e t e r s included limits of detection a n d quantitation,  specificity, p r e c i s i o n , a c c u r a c y , linearity, a n d r u g g e d n e s s . R e c o v e r y a n d stability w e r e e x a m i n e d using the H P L C - U V method only. A s s a y variability w a s m u c h higher for the H P L C - P C m e t h o d d u e to time-dependent c h a n g e s in signal r e s p o n s e c a u s e d by the  d e g r a d a t i o n of the post-column reagent, s o d i u m bisulfite.  T h e more p r e c i s e H P L C - U V  m e t h o d w a s u s e d in the p h a r m a c o k i n e t i c study. The  pharmacokinetic  behaviour  of  carboplatin  c h e m o t h e r a p y in two y o u n g patients w a s investigated.  during  eight  cycles  of  C a r b o p l a t i n c o n c e n t r a t i o n s in  p l a s m a ultrafiltrate w e r e determined by the H P L C - U V m e t h o d , while p l a s m a ultrafiltrate platinum concentrations w e r e d e t e r m i n e d by atomic a b s o r b a n c e . e x a m i n a t i o n of the concentration-time elimination  were  clearly different.  To  B a s e d on visual  profiles, free carboplatin a n d free better  characterize the  magnitude  platinum of  the  differences, A U C e s t i m a t e s derived from free platinum d a t a w e r e c o m p a r e d to t h o s e d e r i v e d from free carboplatin data. T h e free platinum A U C e s t i m a t e s w e r e a s m u c h a s two-fold greater than their free carboplatin counterparts.  H o w e v e r , the o b s e r v e d  d i f f e r e n c e s w e r e m u c h s m a l l e r in o n e patient than in the other. Further study is n e e d e d to m o r e accurately c h a r a c t e r i z e the extent of t h e s e differences a c r o s s a larger patient population.  Currently, w e r e c o m m e n d that H P L C a n d A A a s s a y m e t h o d s not be u s e d  i n t e r c h a n g e a b l y for determination of carboplatin p h a r m a c o k i n e t i c p a r a m e t e r s .  TABLE OF CONTENTS Title  i  Abstract  ii  T a b l e of C o n t e n t s  iv  List of T a b l e s  viii  List of F i g u r e s  x  List of A b b r e v i a t i o n s  xiii  List of C h e m i c a l s  xvi  Acknowledgements  xvii  C H A P T E R 1.  GLOBAL  INTRODUCTION  1.1.  Biological Properties  1  1.2.  Pharmacokinetics and Dose-Adjustment Strategies  8  1.3.  Analytical Methods for Carboplatin Determination  13  1.4.  Hypothesis  19  1.5.  Thesis Objectives and Rationale  19  1.6.  References  20  C H A P T E R 2.  DEVELOPMENT AN HPLC-UV  AND PRELIMINARY ASSAY  METHOD  FOR  EVALUATION  OF  CARBOPLATIN  2.1  Introduction  27  2.2.  Experimental  28  2.2.1.  Buffers a n d mobile p h a s e s  28  2.2.2.  Preparation of p l a s m a ultrafiltrate from blood s a m p l e s  28  2.2.3.  Apparatus  29  2.2.4.  C o l u m n evaluation  29  2.2.5.  E v a l u a t i o n of s o l i d - p h a s e extraction cartridges  30  iv  2.2.6.  C h r o m a t o g r a p h i c optimization (carboplatin in p l a s m a ultrafiltrate)  32  2.2.7.  Determination of limits of detection a n d quantitation  32  2.3.  Results and Discussion  33  2.3.1.  C o l u m n evaluation  33  2.3.2.  Preliminary separation from e n d o g e n o u s p l a s m a c o m p o n e n t s  37  2.3.3.  E v a l u a t i o n of s o l i d - p h a s e extraction cartridges  39  2.3.4.  C h r o m a t o g r a p h i c optimization  40  2.3.5.  Determination of limits of detection a n d quantitation  42  2.4.  Conclusions  43  2.5.  References  44  C H A P T E R 3. DEVELOPMENT AN HPLC-PC  AND PRELIMINARY ASS A Y METHOD  EVALUA TION OF  FOR CARBOPLA  TIN  3.1.  Introduction  45  3.2.  Experimental  48  3.2.1.  Buffers, mobile p h a s e s , a n d p l a s m a ultrafiltrate  48  3.2.2.  Apparatus  48  3.2.3.  Stabilizing the c h r o m a t o g r a p h i c b a s e l i n e  48  3.2.4.  Optimization studies  48  3.2.5.  Determination of limits of detection a n d quantitation  49  3.3.  Results and Discussion  50  3.3.1.  Stabilizing the c h r o m a t o g r a p h i c b a s e l i n e  50  3.3.2.  Optimization of the H P L C - P C s y s t e m  53  3.3.3.  Determination of limits of detection a n d quantitation  58  3.4.  Conclusions  59  3.5.  References  60  C H A P T E R 4.  SYNTHESIS INTERNAL  AND EVALUATION STANDARD  OF  CANDIDATES  4.1.  Introduction  61  4.2.  Experimental  62  4.2.1.  S o l u t i o n s , mobile p h a s e s , a n d p l a s m a ultrafiltrate  62  4.2.2.  Syntheses  62  4.2.3.  Evaluation of s y n t h e s i z e d internal standard c a n d i d a t e s  66  4.3.  Results and Discussion  67  4.3.1.  Syntheses  67  4.3.2.  M e t h M a l , E t h M a l , and M e t h C B D C A a s internal standards  72  4.4.  Conclusions  74  4.5.  References  75  C H A P T E R 5.  VALIDATION  OF HPLC-UV  AND HPLC-PC  ASSAY  METHODS  5.1.  Introduction  76  5.2.  Experimental  78  5.2.1.  P r e p a r a t i o n of mobile p h a s e s  78  5.2.2.  Chromatography  79  5.2.3.  P r e p a r a t i o n of p l a s m a ultrafiltrate  79  5.2.4.  S o l i d - p h a s e extraction ( H P L C - U V a s s a y )  80  5.2.5.  U s e of p e a k height/area ratio v a l u e s  80  5.2.6.  V a l i d a t i o n experiments  80  5.2.7.  Statistical c o m p a r i s o n s  84  5.3.  Results and Discussion  85  5.3.1.  Specificity and selectivity  85  5.3.2.  U s e of p e a k height/area ratio v a l u e s  87  5.3.3.  Reproducibility experiment (precision, a c c u r a c y , and linearity)  88  5.3.4.  Recovery  95  5.3.5.  Stability ( H P L C - U V method)  95  5.3.6.  Ruggedness  98  5.4.  Conclusions  99  5.5.  References  100  C H A P T E R 6.  EVALUATION  OF CARBOPLATIN  PHARMACOKINETICS  6.1.  Introduction  101  6.2.  Experimental  103  6.2.1.  C l i n i c a l protocol  103  6.2.2.  Patient characteristics  103  6.2.3.  G r a p h i t e f u r n a c e atomic a b s o r b a n c e s p e c t r o s c o p y  104  6.2.4.  Sample analysis  105  6.2.5.  P h a r m a c o k i n e t i c evaluation  105  6.3.  Results and Discussion  108  6.3.1.  V i s u a l e x a m i n a t i o n of elimination profiles  108  6.3.2.  P a r a m e t r i c modeling a n d parameter estimation  112  6.3.3.  A s s a y a n d modeling effects o n A U C v a l u e s  114  6.3.4.  R e l e v a n c e of o b s e r v e d A U C differences  116  6.3.5.  F r e e carboplatin/platinum in dose-adjustment strategies  117  6.4.  Conclusions  119  6.5.  References  120  C H A P T E R 7. GLOBAL  SUMMARY  A P P E N D I X . C O M P A R T M E N T A L D A T A FITS  122  126  vii  LIST OF TABLES  Table 1.1  Relative o c c u r r e n c e s of p l a t i n u m - D N A adducts.  7  Table 2.1  P h y s i c a l properties of H P L C c o l u m n s evaluated for carboplatin  29  chromatography. Table 2.2  Retention and efficiency data for carboplatin o n H P L C c o l u m n s  33  evaluated. Table 4.1  Description of products obtained v i a the method of P a s i n i a n d  68  Caldirola. Table 4.2  Retention time data for carboplatin and its M e t h M a l , E t h M a l , a n d  72  M e t h C B D C A analogues. Table 4.3  M e a n percentage recoveries for carboplatin a n d its M e t h M a l ,  74  EthMal, and M e t h C B D C A analogues. Table 5.1  Properties of the H P L C - U V and H P L C - P C a s s a y s .  79  Table 5.2  H P L C - U V data from batch 1 Q C s a m p l e s containing 0.1, 8, a n d  87  4 0 p.g/mL carboplatin. Table 5.3  P e a k a r e a ratio precision (untransformed) d a t a for the H P L C - U V  89  assay. Table 5.4  P e a k height ratio precision (untransformed) d a t a for the H P L C -  89  P C assay. Table 5.5  A c c u r a c y data for Q C s a m p l e s generated by weighted (1/y )  90  2  linear regression ( H P L C - U V a s s a y ) . Table 5.6  A c c u r a c y data for Q C s a m p l e s generated  by weighted  (1/y ) 2  90  linear regression ( H P L C - P C a s s a y ) . Table 5.7  R e c o v e r y of carboplatin following ultrafiltration.  95  Table 5.8  P e a k a r e a ratio data for blood s a m p l e s stored at 4 °C.  96  Table 5.9  P e a k a r e a ratio data for p l a s m a s a m p l e s stored at -70 °C.  96  Table 5.10  Stability of H P L C - U V s a m p l e extracts frozen a t - 7 0 °C.  97  Table 5.11  A u t o s a m p l e r stability of H P L C - U V extracts.  97  Table 6.1  Pediatric studies using free platinum m e a s u r e m e n t s to  101  investigate carboplatin p h a r m a c o k i n e t i c s . Table 6.2  P h y s i c a l characteristics of patients evaluated in the clinical study.  104  Table 6.3  A n a l y t i c a l results s u m m a r y .  106  Table 6.4  M e a n parameter estimates derived from the H P L C a n d A A d a t a .  113  Table 6.5  P a r a m e t r i c a n d trapezoidal A U C estimates (ng/mL * h).  114  Table 6.6  R e c e n t studies describing dose-individualization strategies for  117  carboplatin.  ix  LIST OF FIGURES  Figure 1.1  Structures of platinum c o m p o u n d s a c c e p t e d for g e n e r a l clinical u s e .  1  Figure 1.2  Structures of the cyclobutanedicarboxylato a n a l o g u e s enloplatin  3  a n d D W A 2 1 1 4 R , the 1,2-diaminocyclohexane c o m p o u n d oxaliplatin, the  octahedral  compound  iproplatin,  and  trans-  dichloroamminequinolonoplatinum IV. Figure 1.3  G e n e r a l structures for investigational dinuclear, cationic, a n d redox-  3  activated platinum c o m p l e x e s . Figure 1.4  A q u a t i o n a n d hydrolysis equilibria for carboplatin a n d cisplatin.  5  Figure 1.5  E l e c t r o p h e r o g r a m ( U V 2 0 0 nm) of blank p l a s m a ultrafiltrate a n d  14  p l a s m a ultrafiltrate s p i k e d with 50 jig/mL of enloplatin a n d DWA2114R. Figure 1.6  H P L C - I C P - M S c h r o m a t o g r a m s of blank p l a s m a ultrafiltrate a n d  16  p l a s m a ultrafiltrate containing 20 n g / m L e a c h of carboplatin, enloplatin, a n d D W A 2 1 1 4 R . Figure 1.7  Ultraviolet s p e c t r u m of carboplatin in a q u e o u s solution.  18  Figure 2.1  P r o t o c o l for evaluation of carboplatin retention on normal a n d  31  reverse p h a s e extraction cartridges. Figure 2.2  Effect of k' programming on relative efficiency a n d resolution v a l u e s .  36  Figure 2.3  C h r o m a t o g r a m s ( U V 2 3 0 nm) of blank p l a s m a ultrafiltrate a n d  38  p l a s m a ultrafiltrate s p i k e d with 5 jj.g/ml_ of carboplatin. Figure 2.4  Elution profiles for carboplatin o n normal p h a s e cartridges.  39  Figure 2.5  C h r o m a t o g r a m s ( U V 2 3 0 nm) of s o l i d - p h a s e extracted blank p l a s m a  41  ultrafiltrate a n d p l a s m a ultrafiltrate containing 8 |ag/ml_ carboplatin. Figure 2.6  S t a n d a r d curve (0.1-8 ng/mL) u s e d to estimate the L O Q of the  42  H P L C - U V assay.  x  Figure 3.1  G e n e r a l s c h e m a t i c of the post-column reaction s y s t e m , e m p l o y i n g a  47  knitted tubular reactor, u s e d for the a n a l y s i s of carboplatin in p l a s m a ultrafiltrate. Figure 3.2  Initial H P L C - P C c h r o m a t o g r a m ( U V 2 9 0 nm) of a 5 u.g/mL a q u e o u s  50  carboplatin standard s h o w i n g large fluctuations in the chromatographic baseline. Figure 3.3  C h r o m a t o g r a p h i c b a s e l i n e s o b s e r v e d for mobile p h a s e s p u m p e d  51  into the U V detector with monitoring at 2 9 0 n m . Figure 3.4  A b s o r b a n c e profiles (240-310 nm) for solutions consisting of s o d i u m  52  bisulfite in p h o s p h a t e buffer adjusted to p H 4.5 a n d p H 6.2. Figure 3.5  Effect of post-column reagent p H on b a s e l i n e fluctuations in the  52  H P L C - P C system. Figure 3.6  Effect of post-column reagent flow rate on b a s e l i n e fluctuations in  53  the H P L C - P C s y s t e m . Figure 3.7  H P L C - P C c h r o m a t o g r a m s of blank p l a s m a ultrafiltrate a n d p l a s m a  55  ultrafiltrate containing 8 u.g/mL carboplatin. Figure 3.8  Optimization of the post-column s y s t e m for buffer type, reagent p H ,  57  acetonitrile concentration, a n d bisulfite concentration. Figure 3.9  S t a n d a r d curve (0.1-8 ^ g / m L ) u s e d to estimate the L O Q of the  58  H P L C - P C assay. Figure 4.1  P r o p o s e d s y n t h e s i s of carboplatin a n a l o g u e s using the m e t h o d of  64  P a s i n i a n d C a l d i r o l a , w h i c h h a s b e e n s u c c e s s f u l l y applied to the s y n t h e s i s of carboplatin. Figure 4.2  M e t h o d of C l e a r e et al. u s e d for synthesis of carboplatin a n a l o g u e s .  65  Figure 4.3  H P L C a n a l y s i s of the product of the reaction of cisplatin with c y c l o -  69  hexanedicarboxylic acid.  xi  Figure 4.4  H P L C - U V chromatogram of s o l i d - p h a s e extracted M e t h M a l filtrate.  70  Figure 4.5  H P L C - M S with s e l e c t e d ion recording of parent i o n s f o r m e d u n d e r  71  positive electrospray conditions. Figure 4.6  H P L C c h r o m a t o g r a m s of a n a q u e o u s solution containing  73  carboplatin, M e t h M a l , E t h M a l , a n d M e t h C B D C A under conditions optimized for the H P L C - U V and H P L C - P C m e t h o d s . Figure 5.1  C h r o m a t o g r a m s of an a q u e o u s carboplatin standard e x p o s e d to  85  1 M hydrochloric acid at 0 min a n d 120 min. Figure 5.2  Specificity of the H P L C - U V a s s a y method for carboplatin in the  86  p r e s e n c e of co-administered drugs (50 uxj/mL of e a c h ) . Figure 5.3  B a t c h 3 calibration c u r v e s for the H P L C - U V a s s a y m e t h o d .  93  Figure 5.4  B a t c h 3 calibration c u r v e s for the H P L C - P C a s s a y m e t h o d .  94  Figure 5.5  H P L C - P C r e s p o n s e s o b s e r v e d (peak heights) for M e t h M a l a n d  99  s e l e c t e d carboplatin concentrations after repetitive injection (n=9) of a n a q u e o u s calibration curve. Figure 6.1  T e m p e r a t u r e program u s e d for the a n a l y s i s of platinum in p l a s m a  104  ultrafiltrate. Figure 6.2  C a r b o p l a t i n post-infusion elimination profiles for patient 1.  109  Figure 6.3  C a r b o p l a t i n post-infusion elimination profiles for patient 2.  110  Figure 6.4  P l o t s s h o w i n g residual differences b e t w e e n predicted  111  concentrations from the A A v e r s u s H P L C a s s a y m e t h o d s . Figure 6.5  E x a m p l e s of weighted a n d unweighted two-compartment fits to the  112  A A d a t a from patient 2.  xii  LIST OF ABBREVIATIONS  Platinum  compounds  carboplatin  c/'s-diammine-1,1-cyclobutanedicarboxylatoplatinum II; J M - 8  cisplatin  c/s-diamminedichloroplatinum II  DWA2114R  2-aminomethylpyrrolidine-1-1-cyclobutanedicarboxylatoplatinum  enloplatin  [1,1 -cyclobutanedicarboxylato(2-)-0,0']-(tetrahydro-4H-pyran-4,4dimethanamine-N,N')platinum II  EthMal  bis(ethylamine)malonatoplatinum II  JM-9  dichlorodihydroxybis(isopropylamine)platinum; iproplatin  JM-56  bis(isobutylamine)malonatoplatinum II  MethCBDCA  bis(methylamine)-1,1 -cyclobutanedicarboxylatoplatinum II  MethMal  bis(methylamine)malonatoplatinum II  {JM is the designation  used by Johnson  Matthey for their investigational  Analytical AA  atomic absorption (spectroscopy)  ES  electrospray (ionization)  GC  g a s chromatography  HPLC  high-performance liquid c h r o m a t o g r a p h y  ICP  inductively c o u p l e d p l a s m a (ionization)  MS  m a s s spectrometry  PC  post-column derivatization (with ultraviolet detection)  UV  ultraviolet  compounds)  II  Chromatography  and  Validation  a  selectivity factor  k'  c a p a c i t y factor  LOD  limit of detection  LOQ  limit of quantitation  (N ), N  (effective) n u m b e r of theoretical plates  ODS  o c t a d e c y l s i l a n e (C18)  QC  quality control  S/N  signal-to-noise  tp  void time  t  retention time  eff  r  R  resolution  s  Pharmacokinetics AUC  a r e a under the p l a s m a concentration-versus-time c u r v e  AUMC  a r e a under the m o m e n t c u r v e  Clast  o b s e r v e d concentration of final s a m p l i n g point  CL  clearance  CL B  total b o d y c l e a r a n c e  Cmax  m a x i m a l p l a s m a concentration a c h i e v e d  T  5 1  Cr-EDTA  c h r o m i u m 51-edathamil (radiolabel for G F R m e a s u r e m e n t )  GFR  glomerular filtration rate  i.v.  intravenous (administration)  MRT  m e a n r e s i d e n c e time  ti/  half-life  2  Vss  v o l u m e of distribution at s t e a d y state  xiv  Miscellaneous approximately oc  proportional to (math symbol)  oo  infinity (math symbol)  ANOVA  a n a l y s i s of v a r i a n c e  °C  degrees Celcius  C V (RSD)  coefficient of variation (relative standard deviation)  DNA  d e o x y r i b o n u c l e i c acid  et al.  et alia  eq.  equivalent (number of mole)  g  gravitational acceleration (centrifugation)  h  hour  ICE  ifosfamide-carboplatin-etoposide  i.d.  internal diameter  i.e.  id est (that is)  (k bs), k  (observed) rate constant  min  minute  m/z  m a s s - t o - c h a r g e ratio  MW  m o l e c u l a r weight  p  probability v a l u e (statistics)  r  coefficient of determination  RNA  ribonucleic acid  rpm  revolutions per minute  s  second  SD  standard deviation  0  2  J?  '  LIST OF CHEMICALS  Platinum  compounds  Carboplatin  and  potassium  tetrachloroplatinate  were  obtained  from  Strem  C h e m i c a l s Inc. (Newburyport, M A , U S A ) . Cisplatin w a s obtained from S i g m a C h e m i c a l C o . (St. L o u i s , M O , U S A ) . Gifts of D W A 2 1 1 4 R a n d enloplatin w e r e provided by C h u g a i P h a r m a c e u t i c a l ( S h i z u o k a , J a p a n ) a n d the A m e r i c a n C y a n a m i d C o . ( P e a r l River, N Y , USA),  respectively.  J M - 9 (iproplatin)  a n d J M - 5 6 w e r e d o n a t e d by A n o r M E D Inc.  (Langley, B C , C a n a d a ) .  Drugs co-administered  in clinical trial  E t o p o s i d e injection (20 mg/ml) a n d ifosfamide for injection w e r e from BristolM y e r s S q u i b b (Montreal, P Q , C a n a d a ) , ondansetron injection (2 mg/ml) w a s from G l a x o L a b o r a t o r i e s (Toronto, O N , C a n a d a ) , a n d trimethoprim W e l l c o m e (Kirkland, O N , C a n a d a ) .  p o w d e r w a s from  D e x a m e t h a s o n e , dimenhydrinate,  Burroughs  nystatin, a n d  s u l f i s o x a z o l e p o w d e r s w e r e p u r c h a s e d from S i g m a - A l d r i c h (Oakville, O N , C a n a d a ) .  Solvents and other  chemicals  H P L C g r a d e acetonitrile, m e t h a n o l , a n d tetrahydrofuran w e r e s u p p l i e d by F i s h e r Scientific C o . ( V a n c o u v e r , B C , C a n a d a ) , a s w e r e A C S g r a d e a c e t i c a c i d , perchloric acid 6 0 % , s o d i u m acetate, s o d i u m hydroxide, s o d i u m p h o s p h a t e ( m o n o b a s i c monohydrate a n d d i b a s i c a n h y d r o u s ) , a n d s o d i u m bisulfite (mix of N a H S 0  3  and N a S 0 ) . 2  2  5  Formic  a c i d , hydrochloric a c i d , a n d p o t a s s i u m hydroxide w e r e supplied by B D H C h e m i c a l Inc. (Toronto, O N , C a n a d a ) . Citric acid (monohydrate), silver nitrate, s o d i u m iodide, s o d i u m perchlorate, a n d s o d i u m citrate (trisodium salt, dihydrate) w e r e p u r c h a s e d from S i g m a A l d r i c h C a n a d a , a s w e r e methylamine ( 4 0 % by wt. solution in water) a n d ethylamine ( 7 0 % by wt. solution in water). 1,1-cyclobutanedicarboxylic a c i d , 1,1-cyclopentanediacetic a c i d , 1,1-cyclohexane d i a c e t i c a c i d , 1,1-cyclohexanedicarboxylic a c i d , a n d m a l o n i c acid w e r e obtained from L a n c a s t e r S y n t h e s i s ( C a l e d o n Laboratories, G e o r g e t o w n , O N , C a n a d a ) .  xvi  ACKNOWLEDGEMENTS  F o r e m o s t , I would like to e x p r e s s my gratitude to my supervisor, Dr. L e a n n e E m b r e e , for her patience, support, a n d g u i d a n c e throughout my graduate s c h o o l i n g . I w o u l d a l s o like to thank the other m e m b e r s of my committee: Drs. Frank Abbott, G a i l B e l l w a r d , William C u l l e n , Kath M a c l e o d , Keith M c E r l a n e , a n d W a y n e R i g g s for their input,  evaluation, a n d a s s i s t a n c e .  appreciated.  Their time  spent o n this project  is  gratefully  S p e c i a l thanks g o to Dr. M c E r l a n e for the u s e of laboratory s p a c e a n d to  Dr. B e l l w a r d for the loan of H P L C equipment.  T h e a s s i s t a n c e of S h e i l a Pfiffer (Calgary C h i l d r e n ' s Hospital) a n d T r a c i C o r r (British C o l u m b i a C h i l d r e n ' s Hospital) in obtaining clinical s a m p l e s is a c k n o w l e d g e d , a s is the help of H a n s A d o m a t (British C o l u m b i a C a n c e r A g e n c y ) with atomic a b s o r b a n c e a n a l y s e s a n d Dr. M i c h a e l A b r a m s ( A n o r M E D Inc.) with internal standard s y n t h e s e s . S p e c i a l t h a n k s g o to the ladies at the British C o l u m b i a C a n c e r A g e n c y : N o r m a H u d o n , J e a n H e g g i e , D a r i a Hartley, M a r i a K r s c a n s k i , a n d S h e l l y P r a s a d .  T h a n k y o u for your  encouragement.  I w o u l d like to thank all the individuals within the Faculty w h o provided m e with a s s i s t a n c e or a d v i c e . O f s p e c i a l note are the m a n y n e w friends I m a d e during my time in g r a d u a t e s c h o o l , e s p e c i a l l y V i n c e T o n g , C l a r a K w p k , S u s a n L a m , K a r e n L o , a n d Maggie Li.  B e s t w i s h e s to e a c h of y o u in your future e n d e a v o r s .  Hopefully, the fun  times will continue after graduation.  Finally, I a m grateful to my parents for their support (monetary a n d otherwise) a n d to my girlfriend, for her tolerance in dealing with my unique work s c h e d u l e .  xvii  CHAPTER 1 GLOBAL  INTRODUCTION  .0 CI  H3N  0  H 3 N \  Pt CI  H3N  Carboplatin  Cisplatin  Figure 1.1.  1.1.  O  H 3 N ^  Structures of platinum compounds accepted for general clinical use.  Biological Properties  Overview  and clinical  utility  T h e a c c i d e n t a l d i s c o v e r y of the antitumour activity of platinum c o m p o u n d s by R o s e n b e r g in the late 1 9 6 0 ' s [1] h a s led to m u c h r e s e a r c h into the biological properties of t h e s e c o m p o u n d s .  T h e first platinum c o m p o u n d d e v e l o p e d for w i d e s p r e a d clinical  u s e , cisplatin, w a s introduced clinically in 1 9 7 1 . A n excellent e x a m p l e of the i m p o r t a n c e of cisplatin in m o d e r n c a n c e r therapy is its u s e in a d v a n c e d testicular c a n c e r , w h e r e long-term survival rates h a v e i n c r e a s e d from 5 - 1 0 % before the introduction of cisplatin to g r e a t e r than 8 0 % following treatment with a regimen of cisplatin c o m b i n e d with vinblastine a n d b l e o m y c i n [2].  Unfortunately, the utility of cisplatin is s o m e w h a t limited  d u e to potentially s e r i o u s s i d e effects. nephrotoxicity,  auditory  and  M o r e frequent toxicities of cisplatin include  visual  impairment,  m y e l o s u p p r e s s i o n , a n d n a u s e a a n d vomiting.  peripheral  Cisplatin is primarily eliminated by the  k i d n e y s , consistent with the observation that renal impairment toxicity.  is the  dose-limiting  T h e m e c h a n i s m by w h i c h this renal toxicity o c c u r s r e m a i n s uncertain a n d is  somewhat complex. cellular  neuropathy,  components  T h e inactivation of s p e c i f i c renal b r u s h border e n z y m e s or other may  be  at  least  partially  responsible  [3,4].  While  some  1  investigators h a v e pointed to effects of cisplatin on renal N a , K - A T P a s e activity [5], +  +  others c l a i m that both mitochondrial a n d m e m b r a n e - b o u n d renal N a , K - A T P a s e s are +  not affected [6].  +  R e n a l failure is of particular c o n c e r n in patients that are not properly  hydrated or that h a v e existing glomerular dysfunction. In most patients on conventional cisplatin d o s e s , glomerular filtration rates are r e d u c e d for 2 4 - 4 8 hours p o s t - d o s e , but later return to normal.  In s o m e patients, however, the nephrotoxic, neurotoxic  or  ototoxic s i d e effects of cisplatin may not be completely reversed upon discontinuation of therapy [7]. In the 1970's, various platinum II a n a l o g u e s w e r e d e v e l o p e d in a n effort to find a l e s s toxic but equally efficacious alternative to cisplatin.  T h e most promising of t h e s e  c o m p o u n d s , carboplatin, is the only other platinum c o m p o u n d currently a c c e p t e d for routine clinical u s e in C a n a d a a n d the United States (Figure 1.1).  Carboplatin has a  similar s p e c t r u m of activity a n d incidence of c r o s s - r e s i s t a n c e to cisplatin [8]. in  addition  to  being  less  emetogenic  than  cisplatin, carboplatin  i n c i d e n c e s of nephrotoxicity, neurotoxicity, a n d ototoxicity. toxicities,  m y e l o s u p p r e s s i o n (predominantly  shows  However, reduced  In the a b s e n c e of t h e s e  thrombocytopenia)  b e c o m e s the  dose-  limiting side-effect [9,10]. Both cisplatin a n d carboplatin h a v e demonstrated cytotoxicity in a n u m b e r of s a r c o m a , m e l a n o m a a n d l e u k e m i a tumour m o d e l s a n d h a v e s h o w n clinical efficacy, e s p e c i a l l y in bladder, lung, ovarian, a n d testicular c a r c i n o m a s [11].  Carboplatin has  a l s o s h o w n efficacy in the treatment of children with solid tumours, brain tumours, a n d a c u t e l e u k e m i a [12]. Drug d e v e l o p m e n t efforts continue to find platinum c o m p l e x e s with improved efficacy, d e c r e a s e d toxicity, a n d altered s p e c t r u m of activity to cisplatin a n d carboplatin [13, 14]. compounds,  T h e s e include other "carboplatin-like"  1,2-diaminocyclohexane  compounds,  cyclobutanedicarboxylato  octahedral  compounds,  and  c o m p o u n d s with frans-positioned ligands (Figure 1.2), a s well a s c o m p o u n d s with more novel or interesting structures, s u c h a s dinuclear bis(platinum) c o m p o u n d s , cationic c o m p o u n d s , a n d c o m p o u n d s with redox-activated ligands (Figure 1.3).  W h i l e s o m e of  t h e s e n e w e r a g e n t s are currently undergoing clinical trials, n o n e h a v e to date received a p p r o v a l for w i d e s p r e a d clinical u s e .  2  o  o H N.  / O - C  2  o  ;pt;  H  H N"  S ^ H N  /  P  t  X  0 -  C  V  2  O  DWA2114R  Enloplatin  H N  3  O-C  2  O-C,  2  Oxaliplatin  Figure 1.2.  .NH  3  2  2  3  H N  Ck  CI (CH ) CHNH v I / O H > < (CH ) CHNH/ | ^ O H CI  O  o  2  frans-[PtCl2(NH3)(quin)]  Iproplatin (JM-9)  Structures of the cyclobutanedicarboxylato analogues enloplatin and DWA2114R, the  1,2-diaminocyclohexane  iproplatin, and  compound oxaliplatin, the octahedral  frans-dichloroamminequinolonoplatinum  Ck  .NH  H N. ^Pt;  3  C K "  ^NH (CH ) H hr 2  2  n  IV.  CI  3  ^Pt  compound  CI  2  Dinuclear bis(platinum)  .,  NO, H N 2  H N  O II < .sc R  X "  2  R  CI  Cationic  Figure 1.3.  ,N^.NCH CH OH  Ck  H N 3  2  /  Pt  \  2  CH, CI  Redox-Activated  General structures for investigational dinuclear, cationic, and  redox-activated  platinum complexes.  3  Degradation  and cellular  uptake  C a r b o p l a t i n a n d other s q u a r e planar platinum II c o m p o u n d s undergo nucleophilic substitution reactions a c c o r d i n g to the following g e n e r a l s c h e m e [15]: rate =  * [Pt II c o m p o u n d ] + k * [Pt II c o m p o u n d ] * [nucleophile] 2  T h e c o n s t a n t s ki a n d k are the reaction rates for the nucleophile a n d solvent p a t h w a y s , 2  respectively.  If the nucleophilic s p e c i e s is present in large e x c e s s , then the e x p r e s s i o n  b e c o m e s p s e u d o first order: rate =  + k [nucleophile]) * [Pt II c o m p o u n d ] = k bs * [Pt II c o m p o u n d ] 2  0  In vivo metabolites of platinum c o m p o u n d s are primarily just s i m p l e substitution products.  T h u s , the rate a n d extent of m e t a b o l i s m will be dictated by the reactivity of  the leaving g r o u p s .  In this respect, the chloro ligands of cisplatin are hydrolyzed m u c h  m o r e quickly than the cyclobutanedicarboxylato ligand of carboplatin. For both cisplatin a n d carboplatin, the potential for substitution of ligands with a variety of nucleophiles in solution  obscures  attempts  by  investigators  to  determine  r e s p o n s i b l e for both their d e s i r a b l e a n d toxicologic effects. a q u a t e d , hydroxylated, or chloro-substituted  the  "active"  species  C a r b o p l a t i n d e g r a d e s to  forms (including cisplatin) in biological  fluids a n d other a q u e o u s environments [16,17].  It is generally a c c e p t e d that both  cisplatin a n d carboplatin are hydrolyzed before undergoing reaction with D N A (Figure 1.4).  A t physiological p H , the c h l o r o - h y d r o x y ' a n d dihydroxy c o m p o u n d s are favoured  o v e r the c o r r e s p o n d i n g chloro-aquo a n d d i a q u o c o m p o u n d s [18,19].  It h a s b e e n  h y p o t h e s i z e d that the neutrally c h a r g e d hydroxy s p e c i e s , a l o n g with the parent drug, are r e s p o n s i b l e for diffusion into the cells, while the more reactive a q u o forms are r e s p o n s i b l e for the s u b s e q u e n t D N A - b i n d i n g activity [20].  H o w e v e r , there are s o m e  active cationic platinum c o m p l e x e s which remain positively c h a r g e d in vivo [14].  In  addition, contributions to diffusion from the dihydroxy s p e c i e s a n d to activity from the d i a q u a s p e c i e s are likely to be small, s i n c e the equilibrium c o n s t a n t s are  heavily  w e i g h t e d to the chloro-aquo a n d chloro-hydroxy forms, a n d the rate constants are extremely s m a l l relative to t h o s e for other nucleophiles present in biological fluids.  4  o  H INL  /0-C.  HN  ^o-c;  3  3  O  carboplatin  H2O l  c |  -  H N  CI  H N  O-C, <0  3  3  COOH  H H  OH  H^N  O-C  3  ci-  / ciCI  3  H N^  3  /  OH2  3  Pt H N"  COOH O  ciH N.  2  \  H2O OH2  H NL 3  Pt  Pt' X  CI  H N^  X  3  cisplatin (dichloro)  CI  H N^  /OH  3  Pt H N^ 3  p K a = 5.6  pKai = 6.5  H N^  X  CI  chloro-hydroxy  2  diaquo  chloro-aquo A  OH  X  3  2  H N^  /OH  3  Pt H ^ ^  X  OH2  aquo-hydroxy  OH  H N. 3  -1  0  Pt H N^ 3  X  OH  dihydroxy Figure 1.4.  Aquation and hydrolysis equilibria for carboplatin and cisplatin.  Carboplatin  undergoes conversion to chloro, aquo, and hydroxy forms in vivo. The pK, and p K  2  values are taken from Martin and Lim [18] and Andersson et a/. [19], respectively. The uncharged dichloro, chloro-hydroxy, and dihydroxy compounds may cross cell membranes more readily.  The charged chloro-aquo, diaquo, and aquo-hydroxy  species are approximately 1000 times more reactive than cisplatin [20].  5  Mechanism  of action  A l t h o u g h the cytotoxic activity of carboplatin h a s b e e n l e s s e x t e n s i v e l y studied than that of cisplatin, the active (hydrolysis) s p e c i e s present in the cell are b e l i e v e d to b e similar for both m o l e c u l e s .  W h i l e cisplatin reacts readily with R N A a n d protein,  cross-linking of D N A a p p e a r s to be the predominant m e c h a n i s m by w h i c h it exerts its cytotoxic effects. T h e types of a d d u c t s formed have b e e n reviewed by R e e d a n d K o h n [7].  The  most  common  D N A adduct  generated  (approximately  60%)  involves  intrastrand binding of adjacent g u a n o s i n e r e s i d u e s ( P t - G G ) via the N 7 substituent of their i m i d a z o l e ring.  L e s s frequent are intrastrand c r o s s - l i n k s b e t w e e n a d e n i n e a n d  guanosine  monofunctionally  (Pt-AG),  bound  platinum  (Pt-G),  interstrand c r o s s - l i n k s bridging o n e or s e v e r a l b a s e s ( G - P t - G ) .  and  intrastand  or  W h i l e the interstrand  D N A a d d u c t s m a k e up less than 1% of the total, s o m e investigators h a v e pointed to t h e s e a d d u c t s a s the major c a u s e of cytotoxicity.  T h i s is b a s e d on the o b s e r v a t i o n that  interstrand a d d u c t s h a v e b e e n s h o w n to inhibit D N A replication a n d are not formed to a n y a p p r e c i a b l e extent by transplatin, the less cytotoxic trans  i s o m e r of cisplatin.  H o w e v e r , intrastrand D N A a d d u c t s h a v e b e e n s h o w n to p r o d u c e a distortion of D N A structure w h i c h could interfere with replication.  A l s o , while t h e s e intrastrand a d d u c t s  c a n be r e m o v e d by D N A repair e n z y m e s , the P t - G G a n d P t - A G a d d u c t s s e e m more resistant to e x c i s i o n than d o the G - P t - G a n d m o n o a d d u c t s .  Transplatin is sterically  u n a b l e to form the P t - A G a n d P t - G G a d d u c t s ; the greater n u m b e r of D N A m o n o a d d u c t s a n d G - P t - G a d d u c t s it d o e s form m a y be more readily repaired or, in the c a s e of m o n o a d d u c t s , detoxified by cellular glutathione [13,21]. R e c e n t l y , B l o m m a e r t et al. [22] studied the formation of cisplatin a n d carboplatin D N A a d d u c t s in vitro a n d in C h i n e s e hamster ovary cells. T h e relative a m o u n t s of the v a r i o u s a d d u c t s w e r e determined by atomic absorption s p e c t r o s c o p y or e n z y m e - l i n k e d i m m u n o s o r b e n t ( E L I S A ) quantitation of fractions isolated by e n z y m a t i c digestion a n d column chromatography. both c o m p o u n d s in  W h i l e similar profiles of adduct formation w e r e o b s e r v e d for  vitro,  a  much  greater  i n c i d e n c e of G - P t - G  (intrastrand  interstrand) a d d u c t s w a s o b s e r v e d for carboplatin in vivo (Table 1.1).  and  Furthermore,  1 0 0 - 2 3 0 times more carboplatin than cisplatin w a s required to i n d u c e similar levels of a d d u c t formation, w h e r e a s only a 4 - 2 0 times higher carboplatin d o s e w a s required for  6  similar levels of cytotoxicity.  