UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Evaluation of binding and neutralizing antibodies to interferon beta-1b in multiple sclerosis patients… Paszner, Beatrix M. 1999

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

Item Metadata

Download

Media
831-ubc_1999-0585.pdf [ 9.77MB ]
Metadata
JSON: 831-1.0088986.json
JSON-LD: 831-1.0088986-ld.json
RDF/XML (Pretty): 831-1.0088986-rdf.xml
RDF/JSON: 831-1.0088986-rdf.json
Turtle: 831-1.0088986-turtle.txt
N-Triples: 831-1.0088986-rdf-ntriples.txt
Original Record: 831-1.0088986-source.json
Full Text
831-1.0088986-fulltext.txt
Citation
831-1.0088986.ris

Full Text

EVALUATION OF BINDING AND NEUTRALIZING ANTIBODIES TO INTERFERON BETA-1B IN MULTIPLE SCLEROSIS PATIENTS RECEIVING TREATMENT: DEVELOPMENT AND VALIDATION OF A NEW ELISA COMPARED TO EXISTING ANTIBODY DETECTION TECHNIQUES AND THE SEARCH FOR A CLINICAL CORRELATE by BEATRIX M PASZNER B.Sc. (Biochemistry & Molecular Biology and Pharmacology & Therapeutics) The University of British Columbia, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Experimental Medicine) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1999 © Beatrix M. Paszner, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of 9xyPjr\cttfAo\ Ugiii fXAO The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract In recent years interferon beta has gained widespread use as a first line treatment for relapsing-remitting Multiple Sclerosis (MS). Treatment with interferons can lead to the development of antibodies. Lack of standardization in the measurement of antibodies has made comparison of antibody data, particularly with reference to cut-off points and determination of positivity, impossible. As a result, the frequency of antibody formation to interferon beta in MS patients has been reported to be as low as 5% and as high as 64%. The assays available for antibody detection are divided into 2 classes: binding antibody assays, usually in the form of an enzyme-linked immunosorbent assay (ELISA) and neutralizing antibody bioassays such as the antiviral neutralization bioassay and more recently, the MxA induction assay. In our studies we have developed an ELISA assay to compete with these neutralizing bioassays. However, a major problem in determining the validity of any of the antibody assays is the lack of a true gold standard. The antibody results as measured by ELISA were compared to those obtained previously from samples tested by both the antiviral and MxA induction assays for the detection of antibodies to interferon beta in patients who have received long-term treatment. The results from the assays were compared on a per patient basis, where positivity was determined for a single patient over the entire period of study using defined positivity criteria and on a per sample basis, where positivity was determined within a single sample only. These comparisons showed that the ELISA was both sensitive and specific for detecting antibody positivity. Due to a high number of ELISA binding antibody negative but neutralizing antibody positive results in samples with very low neutralizing antibody titres, we suggest a need to re-evaluate the titre cut-off point used by the neutralization bioassays, as they may be too low. ii Long-term follow-up of the antibody status of patients who had received treatment on average for >6 years, showed that sero-reversion of both binding (ELISA) and neutralizing (MxA induction assay) antibodies occurred. Finally, further work is required to prove that the measurement of antibodies (binding or neutralizing) is important on a clinical level. Although we appear to be able to reliably detect antibodies by each of the assay methods available, no clinical correlation has as yet been established with respect to the presence or absence of these antibodies. As a result, a study has been designed with the primary purpose of validating the use of the three assays in a clinical context. We cannot hope to establish antibody testing as a clinically relevant tool unless it can be related to some clinical effect. Until standardization and validation in these areas is achieved, no definitive conclusions surrounding the issue of the significance of antibodies can be made. Table of Contents Abstract » Table of Contents v List of Tables and Figures vii List of Abbreviations viii Introduction : Interferon Beta Antibodies in the Treatment of Multiple Sclerosis Interferons 1 Interferons in MS 2 Interferon Therapy and the Incidence of Antibodies 3 Past Experience 3 Interferon Alpha 3 Interferon Beta 4 Recent Experience 8 Interferon Beta 1b (Betaseron) Pivotal and Secondary Progressive Trials 8 Interferon Beta 1a (Avonex) Trial .11 Interferon Beta 1a (Rebif) Trial 14 Detection and Measurement of Interferon Antibodies 15 Total Binding Antibody Assays 15 Neutralizing Antibody Assays 17 Antiviral Assay (AVA) 17 MxA Protein Induction Assay (MXA) 18 Anti-Proliferation Assay (APA) 18 Clinical Relevance of Interferon Antibodies 19 Clinical Significance Antibodies on Treatment 19 Implications of Antibodies in Treatment 23 "Short-term" vs. "Long-term" Treatment 23 Selecting Assay Parameters 24 The Anti-Insulin Antibody Experience 26 Remarks 27 Chapter 1: Enzyme-Linked Immunosorbent Assay (ELISA) Materials and Methods Introduction 29 Preparation of Antigen 30 Immobilization of Antigen 30 Washing Protocol 30 Test Serum 31 Enzyme-Substrate System 31 ELISA Setup and Optimization 32 Q Selection of the Solid Phase 33 Immobilization of the Coating Antigen . .36 Blocking Non-Specific Activity 38 Antisera and Assay Controls 38 Selecting an Enzyme-Conjugate and Substrate System 41 Reading and Interpretaion of Results 43 Chapter 2: ELISA Results and Discussion Results 47 Pivotal Trial Study 47 Positivity Criteria 47 Frequency of Antibodies in Betaseron Treated MS Patients 48 Comparison of ELISA results with AVA and MXA results 49 Follow-up Study: Data Gathered During the Quality of Life Review (1997) 52 Reversion 52 Discussion 54 Conclusions 60 Chapter 3: Continuing Studies - "The Berlex-Schering AG Interferon Beta Antibody Study Designing a Study to Correlate Antibodies with Defined Clinical Outcome Measures .. 64 Investigational Plan 64 Objectives 64 Trial Design 65 Trial Population and Sample Size 68 Duration of Trial 68 Trial Methods and Procedures 68 Screening Phase 68 Treatment Phase 69 Design Rationale 71 Clinical Evaluations 72 Injections Site Reactions 72 Questionnaires 73 v Laboratory Evaluations Genetic Markers 74 76 Remarks 77 References 79 Appendix A 87 Appendix B 114 Appendix C 136 vi List of Tables and Figures Tables Table 1.1 Incidence and Clinical Significance of Antibodies to IFN-alpha 5 Table 1.2 Incidence and Clinical Significance of Antibodies to IFN-beta 6 Table 2.1 Frequency of Antibody Positivity in Betaseron Pivotal Trial Patients . . . . 48 Table 2.2 Comparison of Assays based on Per Patient Basis 50 Table 2.3 Comparison of Assays based on a Per Sample Basis 51 Table 2.4 Summary of Results from QoL Follow-up 53 Figures Figure 2.1 Plate Comparisons 35 Figure 2.2 Comparison of Coating Concentrations 37 Figure 2.3 Comparison of Blocked Vs. Unblocked Plates 39 Figure 2.4 Serum Concentration Curves 40 Figure 2.5 Comparison of Dilution Buffers 42 Figure 2.6 Enzyme Concentration Comparisons 44 Figure 2.7 Enzyme Concentration - Comparison of 1:2000 and 1:4000 45 Figure 2.8 Two-by-Two Contingency Table 49 Figure 2.9 Titre Distribution of MxA Assay Positive Samples 56 Figure 2.10 Titre Distribution of AVA Positive Samples 56 Figure 2.11 Time Curve of the Development of Binding vs. NAbs 58 List of Abbreviations ALP Alkaline phosphatase ALT Alanine Amino Transferase APA Antiproliferative Assay AST Aspartate Amino Transferase AVA Antiviral Assay Bili Bilirubin BSA Bovine Serum Albumin CA Condylomata Acuminata CBC Complete blood count CML Chornic Myeloid Leukaemia CNS Central Nervous System DNA Deoxy-ribonucleic acid dsRNA Double stranded ribonucleic acid EDSS Expanded disability status score EIA Enzyme Immunoassay ELISA Enzyme-Linked Immunosorbent Assay eod Every other day FCS Fetal Calf Serum FIA Fluorescence Immunoassay GGT Gamma Glutamyl Transaminase HBV Hepatitis B virus HCL Hairy Cell Leukaemia HCV Hepatitis C virus hD Hodgkin's disease FD High dose HPV Human Papilloma virus FR High responder FFP Horseradish peroxidase [FN Interferon IgG Immunoglobulin G isotype rrt Intramuscular IRMA Immunoradiometric Assay iv Intravenous LD Low dose LR Low responder LU Laboratory units MHC Major Histocompatibility Complex mU Million International Units MR Magnetic Resonance Imaging MS Multiple Sclerosis MSQLI Multiple Sclerosis Quality of Life Indices MTT 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide MxA Myxovirus MXA MxA Induction Assay NAb Neutralizing antibody NHL Non-Hodgkin's Lymphoma nhps Normal human pooled serum N1H National Institute of Health NK Natural killer NR Not reported NSCLC Non-small-cell lung cancer NU Neutralizing units CO Optical density OPD o-Phenylenediamine dihydrochloride RN Positive to negative PBL Peripheral blood lymphocyte PBS Phosphate buffered saline PL Placebo pNPP p-Nitrophenyl phosphate FVWI Pokeweed Mitogen QoL Quality of Life RCC Renal Cell Carcinoma RIA Radioimmunoassay RN Registered Nurse RRMS Relapsing-Remitting Multiple Sclerosis sc Subcutaneous SPMS Secondary Progressive Multiple Sclerosis TCR T Cell receptor UBC University of British Columbia USP United States Pharmacopaeia WHO World Health Organization 2° Secondary Introduction: Interferon Beta Antibodies in the Treatment of MS Interferons (IFNs) are a family of biologically active proteins found to have anti-viral, antiproliferative and immuno-modulatory properties.111 Although the etiology of Multiple Sclerosis (MS) is unknown, evidence linking MS to viral persistence or latency in the central nervous system [ 2" 7 ] and immunoregulatory disturbances18"161 led to clinical trials using recombinant IFN beta. Some of these trials have proven successful in reducing the rate and severity of exacerbations, and the activity and burden of disease as determined by magnetic resonance imaging (MRI).t171 Treatment with both natural and recombinant IFNs has been associated with the appearance of antibodies to IFN. However, the factors affecting the development of these antibodies and their influence on clinical response remain unclear. A significant problem in the area of IFN antibody measurement is a lack of standardization between the assays used by various groups or companies. This makes it impossible to interpret and compare antibody data from different studies. To date, two general categories of assay systems have been used; (i) those that detect binding antibodies, and (ii) those based on neutralization of a particular biological function by antibodies. The "neutralizing antibodies" (NAbs) are a subset of total binding antibodies that are thought to bind to the IFN molecule and block its function. Interferons Interferons were first identified in 1957 by Isaacs and Lindenmann[18] as a protein secreted by virally infected cells to confer viral-resistance to other cells. We now know that IFNs are involved, not only in antiviral activity, but also in modulating the immune system and a number of cellular functions, including cell proliferation and differentiation. Three types of IFNs have been identified; IFN alpha, beta and gamma. Their production is controlled at several different levels. The type of IFN produced is dependent on both the inducer molecule and the target cell being 1 stimulated. IFN alpha is produced by lymphocytes and monocytes in response to viral infection; IFN beta is produced by fibroblasts in response primarily to viral infection but also by in vitro treatment of the cells with dsRNA; and IFN gamma is produced by activated T-cells following antigenic stimulation. Each type of IFN has been localized to distinct areas within the active chronic MS lesion. [ 1 9 ' 2 0 ] The distribution of IFN gamma was found to echo that of class II MHC expression in astroglia, mainly on astrocytes located close to the edge of the lesion, near the normal appearing white matter. IFN alpha was found on macrophages and IFN beta was found on both astrocytes and macrophages. The cellular distribution of these IFNs within the active MS lesion is significant due to the immunomodulatory effects associated with each IFN. IFN gamma is known to enhance immune recognition and antigen presentation by increasing class II MHC expression on target cells. [ 1 9 " 2 2 ] Interferons in Multiple Sclerosis Multiple Sclerosis (MS) is a chronic neurodegenerative disease which is characterized by areas of inflammation and demyelination along neuronal axons in the central nervous system. Clinical manifestations are a reflection of recurrent neurologic dysfunction and are typically associated with progressive physical disability. Variability in the anatomic location and extent of lesion development are usually the primary factors associated with the unpredictability and slow progression of the course of disease. The exact etiology of MS is as yet unknown, however it is thought to involve any or all of genetic, environmental and immunologic factors. The earliest identifiable histological event in MS is believed to be a cellular immune reaction for which the triggering antigen is as yet unknown. Until recently, therapy for MS has involved the use of steroid therapy, to shorten the duration of acute exacerbations; and immunosuppressive drugs, on the assumption that MS may be the result of autoimmunity. The adverse reactions caused by these treatments has