T h e s e results h a v e b e e n attributed to the involvement of  o x y g e n free radicals in activation of carboplatin [23] or to p h a r m a c o k i n e t i c differences b e t w e e n the two platinum c o m p o u n d s [24].  In any event, the studies d e m o n s t r a t e that  carboplatin is not merely a prodrug of or slow-acting substitute for cisplatin, but a unique agent with potential a d v a n t a g e s to be further e x p l o r e d .  Table 1.1.  Relative occurrences of platinum-DNA adducts. Adapted from data presented by Blommaert et al. [22]. Pt-GG  Pt-AG  Pt-G  G-Pt-G  cisplatin (5 (im)  65%  19%  13%  4%  carboplatin (1.35 mM)  65%  14%  17%  3%  cisplatin (40 ^m)  57%  15%  9%  18%  carboplatin (700 ^m)  28%  16%  17%  38%  in vitro  3  in vivo  b  H N 33  \  NH3  /  n  HhN ' \ 3  Pt  /  \  Nhk ^  / Pt  /  i_i M  / Pt  3  \  N 3 H  H N  /  X  x  u N \ 3  /  / Pt  I  N  NFk M  3  H N  NH /  3  \  \  3  /t  .. G -  G -  -  A -  G --  -  G -  G -  - G - - X - G -  -  G - \ G  -  .. C -  C -  -  T -  C -  -  C -  C -  - C - X - G -  -  C -  -  Pt-GG  Pt-AG  Pt-G  G-Pt-G (intrastrand)  C  G-Pt-G (interstrand)  salmon sperm D N A treated in vitro at 37 °C for 16 h Chinese hamster ovary cells incubated with drug for 1 h at 37 °C; post-incubated in drug-free medium for 7 h  7  1.2.  Carboplatin Pharmacokinetics and Dose-Adjustment Strategies  Overview T h e clinical pharmacokinetics of carboplatin in adults have b e e n extensively s t u d i e d a n d excellent reviews written by V a n der Vijgh [25] a n d , most recently, by Duffull a n d R o b i n s o n [26].  S i n c e the oral bioavailability of carboplatin is low (<15%), it  is g i v e n intravenously, most often by 1 h continuous infusion.  Initial protein binding  e x p e r i m e n t s s u g g e s t e d that s o m e carboplatin in p l a s m a is protein-bound. e x p e r i m e n t s by G a v e r et al. [27] using  1 4  However,  C - l a b e l e d carboplatin s h o w e d that this protein  binding w a s essentially irreversible in nature, a n d that carboplatin did not undergo i n s t a n t a n e o u s a n d reversible p l a s m a protein binding.  W h i l e carboplatin did distribute  into blood cells of rats, no carboplatin w a s found a s s o c i a t e d with the cellular fraction in either d o g s or h u m a n s . place  of  whole  blood  This provides a rationale for a n a l y s i s of p l a s m a s a m p l e s in for  human  pharmacokinetic  studies.  Furthermore,  since  irreversibly-bound drug is no longer available for further p h a r m a c o l o g i c a l activity, most s t u d i e s f o c u s on free (ultrafilterable) carboplatin.  M e a s u r e m e n t of this free fraction  c o m m o n l y involves either high-performance liquid c h r o m a t o g r a p h y or a t o m i c absorption spectroscopy.  W h i l e the former t e c h n i q u e is specific for parent d r u g , the  latter  t e c h n i q u e m e a s u r e s a combination of parent drug plus non protein-bound nucleophilic substitution a n d hydrolysis products. Elimination profiles from A A m e a s u r e m e n t of free platinum (parent drug plus substitution products) a n d H P L C m e a s u r e m e n t of free carboplatin (parent drug only) are similar up to 12 h post-administration, after w h i c h the elimination of free platinum b e c o m e s less rapid. T h e c h e m i c a l stability of the cyclobutanedicarboxylato moiety of  carboplatin  a c c o u n t s for its s l o w in vivo degradation to d e c a r b o x y l a t e d degradation products. A s a result, carboplatin is eliminated primarily via the k i d n e y s into the urine w h e r e at least 5 0 % of the platinum is recovered a s intact drug. W h e r e a s renal elimination of cisplatin partially  involves a n active secretory m e c h a n i s m , carboplatin c l e a r a n c e h a s b e e n  s h o w n to approximate the glomerular filtration rate [28].  F o r administration of d o s e s up  to 4 5 0 m g / m , carboplatin p h a r m a c o k i n e t i c parameters s u c h a s c l e a r a n c e a n d A U C 2  8  h a v e b e e n s h o w n to be linearly correlated to d o s e , w h i c h provides the rationale for clinical d o s e - a d j u s t m e n t strategies.  Predictive  models  for carboplatin  Considerable  effort  dosing in adults  has  been  invested  in  understanding  the  clinical  p h a r m a c o k i n e t i c s of carboplatin in adults a n d in d e v e l o p i n g m e t h o d s to u s e this information  to  optimize therapy.  Quantitation  of both free  carboplatin  by  high-  p e r f o r m a n c e liquid chromatography a n d free or total platinum by atomic absorption s p e c t r o s c o p y are p o s s i b l e .  H o w e v e r , lack of a s s a y  methodology  with  sufficient  specificity a n d sensitivity to quantitate the parent drug h a s resulted in the u s e of n o n s p e c i f i c a t o m i c absorption m e t h o d s in most clinical studies. F o r cytotoxic c o m p o u n d s s u c h a s carboplatin, d o s e s b a s e d o n A U C v a l u e s are c o n s i d e r e d to be a better predictor of toxicity or r e s p o n s e than are d o s e s b a s e d simply o n b o d y weight or b o d y surface a r e a [29].  Carboplatin c l e a r a n c e (which d o e s not  involve active secretory or reabsorptive p r o c e s s e s ) is predominantly via glomerular filtration; thus, patients with r e d u c e d renal function are at greater risk for toxicity, most notably m y e l o s u p p r e s s i o n . pretreatment  T h i s w a s proven by E g o r i n et at. [30], w h o s h o w e d that  renal function w a s related to A U C , w h i c h w a s in turn related to the  o b s e r v e d d e g r e e of t h r o m b o c y t o p e n i a .  T h e s e relationships w e r e u s e d to derive a  f o r m u l a from w h i c h the carboplatin d o s e n e e d e d to i n d u c e a particular c h a n g e in platelet count could be predicted using the patient's body s u r f a c e a r e a ( B S A ) , previous history of c h e m o t h e r a p y , a n d creatinine c l e a r a n c e ( C L ) a s v a r i a b l e s : cr  d o s e ( m g / m ) = 0.091 x ( C L 2  C R  / B S A ) x [desired platelet c h a n g e (%)  - a] + 86  w h e r e a w a s 0 or 17 for previously untreated or treated patients, respectively, a n d m e a s u r e m e n t of creatinine c l e a r a n c e b a s e d on 24-hour urine collection w a s u s e d to estimate glomerular filtration.  9  D i s a d v a n t a g e s with the Egorin formula, most notably the a w k w a r d requirement for charting platelet levels, led to the p r o p o s a l of a n alternative formula by C a l v e r t et al. [31], w h i c h related carboplatin A U C directly with the glomerular filtration rate ( G F R ) : d o s e (mg) = d e s i r e d A U C x ( G F R + 25) In the C a l v e r t formula, elimination of (mL/min).  5 1  C r - E D T A [32] is u s e d for G F R estimation  T h e target A U C value c a n be varied a c c o r d i n g to s p e c i f i c institutional or  c h e m o t h e r a p e u t i c protocols, c o m m o n l y u s e d v a l u e s being 5 a n d 7 ( m g / m L * min) for previously treated a n d untreated patients, respectively. T o d a y , C a l v e r t ' s formula h a s b e c o m e the more widely a c c e p t e d . H o w e v e r , the formula d o e s h a v e its o w n d i s a d v a n t a g e s , most notably the requirement for c l e a r a n c e m e a s u r e m e n t s using  5 1  C r - E D T A , w h i c h is not available in all clinics a n d requires that  three b l o o d s a m p l e s b e t a k e n to a d e q u a t e l y c h a r a c t e r i z e elimination of the radiolabel. F o r this r e a s o n , s o m e institutions substitute the C o c k c r o f t - G a u l t formula [33] in p l a c e of 5 1  C r - E D T A c l e a r a n c e m e a s u r e m e n t s , although u s e of the predictive formula h a s b e e n  s h o w n to result in b i a s e d G F R estimates [34,35].  Pediatrics W h i l e the p h a r m a c o k i n e t i c b e h a v i o u r of carboplatin in pediatric patients treated with 175 to 1200 m g / m 2 h a s b e e n reported a s similar to that o b s e r v e d in adults [36,39], the p o o r sensitivity of specific a s s a y m e t h o d s h a s s e v e r e l y limited c o m p l e t e evaluation of the terminal elimination p h a s e for carboplatin in this patient group. P h a r m a c o k i n e t i c s t u d i e s in pediatric patients h a v e predominately substituted ultrafilterable  platinum  determination for parent drug with A U C v a l u e s calculated following fitting of the d a t a to a t w o - c o m p a r t m e n t m o d e l [36-40].  T h e clinical significance of replacing free platinum  determination with free carboplatin is not fully u n d e r s t o o d .  Extrapolation of C a l v e r t ' s  d o s i n g formula to children is c o m p l i c a t e d by the greater variability in b o d y m e t a b o l i s m , a n d other factors, resulting in the potential for significant  size,  inter-patient  differences in non-renal carboplatin elimination within this patient group. T o a c c o u n t for  10  this variability, N e w e l l et al. [37] d e v e l o p e d a modified formula incorporating body m a s s measurements.  Other  investigators  have  modified  C a l v e r t ' s formula  by  simply  d e c r e a s i n g the magnitude of the tissue binding constant to reflect the s m a l l e r a v e r a g e s i z e of pediatric patients a n d by incorporating body surface a r e a m e a s u r e m e n t s into the e q u a t i o n [38]. T o date, no particular dose-adjustment strategy h a s r e c e i v e d w i d e s p r e a d a c c e p t a n c e in this patient population. S t u d i e s o n the relationship between B S A - b a s e d d o s e s a n d A U C h a v e p r o d u c e d conflicting results. F o r e x a m p l e , o n e study of pediatric patients previously not e x p o s e d to platinum c h e m o t h e r a p y [39] demonstrated a strong correlation b e t w e e n d o s e a n d AUC.  T h e r e w a s little interpatient variability in o b s e r v e d A U C a s well a s no i n c r e a s e  following repeated d o s e s . H o w e v e r , another study of 18 patients previously e x p o s e d to c h e m o t h e r a p y a n d receiving h i g h - d o s e carboplatin d e m o n s t r a t e d a two to interpatient variability  in A U C [40].  three-fold  Obviously, more p h a r m a c o k i n e t i c studies  are  n e e d e d to clarify t h e s e o b s e r v a t i o n s .  Novel approaches  for the refinement  of carboplatin  dosing  Chatelut et al. [41] u s e d population pharmacokinetic modeling with the nonlinear m i x e d effects m o d e l ( N O N M E M ) [42] for data from 34 adult patients to p r o p o s e a new formula for predicting carboplatin c l e a r a n c e b a s e d on various patient characteristics: 218 x weight x (1 - 0.00457 x age) x (1 - 0.314 x gender) CL(ml/min) = 0.134 x weight + serum creatinine w h e r e weight is m e a s u r e d in kg, s e r u m creatinine levels are m e a s u r e d in p.mol/L, a n d g e n d e r is 0 for m a l e s a n d 1 for f e m a l e s , respectively.  P r o s p e c t i v e evaluation of the  f o r m u l a in 36 patients g a v e a m e a n bias of +2% a n d a m e a n precision of 1 0 % . comparison,  the  Calvert  formula  and  the  Cockcroft-Gault  formula  gave  By  mean  b i a s / p r e c i s i o n v a l u e s o f - 3 / 1 3 % a n d - 1 7 / 1 8 % , respectively. S o r e n s e n et al. [43] d e v e l o p e d limited s a m p l i n g strategies for estimation carboplatin A U C from o n e or two p l a s m a concentrations.  of  T h e m o d e l s w e r e derived  from 15 ovarian c a n c e r patients a n d prospectively evaluated in nine patients receiving  11  the s a m e treatment.  M e a n bias/precision v a l u e s w e r e - 4 . 4 % / 1 3 . 9 % w h e n using o n e  p l a s m a concentration at 2.75 h post-infusion a n d - 2 . 2 % / 9 . 4 % w h e n using two p l a s m a c o n c e n t r a t i o n s (0.25 h a n d 2.75 h post-infusion).  T h e study w a s unique in that it w a s  the first predictive formula to incorporate H P L C m e a s u r e m e n t of free carboplatin in p l a c e of A A m e a s u r e m e n t of free platinum.  Interestingly, V a n W a r m e r d a m et al. [44]  a l s o prospectively tested this sampling strategy in nine patients, but u s e d free platinum levels rather than free carboplatin.  Bin et al. [45] d e v e l o p e d a dose-individualization  strategy using a population method with B a y e s i a n estimation b a s e d o n o n e or two p l a s m a s a m p l e s . P o p u l a t i o n data w e r e obtained from 2 2 pediatric patients using free platinum concentrations fit by the A D A P T II software p r o g r a m [46].  T h e test d a t a set  c o n s i s t e d of 2 3 patients receiving similar treatment. R e s u l t s from the B a y e s i a n a n a l y s i s w e r e c o m p a r e d to other m e t h o d s of A U C estimation. T h e "true" A U C w a s derived from m o d e l - i n d e p e n d e n t calculation by the trapezoidal rule.  U s i n g o n e d a t a point at 60 min  post-infusion, the m e d i a n bias and precision w e r e - 2 % a n d 3 % , respectively.  These  v a l u e s w e r e better than v a l u e s derived from predictive formulas b a s e d on renal function w h i c h w e r e , in turn, better than dosing b a s e d simply on body s u r f a c e a r e a .  12  1.3.  Analytical Methods for Carboplatin Determination  Overview Investigations of the clinical p h a r m a c o k i n e t i c and p h a r m a c o d y n a m i c b e h a v i o u r of carboplatin are limited by the availability of analytical m e t h o d s with suitable sensitivity and  specificity.  As  described  previously,  carboplatin  degrades  to  aquated,  h y d r o x y l a t e d , or chloro-substituted forms in biological fluids. S t u d i e s reporting sensitive detection of platinum c o m p l e x e s often sacrifice specificity for sensitivity by using atomic a b s o r p t i o n s p e c t r o s c o p y without a c h r o m a t o g r a p h i c c o m p o n e n t [47]. Efforts to preclude interference from degradation products while quantitating the parent c o m p l e x result in the n e e d for chromatographic s e p a r a t i o n s prior to detection.  G a s chromatographic  a s s a y s are not directly useful s i n c e platinum c o m p o u n d s d o not h a v e the required volatility.  Indirect methods, s u c h a s isotope dilution G C - M S [48], d o offer sensitive  detection of platinum but lack specificity owing to the s a m p l e handling p r o c e d u r e s e m p l o y e d . In the p r e s e n c e of degradation products and/or other platinum c o m p l e x e s , s u c h a s s a y m e t h o d s provide total platinum concentrations a n d thus offer no a d v a n t a g e o v e r A A with respect to specificity.  C h r o m a t o g r a p h i c t e c h n i q u e s a p p l i c a b l e to the  a n a l y s i s of platinum c o m p o u n d s include high-performance liquid c h r o m a t o g r a p h y a n d capillary e l e c t r o p h o r e s i s .  Capillary  electrophoresis  First introduced by J o r g e n s o n a n d L u k a c s in 1981 [49], capillary e l e c t r o p h o r e s i s i n v o l v e s the separation of drugs a n d other m o l e c u l e s within 10-100 c m long buffer-filled silica capillaries (25-75 (am internal diameter).  T h e principles are similar to t h o s e of  c o n v e n t i o n a l electrophoresis; however, the a p p a r a t u s is s c a l e d d o w n a n d optimized for drug a n a l y s i s . T h e large surface a r e a of the capillaries allows for g o o d heat dissipation; h e n c e , large potentials c a n be applied a c r o s s the capillary, improving the s e p a r a t i o n efficiency.  Capillary  electrophoresis  is  rapid,  involves  little  buffer  and  c o n s u m p t i o n , a n d d o e s not require equilibration between s a m p l e runs.  solvent Its  major  d r a w b a c k to date h a s b e e n poor sensitivity d u e to the small optical path of the o n capillary detection.  A n u m b e r of m o d e s of electrophoresis are p o s s i b l e , the  most  13  c o m m o n being capillary z o n e electrophoresis, in w h i c h electrophoretic migration rates are determined  by the c h a r g e - t o - m a s s ratio of a given m o l e c u l e .  Unfortunately,  capillary z o n e e l e c t r o p h o r e s i s is not applicable to the s e p a r a t i o n of cisplatin a n d carboplatin, s i n c e the m o l e c u l e s remain u n c h a r g e d in solution.  Micellar electrokinetic  c h r o m a t o g r a p h y , d e v e l o p e d by T e r a b e et al. [50], allows for the s e p a r a t i o n of neutral s p e c i e s v i a their interaction with detergent micelles a d d e d to the b a c k g r o u n d electrolyte. O u r application of this technique to the a n a l y s i s of the carboplatin a n a l o g u e s enloplatin and  D W A 2 1 1 4 R resulted  in  a  rapid  separation  of t h e s e two  e n d o g e n o u s c o m p o n e n t s of p l a s m a ultrafiltrate (Figure 1.5).  compounds  from  Unfortunately, the more  polar carboplatin s h o w e d little affinity for the micelles a n d could not b e s e p a r a t e d using t h e s e conditions. technique  for  A s well, sensitivity limitations  analysis  of  enloplatin  and  only allowed d e v e l o p m e n t of this  DWA211R  in  plasma  ultrafiltrate  at  concentrations b e t w e e n 7.5 a n d 60 ng/ml_ [51]. 2 1  0 Figure 1.5.  6  12  Time (min)  Electropherogram of (A) blank plasma ultrafiltrate and (B) plasma ultrafiltrate spiked with 50 y.g/mL of (1) enloplatin and (2) DWA2114R [51].  Carboplatin  showed little interaction with the micelles, eluting around 6 min. (Capillary: 47 (40) cm, voltage: 12 kV, detection: UV (200 nm), buffer: 150 mM sodium dodecyl sulfate in 5 mM phosphate, pH 7.)  14  High-performance  liquid  chromatography  M a n y platinum c o m p l e x e s , including carboplatin, lack a strong  chromophore  resulting in p o o r sensitivity for H P L C a s s a y s with U V detection [52-54]. T h e s e m e t h o d s typically h a v e detection limits around 0.5-1 jj.g/mL.  F o r the carboplatin d o s e s typically  u s e d , this results in a n inability to quantitate parent drug within a s little a s 6-8 h postadministration.  T h u s , investigators h a v e sought alternative s a m p l e p r o c e s s i n g or  detection s c h e m e s to e n h a n c e their existing H P L C s y s t e m s . H P L C - A A is sensitive but requires collection of eluent fractions prior to s p e c t r o s c o p i c a n a l y s i s , thereby presenting significant reproducibility problems.  S e n s i t i v e H P L C m e t h o d s using e l e c t r o c h e m i c a l  detection h a v e b e e n reported for cisplatin, but ligand effects on the central platinum metal inhibit the electroactivity of, a n d thus sensitivity towards, carboplatin [55]. with on-line differential  pulse polarography  [56]  carboplatin quantitation;  however, the t e c h n i q u e is difficult to  HPLC  has been successfully used reproduce and  for the  n e c e s s a r y e q u i p m e n t not c o m m o n l y available. A variety of m a s s spectrometric detection t e c h n i q u e s are currently a v a i l a b l e a n d c a n be u s e d to improve both the sensitivity a n d selectivity of a given H P L C s e p a r a t i o n . A s p e c i f i c H P L C - M S method with ionization via fast atom b o m b a r d m e n t [57] h a s b e e n d e v e l o p e d to e v a l u a t e the stability of carboplatin in infusion fluids.  HPLC-ICP-MS  m e t h o d s h a v e b e e n reported by Z h a o et al. [58] for determination of cisplatin a n d p o s s i b l e metabolites a n d by C a i r n s et al. [59] for the platinum c o m p o u n d J M - 2 1 6 a n d its d e g r a d a t i o n products.  Recently, Falter a n d W i l k e n [60] applied H P L C - I C P - M S to the  determination of cisplatin a n d carboplatin in environmental s a m p l e s . O u r laboratory h a s a l s o e x a m i n e d the applicability of H P L C - M S t e c h n i q u e s for the a n a l y s i s of carboplatin and  its  cyclobutanedicarboxylato  analogues  enloplatin  and  DWA2114R.  For  carboplatin, detection limits in a q u e o u s solution w e r e only about three-fold better for H P L C - E S - M S [61,62] than for H P L C - U V , but at least twenty-fold better for H P L C - I C P M S [62] than for H P L C - U V , w h e n t h e s e t e c h n i q u e s w e r e c o m p a r e d under equivalent r e v e r s e p h a s e c h r o m a t o g r a p h i c conditions.  Specificity of the H P L C - I C P - M S  method  w a s at least partially demonstrated in the p r e s e n c e of e n d o g e n o u s p l a s m a ultrafiltrate c o m p o n e n t s (Figure 1.6), a n d a n a b s o l u t e detection limit of 20 pg w a s obtained for  15  a q u e o u s carboplatin s t a n d a r d s . H o w e v e r , the technique w a s not p u r s u e d further d u e to its high cost a n d the lack of availability of dedicated H P L C - I C P - M S e q u i p m e n t .  1  0  Figure 1.6.  4  8  12  Time (min)  H P L C - I C P - M S chromatograms of (A) blank plasma ultrafiltrate and (B) plasma ultrafiltrate containing 20 ng/mL each of (1) carboplatin, (2) enloplatin, and (3) DWA2114R [59]. Chromatographic conditions included a Y M C O D S - A Q 150 x 4.6 mm column, mobile phase of 15% methanol in water, and 20 pi. injection volume.  16  Pre-column  derivatization  techniques,  based  on  nucleophilic  substitution  r e a c t i o n s with thiourea [63), o - p h e n y l e n e d i a m i n e [64], or diethyldithiocarbamate provide sensitive detection of platinum II s p e c i e s .  [65],  Unfortunately, all c h r o m a t o g r a p h i c  m e t h o d s using p r e - c o l u m n derivatization lack specificity for the parent platinum c o m p l e x and  are therefore  not  methods  of c h o i c e .  Detection  using on-line  post-column  derivatization c a n provide a s s a y m e t h o d s with both sensitivity a n d specificity generating a product with improved detection characteristics after isolation.  by  chromatographic  Ideally, t h e s e online m e t h o d s e m p l o y reactions w h i c h are rapid, g e n e r a t e a  m i n i m u m b a c k g r o u n d s i g n a l , a n d produce the d e s i r e d c o m p o u n d in high yield.  HPLC-  P C m e t h o d s e m p l o y i n g p o t a s s i u m dichromate activation followed by s o d i u m bisulfite transformation h a v e a l r e a d y b e e n d e v e l o p e d a n d validated for the determination  of  cisplatin in p l a s m a ultrafiltrate [66,67]. K i z u et al. [68] h a v e d e v e l o p e d a method b a s e d o n direct reaction of divalent a n d quadrivalent platinum c o m p l e x e s with s o d i u m bisulfite. U s i n g a 100 jaL injection, the detection limit of this s y s t e m for carboplatin w a s 6 0 n M (22 n g / m L ) , although the authors did not apply the s y s t e m to quantitation of carboplatin in biological s a m p l e s . D e s p i t e the poor limits of quantitation  o b s e r v e d for conventional  HPLC-UV  a n a l y s i s of carboplatin, m e a s u r e s c a n be taken to improve the sensitivity of this technique.  An  ultraviolet  spectrum  of  carboplatin  a b s o r b a n c e of carboplatin is greatest below 2 3 5 n m .  (Figure  1.7)  s h o w s that  the  Unfortunately, m o s t c o m p o u n d s  a b s o r b U V radiation in this wavelength region; h e n c e , a n a l y s i s of carboplatin in the p r e s e n c e of e n d o g e n o u s ultrafiltrate c o m p o n e n t s provides r e s e a r c h e r s with s o m e t h i n g of a c o n u n d r u m .  If a longer analytical w a v e l e n g t h is c h o s e n , then the  sensitivity of the method will be poor.  inherent  H o w e v e r , if a shorter analytical w a v e l e n g t h is  s e l e c t e d in order to m a x i m i z e the carboplatin a b s o r b a n c e , then s a m p l e interferences are problematic. Furthermore, s a m p l e injection v o l u m e s must be s m a l l (typically 20 \xL) to prevent overloading of the c o l u m n a n d s u b s e q u e n t l o s s of resolution  between  carboplatin a n d p l a s m a ultrafiltrate c o m p o n e n t s . The procedures  k e y to improving H P L C - U V a s s a y m e t h o d s lies in the s a m p l e handling employed.  R e m o v a l of  endogenous  interferences  increases  assay  17  sensitivity  by  allowing  for  injection  of  larger  volumes, more  rapid  and  efficient  c h r o m a t o g r a p h i c s e p a r a t i o n s , a n d monitoring using lower U V w a v e l e n g t h s .  Indeed,  A l l s o p p et al. [69] h a v e already d e v e l o p e d a column-switching H P L C a s s a y method e m p l o y i n g U V detection at 2 1 0 n m , w h i c h results in a m u c h improved limit of detection (14 n g / m L ) . Unfortunately, the s y s t e m requires c o m p l i c a t e d e q u i p m e n t not a v a i l a b l e in m o s t laboratories.  A simpler a p p r o a c h to s a m p l e c l e a n - u p involves the u s e of s o l i d -  p h a s e extraction  cartridges to  retain  carboplatin while  interfering  s u b s t a n c e s are  w a s h e d through.  T h e carboplatin c a n then be eluted from the cartridge, dried under  nitrogen g a s , a n d reconstituted into a s m a l l e r s a m p l e v o l u m e to i n c r e a s e concentration. S o l i d - p h a s e extraction without s u b s e q u e n t s a m p l e concentration h a s a l r e a d y b e e n a p p l i e d to a n a l y s i s of the carboplatin a n a l o g u e C I - 9 7 3 in p l a s m a ultrafiltrate [70]. T h u s , d e s p i t e the lack of a n inherently strong c h r o m o p h o r e , m o r e s e n s i t i v e detection of carboplatin by H P L C - U V m e t h o d s is p o s s i b l e .  230  Figure 1.7.  260  290  Wavelength (nm)  Ultraviolet spectrum of carboplatin in aqueous solution.  Carboplatin lacks a  specific chromophore but allows for nonspecific U V detection at wavelengths of 230 nm or less.  18  1.4.  Hypothesis T h e hypothesis of this thesis is that it is f e a s i b l e to u s e H P L C a s s a y m e t h o d s to  quantitate carboplatin in p l a s m a ultrafiltrate in order to c h a r a c t e r i z e its p h a r m a c o k i n e t i c b e h a v i o u r in y o u n g patients.  1.5.  Thesis Objectives and Rationale  Objective #1 -- T o d e v e l o p a n d validate H P L C a s s a y m e t h o d s for carboplatin b a s e d on direct U V detection a n d o n U V detection following p o s t - c o l u m n derivatization. T h e lack of sensitivity of H P L C m e t h o d s u s e d for clinical evaluation h a s resulted in poor characterization of carboplatin elimination.  B a s e d on previously p u b l i s h e d  elimination profiles of carboplatin in adult [56] a n d pediatric [39] patients, a limit of quantitation  of  characterize  approximately  carboplatin  20  elimination  ng/mL up  should to  24  h  provide  sufficient  sensitivity  post-administration.  to  Improved  c o m p a r i s o n s of the elimination profiles of free carboplatin a n d free platinum will then be possible.  Objective  #2 — T o s y n t h e s i z e structural a n a l o g u e s of carboplatin to act a s internal  s t a n d a r d s for the H P L C a s s a y m e t h o d s . Internal s t a n d a r d s are u s e d in c h r o m a t o g r a p h i c a s s a y s to a c c o u n t for l o s s e s during s a m p l e p r o c e s s i n g p r o c e d u r e s a n d to a c c o u n t for other variability present in the m e t h o d or a p p a r a t u s .  S y n t h e s i s of a closely related carboplatin a n a l o g u e is required  b e c a u s e other a n a l o g u e s available are too hydrophobic a n d h a v e inappropriately long retention under c h r o m a t o g r a p h i c conditions required for carboplatin a n a l y s i s .  Objective  #3 -  T o investigate the clinical p h a r m a c o k i n e t i c s of carboplatin in y o u n g  patients. Development  of clinical dosing strategies for  carboplatin  rely  on  accurate  determination of p h a r m a c o k i n e t i c parameters, in particular c l e a r a n c e a n d A U C . S o m e clinical s t u d i e s h a v e a s s u m e d that free carboplatin a n d free platinum m e a s u r e m e n t s are  19  i n t e r c h a n g e a b l e [39,44] a n d therefore provide equivalent p h a r m a c o k i n e t i c parameter e s t i m a t e s of elimination a n d e x p o s u r e . evaluated.  H o w e v e r , this a s s u m p t i o n h a s not b e e n fully  C o m p a r i s o n of parameters determined from both free carboplatin a n d free  platinum m e a s u r e m e n t s is required to demonstrate whether or not H P L C a n d A A a s s a y t e c h n i q u e s c a n be u s e d interchangeably.  1.6.  References  1.  B. Rosenberg, L. Van Camp, and T. Krigas. Inhibition of cell division in Escherichia coli by electrolysis products from a platinum electrode. Nature 205: 698-699 (1965).  2.  P.J. Loehrer and L.H. Einhorn. Cisplatin: diagnosis and treatment. Ann Intern Med 100: 704-713 (1984).  3.  D.L  Bodenner,  P.C.  diethyldithiocarbamate  Dedon,  P.C.  Keng,  and  on c/s-diamminedichloroplatinum  R.F.  Borch.  II-induced  cross-linking, and gamma-glutamyltranspeptidase inhibition.  Effect  cytotoxicity,  of DNA  Cancer Res 46: 2745-2750  (1986). 4.  R. Safirstein, J . Winston, M. Goldstein, D. Moel, S. Dikman, and J . Guttenplan. Cisplatin nephrotoxicity. Amer J Kidney Dis 8: 356 (1987).  5.  J . Uozumi and C.L. Litterst. The effect of cisplatin on renal A T P a s e activity in vivo and in vitro. Cancer Chemother Pharmacol 15: 93-96 (1985).  6.  L.A. Zwelling. Response  Cisplatin and new analogs.  Modifiers  Annual  10.  In Cancer Chemotherapy  and  B.A. Chabner et al. (Editors), Elsevier  Biological Science  Publishing, New York, 64-72 (1988). 7.  E. Reed and K.W. Kohn. Platinum analogues. In Cancer Chemotherapy: Practice.  Principles  and  B.A. Chaber and J.M. Collins (Editors), J.B. Lippincott, Philadelphia, 465-490  (1990). 8.  K.R. Harrap.  Preclinical studies identifying carboplatin as a viable cisplatin alternative.  Cancer Treatment Rev 12: 21-33 (1985). 9.  A.H. Calvert, S.J. Harland, D.R. Newell, Z.H. Siddik, A . C . Jones, T.J. McElwain, K.S. Raju, E. Wiltshaw, I.E. Smith, J.M. Baker, M.J. Peckham, and K.R. Harrap.  Early  20  clinical studies with c/'s-diammine-1,1-cyclobutanedicarboxylate platinum  II.  Cancer  Chemother Pharmacol 9: 140-147 (1982). 10  B.D. Evans, K.S. Raju, A.H. Calvert, S . J . Harland, and E. Wiltshaw. Phase II study of J M 8 , a new platinum analog in ovarian cancer.  Cancer Treatment Rep 67: 997-1000  (1983). 11.  M.K. Wolpert-DeFilippes. Antitumour activity of cisplatin analogs. In Cisplatin: Status and New Developments.  Current  A . W . Prestayko, S T . Crooke, and S.K. Carter (Editors),  Academic Press, New York, 183-191 (1980). 12.  P.S. Gaynon. Carboplatin in pediatric malignancies. Semin Oncol 21 (5 Suppl 12): 65-76 (1994).  13.  L.R. Kelland. New platinum antitumour complexes. Critical Rev Oncol/Hematol  15: 191-  219 (1993). 14.  N. Farrell.  Structurally novel platinum antitumour compounds.  Metal Coordination  Compounds in Cancer Chemotherapy.  In Platinum and Other  S B . Howell (Editor), Plenum  Press, New York, 345-355 (1991). 15.  L. Cattalini. The intimate mechanism of replacements in d square-planar complexes. In 8  Inorganic Reaction Mechanisms.  J.O. Edward (Editor), John Wiley & Sons, New York,  263-327 (1970). 16.  M.A. Allsopp, G . J . Sewell, C.G. Rowland, C M . Riley, and R.L. Schowen.  The  degradation of carboplatin in aqueous solutions containing chloride or other selected nucleophiles. Int J Pharmaceutics 17.  N.D. Tinker, H.L. Sharma,  69: 197-210 (1991).  and C.A. McAuliffe.  Qualitative investigation of the  metabolites formed by cisplatin and paraplatin involving H P L C analysis. In Platinum  18.  Proceedings  and  Other Metal Coordination  Compounds in Cancer Chemotherapy:  of the Fifth  International Symposium.  M. Nicolini (Editor), Martinus Nijhoff, Boston, 144-159 (1987).  M.C. Lim and R.B. Martin. The nature of cis amine P d II and antitumor cis amine Pt II complexes in aqueous solutions. Inorg Nucl Chem 38: 1911 (1976).  19.  A. Andersson, H. Hedenmalm, B. Elfsson, and H. Ehrsson. Determination of the acid dissociation constant for c/'s-diammineaquachloroplatinum II ion. A hydrolysis product of cisplatin. J Pharm Sci 83: 859-862 (1993).  20.  P.T. Daley-Yates, D.C. McBrien. The inhibition of renal A T P a s e by cisplatin and some biotransformation products. Chem Biol Interact 40: 325-334 (1982).  21  21.  A. Meister.  Selective modification of glutathione metabolism.  Science 220: 471-477  (1983). 22.  F.A. Blommaert, H.C. Van Dijk-Knijnenburg, F.J. Dijt, L. Den Engelse, R.A. Baan, F. Berends, and A.M. Fichtinter-Schepman.  Formation of D N A adducts by the anticancer  drug carboplatin: different nucleotide sequence preferences in vitro and in cells.  Biochem  34: 8474-8480 (1995). 23.  M. Tonetti, M. Giovine, A. Gasparini, U. Benatti, and A. De Flora. Enhanced formation of  reactive  species  from  c/'s-diammine-(1,1-cyclobutanedicarboxylato)-platinum  II  (carboplatin) in the presence of oxygen free radicals. Biochem Pharmacol 46: 1377-1383 (1993). 24.  G. Los, E. Verdegaal, H.P. Noteborn, M. Ruevekamp, A. De Graeff, E.W. Meesters, D. T. Huinink, and J.G. McVie.  Cellular pharmacokinetics of carboplatin and cisplatin in  relation to their cytotoxic action. Biochem Pharmacol 42: 357-363 (1991). 25.  W.J. Van der Vijgh.  Clinical pharmacokinetics of carboplatin.  Clin Pharmacokinet  21:  242-261 (1991). 26.  S.B. Duffull and B.A. Robinson. carboplatin. Clin Pharmacokinet  27.  Clinical pharmacokinetics and dose optimisation of  33: 161-183(1997).  R.C. Gaver, A.M. George, and G. Deeb.  In vitro stability, plasma protein binding and  blood cell partitioning of 14C-carboplatin.  Cancer Chemother  Pharmacol  20: 271-276  (1987). 28.  B.T. Sorensen, A. Stromgren, P. Jakobsen, J.T. Nielsen. L.S. Andersen, and A. Jakobsen.  Renal handling of carboplatin.  Cancer Chemother  Pharmacol  30: 317-320  (1992). 29.  M.L. De Lemos.  Application of the area under the curve of carboplatin in predicting  toxicity and efficacy. Cancer Treat Rev 24: 407-414 (1998). 30.  M.J. Egorin, D.A. Van Echo, E.A. Olman, M.Y. Whitacre, A. Forrest, and J . Aisner. Prospective validation  of a pharmacologically based dosing scheme for the c/s-  diamminedichloroplatinum  II  analogue  diamminecyclobutanedicarboxylatoplatinum.  Cancer Res 45: 6502-6506 (1985). 31.  A.H. Calvert, D.R. Newell, L A . Gumbrell, S. O'Reilly, M. Burnell, F.E. Boxall, Z.H. Siddik, I.R. Judson, M.E. Gore, and E. Wiltshaw.  Carboplatin dosage: prospective  evaluation of a simple formula based on renal function. J Clin Oncol 7: 1748-1756 (1989).  22  32.  C. Chantler, E.S. Garnett, V. Parsons, and N. Veall.  Glomerular filtration rate  measurement in man by the single injection method using 51Cr-EDTA. J Clin Sci 37: 169190 (1969). 33.  D.W.  Cockcroft  and M.H. Gault.  Prediction of creatinine clearance from  serum  creatinine. Nephron 16: 31-41 (1976). 34.  M.J. Millward, L.K. Webster, G.C. Toner, J.F. Bishop, D. Rischin, K.H. Stokes, V.K. Johnston, and R. Hicks. Carboplatin dosing based on measurement of renal function: experience at the Peter MacCallum Cancer Institute. Aust NZ J Med 26: 372-379 (1996).  35.  L.J Van Warmerdam, S. Rodenhuis, W.W. Ten Bokkel Huinink, R.A. Maes, and J.H. Beijnen. Evaluation of formulas using the serum creatinine level to calculate the optimal dosage of carboplatin. Cancer Chemother Pharmacol 37: 266-270 (1996).  36.  F. Doz, L. Brugieres, G. Bastian, E. Quintana, J . Lemerle, and J.-M. Zucker. Clinical trial and pharmacokinetics of carboplatin 560 m g / m in children. Med Ped Oncol 18: 4592  465 (1990). 37.  D.R. Newell, A.D. Pearson, K. Balmanno, L. Price, R.A. Wyllie, M. Keir, A.H. Calvert, I.J. Lewis, C.R. Pinkerton, and M.C. Stevens. Carboplatin pharmacokinetics in children: the development of a pediatric dosing formula. J Clin Oncol 11: 2314-2323 (1993).  38.  N.M. Marina, J.H. Rodman, D.J. Murry, S.J. Shema, L.C. Bowman, D.P. Jones, W. Furman, W.H. Meyer, and C.B. Pratt.  Phase I study of escalating targeted doses of  carboplatin combined with ifosfamide and etoposide in treatment of newly diagnosed pediatric solid tumors. J Nat Cancer Inst 86: 544-548 (1994). 39.  R. Riccardi, A. Riccardi, A. Lasorella, C. Di Rocco, G. Carelli, A. Tornesello,  T.  Servidei, A. lavarone, and R. Mastrangelo. Clinical pharmacokinetics of carboplatin in children. Cancer Chemother Pharmacol 33: 477-483 (1994). 40.  T. Madden, M. Sunderland, V.M. Santana, and J.H. Rodman. The pharmacokinetics of high-dose carboplatin in pediatric patients with cancer. Clin Pharmacol  Ther51:  701-707  (1992). 41.  E. Chatelut, P. Canal, V. Brunner, C. Chevreau, A. Pujol, A. Boneu, H. Roche, G. Houin, and R. Bugat.  Prediction of carboplatin from standard morphological and  biological patient characteristics. J Nat Cancer Inst 87: 573 (1995). 42.  L.B. Sheiner and  S.L. Beal.  Evaluation of methods for estimating  population  pharmacokinetics parameters. I. Michaelis-Menten model: routine clinical pharmacokinetic data. J Pharmacokinet  Biopharm 8(6): 553-571 (1980).  23  43.  B.T. Sorensen, A. Stromgren, P. Jakobsen, and A. Jakobsen. strategy for estimation of carboplatin area under the curve.  A limited sampling Cancer  Chemother  Pharmacol 31: 324-327 (1993). 44.  L.J. Van Warmerdam, S. Rodenhuis, O. Van Tellingen, R.A. Maes, and J.H. Beijnen. Validation of a limited sampling model for carboplatin in a high-dose chemotherapy combination. Cancer Chemother Pharmacol 35: 179-181 (1994).  45.  P. Bin, A.V. Boddy, M. Cole, A.D. Pearson, E. Chatelut, H. Rubie, and D.R. Newell. Comparison of methods for the estimation of carboplatin pharmacokinetics in paediatric cancer patients. Eur J Cancer 31 A: 1804-1810(1995).  46.  D.Z. and A. Schumitzky.  A D A P T II user's guide: pharmacokinetic/pharmacodynamic  systems analysis softward. Biomedical simulations resource, Los Angeles, C A , U S A . 47.  D.A. Hull, N. Muhammad, J . G . Lanese, S.D. Reich, T.T. Finkelstein, and S. Fandrich. Determination of platinum in serum and ultrafiltrate by flameless atomic absorption spectrophotometry. J Pharm Sci 70: 500-502 (1981).  48.  S.K. Aggarwal, N.W. Gemma, M. Kinter, J . Nicholson, J.R. Shipe, and D.A. Herold. Determination of platinum in urine, ultrafiltrate, and^whole plasma by isotope dilution gas chromatograph compared to electrothermal atomic absorption spectrometry.  Anal  Biochem 210: 113-118(1993). 49.  J.W. Jorgenson  and  K.D. Lukacs.  Zone electrophoresis in open-tubular  glass  capillaries. Anal Chem 53: 1298-1302(1981). 50.  S. Terabe. Selectivity manipulation in micellar electrokinetic chromatography. BiomedAnal  51.  10: 705-715(1992).  R.B. Burns and L. Embree.  Comparison of high-performance liquid chromatographic  and capillary eiectrophoretic analysis of DWA2114R in plasma ultrafiltrate. Meeting Proceedings 52.  J Pharm  AFPC  Annual  (1996).  S.J. Harland, D.R. Newell, Z.H. Siddik, R. Chadwick, A.H. Calvert, and K.R. Harrap. Pharmacokinetics of c/s-diammine-1,1-cyclobutanedicarboxylate platinum II in patients with normal and impaired renal function. Cancer Res 44: 1693-1697 (1984).  53.  R.C. Gaver and G. Deeb.  High-performance liquid chromatographic procedures for the  analysis of carboplatin in human plasma ultrafiltrate.  Cancer Chemother  Pharmacol  16:  201-206(1986).  24  54.  R.C. Gaver, A.M. George, G.F. Duncan, A.D. Morris, G. Deeb, H.C. Faulkner, and R.H. Farmen. The disposition of carboplatin in the beagle dog. Cancer Chemother  Pharmacol  21: 197-202 (1988). 55.  I.S. Krull, X.-D. Ding, S. Braverman, and C. Selavka. anticancer drugs using L C E C . J Chromatogr Sci'21:  56.  F. Elferink, W.J. Van  der Vijgh, and  H.M.  Trace analysis for c/s-platinum  166-173 (1983). Pinedo.  On-line differential  polarographic detection of carboplatin in biological samples after  pulse  chromatographic  separation. Anal Chem 58: 2293-2296 (1986). 57.  J.A. Hadfield, A.T. McGown, M.J. Dawson, N. Thatcher, and B.W. Fox. The suitability of carboplatin solutions for 14-day continuous infusion by ambulatory pump: an H P L C dynamic F A B study. J Pharm Biomed Anal 11: 723-727 (1993).  58.  Z. Zhao, K. Tepperman, J.G. Dorsey, and R.C. Elder.  Determination of cisplatin and  some possible metabolites by ion-pairing chromatography with inductively coupled plasma mass spectrometric detection. J Chromatogr 615: 83-89 (1993). 59.  W.R. Cairns, L. Ebdon, and S.J. Hill. Development of an H P L C - I C P - M S method for the determination of platinum species from new antitumour drugs.  Anal Proc 31: 295-297  (1994). 60.  R. Falter and R.-D. Wilken.  Determination of carboplatinum and cisplatinum by  interfacing H P L C with ICP-MS using ultrasonic nebulisation. Sci Total Environ 225: 167176 (1999). 61.  R.B. Burns, R.W. Burton, S.A. Albon, and L. Embree.  Liquid chromatography-mass  spectrometry for the detection of platinum antineoplastic complexes. J Pharm  Biomed  Anal 14: 367-372 (1996). 62.  R.B. Burns and L. Embree. Analysis of antineoplastic platinum complexes by highperformance liquid chromatography with inductively coupled plasma and electrospray mass spectrometric detection. Pharm Research 12: S 7 0 (1995).  63.  J.D. Woollins, A. Woollins, and B. Rosenberg. fr-a/7S-Pt(NH ) CI 3 2  2  in the  presence of  The detection of trace amounts of  c/s-Pt(NH ) Cl2. 3  2  A  high-performance  liquid  chromatographic application of Kurnakow's test. Polyhedron 2: 175-178 (1983). 64.  H.  Hasson  and  A.  Warshawsky.  High-performance  liquid  chromatographic  determination of c/s-diamminedichloroplatinum II (cisplatin) as the o-phenylenediamine complex. J Chromatogr 53: 219-221 (1990).  25  65.  R. Goel, P.A. Andrews, C.E. Pfeifle, I.S. Abramson, S. Kirmani, and S.B. Howell. Comparison of the pharmacokinetics of ultrafilterable cisplatin species detectable by derivatization with diethyldithiocarbamate or atomic absorption spectroscopy.  Eur J  Cancer 26: 21-27 (1990). 66.  H.H. Farrish, P.-H. Hsyu, J.F. Pritchard, K.R. Brouwer, and J . Jarrett. Validation of a liquid chromatography post-column derivatization assay for the determination of cisplatin in plasma. J Pharm Biomed Anal 12: 265-271 (1994).  67.  M. Kinoshita, N. Yoshimura, and H. Ogata.  High-performance liquid chromatographic  analysis of unchanged c/s-diamminedichloroplatinum (cisplatin) in plasma and urine with post-column derivatization. J Chromatogr 529: 462-467 (1990). 68.  R. Kizu, T. Yamamoto, T. Yokoyama, M. Tanaka, and M. Miyazaki.  A sensitive  postcolumn derivatization/UV detection system for H P L C determination of antitumor divalent and quadrivalent platinum complexes. Chem Pharm Bull 43: 108-114 (1995). 69.  M.A.  Allsopp,  G.J.  Sewell,  and  C.G.  Rowland.  A  column-switching  chromatographic assay for the analysis of carboplatin in plasma ultrafiltrate.  liquid  J Pharm  Biomed Anal 10: 375-381 (1992). 70.  W.W. Bullen, L.D. Andress, T. Chang, L.R. Whitfield, M.L. Welch, and R.A. Newman. A high-performance liquid chromatographic assay for CI-973, a new anticancer platinum diamine complex, in human plasma and urine ultrafiltrates. Cancer Chemother  Pharmacol  30:193-198(1992).  26  CHAPTER 2 DEVELOPMENT AN HPLC-UV  2.1.  AND PRELIMINARY ASSAY  METHOD  EVALUATION  FOR  OF  CARBOPLATIN  Introduction Quantitation of carboplatin in the p r e s e n c e of its in vitro a n d in vivo degradation  products requires liquid chromatographic s e p a r a t i o n s prior to detection. D u e to its polar nature, the first H P L C a s s a y s for determination of carboplatin in biological s a m p l e s e m p l o y e d normal p h a s e s e p a r a t i o n s on silica [1], diol [2], or a m i n o [3] c o l u m n s . T h e s e m e t h o d s are limited by poor sensitivity, providing detection limits a r o u n d 0.5 n g / m L , a n d a l s o require significant a m o u n t s of o r g a n i c solvents in mobile p h a s e preparation.  In  contrast, H P L C a s s a y m e t h o d s b a s e d on r e v e r s e p h a s e s e p a r a t i o n of carboplatin from p l a s m a c o m p o n e n t s h a v e the a d v a n t a g e of r e d u c e d o r g a n i c solvent c o n s u m p t i o n . Furthermore, the greater efficiency of reverse p h a s e v e r s u s normal p h a s e s e p a r a t i o n s results in i n c r e a s e d a s s a y sensitivities. carboplatin  make  development  of  H o w e v e r , the p h y s i c o c h e m i c a l properties of  reverse  phase  methods  particularly  difficult.  C a r b o p l a t i n h a s minimal k' v a l u e s on most O D S c o l u m n s , while the a b s e n c e of a s p e c i f i c c h r o m o p h o r e n e c e s s i t a t e s U V detection in the n o n s p e c i f i c a b s o r b a n c e region (below 2 3 0 nm).  Unfortunately, p l a s m a - b a s e d s a m p l e s contain a myriad of polar  constituents that are a l s o detected at t h e s e w a v e l e n g t h s .  A viable r e v e r s e p h a s e  m e t h o d for carboplatin requires either s a m p l e pretreatment or more selective detection t e c h n i q u e s . T o further complicate matters, carboplatin is not sufficiently h y d r o p h o b i c to be retained o n m o s t s o l i d - p h a s e extraction cartridges a n d will not partition extensively into w a t e r - i m m i s c i b l e s o l v e n t s , which are m e t h o d s c o m m o n l y u s e d for removal of interferences.  T h u s , investigators h a v e relied mainly on a d v a n c e m e n t s in  HPLC  t e c h n o l o g y or alternative detection strategies to improve the sensitivity a n d specificity of traditional  HPLC  methods  for  carboplatin  quantitation.  T h e s e alternative  HPLC  m e t h o d s h a v e a l r e a d y b e e n reviewed in C h a p t e r 1, including H P L C - U V with c o l u m n switching t e c h n o l o g y [4], H P L C with on-line differential p u l s e polarography [5], a n d HPLC-MS  [6,7].  The  utility of H P L C - U V  following  p o s t - c o l u m n derivatization  of  carboplatin with s o d i u m bisulfite is fully e v a l u a t e d a n d d i s c u s s e d in C h a p t e r 3.  27  T h i s chapter d e s c r i b e s the d e v e l o p m e n t of a n H P L C - U V a s s a y method a n a l y s i s of carboplatin  in p l a s m a ultrafiltrate.  The chromatographic  for  behaviour  of  carboplatin on a n u m b e r of reverse p h a s e c o l u m n s a n d s u b s e q u e n t s e p a r a t i o n of carboplatin from its nucleophilic substitution products a n d e n d o g e n o u s c o m p o n e n t s of p l a s m a ultrafiltrate is reported.  A s well, the utility of various s o l i d - p h a s e extraction  cartridges to facilitate s a m p l e c l e a n - u p is e x a m i n e d . Following preliminary d e v e l o p m e n t of c h r o m a t o g r a p h i c  conditions for quantitative  analysis, estimates of the  limits  of  detection a n d quantitation of the a s s a y are m a d e .  2.2.  Experimental  2.2.1. Buffers and mobile phases W a t e r u s e d for all s a m p l e s , buffers, a n d mobile p h a s e s w a s H P L C - g r a d e a n d p r o d u c e d on-site by reverse o s m o s i s a n d s u b s e q u e n t filtration using a Milli-Q s y s t e m (Millipore, B e d f o r d , M A , U S A ) . Buffers u s e d during d e v e l o p m e n t of the H P L C - U V method w e r e adjusted to the d e s i r e d p H v i a addition of a n equimolar concentration of the b a s e form of a particular buffer salt to the acid form until the d e s i r e d p H w a s r e a c h e d , a s indicated by monitoring o n a n A c c u m e t m o d e l 2 2 0 p H meter (Fisher Scientific, N e p e a n , O N , C a n a d a ) . M o b i l e p h a s e s w e r e filtered through a 0.45 (im Nylaflow nylon m e m b r a n e filter ( G e l m a n S c i e n c e s , A n n Arbor, M l , U S A ) a n d d e g a s s e d under v a c u u m prior to u s e .  2.2.2. Preparation of plasma ultrafiltrate from blood samples Blood  obtained  from  adult  volunteers  was  first  centrifuged  at  ambient  temperature for 10 min at 3 0 0 0 rpm in a G P Centrifuge ( B e c k m a n Instruments,  Palo  Alto, C A , U S A ) . T o obtain p l a s m a ultrafiltrate, a n aliquot (0.7-1.0 mL) of the p l a s m a w a s s u b s e q u e n t l y p l a c e d within a Centrifree micropartition unit ( A m i c o n Inc., D a n v e r , M A , U S A ) a n d centrifuged for 60 min at 15 °C a n d 4 5 0 0 rpm (2000 g) in a B e c k m a n J 2 21 ultracentrifuge.  28  2.2.3. Apparatus F o r preliminary evaluation of carboplatin chromatography (sections 2.2.4 through 2.2.6),  a  Hewlett-Packard  model  1050 liquid  Avondale, P A , U S A ) w a s employed.  chromatograph  (Hewlett  Packard,  This s y s t e m c o n s i s t e d of a quaternary p u m p ,  v a r i a b l e - w a v e l e n g t h detector, a n d model 3 3 9 6 A integrator.  S a m p l e injection e m p l o y e d  a R h e o d y n e 7 1 2 5 injector (Alltech A s s o c i a t e s , Deerfield, IL, U S A ) with a 2 0 |al_ s a m p l e loop. S u b s e q u e n t e x p e r i m e n t s w e r e performed o n a W a t e r s H P L C s y s t e m (Waters Limited, Milford, M A , U S A ) , w h i c h c o n s i s t e d of a m o d e l 5 1 0 p u m p , a m o d e l 7 1 2 W I S P autoinjector, a n d a m o d e l 4 8 4 variable wavelength U V detector.  For data analysis,  p e a k integrations w e r e performed using the W a t e r s M a x i m a 8 2 0 c o m p u t e r software. P e a k height or a r e a v a l u e s w e r e then exported to E x c e l (Microsoft C o r p o r a t i o n , R e d m o n d , W A , U S A ) for d a t a manipulation a n d statistical a n a l y s e s . 2.2.4. C o l u m n evaluation C h r o m a t o g r a p h i c properties of carboplatin o n s e v e r a l c o m m e r c i a l r e v e r s e p h a s e H P L C c o l u m n s w e r e e v a l u a t e d . T h e s e included W a t e r s ^ B o n d a p a k a n d N o v a p a k O D S c o l u m n s , a V y d a c 2 0 1 T P O D S c o l u m n (The S e p a r a t i o n s G r o u p , H e s p e r i a , C A , U S A ) , a n d a Y M C O D S - A Q c o l u m n (Waters).  F o r c o m p a r i s o n , a normal p h a s e |aBondapak  a m i n o ( N H ) c o l u m n w a s a l s o e v a l u a t e d . T a b l e 2.1 lists the p h y s i c a l properties of t h e s e 2  c o l u m n s , including the s h a p e , diameter, pore s i z e , a n d percent c a r b o n loading of e a c h p a c k i n g material. Table 2.1. Physical properties of H P L C columns evaluated for carboplatin chromatography. Particle Shape  Particle Diameter  Pore Size  Carbon Load  I^Bondapak N H (3.9 x 300 mm)"  irregular  10 nm  125 A  3.5%  (iBondapak O D S (3.9 x 150 mm)  irregular  10 |im  125 A  10%  Vydac 201 T P O D S (4.6 x 150 mm)"  spherical  5 |am  300 A  unknown  Novapak O D S (3.9 x 150 mm)  spherical  4 |am  60 A  7%  Y M C O D S - A Q (4.6 x 150 mm)  spherical  3 p.m  120 A  14%  Column (Dimensions) 2  " non-endcapped columns  29  T o m a x i m i z e the chromatography o b s e r v e d o n the H P L C c o l u m n s , efficiency a n d c a p a c i t y factor v a l u e s for the reverse p h a s e c o l u m n s w e r e determined using a mobile p h a s e of 1 0 0 % water.  F l o w rates w e r e c h o s e n within t h e optimal flow-rate range listed  by the manufacturers a n d w e r e 0.5 m L / m i n for the 3.9 x 150 m m c o l u m n a n d 0.7 m L / m i n for the 4 . 6 x 150 m m c o l u m n s .  F o r the a m i n o c o l u m n , c h r o m a t o g r a p h i c  conditions w e r e a s reported by G a v e r a n d D e e b [3] for a n a l y s i s of carboplatin in d o g plasma  ultrafiltrate,  with  a mobile  phase  of acetonitrile/methanol/5  m M sodium  perchlorate, p H 2.4 (75/15/10) p u m p e d isocratically at 1.5 m L / m i n . Efficiency v a l u e s w e r e calculated in terms of both total (A/) a n d effective (A/ /r) e  n u m b e r of theoretical plates, a c c o r d i n g to the b a s e l i n e tangent (4-sigma) m e t h o d [8]: tr  A/  =  16 * W  (tr - to)  2  N  2  eff  =  16 * W  b  2  2 b  w h e r e t is t h e void time, t is the retention time of carboplatin, a n d w is the width at the 0  r  b  b a s e l i n e intercept. Alternatively, N a n d N  eff  c a n b e related by capacity factor {k):  k' Neff =  N * (  t -1 r  )  w h e r e k'  2  0  =  ^+k'  to  2.2.5. Evaluation of solid-phase extraction cartridges Several evaluated,  r e v e r s e p h a s e ( O D S ) extraction  specifically  Sep-Pak  €  C18  cartridges  (Waters),  (1 m L v o l u m e s )  Supelclean  C18  were  (Supelco  C h r o m a t o g r a p h y P r o d u c t s , S i g m a - A l d r i c h C a n a d a , Oakville, O N , C a n a d a ) , a n d B o n d Elut  C 1 8 (Varian  employed  Canada,  R i c h m o n d , B C , C a n a d a ) cartridges.  is d e s c r i b e d in M e t h o d  1 of F i g u r e 2 . 1 .  T h e procedure  Briefly, the cartridge w a s  preconditioned with 1 m L methanol followed by 3 x 1 m L water. A n aliquot (200 (iL) of a 15 fag/mL a q u e o u s carboplatin standard w a s then pulled under v a c u u m onto the sorbent b e d of the cartridge, following w h i c h the sorbent w a s rinsed with s u c c e s s i v e fractions (1 m L ) of water.  T h e a m o u n t of drug in e a c h of t h e s e fractions w a s d e t e r m i n e d from p e a k  a r e a s obtained after injection onto the H P L C s y s t e m .  30  N o r m a l p h a s e extraction cartridges evaluated w e r e t h o s e containing S u p e l c l e a n a m i n o , c y a n o , silica, a n d diol s o r b e n t s .  Slight modifications to the extraction protocol  w e r e m a d e a s s h o w n in M e t h o d 2 of Figure 2.1.  Briefly, the preconditioning solvent  c o n s i s t e d of acetonitrile/water (95/5). T h e carboplatin standard (15 jag/mL, 100 u.L) w a s diluted with 1 ml_ acetonitrile prior to addition to the cartridge, a n d the first fraction c o l l e c t e d actually c o r r e s p o n d e d to this 1 ml_ of drug solution being pulled to the top of the sorbent b e d .  S u c c e s s i v e rinses of the sorbent w e r e m a d e with 1 ml_ aliquots of  acetonitrile/water (90/10), c o r r e s p o n d i n g to collected fractions 2 through 4 .  A l l four of  the collected fractions w e r e then dried under nitrogen g a s a n d reconstituted with water prior to injection onto the H P L C s y s t e m . (Column:  YMC ODS-AQ  flow rate: 0.7 mUmin;  4.6 x 150 mm (3 jum); mobile phase:  detection:  UV 230  Method 1 Reverse Phase Cartridges _„ . . . . 200 nL standard  3% acetonitrile  in water;  nm.)  Method 2 Normal Phase Cartridges dilute 100 u.L standard tonitrile w j t h  1  m  L  ace  add to sorbent preconditioned with methanol (1 mL) and water (3 x 1 mL) (pull under vacuum to top of sorbent bed)  add to normal phase sorbent preconditioned with acetonitrile/water (95/5) (pull under vacuum to top of sorbent bed)  rinse with successive fractions of 1 mL water (pull under vacuum to top of sorbent bed)  rinse with successive fractions of 1 mL acetonitrile/water (90/10) (pull under vacuum to top of sorbent bed)  assay fractions for carboplatin  dry under nitrogen gas at 40 o c reconstitute in 200 \i.L of water  assay fractions for carboplatin Figure 2.1.  Protocol for evaluation of carboplatin retention on normal and reverse phase extraction cartridges.  31  2.2.6. Chromatographic optimization (carboplatin in plasma ultrafiltrate) After s e l e c t i o n of the most appropriate H P L C c o l u m n a n d s o l i d - p h a s e extraction cartridge, a s s a y conditions w e r e optimized for the quantitation of carboplatin in p l a s m a ultrafiltrate.  T h e c h r o m a t o g r a p h i c behaviour of blank p l a s m a ultrafiltrate a n d p l a s m a  ultrafiltrate containing carboplatin w a s evaluated under v a r i o u s mobile p h a s e conditions at flow rates b e t w e e n 0 . 5 a n d 1 . 0 m L / m i n . T h e effects of o r g a n i c modifiers (methanol, acetonitrile, a n d tetrahydrofuran) w e r e e v a l u a t e d , a s w e r e the effects of p h o s p h a t e , a c e t a t e , a n d perchlorate buffers in the p H range 3 - 7 . (Column:  YMC ODS-AQ  4.6 x 150 mm (3 jum) or 4.6 x 250 mm (5 /urn); mobile  varied; flow rate: varied (0.5-1.0 mUmin);  detection:  UV 230  phase:  nm.)  2.2.7. Determination of limits of detection and quantitation E s t i m a t e s of both the a s s a y  L O D and L O Q [ 9 , 1 0 ] were  optimized c h r o m a t o g r a p h i c a n d extractions conditions. p l a s m a ultrafiltrate to p r o d u c e final concentrations of 2 , a n d 8 |ag/mL.  m a d e under  the  C a r b o p l a t i n w a s a d d e d to  0.013, 0.025, 0.05, 0.075, 0.1, 0.5,  Following s o l i d - p h a s e extraction, duplicate injections ( 6 0 JJL) w e r e  m a d e onto the H P L C s y s t e m .  T h e L O D w a s defined simply a s the concentration  resulting in a signal-to-noise v a l u e of 3 : 1 .  T o estimate the L O Q , m e a n p e a k height  v a l u e s from the s a m p l e s providing signal-to-noise ratios greater than 1 0 : 1 w e r e u s e d to construct  a  standard  curve  using  1/y  2  weighted  linear  regression.  Predicted  concentration v a l u e s w e r e then determined from the m e a n p e a k height r e s p o n s e s o b s e r v e d for e a c h s a m p l e concentration below the lowest concentration in that standard curve.  T h e L O Q w a s the concentration at w h i c h the m e a n b i a s (difference b e t w e e n  e x p e c t e d a n d predicted concentrations) a p p r o a c h e d but did not e x c e e d 2 0 % .  32  2.3.  Results and Discussion  2.3.1. C o l u m n evaluation Table 2.2.  Retention and efficiency data for carboplatin on H P L C columns evaluated. to  tr  k'  N  N  uBondapak N H *  2.00  9.80  3.90  3360  2130  uBondapak O D S  1.96  4.92  1.51  2180  1130  Vydac 201TP O D S  2.36  5.35  1.27  4960  2200  Novapak O D S  1.81  4.39  1.43  6460  2620  YMC ODS-AQ  2.63  8.31  2.16  14400  7050  2  *  eff  determined from chromatograms presented by Gaver et al. [3]  T a b l e 2.2 lists the retention evaluated H P L C columns.  a n d efficiency d a t a obtained for e a c h of  the  T h e ability to resolve two c o m p o u n d s in H P L C is directly  related to selectivity, capacity factor, a n d efficiency a s follows:  R  s  =  1 4  (  a - 1  ) (  a  k' 2  }  *  /v°s  1 + k' 2  w h e r e selectivity (a) = k ' I ki a n d capacity factor {k) = (t - to) / to 2  r  T h u s , efficiency is a n important parameter to c o n s i d e r w h e n attempting to r e s o l v e a particular analyte from other s a m p l e c o m p o n e n t s . Efficiency is approximately constant (i.e. independent of retention values) for a particular H P L C c o l u m n , provided that the solvent composition a n d flow rate of the mobile p h a s e are not significantly altered. T h u s , the parameter N is s o m e t i m e s referred to a s c o l u m n efficiency a n d c a n be directly related to the physical properties of the H P L C c o l u m n itself.  Reduced  efficiencies are the result of analyte dilution by the mobile p h a s e (band spreading). T h r e e major p r o c e s s e s h a v e b e e n d e s c r i b e d : e d d y diffusion, m o l e c u l a r diffusion, a n d m a s s transfer.  H P L C conditions which minimize t h e s e contributions to b a n d s p r e a d i n g  will result in i n c r e a s e d efficiencies.  Efficient H P L C c o l u m n s typically yield  10,000  theoretical plates or more.  33  F o r the normal p h a s e separation of carboplatin on the ^ B o n d a p a k a m i n o c o l u m n , the c a l c u l a t e d efficiency w a s very low.  T h i s poor efficiency is easily e x p l a i n e d by the  p h y s i c a l properties of the c o l u m n itself.  Its large and irregularly s h a p e d particles  contribute to a significant a m o u n t of e d d y diffusion, while the large pore s i z e of the particles results in i n c r e a s e d m a s s transfer effects d u e to diffusion of the analyte within stagnant mobile p h a s e contained within the pores. In c o m p a r i s o n to normal p h a s e separations, reverse p h a s e s e p a r a t i o n s are usually m o r e efficient d u e to the very nature of the interactive f o r c e s involved.  Adsorptive  t e c h n i q u e s s u c h a s normal p h a s e i n c r e a s e stationary p h a s e m a s s transfer effects b e c a u s e the i n c r e a s e d e n e r g y of the hydrogen b o n d s results in i n c r e a s e d r e s i d e n c e times at the site of adsorption.  B y c o m p a r i s o n , reverse p h a s e c h r o m a t o g r a p h y simply  involves partitioning of m o l e c u l e s b e t w e e n the mobile a n d stationary p h a s e s , a p r o c e s s w h i c h is inherently m o r e rapid a n d efficient. T h i s is d e m o n s t r a t e d by direct c o m p a r i s o n of c o l u m n efficiency v a l u e s for the j i B o n d a p a k amino a n d O D S c o l u m n s .  Accounting  for the differences in length of t h e s e c o l u m n s , the c o l u m n efficiency w a s 3 0 % greater for the O D S v e r s u s the a m i n o c o l u m n .  Unfortunately, this i n c r e a s e is of little practical  i m p o r t a n c e s i n c e the efficiency v a l u e s t h e m s e l v e s are s o poor. When increased  moving from the column  ^ B o n d a p a k to V y d a c to N o v a p a k O D S c o l u m n s , the  efficiencies  observed  are  primarily  a  result  characteristics of the p a c k i n g materials u s e d in t h e s e c o l u m n s .  of  the  improved  T h e m a i n difference  w a s in the d i a m e t e r a n d s h a p e of the particles, the s m a l l e r s p h e r i c a l l y - s h a p e d particles resulting in m o r e efficient p a c k i n g a n d thus l e s s e d d y diffusion. A m a r k e d i n c r e a s e in efficiency w a s o b s e r v e d for the Y M C O D S - A Q c o l u m n a s c o m p a r e d to the N o v a p a k O D S c o l u m n .  T h e s i z e of this i n c r e a s e is s o m e w h a t  surprising, a s i m p r o v e m e n t s in efficiency d u e to the s m a l l e r particle s i z e (3 v e r s u s 4 (j.m) s h o u l d be approximately offset by the larger pore s i z e (120 A v e r s u s 6 0 A) of this c o l u m n . T h e difference is likely attributable to two major factors. First, the Y M C c o l u m n h a s a proprietary hydrophilic e n d c a p p i n g w h i c h e n h a n c e s interaction in highly polar mobile p h a s e s by k e e p i n g the O D S c h a i n s erect.  With c o n v e n t i o n a l trimethylsilane  e n d c a p p i n g , the h y d r o p h o b i c s i d e - c h a i n s m a y " s a g " a n d overlap d u e to repulsion from  34  the mobile p h a s e .  S e c o n d , the Y M C c o l u m n h a s a greater c a r b o n load ( 1 4 % v e r s u s  7%), w h i c h is important d u e to the apparent reluctance of carboplatin to l e a v e the polar e n v i r o n m e n t of the mobile p h a s e .  Indeed, for efficiency tests of a more h y d r o p h o b i c  m o l e c u l e (toluene), efficiency v a l u e s o n the Y M C c o l u m n (N = 14000) w e r e only slightly greater than t h o s e o n the N o v a p a k c o l u m n (N = 12000), consistent with the previously d i s c u s s e d differences (particle s i z e , pore s i z e , a n d c o l u m n diameter) b e t w e e n t h e s e columns. W h i l e t h e c o l u m n efficiency or total n u m b e r of theoretical plates (A/) is a c o n v e n i e n t m e a n s of c o m p a r i n g the physical properties of H P L C c o l u m n s , t h e effective n u m b e r of theoretical plates (N f), or separation efficiency, is a more useful parameter ef  when  dealing  with  a particular  s e t of a s s a y  conditions.  Separation  efficiency  incorporates both the c o l u m n efficiency a n d the influence of capacity factor/retention o n the s e p a r a t i o n .  A s s u m i n g that selectivity is approximately constant for a particular  stationary p h a s e a n d mobile p h a s e constituent  mixture,  resolution a n d s e p a r a t i o n  efficiency are related a s follows: k' R  oc (  s  )  * A/  or  0 5  R  s  oc  N  a5 eff  1 +k' T h e a b o v e equation gives rise to the idea of k' p r o g r a m m i n g . F i g u r e 2.2 s h o w s a plot relating W a n d relative R (or A/ ff) v a l u e s . T h e plot c a n b e roughly divided into three s  p s e u d o - l i n e a r regions. retention.  e  B e l o w k' = 3, R  s  v a l u e s improve markedly with increasing  F o r k' v a l u e s of 3-10, more m o d e s t i n c r e a s e s are o b s e r v e d . A b o v e k' = 1 0 ,  the R is approximately constant a s N tf a p p r o a c h e s A/. s  e  With  r e s p e c t to separation of carboplatin  o n the r e v e r s e p h a s e  columns  e x a m i n e d , t h e actual separation efficiency a c h i e v e d w a s in all c a s e s m u c h lower than the inherent c o l u m n efficiencies d u e to poor retention of t h e drug.  S i n c e t h e mobile  p h a s e s e m p l o y e d w e r e the most polar p o s s i b l e , no further i m p r o v e m e n t s in efficiency c a n b e a c h i e v e d through k' p r o g r a m m i n g . B e l o w k' v a l u e s of 3, any e n h a n c e m e n t in k' brings about large i n c r e a s e s in A W T h u s , the greater retentivity (k' = 2 . 1 6 ) of the Y M C ODS-AQ  c o l u m n v e r s u s the otrW c o l u m n s is of particular importance.  From a  35  theoretical standpoint, improved separation efficiencies for carboplatin could readily be a c h i e v e d by producing a more retentive c o l u m n .  With reference to the p h y s i c a l  properties listed in T a b l e 2.1, the most influential parameter on drug retention is that of c a r b o n load, w h i c h is itself a c o m p l e x function of the particle s i z e , pore s i z e , a n d density of O D S c o v e r a g e of the p a c k i n g material. A m o r e c o m p r e h e n s i v e review of b a n d s p r e a d i n g a n d the r e l e v a n c e of efficiency, c a p a c i t y factor, a n d other variables to resolution in c h r o m a t o g r a p h i c s e p a r a t i o n s c a n be found in Hamilton a n d S e w e l l [11].  3  6  9  12  15  18  21  k'  Figure 2.2.  Effect of k' programming on relative efficiency and resolution values.  The data  points are theoretical and assume that selectivity (a) is constant for a given H P L C column and are thus not affected by the change in k' values.  Below k' = 3,  resolution and efficiency values decrease dramatically. Above k' = 1 0 , retention times are inconveniently long. Adapted from data in Hamilton and Sewell [11].  36  2.3.2. Preliminary separation from endogenous plasma components G i v e n the m o d e s t improvement in retention times o b s e r v e d for carboplatin o n e v e n the Y M C stationary p h a s e , w e b e g a n by evaluating a longer (4.6 x 2 5 0 mm) Y M C O D S - A Q c o l u m n in order to i n c r e a s e the retention of carboplatin, thereby i n c r e a s i n g the ability of the c o l u m n to s e p a r a t e carboplatin from e n d o g e n o u s p l a s m a ultrafiltrate components.  U s i n g a mobile p h a s e of 1 0 0 % water, c h r o m a t o g r a m s obtained after  injection of ultrafiltrate s a m p l e s s h o w e d a n u m b e r of c o m p o u n d s w h i c h interfered with the detection of the analyte.  