encouraged the need for more effective and less toxic forms of immunotherapy. Other forms of treatment have targeted symptomatic relief, and therefore offer no neuro-protection against disease 2 progression, let alone a reversal in disease course. In recent years, the use of IFNs as a treatment for MS has been introduced. The rationale for using IFNs in MS was initially based on the belief that the disease was coupled to persistent and/or latent viral CNS infection1671 However in later years there was also evidence that patients with MS had immunoregulatory disturbances as seen by decreased suppressor cell activity, NK activity and IFN synthesis. [ 1°" 12,23] other immune abnormalities include the upregulation of class II major histocompatibility complex (MHC) associated antigen expression and abnormal cytokine production.'21'241 Although the IFNs were identified and isolated in the 1950's, clinical application had to wait until the 1980's for recombinant DNA technology which would permit the cloning of the IFN genes and large scale techniques for producing purified IFN products. Since then various studies in MS patients have been conducted using all of the IFNs ( alpha, beta and gamma). Trial results from the use of IFN gamma showed that it increased the relative frequency of exacerbations [24,25] and results from IFN alpha were inconclusive. Only in the IFN beta trials was a positive treatment effect seen. Interferon Therapy and the Incidence of Antibodies Past Experience The development of antibodies in response to IFN treatment should have been expected. Past experiences indicate that in many cases, parenteral administration of a therapeutic protein is followed by the appearance of antibodies. Examples of antibody responses to foreign protein abound: bovine or porcine insulin,126"291 proteins secreted by human cells, [ 3 0 ' 3 1 ] blood factors,1321 or proteins produced by recombinant DNA techniques.133341 It is thus not surprising that treatment with both natural and recombinant IFNs leads to the development of anti-IFN antibodies. Interferon Alpha Anti-IFN antibodies were first reported by Vallbracht et al [ 3 5 ] in 1981 in a patient receiving 3 fibroblast IFN beta therapy. However, IFN alpha therapy was used first in trials for a variety of diseases, predominantly malignancies and viral disease, and thus early data on antibodies are related to IFN alpha trials. Antibody formation has been inconsistent and an array of confounding variables can be recognized. [Table 1.1] Levels of antibodies in any given situation are a function of many factors. The route of injection, dose and frequency of administration, along with the duration of treatment play an important role in the degree to which the antibody response is stimulated. The timing and frequency of the drawing of the sample, the standardization of the assay method and the units used in reporting the results, all affect the sensitivity and specificity of the antibody detection technique. The immunogenetic make-up of the individual and the antigenicity of the protein will affect the level of immune activation, rate of production and binding affinities of the antibodies. In addition, the disease(s) being treated as well as the concomitant medications used may also affect the antibody response. As a result, comparing the various trials is difficult, with several reports suggesting that possible treatment benefits are abolished in some antibody positive pa^jen^s [37,40,42,44,45,47,48,52,53,55,56] IFN alpha has been infrequently used in MS trials: Only six trials have been completed which have studied the effect of IFN alpha in MS; 4 on relapsing-remitting MS [ 5 8 ' 6 2 ] and 2 on progressive MSJ63.64]. Antibodies were measured in only one of those studies.1581 The authors report finding no anti-IFN alpha antibodies in the 24 relapsing-remitting patients receiving 6 months of systemic IFN alpha treatment. Interferon Beta Use of IFN beta therapy followed IFN alpha closely. [Table 1.2] As with IFN alpha, IFN beta has been used in the treatment of viral disease and malignancies with variable success; but IFN beta's major success has been in the treatment of MS. Comparing the effect of antibodies to IFN beta among the different diseases provides little information. Unlike studies involving both viral 4 Table 1.1 Incidence and Clinical Significance of Antibodies to IFN-alpha Incidence of Reduced Reference Disease Assay Antibodies 3 Clinical [% (number)] Response13 Interferon-alpha 2a Gutterman et al.t3 6! Mixed Cancer EIA/AVA 19 (3/16) no Quesada et al.t 3 7 ' R C C EIA/AVA 38 (20/53) yes Figlin et al.l 3 8J R C C EIA/AVA 32 (6/19) no Itri et a l . l 3 9 ! Mixed Cancer EIA/AVA 25 (152/617) no Quesada et a l . ! 4 0 ! HCL EIA/AVA 18 (14/78) yes Gauci et al.t 4 1! Mixed Cancer EIA/AVA 11 (50/467) no Steis et al.t 4 2l HCL EIA/AVA 31 (16/51) yes Itri et a l . l 4 3 ! Mixed Cancer EIA/IRMA/AVA 25 (160/653) no Porres et a l . ' 4 4 ! Chronic HBV EIA/RIA 26 (12/46) yes Lok et a l . l 4 5 ! Chronic HBV EIA/AVA 39 (21/54) yes Thomas et al.t4 6! Chronic HBV EIA/AVA 16 (29/180) yes Gianelli et al . l 4 7 ! Chronic HCV EIA/AVA 23 (15/63) yes Antonnelli et al.t 4 8 ' Chronic HBV EIA/AVA 23 (85/375) yes Interferon-alpha 2a, 2b, lymphoblastoid Craxi et al. 1988 l 4 9 l Hepatitis EIA 11 (4/35) no Interferon-alpha 2b Spiegel et al.l5°l Mixed Cancer IRMA/AVA 2 (13/537) no Spiegel! 5 1! Mixed Cancer IRMA/AVA 3 (16/630) no Leavitt et al.t 5 2 ' NHL/HD IRMA/AVA 3 (2/64) yes VonWussow et a l . ' 5 3 ! CML EIA/RIA/AVA 20 (5/25) yes Golomb et al. t 5 4 l HCL IRMA/AVA 0 (0/82) -Oberg et a l . ! 5 5 ! Carcinoid EIA/RIA/AVA 15 (5/35) yes Freund et al.t5 6! CML ELISA/IRMA/AVA 30 (8/27) yes Inteferon-alpha n1 Week et a l . ! 5 7 ! HPV A \ / A 3 (13/391) Mixed Cancer M V M 0 (0/221) no Interferon-alpha (natural) Knobler et al.t5 8! RRMS Not Specified 0 (0/24) -a Incidence determined from last assay shown (ie. EIA/AVA = incidence determined from AVA) if results from both assays did not concur. b Not always statistically significant AVA = anti-viral assay; CML = chronic myeloid leukaemia; EIA = enzyme immunoassay; ELISA = enzyme-linked immunosorbent assay; HBV = hepatitis B virus; HCL = hairy cell leukaemia; HCV = hepatitis C virus; HD = Hodgkin's disease; HPV = human papilloma virus; IRMA =immunoradiometric assay; NHL = non-Hodgkin's lymphoma; RCC = renal cell carcinoma; RIA = radioimmunoassay; RRMS = relapsing-remitting multiple sclerosis. 5 Table 1.2 Incidence and Clinical Significance of Antibodies to IFN-beta Reference Disease Assay Incidence of Antibodies3 [% (number)] Reduced Clinical Response13 Interferon-beta 1a (Avonex®) Jacobs et a l . ! 6 5 ! RRMS Not Specified 22 (19/85) -Rudick et a l . l 6 6 l RRMS ELISA/AVA 6 (4/70) no Interferon-beta 1a (Rebif®) Byhardt et al.t6 7! NSCLC ELISA 18 (2/11) -Bornstein et a l . ' 6 8 ! HPV-CA ELISA 8 (5/60) no Dinsmore et a l . ' 6 9 ' HPV-CA ELISA 0 (0/124) -Abdul-Ahad et al . l 7 0 l RRMS, SPMS ELISA/AVA 7 (23/334)c 14 (48/334)d -Abdul-Ahad et a l . l 7 1 l RRMS, SPMS ELISA/AVA 5 (14/266)c 11 (28/266)d -PRISMS Study Group! 7 2! RRMS ELISA/AVA 22ug 24 (45/189) 44 ug 13 (23/184) no Antonelli et a l . ! 7 3 ' RRMS AVA 16 (10/63) yes Interferon-beta 1 b (Betaseron( Rinehart et a l . ' 7 4 ! R C C ELISA 38 (6/16) no Konrad et a l . l 7 5 l Mixed Cancer ELISA/AVA 80 (24/30) no Larocca et a l . ' 7 6 ' R C C MS Mixed Cancer ELISA/AVA 1.(3/332) 64 (9/14) 8(4/53) no Knobler et a l J 7 7 ! RRMS AVA 58 (14/24) no IFNB MS Study Group!78) RRMS ELISA/AVA 35 (43/124) yes Kivisakk et al . l 7 9 l RRMS FIA/AVA 42 (8/19) -EuropeanStudy Group' 8 0! SPMS MxA Assay 28 (100/360) yes Fierlbeck et aU 8 i ] (Betaferon®) Melanoma AVA/APA 28 (2/7) no Interferon-beta (recombinant) Niijima et a l . l 8 2 ! Bladder Cancer Not Specified 0 (0/51) -Interferon-beta (natural) Dummer et al . ' 8 3 l Melanoma AVA SC 92 (13/14) IV 30 (6/20) no Fierlbeck et a l . [ 8 1 l Melanoma AVA/APA 95 (20/21) no a Incidence determined from last assay shown (ie. EIA/AVA = incidence determined from AVA) if results from both assays did not concur. b Not always statistically significant c NAB definition of 2 consecutive positive titres d NAB definition of 1 positive titre at any time APA = antiproliferative assay; AVA = anti-viral assay; CA = condylomata acuminata; ELISA = enzyme4inked immunosorbent assay; FIA = fluorescence immunoassay; HPV = human papilloma virus; IV = intravenous; MS = multiple sclerosis; NR = not reported; NSCLC = non-small-cell lung cancer; RCC = renal cell carcinoma; RRMS = relapsing-remitting multiple sclerosis; SC = subcutaneous; SPMS = secondary progressive multiple sclerosis. 6 disease and malignancy which have well-defined endpoints, studies involving MS rely on endpoints which are more difficult to measure. The course of MS is complex and involves a multitude of parameters which are more difficult to control for. It is this "fuzziness" which makes it difficult to evaluate how IFN antibodies affect a therapeutic effect which in itself is far from being obvious. Several large, well designed placebo-controlled, double blind, randomized clinical trials have compared placebo to IFN beta in relapsing-remitting M S . [ 1 7 ' 6 5 , 7 7 ' 7 2 1 These trials have demonstrated decreased exacerbation rates, reduced severity of exacerbations, slowed progression of disability and reduced MRI activity. However, each study has focused on a different primary end-point and results have varied depending on the type of IFN beta used. While Betaseron® (IFN beta-1b) decreased annual exacerbation rates and MRI activity with no significant effects on the progression of disability, Avonex®(IFN beta-1a) was found to slow disability as measured by EDSS, and decrease the number of new and enlarging lesions. Only Rebif® (another IFN beta-1a) reduced exacerbations, slowed progression and reduced MRI activity. Unfortunately not all trials measured or reported results of antibody levels. The results which have been published, are not straightforward or consistent. Early studies with IFN beta reported that the development of antibodies may be route dependent;175,761 i.v. administration is associated with a lower incidence of antibody formation than i.m. or s.c. route. In one study by Larocca et a l , [ 7 6 ] where patients with different diseases were evaluated for antibody response, 9 out of 23 (39%) MS patients receiving s.c. IFN beta, developed antibodies. This was more frequent than in any of the other disorders treated and confirms that MS patients exhibit a brisker immune response than other patients tested. Higher antibody production by MS patients has been well documented with a higher incidence of antinuclear [ 8 4 1 and antiviral antibodies184"901 in the serum of patients with MS than in both controls and other neurological diseases (for review see Oger, Roos, Antel 1983[911). 7 Recent Experience The approval of Betaseron® Avonex® and Rebif® by regulatory authorities for the treatment of relapsing-remitting MS has opened an enormous market for use of immunomodulatory molecules in Multiple Sclerosis. The different molecules marketed: Betaseron®, Avonex® and Rebif® show only subtle differences in their effectiveness, side effect profile and antigenicity, but these characteristics are being used as a marketing argument. The financial stakes render the subject of the effect of antibodies on benefits of IFN therapy quite a sensitive one. Interferon beta-1b (Betaseron®) Pivotal and Secondary Progressive MS Trials The approval of recombinant human IFN beta-1b (Betaseron®) for the treatment of relapsing-remitting MS was based upon results obtained during a 5 year multicentre trial in both the US and Canada. [ 1 7 J Data were gathered on 372 clinically definite relapsing-remitting MS patients during this carefully designed, blinded, randomized and placebo controlled study. Early results reported after 3 years of study, showed a 34% reduction in exacerbation rates between the high dose (8 million international units (mlU)) treatment arm and the placebo with a 50% reduction in the frequency of moderate and severe attacks [17] and an 83% reduction in disease activity on MRI with a 23% reduction in MRI lesion burden. [92] At the time of the initial reports, using an ELISA and an anti-viral neutralization bioassay (AVA) NAbs were reported in 11% of the placebo sera, 47% of low dose group (1.6 mlU) and 45% of high dose group (8 mlU). [ 17 ] No significant effect of these antibodies on exacerbation severity or frequency was reported. The final 5 year outcome report confirmed the original results obtained by the third year of study. [93] The one-third reduction in exacerbation rates was sustained to the end of the study. The severity of exacerbations was still significantly less in the treated groups. MRI studies showed no increase in the burden of disease in the high dose group as compared with baseline, 8 while the burden of disease was significantly increased in the placebo group. Measurement of NAbs in treated patients using redefined criteria for positivity resulted in 35% (43/124) positivity among the patients in the high dose treatment group. However, by cross-sectional analysis a relationship between the presence of NAbs and exacerbation rate was now found, with exacerbation rates in NAb positive patients twice those in NAb negative patients within the high dose treatment group. To address the incidence and significance of the NAbs detected during the pivotal trial, a report was published to provide and discuss the NAb data generated during the study.1781 Information regarding the validation of the NAb detection assay system along with additional clinical and MRI data was included. Validation of the original AVA involved the retesting of selected sera from the initial 3 years of the trial using an independent reference laboratory. Some positivity was recorded in placebo patients (11%) which led the investigators to redefine the criteria for NAb positivity. The definition for NAb positivity was changed to require that two consecutive serum samples from a patient exhibit a titre of at least 20 neutralizing units (NU)/ml; such a patient was considered positive at all times subsequent to the first titre greater than 20 NU/ml: "once positive, always positive". Therefore a patient with at least two consecutive titres greater than 20 NU/ml would be classified as NAb positive, even if they became negative later. The cut-off titre of 20 NU/ml was originally based on the maximum sensitivity (lowest titre detectable) of the AVA . The criteria for two consecutive positive titres was selected empirically as giving the lowest possible positive rate among placebo patients. This definition was also used in an attempt to attribute the maximum possible significance of NAb status on the primary outcome measure. With these new definitions the NAb positive rate in placebo patients fell from 11% to 2% while only reducing the positive rate from 45% to 35% in the 8 mlU group. No clear dose-response effect was seen between the high and low dose groups with respect to NAb formation: in each group approximately the same proportion of patients became positive within the same time intervals. The attenuation of treatment effect in NAb positive patients was found to be most noticeable with 9 respect to relapse rate and MRI burden of disease activity. No evidence of an association between NAb positivity and increased disability as measured by EDSS over the period of study was seen. Results from the trial conducted using IFN beta-1b in secondary progressive MS have recently been published1801 and indicate in patients with secondary progressive disease the beneficial effect observed in relapsing-remitting patients. The primary endpoint for this trial was the time to confirmed progression of disability (as measured by a sustained increase in EDSS of 1.0 point over 3 months or by a 0.5 point increase in patients with a baseline EDSS of 6.0 or 6.5). Seven-hundred-eighteen patients were randomized into 1 of 2 groups: 358 were given placebo and 360 were treated with 8 mlU for up to 3 years. An interim analysis of the results after all patients had completed 2 years of the study showed a highly significant effect on delaying the progression of disability in treated patients leading to early study termination. 50% of the placebo group had confirmed progression during the trial (40th percentile=549 days) whereas 39% of the IFN beta-lb treated group had confirmed progression (40th percentile=893 days). This delay in time to progression represents a 22% reduction in the proportion of patients with progression over the total period of study. Among the secondary endpoints, there was a 32% reduction in treated patients becoming wheelchair bound (corresponding to a change in EDSS from 6.5-7.0) versus placebo. The mean annual relapse rate was reduced by 30%, the time to first relapse was prolonged and there was a decrease in the severity of the relapses among the treated group. MRI outcome measures were also positively affected by IFN beta-1b treatment. There was an 8% decrease in the mean MRI T2 lesion volume from baseline of the treated group and a 5% increase from baseline in the placebo group during the study. A frequent MRI cohort showed that newly active lesions were also decreased during treatment. NAbs were found in 28% (100/360) of patients receiving active treatment; 66 of which had become positive during the first 6 months of treatment. Reversion to negative with at least one negative NAb titre was seen in 71% (47/66) of the positive patients, with sustained reversion in 79% (37/47). The only information on NAb effects provided in this initial report is a statement that longitudinal analyses, 10 based on the data for NAb positive patients, showed a significant increase in exacerbation rate associated with the switch in status from NAb negative to NAb positive, but no such effect was seen in the EDSS scores. Interferon Beta-1a (Avonex®) Trial The initial outcome report of intramuscular administration of IFN beta-1a (Avonex®, Biogen) was published in 1996. [65] Three-hundred and one patients with clinically definite relapsing-remitting MS were enrolled in this 2 year multicentre, double-blinded, placebo controlled randomized trial. Treated patients received 6 mlU/dose of Avonex® intramuscularly once a week. The primary endpoint was the time to onset of sustained worsening in disability as defined by an increase from baseline of at least 1.0 EDSS point confirmed at 6 months. Results showed that the time to sustained progression was significantly longer in the treated group with a mean increase in EDSS scores at 2 years of 0.25 versus 0.74 in the placebo group. Among patients treated for 2 years (a sub-population), 33.3% of the placebo group and 21.1% in the treated group exhibited sustained progression of disability. Secondary outcome variables included the frequency of attacks, MRI plaque load and the number and size of new enhancing lesions. The annual exacerbation rate of all treated patients was reduced by 18% relative to the placebo group (32% in patients treated for 2 years). Of statistical significance was a greater than 50% reduction in the number of treated patients with 3 or more exacerbations. In the first year of study the number of positive MRI scans and the number and volume of gadolinium enhancing lesions in treated patients were significantly less than in placebo patients. In a later report focusing on the MRI results from the Avonex® trial, [94 ] a similar trend was shown by a significant decrease in accumulated number of new, enlarging and new plus enlarging lesions after 2 years of treatment when compared with placebo patients. In this trial, neutralizing antibodies in the treated group were found in 14% of patients after 1 year of treatment and in 22% of treated patients after 2 years. Neutralizing activity was also seen in 11 4% of the placebo patients. No association was reported between the incidence of neutralizing activity and a possible attenuation of treatment effect. The method used to assay for anti-IFN neutralizing activity was not specified in the original publication.[65] The subject has been recently revisited by Rudick et a l [ 6 6 ] using a 2 step assay technique with samples first screened by ELISA and then assayed for NAbs using AVA. Two cut-off points were used: 5 units NU/ml and 20 NU/ml and a single positive result was needed to meet positivity criteria (in contrast with the need for 2 consecutive positive in the previous study). The authors confirmed that NAbs to IFN beta-l a increase with time on drug: 1% and 0% (> 5 NU/ml and > 20 NU/ml respectively) at 6 months, 7% and 4% at 12 months, 6% and 6% at 18 months and then decrease again to 3% and 3% at 2 years. These numbers were generated from 84 new patients put on drug in an open-label study and are considerably lower than the numbers reported for the original 301 patient trial of Avonex®; 14% at 1 year and 22% at 2 years (not all patients completed 2 years). [ 6 5 ] This difference was attributed to the fact that the 84 new patients were being treated with a new formulation of the drug meant to reduce in vitro aggregation of the IFN beta-1a molecules. Furthermore, it was shown that the neopterin response (an IFN induced activation of macrophages to secrete the protein neopterin) to IFN beta-1a injection is blunted when NAbs are present; however no effect of NAb positivity on MRI and clinical parameters was found. If anything, the NAb positive patients appeared to fare slightly better than the NAb negative patients; the numbers were so low however that no difference was to be expected. Rudick et al also studied NAb levels in 118 patients who were in the process of switching from IFN beta-1b (Betaseron®) to IFN beta-1a (Avonex®).[66] They found 22% and 0% positivity against IFN beta-1a (with a cut-off at > 5 NU/ml and > 20 NU/ml, respectively) for patients who had been treated with IFN beta-1b for 0 to 6 months, 31% and 10% for 7 to 11 months, 35% and 23% for 12 to 18 months and 39% and 26% for > 18 months. These patients had made the decision to switch treatment from IFN beta-1b to IFN beta-1a primarily because of side-effects, however some may have known that they were NAb positive, thus possibly introducing a bias in 12 recruitment. Only a minority of these patients (5/118) reported having switched as a result of disease worsening. As these numbers are higher than those discussed above following IFN beta-1 a injections, Rudick et al suggest that IFN beta-1 b is more immunogenic than IFN beta-1 a. It should be stressed that, in this study, sera from patients having received IFN beta-1 b were tested for presence of neutralizing antibodies using IFN beta-1 a as the assay substratum; this implies that what the authors studied is cross-reactivity between IFN beta-1 a as antigen and antibodies generated by IFN beta-1 b treatment. This cross-reactivity has not, however, been studied systematically neither by Rudick et al, or any other group. The conclusion about the relative immunogenicity of the 2 IFN beta preparations therefore relies on the assumption that antibodies to IFN beta-1 b are not more potent in binding to and neutralizing IFN beta-1 a than antibodies to IFN beta-1 a itself, an assumption for which there is currently no experimental support. This conclusion about relative immunogenicity is also based on the elapsed time to antibody appearance and does not directly compare patients receiving IFN beta-1 a versus those receiving IFN beta-1 b in terms of the kinetics of antibody appearance and their affinity. We, and the group in London (Ont.), have shown that when injections of IFN beta-1 b are continued in antibody positive patients, antibodies disappear with longer follow-up (in 11 of the 11 NAb-positive patients for Rice et al . [ 9 5 ] and 11 of 14 NAb-positive patients for Oger et al. [ 96 ]). We wonder if this phenomenon could be dose-dependant and suggest that this problem should be analyzed in terms of total amount of IFN received rather than time to antibody positivity. A point should also be made on methodology: The laboratory units used between the different groups are extremely different and 20 NU in the Biogen assay are clearly different in terms of number of antibody molecules from 20 NU in the Berlex assay (and also from the Serono assay; see below). Practically speaking, we believe that it is impossible to compare the results presented by Rudick et with those reported by the IFNB MS Study Group.!78! The need for 2 consecutive positive titres and the operational terminology "once positive, always positive" were 13 used to define positivity only in the Berlex studies. This definition leaves no place for sero-reversion to negative and unfortunately no account of reversion to NAb negative status is given in Rudick et al. [ 6 6 ] It is now known that most patients who have developed antibodies to IFN beta-1 b revert to NAb negative on follow-up for >5 years. [ 9 5"9 7 ] Interferon Beta-1a (Rebil®) Trial Preliminary findings have recently been reported by Ares-Serono regarding the incidence of NAbs during the use of IFN beta-1a (Rebif®).