Slight improvements w e r e o b s e r v e d for mobile p h a s e s  containing s m a l l a m o u n t s of methanol or acetonitrile; however, the effectiveness of this a p p r o a c h w a s limited b e c a u s e the retention times of carboplatin quickly a p p r o a c h e d t h o s e of the solvent front w h e n e v e n small p e r c e n t a g e s of o r g a n i c modifier w e r e a d d e d . T h e influence of p H on carboplatin chromatography w a s initially e x a m i n e d v i a addition of s o d i u m p h o s p h a t e buffer, p H 3-7, to the mobile p h a s e . T h e retention time of carboplatin itself, w h i c h is u n c h a r g e d in a q u e o u s solution, w a s not affected by c h a n g e s in m o b i l e p h a s e p H .  H o w e v e r , the retention times of m a n y e n d o g e n o u s p l a s m a  ultrafiltrate c o m p o n e n t s w e r e sensitive to p H c h a n g e s . T h e best c h r o m a t o g r a p h y w a s o b s e r v e d for  a mobile  p h a s e consisting of  p h o s p h a t e (20 mM; p H 4.5).  1%  methanol  in m o n o b a s i c  sodium  Unfortunately, carboplatin w a s not completely r e s o l v e d  from e n d o g e n o u s p l a s m a interferences (Figure 2.3).  T h e u s e of alternative  buffer  s y s t e m s containing acetate or perchlorate ions did not result in significant i m p r o v e m e n t s to the s e p a r a t i o n , although slight differences w e r e o b s e r v e d in the elution patterns of e n d o g e n o u s interferences.  Unfortunately, the large n u m b e r of t h e s e interferences  p r e s e n t in the c h r o m a t o g r a m s m a d e their separation from carboplatin unlikely under a n y mobile p h a s e conditions applied. T h u s , w e c o n c l u d e d that r e m o v a l of interferences by  s o l i d - p h a s e extraction  was  a  n e c e s s a r y prerequisite  to  obtaining  a  usable  chromatographic separation.  37  0  Figure 2.3.  10  20  Time (min)  Chromatograms (UV 230 nm) of (A) blank plasma ultrafiltrate, and (B) plasma ultrafiltrate spiked with 5 ng/mL of carboplatin. Column: Y M C O D S - A Q 4.6 x 250 mm; mobile phase: 1% methanol in 5 mM monobasic sodium phosphate; flow rate: 0.6 mL/min.  38  2.3.3. Evaluation of solid-phase extraction cartridges F o r the four O D S - t y p e extraction cartridges e v a l u a t e d , all drug eluted within the first rinse fraction. changes were  S i n c e the eluent u s e d w a s the w e a k e s t possible ( 1 0 0 % water), no  p o s s i b l e in order to  improve the  retentivity of the  cartridges  for  carboplatin. F o r the normal p h a s e cartridges, Figure 2.4 s h o w s the elution profiles obtained using M e t h o d 2 from Figure 2 . 1 .  All four sorbent types e v a l u a t e d (amino,  c y a n o , silica, a n d diol) s h o w e d s o m e retention of carboplatin; however, retention o n the a m i n o cartridge w a s by far the greatest. F o r the other cartridges, drug w a s o b s e r v e d in the very first fraction, which c o r r e s p o n d e d to the addition of drug onto the cartridge. E v e n for the a m i n o cartridge the binding of drug w a s s o m e w h a t less than d e s i r e d , s i n c e a conventional extraction p r o c e d u r e s u s e s at least o n e rinse step following addition of the drug onto the cartridge.  1 oo - i  FRACTION  Figure 2.4.  Elution profiles for carboplatin on normal phase cartridges. Fraction 1 corresponds to drug addition to the pre-conditioned cartridge, fractions 2-4 to successive rinses with 1 mL of acetonitrile/water (90/10).  Of the cartridges examined, the amino  cartridge is the most retentive for carboplatin.  39  2.3.4. Chromatographic optimization T o e n s u r e a d e q u a t e recovery of carboplatin w e converted to 3 m L S u p e l c l e a n a m i n o cartridges.  W e followed the identical procedure to that s h o w n in M e t h o d 2 of  Figure 2 . 1 , e x c e p t that 2 0 0 u.L of p l a s m a ultrafiltrate w a s diluted to 2 m L with acetonitrile prior to addition to the larger cartridge. ultrafiltrate  samples were  much  C h r o m a t o g r a m s obtained for the extracted  c l e a n e r than  those  obtained  previously  without  extraction, although a n u m b e r of e n d o g e n o u s c o m p o u n d s w e r e still o b s e r v e d .  Re-  evaluation of acetonitrile a n d methanol a s organic modifiers d e m o n s t r a t e d that the u s e of acetonitrile o v e r methanol resulted in improved (sharper) p e a k s h a p e s a n d a better separation.  U s i n g a mobile p h a s e of 1.3% acetonitrile in 2 0 m M m o n o b a s i c s o d i u m  p h o s p h a t e , w e obtained resolution of carboplatin from ultrafiltrate c o m p o n e n t s , e v e n on the 4.6 x 150 m m Y M C O D S - A Q analytical c o l u m n (Figure 2.5).  H o w e v e r , small  c h a n g e s (as little a s 0.1%) in the proportion of organic modifier present in the mobile p h a s e resulted in noticeable c h r o m a t o g r a p h i c differences.  T o e n s u r e reproducible  c h r o m a t o g r a p h y , mobile p h a s e s w e r e prepared by pipetting the correct v o l u m e of acetonitrile into a volumetric flask with s u b s e q u e n t dilution to the appropriate v o l u m e , rather  than  by  separate  volume  measurements  of  the  acetonitrile  and  buffer  components.  40  Q.  o  -Q i  re O  B  7V__  10  Figure 2.5.  ~_  Time (min)  Chromatograms (UV 230 nm) of solid-phase extracted (A) blank plasma ultrafiltrate and (B) plasma ultrafiltrate containing 8 ^g/mL carboplatin, which eluted at 6.5 min. Column: Y M C O D S - A Q 4.6 x 150 mm (3 mm); mobile phase: 1.3% acetonitrile in 20 mM monobasic sodium phosphate; flow rate: 0.7 mL/min.  41  2.3.5. Determination of limits of detection and quantitation A l l o w i n g for 2 0 % C V in a c c u r a c y a n d precision, L O Q v a l u e s typically c o r r e s p o n d to c o n c e n t r a t i o n s with signal-to-noise ratios of 10:1 or s m a l l e r [9,10].  U n d e r the  o p t i m i z e d H P L C - U V conditions, m e a n p e a k height r e s p o n s e s w e r e 113, 2 5 1 , 3 7 7 , 6 1 3 , 3101,  1 1 7 3 1 , a n d 5 0 6 2 8 u V at carboplatin concentrations of 0.025, 0.05, 0.075, 0.1,  0.5, 2, a n d 8 u.g/mL. N o r e s p o n s e w a s o b s e r v e d from the 0.013 u,g/mL s a m p l e s . S i n c e 0.1  u.g/mL  injections  g a v e signal-to-noise v a l u e s greater than  1 0 : 1 , four  sample  c o n c e n t r a t i o n s from 0.1 to 8 p.g/mL w e r e u s e d to construct a s t a n d a r d c u r v e (Figure 2.6) a s d e s c r i b e d in the experimental (section 2.2.7).  T h e weighted (1/y ) 2  p r o c e d u r e w a s u s e d in order to m a x i m i z e the quantitation limit attained [12].  regression T h e 0.05  u.g/mL s a m p l e , which resulted in a m e a n bias v a l u e of 2 0 . 0 % , c o r r e s p o n d e d to the best e s t i m a t e of a s s a y L O Q . F o r p l a s m a ultrafiltrate s a m p l e s , injection of the 0.025 u.g/mL s a m p l e s provided s i g n a l - t o - n o i s e v a l u e s slightly greater than 3:1. T h i s concentration w a s then verified to be the L O D by further injection of 0.02 u.g/mL s a m p l e s , which did not provide detectable responses.  60 50 •  o -  40 -  r 0.9999 slope 6.24 ntercept -0.013 2  *  ~  30 -  f 20 0! a> Q_ 10 0 0  2  4  6  8  10  Concentration (ng/mL)  Figure 2.6.  Standard curve (0.1-8 u.g/mL) used to estimate the L O Q of the H P L C - U V assay. Predicted concentrations for sample concentrations below 0.1 u.g/mL were then determined using the best-fit equation from weighted (1/y ) regression. 2  42  2.4.  Conclusions T h e polar structure of carboplatin resulted in very poor retention characteristics  on m a n y c o n v e n t i o n a l reverse p h a s e ( O D S ) H P L C c o l u m n s .  Better retention w a s  o b s e r v e d on a Y M C O D S - A Q c o l u m n , which c a n be attributed to the s m a l l e r particle s i z e , i n c r e a s e d c a r b o n loading, a n d unique hydrophilic e n d c a p p i n g of this c o l u m n a s c o m p a r e d to the other c o l u m n s e v a l u a t e d .  Still, s e p a r a t i o n of carboplatin  from  e n d o g e n o u s c o m p o n e n t s of p l a s m a ultrafiltrate required s o l i d - p h a s e extraction prior to H P L C analysis.  Little or no retention w a s o b s e r v e d w h e n carboplatin w a s applied to  r e v e r s e p h a s e extraction cartridges; thus, a p r o c e d u r e utilizing normal p h a s e (amino) extraction cartridges w a s d e v e l o p e d . This combination of normal p h a s e s a m p l e c l e a n up with r e v e r s e p h a s e H P L C a n a l y s i s resulted in improved sensitivity c o m p a r e d to previously reported a s s a y m e t h o d s b a s e d on normal p h a s e H P L C , the L O D a n d L O Q e s t i m a t e s for this n e w method being 25 a n d 50 n g / m L , respectively. c l o s e to our initial g o a l of 20 ng/mL.  T h e s e limits are  T h e potential of p o s t - c o l u m n derivatization to  improve upon the o b s e r v e d sensitivity is d i s c u s s e d in C h a p t e r 3.  With this H P L C - U V  m e t h o d , full optimization still requires identification of a n appropriate internal standard m o l e c u l e in order to control imprecision introduced by variability  in the  extraction  p r o c e d u r e ( C h a p t e r 4).  43  2.5.  References  1.  S.J. Harland, D.R. Newell, Z.H. Siddik, R. Chadwick, A.H. Calvert, and K.R. Harrap. Pharmacokinetics of c/s-diammine-1,1-cyclobutanedicarboxylate platinum II in patients with normal and impaired renal function. Cancer Res 44: 1693-1697 (1984).  2.  R.C. Gaver and G. Deeb.  H P L C procedures for the analysis of carboplatin in human  plasma ultrafiltrate. Cancer Chemother Pharmacol 16: 201-206 (1986). 3.  R.C. Gaver, A.M. George, G.F. Duncan, A.D. Morris, G. Deeb, H.C. Faulkner, and R.H. Farmen. The disposition of carboplatin in the beagle dog. Cancer Chemother  Pharmacol  21:197-202 (1988). 4.  M.A.  Allsopp,  G.J.  Sewell,  and  C.G.  Rowland.  A  column-switching  chromatographic assay for the analysis of carboplatin in plasma ultrafiltrate.  liquid  J Pharm  Biomed Anal 10: 375-381 (1992). 5.  F. Elferink, W.J. Van  der Vijgh,  and  H.M.  Pinedo.  On-line differential  polarographic detection of carboplatin in biological samples after  pulse  chromatographic  separation. Anal Chem 58: 2293-2296 (1986). 6.  R.B. Burns, R.W. Burton, S.A. Albon, and L. Embree.  Liquid chromatography-mass  spectrometry for the detection of platinum antineoplastic complexes. J Pharm  Biomed  Anal 14: 367-372(1996). 7.  R.B. Burns and L. Embree. Analysis of antineoplastic platinum complexes by highperformance liquid chromatography with inductively coupled plasma and electrospray mass spectrometric detection. Pharm Research 12: S 7 0 (1995).  8.  C. Horvath and W.R. Melander. Fundamental  and Applications  Theory of chromatography.  of Chromatographic  In  and Electrophoretic  Chromatography: Methods.  E.  Heftmann (Editor), Elsevier Scientific, New York, A28-A130 (1983). 9.  I. Krull and Swartz M. Determining limits of detection and quantitation. LC-GC.  16: 922-  923 (1998). 10.  H.T. Karnes, G. Shiu, and V.P. Shah.  Validation of bioanalytical methods.  Pharm  Research 8 (1991). 11.  R.J. Hamilton and P.A. Sewell.  Introduction to Modern  Liquid  Chromatography.  Chapman and Hall, London, 17-25 (1982). 12.  G.K. Szabo, H.K. Browne, A. Ajami, and E.G. Josephs. Alternatives to least squares linear regression analysis for computation of standard curves for quantitation by H P L C : applications to clinical pharmacology. J Clin Pharmacol 34: 242-249 (1994).  44  CHAPTER 3 DEVELOPMENT AN HPLC-PC  3.1.  AND PRELIMINARY ASSAY  METHOD  FOR  EVALUATION  OF  CARBOPLATIN  Introduction A s d i s c u s s e d in C h a p t e r 1, pre-column derivatization t e c h n i q u e s are not m e t h o d s  of c h o i c e for carboplatin, a s multiple platinum nucleophilic substitution products m a y give rise to the s a m e derivatized product, h e n c e w o u l d not be distinguishable upon injection onto the H P L C c o l u m n . B y performing the derivatization reaction p o s t - c o l u m n (following the c h r o m a t o g r a p h i c separation of carboplatin from its degradation products), specificity of the method for the parent c o m p o u n d is e n s u r e d . F r e i et al. [1] h a v e reviewed the utility of p o s t - c o l u m n reaction detection in H P L C methods.  A g e n e r a l s c h e m a t i c of a n H P L C - P C s y s t e m is s h o w n in F i g u r e 3.1.  The  e q u i p m e n t is similar to that e m p l o y e d for c o n v e n t i o n a l H P L C s y s t e m s ; h o w e v e r , a n additional p u m p is u s e d to a d d a c h e m i c a l reagent to the c h r o m a t o g r a p h i c eluent prior to its r e a c h i n g the detector (most c o m m o n l y U V , f l u o r e s c e n c e , or electrochemical). T h e a d v a n t a g e of this a p p r o a c h is the potential to p r o d u c e a c h e m i c a l derivative with e n h a n c e d a n d / o r more selective detection characteristics a s c o m p a r e d to the original analyte.  P r o v i d e d the reaction is reproducible, it n e e d not g o to completion or e v e n  p r o d u c e a single, stable product.  D i s a d v a n t a g e s of on-line p o s t - c o l u m n  reaction  include the requirements for additional equipment, problems^ resulting from i n a d e q u a t e reagent mixing, a n d l o s s of c h r o m a t o g r a p h i c resolution d u e to b a n d - b r o a d e n i n g within the reactor. detection  of  P o s t - c o l u m n derivatization reactions h a v e already b e e n e m p l o y e d for the many  classes  of  compounds,  including  amino  acids,  barbiturates,  c a t e c h o l a m i n e s , a n d carbohydrates. T w o m a i n types of post-column reactors h a v e b e e n d e v e l o p e d , o p e n tubular reactors a n d p a c k e d b e d reactors [2]. O p e n tubular reactors are most c o m m o n a n d are the s i m p l e s t to construct, consisting of a length of steel o r tubing, typically 0.3-0.8 m m in diameter.  polytetrafluoroethylene  T h e y are usually e m p l o y e d for relatively fast  reactions (~ 30 s), s i n c e b a n d broadening within the reactor will lead to a l o s s of  45  c h r o m a t o g r a p h i c resolution.  T h i s b a n d broadening c a n be significantly r e d u c e d by  coiling or knitting of the tubing.  V a r i o u s geometries for knitting reactor tubing are  p o s s i b l e , e a c h with different a d v a n t a g e s in terms of simplicity of construction, mixing efficiency, a n d reduction in overall reactor length.  Additionally, knitting of tubular  reactors provides better radial mixing of the eluent, which is significant b e c a u s e a c o m m o n problem with tubular reactors is a n unstable c h r o m a t o g r a p h i c b a s e l i n e d u e to i n c o m p l e t e mixing of the mobile p h a s e a n d post-column reagent.  In c a s e s w h e r e  knitting of the reactor d o e s not o v e r c o m e band broadening or poor mixing, the u s e of p a c k e d b e d reactors in p l a c e of tubular reactors is a rational alternative. The  reaction of s o d i u m bisulfite with platinum c o m p o u n d s w a s first noted by  H u s s a i n et al. [3], w h o reported that addition of bisulfite to solutions of cisplatin resulted in e n h a n c e d a b s o r b a n c e around 2 9 0 n m .  T h e potential u s e of this reaction in post-  c o l u m n s y s t e m s w a s then e x p l o r e d by M a r s h et al. [4], w h o investigated the effect of v a r i o u s conditions on the rate a n d extent of the bisulfite-cisplatin reaction. T h e authors o b s e r v e d that prior addition of p e r m a n g a n a t e a n d dichromate to the reaction mixture resulted in a more rapid reaction rate. employing  sequential dichromate  and  Validated H P L C - P C bisulfite  addition  a s s a y s for  were  later  cisplatin  developed  by  bisulfite  for  K i n o s h i t a et al. [5] a n d Farrish et al. [6]. Development  of  post-column  techniques  utilizing  sodium  c y c l o b u t a n e d i c a r b o x y l a t o a n d malonato c o m p o u n d s is c o m p l i c a t e d by the reactivity of t h e s e c o m p o u n d s , w h i c h is less than that of cisplatin. simplified  HPLC-PC  system  p o t a s s i u m dichromate)  utilizing  sodium  bisulfite  K i z u et al. [7] d e v e l o p e d a (without  a s the only post-column reagent.  prior  reaction  with  T h e reaction rate w a s  e n h a n c e d by increasing the bisulfite concentration approximately 10-fold o v e r previous m e t h o d s applied to cisplatin, a s well a s by elevation of the p o s t - c o l u m n  reaction  temperature to 6 0 °C. U n d e r a unified set of reaction conditions, the authors a c h i e v e d detection  limits of nearly 2 0 n g / m L for a q u e o u s solutions containing  oxaliplatin, or tetraplatin.  T h e method w a s then applied to the  evaluation of oxaliplatin in rabbit  p l a s m a a n d urine.  investigated the utility of the method for determination  carboplatin,  pharmacokinetic  H o w e v e r , the authors  only  of carboplatin in a q u e o u s  46  standards.  Direct c o m p a r i s o n s of published c h r o m a t o g r a m s of a q u e o u s carboplatin  injections a n d injections of the rabbit p l a s m a ultrafiltrate a n d urine d e m o n s t r a t e that the Inertsil O D S - 2 c o l u m n (4.6 x 2 5 0 mm) e m p l o y e d for oxaliplatin a n a l y s i s did not provide sufficient  retention  (k'  ~ 0.6)  or efficiency (N f ef  < 2000) to provide resolution  of  carboplatin from biological interferences. T h i s c h a p t e r d e s c r i b e s the d e v e l o p m e n t of a post-column reaction detection s y s t e m for a n a l y s i s of carboplatin in p l a s m a ultrafiltrate. employs a  knitted tubular  detection at 2 9 0 n m .  This H P L C - P C  reactor for analyte derivatization  method  with s u b s e q u e n t  UV  C o n d i t i o n s required to eliminate the p r o b l e m of a fluctuating  b a s e l i n e are d e s c r i b e d , including the effects of controlling p H a n d mobile p h a s e a n d p o s t - c o l u m n reagent flow rates, a s well a s the effectiveness of p u l s e d a m p e n i n g d e v i c e s a d d e d to the s y s t e m .  Optimization of both c h r o m a t o g r a p h y (mobile p h a s e  composition) a n d s i g n a l r e s p o n s e (post-column reagent composition) are d i s c u s s e d . With r e s p e c t to the post-column reagent, the influence of c h a n g e s in buffer type, p H , acetonitrile concentration, and s o d i u m bisulfite concentration are all e x p l o r e d .  Finally,  e s t i m a t e s of both the limits of detection a n d quantitation of the optimized m e t h o d are m a d e a n d potential a d v a n t a g e s a n d d i s a d v a n t a g e s of the H P L C - P C v e r s u s H P L C - U V t e c h n i q u e s are s u m m a r i z e d .  t-connector  Pump 1 mobile p h a s e  ^  ^  H P L C Column  Pump 2 P C reagent  1 P T F E tubing  U V Detector  Figure 3.1.  General schematic of the post-column reaction system, employing a knitted tubular reactor, used for the analysis of carboplatin in plasma ultrafiltrate.  47  3.2.  Experimental  3.2.1. Buffers, mobile phases, and plasma ultrafiltrate S e e s e c t i o n 2.2.1 regarding preparation of buffers, mobile p h a s e s a n d p l a s m a ultrafiltrate. Buffers u s e d for the H P L C - P C method w e r e adjusted to the d e s i r e d p H via d r o p w i s e addition of a 1 M s o d i u m hydroxide solution to the a c i d i c form of that particular buffer (as o p p o s e d to mixing equimolar concentrations of a c i d / b a s e forms of the buffer). 3.2.2. Apparatus Chromatography  was  d e s c r i b e d in section 2.2.3.  performed  on  the  Waters  HPLC  system  previously  A s e c o n d m o d e l 510 pump w a s u s e d to deliver the post-  c o l u m n reagent. T h e analytical wavelength w a s 2 9 0 nm. C h r o m a t o g r a p h i c s e p a r a t i o n s w e r e performed using the Y M C O D S - A Q 4.6 x 150 m m (3 nm) c o l u m n with a 4 x 2 3 m m g u a r d c o l u m n containing the s a m e 3 ^im p a c k i n g material.  T h e p o s t - c o l u m n reactor  w a s p r e p a r e d from a narrow diameter (0.5 m m i.d. x 13.2 m)  polytetrafluoroethylene  coil, knitted to provide a torturous pathway for better reagent mixing. uniformity,  model  L P - 2 1 pulse d a m p e n i n g  units  T o improve flow  ( M a n d e l Scientific, G u e l p h , O N ,  C a n a d a ) w e r e p l a c e d b e t w e e n the post-column reagent p u m p a n d the p o s t - c o l u m n reactor a n d b e t w e e n the mobile p h a s e p u m p a n d the autoinjector. 3.2.3. Stabilizing the chromatographic baseline During our preliminary evaluation of the post-column s y s t e m , w e e x a m i n e d the effect of the following parameters with respect to stability of the  chromatographic  b a s e l i n e : the addition of pump d a m p e n i n g d e v i c e s , the mobile p h a s e a n d p o s t - c o l u m n reagent flow rates, a n d the p H of the eluent entering the detector. 3.2.4. Optimization studies Chromatography In order to optimize the c h r o m a t o g r a p h i c conditions for the H P L C - P C  assay  m e t h o d , w e maintained a fixed p o s t - c o l u m n reagent c o m p o s i t i o n a n d then e v a l u a t e d the effect of mobile p h a s e c h a n g e s on the o b s e r v e d c h r o m a t o g r a p h y . separation  of  carboplatin  from  endogenous  plasma  ultrafiltrate  T o achieve  components,  we  48  primarily e x a m i n e d the effect of mobile p h a s e p H a n d addition of acetonitrile on the separation.  M e t h a n o l w a s not evaluated a s a n organic modifier s i n c e previous studies  by M a r s h et al. [4] a n d K i z u et al. [7] had d e m o n s t r a t e d that it h a d a m a r k e d inhibitory effect o n the p o s t - c o l u m n reaction, e v e n at concentrations a s low a s 0 . 1 % . (Mobile  phase:  phosphate  varied  composition;  post-column  (final pH 5.5); flow rates: 0.7 mUmin  reagent:  40 mM bisulfite  for both mobile phase  and  in 20  mM  post-column  reagent.)  Post-column  reagent  T h e c o m p o s i t i o n of the mobile p h a s e w a s fixed a n d c o n s i s t e d of the optimized chromatographic system.  Optimization studies w e r e c o n d u c t e d on-line with c h a n g e s  m a d e to the p o s t - c o l u m n reagent only, the exception being to e n s u r e that identical buffer salts (phosphate, acetate, or citrate) w e r e u s e d in both mobile p h a s e a n d postcolumn reagents.  Initially, w e e m p l o y e d the s a m e p o s t - c o l u m n reagent (20  mM  p h o s p h a t e a n d 4 0 m M bisulfite, p H 5.5) that w e u s e d to e v a l u a t e the effect of mobile p h a s e c h a n g e s on chromatography. C o m p o s i t i o n of the p o s t - c o l u m n reagent w a s then altered  to  examine  the  effect  of  the  following  variables  on  detector  response  ( a b s o r b a n c e at 2 9 0 nm): buffer type, buffer p H , acetonitrile concentration, a n d bisulfite concentration.  Following equilibration of the mobile p h a s e a n d p o s t - c o l u m n reagents,  m e a n p e a k heights w e r e recorded after triplicate injections (30 u.L) of a 5 u,g/mL a q u e o u s carboplatin s t a n d a r d . (Mobile  phase:  20 mM  phosphate,  reagent:  varied; flow rates: 0.7 mUmin  acetate,  or citrate  buffer, pH  4.5;  post-column  for both mobile phase and post-column  reagent.)  3.2.5. Determination of limits of detection and quantitation T h e limits of detection a n d quantitation w e r e determined a s d e s c r i b e d in section 2.2.7 for the H P L C - U V a s s a y m e t h o d , with p l a s m a ultrafiltrate s a m p l e s injected directly (without extraction) into the optimized H P L C - P C s y s t e m . (Mobile  phase:  20 mM monobasic  bisulfite in 20 mM phosphate, and post-column  sodium  phosphate;  post-column  final pH 5.4; flow rates: 0.7 mUmin  reagent:  for both mobile  40  mM phase  reagent.)  49  3.3.  Results and Discussion  3.3.1. Stabilizing the chromatographic baseline A s reported previously [3-7], the reaction of platinum c o m p o u n d s with s o d i u m bisulfite p r o d u c e s c h r o m o p h o r i c products with m a x i m a l a b s o r b a n c e a r o u n d 2 9 0 n m . Unfortunately, s o d i u m bisulfite itself exhibits a significant amount of a b s o r b a n c e at this w a v e l e n g t h , w h i c h results in a high b a c k g r o u n d . T h i s c a n produce large fluctuations in the c h r o m a t o g r a p h i c b a s e l i n e if a s t e a d y flow into the U V detector or uniform mixing of p o s t - c o l u m n reagent a n d mobile p h a s e are not a c h i e v e d .  U n d e r the initial conditions  e v a l u a t e d , injection of a n a q u e o u s carboplatin solution resulted in the formation of a reasonable peak response.  H o w e v e r , very large fluctuations in t h e ^ c h r o m a t o g r a p h i c  b a s e l i n e w e r e a l s o o b s e r v e d (Figure 3.2).  Figure 3.2.  Initial H P L C - P C chromatogram (UV 290 nm) of a 5 ng/mL aqueous carboplatin standard showing large fluctuations in the chromatographic baseline.  50  T o study the effect of p u l s e d a m p e n i n g d e v i c e s on the s y s t e m , w e p u m p e d either the mobile p h a s e or p o s t - c o l u m n reagent directly into the U V detector (Figure 3.3). W h e n mobile p h a s e w a s p u m p e d with no pulse d a m p e n i n g d e v i c e s in p l a c e , a periodic negative s p i k e in the b a s e l i n e w a s o b s e r v e d .  T h i s s p i k e w a s r e m o v e d a n d a stable  b a s e l i n e obtained w h e n the p u l s e d a m p e n i n g d e v i c e or the H P L C c o l u m n w a s p l a c e d b e t w e e n the p u m p a n d the detector.  H o w e v e r , w h e n the p o s t - c o l u m n reagent w a s  p u m p e d into the detector, a w a v e - l i k e fluctuation w a s o b s e r v e d . T h i s fluctuation w a s not affected by the p r e s e n c e or a b s e n c e of p u l s e d a m p e n i n g d e v i c e s . It s e e m s likely, therefore, that the o b s e r v e d b a s e l i n e fluctuations in our initial c h r o m a t o g r a m w e r e d u e primarily to incomplete mixing of the mobile p h a s e a n d p o s t - c o l u m n r e a g e n t s a n d not a result of non-uniform flow into the detector.  F o r s u b s e q u e n t e x p e r i m e n t s , a pulse  d a m p e n i n g d e v i c e w a s p l a c e d b e t w e e n the post-column reagent a n d reaction coil but not b e t w e e n the mobile p h a s e a n d H P L C c o l u m n , s i n c e the c o l u m n itself w a s s h o w n to act a s a p u l s e d a m p e n i n g d e v i c e . A  B  c  Figure 3.3.  Chromatographic baselines observed for mobile phases pumped into the U V detector with monitoring at 290 nm.  (A) mobile phase of 20 m M monobasic  sodium phosphate, no pulse dampening devices present (B) mobile phase of 20 mM monobasic sodium phosphate, pulse dampening unit or H P L C column present (C) mobile phase of 40 mM sodium bisulfite in 20 mM monobasic sodium phosphate (with or without pulse dampening).  Next, w e e x a m i n e d the influence of p H on the magnitude of b a s e l i n e fluctuations observed.  U V s c a n s of s o d i u m bisulfite in p h o s p h a t e buffer h a d s h o w n that the  b a c k g r o u n d a b s o r b a n c e at 2 9 0 n m w a s lower w h e n the p H of the buffer w a s i n c r e a s e d (Figure 3.4).  H o w e v e r , the effect of p H c h a n g e s on the r e s p o n s e s o b s e r v e d a n d thus  51  o n the signal-to-noise v a l u e s w a s not known.  F o r c u v e t t e - b a s e d studies, conflicting  results a s to the optimal p H of the cisplatin/bisulfite had b e e n o b t a i n e d , with M a r s h et al. [4] reporting a n optimal p H of 4.5 a n d K i z u et al. [7] reporting a n optimal p H of 5.5-6.0. O u r on-line p o s t - c o l u m n studies demonstrate that adjustment of the eluent p H to 6.2 instead of 4 . 5 results in improvements to both the c h r o m a t o g r a p h i c b a s e l i n e (Figure 3.5) a n d p e a k r e s p o n s e s o b s e r v e d (not shown).  A b s (290 nm) O C CO -O I  o (/)  <  240  275  310  Wavelength (nm) Figure 3.4.  Absorbance profiles (240-310 nm) for solutions consisting of sodium bisulfite in phosphate buffer adjusted to (A) pH 4.5 and (B) pH 6.2. employed  for  post-column  reaction  detection  (290  nm),  At the wavelength the  background  absorbance is much lower at the higher pH value (6.2).  Figure 3.5.  Effect of post-column reagent pH on baseline fluctuations in the H P L C - P C system. The mobile phase consisted of 20 mM monobasic sodium phosphate, while the P C reagent consisted of 40 mM sodium bisulfite in 20 mM phosphate buffer at (A) pH 4.5 and (B) pH 6.2. Mobile phase and post-column reagent flow rates were 0.7 mL/min and 0.3 mL/min, respectively.  52  Finally, to improve the mixing of the mobile p h a s e a n d p o s t - c o l u m n r e a g e n t s , w e e x a m i n e d the effects of c h a n g e s to the flow rate of the p o s t - c o l u m n reagent. T o e n s u r e optimal c h r o m a t o g r a p h i c performance, mobile p h a s e flow rates w e r e maintained at 0.7 mL/min.  F o r other H P L C - P C m e t h o d s previously reported [4-7], p o s t - c o l u m n reagent  flow rates w e r e kept to a minimum (0.1 to 0.3 mL/min) in order to l e s s e n the dilution of the eluent from the H P L C c o l u m n , thereby maximizing the a s s a y sensitivity.  However,  to further r e d u c e the magnitude of the b a s e l i n e fluctuations o b s e r v e d w e found that e q u a l i z a t i o n of mobile p h a s e a n d p o s t - c o l u m n reagent flow rates at 0.7 m L / m i n w a s a n e c e s s a r y condition (Figure 3.6).  0.3 mL/min  Figure 3.6.  Effect of post-column reagent flow rate on baseline fluctuations in the H P L C - P C system. The mobile phase consisted of phosphate buffer pumped at 0.7 mL/min. The P C reagent consisted of 40 m M sodium bisulfite in phosphate buffer, pumped at the indicated flow rates.  3.3.2. Optimization of the H P L C - P C system P r e v i o u s studies [4,7] on the reaction of s o d i u m bisulfite with carboplatin a n d other platinum c o m p l e x e s h a v e d e m o n s t r a t e d that the reaction kinetics are c o m p l e x . F a c t o r s affecting the reaction include reaction time, temperature, p H , a n d the p r e s e n c e  53  of buffer salts, metal ions, o x y g e n , a n d o r g a n i c modifiers.  F o r the reaction of s o d i u m  bisulfite with cisplatin, M a r s h et al. [4] w e r e able to s e p a r a t e chromatographically one major a n d s e v e r a l minor products, albeit the structural identities of t h e s e products w e r e not e s t a b l i s h e d . H o w e v e r , the o b s e r v e d U V spectral c h a n g e s are consistent with the formation of a higher affinity sulfur-platinum linkage, resulting in a d e c r e a s e d e n e r g y difference b e t w e e n the o c c u p i e d a n d u n o c c u p i e d e n e r g y levels of the s q u a r e planar platinum.  In a n y event, the products formed by the reaction s h o w limited stability, a n d  proper optimization a n d timing of the reaction is n e c e s s a r y to m a x i m i z e the c h a n g e s in a b s o r b a n c e o b s e r v e d at 2 9 0 n m . studies on-line.  F o r this r e a s o n , w e c o n d u c t e d our optimization  S i m i l a r results m a y be a c h i e v a b l e with different c o m b i n a t i o n s of post-  c o l u m n reagent additives and/or reaction conditions.  Chromatography F i g u r e 3.7 s h o w s s a m p l e c h r o m a t o g r a m s following injection of blank p l a s m a ultrafiltrate a n d p l a s m a ultrafiltrate containing carboplatin.  W h i l e c h a n g e s in mobile  p h a s e p H had no effect on the retention of carboplatin (11.5 min), large shifts in the retention time of the most prominent e n d o g e n o u s p e a k w e r e o b s e r v e d .  For example,  this p e a k h a d retention times of 9.5, 12.5, 13.0 a n d 16.5 min w h e n mobile p h a s e p H v a l u e s w e r e 6.0, 5.5, 5.0, a n d 4 . 5 , respectively.  O b v i o u s l y , mobile p h a s e s with p H  v a l u e s b e t w e e n 5.0 a n d 6.0 m a y be problematic, s i n c e slight errors in p H adjustment c o u l d result in co-elution of carboplatin a n d this p l a s m a ultrafiltrate c o m p o n e n t .  Since  solutions of m o n o b a s i c s o d i u m p h o s p h a t e h a v e a p H a r o u n d 4 . 5 , u s e of this buffer salt (without p H adjustment)  e n s u r e d reproducible c h r o m a t o g r a p h y a n d m a x i m i z e d the  s e p a r a t i o n b e t w e e n carboplatin a n d the m o s t prominent e n d o g e n o u s interference.  Post-column Buffer  reagent  composition  type T h e buffers e x a m i n e d w e r e citrate, acetate, a n d p h o s p h a t e (20 m M ) in both the  eluent a n d the p o s t - c o l u m n reagent.  T h e s e salts w e r e c h o s e n b e c a u s e they are  c o m m o n l y u s e d in H P L C a n d h a v e buffering capacity in the p H region n e e d e d for the  54  a s s a y (pH 4.5 to 6.0).  M e a n p e a k heights obtained for p h o s p h a t e w e r e greater than  t h o s e obtained for acetate or citrate (Figure 3.8A); h e n c e , p h o s p h a t e buffer w a s utilized in s u b s e q u e n t optimization studies.  Figure 3.7.  HPLC-PC  chromatograms of (A)  blank plasma ultrafiltrate and (B)  ultrafiltrate containing 8 u.g/mL carboplatin.  plasma  Use of the pH 4.5 mobile phase  resulted in excellent separation between carboplatin (11.5 min) and the major interference (16 min) present in plasma ultrafiltrate samples. Column: Y M C O D S A Q 4.6 x 150 mm (3 urn); mobile phase: 20 mM monobasic sodium phosphate; post-column reagent: 40 mM sodium bisulfite in 20 mM monobasic sodium phosphate, pH 5.5; flow rates: 0.7 mL/min in both pumps; detection: U V 290 nm.  55  QH  T h e effect of p H on the reaction w a s e x a m i n e d by adjusting the p H of the postc o l u m n reagent with s o d i u m hydroxide to final v a l u e s of 5.1, 5.3, 5.5, 5.7, or 5.9.  