[70"72] This double-blind, placebo-controlled trial is the largest relapsing-remitting MS study completed to date: it involved 560 patients with an entry EDSS of 0 to 5.5 followed for 2 years. lnterferon-beta-1a was administered subcutaneously 3 times per week at 1 of 2 doses: 6 mlU (22(ig) or 12 mlU (44ug). The first quartile of time to confirmed progression of disability was doubled in the high dose group; overall disability was also reduced, as were the number and severity of relapses. A dramatic reduction in disease activity and disease burden was also seen in a dose-response fashion J 7 2' 9 8! Sera were tested for neutralizing activity pre-treatment, and every 6 months using an ELISA pre-screen followed by the AVA assay for detection of NAbs. To reduce interference in the AVA caused by circulating levels of IFN, a minimum time of 24 hours post-dose was required prior to drawing of blood samples. NAb results were analyzed using 2 definitions for NAb positivity. f701 The first definition required evidence of only one positive sample at any point during the time of treatment. The other definition was based on the criteria of sustained positivity as established during the Betaseron® trials. Of the 303 patients on active drug treated for 2 years, NAb positivity at the end of the trial was 24% for patients receiving 22 ug (with 4 of the 49 NAb+ patients having reverted) and only 13% for patients receiving 44 ug (with 9 of 32 NAb+ having reverted). In this study it appears that reversal of NAb positivity can occur for IFN beta-1a as has been shown for IFN beta-1b. The kinetics of the appearance and disappearance of antibodies seems to be dose-dependent both in IFN beta-1a and 1b. The lower rate of positivity 14 in Avonex® (at 2 years) may in fact represent a delayed response secondary to the use of a lower dose (30 pg/week) rather than the effect of drug formulation and route of injection. If this were the case, one would expect that NAbs in Avonex® treated patients would continue to rise before they start to revert to negative. In the Rebif® study, the antibodies did not appear to be of clinical significance but no detail was given. [72] Detection and Measurement of Interferon Antibodies To gain a clearer understanding of the issues surrounding the question of antibody relevance, it is necessary to review the assays used in antibody determination. Antibodies are heterogeneous by nature; in any given individual antibodies will exhibit different binding affinities, avidities and specificities. In any given individual these binding characteristics will also vary with time. The various types of immunoassays available will selectively detect antibodies sharing the same set of binding characteristics. The immunoassay the clinician needs should detect those antibodies which have clinical relevance. Currently many different methods are used for the detection and measurement of IFN beta antibodies and no attempt at a consensus has been made. These assays can be divided into two broad categories: a) the direct detection of binding antibody (total antibodies) and b) the indirect assay for neutralizing activity of antibodies (neutralizing antibodies). Among the neutralizing assays, the anti-viral neutralization assay (AVA), the standard method recommended by the World Health Organization (WHO), [99] has been the most widely used. More recently, the use of a non-standardized MxA induction assay defined "in house" by Berlex Laboratories has been reported [ 1 0 0 ] and tested for use in the detection of NAbs. The effect to be neutralized by the NAbs is the induction of the MxA protein. Absence of standardization of the assay and limits placed on attempts at external validation, as consequences of trade-marking the MxA induction assay, have rendered the field most difficult to evaluate and have generated some strong feelings such as expressed by Pachner. [ 101 ] 15 Total Binding Antibody Assays Total binding antibodies can be measured in two ways; by enzyme linked immunosorbent assay (ELISA) or by immunoblotting; ELISA is easier and less costly. These assays have now been relegated as a pre-screen to the neutralization bioassays, as they are based on the binding of antibodies to any epitope of the IFN molecule, even those with low binding affinity. Immunoblotting or Western Blotting is a qualitative assay rarely performed outside of a research setting. The enzyme immunoassay (EIA) is an indirect assay which detects serum factors that bind to IFN. Antibodies to IFN are detected by measuring the binding of one antibody molecule to two IFN molecules; creating a bridge between IFN which is coupled to a solid phase and IFN which is conjugated to an enzyme. Substrate is added and the colorimetric reaction quantitated by spectrophotometry. Two types of radioimmunoassays have also been used to detect IFN antibodies. One radioimmunoassay (RIA) is identical in principle to the EIA. But the second assay is a competitive RIA or immunoradiometric assay (IRMA) which tests the ability of sera to compete with radiolabelled IFN antibody for the binding of IFN. [ 1 0 2 ] In principle, the ELISA detects and quantifies any antibody which binds to IFN using an indirect, non-competitive, antigen coated plate system. The IFN molecule is adsorbed to a solid phase (the plate) after which serum is then added. Any antibodies in the serum will bind to the IFN on the plate, forming an "IFN-antibody" complex. A washing procedure removes any unbound serum factors. A secondary (2°) antibody specific for the human anti-IFN antibody will then bind to the "IFN-antibody" complex on the plate. This secondary antibody is generally conjugated to an enzyme such that when substrate is added the enzyme-substrate reaction produces a chromogenic reaction. The colour intensity of this reaction, which can be quantified using spectrophotometric analysis, is based on the number of "IFN-antibody-2° antibody" complexes ! 16 which have been formed. Neutralizing Antibody Assays Antiviral Assay The antiviral neutralization bioassay (AVA) is the standard method recommended by the WHO [ 9 9 ] for the measurement of neutralization of the biological activity of IFN, based on the measurement of cytopathic effect. Another neutralization assay, the MxA induction assay (MxA Assay) has been reported1 1 0 0 1 0 3 1 and tested for use in the detection of NAbs and occasionally one may come across the use of an anti-proliferative assay (APA). Neutralization assays are functional assays which measure the extent by which IFN antibodies inhibit the biological activity of IFN. The AVA specifically measures the neutralization by antibodies of IFN's ability to inhibit viral mediated cell lysis. This assay method is also commonly referred to as the cytopathic endpoint assay, as results are a reflection of the cytopathic effect caused by a virus on a target cell. Serum samples are diluted into several concentrations and incubated with a predetermined amount of IFN known to protect an IFN sensitive cell line from the cytopathic effect of a viral line. The serum-IFN sample is then added to the cell line, incubated and then infected with virus. After sufficient viral challenge, the cultures are scored for cytopathic effect and compared to positive and negative control. Cells incubated with sera free from NAbs would exhibit no cytopathic effect; whereas cells incubated with sera containing NAbs would show some degree of cytopathic effect. Quantitation of these results is a reflection of the neutralizing potency of a patient's serum; more specifically it has been defined as the dilution of serum that is capable of reducing 10 laboratory units (LU)/mL of IFN to 1 LU7mU1 0 4 1 A major drawback to neutralization assays is that non-antibody (or non-IgG) serum factors may be responsible for the inhibition of the antiviral activity of IFN, thus leading to false positive results. As well, human serum can cause direct cytotoxicity to the cells used in these cytopathic 17 assays and skew results in favour of a higher cytopathic effect with increasing false positive results and decreasing specificity. MxA Protein Induction Assay Berlex Laboratories has set up a MxA induction assay for the measurement of NAbs and licensed it to Immuno-Diagnostic Laboratories. The MxA assay is a variant of the AVA assay: IFN after binding to its receptor activates the MxA gene (Myxovirus activation) and increases transcription of the MxA protein which is then measured. The MxA assay is a two step assay. The first step involves the incubation and co-culture of patient sera, IFN and a cell line sensitive to induction of MxA protein by IFN. The second part of the assay is a two site immunochemiluminescent assay specific for the MxA protein, somewhat similar to a non-competitive antibody bound ELISA. [ 1 0 5 ] MxA protein in the cell lysate is quantitated by comparing the sample results against standards of purified recombinant MxA protein.11001 Anti-Proliferation Assay The antiproliferation assay (APA) is rarely used for the measurement of NAbs; however it is mentioned here because it has been previously used in some studies. The test principle is based on the anti-proliferative effect of IFN on cell growth. This assay uses an MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) test which indirectly measures cell proliferation.[106' Results are expressed as a percentage of proliferation as defined by control samples. The biggest problem with most bioassays is the high intra-assay and inter-laboratory variability. This variability can give rise to different results based on the different sensitivities of each assay. Many labs have their own preferred cell line, preferred viral antigen and preference for methods of diluting IFN, as opposed to the conventional dilution of serum, to obtain the desired results. Endpoints may also vary; some laboratories use cytopathic effect while others estimate 18 viral growth or induction of specific viral or cellular enzymes or proteins. Each laboratory defines its own "Laboratory units" (LU). These units are laboratory specific and are derived from the ability of a certain dilution of serum to neutralize a laboratory defined (LU) amount of the IFN beta being studied. In the IFN beta field, this is rendered even more complex due to the absence of standardization between the different molecules in terms of activity per microgram. No comparison of the 3 IFN betas presently used in MS has been made with respect to a unique common laboratory standard. Finally, all the neutralizing antibody measurements reported above are based on the inhibition of some function of the IFN beta. However, as the specific mechanism(s) by which IFN beta influences the course of MS has not yet been identified, all of the neutralizing assays are one step removed from direct clinical relevance. Clinical Relevance of Interferon Antibodies Because an association with clinical effect relies heavily upon the detection of antibodies; the evaluation of each study, protocol testing and method of data analysis are important. Until the issue of assay standardization is resolved, few firm conclusions can be made with respect to clinical relevance; and yet it is futile to have an assay that measures antibody positivity without attempting to correlate it with clinical effectiveness. Clinical Significance of Antibodies on Treatment To date, four reports 1 6 5 ' 6 6 ' 7 2 ' 7 8 , 8 0 1 have looked at the association between the incidence and clinical significance of NAbs during the treatment of MS with IFN beta. However, none of the studies involved has focussed extensively on solving the "dilemma" of antibodies; but rather have used the data already gathered for endpoint determinations and applied them to look for significance retrospectively. The Rebif®[72] and Avonex®165661 studies did not find any evidence that NAbs influence clinical efficacy. With Betaseron®, the tendency for MS patients to develop 19 more antibodies could be traced to the article by Larocca et a l . [ 7 6 ] in 1989; however the hypothesis that a difference in clinical course could exist between NAb positive and NAb negative patients was not fully investigated until final results were published in 1995. [93 ] In 1996, a more detailed report was published1781 discussing specifically what effect NAbs exerted during the initial 3 years of the pivotal trial. And in 1998, the European Secondary Progressive Study also produced data pointing toward a NAb effect in treated patients.[80] Comparison of available baseline characteristics showed no significant differences between the NAb positive and NAb negative groups prior to study entry. It is possible, however, that the baseline characteristics collected are neither relevant nor sensitive towards predicting NAb development. Nevertheless, we have generated data suggesting that patients destined to become NAb positive had a higher in vitro T cell dependent B cell IgG secretion even before treatment.11071 In 50 patients from the Vancouver Cohort of the Betaseron pivotal trial, we found that the level of IgG secretion of individual patients in response to in vitro stimulation may be a predictor for future NAb development.11071 IgG secretion, as measured in 7 day lymphocyte cultures in response to pokeweed mitogen (PWM) stimulation showed that 73% (11/15) of patients exhibiting high IgG secretion levels at baseline eventually tested NAb positive, as opposed to only 25% (4/16) from a low responder group (p<0.05). These results suggested that high in vitro secretion may be a predictive marker of patients prone to developing NAbs. NAb positivity was not a dose-dependent phenomenon in the pivotal Betaseron® trial: approximately the same proportion of low dose and high dose recipients developed NAbs. After 18 months of treatment, the NAb positive high dose sub-group exhibited a statistically significant increase in exacerbation rate, (when compared to the NAb negative high dose sub-group) resembling that of the placebo group. This difference in exacerbation rates was not accompanied by an increase in disability. NAb positive patients receiving low dose treatment, on the other hand, showed lower relapse rates than NAb negative patients. NAb positivity was shown not to play a major role in discontinuation of therapy nor did it show any difference in the 20 frequency of adverse events. NAb positivity was accompanied by reduced treatment effect as judged by MRI only in the high dose group. The annual incidence of enlarging lesions was significantly greater in the antibody positive group than in the NAb negative group, however it did not exceed that of the placebo group. Among the high dose group, NAb positive patients showed a trend towards greater incidence of new lesions than in the NAb negative group. The higher MRI burden of disease seen in the NAb positive patients in both treatment groups did not attain significance.1781 This evidence for reduced treatment efficacy in the IFN beta-1 b trial has been scrutinized more closely by Petkau and White of the UBC Biostatistics Research Group. In a report presented at ACTRIMS 1997, [1081 they suggest that the data generated during the pivotal trial indicating reduced treatment effect in NAb positive patients may have been over-interpreted. The findings had been based on cross-sectional analysis of the data and therefore could be due to factors other than the difference in NAb status. Results based on up to 5 years of follow-up data showed no significant effect of the switch in NAb status in the high dose group when more sophisticated longitudinal analyses were applied.11081 Cross-sectional comparisons attribute differences between the two groups to their change in antibody status when other factors not controlled for in the design or in the analysis may have been at work. Longitudinal analyses provide a direct assessment of association between the switch in NAb status and the clinical results of IFN treatment. The data have been analyzed with respect to both primary and secondary outcome responses: exacerbation rates and EDSS scores and burden of disease by MRI. These longitudinal analyses allow for trends seen naturally over the course of the trial thus accounting for built-in biases. One potential source of such a bias is the spontaneous reduction in attack rate which is known to occur over time in all MS patients and is clear in the placebo arm of the Betaseron® pivotal study. Petkau and White took into account the reversion to negative NAb status in their analyses in 21 addition to using the "once positive, always positive" definition from the IFNB Study Group. Incorporating reversion to NAb negative thus created a second view of the same dataset. Using NAb data obtained from the AVA assay a significant association was detected only for one outcome response: there was a significant association between the change from NAb negative to NAb positive status and increased annual exacerbation rate in the low dose group. Using the MxA assay data, associations with increased exacerbation rates were found in both low and high dose groups when the "revert to negative" definition was used, but not when the "once positive, always positive" definition was used. [ 1 0 8 ] Petkau and White could not make an unequivocal statement regarding the impact of the switch in NAb status on the efficacy of IFN beta and thus concluded that there was a need for further clarification of the effects of NAbs on the efficacy of treatment. The same methods of longitudinal analyses applied to the data obtained with the MxA assay in the European Study of IFN beta-1b in S P M S [ 8 0 ] indicated a significant increase in relapse rate associated with the switch from NAb negative to NAb positive status; no such effect was seen in the EDSS scores. In an effort to simplify data analyses and interpretation of results, we have only broached the topic of reversion to negative and we will now address this more thoroughly. The definitions and criteria for NAb positivity set forth in the IFN beta-1b final reports have gone unquestioned until recently. Reversion to negative arises when NAb positive patients are reclassified as NAb negative when any subsequent serum titre falls below 20 NU/ml (the positive cutoff). In some cases, reversion occurred several times during the course of therapy; such multiple reversions can occur when a patient who has reverted to negative becomes positive again. During the IFN beta-1 b study, it was found that 60% of the NAb positive patients reverted to negative.^81 In the longitudinal analysis by Petkau and White, using the AVA results, it was found that 51% (22/43) of patients in the high dose treatment arm reverted to negative at least once, and 36% (8/22) of reverting patients exhibited multiple reversions.11081 Results discussed at an International Workshop on Interferon Antibodies (New York, NY, 22 1997) [ 1 0 9 ] showed similar variability in reversion responses. McFarland reported that in an ongoing MRI study of 33 patients treated with IFN beta-1 b (Betaseron®) at the NIH where 11/33 (33%) of treated patients developed NAbs, 4/11 (36%) reverted to negative by the end of 2 years'1091. Similarly in a 2 year study using IFN beta-1a (Rebif®), Pozzilli reported a 28% (17/62) incidence of NAbs (based on an NAb positivity definition requiring only one positive titre) with reversion occurring in 65% (11/17) of them. A 100% reversion rate has been reported by Rice in the long-term (> 5 years) follow-up of NAb positive patients from the Western Ontario cohort from the IFN beta-1 b (Betaseron®) pivotal trial. [95 ] We have performed long-term follow-up of the Vancouver cohort and find that after 7 to 9 years of treatment, 11 of the 14 patients who had been positive during the pivotal Betaseron® trial are now negative, both by ELISA and MxA assay. [ 9 6 ] Suggestions have been made [ 9 8 ] that a similar trend is appearing with patients treated with beta-1 a (Rebif®) and that this effect appears to be dose dependent. The Rebif® report also suggested that a threshold in NAb levels exists and that most of the patients with low titres (< 20 units) revert which is not the case for patients with highest titres (>500 units). No increase in the exacerbation count (the primary end-point) was found in the NAb positive patients. Reversal of NAb positivity has not been reported for the Avonex® study.[66] Implications of Antibodies in Treatment At this point in time there are too many inconsistencies to conclude that the measurement of serum antibody titres is clinically relevant; still the issue of the influence of antibodies on treatment effect begs for an answer. We think further studies are warranted, bearing in mind the many confounding factors. As mentioned above, these factors involve many aspects associated with drug dosing, test sampling, sensitivities and specificities of detection methods and immunogenetic responses of the patients being studied, particularly as they pertain to the IgG mediated immune mechanisms which contribute to antibody response. "Short-term" vs. "Long-term" Treatment 23 Treatment duration is one such variables. In a preliminary report from the IFN beta-1a (Rebif®) trials,1701 it was found that none of 403 patients participating in "short term" trials (where treatment did not exceed 4 months) exhibited NAbs. On the other hand, analyses of the "long term trials" (> 6 months) showed NAbs detectable in 48 of 390 (12.3%) patients. It is interesting to note that all 48 of these NAb positive patients came from the MS group. The incidence of NAbs in these MS patients was estimated to be 14.4% (48 of 334). We conclude that long-term treatment with IFN beta was indicative of a greater probability of NAb development and MS patients are more prone to develop them. Rice et a l [ 9 5 , 9 7 ] and Oger et a l [ 9 6 ' 9 7 ] have reviewed patients from the pivotal Berlex study who had been treated from 5 to 9 years and found that most (22/25) of the previously positive patients had reverted to negative both by MxA assay and by ELISA. Not all patients could however be contacted and the cohort was therefore assessed incompletely. An evaluation by Larocca et al . [ 7 6 ] of 471 patients receiving IFN beta-1b therapy for a variety of diseases, noted that treatment duration seemed to be a major factor. In fact, for patients receiving subcutaneous injections, the overall mean time on study was 406 days for the NAb positive patients versus 188 days for NAb negative patients. Assessment of the assay methods used, showed that on average, total binding antibodies, as measured by ELISA were detectable at 50 days after treatment initiation versus neutralizing activity as measured by AVA, which started to be detected significantly later at 159 days. Selecting Assay Parameters As mentioned above, the assay system and the definition for positivity are essential parameters to consider. Criteria for NAb positivity was originally defined based on the maximum sensitivity of the AVA assay (at 20 NU/ml); however as new assays are developed, the question of which cut-off best reflects greater sensitivity and clinical significance needs to be answered. The empirical decision to use the 20 NU/ml as a cutoff point did not rest on significance. We 24 independently re-tested sera obtained from the Vancouver cohort during the Betaseron pivotal trial using an ELISA detection technique, and our findings suggest that the cutoff point for NAb positivity of 20 NU/ml may have been too low. In a study to evaluate the sensitivity and specificity of a new ELISA technique for the detection of anti-IFN beta antibodies [ 9 6 ] we compared results of 289 samples assayed by both MxA assay and this new ELISA. Results from these comparisons indicated a high false negative rate by ELISA. This finding was unexpected as ELISA is the more sensitive test. These false negative samples showed MxA assay titres less than or equal to 184 NU/mL and 95% (34/36) of the titres fell below 66 NU/ml. These results suggest that in some cases NAb titres may be due to the inhibition of MxA protein by other serum factors which may not be antibodies. Clinical significance can only be addressed if one can establish a correlation between antibody titre and clinical response. A cut-off point for the assay must first be re-defined. It should provide a reliable indication of impending reduction of therapeutic effect. Several studies have used the terms high and low titres, based on secondary cut-offs created for the purpose of distinguishing between clinically irrelevant and potentially relevant titres. With so many methods of interpretation and with intra-assay and inter-laboratory variability, it is not surprising that so many reports on NAb significance are inconclusive. During the Betaseron pivotal trial, total binding antibody assays by ELISA were found to show an extremely high sensitivity; 97% of the treatment group were found to be antibody positive.1781 ELISA was regarded as an ineffective tool for determining the incidence of 'significant' antibody levels, particularly when compared to the AVA and MxA assay. However, in the past, total binding antibodies have shown good correlation with loss of effectiveness for IFN alpha.[Table 1.1] ELISA is indeed far too sensitive with a 97% positivity rate, but titres might be more reflective of potential clinical significance. We suggest that low levels of anti-IFN antibodies are probably clinically irrelevant but that high levels probably aggregate or inactivate the IFN beta molecule. This hypothesis however does not take into account the specificity that the route of 25 injection imparts on the antigen-antibody reaction in vivo. We expect i.v. injected IFN to be rapidly bound and inactivated by antibodies. Evidence has been presented that the presence of antibodies may lengthen the circulating time of the IFN. [ 1 1 0 ] When IFN is injected s.c, however, its pharmacology may be greatly modified. It is not directly measurable in serum, even in the absence of antibodies, and its action may be through a secondary local messenger (possibly a cytokine secreted by dermal cells). If antibodies precipitate IFN beta at the skin level, this might also result in deposition of antigen-antibody complexes which end up being transported by macrophages and stellar cells in the skin, either to be destroyed or to generate a "depot" effect. The antibodies that are generated in response to IFN beta are heterogeneous, with differing affinities and avidities and change in the affinities of antibodies is known to occur as time passes resulting in clinically relevant correlations.'1111 The characteristics of the antigen-antibody interaction are often dependent on the epitope recognized. The contrast between functional and general antibodies is probably dependent on the epitope involved. Functional binding will affect the ability of the IFN molecule to interact with its target cell. These "functional" antibodies can inhibit function or neutralize activity by directly blocking IFN and receptor interaction, inducing a conformational change such that the active site of the IFN molecule is no longer recognized by the receptor. General binding antibodies may or may not exhibit a direct effect on the function of the IFN molecule, however, they most probably alter the bioavailability of the IFN molecule (formation of immune complexes, opsonization, change in antigen presentation) These factors are probably most relevant when the antigen is delivered by a sub-cutaneous route. The Anti-Insulin Antibody Experience Review of similar, well-established therapeutic systems has provided no greater insight into the role that NAbs may play in the therapeutic efficacy of IFN treatment. The treatment of diabetes with insulin is one such well studied system. Unlike the IFNs, no difference has been made among anti-insulin antibodies between total binding and neutralizing antibodies; indeed all reports describe antibodies detected and measured using enzymatic or radiometric methods. Since their 26 initial discovery by Banting et a l [ 2 6 ] in 1938, anti-insulin antibodies have been carefully studied to determine their influence on the pharmacokinetics of insulin treatment. To date it is generally accepted that high titres of these antibodies can cause insulin resistance 1 1 1 2 , 1 1 3 1 however, the effect of low to moderate titre antibodies remains controversial. Several studies have suggested that these low to moderate titre antibodies may prolong the effect of insulin by acting as a buffer, gradually releasing "free" insulin into circulation and in effect prolonging the half-life of insulin.'1 1 4"1 1 6 1 The clinical effect produced by these antibodies is affected by the relative affinities, association and dissociation constants of both receptors and antibodies for the insulin. Hypoglycemia is only seen when the equilibrium reaction favours dissociation of insulin from the antibodies for the higher affinity interaction of insulin with its receptors. The relevance of this phenomenon is limited by the fact that all patients receiving commercial insulin develop antibodies and only in a small number of cases will