The  p H of the eluent ( m o n o b a s i c s o d i u m p h o s p h a t e , p H 4.5) w a s not adjusted for this a n d for s u b s e q u e n t e x p e r i m e n t s , s i n c e w e had already e s t a b l i s h e d that m o r e reproducible c h r o m a t o g r a p h y w a s a c h i e v e d at lower p H v a l u e s .  With respect to p e a k  height  r e s p o n s e s , the optimal post-column reagent p H occurred s o m e w h e r e b e t w e e n p H 5.3 a n d 5.5 (Figure 3.8B). A post-column reagent p H of 5.4 w a s s u b s e q u e n t l y u s e d .  Acetonitrile  concentration  C u v e t t e studies by K i z u et al. [7] s h o w e d that the p r e s e n c e of e n h a n c e d the  reaction  of platinum  c o m p o u n d s with s o d i u m  bisulfite.  acetonitrile However,  e x a m i n a t i o n of acetonitrile concentrations from 2 % to 1 0 % in the p o s t - c o l u m n reagent s h o w e d no o b v i o u s relationship b e t w e e n concentration a n d m e a n p e a k height r e s p o n s e (Figure 3 . 8 C ) . A s well, further c o m p a r i s o n of post-column reagents with a n d without 4 % acetonitrile s h o w e d no significant difference b e t w e e n the two. S i n c e acetonitrile h a s the potential to c a u s e additional mixing or diffusional effects b e t w e e n the post-column reagent a n d the mobile p h a s e , w h i c h d o e s not contain acetonitrile, it w a s not utilized in further optimization studies.  Bisulfite  concentration Bisulfite concentrations in the 3 0 - 9 0 m M range w e r e e x a m i n e d . A s illustrated in  F i g u r e 3 . 8 D , m e a n p e a k height r e s p o n s e s obtained i n c r e a s e d from 30 to 70 m M bisulfite; h o w e v e r , the b a s e l i n e noise a n d the range of r e s p o n s e s o b s e r v e d a l s o i n c r e a s e d at higher bisulfite concentrations. With respect to signal-to-noise v a l u e s the optimal bisulfite concentration w a s 4 0 m M . T h i s observation w a s s u p p o r t e d by a further study w h i c h c o m p a r e d post-column reagents containing 4 0 or 50 m M bisulfite with r e s p e c t to p e a k height r e s p o n s e s obtained over a range of carboplatin concentrations (0.1-25 iag/mL) in p l a s m a ultrafiltrate.  Both the analytical sensitivity (slope of the  s t a n d a r d curve) a n d the peak height obtained for the least c o n c e n t r a t e d s a m p l e (0.1  56  u.g/mL) w e r e better w h e n the post-column reagent contained 4 0 m M rather than 50 m M s o d i u m bisulfite.  B  g> CD X  30000  30000  20000  20000  10000  10000  CO CD  CL  c  CO CD  I  CL  Buffer  PC reagent pH  30000  30000  20000  20000  10000  10000  i 1 Peak Height f W I S/N X 50  g> 'CD  I j*: CO CD  CL  c  (0 CD  2%  4%  6%  8%  % Acetonitrile  Figure 3.8.  10%  30 40  50 60 70 80 90  Bisulfite Concentration (mM)  Optimization of the post-column system for: (A) buffer type, (B) reagent pH, (C) acetonitrile concentration, and (D) bisulfite concentration.  57  3.3.3. Determination of limits of detection and quantitation T h e a s s a y L O D of the H P L C - P C method for p l a s m a ultrafiltrate s a m p l e s w a s similar to that of the H P L C - U V method.  Injection of the 0.025 iag/mL s a m p l e s provided  a m e a n p e a k height value of 117 u V with a n approximate signal-to-noise v a l u e of 3:1. T o d e t e r m i n e the a s s a y L O Q , m e a n p e a k height r e s p o n s e s of 191,, 4 2 3 , 6 0 6 , 8 5 5 , 4 0 8 2 , 1 4 9 0 5 , a n d 5 2 5 3 7 (iV w e r e o b s e r v e d at carboplatin concentrations of 0.025, 0.05, 0 . 0 7 5 , 0.1, 0.5, 2, a n d 8 n g / m L , respectively. a g a i n u s e d to construct a 1/y  2  C o n c e n t r a t i o n s a b o v e 0.1 ^ g / m L w e r e  weighted standard curve (Figure 3.9), from w h i c h bias  v a l u e s of 2 . 6 % , 1 9 % , a n d 4 8 % w e r e o b s e r v e d at carboplatin concentrations of 0.075, 0.05, a n d 0.025 n g / m L , respectively. T h u s , the 0.05 p.g/mL s a m p l e c o r r e s p o n d e d to the best estimate of L O Q , although the analytical sensitivity (slope of the calibration curve) w a s slightly better for the H P L C - P C m e t h o d (6.75 ^iV * n g / m L ) than for the H P L C - U V -1  m e t h o d (6.24 |xV * n g / m L ) . 1  0  2  4  6  8  10  Concentration (jig/mL)  Figure 3.9.  Standard curve (0.1-8 fig/mL) used to estimate the L O Q of the H P L C - P C assay. Predicted concentrations for sample concentrations  below 0.1  |a.g/mL were  determined using the best fit equation determined by weighted (1/y ) regression. 2  58  Injection of s o l i d - p h a s e extracts in lieu of direct injection of p l a s m a ultrafiltrate d o e s provide the possibility of obtaining a slightly lower detection limit for the H P L C - P C a s s a y m e t h o d ; however, the utility of this a p p r o a c h is limited in terms of the a m o u n t of p l a s m a ultrafiltrate obtainable from clinical blood s a m p l e s (~2 m L ) a n d the concentration factor (less than two-fold) a c h i e v a b l e with the 3 m L amino extraction cartridges.  3.4.  Conclusions During initial d e v e l o p m e n t of the  HPLC-PC  m e t h o d , c o n c e r n s regarding  a  fluctuating b a s e l i n e d u e to p o o r mixing of the mobile p h a s e a n d P C reagents w e r e r e s o l v e d primarily by i n c r e a s i n g the p H of the eluent entering the U V detector a n d by e q u a l i z i n g the mobile p h a s e a n d post-column reagent flow rates.  Interestingly,  the  p r e s e n c e or a b s e n c e of p u l s e d a m p e n i n g d e v i c e s s e e m e d to h a v e little effect o n b a s e l i n e s o b s e r v e d for the H P L C - P C s y s t e m . On-line optimization of the p o s t - c o l u m n reaction conditions w a s required s i n c e the products formed by the reaction s h o w limited stability. The method.  H P L C - P C a s s a y m e t h o d provides s e v e r a l a d v a n t a g e s o v e r the  HPLC-UV  T h e improved selectivity afforded by U V monitoring of the bisulfite reaction  product at 2 9 0 n m r e m o v e s the n e e d for s a m p l e c l e a n - u p of p l a s m a ultrafiltrate prior to a n a l y s i s , resulting in both time a n d cost s a v i n g s . inherently  W h i l e the H P L C - P C  method is  m o r e sensitive with respect to s i g n a l r e s p o n s e s o b s e r v e d , the  greater  b a c k g r o u n d (baseline fluctuations) results in similar limits of detection a n d quantitation for both H P L C - P C a n d H P L C - U V m e t h o d s . D u e to the a f o r e m e n t i o n e d instability of the p o s t - c o l u m n reagent, s y n t h e s i s of a n appropriate internal standard c o m p o u n d for the HPLC-PC  a s s a y is required to a c c o u n t for any t i m e - d e p e n d e n t c h a n g e s in signal  r e s p o n s e s that m a y b e o b s e r v e d .  I 59  3.5.  References  1.  R.W. Frei, H. Jansen, and U.A. Brinkman.  Post-column reaction detectors for H P L C .  Anal Chem 57: 1529A-1539A (1985). 2.  J.F. Huber, K.M. Jonker, and H. Poppe.  Optimal design of tubular and packed-bed  homogenous flow chemical reactors for column liquid chromatography. Anal Chem 52: 29 (1980). 3.  A.A. Hussain, M. Haddadin, and K. Iga.  Reaction of c/s-platinum with sodium bisulfite.  J Pharm Sci 69: 364-365 (1980). 4.  K.C. Marsh, L.A. Sternson, and A . J . Repta. Post-column reaction detector for platinum II antineoplastic agents. Anal Chem 56: 491-497 (1984).  5.  M. Kinoshita, N. Yoshimura, and H. Ogata.  High-performance liquid chromatographic  analysis of unchanged c/s-diamminedichloroplatinum (cisplatin) in plasma and urine with post-column derivatization. J Chromatogr 529: 462-467 (1990). 6.  H.H. Farrish, P.-H. Hsyu, J.F. Pritchard, K.R. Brouwer, and J . Jarrett. Validation of a liquid chromatography post-column derivatization assay for the determination of cisplatin in plasma. J Pharm Biomed Anal 12: 265-271 (1994).  7.  R. Kizu, T. Yamamoto, T. Yokoyama, M. Tanaka, and M. Miyazaki. A sensitive postcolumn derivatization/UV detection system for H P L C determination of antitumour divalent and quadrivalent platinum complexes. Chem Pharm Bull 43: 108-114 (1995).  60  CHAPTER 4 SYNTHESIS  4.1.  AND EVALUATION  OF INTERNAL  STANDARD  CANDIDATES  Introduction Internal s t a n d a r d s w e r e required for the H P L C a n a l y s i s of carboplatin in order to  m i n i m i z e variability d u e to the extraction procedure ( H P L C - U V method) or d u e to instability of the p o s t - c o l u m n reagent ( H P L C - P C method).  Furthermore, b a s e d o n the  complexity of the s a m p l e treatment a n d detection s y s t e m s u s e d for the H P L C - U V a n d H P L C - P C a s s a y m e t h o d s , t h e s e internal standard c o m p o u n d s n e e d e d to b e closely related  structurally  to  carboplatin.  Chromatographic  analysis  of  the  c y c l o b u t a n e d i c a r b o x y l a t o a n a l o g u e s enloplatin a n d D W A 2 1 1 4 R had d e m o n s t r a t e d that these  compounds  were  too  lipophilic  (with  very  large  k'  values  under  the  c h r o m a t o g r a p h i c conditions e m p l o y e d for carboplatin analysis) to b e of practical utility. A similar p r o b l e m w a s o b s e r v e d for two other platinum c o m p o u n d s e v a l u a t e d , J M - 9 (iproplatin) a n d J M - 5 4 .  T h u s , s y n t h e s i s of a n appropriate structural a n a l o g u e w a s  required. A s d e s c r i b e d by A b r a m s et al.  [1],  malonato a n d  c o m p l e x e s c a n be m a d e by three m e t h o d s .  cyclobutanedicarboxylato  T h e s e include direct reaction of the  p o t a s s i u m salt of the cyclobutanedicarboxylato or malonato ligand with cisplatin in d i m e t h y l f o r m a m i d e [2], reaction of c/s-diamminesulfatoplatinum II with the barium salt of the malonato ligand [3], a n d replacement of the c / s - d i a m m i n e d i a q u o p l a t i n u m II s p e c i e s with a n alkali metal salt of 1,1-cyclobutanedicarboxylic a c i d  [4].  W e b e g a n by  evaluating the first method [2], a s the s y n t h e s i s required only a single step a n d w e a l r e a d y h a d the starting material (cisplatin) available in our laboratory.  Furthermore,  o w i n g to the polar structure a n d g o o d a q u e o u s solubility of carboplatin, w e anticipated that the s y n t h e s i z e d c o m p o u n d s might b e more readily precipitated if the reactions w e r e performed in a n o r g a n i c solvent (dimethylformamide) instead of water.  61  4.2. Experimental 4.2.1. Solutions, mobile phases, and plasma ultrafiltrate A l l solutions a n d mobile p h a s e s w e r e freshly p r e p a r e d using H P L C - g r a d e water ( s e e s e c t i o n 2.2.1).  4.2.2.  Syntheses  Method  of Pasini and Caldirola Synthesis  [2]  of carboplatin w a s first attempted  v i a direct  substitution  of the  c y c l o b u t a n e d i c a r b o x y l a t o moiety by 1,1-cyclopentanediaceto, 1 , 2 - c y c l o h e x a n e d i a c e t o , or 1,2-cyclohexanedicarboxylato moieties (Figure 4.1). dimethylformamide  Cisplatin w a s d i s s o l v e d in  with heating followed by c y c l o b u t a n e d i c a r b o x y l i c acid (or other  ligand) a n d a q u e o u s p o t a s s i u m hydroxide (0.1 N) in a 1:2:1 m o l a r ratio.  T h e reaction  w a s continued for approximately 2 4 h at 6 0 ° C . Following addition of ether, the mixture w a s c o o l e d to 4 °C for 4 8 h to induce product precipitation. Method  of Cleare et al. [4] M e t h C B D C A , M e t h M a l , a n d E t h M a l w e r e p r o d u c e d in a two-step s e q u e n c e , the  first s t e p being the c o n v e r s i o n of tetrachloroplatinate to diiodoplatinum II c o m p o u n d s , the s e c o n d step being the r e p l a c e m e n t of the two chloro ligands with  bidentate  c y c l o b u t a n e d i c a r b o x y l a t o or malonato ligands (Figure 4.2).  Step 1 P o t a s s i u m tetrachloroplatinate (1 eq.) a n d s o d i u m iodide (4 eq.) w e r e d i s s o l v e d s e p a r a t e l y in water then mixed in the dark for 1 m i n , resulting in rapid formation of unstable tetraiodoplatinate. reaction  continued,  M e t h y l a m i n e or ethylamine (2 eq.) w a s a d d e d a n d the  with  bis(methylamine)diiodoplatinum  mustard-coloured II  or  products  (assumed  bis(ethylamine)diiodoplatinum  II)  to  be  starting  to  precipitate within a f e w minutes. After continued stirring for 2 h, the mixture w a s c o o l e d to 4 ° C a n d the products r e m o v e d by filtration.  62  Step 2 S i l v e r nitrate (1.9 eq.) w a s d i s s o l v e d in water followed by the product (1 eq.) isolated from the first step. T h e mixture w a s allowed to stir for 3 h at r o o m temperature a n d precipitated silver iodide w a s then r e m o v e d by filtration.  After addition of e x c e s s  (1.5 eq.) c y c l o b u t a n e d i c a r b o x y l i c or malonic a c i d , the p H w a s adjusted to 6-7 with p o t a s s i u m hydroxide, then the reaction continued at 60 °C for 2 h. After cooling at 4°C for  24  h,  residual silver iodide  precipitate  was  removed  by filtration.  For  the  M e t h C B D C A a n d E t h M a l products, further cooling of the mixture at 4 ° C for s e v e r a l d a y s resulted in precipitation of white crystals.  F o r the M e t h M a l product, no precipitate  formation w a s o b s e r v e d likely d u e to high water solubility for this c o m p o u n d . Instead, product purification w a s a c c o m p l i s h e d v i a s o l i d - p h a s e extraction of the filtrate.  Purification  of MethMal  Product  Following conditioning of a 6 m L Y M C O D S - A Q (1000 mg) extraction cartridge with 3 m L of methanol a n d 2 x 2 m L of water, a 0.5 m L aliquot of the M e t h M a l product w a s d r a w n under v a c u u m onto the sorbent b e d a n d s u b s e q u e n t l y w a s h e d with 2 m L of water. A n additional 4 m L of water w a s then drawn through the cartridge a n d c o l l e c t e d .  Structural  confirmation  using  HPLC-MS  T h e H P L C - M S s y s t e m w a s c o m p o s e d of a H e w l e t t - P a c k a r d m o d e l 1 0 9 0 liquid c h r o m a t o g r a p h c o u p l e d to a V G Quattro q u a d r o p o l e m a s s s p e c t r o m e t e r ( F i s o n s , A l t r i n c h a m , U K ) . T h e identity of the s y n t h e s i z e d c o m p o u n d s w a s investigated using positive electrospray ionization conditions that had b e e n previously o p t i m i z e d for the detection of carboplatin [5], including a s o u r c e temperature of 80 °C a n d a c o n e voltage of 2 4 V .  F o r characterization of the  products of the  reaction of cisplatin with  c y c l o p e n t a n e d i a c e t i c a c i d , c y c l o h e x a n e d i a c e t i c acid, a n d c y c l o h e x a n e d i c a r b o x y l i c acid, a 4.6 x 150 m m (5 u,m) Y M C O D S - A c o l u m n w a s u s e d , a n d the mobile p h a s e c o n s i s t e d of 1 0 0 % water p u m p e d at 1.0 m L / m i n . F l o w from the c o l u m n w a s split, with one-tenth the c h r o m a t o g r a p h i c eluent entering the m a s s spectrometer. F o r characterization of the M e t h M a l , E t h M a l , a n d M e t h C B D C A platinum a n a l o g u e s , a 3.0 x 150 m m (5 urn) Y M C  63  O D S - A Q c o l u m n w a s e m p l o y e d , with a mobile p h a s e of 3 % acetonitrile in 0 . 1 % formic acid p u m p e d directly into the m a s s spectrometer at 0.2 m L / m i n .  Synthesis of carboplatin [2]  o  O HN  Cl  3  Pt  HO-C  H3N  x  ci  DMF/KOH heat 60 °C  HO- C  o-c  H3N-  Pt H3N-  "O-  C  o  o carboplatin  1,1-cyclobutanedicarboxylic acid  Proposed synthesis of analogues  o  O II  HO—C  H3N*  HO—C  H3N  II  ii  Pt  o - c  II  o  o  1,1-cyclopentanediacetic acid  o  O II  HO—C  H 3 N \  HO—C  H3N  II  O  o-c  II  pt:  o-c  ^o-c  II  o  1,1-cyclohexanediacetic acid  H O - C  H O - C 1,2-cyclohexanedicarboxylic acid  Figure 4.1.  Proposed synthesis of carboplatin analogues using the method of Pasini and Caldirola [2], which has been successfully applied to the synthesis of carboplatin.  64  -Cl  cu  4Nal  ;pt; cr  ;pt;  dark  ^ci  Tetrachloroplatinate 2CH3CH2NH2  2CH3NH2  CH CH NH \^ 3  2  CH NH ^  2  3  2  Pt' CH CH NH 3  2  Pt CH NH;f  2  3  Bis(ethylamine)diiodoplatinum II  Bis(methylamine)diiodoplatinum II  2AgN03  2AgN0  Agl(s) CH CH NH ^ 3  2  CH CH NH ^ 2  CH NH \  2  Pt^ 3  Agl(s)  H 0  2  3  20  CH NH ^ 3  2  Cyclobutanedicarboxylic acid  KOH  KOH  2  CHaNhU  /0~  Bis(ethylamine) malonatoplatinum II EthMal  CH NH " /  3  2  -0-C  CHaNH,^  C  Pt  Pt  ^o-c  2  o  O  Pt 3  3  2  .0  CH CH NH -'  (NQ )2  H Q  Malonic acid  /  CHaChfeNH^  2  3  Malonic acid, KOH  ^H 0 Pt^  2  (NQ )2 H  2  3  CH NH ' /  " ^ O -C  O  3  2  *0- C  o  O Bis(methylamine) malonatoplatinum II MethMal  Bis(methylamine)cyclobutane dicarboxylatoplatinum II MethCBDCA  Figure 4.2. Method of Cleare et al. [4] used for synthesis of carboplatin analogues.  65  4.2.3. Evaluation of synthesized internal standard candidates E v a l u a t i o n of the s y n t h e s i z e d c o m p o u n d s w a s b a s e d o n two criteria.  Firstly, a  suitable internal standard n e e d e d to h a v e a n appropriate retention time under the c h r o m a t o g r a p h i c conditions e m p l o y e d by the a s s a y , eluting within a region of b a s e l i n e free  from e n d o g e n o u s  components.  S e c o n d l y , a n appropriate  internal  standard  required similar extraction characteristics to those of carboplatin. Retention time properties on the Y M C O D S - A Q c o l u m n w e r e e v a l u a t e d following injection  of  a  sample  containing  carboplatin  and  the  MethMal,  EthMal,  and  M e t h C D B D C A a n a l o g u e s . M o b i l e p h a s e s u s e d c o n s i s t e d of 1.3% acetonitrile in 20 m M monobasic  sodium  phosphate  and  20  mM  monobasic  sodium  c o r r e s p o n d i n g to t h o s e optimized for the H P L C - U V (section 2.3.4)  phosphate,  and  HPLC-PC  (section 3.3.2) a s s a y m e t h o d s , respectively. T h e s o l i d - p h a s e extraction p r o c e d u r e previously d e s c r i b e d in s e c t i o n s 2.2.5 a n d 2.3.3 w a s u s e d to evaluate the  retention  characteristics of the  internal  standard  c a n d i d a t e s o n the S u p e l c l e a n a m i n o ( N H ) extraction cartridges. A n a q u e o u s standard 2  containing carboplatin (20 fj.g/mL) a n d its M e t h C B D C A , M e t h M a l , a n d E t h M a l a n a l o g u e s w a s p r e p a r e d to provide similar p e a k a r e a v a l u e s for e a c h a n a l o g u e . A l i q u o t s (200 jaL) of this solution w e r e diluted with 2 m L of acetonitrile a n d then a d d e d directly to the extraction cartridges, rinsed, a n d eluted.  After drying a n d s a m p l e reconstitution, p e a k  a r e a v a l u e s for e a c h a n a l o g u e w e r e determined by injection onto the H P L C s y s t e m . F o r c o m p a r i s o n , recovery after  evaporation/reconstitution  (without extraction)  determined by adding 2 m L of eluting solvent to a 2 0 0 |aL aliquot of the c o m p o u n d mixture a n d p r o c e e d i n g directly to the drying step.  was  platinum  Experiments were  performed in quintuplicate.  66  4.3.  Results and Discussion  4.3.1. Syntheses O u r initial attempts  to s y n t h e s i z e carboplatin  a n a l o g u e s involved the  direct  reaction of cisplatin with 1,1-cyclobutanedicarboxylic acid in dimethylformamide.  This  m e t h o d had b e e n s u c c e s s f u l l y utilized by P a s i n i a n d C a l d i r o l a [2] for the s y n t h e s i s of a variety of cyclobutanedicarboxylato, malonato, a n d hydroxymalonato c o m p o u n d s .  We  applied this method first to the s y n t h e s i s of carboplatin from cisplatin, leaving the a m m i n e ligands o n o n e half of the m o l e c u l e a n d replacing the chloro ligands with the bidentate  cyclobutanedicarboxylato  ligand.  Similar  reaction  of  cisplatin  with  c y c l o p e n t a n e - or c y c l o h e x a n e - l i n k e d a c i d s should therefore result in structurally similar a n a l o g u e s of carboplatin, a n a l o g u e s p o s s e s s i n g suitable properties for u s e a s H P L C internal s t a n d a r d s .  Unfortunately, w e w e r e unable to obtain the d e s i r e d products.  Instead, p h y s i c a l descriptions of the reaction products obtained are given in T a b l e 4 . 1 . F o r the cyclobutanedicarboxylato reactant, the yellow crystal precipitate  was  identified by H P L C - U V retention time a n a l y s i s to be residual unreacted cisplatin, while a similar a n a l y s i s of the filtrate identified the p r e s e n c e of the d e s i r e d product (carboplatin). H o w e v e r , e x t e n d e d cooling did not precipitate carboplatin from the product mixture.  For  the c y c l o p e n t a n e d i a c e t o reaction, no significant p e a k s w e r e o b s e r v e d in either the precipitate  or  filtrate  upon  HPLC-UV  analysis.  Solids  obtained  from  the  c y c l o h e x a n e d i a c e t o a n d c y c l o h e x a n e d i c a r b o x y l a t o reactions w e r e subjected to m a s s spectral a n a l y s e s .  F o r both c o m p o u n d s , two  p e a k s of  interest  were  observed,  c o r r e s p o n d i n g to [M+1] (parent ion) at m/z 4 2 7 / 4 0 0 a n d [M+18] at m/z 4 4 6 / 4 1 8 for the +  +  c y c l o h e x a n e d i a c e t o / c y c l o h e x a n e d i c a r b o x y l a t o derivatives, respectively.  W e attributed  the [M+18] p e a k s to be products with monodentate links rather than bidentate links of +  the c y c l o h e x a n e d i a c e t o or c y c l o h e x a n e d i c a r b o x y l a t o reactants to the platinum, thereby a c c o u n t i n g for a n additional water moiety. T h i s hypothesis w a s supported by the faster elution times a n d significant tailing o b s e r v e d for t h e s e p e a k s . Unfortunately, it w a s the m o n o d e n t a t e - l i n k e d product w h i c h w a s the major peak in both c a s e s , with the d e s i r e d parent ion present in m u c h lower a b u n d a n c e , a s s h o w n in Figure 4 . 3 for the product of the c i s p l a t i n - c y c l o h e x a n e d i c a r b o x y l a t o reaction.  67  Table 4.1.  Description of products obtained via the method of Pasini and Caldirola [2].  Reactant ligand  Solid product obtained  Comments  cyclobutanedicarboxylato  distinct yellow crystals (cisplatin)  carboplatin in filtrate  cyclopentanediaceto  indistinct brown precipitate  no peaks upon H P L C analysis  cyclohexanediaceto  fine grey/green precipitate  H P L C and M S analyses done  cyclohexanedicarboxylato  indistinct dark green precipitate  H P L C and M S analyses done  The  inability  of  the  cyclopentanediaceto,  cyclohexanediaceto,  and  c y c l o h e x a n e d i c a r b o x y l a t o ligands to p r o d u c e the desired products m a y b e d u e to steric c o n s i d e r a t i o n s in their binding to the central platinum  atom.  This hypothesis  s u p p o r t e d by H P L C a n a l y s i s of the c y c l o h e x a n e d i c a r b o x y l a t o product (Figure 4.3). major c h r o m a t o g r a p h i c property  is  The  p e a k o b s e r v e d around 9 min s h o w e d significant tailing, a  not o b s e r v e d for cyclobutanedicarboxylato  a n a l o g u e s but c o n s i s t e n t with  interaction of a carboxylic acid moiety with silanol groups o n the silica stationary p h a s e . M a s s s p e c t r a l a n a l y s i s of this peak fraction  p r o d u c e d a parent  ion at m/z  418,  c o n s i s t e n t with the p r e s e n c e of a monodentate linkage (and thus free c a r b o x y l i c acid moiety)  in  place  of  the  bidentate  linkage  which  is  present  c y c l o b u t a n e d i c a r b o x y l a t o ligand a n d s q u a r e - p l a n a r platinum atom.  between  the  Figure 4.3 also  d e m o n s t r a t e s the p r e s e n c e of a small a m o u n t of the d e s i r e d bidentate product (minor p e a k at 13 min), w h i c h w e w e r e unable to isolate.  68  5  Figure 4.3.  HPLC  analysis  of  10  the  product  of  the  Time (min)  reaction  of  cisplatin  with  cyclohexanedicarboxylic acid. Column: Y M C O D S - A 4.6 x 150 mm (5 um); mobile phase: 100% water; flow rate: 0.7 mL/min; detection: U V 230 nm; injection volume: 15 mL. Structures shown were determined by M S analysis of the eluent fractions collected from the major peak at 9 min and the minor peak at 13 min.  69  Following the inability of the n o n - a q u e o u s s y n t h e s i s a p p r o a c h d i s c u s s e d a b o v e to  yield  cyclopentanediaceto,  analogues,  we  decided  to  cyclohexanediaceto,  return to  the  or  cyclohexanedicarboxylato  cyclobutanedicarboxylato  and  malonato  c o m p o u n d s a s a b a s i s for s y n t h e s i s of a w o r k a b l e c o m p o u n d for u s e a s a n internal s t a n d a r d . T h e a q u e o u s - b a s e d method of C l e a r e et al. [3] w a s s e l e c t e d in p l a c e of the m e t h o d of P a s i n i a n d C a l d i r o l a [2], s i n c e our attempts to s y n t h e s i z e carboplatin with the latter ( n o n a q u e o u s ) method had resulted in s u c c e s s f u l product formation but a n inability to precipitate the product from solution. With this n e w m e t h o d , w e readily a c c o m p l i s h e d our d e s i r e d s y n t h e s e s , although  we  still h a d s o m e difficulty  in inducing  product  precipitates. F o r the M e t h C B D C A a n d E t h M a l products, solids w e r e obtained only after cooling the product mixtures for s e v e r a l d a y s . evaporation  of the carboplatin  S i n c e e x t e n d e d cooling a n d  a n d M e t h M a l products failed to yield  purification w a s instead a c c o m p l i s h e d by s o l i d - p h a s e extraction. 4.4,  HPLC-UV  a n a l y s i s of the  purified  partial  precipitates,  A s s h o w n in Figure  filtrate yielded only a single major  peak.  F u r t h e r m o r e , the identities a n d purity of all four s y n t h e s i z e d platinum c o m p o u n d s w e r e confirmed by H P L C - M S (Figure 4.5).  3 Figure 4.4.  Time (min)  H P L C - U V chromatogram of the solid-phase extracted MethMal filtrate. Column: Y M C O D S - A Q 4.6 x 150 mm (3 |im); mobile phase: 1.3% acetonitrile in sodium phosphate; flow rate: 0.7 mL/min; detection: U V 230 nm; injection volume: 10 \xL.  70  MethCBDCA m/z 400  I  20  10  30  EthMal m/z 388  I J ^  10  30  20  Carboplatin m/z 372  10  30  20  MethMal m/z 360  10  20  Time (min)  30  Figure 4.5. H P L C - M S with selected ion recording of parent ions formed  under  positive  electrospray conditions. Conditions are as described in section 4.2.2.  71  4.3.2. MethMal, EthMal, and M e t h C B D C A as internal standards F i g u r e 4.6 s h o w s c h r o m a t o g r a m s obtained following containing a mixture  injection  of a s a m p l e  of carboplatin a n d its M e t h M A L , E t h M A L , a n d  MethCBDCA  a n a l o g u e s , while T a b l e 4.2 s u m m a r i z e s the retention time d a t a obtained.  Interestingly,  the M e t h M a l a n a l o g u e eluted prior to carboplatin w h e n 1.3% acetonitrile w a s a d d e d to the eluent, but later than carboplatin w h e n the eluent c o n s i s t e d of 1 0 0 % buffer.  This  resulted in its co-elution with e n d o g e n o u s p l a s m a ultrafiltrate c o m p o n e n t s under the chromatographic  conditions  employed  by  the  HPLC-UV  assay,  but  reasonable  s e p a r a t i o n under the H P L C - P C a s s a y conditions.  Table 4.2.  Retention time data for carboplatin and its MethMal, EthMal, and M e t h C B D C A analogues.  Compound  Retention time  Retention time  (UV assay conditions )  (PC assay conditions' )  Carboplatin  6.5 min  8.0 min  MethMal  6.0 min  9.5 min  EthMal  29 min  65 min  MethCBDCA  44 min  98 min  3  a  1.3% acetonitrile in phosphate buffer (20 mM; pH 4.5)  b  20 mM phosphate buffer, pH 4.5  Peak  a r e a recovery data for carboplatin a n d  1  its a n a l o g u e s from  s t a n d a r d s following s o l i d - p h a s e extraction is provided in T a b l e 4 . 3 .  aqueous  R e c o v e r i e s of the  two c y c l o b u t a n e d i c a r b o x y l a t o c o m p o u n d s , carboplatin a n d M e t h C B D C A , w e r e similar (94.4%  and  93.9%)  following  extraction.  However,  r e c o v e r i e s of the  malonato  c o m p o u n d s , e s p e c i a l l y E t h M a l , w e r e significantly lower a n d m o r e variable. S i n c e only a slight difference w a s o b s e r v e d in recovery of malonato v e r s u s c y c l o b u t a n e d i c a r b o x y l a t o u p o n nitrogen e v a p o r a t i o n a n d reconstitution, the poorer recovery of the  malonato  a n a l o g u e s must be d u e to greater loss of drug during the s o l i d - p h a s e extraction procedure.  72  2  I  10  10  20  30  40  B  Time (min)  12  i—l*—'  20 Figure 4.6.  40  60  80  Time (min)  H P L C chromatograms of an aqueous solution containing (1) carboplatin,  (2)  MethMal, (3) EthMal, and (4) M e t h C B D C A under conditions optimized for the (A) H P L C - U V and (B).HPLC-PC methods. Magnified views of the early time periods are shown in the insets.  73  Table 4.3.  Mean percentage recoveries for  carboplatin  and  its  MethMal, EthMal, and  M e t h C B D C A analogues. % C V values are given in parentheses. Carboplatin  MethMal  EthMal  MethCBDCA  Evaporated* (n=5)  98.0 (1.1)  96.0 (2.5)  95.6(1.0)  98.8 (0.5)  Overall** (n=5)  94.4 (1.5)  88.5 (4.1)  79.0 (3.0)  93.9(1.0)  Sample  * Samples were simply evaporated and reconstituted; recoveries are relative to unevaporated standards ** Samples were subjected to both solid-phase extraction and evaporation reconstitution; recoveries are versus unextracted/unevaporated standards  4.4.  prior  to  Conclusions W e e v a l u a t e d the retention times of e a c h of the internal s t a n d a r d c a n d i d a t e s  under the c h r o m a t o g r a p h i c conditions u s e d in both the H P L C - U V a n d H P L C - P C a s s a y methods.  A s well, w e  e x a m i n e d the  s o l i d - p h a s e extraction  r e c o v e r i e s of  each  c o m p o u n d . F o r the H P L C - U V a s s a y , only M e t h C B D C A (with a retention time of 4 5 min) p o s s e s s e d similar extraction characteristics to carboplatin. A l t h o u g h the elution time of this c o m p o u n d is s o m e w h a t lengthy, the H P L C - U V a s s a y requires a long run time in a n y e v e n t d u e to the p r e s e n c e of late-eluting e n d o g e n o u s c o m p o u n d s .  Extraction  differences are not of c o n c e r n with the H P L C - P C m e t h o d , a s no s a m p l e c l e a n - u p is required.  F o r this m e t h o d , M e t h M a l had theiiimost appropriate retention time (9.5 min),  eluting shortly after carboplatin (8.0 min), while the retention times of E t h M a l a n d M e t h C B D C A w e r e m u c h longer.  74  4.5.  References  1.  M.J. Abrams.  The chemistry of platinum antitumour agents.  In The Chemistry  of  Antitumour Agents, Chapman and Hall, New York (1990). 2.  Pasina and C. Caldirola.  A new synthetic method for diaminomalonatoplatinum type  complexes and the unexpected behaviour of [PtCI (trans-dach)]. Inorganica Chimica Acta 2  150: 19-20 (1988). 3.  R.C. Harrison, C.A. McAuliffe, and A.M. Zaki. A n efficient route for the preparation of highly soluble platinum II antitumour agents. Inorg Chim Acta 46: L15 (1980).  4.  M.J. Cleare, J.D. Hoeschele, B. Rosenberg, and L.L. Van Camp.  Malonato platinum  anti-tumor compounds. US Patent 4,140,707 (1979). 5.  R.B. Burns, R.W. Burton, S.P. Albon, and L. Embree.  Liquid chromatography-mass  spectrometry for the detection of platinum antineoplastic complexes. J Pharm AnalU:  Biomed  367-372(1996).  75  CHAPTER 5 VALIDATION  5.1.  OF HPLC-UV  AND HPLC-PC  ASSAY  METHODS  Introduction R e v i e w s c o n c e r n i n g the validation of bioanalytical m e t h o d s h a v e b e e n written by  K a r n e s et al. [1] a n d by L a n g a n d Bolton [2].  A c o m p l e t e validation  encompasses  parameters:  characterization  of  the  following  procedure  specificity/selectivity,  sensitivity (limits of quantitation a n d / o r detection), reproducibility (precision, a c c u r a c y , a n d linearity), recovery, a n d r u g g e d n e s s / r o b u s t n e s s . A s s a y validation implies a fixed concentration r a n g e o v e r w h i c h a method is e v a l u a t e d .  T h e concentration  range  c h o s e n for validation m a y be b a s e d on estimates of the a s s a y ' s linear r a n g e (upper a n d lower limits of a s s a y linearity) or simply be b a s e d on the experimenter's n e e d s . Specificity a n d selectivity refer to the ability to distinguish the a n a l y t e f r o m other c o m p o u n d s present in a s a m p l e . T h e two terms are frequently u s e d interchangeably, although there exists a subtle difference in their m e a n i n g . W h e r e a s selectivity implies the ability of a c h r o m a t o g r a p h i c separation to s e p a r a t e the analyte f r o m p o s s i b l e interferences, specificity implies that only the analyte p r o d u c e s a d e t e c t a b l e r e s p o n s e . Specificity is d e m o n s t r a t e d by c o m p a r i s o n of "blank" c h r o m a t o g r a m s from the relevant biological matrix with c h r o m a t o g r a m s from the s a m e matrix containing the analyte. Selectivity of a s e p a r a t i o n is s h o w n by characterizing the c h r o m a t o g r a p h y of the drug together with that of c l o s e l y related c o m p o u n d s (e.g. metabolites or  degradation  products). T h e limits of detection a n d quantitation are the lowest analyte c o n c e n t r a t i o n s that c a n be reliably distinguished from a blank a n d that c a n b e reliably  quantitated,  respectively. T h e s e terms are frequently u s e d to d e s c r i b e the "sensitivity" of a n a s s a y , although the analytical sensitivity is a l s o defined a s the s l o p e of the r e s p o n s e function [3].  