the severity of the hypoglycemia make aggressive treatment necessary.'117"119] Remarks Despite the overwhelming attention that antibodies have received in the past few years, particularly with the focus created after many of the pivotal trials were completed, we are still left with many unanswered questions concerning antibody relevance. Due to the nature of many of these pivotal trials, longitudinal analyses of the type of data generated has been complex and have made it difficult to come to any definite conclusions regarding antibody significance. The possibility that antibodies may have an adverse effect on treatment is a concern which needs to be allayed. The studies have readily published information pertaining to the incidence of antibody formation and even suggestions regarding which assay methods are best for measuring antibodies. Even a set of guidelines concerning antibody testing and interpretation of results has been published based on the assumption that antibodies may have some significance.'1201 Despite the apparent abundance of information and data already available further work is still required before we can definitively say whether or not antibodies are 27 clinically significant. Chapter 1: Enzyme-Linked Immunosorbent Assay (ELISA) Materials and Methods Introduction Immunoassays are used for the quantitation and characterization of the immunologic properties of a wide variety of analytes. It is a sensitive and specific technique dependent on the antigen-antibody reaction for the detection and measurement of these analytes. Its versatility stems from the fact that many of these analytes are antibodies or antigens, and thus high specificity and affinity assay antibodies can be generated against them. The first generation of immunoassays used radioisotopic labels, however since then the properties of a variety of non-isotopic labels have been tested, including enzyme labels. Enzyme immunoassays (ElA's) have found widespread application since their introduction.1121'1221 Enzymes are capable of functioning as labels due to their catalytic properties which allow for the detection and quantitation of even small levels of labelled antibodies or antigens. Numerous EIA techniques have been developed, however common to all is the interaction between antibody and ligand, which in a quantitative fashion, directly affects the activity of the enzyme. This assay system can be broadly divided into two classes; heterogeneous and homogeneous assays. In the former, a labelled antigen or antibody reacts in one or more ways with an analyte and other antibody/antigen reactants. The concentration, presence or absence of an analyte is determined by separating and measuring the fraction of the label which has become bound to a solid phase. A homogeneous assay is defined as a system in which both the antigen-antibody reaction and the measurement of the extent of the reaction are performed in solution without the separation of the free and antibody-bound components. In this case, the extent of the antigen-antibody reaction can be determined by alterations in enzyme activity as seen by the signal of the marker molecules. The enzyme-linked immunosorbent assay (ELISA) is an example of the homogeneous assay system. Antibodies binding to IFN beta-1b are detected and quantified using a non-competitive, antigen 29 coated plate, indirect ELISA. Serum, obtained from both patients and healthy controls is diluted and incubated with IFN beta-1 b coated 96 well microtitre plates. Bound IFN beta-1 b antibodies are detected using goat anti-human IgG(gamma) conjugated horseradish peroxidase (HRP), followed by the addition of o-phenylenediamine dihydrochloride (OPD)/ hydrogen peroxide substrate. Results are obtained via spectrophotometric analysis and are presented in units of optical density (OD). Preparation of Antigen The antigen used to coat plates consists of purified, therapeutic grade IFN beta-1 b (Betaseron®, Berlex Laboratories, Richmond CA). The antigen is supplied in single use vials as a lyophilized protein (~ 18.5 kDa) containing 0.3 mg (9.6 million IU) of IFN beta-1 b, as well as 15 mg each of Dextrose USP and Albumin Human USP (~ 68 kDa) as stabilizers. Each vial is reconstituted with 0.9 % saline to produce a stock solution with a concentration of 40 jig/mL of IFN beta-1 b, which is then stored at -70° C in ~1 mL aliquots. Immobilization of Antigen Ninety-six well, flat bottomed, high binding E.I.A. microtitre plates (Costar3590, Cambridge, MA) are coated with 110 uL/well of a 10 |ig/mL solution of antigen. This 10 u,g/mL solution of IFN beta-1 b is prepared using 1 mL of the 40 ug/mL IFN beta-1 b (stock solution) to every 3 mL of 0.1 M carbonate-bicarbonate coating buffer, pH 9.6. The coated plates are then incubated overnight at 4°C, in a humidified chamber, to facilitate passive adsorption of the IFN beta-1 b to the plastic surface. Washing Protocol Washing the plates is carried out manually using a washing buffer of 0.5 M Tris/HCI pH 8.0 (Sigma Biosciences, St. Louis, MO) containing a non-ionic detergent, 0.05% Tween 20 (VWR Scientific of Canada Ltd., Toronto, ON.). Washing is accomplished by emptying each well; filling the well with washing buffer and then removing the washing fluid from each well. This cycle is 30 repeated three times, followed by one cycle using dH 2 0 and the wells are emptied at the end of the final wash cycle. Test Serum Serum samples used for ELISA are obtained from the Multiple Sclerosis Clinical Trials serum bank in Dr. Oger's Lab. Serum banking was maintained over the duration of the study from patients participating in the Betaseron Pivotal Trial and various non-MS healthy donors. Test samples are diluted 1 in 400 using 0.5 M Tris/HCI buffer pH 8.0 with 1% normal human pooled sera (nhps) dilution buffer. 100 uL of each sample dilution is added in triplicate to the coated plate; leaving the first column as a blank control and the bottom three rows blank for the positive and negative controls. A positive curve is made from serial dilutions, 1 in 50 to 1 in 12800, of pooled positive antisera. The positive curve is made up of pooled sera from three antibody positive patients as determined independently using ELISA and AVA by Berlex Laboratories Inc. during the pivotal trial. The positive curve, along with two negative controls are added in triplicate to each plate being assayed. The plates are then incubated at room temperature (22° C) for 1 hour and then washed prior to adding the enzyme-conjugate. Enzyme-Substrate System The enzyme antibody-conjugate system used in these ELISA's consists of goat anti-human IgG(gamma) conjugated horseradish peroxidase (GibcoBRL, Life Technologies, Gaithersbeurg, MD). The HRP-conjugate is provided as 1.0 mg of lyophilized protein, with goat serum as a stabilizer. It is rehydrated using 1 mL of 50% glycerol solution to give a concentration of 1.0 mg/mL. The enzyme is diluted to a stock solution of 1 in 10 using PBS (no azide) and stored in 150 uL aliquots at -70° C until needed. During the incubation of the plates with sera, the enzyme is diluted with 0.5 M Tris/HCI buffer pH 8.0 to give a lot-specific working concentration of 1 in 4000 (0.025 ug/mL). 110 uL of enzyme-conjugate is then added to each well, and the plates are incubated at room temperature (22° C) for 1 hour in the absence of light. The plates are then 31 washed prior to the addition of substrate. Ort/7o-phenylenediamine dihydrochloride (OPD) is the chromogenic substrate most often used in ELISA procedures which utilize HRP conjugates. It produces a soluble end product that can be read spectrophotometrically at 450 nm. During the incubation of the enzyme-conjugate, OPD (Sigma Biosciences, St. Louis, MO) is prepared in 0.05 M citrate-phosphate buffer, pH 5.0 to a final concentration of 0.4 mg/mL OPD. Immediately prior to use, 30% hydrogen peroxide (H202) (Fisher Scientific, Fair Lawn, NJ) is added to the substrate buffer solution to give a final concentration of 0.4 uUmL H 20 2. 110 u l of the substrate solution is then added to the washed plates, and allowed to incubate at room temperature (22° C) for exactly 20 min to yield optimal colour development. The optical density (OD) is then assessed on an MR600 Microplate Reader (Dynatech, Alexandria, VA), blanked against the first row of empty wells, using a 450 nm filter. Sample data is analyzed by calculating the mean value from the OD readings of each set of triplicate wells. A sample is considered positive for total binding antibodies if the mean OD value is greater than twice the mean value of the negative controls. ELISA Setup and Optimization Since ELISA's were first described in the early seventies 1 1 2 1" 1 2 3 1 they have replaced more conventional systems, such as radioimmunoassay due to their ease of use and their greater sensitivity and specificity. In principle the ELISA uses antibodies or antigens that are covalently coupled to an enzyme, so that bound enzyme activity is measured. There are many different ways in which the ELISA may be configured, however the choice often depends on the nature of the sample (antibody or antigen) to be measured, availability of reagents and the precision and sensitivity desired (quantitative or qualitative). The indirect, non-competitive sandwich antibody method is the simplest way to detect and measure specific antibodies in unknown serum. Antigen is non-covalently attached to the solid phase support, usually the wells of a plastic microtitre plate. Excess free antigen is washed off, and the wells are incubated with a blocking solution to prevent the binding of non-specific proteins. The test serum is then added to each 32 coated well, and any antibody that is present in the serum binds to the solid phase antigen. The plates are washed to remove unbound antibody and then the enzyme-immunoglobulin conjugate is added. This should bind to any specific antibody attached to the solid phase. The solid phase is washed again, to remove any unbound enzyme-immunoglobulin conjugate and a colorimetric substrate is added, which is measured as a change in absorbance. This simple method proved to be ideal for the detection of total binding anti-IFN beta antibodies in the serum of patients receiving IFN beta-1b (Betaseron®) immunotherapy. The test principle was the antigen coated plate sandwich ELISA which entailed the immobilization of IFN beta-1b to a plastic microtitre plate. The antibodies, which were found in the patient serum, should be specific for and bind with good affinity to the antigen, as attached to the solid phase. The enzyme system of choice was an anti-human immunoglobulin gamma (IgG) - horseradish peroxidase-conjugated system that catalyzed a colorimetric substrate reaction with o-phenylenediamine hydrochloride and hydrogen peroxide. The most important factor that determines the performance of an ELISA is the extent to which it has been optimized. Therefore, it is important that the assay be set up such that each step is carefully tested and optimized to ensure accurate and reproducible results. There are numerous variations to the basic method of ELISA, each of which are dependent on the specific nature of the antigen, antibody and assay system used. Four very important assay conditions include; the pH of the various buffers and reagents used throughout the assay; the incubation times and conditions (ie. temperature, humidity, light sensitivity) associated with each phase of the assay; and the concentrations of the different components within the assay. Selection of the Solid Phase The solid phase of an ELISA is defined as the solid support material to which the antigen or antibody is bound. The solid phase can be made up of a variety of different materials that may be specially treated or left untreated depending on the potential antigen which it needs to bind. The wide variation in the properties of the solid phase are often associated with a significant 33 variation in binding capacity, therefore making it necessary to make comparisons among the various types of plates. Specially marketed ELISA plates are generally controlled for a specific binding capacity and uniformity in binding, usually regulating intra-plate variability to less than 5-10%. Three important requirements for selecting a suitable solid phase are; that the coating material actually bind the plate; having bound it should not fall off; and having bound firmly it should retain as much of its immunological activity as possible. In these studies a set of seven different ELISA or EIA plates were tested to determine their binding capacity for the antigen - IFN beta. These 96 well microtitre plates came from a variety of different manufacturers, all of which were recommended for high capacity binding of proteins. ELISA's using a healthy control, predetermined antibody negative and antibody positive sample from 3 different patient sources (Test Group 1, 2 and 3) were run in parallel on all 7 plates following the methods already described. The ELISA results which are expressed in optical density (O.D.) units were adjusted to remove background or aspecific noise. The results were analyzed by calculating the mean and standard deviation for each test sera on each plate and graphed to allow empirical evaluation. Prior to assaying, it was determined that the best solid phase support should maximize the binding capacity of the antigen, while minimizing nonspecific background binding. Sera with a relatively high predetermined IFN beta antibody titre was used to characterize the binding capacity of each of the plates. An elevated O.D. value was characteristic of increased binding of the serum antibodies to the plate-bound antigen, thus indirectly representing the extent of antigen-solid phase binding. Sera obtained from both healthy control and negative antibody (pre-treatment) patients was used to establish a value for background or non-specific binding. In terms of these criteria, the best solid phase support would be expected to exhibit a relatively high value for positive sera as compared with the healthy control (negative) sera. Results (Figure 2.1) showed only one of the plates tested to be significantly better in terms of high binding and low background; the Costar E.I.AJR.I.A. flat bottom high binding plate (Costarcorporation, Cambridge, MA). 34 Figure 2.1: Plate Comparisons 1.000 i 1 : 0.300 f 0.200 -I 1 1 1 + - + - + TEST GROUP 1 TEST GROUP 2 TEST GROUP 3 CONTROL SAMPLE —•—Gibco Nunc maxisorp — a — Corning 25805-96 —•—Corning 25801 —.— Gibco 2-69620 — A — Dynatech —A—Linbro EIA II Plus 76-181-00 —•— Costar Figure 2.1 : A comparison of 7 different EIA plates to determine the best binding capacity for IFN beta-1b. Test Group samples are from 3 treated patients. "-" denotes a pre-treatment or antibody negative samples. "+" denotes an antibody positive sample. The control sample was obtained from an untreated, non-MS healthy individual. 35 Immobilization of the Coating Antigen The antigen is passively adsorbed to the solid phase. Two main factors affect the effectiveness of this capture layer: 1) The integrity of the bound molecule and 2) the avidity by which it binds. The integrity of the bound molecule is important in assuring the stability and epitope display; while the binding avidity determines the dissociation or desorptive capacity of the antigen to the solid phase. The mechanism by which proteins bind to plastic is as yet not completely understood, however charge and hydrophobicity are known to play an important role in determining the adsorptive capability of the protein antigen to the plate. In general the polystyrene microtitre plates hold a negative charge. The charge of the antigen is dependent on the pH of the coating buffers used and can therefore be used to aid in the antigen-plate interaction. The capacity of the plastic microtitre plates to bind protein is limited. At high antigen concentrations there is a tendency for protein molecules to bind to each other because there is no space left on the surface of the plastic. These protein-protein interactions are generally weaker than those between protein and plastic and can result in dissociation of bound protein during the assay. Low antigen concentrations leave more space on the plastic surface, thus encouraging (if not blocked) aspecific binding. The optimum dilution for the capture layer, as well as for the sample layer, is the highest dilution of capture antigen which gives the highest positive to negative sample ratio (P:N) and/or the steepest dilution curve. In the case of proteins it is recommended that the coating antigen should be used at a concentration between 1 and 10 ug/mL. The best concentration to use is the one that gives the steepest dilution curve; when the slope is steep, a small change in concentration is reflected by a large change in colour (absorbance). A comparison of four IFN beta-1 b coating concentrations (Figure 2.2) ranging from 2.5 ug/mL to 10 ug/mL shows that the highest concentration tested (10 ug/mL) is the most suitable for assay conditions. These results are not unexpected due to the nature of the coating antigen, in that there is approximately a 1:14 molar ratio between the IFN 36 Figure 2.2: Comparison of Coating Concentrations 0.6 -i Serum Concentration —•—2.5mg/ml — a— 5.0 mg/ml —*—7.5 mg/ml — •— 10mg/ml Figure 2.2 : A comparison of 4 different IFN beta-1b coating concentrations. Background O.D. values have been subtracted from each data point. 37 beta-1b and human albumin. Blocking Non-Specific Activity Non-ionic detergents and low molecular weight proteins in washing and dilution buffers are used to minimise non-specific interactions in ELISAs. Wetting agents such as Tween 20 are added to washing buffers to reduce non-specific binding to the solid phase, thus reducing variability of results and increasing the P:N ratio. Small proteins such as bovine serum albumin (BSA) are used to block plates against aspecific binding of other ELISA components. The protein acts to prevent adsorption of other molecules such as serum proteins or antibodies from the enzyme-conjugate system. The high human albumin content of the IFN beta-1 b questioned the necessity of a blocking step. A comparison of plates that were blocked and unblocked was performed (Figure 2.3) and no significant difference in results was observed. Antisera and Assay Controls To determine optimal conditions for an assay it is necessary to select suitable reagents, including positive and negative samples. In any antigen-antibody reaction there is a limit to the range of concentrations which will bind effectively; hence there is a limit to the amount of antibody which can be accurately detected within a sample. At very high antibody concentrations there may be a tendency for lower affinity antibodies to bind; as well there will be an increase in the percentage of aspecific binding to both the plate and capture layer. To determine a serum concentration suitable for qualitative assay, we ran titration curves using both positive pooled and negative control sera (Figure 2.4). Two curves were generated from the serial dilution of both samples (1:50-1:12800). Ideally, the best concentration for test sera should fall within the mid-range of the slope of the titration curve. Knowing beforehand, from published results, [78] that the previous ELISA's were very insensitive, a serum dilution of 1:400 was selected. The negative controls were selected randomly from a pool of non-MS healthy control sera maintained by the Neuroimmunology Lab and were diluted to the same concentration as the test 38 Figure 2.3: Comparison of Blocked vs. Unblocked Plates 1.200 n Serum Dilutions —•—Unblocked/Positive Curve —•— Blocked/Positive Curve —•—Unblocked/Negative Curve —•— Blocked/Negative Curve Figure 2.3 : A comparison of both positive and negative standard curves with and without blocking of plates. "Blocked" plates were incubated for 1 hour with 0.1% BSA in 0.5M Tris/HCI buffer, pH 8.0 after immobilization of the IFN beta-1 b overnight; prior to the addition of test samples. 39 Figure 2.4 Serum Concentration Curves 1.200 ! 1.000 -f 0.800 --? Serum Concentration — • — Positive Curve — • — Negative Curve Figure 2.4 : An example of an antibody positive and antibody negative serum concentration curve based on the final, optimized assay conditions. 40 sera. The use of non-MS controls was based on the fact that many MS patients have naturally occurring autoantibodies against a variety of "self proteins and thus may show anti-IFN antibody positivity in the absence of IFN treatment.'911 The positive dilution used to determine optimum serum concentration, was also replicated on each test plate. No international standard for human anti-IFN antibodies existed at the time of assay. Hence these positive curves would provide not only an internal reference by which intra- and inter- assay variability could be monitored but also as a means of evaluating results in a semi-quantitative fashion. Sample dilutions were carried out using the selected Tris/HCI buffer system. Additives containing serum proteins are often added to the dilution buffer to minimize aspecific and cross-reactive binding. Normal human pooled sera (1%), fetal calf serum (FCS) and a no serum control were tested (Figure 2.5). The highest P:N ratio was obtained using normal human pooled serum. Selecting an Enzyme-Conjugate and Substrate System A large number of different enzymes have been used in immunoassays, however, by far the two most common are alkaline phosphatase (ALP) and horseradish peroxidase (HRP), both of which fulfil the basic criteria for the ideal enzyme label. They have high specific activity, are stable in solution, use substrates which are easily and precisely detected, are compatible with a wide range of sample media and assay conditions, are easily conjugated, have low non-specific binding to capture layers, are readily available and are relatively inexpensive. ALP catalyzes the hydrolysis of para-Nitrophenyl Phosphate (p-NPP). A major drawback to using ALP in an ELISA designed to detect proteins in serum, is that ALP is normally present at a low level in serum. This problem is even more significant when one takes into account that as with other drug therapies, some patients who are treated with IFN beta may have elevated liver enzymes, such as ALP, thus risking false positive results. Another common problem with ALP-ELISA systems is that the commonly used PBS buffers can act as a competitive inhibitor (due to free phosphate ions) of the enzyme and can therefore lead to reduced or variable rates of product formation. To avoid 41 Figure 2.5: Comparison of Dilution Buffers 1.000 -, Serum Dilutions — • — 1 % Normal Human Pooled Serum/Negative Curve — • — 1% Normal Human Pooled Serum/Positive Curve —-•—-1 % Fetal Calf Serum/Negative Curve — • — 1% Fetal Calf Serum/Positive Curve — @ — No Additive/Negative Curve — • — No Additive/Positive Curve Figure 2.5 : A comparison of both standard positive and negative curves produced using different serum dilution buffers. All 3 dilution buffers were made from 0.5M Tris/HCI, pH 8.0, each containing 1 of 3 additives: 1% normal human pooled serum, 1% fetal calf serum or no additive. All other assay conditions were the same between the three test groups. 42 these possible problems the HRP-OPD detector system using Tris/HCI buffers was selected. HRP is the most commonly used enzyme label in ElAs. The enzyme substrate reaction involves both a primary and secondary substrate. The HRP enzyme first reacts with hydrogen peroxide (H202) to form an oxidized compound. The chromogenic reaction occurs when the secondary substrate (OPD) becomes oxidized by reducing the H202« back to its original state. The amount of oxidized OPD substrate or chromogen is measured spectrophotometrically. The greatest disadvantage to the HRP system is the stability of the enzyme; however careful design of the assay can circumvent factors which may inactivate it. The concentration of enzyme detector needed was determined by titration of the enzyme conjugate in a series of parallel serum dilution curves (Figure 2.6 and 2.7). Enzyme was prepared in six different concentrations (1:250 - 1:8000), including the manufacturers recommended working concentration (1:1000) to determine the optimum assay concentration. The curve with the greatest P:N ratio and slope was selected (1:4000). Reading and Interpretation of Results The chromogenic reaction which occurs with the addition of substrate is read using a spectrophotometer. It is common, though not essential, to add a stopping reagent after the enzyme-substrate incubation period is finished. The stopping reagent causes a change in the pH of the reaction, thereby inhibiting further hydrolysis of the substrate. This reaction is often accompanied by an upward shift in the absorbance (from 450nm to 490nm). Due to the limitations in the dynamic range of the microplate reader used for these assays, a stopping reagent was not used. Due to the lack of international standards, quantitative evaluation could not be performed. The addition of reference curves to each plate act as an internal standard by which semi-quantitative results can be derived based on "in-house" standards or laboratory units. For the purpose of this study results were expressed only in a qualitative, positive/negative fashion. The cut-off value between positive and negative samples depended on the OD values obtained from the negative controls run on each plate. The rationale for determining the cut-off point in 43 Figure 2.6: Enzyme Concentration Comparisons 1.200 Serum Dilutions —* — 1 250- Positive Curve — • - 1 250- Negative Curve —ft— 1 500- Positive Curve — a — •1 500- Negative Curve —*— 1 1000 - Positive Curve H— -1 1000 - Negative Curve —»— 1 2000 - Positive Curve — a — 1 2000 - Negative Curve —•— 1 4000 - Positive Curve —•— •1 4000 - Negative Curve —•— 1 8000 - Positive Curve —•— 1 8000 - Negative Curve Figure 2.6 : A comparison of 6 different enzyme concentrations in both positive and negative standard curves. 6 serial dilutions of the enzyme-conjugate in 0.5M Tris/HCI buffer, pH 8.0 were tested. All other assay conditions were the same in each test group. 44 Figure 2.7: Enzyme Concentration - Comparison of 1:2000 and 1:4000 1.000 0.900 4-Serum Dilutions —«—1:2000- Negative Curve —•—1:2000- Positive Curve —•—1:4000- Negative Curve — 1:4000- Positive Curve Figure 2.7 : A comparison between 1:2000 and 1:4000 enzyme concentrations in both negative and positive standard curves. All other assay conditions were the same. 45 these assays was based on previous study results, and was confirmed by comparisons between several different cut-off criteria. 