In c o n t e m p o r a r y practice it is c o m m o n for the lowest calibration c u r v e standard to  b e referred to a s the limit of quantitation [4], s i n c e it is the lowest concentration that h a s operationally  been  demonstrated  to  yield  accuracy and  precision  values  within  predefined limits d e e m e d " a c c e p t a b l e . " T h i s c a n be c o n f u s i n g , s i n c e reported limits of  76  quantitation m a y not be c l o s e to the lowest concentrations that c o u l d potentially be reliably quantitated.  Alternatively, limits of quantitation m a y be estimated using s i g n a l -  t o - n o i s e v a l u e s or extrapolated f o r m u l a e [5],  from  calibration  curve  data  using  computational  T h e performance of the a s s a y at the lower e n d of the calibration range  c a n then be stated simply with reference to the lowest Q C standard e v a l u a t e d . Precision,  accuracy, and  c o n s i d e r e d together.  linearity  are  interdependent  terms  and  are  thus  In practice, precision, a c c u r a c y , a n d linearity are e s t a b l i s h e d by  a s s a y i n g replicate standard curve a n d Q C s a m p l e s during different d a y s , or b a t c h e s , of validation.  P r e c i s i o n refers to the c l o s e n e s s of repetitive m e a s u r e m e n t s a n d c a n be  further divided into precision of the apparatus (injection, c h r o m a t o g r a p h i c , a n d p e a k integration variability), precision of the method (the previous variability plus s a m p l e handling  variability),  and  precision  of  results  (similar  to  method  precision  but  incorporating the effect of data treatment on the results). P r e c i s i o n v a l u e s are typically expressed  as  the  measurements.  relative  standard  deviation  (%RSD,  %CV)  of  a  series  of  F o r s a m p l e s taken over multiple validation b a t c h e s , m e t h o d precision  is usually poorer than precision of the transformed d a t a b e c a u s e the s t a n d a r d c u r v e s run with e a c h batch help a c c o u n t for a s s a y variability. difference  A c c u r a c y , or b i a s , is the  b e t w e e n predicted concentrations a n d the "true"  concentration,  usually  obtained by spiking the analyte with a known concentration of drug. T h e a c c u r a c y at a particular concentration is a function of the goodness-of-fit of both the c h o s e n r e s p o n s e function a n d weighting s c h e m e . In this respect, a c c u r a c y a n d linearity are linked, s i n c e a linear r e s p o n s e function c a n not be d e e m e d appropriate if it results in poor a c c u r a c y . In practice, however, linearity is often d i s c u s s e d simply with reference to the r v a l u e s 2  obtained from a s e r i e s of calibration c u r v e s . Alternatively, investigators m a y refer to the "linear r a n g e " of a n a s s a y , roughly c o r r e s p o n d i n g to the upper a n d lower concentration limits w h i c h provide r v a l u e s of 0.99 or greater. 2  A n a l y t e recovery is determined by c o m p a r i s o n of r e s p o n s e s from  extracted  s a m p l e s v e r s u s s t a n d a r d s into w h i c h the analyte is s p i k e d following extraction of a matrix blank.  T h i s blank extraction is n e c e s s a r y to c o m p e n s a t e for the possibility that  the matrix itself m a y h a v e s o m e effect on the o b s e r v e d detector r e s p o n s e . T h e internal s t a n d a r d is not extracted but a d d e d just prior to injection in order to a c c o u n t for  77  a p p a r a t u s variability. Optimal recovery v a l u e s should be a s c l o s e a s p o s s i b l e to 1 0 0 % ; h o w e v e r , it is far more important that the recovery be reproducible.  In c a s e s w h e r e  r e c o v e r y variability is significant, the u s e of an internal standard with similar extraction properties to the analyte is n e c e s s a r y . S a m p l e stability of the drug s h o u l d be a s s e s s e d under the s t o r a g e conditions to be  u s e d for the unknown s a m p l e s .  T h i s includes both the relevant m a t r i c e s a n d  t e m p e r a t u r e s to w h i c h the analyte is e x p o s e d and the e x p o s u r e time. S a m p l e instability is a significant factor contributing to r u g g e d n e s s p r o b l e m s .  R u g g e d n e s s of a method  refers to the reproducibility of results under c h a n g i n g experimental conditions. includes  changes  in  laboratory/instrument/operator  time  (within-day  and  day-to-day  This  variability),  conditions, reagent lots, a n d stability of s t a n d a r d s or  s a m p l e s . C l o s e l y related to r u g g e d n e s s is robustness, w h i c h relates to the ability of a n a s s a y m e t h o d to remain unaffected by deliberate variations in method p a r a m e t e r s . T h i s c h a p t e r d e s c r i b e d the validation of both H P L C - U V a n d H P L C - P C m e t h o d s previously optimized for quantitation of carboplatin in p l a s m a ultrafiltrate. and  interpretation  of all the  relevant validation  parameters  Derivation  are d i s c u s s e d .  The  validation results are then e m p l o y e d to select the most appropriate a s s a y m e t h o d to be s u b s e q u e n t l y u s e d for the p h a r m a c o k i n e t i c evaluation of blood s a m p l e s obtained from pediatric patients (as d e s c r i b e d in C h a p t e r 6).  5.2.  Experimental  5.2.1. Preparation of mobile phases S e e s e c t i o n 2.2.1 regarding preparation of buffers a n d m o b i l e p h a s e s . T o m a k e 1 L of H P L C - U V  mobile p h a s e , 2.76 g m o n o b a s i c s o d i u m p h o s p h a t e a n d 13 m L  acetonitrile w e r e mixed a n d brought to v o l u m e with water in a 1 L E r l e n m e y e r flask. The  H P L C - P C mobile p h a s e w a s similarly p r e p a r e d , e x c e p t that no acetonitrile w a s  added.  T o m a k e 1 L of H P L C - P C post-column reagent, 2.76 g m o n o b a s i c s o d i u m  p h o s p h a t e a n d 4.16 g of s o d i u m bisulfite w e r e d i s s o l v e d together in approximately 500 m L water.  T h e mixture w a s adjusted to p H 5.4 via addition of 1 M a q u e o u s s o d i u m  hydroxide, then brought to v o l u m e (1 L) with water.  78  5.2.2. Chromatography T h e W a t e r s H P L C s y s t e m , a s d e s c r i b e d previously in s e c t i o n s 2.2.3 a n d 3.2.2, c o n s i s t e d of two m o d e l 5 1 0 P u m p s , a m o d e l 7 1 2 W I S P autoinjector, a n d a m o d e l 4 8 4 v a r i a b l e w a v e l e n g t h U V detector. F o r both m e t h o d s , c h r o m a t o g r a p h i c s e p a r a t i o n s w e r e performed o n a Y M C O D S - A Q 4.6 x 150 m m (3 u,m) c o l u m n with a 4 x 2 3 m m guard c o l u m n containing the s a m e 3 u.m packing material.  F o r the H P L C - U V a s s a y , the  mobile p h a s e c o n s i s t e d of 1.3% acetonitrile in 2 0 m M m o n o b a s i c s o d i u m p h o s p h a t e , p u m p e d isocratically at 0.7 m L / m i n . F o r the H P L C - P C a s s a y , the mobile p h a s e w a s 2 0 m M m o n o b a s i c s o d i u m p h o s p h a t e , while the post-column reagent w a s 4 0 m M s o d i u m bisulfite in 2 0 m M s o d i u m phosphate (final p H adjusted to 5.4 with s o d i u m hydroxide). Both the mobile p h a s e a n d post-column reagent flow rates w e r e 0.7 m L / m i n . T h e postc o l u m n reactor w a s kept at ambient temperature, while the post-column reagent w a s protected from light a n d c o o l e d in a n ice/water bath at 0 °C to minimize s o d i u m bisulfite d e g r a d a t i o n . O t h e r a s s a y properties w e r e a s d e s c r i b e d in T a b l e 5.1 below.  Table 5.1. Properties of the H P L C - U V and H P L C - P C assays. HPLC-UV Assay  HPLC-PC Assay  230 nm  290 nm  MethCBDCA  MethMal  t (carboplatin)  6.5 min  11.5 min  t (internal standard)  45 min  13 min  sample run time  52 min  26 min  analytical wavelength internal standard r  r  5.2.3. Preparation of plasma ultrafiltrate Both  the  HPLC-UV  and H P L C - P C  assay  methods  were  used  for the  determination of u n b o u n d carboplatin. T h i s protein-free fraction w a s obtained by ultracentrifugation  of p l a s m a (0.7-1 m L ) in Centrifree ultrafiltration units a s d e s c r i b e d in  s e c t i o n 2.2.2.  F o r the H P L C - P C a s s a y , a 150 u L aliquot w a s s p i k e d with the internal  standard (MethMal) a n d injected directly onto the H P L C c o l u m n .  F o r the H P L C - U V  79  a s s a y , a n aliquot of p l a s m a ultrafiltrate w a s extracted using the a m i n o s o l i d - p h a s e cartridges prior to injection.  5.2.4. Solid-phase extraction (HPLC-UV assay) P l a s m a ultrafiltrate (200 (aL) w a s s p i k e d with internal standard ( M e t h C B D C A ) , diluted with 2 m L of acetonitrile, a n d a d d e d directly to the extraction cartridge (3 m L S u p e l c l e a n a m i n o ; 5 0 0 m g sorbent) w h i c h w a s pre-conditioned with 2 m L of 9 5 / 5 acetonitrile/water.  Following a w a s h step with 2 m L of 90/10 acetonitrile/water, 2 m L of  5 0 / 2 5 / 2 5 acetonitrile/methanol/water w a s u s e d to elute carboplatin w h i c h w a s collected and dried under N g a s at 4 0 ° C . O n c e dry, the s a m p l e w a s reconstituted with 150 |uL 2  of mobile p h a s e a n d 6 0 [iL injected onto the H P L C c o l u m n .  5.2.5. U s e of peak height/area ratio values C a r b o p l a t i n a n d internal standard p e a k s w e r e integrated  using the W a t e r s  M a x i m a 8 2 0 c h r o m a t o g r a p h y software.  F o r the H P L C - U V m e t h o d , w e e x a m i n e d the  performance  height  of  plots  of  both  peak  ratio  and peak  (carboplatin/internal standard) v e r s u s carboplatin concentration.  area  ratio  values  F o r the H P L C - P C  m e t h o d , only p e a k height ratios w e r e e m p l o y e d d u e to the c l o s e elution (R 0.8-1.0) of s  carboplatin a n d a n e n d o g e n o u s c o m p o n e n t of p l a s m a ultrafiltrate (as d i s c u s s e d further in s e c t i o n 5.3.2).  5.2.6. Validation experiments Specificity  and  Selectivity  S m a l l aliquots ( - 0 . 5 m L ) of p l a s m a (n=14) w e r e obtained from pediatric patients at C a l g a r y C h i l d r e n ' s Hospital. A l l patients w e r e o n c h e m o t h e r a p e u t i c r e g i m e n s that did not include carboplatin a n d the s a m p l e s w e r e collected in h e p a r a n i z e d t u b e s .  Eight of  t h e s e fourteen "blank" s a m p l e s w e r e a n a l y z e d by the H P L C - U V m e t h o d , the remaining six by the H P L C - P C m e t h o d . Additionally, p l a s m a ultrafiltrate w a s p r e p a r e d from blood that w a s collected from healthy adult volunteers (n=5) a n d a n a l y z e d by both H P L C - U V and H P L C - P C m e t h o d s .  80  Interference from drugs co-administered with carboplatin w a s a l s o investigated. A l i q u o t s of p l a s m a ultrafiltrate, s p i k e d with e a c h c o m p o u n d at a concentration of 50 u g / m L , w e r e a s s a y e d by both H P L C - U V a n d H P L C - P C m e t h o d s . Selectivity of the Y M C O D S - A Q c o l u m n for carboplatin in the p r e s e n c e of its chloro-substituted  degradation  products w a s  examined  by  exposing  an  aqueous  carboplatin standard (15 u.g/mL) to 1 M hydrochloric acid over 2 h a n d characterizing the mixture by injection of a n aliquot (20 u.L) onto the H P L C - U V s y s t e m at 0, 10, 2 0 , 30, 4 0 , 6 0 , 9 0 , a n d 120 min.  Limits of detection  and  quantitation  A s d e s c r i b e d in s e c t i o n s 2.3.5 a n d 3.3.3, limits of detection a n d quantitation for both m e t h o d s w e r e previously estimated to be 0.025 and 0.05 u,g/mL, respectively. T h e latter v a l u e provided the b a s i s for the lower limit (0.05 uxj/mL) of the validation range. T h e u p p e r concentration limit, 4 0 u,g/mL, w a s c h o s e n after e x a m i n a t i o n of reported elimination profiles of carboplatin following 1 h infusion of 1 7 5 - 6 0 0 m g / m doses  [6].  T h e major  objective  of the validation  2  carboplatin  p r o c e d u r e w a s to a s s e s s  the  p e r f o r m a n c e of the H P L C - U V a n d H P L C - P C a s s a y m e t h o d s o v e r this concentration range.  Reproducibility  experiment  (precision,  accuracy,  and  linearity)  T h r e e b a t c h e s of p l a s m a ultrafiltrate s a m p l e s w e r e a s s e s s e d , e a c h consisting of a s t a n d a r d c u r v e a n d five sets of Q C s a m p l e s . F o r e a c h standard c u r v e or Q C set a s e p a r a t e w e i g h i n g of p o w d e r w a s m a d e followed by serial dilutions to the appropriate concentrations.  A  standard  curve  consisted  of  the  following  nine  carboplatin  concentrations: 0.05, 0.1, 0.2, 0.5, 2, 8, 15, 2 5 , a n d 4 0 u g / m L . Q C s e t s c o n s i s t e d of five concentrations: 0.05, 0.1, 0.2, 8, a n d 4 0 u.g/mL. T o e n s u r e a s s a y r u g g e d n e s s , s a m p l e b a t c h e s w e r e not run o n c o n s e c u t i v e d a y s , but w e r e s e p a r a t e d by s e v e r a l w e e k s . During batch 3 of validation, additional s a m p l e s w e r e run at e a c h s t a n d a r d curve concentration not part of the Q C sets s u c h that six c o m p l e t e s t a n d a r d c u r v e s w e r e run on that d a y .  T h i s a l l o w e d for a within batch c o m p a r i s o n of r e g r e s s i o n e q u a t i o n s  81  obtained from multiple calibration c u r v e s .  M o r e o v e r , injections of the batch 3 Q C  s a m p l e s (but not the standard curve) w e r e r a n d o m i z e d to e n s u r e that there w e r e no p r o b l e m s with carry-over b e t w e e n injections or c h a n g e s in s i g n a l r e s p o n s e with time. With r e s p e c t to the generation of standard c u r v e s , the a c c u r a c y of s e v e r a l linear r e g r e s s i o n f o r m u l a s w a s a s s e s s e d in addition to the standard l e a s t - s q u a r e s ( P e a r s o n ) f o r m u l a . T h e s e included weighted (1/y a n d 1/y ) v e r s i o n s of the l e a s t - s q u a r e s formula, 2  a s well a s a non-parametric form of linear r e g r e s s i o n ( P a s s i n g / B a b l o k ; S M E Statistics™ Inc., V a n c o u v e r , B C , C a n a d a ) .  P a s s i n g / B a b l o k r e g r e s s i o n u s e s a n iterative p r o c e d u r e ,  m a k i n g a large n u m b e r of s l o p e calculations from w h i c h a best-fit line is d e t e r m i n e d . A s c o m p a r e d to l e a s t - s q u a r e s r e g r e s s i o n methods, the P a s s i n g / B a b l o k a p p r o a c h is m o r e resistant to the impact of outlying data points [7].  Standard least-squares regression  a s s u m e s that there is n o variability present in concentration v a l u e s a n d that the m a g n i t u d e of y-directional v a r i a n c e s (signal r e s p o n s e s ) d o not vary with concentration, a s s u m p t i o n s that a r e usually untrue for c h r o m a t o g r a p h i c m e t h o d s .  A more common  situation is that of proportional error, w h e r e b y the C V r e m a i n s approximately constant a c r o s s m o s t of the range of concentration v a l u e s , but sharply i n c r e a s e s at very low concentration v a l u e s .  In that situation, weighted  r e g r e s s i o n p r o c e d u r e s result in  s u p e r i o r a c c u r a c y at low analyte concentrations, thereby improving the quantitation limit of the a s s a y , without sacrificing significant a c c u r a c y at higher analyte concentrations [8]. T h e c o m p u t a t i o n a l formulae e m p l o y e d by the standard a n d weighted least s q u a r e s linear r e g r e s s i o n p r o c e d u r e s w e r e p r o g r a m m e d into Microsoft E x c e l a n d w e r e a s follows: Standard (Pearson) Regression E x = s u m ([Xj - x  avg  ] ) w h e r e xi ( x  avg  ) are.the ith (mean) concentration v a l u e s  E y = s u m ([yi - y  avg  ] ) w h e r e yi ( y  avg  ) a r e the ith (mean) height (or area) ratio v a l u e s  2  2  2  2  E x y = s u m ([XJ - x slope = E x y / E x intercept = y  a v g  a v g  ] * [y - y  avg  ])]  '  2  - slope * (x  avg  )  r = ( E x y ) / ( E x * Ey ) 2  2  2  2  82  W e i g h t e d (1/v a n d 1/y ) R e g r e s s i o n 2  W| = 1/y; or 1/y  2  (weighting factor)  s l o p e = E(Wj*xy) / E(Wj*x ) 2  intercept = E ( w * y ) / E(W|) - s l o p e * E(w *x) / E(Wj) t  2  = (E(Wi*xy)) / ( E ( * x ) E(Wi*y)) 2  Wi  Recovery R e c o v e r y of carboplatin following ultrafiltration w a s e v a l u a t e d at c o n c e n t r a t i o n s of 0.5, 8, a n d 2 5 u,g/mL. A q u e o u s carboplatin s t a n d a r d s w e r e p r e p a r e d in quintuplicate at e a c h concentration a n d the p e a k a r e a s obtained upon injection onto the s y s t e m c o m p a r e d before a n d after  ultrafiltration.  R e c o v e r y following  HPLC-UV  solid-phase  extraction w a s a l s o e v a l u a t e d at similar concentration v a l u e s . P e a k a r e a ratios of five extracted ultrafiltrate s a m p l e s w e r e c o m p a r e d to t h o s e from five s a m p l e s p r e p a r e d by a d d i n g carboplatin to a n extract of blank p l a s m a ultrafiltrate.  T h e internal s t a n d a r d w a s  a d d e d after the extraction p r o c e d u r e ; this s e r v e d simply to a c c o u n t for injection a n d c h r o m a t o g r a p h i c variabilities.  Stability Stability of carboplatin in b l o o d , p l a s m a , p l a s m a ultrafiltrate, extracts of p l a s m a ultrafiltrate w a s e x a m i n e d using the  HPLC-UV  and solid-phase a s s a y method.  Stability of blood at 4 °C w a s e v a l u a t e d over a 24 h period at c o n c e n t r a t i o n s of 0.5, 8, a n d 2 5 u.g/ml.  A t times 0, 12, a n d 2 4 h, three aliquots w e r e r e m o v e d from a large-  v o l u m e s t a n d a r d at e a c h concentration, centrifuged to obtain p l a s m a , a n d a s s a y e d . Stability of p l a s m a , p l a s m a ultrafiltrate, a n d s o l i d - p h a s e extracts of p l a s m a ultrafiltrate was also evaluated.  P l a s m a s t a n d a r d s (0.5 a n d 2 5 u.g/mL) w e r e s u b j e c t e d to f r e e z e -  t h a w c y c l e s , with multiple a s s a y i n g (n=5) of e a c h s t a n d a r d being performed prior to f r e e z i n g at -70 °C a n d a g a i n after o n e a n d two months.  For plasma  ultrafiltrate,  carboplatin w a s a d d e d to multiple s a m p l e s (n=10), then five s a m p l e s w e r e extracted immediately a n d five w e r e c o o l e d for 1 h at 15 °C prior to extraction.  For solid-phase  83  extracts of p l a s m a ultrafiltrate, injections w e r e m a d e immediately following extraction a n d repeated after the s a m p l e s (n=5) had remained in the a u t o s a m p l e r tray for 60 h, a s well a s both before a n d after freezing at -70 °C for 3 months.  Ruggedness In order to a s s e s s the variability of the H P L C - U V a n d H P L C - P C a s s a y m e t h o d s following r e p e a t e d s a m p l e injections under the finalized a s s a y conditions, s u c c e s s i v e calibration  c u r v e s (n=9)  were  run  using  unextracted  aqueous standards.  Each  calibration c u r v e c o n s i s t e d of the s a m e (0.05, 0.1, 0.2, 0.5, 2, 8, 15, 2 5 , a n d 4 0 (xg/mL carboplatin) s a m p l e s plus internal s t a n d a r d .  P e a k a r e a ( H P L C - U V ) a n d p e a k height  ( H P L C - P C ) v a l u e s of both carboplatin a n d the internal standard w e r e then recorded to e v a l u a t e w h e t h e r a n y time-dependent c h a n g e s in signal r e s p o n s e w e r e o b s e r v e d .  5.2.7. Statistical comparisons C o m p a r i s o n of m e a n s of data sets w e r e m a d e using the single factor o n e - w a y a n a l y s i s of v a r i a n c e ( A N O V A ) program a n d t-tests contained in Microsoft E x c e l .  F o r all  c o m p a r i s o n s , a p v a l u e l e s s than 0.05 w a s c o n s i d e r e d statistically significant.  84  5.3.  Results and Discussion  5.3.1. Specificity and selectivity The  retention  time  of  carboplatin  was  initially  identified  under  the  c h r o m a t o g r a p h i c conditions of both H P L C - U V a n d H P L C - P C a s s a y s v i a injection of freshly p r e p a r e d a q u e o u s s a m p l e s containing carboplatin, w h i c h provided only a single major p e a k at the retention times indicated in T a b l e 5.1.  Identification of t h e s e p e a k s  w a s confirmed by similar a n a l y s i s of carboplatin p o w d e r obtained from a n alternative s o u r c e ( J o h n s o n - M a t t h e y , C h e s t e r , P A , U S A ) a n d by H P L C - M S a n a l y s i s a s d e s c r i b e d in s e c t i o n 4 . 2 . 2 .  Figure 5.1.  Chromatograms of an aqueous carboplatin standard exposed to 1 M hydrochloric acid at (A) 0 min and (B) 120 min. Impurities in the acid were responsible for the large solvent front around 3 min. Injection of cisplatin resulted in a peak with a retention time less than 4 min (not shown).  Selectivity of the c h r o m a t o g r a p h i c s y s t e m for carboplatin w a s d e m o n s t r a t e d by f o r c e d d e g r a d a t i o n of a n a q u e o u s carboplatin standard with hydrochloric acid (Figure 5.1).  A s the chloro-substituted degradation products are f o r m e d , the carboplatin p e a k  at 6.5 min d i s a p p e a r s a n d at least o n e n e w p e a k (at 3.5 min) a p p e a r s .  D u e to the  c o m p l e t e d i s a p p e a r a n c e of the carboplatin p e a k , w e c o n c l u d e that the Y M C O D S - A Q  85  c o l u m n is i n d e e d selective for carboplatin in the p r e s e n c e of t h e s e m o r e hydrophilic d e g r a d a t i o n products. For both H P L C - U V a n d H P L C - P C m e t h o d s , no interferences w e r e noted for p l a s m a ultrafiltrate s a m p l e s containing the following drugs frequently c o - a d m i n i s t e r e d in the  ifosfamide-carboplatin-etoposide  Chapter  6):  dexamethasone,  (ICE)  chemotherapy  dimenhydrinate,  protocol  etoposide,  o n d a n s e t r o n , s u l f i s o x a z o l e , a n d trimethoprim (Figure 5.2).  (described  ifosfamide,  in  nystatin,  N o significant p e a k s w e r e  found at the retention time of carboplatin in control p l a s m a obtained from healthy adult volunteers (n=5) a n d from pediatric patients (n=14) receiving c h e m o t h e r a p e u t i c agents other than carboplatin.  S i n c e c o m p o u n d s s u c h a s t h e s e c o - a d m i n i s t e r e d drugs are  likely m o r e n o n p o l a r t h a n carboplatin, they m a y remain on c o l u m n during a s e q u e n c e of injections.  T h i s provides a further rationale for both proper validation of the analytical  m e t h o d s o v e r a n e x t e n d e d time period (including m a n y s u c c e s s i v e injections), a n d for inclusion of Q C s a m p l e injections before and after e a c h batch of clinical s a m p l e s evaluated.  n  o  ~r 0  3  6  9  T i m e (min)  Figure 5.2. Specificity of the H P L C - U V a s s a y method for carboplatin in the p r e s e n c e of c o - a d m i n i s t e r e d drugs (50 u.g/mL of e a c h ) . C h r o m a t o g r a m s are of p l a s m a ultrafiltrate s a m p l e s containing carboplatin (8 u.g/mL), a n d similar s a m p l e s with ifosfamide a n d e t o p o s i d e , a n d with d e x a m e t h a s o n e , dimenhydrinate, nystatin, o n d a n s e t r o n , s u l f i s o x a z o l e , a n d trimethoprim.  86  5.3.2. Use of peak height/area ratio values F o r the H P L C - U V method, the u s e of p e a k a r e a rather than p e a k height ratios resulted  in  improved  assay  precision.  We  observed  a  slight  broadening  in  c h r o m a t o g r a p h i c p e a k s h a p e s during the validation p r o c e d u r e s , a n effect that w a s more p r o n o u n c e d on the M e t h C B D C A p e a k at 4 5 min than o n the carboplatin p e a k at 6.5 min.  Both p e a k height a n d p e a k a r e a ratio v a l u e s t e n d e d to i n c r e a s e with s u b s e q u e n t  injections, but a r e a ratios w e r e more resistant to this effect a n d thus g a v e better p r e c i s i o n v a l u e s at all carboplatin concentrations studied (Table 5.2). F o r the H P L C - P C  method, only p e a k height ratio v a l u e s w e r e u s e d .  The  b r o a d e n i n g effect noted a b o v e w a s not a s problematic for this a s s a y d u e to the similar retention times of carboplatin (11.5 min) a n d the M e t h M a l internal standard (13 min).  In  addition, resolution v a l u e s of 0.8-1.0 for carboplatin a n d a c l o s e l y eluting e n d o g e n o u s ultrafiltrate c o m p o n e n t m a d e the u s e of p e a k a r e a ratio v a l u e s u n a c c e p t a b l e [10].  Table 5.2.  H P L C - U V data from batch 1 Q C samples containing 0.1, 8, and 40 ^g/mL carboplatin. 0.1 uxj/mL Carboplatin  8 ).ig/mL Carboplatin  40 |ig/mL Carboplatin  Height Ratio  Area Ratio  Height Ratio  Area Ratio  0.0018 4  1.0570  0.1796  5.6521  0.9719  0.0125  0.0020  1.0606  0.1804  5.2516  0.9162  3  0.0121  0.0021  1.0410  0.1779  5.4636  0.9363  4  0.0161  0.0021  1.1067  0.1869  5.5161  0.9408  5  0.0137  0.0021  1.1512  0.1955  5.7622  0.9701  mean  0.0132  0.0020  1.0833  0.1841  5.5291  0.9470  %CV  13  6.3  4.2  3.9  3.5  2.5  QC Set  Height Ratio  1  0.0118  2  Area Ratio  87  5.3.3. Reproducibility experiment (precision, accuracy, and linearity) Precision For a s s a y m e t h o d s u s e d for p h a r m a c o k i n e t i c studies, S h a h et al. [4] defined a c c e p t a b l e precision a n d a c c u r a c y v a l u e s a s not more than  have  1 5 % C V in  r e s p o n s e for precision a n d not more than 1 5 % deviation from e x p e c t e d concentration (i.e. 1 5 % bias) for a c c u r a c y .  A t the L O Q , however, 2 0 % C V or bias is a c c e p t a b l e .  M e a n p e a k a r e a or p e a k height ratios (with % C V v a l u e s in p a r e n t h e s e s ) obtained o v e r the three d a y s of validation are s h o w n in T a b l e 5.3 ( H P L C - U V a s s a y ) a n d T a b l e 5.4 ( H P L C - P C a s s a y ) . F o r the H P L C - U V m e t h o d , all C V v a l u e s (both intra-batch a n d total) w e r e within a c c e p t a b l e limits, with the imprecision rising steeply b e t w e e n the 0.2 a n d 0.05 u.g/mL concentrations.  F o r the H P L C - P C m e t h o d , m a n y of the C V v a l u e s for Q C  c o n c e n t r a t i o n s of 0.05, 0.1, a n d e v e n 0.2 u.g/mL w e r e greater than 1 5 % .  Accuracy For both H P L C - U V a n d H P L C - P C a s s a y m e t h o d s , the most f a v o u r a b l e results w e r e g e n e r a t e d using the 1/y  2  weighted a p p r o a c h .  E q u a t i o n s g e n e r a t e d from this  r e g r e s s i o n p r o c e d u r e w e r e u s e d to calculate predicted concentration v a l u e s for e a c h of the Q C s a m p l e s , from w h i c h bias v a l u e s (differences b e t w e e n s p i k e d a n d d e t e r m i n e d concentrations) w e r e in turn d e r i v e d . F o r the H P L C - U V m e t h o d , a c c e p t a b l e m e a n bias v a l u e s w e r e a c h i e v e d at all carboplatin concentrations e x a m i n e d (Table 5.5). F o r the H P L C - P C m e t h o d , most but not all of the m e a n b i a s v a l u e s for the predicted concentrations w e r e within a c c e p t a b l e limits (Table 5.6).  M o r e o v e r , the bias  v a l u e s o b s e r v e d w e r e generally larger than those o b s e r v e d for the H P L C - U V method and did not improve a s m u c h at the higher carboplatin Q C concentrations e x a m i n e d .  88  Table 5.3.  Peak area ratio precision data for the H P L C - U V assay.  Cone. (ng/mL)  Batch 1 (n=5) mean (CV)  Batch 2 (n=5) mean (CV)  Batch 3 (n=5) mean (CV)  Total (n=15) mean (CV)  0.05  0.0011 (8.7%)  0.0008 (10%)  0.0008 (17%)  0.0009 (20%)  0.100  0.0020 (6.3%)  0.0017 (11%)  0.0017 (11%)  0.0018 (11%)  0.200  0.0044 (3.2%)  0.0036 (5.3%)  0.0041 (9.2%)  0.0040 (10%)  8.00  0.1841 (3.9%)  0.1671 (6.1%)  0.1699 (4.5%)  0.1741 (6.3%)  40.0  0.9471 (2.5%)  0.8322 (3.1%)  0.9106(3.0%)  0.8966 (6.1%)  w  Table 5.4. Peak height ratio precision data for the H P L C - P C assay. Cone. (ng/mL)  Batch 1 (n=5) mean (CV)  Batch 2 (n=5) mean (CV)  Batch 3 (n=5) mean (CV)  Total (n=15) mean (CV)  0.05  0.0062 (4.8%)  0.0046 (18%)  0.0095 (13%)  0.0068 (34%)  0.100  0.0134(11%)  0.0121 (13%)  0.0173(12%)  0.0143(19%)  0.200  0.0268(16%)  0,0262(11%)  0.0360(17%)  0.0295 (21%)  8.00  1.268 (2.5%)  1.524 (9.3%)  1.701 (11%)  1.519 (15%)  40.0  7.346 (7.0%)  8.412(9.9%)  9.053 (8.4%)  8.276(12%)  89  Table 5.5. Accuracy data for Q C samples generated by weighted (1/y ) linear regression 2  ( H P L C - U V assay). Shown are the mean predicted concentrations with %bias and % C V values. Cone. uxj/mL  Batch 1 (n=5) mean bias CV  Batch 2 (n=5) mean bias CV  Batch 3 (n=5) mean bias CV  Total(n=15) mean bias CV  0.050  0.058 +18%  6.4%  0.060 +20% 6.4%  0.045 -10%  13%  0.055 +9.1% 15%  0.100  0.095 -4.8%  5.3%  0.102 +2.2% 8.5%  0.086 -14%  9.3%  0.095 -5.4% 10%  0.200  0.190 -5.0%  2.9%  0.186 -7.2% 4.7%.:  0.193 -3.6% 4.5%  0.190 -5.2% 5.8%  8.00  7.38  -7.7%  3.9%  7.67  -4.1% 6.1%  7.63  -4.6% 4.5%  7.55  -5.6% 4.8%  40.0  37.9  -5.2%  2.5%  38.1  -4.6% 3.1%  40.8 +2.1% 3.0%  39.0  -2.6% 4.4%  Table 5.6. Accuracy data for Q C samples generated by weighted (1/y ) linear regression 2  ( H P L C - P C assay). Shown are the mean predicted concentrations with %bias and % C V values. Cone. (ig/mL  Batch 1 (n=5) mean bias CV  Batch 2 (n=5) mean bias CV  Batch 3 (n=5) mean bias CV  Total(n=15) mean bias CV  0.050  0.051 +1.2% 3.4%  0.045 -9.3% 9.3%  0.054 +7.4% 11%  0.050 -0.3% 11%  0.100  0.091 - 9 . 1 %  9.4%  0.084 -16%  9.5%  0.092 -8.2% 1 1 %  0.089 - 1 1 %  10%  0.200  0.167 -17%  14%  0.158 - 2 1 %  9.8%  0.184 -7.9% 16%  0.169 -15%  15%  8.00  7.24  -9.5%  2.5%  8.32  +4.0% 9.3%  8.40  +5.0% 11%  7.96  -0.4% 11%  40.0  41.6  +3.9%  7.0%  43.8  +9.5% 9.9%  44.7  +12% 8.4%  43.3  +8.3% 8.5%  90  F o r both H P L C - U V a n d H P L C - P C a s s a y m e t h o d s , statistical a n a l y s i s ( A N O V A ) s h o w e d that the m e a n peak height ratios (untransformed data) a n d m e a n predicted concentration v a l u e s (transformed data) for at least s o m e of the Q C concentrations studied w e r e different over all three b a t c h e s of validation s a m p l e s . T h i s d e m o n s t r a t e s the n e e d to run a standard curve for e a c h n e w batch of s a m p l e s a n a l y z e d . E x a m i n a t i o n of the precision ( % C V ) v a l u e s s h o w n in T a b l e s 5.3 a n d 5.4 (untransformed data) v e r s u s similar v a l u e s in 5.5 a n d 5.6 (transformed data) d e m o n s t r a t e s that the precision of the d a t a s e t s is greatly improved following application of the 1/y weighted linear regression 2  transformation.  Not only w e r e the % C V v a l u e s better for the c o m b i n e d d a t a s e t s (n=15)  at all concentrations e v a l u a t e d , they w e r e a l s o greatly improved with respect to the intra-batch precision v a l u e s o b s e r v e d at the lowest three Q C concentrations (0.05 to 0.2 l^g/mL).  Linearity Linearity w a s e x a m i n e d both in terms of the goodness-of-fit of the d a t a to the linear  equation  used  and  its  ability  to  accurately  predict  concentration  values.  Calibration c u r v e s beginning e a c h d a y of validation a n d multiple calibration c u r v e s run during batch 3 of validation allow for examination of both within batch a n d b e t w e e n b a t c h variability in detector r e s p o n s e . U s i n g the 1/y weighted r e g r e s s i o n a p p r o a c h , r 2  v a l u e s w e r e in all c a s e s greater than 0.999 for both a s s a y m e t h o d s .  2  M e a n regression  e q u a t i o n s obtained w e r e a s follows:  H P L C - U V assay (within batch; n=3)  y = 0 . 0 2 2 5 (+/- 2.4%) * x - 0.0004 (+/-35%)  (between batch; n=6)  y= 0.0236'(+/- 7.4%) * x - 0.0004 (+/-34%)  H P L C - P C assay (within batch; n=6)  y = 0.2120 (+/-4.9%) * x - 0.0022 (+/- 4 8 % )  (between batch; n=3)  y = 0.1906 (+/- 6.8%) * x - 0.0027 (+/- 51%)  91  T h e large variability in the y-intercept v a l u e s obtained is likely more indicative of their c l o s e n e s s to z e r o than to variability in the a s s a y m e t h o d s t h e m s e l v e s .  For  e x a m p l e , the inter-batch intercepts of -0.0004 a n d - 0 . 0 0 2 7 for the H P L C - U V a n d H P L C P C m e t h o d s roughly c o r r e s p o n d to concentration v a l u e s of -17 n g / m L a n d -14 n g / m L , v a l u e s w h i c h are m u c h lower than the lowest Q C concentration e v a l u a t e d (50 ng/mL). W h i l e the high r  2  a n d r e a s o n a b l e s l o p e precision v a l u e s for the H P L C - U V a n d  H P L C - P C a s s a y s s u g g e s t the p r e s e n c e of a similar a m o u n t of variability in the two m e t h o d s , t h e s e v a l u e s are s o m e w h a t m i s l e a d i n g .  T h e precision a n d a c c u r a c y data  s h o w n in T a b l e s 5.3 to 5.6 clearly s h o w that the variability of the H P L C - P C method w a s c o n s i d e r a b l y greater than that of the H P L C - U V m e t h o d . A s d e s c r i b e d in section 5.2.6, the injection of the batch 3 Q C s a m p l e s w a s r a n d o m i z e d , while the initial standard c u r v e s a m p l e s w e r e injected in order of increasing carboplatin concentration.  While sample  r a n d o m i z a t i o n h a d little or no effect on the H P L C - U V a s s a y (Figure 5.3), it w a s very problematic for the H P L C - P C a s s a y (Figure 5.4). Nearly e v e r y data point for the H P L C U V calibration c u r v e s falls c l o s e to the best-fit line; however, m a n y of the H P L C - P C calibration d a t a points fall well a b o v e or below this line. T h e e x c e p t i o n is for the initial H P L C - P C calibration curve, where the injection order w a s not r a n d o m i z e d .  If w e  c o n s i d e r that e a c h of t h e s e data points could in fact represent a clinical s a m p l e then the variability present in the H P L C - P C method is s o m e w h a t alarming. T h i s variability could be lowered to s o m e extent by making multiple injections of e a c h clinical s a m p l e a n d t h e n using the m e a n v a l u e s obtained. H o w e v e r , this is not a convenient solution d u e to the limited s a m p l e v o l u m e s available a n d the fact that additional s a m p l e injections w o u l d i n c r e a s e the total time n e e d e d to a s s a y a set of s a m p l e s , thereby resulting in further d e g r a d a t i o n of the post-column reagent.  92  Figure 5.3.  Batch 3 calibration curves for the H P L C - U V assay method. The standard curve was prepared from nonrandomized samples, while the injection order for the samples in the remaining Q C sets was randomized. Note the tight scatter of data points around the best-fit lines.  93  Figure 5.4.  Batch 3 calibration curves for the H P L C - P C assay method. The standard curve was prepared from nonrandomized samples, while the injection order for the samples in the remaining Q C sets was randomized.  Increased scatter of data  points around the best-fit lines of the randomized samples was observed.  94  5.3.4.  Recovery Following the s o l i d - p h a s e extraction p r o c e d u r e , m e a n carboplatin r e c o v e r i e s  (n=5) w e r e 8 3 . 4 % , 8 6 . 2 % , a n d 8 8 . 5 % for carboplatin concentrations of 0.5, 8, a n d 25 u,g/mL  T h e s e results w e r e lower than t h o s e o b s e r v e d previously for a q u e o u s extracts  (section 4.3.2) but in a g r e e m e n t with t h o s e s e e n for p l a s m a ultrafiltrate s a m p l e s during m e t h o d d e v e l o p m e n t , in w h i c h o b s e r v e d r e c o v e r i e s ranged from 8 0 - 9 0 % . L o s s e s m a y reflect  nonspecific binding  to  the  cartridges,  instability  of  carboplatin  under  the  extraction conditions, or incomplete drug transfer between extraction s t e p s . Binding of carboplatin to the A m i c o n Centrifree filtration units w a s minimal, with r e c o v e r i e s of greater than 9 8 % obtained at all concentrations e v a l u a t e d (Table 5.7). S i n c e binding of carboplatin to p l a s m a proteins, D N A , a n d R N A is irreversible, no c h a n g e s in the b o u n d / u n b o u n d equilibrium w o u l d be e x p e c t e d during ultrafiltration a n d l o s s e s result from Centrifree unit itself.  n o n s p e c i f i c binding b e t w e e n the drug a n d m e m b r a n e filter or Furthermore, the extent of drug binding o b s e r v e d for a q u e o u s  solutions s h o u l d be at least a s great a s that o b s e r v e d for p l a s m a ultrafiltrate s a m p l e s . Table 5.7.  Recovery of carboplatin following ultrafiltration.  Concentration (ng/mL)  Mean Area Standard (uV*s)  Mean Area Ultrafiltrate (uV*s)  Recovery (%)  0.500  10915  11062  98.7  8.00  212662  215418  98.7  25.0  661794  669601  98.8  5.3.5. Stability (HPLC-UV method) U s i n g a n H P L C - U V a s s a y d e v e l o p e d for quantitation of carboplatin from 1 to 50 u.g/mL in h u m a n p l a s m a , a thorough examination of the stability of carboplatin in p l a s m a a n d p l a s m a ultrafiltrate w a s m a d e by G a v e r a n d D e e b [9].  T h e y found that carboplatin  h a d limited stability in p l a s m a w h e n stored at - 2 5 °C, degrading with a half-life of about 50 d a y s .  T h i s effect a p p e a r e d to s h o w a concentration d e p e n d e n c e a n d w a s e v e n  m o r e p r o n o u n c e d for ultrafiltrates of p l a s m a stored at similar t e m p e r a t u r e s . Following a period of 6 d a y s in w h i c h no c h a n g e s in concentration w e r e o b s e r v e d , d e g r a d a t i o n half-  95  lives of 17 a n d 36 d a y s w e r e reported for p l a s m a ultrafiltrate carboplatin concentrations of 5 a n d 4 0 |ag/ml_, respectively. Similarly, a study by E r k m e n et al. [10] reported loss of ultrafilterable platinum from p l a s m a s a m p l e s containing carboplatin stored at - 2 5 °C. H o w e v e r , the s a m e study d e m o n s t r a t e d that p l a s m a s a m p l e s w e r e stable w h e n stored at -70 °C. Blood M e a n p e a k a r e a ratios ( c a r b o p l a t i n / M e t h C B D C A ) h a d d e c r e a s e d after 12 a n d 2 4 h at both concentrations studied; however, the d e c r e a s e s o b s e r v e d w e r e s m a l l (<5%) a n d statistical a n a l y s i s ( A N O V A ) did not s h o w a significant difference b e t w e e n the m e a n p e a k a r e a ratios at the different time points (Table 5.8). Table 5.8. Time (h)  Peak area ratio data for blood samples stored at 4 °C (n=3 at each concentration). 0.5 u,g/ml_ SD mean  %CV  8 u.g/mL SD mean  %cv  25 u^g/mL SD mean  %cv  0  0.00584  0.00041  7.0  0.1425  0.0059  4.1  0.438  0.0079  1.8  12  0.00582  0.00063  11  0.1372  0.0050  3.6  0.438  0.0052  1.2  24  0.00583  0.00054  9.3  0.1362  0.0058  4.3  0.435  0.0101  2.3  Plasma After two m o n t h s , the m e a n p e a k a r e a ratios of the f r o z e n p l a s m a s a m p l e s had d e c l i n e d by 3 . 5 % a n d 4 . 9 % at concentrations of 0.5 uxj/mL a n d 2 5 n g / m L s a m p l e s , respectively (Table 5.9). Table 5.9.  Peak area ratio data for plasma samples stored at -70 °C (quintuplicate assay of large volume standards).  mean  25 ng/mL SD  %CV  3.0  0.4938  0.0060  1.2  0.00062  9.2  0.4690  0.0123  2.6  0.00022  2.9*  0.4700  0.0099  2.1  mean  0.5 ng/mL SD  %cv  0  0.00796  0.00036  1  0.00680  2  0.00778  Month  * statistically different from 0 months as determined by one-tailed t-test (P=0.045)  96  Plasma  ultrafiltrate No  significant  differences  in a r e a ratio v a l u e s w e r e  o b s e r v e d for  plasma  ultrafiltrate s a m p l e s c o o l e d at 15 °C for 1 h (the conditions u s e d for preparation of p l a s m a ultrafiltrate).  F o r the reconstituted extracts of p l a s m a ultrafiltrate stored at -70  °C for three months, p e a k a r e a ratio v a l u e s w e r e stable (no statistically significant differences, T a b l e 5.10).  F o r the extracts left in the a u t o s a m p l e r for 60 h, s m a l l but  statistically significant differences w e r e o b s e r v e d (Table 5.11). Table 5.10.  Stability of H P L C - U V sample extracts frozen at -70 °C (n=5 at each concentration). After 3 months  Prior to freezing Cone. (ng/mL)  Mean Area Ratio  SD  %CV  Mean Area Ratio  SD  %CV  % Change in Area Ratio  0.2  0.0043  0.0004  9.0  0.0042  0.0003  7.9  -2.7  8  0.1850  0.0077  4.2  0.1779  0.0101  5.7  -3.8  40  0.9423  0.0249  2.6  0.9316  0.0143  1.5  -1.1  Table 5.11 . Autosampler stability of H P L C - U V extracts (n=5 at each concentration). 60 h  0 h Cone. (ng/mL)  Mean Area Ratio  SD  0.5  0.0080  9.4x10"  25  0.4920  0.0064  5  %CV  Mean Area Ratio  SD  1.2  0.0076  9.4x10"  ' 1.3  0.4829  0.0104  %CV  % Change in Area Ratio  1.2  -6.5*  2.2  -1.8*  5  * statistically significant difference as determined by paired sample t-test  In  summary,  the  stability  data demonstrate  that  carboplatin o c c u r r e d in p l a s m a s a m p l e s stored at -70 ° C .  minimal  degradation  of  W h i l e still relatively stable,  extracts left in the a u t o s a m p l e r for 6 0 h s h o w e d greater degradation than any of the c o o l e d or f r o z e n matrices.  H o w e v e r , s o l i d - p h a s e extracts of p l a s m a ultrafiltrate w e r e  c o m p l e t e l y s t a b l e for 3 months at - 7 0 °C. W e therefore r e c o m m e n d that clinical p l a s m a s a m p l e s be extracted a s s o o n a s p o s s i b l e after being r e c e i v e d .  If n e c e s s a r y , t h e s e  extracts (with internal standard) c a n be f r o z e n at -70 °C a n d a s s a y e d at a later date.  97  5.3.6.  Ruggedness With respect to the validation d a t a presented in this chapter, the  m e t h o d w a s not rugged.  HPLC-PC  It had high intra-batch a n d b e t w e e n batch variability a n d the  o b s e r v e d variability i n c r e a s e d w h e n s a m p l e injections w e r e r a n d o m i z e d . variability c a n be attributed to instability of the post-column reagent.  T h i s high  D e g r a d a t i o n of  s o d i u m bisulfite results in c h a n g e s in the rate or extent of its reaction with carboplatin. S i m i l a r but not identical c h a n g e s are a l s o o b s e r v e d in the reaction of s o d i u m bisulfite with the internal standard (MethMal), resulting in time-dependent c h a n g e s in the p e a k height or p e a k a r e a ratios o b s e r v e d . T h e s e c h a n g e s w e r e evident following injection of multiple (n=9) a q u e o u s calibration c u r v e s o v e r a 35 h time period. Figure 5.5 s h o w s the p e a k height v a l u e s obtained for s e l e c t e d carboplatin c o n c e n t r a t i o n s (0.5, 2, a n d 8 u.g/mL) from e a c h calibration curve, a s well a s p e a k height v a l u e s for the internal standard (MethMal).  P e a k heights for both carboplatin a n d the  internal standard i n c r e a s e d upon s u c c e s s i v e injections for approximately the first 24 h, t h e n s u b s e q u e n t l y d e c l i n e d . Furthermore, the decline profiles w e r e different, with the internal standard p e a k heights falling off rapidly a n d later recovering a n d the carboplatin p e a k heights maintaining a plateau in r e s p o n s e before falling off. T h e potential utility of the p o s t - c o l u m n a s s a y is thus limited to the time period during w h i c h c h a n g e s in the M e t h M a l internal standard mimic t h o s e of carboplatin (less than 2 4 h). In contrast to the H P L C - P C m e t h o d , the H P L C - U V method w a s r u g g e d . S a m p l e stability w a s d e m o n s t r a t e d a s d i s c u s s e d in section 5.3.5.  S a m p l e extractions w e r e  performed using four lots of S P E cartridges (randomly mixed) a n d g a v e r e a s o n a b l e a n d c o n s i s t e n t recoveries. T h e variability in p e a k a r e a precision a n d a c c u r a c y v a l u e s w a s a c c e p t a b l e a c r o s s the entire concentration range e x a m i n e d (0.05 to 4 0 L i g / m L ) . A s well, no t i m e - d e p e n d e n t c h a n g e s in r e s p o n s e w e r e o b s e r v e d after injection of s u c c e s s i v e calibration c u r v e s (n=9) over more than a 7 0 h time period.  98  • • A •  60000 -  MethMal 0.5 ng/mL 2 ng/mL 8 ng/mL  •jj? 40000 x  0) CO  CL  20000 -  0 0  10  20  30  Time (h)  Figure 5.5.  HPLC-PC  responses observed  (peak  heights)  for  MethMal  and  selected  carboplatin concentrations after repetitive injection (n=9) of an aqueous calibration curve. 5.4.  Conclusions In our validation  study, w e  o b s e r v e d a s m a l l albeit statistically  significant  d e g r a d a t i o n of carboplatin in p l a s m a a n d p l a s m a ultrafiltrate extracts o v e r time p e r i o d s e x t e n d i n g past t h o s e typically e m p l o y e d during routine a n a l y s e s . H o w e v e r , within the time p e r i o d s u s e d for routine a n a l y s e s , no degradation of carboplatin w a s o b s e r v e d . During reproducibility e x p e r i m e n t s , variability for the H P L C - P C  method  was  m u c h higher than anticipated, d e s p i t e cooling of the p o s t - c o l u m n reagent in a n icew a t e r bath maintained at 0 °C.  Indeed, for the H P L C - P C  method m a n y of Q C  c o n c e n t r a t i o n s s t u d i e s yielded precision a n d a c c u r a c y v a l u e s b e y o n d the a c c e p t a b l e limits p r o p o s e d by S h a h et al. [4].  T h u s , despite the a d v a n t a g e s of the  HPLC-PC  m e t h o d , w h i c h include a shorter run time a n d lack of s a m p l e handling requirements, the H P L C - U V method w a s s u b s e q u e n t l y utilized for determination of free carboplatin levels in the p h a r m a c o k i n e t i c study ( C h a p t e r 6).  99  5.5. 1.  References H.T. Karnes, G. Shiu, and V.P. Shah. Validation of bioanalytical methods. Pharm Res 8: 421-426 (1991).  2.  J.R. Lang and S.M. Bolton.  A comprehensive method validation strategy  bioanalytical applications in the pharmaceutical industry.  for  J Pharm Biomed Anal 9(5):  357-361 (1991). 3.  International Union of Pure and Applied Chemistry.  Nomenclature, symbols, units,  and their usage in spectrochemical analysis II. Anal Chem 48: 2294-2296 (1976). 4.  V.P. Shah, K.K. Midha, S. Dighe, I.J. McGilveray, J.P. Skelly, A. Yacobi, T. Layloff, C.T. Viswanathan, C.E. Cook, R.D. McDowall, K.A. Pittman, and S. Analytical methods validation: bioavailability,  Spector.  bioequivalence, and pharmacokinetic  studies. Pharm Res 9: 588-592 (1992). 5.  I. Krull and M. Swartz.  Determining limits of quantitation and detection.  LC-GC  16:  922-923(1998). 6.  R. Riccardi, A. Riccardi, A. Lasorella, C. Di Rocco, G. Carelli, A. Tornesello, T. Servidei, A. lavarone, and R. Mastrangelo. Clinical pharmacokinetics of carboplatin in children. Cancer Chemother Pharmacol 33: 477-483 (1994).  7.  W. Bablok and H. Passing. A general regression procedure for method transformation: application of linear regression procedures for method comparison studies in clinical chemistry, part III. J Clin Chem Clin Biochem 26: 783 (1988).  8.  G.K. Szabo, H.K. Browne, A. Ajami, and E.G. Josephs. Alternatives to least squares linear regression analysis for computation of standard curves for quantitation by high performance liquid chromatography: applications to clinical pharmacology.  J Clin  Pharmacol 34: 242-249 (1994). 9.  L.R. Snyder and J.J. Kirkland (Editors). Introduction to modern liquid chromatography (2nd edition). John Wiley and Sons, New York, 15-82 (1979).  10.  R.C. Gaver and G. Deeb. High-performance liquid chromatographic procedures for the analysis of carboplatin in human plasma and urine. Cancer Chemother Pharmacol  16:  201-206 (1986). 11.  K.E. Erkmen, M.J. Egorin, L.M. Reyno, R. Morgan, and J.H. Doroshow.  Effects of  storage on the binding of carboplatin to plasma proteins. Cancer Chemother  Pharmacol  35: 254-256 (1995).  100  CHAPTER 6 EVALUATION  6.1.  OF CARBOPLATIN  PHARMACOKINETICS  Introduction C o m p a r e d to cisplatin, carboplatin h a s a similar s p e c t r u m of a n t i c a n c e r activity  but r e d u c e d i n c i d e n c e s of nephrotoxicity a n d neurotoxicity.  In the a b s e n c e of t h e s e  toxicities, the dose-limiting toxicity b e c o m e s myelotoxicity, primarily t h r o m b o c y t o p e n i a [1,2].  Considerable  effort  has  been  invested  in  understanding  the  clinical  p h a r m a c o k i n e t i c s of carboplatin in adults a n d in d e v e l o p i n g m e t h o d s to u s e this information to optimize therapy.  M a n y carboplatin studies involve multiple  drugs.  A l t h o u g h carboplatin h a s not b e e n s h o w n to interact p h a r m a c o k i n e t i c a l l y with any c h e m o t h e r a p e u t i c agent p r o l o n g e d treatment.  in particular  [3], the  potential for toxicity  P h a r m a c o k i n e t i c studies of carboplatin  i n c r e a s e s with  have  utilized  both  ultrafilterable carboplatin (parent drug) a n d ultrafilterable platinum (parent drug plus non protein-bound degradation products).  T h e lack of a s s a y m e t h o d o l o g y with sufficient  specificity a n d sensitivity to quantitate the parent drug h a s resulted in the u s e of n o n s p e c i f i c a s s a y m e t h o d s in m o s t clinical studies, thereby complicating p h a r m a c o k i n e t i c e v a l u a t i o n . S t u d i e s s h o w i n g that levels of free platinum a n d free parent drug are nearly identical for 4 - 1 2 h after i.v. administration of carboplatin h a v e led s o m e investigators to c o n c l u d e that s p e c i f i c a n d non-specific a s s a y m e t h o d s are i n t e r c h a n g e a b l e .  However,  this a s s e r t i o n h a s yet to b e d e m o n s t r a t e d . A s d i s c u s s e d in C h a p t e r 1, the correlation b e t w e e n carboplatin elimination a n d renal  function  (GFR)  allow  prediction  of  AUC  values  from  renal  clearance  m e a s u r e m e n t s . T h i s relationship is the b a s i s of the adult d o s i n g formula d e v e l o p e d by C a l v e r t et al. [4], w h i c h is e m p l o y e d to calculate the carboplatin d o s e n e c e s s a r y to a c h i e v e a target A U C .  D o s i n g b a s e d on target A U C v a l u e s , w h i c h a c c o u n t s for  differences in renal function, is c o m m o n in the adult population s i n c e up to a three-fold variation h a s b e e n o b s e r v e d in A U C v a l u e s resulting from a g i v e n d o s e b a s e d o n body s u r f a c e a r e a [5].  C o n s i d e r i n g the  narrow therapeutic  margin  between  treatment  s u c c e s s a n d toxicity, this variation in A U C m a y h a v e s e v e r e c o n s e q u e n c e s .  101  A brief review of s o m e important pediatric pharmacokinetic studies on carboplatin (Table 6.1) d e m o n s t r a t e s that literature A U C calculations in this patient group h a v e to d a t e b e e n mainly b e e n b a s e d o n 2 4 h A A d a t a using a variety of d a t a points to c h a r a c t e r i z e the elimination of free platinum. clearly differentiate  between  Unfortunately, s o m e investigators d o not  parameters derived from free carboplatin from  d e r i v e d from free platinum m e a s u r e m e n t s .  those  F o r e x a m p l e , R i c c a r d i et al. [6] c o n c l u d e d  that the two w e r e interchangeable, s i n c e m e a s u r e m e n t s taken to 8 h post-infusion g e n e r a t e d similar results for free platinum a n d free carboplatin.  T h i s h a s led other  authors, s u c h a s D o z et al. [7], to report on the p h a r m a c o k i n e t i c s of free carboplatin w h e n their parameter estimates are in reality b a s e d on free platinum m e a s u r e m e n t s .  Table 6.1.  Pediatric studies using free platinum measurements to investigate carboplatin pharmacokinetics.  Note the differences in sampling times, weighting schemes,  and methods used for calculation of A U C values.  1  Ref.  Doses (mg/m )  Sampling (post-infusion)  Weighting  AUC  6  200-1200  0, 1,2, 4, 6, 8, 12, 24 h  none  trapezoidal  7  560  0, 0.25, 0.5, 1, 2, 6, 12, 24 h  none  trapezoidal  8  escalated  1, 3, 6, 12, 24 h  inverse of variance  parametric  1  9  400-700  0.5, 1, 3, 6, 12, 18, 24 h  inverse of variance  parametric  2  2  one-compartment and two-compartment data fits ( A D A P T II software) 2  T h i s chapter e v a l u a t e s carboplatin  pharmacokinetics  in two  receiving carboplatin in combination c h e m o t h e r a p e u t i c r e g i m e n s .  young  patients  C o m p a r i s o n s are  m a d e b e t w e e n the elimination profiles of free carboplatin a n d free platinum in t h e s e patients to determine if they are interchangeable. platinum  A U C estimates b a s e d o n different  s p e c i e s , s a m p l i n g intervals, a n d calculation m e t h o d s are c o m p a r e d  r e c o m m e n d a t i o n s are m a d e regarding carboplatin a n a l y s i s a n d derivation  and  of A U C  estimates.  102  6.2. Experimental 6.2.1. Clinical protocol Carboplatin-containing Hospital  as  part  of  the  samples  protocol  were  entitled  obtained  from  Vancouver  "Clinical P h a r m a c o l o g y of  Children's Ifosfamide,  C a r b o p l a t i n , a n d E t o p o s i d e (ICE) C h e m o t h e r a p y in Pediatric Patients". P a t i e n t s within this protocol r e c e i v e d carboplatin (i.v.-infused o v e r 1 h) on d a y 1 a n d a g a i n o n d a y 2 a s part of a monthly c h e m o t h e r a p e u t i c c y c l e .  B l o o d s a m p l e s w e r e collected in 2 m L  heparin V a c u t a i n e r s a n d w e r e d r a w n pre-infusion a n d at 0.5, 1, 1.5, 2, 3, 4, 5, 6, a n d 8 h post-infusion.  Additional s a m p l e s w e r e s c h e d u l e d for midnight, 6 a . m . , a n d n o o n ,  r e g a r d l e s s of the  actual infusion  time.  O n c e d r a w n , the  blood  samples  were  refrigerated at 4 °C for up to 12 h prior to the preparation of p l a s m a ultrafiltrate.  As  detailed in section 6.2.2, a total of eight d o s e s of carboplatin (four d o s e s e a c h from two patients) w e r e e v a l u a t e d . 6.2.2. Patient characteristics Table 6.2 provides  a brief summary  of the physical  characteristics  of patients  1 and 2.  Patient 1 w a s a 19-year-old white m a l e with o s t e o g e n i c s a r c o m a of the right femur.  After  diagnosis, he  was  initially  treated  with two  c o u r s e s of  cisplatin,  d o x o r u b i c i n , a n d h i g h - d o s e methotrexate followed by tumour e x c i s i o n in J u n e 1998. I C E c h e m o t h e r a p y w a s b e g u n in J u l y 1998 b e c a u s e tumour histology d e m o n s t r a t e d a low p e r c e n t a g e of n e c r o s i s in the e x c i s e d tumour,  s u g g e s t i n g that the  c h e m o t h e r a p y w a s insufficient to e r a d i c a t e all c a n c e r o u s cells present.  previous  S i x c y c l e s of  I C E c h e m o t h e r a p y w e r e g i v e n , the last in D e c e m b e r 1998. In J a n u a r y 1 9 9 9 , a 7th c y c l e of carboplatin w a s administered  in combination with thiotepa  and  melphalan  as  preparative treatment for a u t o l o g o u s b o n e marrow transplantation. B l o o d s a m p l e s w e r e obtained following both carboplatin d o s e s given during c y c l e 5 a n d c y c l e 7. Patient 2 w a s a 5-year-old f e m a l e d i a g n o s e d with bilateral W i l m s tumour (renal) in M a y 1997.  S h e h a d a right n e p h r e c t o m y in S e p t e m b e r 1997 a n d left renal tumour  resection in F e b r u a r y 1998.  S h e c o m p l e t e d a n initial round of c h e m o t h e r a p y in M a y  1998, but w a s d i a g n o s e d with a recurrent tumour in A u g u s t 1 9 9 8 .  ICE chemotherapy  w a s b e g u n in S e p t e m b e r 1998. S h e received six c o u r s e s of I C E , the last in F e b r u a r y  103  1999.  B l o o d s a m p l e s w e r e obtained during c y c l e 3 ( d o s e s 1 a n d 2), c y c l e 4 (dose 1  only), a n d c y c l e 5 ( d o s e 2 only). Table 6.2.  Physical characteristics of patients evaluated in the clinical study. Patient 1  Patient 2  sex  male  female  weight  70 kg  17 kg  1.93 m  body surface area  0.70 m  2  osteogenic sarcoma  diagnosis  6.2.3. Graphite furnace atomic absorbance  2  bilateral renal Wilms tumour  spectroscopy  F r e e platinum concentrations in p l a s m a ultrafiltrate w e r e d e t e r m i n e d with a S p e c t r A A 3 0 0 (Varian C a n a d a , R i c h m o n d , B C , C a n a d a ) e m p l o y i n g a graphite tube a t o m i z e r a n d a u t o m a t e d s a m p l e introduction.  T h e lamp current w a s 10 m A , the slit  width 0.2 n m , a n d the signal integration time 1.0 s.  Aliquots (20 u.L) of p l a s m a  ultrafiltrate w e r e introduced at 90 °C, a s h e d at 1100 °C, then a t o m i z e d at 2 8 0 0 °C (Figure 6.1). T h e a r g o n g a s flow w a s 3 L/min, e x c e p t during atomization during w h i c h the flow w a s s t o p p e d a n d the Z e e m a n - c o r r e c t e d a b s o r b a n c e of the 2 6 5 . 9 nm Pt line r e a d . A calibration c u r v e w a s prepared following injection of s t a n d a r d s containing 2 0 0 , 4 0 0 , 6 0 0 , a n d 8 0 0 n g / m L platinum. P l a s m a ultrafiltrate s a m p l e s w e r e read in duplicate. S a m p l e s containing g r e a t e r | h a n 800 n g / m L platinum w e r e diluted with distilled water a n d reinjected s o a s to produce a b s o r b a n c e v a l u e s within the calibration c u r v e range. 3000  o CD  3 CO  2500 2000 1500  i  CD CL  1000  CD  500  E  1 - drying 2 - ashing 3 - atomization  0 10  20  30  40  50  60  Time (s)  Figure 6.1.  Temperature program used for the analysis of platinum in plasma ultrafiltrate.  104  6.2.4. Sample analysis Following collection of patient blood s a m p l e s a s d e s c r i b e d in s e c t i o n 6.2.1, p l a s m a ultrafiltrate w a s prepared ( s e e section 2.2.2) with the ultracentrifuge maintained at a temperature of 15 °C to r e d u c e the potential for carboplatin d e g r a d a t i o n by nucleophilic s p e c i e s present in solution. S i n c e the binding of carboplatin to D N A , R N A , and  protein is essentially irreversible in nature, no c h a n g e s in the b o u n d / u n b o u n d  carboplatin  ratio  are expected even  after  prolonged  ultrafiltration.  The plasma  ultrafiltrate s a m p l e s w e r e then a n a l y z e d for carboplatin b y the H P L C - U V  method  ( d e s c r i b e d in section 5.2.2) a n d for platinum by the graphite furnace A A method ( d e s c r i b e d in s e c t i o n 6.2.3 a b o v e ) . F o r c o m p a r i s o n s , platinum c o n c e n t r a t i o n s from the A A m e t h o d w e r e converted to carboplatin equivalents by multiplying by the ratio of the MW  of carboplatin v e r s u s the M W of the platinum atom.  T h e p l a s m a ultrafiltrate  concentration d a t a obtained from the two m e t h o d s is s u m m a r i z e d in T a b l e 6 . 3 .  6.2.5. Pharmacokinetic evaluation F o r e a c h d o s e obtained, concentration-time d a t a for both free carboplatin a n d free platinum w e r e u s e d to construct elimination profiles. T h e s e profiles w e r e fit using n o n l i n e a r least s q u a r e s regression a n a l y s i s with the standard c o m p a r t m e n t a l  i.v.-  infusion m o d e l s present in V e r s i o n 1.5 of W i n N o n l i n (Scientific C o n s u l t i n g Inc., A p e x , N C , U S A ) . E v a l u a t i o n of the goodness-of-fit of a particular m o d e l w a s predominantly b a s e d o n e x a m i n a t i o n of A k a i k e ' s information criterion [10], the pattern of r e s i d u a l s (the difference b e t w e e n o b s e r v e d concentrations a n d concentrations predicted from the m o d e l fit), a n d the residual s u m of s q u a r e s v a l u e s .  Compartmental  e s t i m a t e s w e r e derived from both the H P L C - U V a n d A A d a t a .  parameter  In addition to A U C  estimation from the fitted d a t a (parametric estimation), A U C v a l u e s w e r e obtained by n o n c o m p a r t m e n t a l m e t h o d s , specifically application of the linear t r a p e z o i d a l rule with extrapolation to infinity using terminal elimination rate constant e s t i m a t e s . C o m p a r i s o n s b e t w e e n m e a n A U C v a l u e s w e r e m a d e using the statistical p a c k a g e p r e s e n t in Microsoft E x c e l (p < 0.05 c o n s i d e r e d statistically significant).  T h e e q u a t i o n s u s e d for  calculation of A U C , C L B , a n d V s s v a l u e s a r e s h o w n o n p a g e 1 0 7 . T  105  o _i  00  o  Q. X  CM CO  o  CM  o _J a. x  CO CD CL  CD CM  O  CM  iri  ^  m in  CN  oo  00  CD  CD CO  T —  O  O  m  co  CO CD CL  O  O 1=  CD CL  c  CD  CD CO  10  O  Q , r^-~ •  CD 0 0  oi  CO  T —  CM 0 0  -a-  JD  o >s O  o  CO CD CO  c-  CD CO  a  o  o  in  <  CM  ^  CM C D i n  "tf"  CD  JD  in o m  CD  ob  co T— CD r- o CD CM CM 1— O  H  CN]  O _l  to in s  CO  a.  00  oc> 0 0  m CD  CM  o -  CM  CM CO T — CD o lO T o ~  CO CD CL  CO CO CO  CO CO CM  t  H  O —I  0_ X  CD  o  CD CJ> t -  w co s  in CM  oo CM  CD  CM  00  O  CD CD  CD  CD O  in 00  T—  iri C O C M  CO  T—  X—  v -  CD CM i n CM CD O CD r-CD CM  T—  in o  3  J D: <  o >^ O  E  o" CM  o  T —  in  m  CM  d  co  O  JZ c.  3  o  CD  o Q  CD 00  CM  0  o >. O in  tN t o  in s  ^~  CM  CO  \—  in  CO  CD  to CD  o  E E  E  co  H  JC  CO CD CL  to  ho^  to CD  E o  CD  CO  O  o CO  a> E  O  o  -4—'  in  cb  c g  co 3  CO  o >. O  E  a  t--  in  CD  O) zi.  I— C D  c  CD  O a.  CM CD „ O ^ T -  00  ^  . o co  (D  in  N CO CM C D O CM  CM  i -  co  CD CL  o _l  co  CD CL  CL X  O  T -  CM  CM  CD  CD CO  O  a ,  m m  m" *  T—  00  a> o >< O  CM  CO  00  00 -4-  — rC O  1^  iri  CO  00  in  7— 00 00 • X— C D CO CD 00  CM  T  in CM CM  T—  -  N-  CD  CO  CD  O  i_ CO  co  to  CD CL  c o  O  -t—>  c  CO I CD O  O  CO CM  a ,  co o  ^ N n 6 6 6 - i -  co" '  0 0 CO CD CO in 00 o CD T— CD CM CO c d x X CO CM r->: o o -  Q)  c o O  -  O >»  £r  o  CD  m co  E  ro  N  E  CD CM • c - CM  CO  CD • CM  co \ t  i n CD N  CB  H  E E  CM CM  to to ^ T - c o c M C N o o c n c n c o m  r  L  c b c N d  (  N ^  c  !  c  9 '  n  «  co co CD CD CL  in  0. X  CL  y T - T - T - C O l f C N ' - O O CD to o  oo C O C O C O i n  co  CD  CO  CD  CD  CD  CM  x  x  CD  T—  0 0 CD  CM  in CD c\i  cb  CM  00  x  -  CM  x  -  -  -  CD CL  o  CD  3 CO CD  "co o  CD  Q , in • _CD o >. O  c  CM  CM CM CD 00 in T — T— 0 0  CD CD 0 0 T— CD CD CO CM iri  to o  a r-  o  co  o CM  CD  E  H  in  co i  CD  T  o  1- CM  oo" o o >» O  r-~ co oo  ro  CO  C 0)  <  CM  o  in  t  in  CD  s  o  CM CM  CD  o Si  CO  106  Noncompartmental  AUC calculations  (linear trapezoidal  rule)  (Cn + C n - 1 ) ( t n - t n - 1 ) A U C (O-oo) = A U C (0-t) + A U C (t-oo)  =  I  +  2 T h e terminal  elimination  constant (Xz)  Clast Xz  w a s estimated  by performing  a  linear  r e g r e s s i o n of the natural logarithm of o b s e r v e d concentrations v e r s u s s a m p l i n g times, r e p e a t e d using the last three s a m p l i n g points, last four s a m p l i n g points, etc. until a n optimal r v a l u e w a s obtained. 2  Parametric  calculations Two-Compartment Model  One-Compartment Model infusion  k2l  infusion  K10  M2 K10!  F o r both m o d e l s , primary estimated p a r a m e t e r s include k-m a n d the  apparent  v o l u m e of distribution (V). F o r the two-compartment m o d e l , A a n d B c o r r e s p o n d to z e r o time intercepts a n d a r e d e p e n d e n t o n the distribution a n d elimination rate c o n s t a n t s a a n d p, a s well a s the m i c r o c o n s t a n t s k i , k , a n d k i . 0  12  T h e s e p a r a m e t e r s a r e then  2  u s e d collectively to c a l c u l a t e the following: A U C , A U M C , C L B , M R T , a n d V s s . T  Parametric  AUC dose / V  A U C (O-oo)  A U C (O-oo)  =  kio  B  a  P  (2-compartment model)  (1-compartment model)  Total body clearance  A =  (CLTB)  C L T B = d o s e / A U C (O-oo)  Volume Vss  of distribution =  CLJB  *  at steady-state M R T  where  (Vss) M R T  =  [(AUMC / A U C )  - (infusion time / 2)]  107  6.3.  Results and D i s c u s s i o n  6.3.1. Visual examination of elimination profiles F i g u r e s 6.2 (patient 1) a n d 6.3 (patient 2) s h o w a g r a p h i c a l c o m p a r i s o n of the elimination  profiles of free platinum  and free carboplatin.  In both  patients,  elimination of free carboplatin a n d free platinum a p p e a r similar o v e r the first 8 h.  the Two  p h a s e s of elimination are apparent, a distribution p h a s e occurring primarily within the first few hours post-infusion a n d a s u b s e q u e n t elimination p h a s e occurring thereafter. H o w e v e r , c o m p a r t m e n t a l a n a l y s i s m a y result in the grouping of t h e s e p h a s e s into a s i n g l e compartment, s i n c e for s o m e d o s e s there are likely insufficient s a m p l e s taken immediately  post-infusion or the rate constants are too similar to distinguish  the  c o m p a r t m e n t s mathematically. After 8 h, the rapid elimination of free carboplatin c o n t i n u e s , with carboplatin c o n c e n t r a t i o n s quickly falling b e l o w the limit of quantitation of the H P L C a s s a y .  In  contrast to free carboplatin, the p h a r m a c o k i n e t i c behaviour of free platinum up to 24 h post-infusion a p p e a r s to be triphasic, with a third c o m p a r t m e n t b e t w e e n 8 a n d 24 h post-infusion.  occurring  roughly  F o r the eight d o s e s of c h e m o t h e r a p y given, a  c o m b i n e d total of 14 s a m p l e s w e r e obtained more than 12 h post-infusion.  Of these,  only three c o n t a i n e d carboplatin concentrations sufficiently large for quantitation v i a the HPLC-UV  a s s a y method.  B y c o m p a r i s o n , all 14 s a m p l e s c o n t a i n e d m e a s u r a b l e  a m o u n t s of free platinum, quantitation by A A being p o s s i b l e e v e n at 2 4 h post-infusion. In s u m m a r y , the elimination profiles for free platinum a n d free carboplatin are clearly different after 8 h post-infusion. T h i s c o n c l u s i o n is supported not only by visual e x a m i n a t i o n of the profiles t h e m s e l v e s , but by careful e x a m i n a t i o n of residual plots of the p e r c e n t a g e difference in concentration v a l u e s predicted from the A A a s s a y v e r s u s the H P L C - U V a s s a y (Figure 6.4).  108  Patient 1 Cycle 5, Dose 2 (300 mg/m ; 580 mg)  Cycle 5, Dose 1 (300 mg/m ; 580 mg)  2  2  10  10 -J  cn  free platinum free carboplatin  c  o o 5 1  1  -*—'  c CD  o c o o  0.1  0.1  HPLC-UV assay LOQ —  6  12  18  r  6  24  1  1  12  18  24  Time (h)  Time (h)  Cycle 7, Dose 2 (700 mg/m ; 1320 mg)  Cycle 7, Dose 1 (700 mg/m ; 1320 mg)  2  2  10 -4  10  C O 1 c CD O c  o o  0.1  0.1 —  12  18  r 24  — r ~  6  Time (h)  Figure 6.2.  Post-infusion elimination profiles for patient 1.  12 Time (h)  18  24  Patient 2  0  6  12  18  24  0  Time (h)  Figure 6.3.  Post-infusion elimination profiles for patient 2.  5  10  15  Time (h)  20  25  300  -i  0  2  4  6  8  10  Time (h)  Figure 6.4.  Plots showing residual differences between predicted concentrations from the A A assay method versus the H P L C - U V assay method.  If the methods provide  equivalent results, then the residuals should be randomly distributed about zero.  111  6.3.2. Parametric modeling and parameter estimation T h e H P L C data from patients 1 a n d 2 w e r e best d e s c r i b e d by o n e a n d twocompartment  m o d e l s , respectively.  F o r the A A data, a two-compartment  model  provided a better overall fit to the entire data set, w h i c h m a y be d u e to a lack of sufficient  time-points  past  8 h post-infusion  to  adequately  characterize a  three-  compartment model. T o e n s u r e that the biexponential m o d e l did not "misfit" the data by ignoring the late time-points a n d placing two p h a s e s within the first 8 h, weighting (1/y ) 2  of the later points w a s n e c e s s a r y (Figure 6.5).  Predicted fits from the H P L C a n d A A  data for e a c h of the d o s e s e v a l u a t e d are s h o w n in the A p p e n d i x . •  Observed  — Predicted  Time (h)  0.01  +  1  0  5  1  1  1  1  10  15  20  25  1 30  Time (h)  Figure 6.5. Examples of unweighted (top panel) and weighted (bottom panel) twocompartment fits to the A A data from patient 2. The data is from cycle 5, dose 2.  112  M e a n p h a r m a c o k i n e t i c parameter estimates derived from the H P L C a n d A A data are s h o w n in T a b l e 6.4. F o r patient 1, C m a x v a l u e s determined from the A A data w e r e m u c h larger than t h o s e determined from the H P L C data.  T h i s effect w a s e s p e c i a l l y  p r o n o u n c e d at the larger (700 m g / m ) d o s e s a s c o m p a r e d to the lower (300 2  doses. mg/m  2  mg/m ) 2  F o r the H P L C data, derived parameter estimates of C L a n d V s s for the 7 0 0 d o s e w e r e outside the range of literature v a l u e s ( C L : 4 4 - 2 2 5 m L / m i n / 1 . 7 3 m ) , 2  V s s : 17-30 L) previously reported for t h e s e parameters [3,11]. W h i l e the d i s c r e p a n c y b e t w e e n C m a x v a l u e s determined from H P L C a n d A A data could a c c o u n t for the o b s e r v e d V s s a n d C L v a l u e s , no o b v i o u s explanation a s to the magnitude of t h e s e C m a x differences is apparent. Table 6.4.  Mean parameter estimates derived from the  Parameter  HPLC  and  AA  HPLC  AA  Patient 1 (300 mg/m ) (n = 2 doses) Cmax (u.g/mL) t (min) V s s (L/1.73 m ) C L (mL/min/1.73 m )  19  25  80 22 178  73 24 130  Patient 1 (700 mg/m ) (n = 2 doses) Cmax (u.g/mL)  18  43  102 55 323  56 47 186  39  37  41 1.8 14 142  61 7.7 25 136  data.  2  1/2  2  2  T B  2  ti/ (min) V s s (L/1.73 m ) C L T B (mL/min/1.73 m ) 2  2  2  Patient 2 (400 mg/m ) (n = 4 doses) Cmax (ng/mL) 2  t alpha (min) t beta (h) V s s (L/1.73 m ) C L (mL/min/1.73 m ) 1/2  1/2  2  2  T B  F o r patient 2, C m a x estimates derived from both g r o u p s of d a t a w e r e similar. T h e greater V s s o b s e r v e d for free platinum c o m p a r e d to free carboplatin h a s b e e n previously noted in the literature, the difference p e r h a p s c o r r e s p o n d i n g to carboplatin distribution with the extracellular fluid v e r s u s platinum distribution throughout total b o d y w a t e r [12].  A s previously m e n t i o n e d , the greater elimination half-lives of free platinum  113  are o b v i o u s from v i s u a l examination of the elimination profiles.  F o r this  patient,  h o w e v e r , this p r o l o n g e d elimination p h a s e of free platinum resulted in l e s s than a 5 % i n c r e a s e in the m e a n parametric A U C value.  6.3.3. A s s a y and modeling effects on A U C values AUC estimation  by parametric  versus trapezoidal  methods  T a b l e 6.5 provides the parametric and trapezoidal A U C e s t i m a t e s derived from the free platinum a n d free carboplatin data.  U p o n only brief e x a m i n a t i o n of this d a t a ,  the similarity b e t w e e n t h e s e m e t h o d s of A U C calculation in this study is apparent. Indeed, paired s a m p l e t-tests s h o w e d that A U C v a l u e s derived from the two m e t h o d s w e r e not significantly different.  F o r the H P L C data, the m e a n parametric A U C v a l u e  w a s only 1.5% greater than its trapezoidal counterpart.  F o r the A A d a t a , the m e a n  t r a p e z o i d a l estimate w a s 3 . 5 % greater than its parametric counterpart.  T h i s difference  fell to 2 . 0 % after e x c l u s i o n of the patient 2, c y c l e 4, d o s e 1 d a t a , for w h i c h the t r a p e z o i d a l A U C estimate w a s likely high d u e to the a b s e n c e of a s a m p l e point b e t w e e n 8 a n d 17 h post-infusion. Table 6.5.  Parametric and trapezoidal A U C estimates (u.g/mL * h). HPLC data trapezoidal parametric  AA data trapezoidal parametric  Patient 1 cycle 5, dose 1  53.0  51.9  67.7  65.7  cycle 5, dose 2  44.9  46.0  65.9  66.2  cycle 7, dose 1  56.1  57.4  112  109  cycle 7, dose 2  66.8  66.0  101  127  cycle 3, dose 1  78.9  76.7  79.2  75.7  cycle 3, dose 2  72.2  68.0  82.9  77.9  cycle 4, dose 1  82.3  86.5  88.6  101  cycle 5, dose 2  78.6  72.2  75.0  69.2  Patient 2  114  AUC  estimates As  from HPLC versus AA data  d i s c u s s e d previously, the elimination  profiles of free platinum  a n d free  carboplatin are generally c o n s i d e r e d to be equivalent for at least 8 h following  i.v.  administration of this drug. C u r s o r y examination of the profiles s h o w n in section 6.3.1 for the two patients in this study supported this h y p o t h e s i s , a s little difference w a s o b s e r v e d visually b e t w e e n the elimination of free platinum a n d free carboplatin o v e r this time interval. species.  After 8 h, there w a s a n o b v i o u s d i v e r g e n c e in the profiles of t h e s e two  H o w e v e r , the effect of this d i v e r g e n c e on the magnitude of clinically relevant  p a r a m e t e r s s u c h a s A U C is uncertain d u e to the s m a l l concentration levels present during this time p e r i o d .  T h u s , direct c o m p a r i s o n of A U C v a l u e s c a l c u l a t e d from free  platinum a n d free carboplatin m e a s u r e m e n t s is required. In patient 1, A U C v a l u e s derived from the A A data v e r s u s the H P L C d a t a w e r e 28-100%  greater  for  e s t i m a t e s , respectively.  parametric  estimates  and  27-92%  greater  for  trapezoidal  E x a m i n a t i o n of the analytical d a t a set in Figure 6.2 s u g g e s t s  that the major c a u s e for this difference is the larger concentrations of free platinum a s c o m p a r e d to free carboplatin immediately following the drug infusion period. W h i l e their d o e s exist the possibility of effects d u e to imperfect correlations b e t w e e n concentrations d e t e r m i n e d from the two a s s a y m e t h o d s , the c l o s e n e s s of the m e t h o d s for s a m p l e s in the 2-6 h post-infusion time-period s u g g e s t s that a s s a y differences are not likely the c a u s e of this d i s c r e p a n c y .  A more likely explanation is that a rapid c o n v e r s i o n of a  large a m o u n t of carboplatin to its nucleophilic substitution products o c c u r r e d during a n d immediately following d o s i n g in patient 1.  Interestingly, the differences b e t w e e n free  platinum a n d free carboplatin concentrations in patient 1 w e r e e v e n more p r o n o u n c e d w h e n a greater d o s e w a s administered. O b v i o u s l y , the u s e of free carboplatin v e r s u s free platinum evaluation could have a great impact o n d o s i n g d e c i s i o n s m a d e for this patient if A U C is u s e d a s a clinical endpoint.  Further study is definitely n e e d e d to  e v a l u a t e if the differences o b s e r v e d for this patient are merely a n aberration or are typical within this patient population. In patient 2, A U C v a l u e s derived from A A v e r s u s H P L C d a t a w e r e - 4 . 5 % to 1 5 % ( m e a n : +4.5%) greater for parametric e s t i m a t e s a n d - 4 . 2 % to 1 7 % ( m e a n : +6.5%)  115  greater for trapezoidal estimates. With this patient, C m a x v a l u e s w e r e similar for both free carboplatin a n d free platinum data. T h u s , the small i n c r e a s e in A U C o b s e r v e d for free platinum in this patient c a n be attributed to the additional contributions o b s e r v e d after 8 h w h e n the elimination profiles diverge. T h e large A U C differences in patient 1 m a y therefore be more the result of the o b s e r v e d C m a x differences than any additional contribution to A U C from i n c r e a s e d free platinum levels b e t w e e n 8 a n d 2 4 h postinfusion.  6.3.4. Relevance of observed A U C differences T a b l e 6.6 s u m m a r i z e s s o m e recent p a p e r s c o n c e r n i n g d o s e optimization  of  carboplatin in adult patients. W h i l e r e s e a r c h p a p e r s dealing with d e v e l o p m e n t of d o s i n g f o r m u l a s in both adults a n d pediatrics relied almost e x c l u s i v e l y o n m e a s u r e m e n t s of free platinum, t h e s e more recent p a p e r s h a v e u s e d a variety of monitoring s c h e m e s involving both free platinum a n d free carboplatin. P a r a m e t e r e s t i m a t e s b a s e d on t h e s e different s p e c i e s continue to b e u s e d interchangeably, a n d c o m p a r i s o n s of the relative p e r f o r m a n c e of dosing formulas a n d optimization strategies h a v e m a d e no attempt to c h a r a c t e r i z e the contribution of variability in study d e s i g n to the o b s e r v e d differences. F o r e x a m p l e , the p e r f o r m a n c e of the C a l v e r t formula for A U C prediction to A U C v a l u e s c a l c u l a t e d v i a the limited s a m p l i n g method of S o r e n s e n et al. [12] w e r e e v a l u a t e d in a s u b s e q u e n t study by V a n W a r m e r d a m et al. [13], but the latter authors substituted free platinum m e a s u r e m e n t s for free carboplatin. In light of the differences in C m a x v a l u e s a n d resulting A U C e s t i m a t e s o b s e r v e d for o n e of the two patients e v a l u a t e d in this study, this practice of using free platinum a n d free carboplatin v a l u e s interchangeably is not a d v i s e a b l e .  116  Table 6.6.  Recent studies describing dose-individualization strategies for carboplatin.  Ref.  Assay of  Study  12  free carboplatin  Developed a limited sampling strategy based on optimal (one or two) time points for prediction of A U C values.  free platinum  13  Compared interpatient variability in A U C values determined by Calvert formula versus variability in values determined by a limited sampling strategy [12].  free platinum  14  Compared a Bayesian dose-individualization method for A U C estimation versus conventional renal function and body surface area methods.  free carboplatin  15  Developed a sequential Bayesian algorithm for prediction of A U C values and compared predictions to those from both the Calvert formula and a regular Bayesian method.  6.3.5. Free carboplatin versus free platinum in dose-adjustment strategies T h e first d o s e - a d j u s t m e n t strategies, d e v e l o p e d by C a l v e r t et al. [4] a n d E g o r i n et al.  [16],  relied  on  pharmacokinetics.  free  platinum  measurements  to  characterize  carboplatin  T h i s w a s d o n e primarily out of necessity, a s early H P L C  assay  m e t h o d s did not h a v e the required sensitivity to adequately c h a r a c t e r i z e free carboplatin elimination.  A s e v i d e n c e d from the p a p e r s listed in T a b l e 6.6, there is a n i n c r e a s i n g  trend in recent studies towards the u s e of free carboplatin m e a s u r e m e n t s in p l a c e of free  platinum.  T h i s trend  may  reflect  chromatographic  a d v a n c e s resulting in m o r e sensitive H P L C methods. Bayesian feedback  approaches now  being  and  sample  preparation  A l s o , the limited s a m p l i n g a n d  d e v e l o p e d e n a b l e derivation  p h a r m a c o k i n e t i c estimates b a s e d on fewer s a m p l e points.  of  good  U s u a l l y , the s a m p l e points  u s e d are r e a s o n a b l y c l o s e to the time of drug administration; h e n c e , a s s a y m e t h o d s e m p l o y e d n e e d not be a s sensitive. O f c o u r s e , initial evaluation of the p e r f o r m a n c e of t h e s e novel m e t h o d s is b a s e d on more complete (and thereby a s s u m e d accurate) p h a r m a c o k i n e t i c p a r a m e t e r e s t i m a t e s . T h i s h a s resulted in the potential pitfall of using  117  2 4 h free platinum data a s the "true" v a l u e to w h i c h dosing formulas b a s e d on free carboplatin m e a s u r e m e n t s are c o m p a r e d .  S u c h a n a p p r o a c h is only rational if free  carboplatin a n d free platinum m e a s u r e m e n t s are interchangeable. T h e d e s i r e of s o m e r e s e a r c h e r s to u s e free carboplatin instead of free platinum c o n c e n t r a t i o n s is b a s e d primarily on the logical principle that monitoring of parent drug s h o u l d provide a better indicator of remaining p h a r m a c o l o g i c a l activity than  does  monitoring of a mixture of parent drug a n d its metabolites. This m a y not be the c a s e for carboplatin, h o w e v e r , s i n c e the available p h a r m a c o k i n e t i c e v i d e n c e (predominantly the greater V s s v a l u e s for free platinum c o m p a r e d to free carboplatin) s e e m s to s u g g e s t that carboplatin is essentially a pro-drug, with its platinum-containing products  degradation  r e s p o n s i b l e for entry into cells a n d s u b s e q u e n t D N A - a l k y l a t i n g  activity.  Furthermore, a s free platinum m e a s u r e m e n t s are the b a s i s of the d o s i n g formulas currently u s e d in m o s t clinical settings, c o m p a r a t i v e evaluation of novel d o s i n g s c h e m e s is m o r e e a s i l y m a d e using free platinum m e a s u r e m e n t s .  W h e r e a s nearly all free  platinum p h a r m a c o k i n e t i c studies h a v e u s e d a 24 h m e a s u r e m e n t period, no consistent time period h a s b e e n u s e d for determination of free carboplatin. With r e s p e c t to equipment availability, the p r e s e n c e of H P L C e q u i p m e n t in both clinical a n d a c a d e m i c laboratories h a s b e c o m e s o m e w h a t c o m m o n p l a c e . Duffull et al. [16] indicated this to be their primary motivation for using H P L C m e a s u r e m e n t of free carboplatin in lieu of that of free platinum.  H o w e v e r , for laboratories p o s s e s s i n g A A  e q u i p m e n t , free platinum m e a s u r e m e n t s are more convenient, not only a s a result of the faster a n a l y s i s times a n d limited s a m p l e handling required by this t e c h n i q u e , but a l s o d u e to the  inherent stability  of nonvolatile  platinum  s p e c i e s v e r s u s that  of  carboplatin itself, w h i c h is subject to nucleophilic degradation. A  number  of  studies will h a v e to  be  performed  in order  to  assess  the  a p p r o p r i a t e n e s s of free platinum v e r s u s free carboplatin m e a s u r e m e n t s . First of all, a n e x p a n d e d study involving a larger patient group n e e d s to be c o n d u c t e d to determine w h e t h e r the differences o b s e r v e d in our study are o b s e r v e d a c r o s s a wider patient population.  If significant differences b e t w e e n p h a r m a c o k i n e t i c p a r a m e t e r s b a s e d o n  free platinum/carboplatin do in fact exist, then c o m p a r a t i v e studies of the relationships  118  b e t w e e n free platinum/carboplatin pharmacokinetic parameters a n d p h a r m a c o l o g i c a l effect (clinical o u t c o m e s ) or toxicity are n e c e s s a r y .  Unfortunately, s u c h studies are  c o m p l i c a t e d by a myriad of factors, including heterogeneity in patient populations a n d institutional  protocols.  F o r free platinum m e a s u r e m e n t s , only a few s t u d i e s have  m a n a g e d to link specific A U C v a l u e s with treatment s u c c e s s e s or failures [17]. potential  The  of free carboplatin to provide clearer p h a r m a c o k i n e t i c - p h a r m a c o d y n a m i c  relationships r e m a i n s largely unexplored; thus, further r e s e a r c h relying on H P L C a s s a y m e t h o d s is justified.  6.4.  C o n c l u s i o n s and recommendations In the two patients e v a l u a t e d , differences in the elimination profiles of free  platinum a n d free carboplatin w e r e o b s e r v e d within 8 h post-administration.  W h i l e no  statistically significant differences w e r e noted b e t w e e n parametric a n d trapezoidal m e t h o d s of A U C estimation, A U C v a l u e s w e r e larger w h e n d e t e r m i n e d from free platinum a s o p p o s e d to free carboplatin data. w e r e larger in o n e patient.  Interestingly, the o b s e r v e d differences  In this patient, end-of-infusion  ( C m a x ) free  platinum  concentrations w e r e also larger than the c o r r e s p o n d i n g free carboplatin concentrations. D u e to the d i s c r e p a n c y in free carboplatin/free platinum A U C differences o b s e r v e d for the two patients e v a l u a t e d , a further study in a larger patient population is warranted. Currently, w e r e c o m m e n d that free platinum a n d free carboplatin d a t a not be u s e d interchangeably.  F o r laboratories p o s s e s s i n g A A equipment, m e a s u r e m e n t of free  platinum levels is more convenient a n d provides d a t a w h i c h are m o r e readily c o m p a r e d to m u c h of the existing literature. studies  establishing  clear  H o w e v e r , it must be r e c o g n i z e d that the lack of  relationships  between  carboplatin  pharmacokinetic  p a r a m e t e r s s u c h a s A U C a n d clinical o u t c o m e s m a y be, in part, d u e to the practice of using free platinum d a t a in p l a c e of free carboplatin.  Theoretically, it is better to  e v a l u a t e a single s p e c i e s , s u c h a s free carboplatin, w h e n evaluating p h a r m a c o k i n e t i c a n d p h a r m a c o d y n a m i c relationships. evaluate  the  carboplatin  clinical  W e r e c o m m e n d that studies be c o n d u c t e d to  pharmacokinetic  a n d , if indicated  by the  and  pharmacodynamic  behaviour  p h a r m a c o d y n a m i c relationships, that  of  free  dosing  f o r m u l a s be d e v e l o p e d b a s e d exclusively on m e a s u r e m e n t of free carboplatin.  119  6.5.  References  1.  A.H. Calvert, S.J. Harland, D.R. Newell, Z.H. Siddik, A . C . Jones, T.J. McElwain, K.S. Raju, E. Wiltshaw, I.E. Smith, J.M. Baker, M.J. Peckham, and K.R. Harrap. clinical studies with c/s-diammine-1,1-cyclobutanedicarboxylate platinum II.  Early Cancer  Chemother Pharmacol 9: 140-147 (1982). 2.  B.D. Evans, K.S. Raju, A.H. Calvert, S.J. Harland, and E. Wiltshaw. Phase II study of J M 8 , a new platinum analog in ovarian cancer.  Cancer Treatment Rep 67: 997-1000  (1983). 3.  S.B. Duffull and B.A. Robinson. carboplatin. Clin Pharmacokinet  4.  A.H.Calvert.  Clinical pharmacokinetics and dose optimisation of  33: 161-183 (1997).  Dose optimization of carboplatin in adults. Anticancer Res 14: 2273-2278  (1994). 5.  A.H. Calvert, D.R. Newell, and M.E. Gore.  Future directions with carboplatin: can  therapeutic monitoring, high-dose administration, and hematological support with growth factors expand the spectrum compared with cisplatin? Semin Oncol 19 (1 Suppl 2): 155163 (1992). 6.  R. Riccardi, A. Riccardi, A. Lasorella, C. Di Rocco, G. Carelli, A. Tornesello, T. Servidei, A. lavarone, and R. Mastrangelo. Clinical pharmacokinetics of carboplatin in children. Cancer Chemother Pharmacol 33: 477-483 (1994).  7.  F. Doz, L. Brugieres, G. Bastian, E. Quintana, J . Lemerle, and J.-M. Zucker. Clinical trial and pharmacokinetics of carboplatin 560 m g / m in children. Med Ped Oncol 18: 4592  465(1990). 8.  T. Madden T, M. Sunderland, V.M. Santana, and J.H. Rodman. The pharmaco-kinetics of high-dose carboplatin in pediatric patients with cancer. Clin Pharmacol  Ther 51: 701-  707 (1992). 9.  N.M. Marina, J.H. Rodman, D.J. Murry DJ, S.J. Shema, L.C. Bowman, D.P. Jones, W. Furman, W.H. Meyer, and C B . Pratt.  Phase I study of escalating targeted doses of  carboplatin combined with ifosfamide and etoposide in treatment of newly diagnosed pediatric solid tumors. J Nat Cancer Inst 86: 544-548 (1994). 10.  K. Yamaoka, T. Nakagawa, and T. Uno.  Application of Akaike's information criterion  (AIC) in the evaluation of linear pharmacokinetic equations. J Pharmacokinet  Biopharm  6(2): 165-175 (1978).  120  11.  B. Peng, A.V. Boddy, M.W. English, A.D. Pearson, L. Price, M.J. Tilby, and D.R. Newell.  The comparative pharmacokinetics and pharmacodynamics of cisplatin and  carboplatin in paediatric patients: a review. Anticancer Res 14: 2270-2284 (1994). 12.  B.T. Sorensen, A. Stromgren, P. Jakobsen, and A. Jakobsen.  A limited sampling  strategy for estimation of the carboplatin area under the curve.  Cancer  Chemother  Pharmacol 31: 324-327 (1993). 13.  L.J. Van Warmerdam, S. Rodenhuis, O. Van Tellingen, R.A. Maes, and J.H. Beijnen. Validation of a limited sampling model for carboplatin in a high-dose chemotherapy combination. Cancer Chemother Pharmacol 35: 179-181 (1994).  14.  B. Peng, A.V. Boddy, M. Cole, A.D. Pearson, E. Chatelut, H. Rubie, and D.R. Newell. comparison of methods for the estimation of carboplatin pharmacokinetics in paediatric cancer patients. Eur J Cancer 31 A( 11): 1804-1810 (1995).  15.  S.B. Duffull, E.J. Begg, B.A. Robinson, and J . J . Deely.  A sequential Bayesian  algorithm for dose individualisation of carboplatin. Cancer Chemother Pharmacol 39: 317326 (1997). 16.  M.J. Egorin, D.A. Van Echo, E.A. Olman, M.Y. Whitacre, A. Forrest, and J . Aisner. Prospective  validation  of  diamminedichloroplatinum  a II  pharmacogically analogue  based dosing scheme for  the  cis-  diamminecyclobutanedicarboxylatoplatinum.  Cancer Res 45: 6502-6506 (1985). 17.  M.L. De Lemos.  Application of the area under the curve of carboplatin in predicting  toxicity and efficacy. Cancer Treatment Rev 24: 407-414 (1998).  121  CHAPTER 7 GLOBAL  SUMMARY  P r e v i o u s literature reports evaluating carboplatin p h a r m a c o k i n e t i c s in both adult a n d h u m a n patients relied mainly upon m e a s u r e m e n t of free platinum levels in p l a c e of free carboplatin.  T h i s w a s d o n e out of necessity, a s H P L C a s s a y s for carboplatin  g e n e r a l l y l a c k e d the required sensitivity d u e to the rapid p l a s m a elimination d i s p l a y e d by this c o m p o u n d . W h i l e t h e s e H P L C a s s a y s typically p o s s e s s e d detection limits of 0.5 (ag/mL or poorer, a limit of quantitation in the lower n g / m L range is required in order to follow carboplatin for m o r e than 12 h post-administration. T h i s h a d b e e n d e m o n s t r a t e d in s t u d i e s by Elferink et al. [1] a n d A l l s o p p et al. [2], the former investigators utilizing a n H P L C a s s a y with differential p u l s e polarography, the latter investigators using a c o l u m n switching H P L C a s s a y .  O u r g o a l w a s to d e v e l o p a n d validate a n H P L C a s s a y with  similar sensitivity to t h e s e m e t h o d s (limit of quantitation: 20 n g / m L ) , but with H P L C e q u i p m e n t that w o u l d b e readily available in m o s t laboratories (i.e. e m p l o y i n g a U V detector a n d no column-switching technology).  With this n e w H P L C method a n d a n  existing A A m e t h o d , the e q u i v a l e n c y of free carboplatin a n d free platinum elimination c o u l d be e x a m i n e d by pharmacokinetic evaluation of clinical s a m p l e s obtained from pediatric patients receiving carboplatin. A s d i s c u s s e d in C h a p t e r 2, the polar structure of carboplatin c o m b i n e d with its lack of a significant c h r o m o p h o r e m a k e d e v e l o p m e n t of a sensitive H P L C a s s a y method extremely  challenging.  Our  development  strategy  centered  around  two  ideas:  optimizing the c h r o m a t o g r a p h y through careful selection of the m o s t efficient stationary p h a s e for carboplatin a n a l y s i s a n d employing s o l i d - p h a s e extraction t e c h n i q u e s to r e m o v e the majority of s a m p l e interferences.  Testing of a s e r i e s of O D S c o l u m n s  d e m o n s t r a t e d that the Y M C O D S - A Q column provided effective n u m b e r s of theoretical plates that w e r e nearly three times greater than the next best c o l u m n .  T h i s efficiency  resulted not just from the p h y s i c a l properties of the p a c k i n g material itself (e.g. particle d i a m e t e r a n d pore size) but a l s o from the unique retentivity of the Y M C p a c k i n g material for carboplatin.  Still, a d e q u a t e resolution of carboplatin from s a m p l e c o m p o n e n t s w a s  not a c h i e v e d on the Y M C c o l u m n following direct injection of p l a s m a ultrafiltrate. T h u s ,  122  a  s o l i d - p h a s e extraction  developed  to  provide  procedure,  sufficient  carboplatin on this c o l u m n . provided  employing  sample  amino  clean-up  to  extraction allow  specific  separations.  was  detection  of  In this manner, w e obtained the i n c r e a s e d selectivity  by normal p h a s e separation of polar c o m p o u n d s while  efficiency benefits  cartridges,  maintaining  (not to mention the solvent s a v i n g s ) of r e v e r s e p h a s e  E m p l o y i n g this unique a p p r o a c h , the L O Q of the optimized  the  HPLC  HPLC-UV  m e t h o d w a s e s t i m a t e d to b e 5 0 ng/mL. P o s t - c o l u m n derivatization w a s e m p l o y e d in a n attempt to further improve the sensitivity a n d specificity of the technique. S o d i u m bisulfite w a s utilized a s a derivitizing reagent, allowing for carboplatin detection at a longer analytical w a v e l e n g t h (290 nm). T h i s i m p r o v e d selectivity s h o r t e n e d the required run time a n d eliminated the n e e d for s o l i d - p h a s e extraction prior to p l a s m a ultrafiltrate injection.  W h i l e a slight i n c r e a s e in  analytical sensitivity w a s o b s e r v e d , i n c r e a s e d b a c k g r o u n d a n d variability of r e s p o n s e resulted in a similar estimate of L O Q (50 ng/mL) for this H P L C - P C m e t h o d . Prior  to  method  validation,  identification  of  appropriate  internal  standard  c o m p o u n d s w a s n e e d e d for the H P L C - U V a s s a y to a c c o u n t for variability d u e to the solid-phase  extraction  procedure  and  for  d e g r a d a t i o n of the post-column reagent.  the  HPLC-PC  assay  due  to  S e v e r a l platinum c o m p o u n d s e v a l u a t e d a s  p o s s i b l e internal standard c a n d i d a t e s w e r e too lipophilic, p o s s e s s i n g long t i m e s o n the analytical c o l u m n . EthMal,  and  MethCBDCA  was  possible  retention  S y n t h e s i s of the carboplatin a n a l o g u e s M e t h M a l , readily  a c c o m p l i s h e d from  a q u e o u s solution v i a a two-step p r o c e s s [3], albeit in low yield.  tetrachloroplatinate  in  T h e suitability of the  c o m p o u n d s a s internal s t a n d a r d s w a s then e v a l u a t e d in terms of their  retention  characteristics ( H P L C - U V a n d H P L C - P C methods) a n d their extraction b e h a v i o u r on the a m i n o cartridges ( H P L C - U V method only).  F o r the H P L C - P C a s s a y m e t h o d , M e t h M a l  provided the m o s t appropriate retention, eluting shortly after carboplatin.  W h i l e both  M e t h M a l a n d E t h M a l a n a l o g u e s provided poor extraction r e c o v e r i e s on the  amino  cartridges a n d M e t h M a l co-eluted with e n d o g e n o u s c o m p o n e n t s of p l a s m a ultrafiltrate, this w a s not the c a s e for the M e t h C B D C A a n a l o g u e , w h i c h provided very similar recovery to that of carboplatin.  M e t h C B D C A w a s a l s o the most lipophilic of the three  123  a n a l o g u e s , resulting in a retention time of 4 5 min a n d a n overall s a m p l e run time of 52 min for the H P L C - U V a s s a y . S i n c e only a single H P L C method w a s required for the p r o p o s e d clinical study, a c o m p a r a t i v e validation of both H P L C a s s a y m e t h o d s w a s undertaken in order to c h a r a c t e r i z e the performance of the m e t h o d s .  T h e validated concentration range w a s  0.05-40 n g / m L carboplatin in p l a s m a ultrafiltrate.  T h e lower limit c o r r e s p o n d e d to the  e s t i m a t e d L O Q of the m e t h o d s , a n d the upper limit w a s b e l i e v e d to e n c o m p a s s the r a n g e of C m a x concentrations likely to be o b s e r v e d in y o u n g patients at the d o s e s of carboplatin normally g i v e n . Certainly, this nearly 1000-fold concentration r a n g e is m u c h larger than a n y previously validated for carboplatin H P L C a s s a y s . parameters  included specificity  and  selectivity,  Validated assay  precision, a c c u r a c y , linearity,  and  r u g g e d n e s s . R e c o v e r y a n d stability w e r e e x a m i n e d for the H P L C - U V m e t h o d only. T h e o b s e r v e d a s s a y variability w a s m u c h higher for the H P L C - P C method d u e to timed e p e n d e n t c h a n g e s in signal r e s p o n s e c a u s e d by the d e g r a d a t i o n of s o d i u m bisulfite. Thus,  the  more  precise H P L C - U V  p h a r m a c o k i n e t i c study.  method  w a s s e l e c t e d for  u s e in the  clinical  C o m p a r e d to previous H P L C - U V a s s a y m e t h o d s e m p l o y i n g  c o n v e n t i o n a l H P L C equipment, this n e w a s s a y is 10-fold m o r e sensitive a n d is the first to e m p l o y a n internal standard for carboplatin quantitation. T h e p h a r m a c o k i n e t i c study involved evaluation of blood s a m p l e s from two y o u n g patients  receiving  carboplatin  as  part  of  ICE chemotherapy.  Eight  cycles  of  c h e m o t h e r a p y (four c y c l e s from e a c h patient) w e r e e v a l u a t e d . P l a s m a ultrafiltrate w a s p r e p a r e d from the blood a n d this ultrafiltrate w a s a n a l y z e d for free platinum by a n existing A A m e t h o d a n d for free carboplatin by the H P L C - U V method d i s c u s s e d a b o v e . O u r g o a l w a s to c o m p a r e the elimination profiles of platinum a n d carboplatin in this patient group. V i s u a l c o m p a r i s o n s s h o w e d that t h e s e profiles w e r e clearly different, and a n attempt w a s m a d e to determine the magnitude of this difference with r e s p e c t to the derivation of clinically significant p h a r m a c o k i n e t i c p a r a m e t e r s , most notably A U C .  In  both patients, A U C v a l u e s calculated from platinum levels w e r e larger than t h o s e c a l c u l a t e d from carboplatin levels. H o w e v e r , the o b s e r v e d differences w e r e s m a l l e r in o n e patient than in the other. W h i l e further study in a m u c h larger patient population is  124  required, the results of our clinical study s u g g e s t that the contemporary practice of interchanging carboplatin p h a r m a c o k i n e t i c p a r a m e t e r s b a s e d o n free platinum a n d free carboplatin  is  invalid.  pharmacodynamic  More  studies  relationships  based  investigating on  free  carboplatin  carboplatin  or  pharmacokineticcombined  free  On-line differential  pulse  carboplatin/free platinum determination are n e e d e d .  References 1.  F. Elferink, W.J. Van  der Vijgh,  and  H.M.  Pinedo.  polarographic detection of carboplatin in biological samples after  chromatographic  separation. Anal Chem 58: 2293-2296 (1986). 2.  M.A.  Allsopp,  G.J.  Sewell,  and  C.G.  Rowland.  A  column-switching  chromatographic assay for the analysis of carboplatin in plasma ultrafiltrate.  liquid  J Pharm  Biomed Anal 10: 375-381 (1992). 3.  M.J. Cleare, J.D. Hoeschele, B. Rosenberg, and L.L. Van Camp.  Malonato platinum  anti-tumor compounds. US Patent 4,140,707 (1979).  125  APPENDIX COMPARTMENTAL DATA FITS  Patient 1 Cycle 5, Dose 1 HPLC Data  50  10 i E c o £ c  -6- Observed — Predicted  03 o  c  o O  0.11  0.01 » 0  10  Time (h)  50  T  Patient 1 Cycle 5, Dose 1 AA Data  -©- Observed — Predicted  Time (h)  126  0.01 © 0  1  1  1  1  1  2  4  6  8  10  1  1  1  15  20  25  —  •  Time (h)  0.01 © 0  •  1  1  5  10 Time (h)  127  Patient 1 Cycle 7, Dose 1 HPLC Data  Observed — Predicted 0.1  i  0.01 © 0  1  1  1  1  1  2  4  6  8  10  Time (h)  o  o.i 4  0.01 *  1  1  1  1  0  5  10  15  20  1  25  Time (h)  128  Patient 1 Cycle 7, Dose 2 HPLC Data  50  10  £ c o  15  -©• Observed — Predicted  c o O  0.1  0.01 © 0  1  1  1  1  1  2  4  6  8  10  Time (h)  0.01 © 0  1  1  1  1  1  5  10  15  20  25  Time (h)  129  Patient 2  -©- Observed — Predicted  0.01 © 0  1  1  1  1  1  2  4  6  8  10  Time (h)  •©- Observed — Predicted  0.01 ©  1  i  h  1  1  0  5  10  15  20  25  Time (h)  130  Patient 2 Cycle 3, Dose 2 HPLC Data  -©- Observed — Predicted  Time (h)  Patient 2 Cycle 3, Dose 2 AA Data  -©- Observed — Predicted  10  15  20  25  Time (h)  131  0.01 ®  1  0  2  1 4  1  !  1  6  8  10  Time (h)  0.01  ffl 0  i  1  1  1  5  10  15  20  Time (h)  132  0.01 © 0  1  1  1  1  \  1  5  10  15  20  25  30  Time (h)  133  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0089694/manifest

Comment

Related Items