46 Chapter 2: ELISA Results and Discussion Results Pivotal Trial Study The comparisons discussed below focus primarily on the Vancouver Cohort of the Betaseron Pivotal trial. Details of the general methodology of the US/Canadian multicentre trial from 1988-1993 have been outlined in previous publications.'171 Sera from 52 patients participating in the Vancouver frequent MRI cohort of the trial was collected serially between 1988 and 1993. Interferon beta-1 b (Betaseron®) was administered subcutaneously in one of two doses: 1.6 mlU [50|xg IFN] (n=18) or 8 mlU [250u.g IFN] (n=17) every other day for up to 5 years. Serum was obtained from each patient at every 6 week visit during the trial. Sera were sent to Berlex Laboratories for antibody measurement by AVA and subsequently MxA assay. Sera were also assayed "in house" after trial termination for binding antibodies using the ELISA described above (Appendix A). The results from the AVA and MxA assay were made available to us from Berlex (Appendix B) for comparison with this new ELISA. Positivity Criteria The criteria used to establish antibody positivity within each of the NAb assay systems has been previously published in the early results from some of the clinical trial.'781 The ELISA by Berlex used during the trial was established only to screen samples for antibodies in a qualitative fashion. The cut-off point for positivity was determined by doubling the mean optical density values of a set of negative controls within each plate and/or assay. From the information already made available with regards to ELISA cut-off points, we were able to confirm, by comparison of several different cut-off points, that a cutoff of two times the mean was optimal for determining positivity. This definition allowed for the greatest number of positive treated 47 patients while minimizing the number of false positive results to be found in the placebo group. As mentioned previously, the definition for NAb positivity for both the AVA and MxA Assay per sample required an NAb titre of at least 20 NU/ml. As well, per patient antibody positivity requires the patient to have 2 consecutive serum samples >20 NU/ml. In the data obtained for the following comparisons the positive cut-off point for the neutralizing assays is 20 NU/ml, for the ELISA, twice the mean OD value of the healthy controls was used. Positivity can be defined in two ways, either on a per patient basis or on a per sample basis. The definitions for positivity in each of these cases depend on the assay type and comparison to be made. Criteria for overall patient positivity as used in the per patient analyses, requires a minimum of two consecutive positive titres for the AVA and MxA assay but only one positive sample for the ELISA. Positivity on a per sample basis evaluates single samples as either positive or negative independent of any additional samples taken from that same patient. Frequency of Antibodies in Betaseron Treated MS Patients Using the defined criteria for positivity we can show the frequency of antibody positivity as measured by the UBC ELISA vs. the other assay systems. To determine this frequency it was assumed that only treated patients can be antibody positive, and that any placebo patients with positive results are false. A summary of the results is found in Table 2.1. Table 2.1 Frequency of antibody positivity in Patients from the Betaseron Pivotal Trial as Determined by the Various Assay Techniques Frequency of Antibody Formation in Patients from the Vancouver Cohort3 Berlex ELISA UBC ELISAd AVAe MxA assay6 Treated Groups (n=35) 100% (35/35) 51% (18/35) 46% (16/35) 60% (21/35) Placebo Group (n=17) 88% (15/17) 24% (4/17) 0% (0/17) 0% (0/17) 48 Frequency of Antibody Formation in Patients from the Trial Cohort Berlex ELISA UBC ELISAd AVAe MxA assay0 Vancouver Cohorts (n=52) 96% (50/52) 42% (22/52) 31% (16/52) 40% (21/52) Study Cohort (n=372) 97% n/af 38% n/a c a results summarized from data in Appendix C t> includes patients from both the low dose (1.6mlU) and high dose (8mlU) treatment groups c results obtained from publication's] d antibody positive definition of 1 positive sample OD value at any time, "once positive always positive" e neutralizing antibody positive definition of at least 2 consecutive positive titres > 20 NU/ml, "once positive always positive" f results not available Comparison of ELISA results with AVA and MxA assay results Serum samples from the 52 patients participating in the Vancouver Cohort of the Betaseron Pivotal trial were assayed by the UBC ELISA. The occurrence of antibodies as measured by the UBC ELISA, was compared to results obtained from samples previously tested by the AVA and MXA assay systems. To compare the new ELISA to these other assays, results were analyzed using a two-by-two contingency table method for determining sensitivity, specificity, false positive and false negative values. The formulas used are found in Figure 2.8. Figure 2.8 2x2 Contingency Table "Gold Standard" (AVA/MxA assay) Positive Negative Test Assay Positive A 1 (A + B) (ELISA) Negative ' (C + D) (A + C) (B + D) Total 49 The variables are defined as follows: Sensitivity = A/(A+C) = proportion of samples testing positive by both standard and test assays Specificity = D/(B+D) = proportion of samples testing negative by both standard and test assays False negative = 1 - sensitivity = C/(A+C) = proportion of samples testing negative in the test assay but positive in the "gold standard" assay False positive = 1 - specificity = B/(B+D) = proportion of samples testing positive in the test assay but negative in the "gold standard" assay Results from the assays performed have been assessed in two ways; (i) on a per patient basis, where positivity was determined by looking at all the samples of a single patient over the entire period of study and (ii) on a per sample basis, where positivity was determined for a single sample. The standard assay used in each comparison was either the AVA or the MxA assay. The first set of comparison results are from per patient positivity between the different assays. The results are summarized in Table 2.2. Table 2.2 Summary of Antibody Results from the Comparison of the Various Assays Based on Individual Patient Positivity • Standard Assay'1 AVA MxA assay AVA "Test Assay'1 Berlex ELISA UBC ELISA Berlex ELISA UBC ELISA Sensitivity 100% 81% 100% 76% 76% Specificity 8% 75% 10% 81% 100% False Negative — 19% ~ 24% 24% False Positive 92% 25% 90% 19% — Sample Size n=52 (Results obtained from data in Appendix C) 50 The Berlex ELISA shows excellent sensitivity at 100% but has a very low specificity for NAbs when compared to the AVA or MXA assay. When compared to the MxA assay, both the UBC ELISA and the AVA exhibit a similar degree of sensitivity (76%). The UBC ELISA found 19% of patients to be positive by ELISA but negative by MxA assay ("false positive") whereas the AVA showed no samples to be positive when the MxA assay was negative. The second set of results comes from the comparison of assay data among single samples (per sample). The results are summarized in Table 2.3. Table 2.3 Summary of Antibody Results from the Comparison of the Various Assays Based on Single Sample positivity "Test Assay" Berlex ELISA UBC ELISA Berlex ELISA UBC ELISA AVA Sensitivity 100% 72% 93% 53% 59% Specificity 58% 83% 63% 84% 99% False Negative — 28% 7% 47% 41% False Positive 42% 17% 37% 16% 1% Sample Size n=569 n=285 n=565 (Results obtained from data in Appendix B) n=289 n=724 On a per sample basis, comparisons between the UBC ELISA and AVA, and the UBC ELISA and MxA assay show a marginal increase in specificity while the sensitivity for AVA-NAbs (72%) vs. MxA-NAbs (53%) is greatly decreased. Closer observation of this data subset reveals that only 36 of 289 samples (12% of all samples tested by both UBC ELISA and MxA assay) were UBC ELISA negative but MxA assay positive samples. The distribution of the mxA assay titres for both concordant (UBC ELISA positive/MxA assay positive) and discordant (UBC ELISA negative/MxA assay positive) samples showed a range of positive titres within the MxA data set from 20 to 17,080 NU/ml. Discordant samples however, had lower NAb titres ranging from 20 to 51 184 NU/ml (95% at <66 NU/ml). Thirty-four samples out of 289, were "false positive" samples (UBC ELISA positive but MxA-NAb negative). Of these 34 samples, 25 were from patients who subsequently tested positive of MxA-NAbs. Of the remaining 9 samples, 3 were from patients receiving treatment and 6 were from patients receiving placebo. These 9 reflect a new "false positive" rate of 5%. Follow-up Study: Data Gathered During Quality of Life Review (1997) After IFN beta-1b was given regulatory approval in 1993, all patients who had participated in the pivotal trial, treated (low dose, high dose) or placebo, were offered to continue treatment on the approved dose of 8 mlU (250ug)(high dose equivalent) of IFN beta-1 b eod. As a part of a follow-up study for "Quality of Life" (QoL), 37 of the original 52 patients from Vancouver cohort of the Berlex pivotal trial were contacted in the summer of 1997 (15 were lost to follow-up). At this time serum was obtained from 28 (of the 32 patients consenting to participation in the QoL) and assayed in a blinded fashion using both the UBC ELISA and MxA-NAb assays. These assays were also performed on 31 patients seen for a similar follow-up as part of the London (Ontario) cohort. The combined results from these two cohorts is reported elsewhere.'971 Reversion From the 28 patients tested for antibodies, 8 patients had received low dose (1.6 mlU), 12 had received high dose (8 mlU) and 8 had been on placebo treatment. Both the low dose and placebo groups rolled onto active treatment in 1993, and since then were treated using the approved 8 mlU s.c. eod. The mean treatment duration for these 28 patients was 6.3 years. Seventeen of the 28 patients were on active treatment at the time of the QoL study. The results are summarized in Table 2.4. 52 Table 2.4 Summary of Antibody Data Collected from Patients Participating in the Follow-up Quality of Life (QoL) Study Patients tested from QoL Study 28 Patients on active treatment during pivotal trial Patients on active treatment at time of QoL Study Antibody Results in Patients Who Received Active Treatment During the Pivotal Trial (n=20) ELISA MxA assay Sero-conversionb during the pivotal trial ;.>: 12 Sero-reversionc prior to trial termination 7 Sustained Sero-conversion at QoL (1997)d .:; ... 2y •:•:•[. Sero-reversion post-trial/pre-QoLe 5 Sero-reversion by the end of QoLf a includes patients receiving placebo at time of b defined as a change in antibody status from negative to positive c defined as a change in antibody status from positive back to negative d defined for patients for which no reversion was seen by the end of the pivotal trial or by the time of QoL follow-up e defined as the interim period after pivotal trial termination and before QoL follow-up f defined as overall reversion, taking place at any time between trial initiation and QoL follow-up Twenty of the 28 patients participating in the QoL study had received treatment during the trial, 14 (of 28) had tested positive for NAbs by MxA assay and 12 (of 28) had been positive by ELISA. Of these positive patients 10 shared positivity between both assays. Sero-reversion at the end of the QoL study was seen in 86% (12/14) of the MxA assay positive patients and in 83% (10/12) of the ELISA positive patients. Reversion occurred in 50% (7/14) of NAb positive patients at some point during the course of the pivotal trial, and in 58%) (7/12) of ELISA positive patients. Among the patients reverting in the interim of post pivotal trial and pre-QoL study, only 2 of the 5 by MxA assay and 1 of the 3 by ELISA could be attributed to discontinuation of IFN treatment. Concordance between assays in detecting reversion was 53 seen in 3 of the 7 reverted patients. When the serum obtained during the QoL study was assayed for antibodies, 3 of the 28 samples assayed were found to be positive in each assay, with 2 patients concordantly positive in both ELISA and MxA assay, 1 patient positive by ELISA and 1 patient positive by MxA assay. Sustained sero-conversion (positive between the time of pivotal rial termination and QoL Study follow-up) was seen in 2 patients. Discussion The comparisons performed were based on the assumption that the true state of antibodies in each patient could be definitively ascertained. However, our conclusions regarding the sensitivity and specificity of new assays are tainted by the lack of a true gold standard. And although in the following comparisons we use the neutralizing antibody assays as the gold standard to which we compare the ELISA, this by no means assumes its accuracy in detecting anti-IFN beta antibodies. The Berlex ELISA typifies the screening ELISAs used to detect total binding antibodies prior to testing for neutralizing activity. The resulting high false positive rate explains why early reports of anti-IFN beta antibodies stated that as many as 97% of study patients exhibit antibodies'781. With the UBC ELISA, both sensitivity and specificity are maintained; and when compared to the MxA assay, the ELISA exhibits a similar degree of sensitivity as the AVA (76%). The fact that the UBC ELISA finds 19% of patients to be positive by ELISA but negative by MxA assay ("false positive) whereas the AVA showed no samples to be positive when the MxA assay was negative is likely due to the differences in binding affinity and avidity of the antibodies being detected. Remembering that both the AVA and MxA assay detect a fraction of the total antibodies present, one would expect the ELISA, which detects all antibodies, to show positive results when the AVA and MxA assay are still negative. Hence the UBC ELISA may in fact be more specific than indicated by the comparison to either neutralizing activity assay. 54 On a per sample basis, comparisons between the UBC ELISA/AVA and the UBC ELISA/MxA show a marginal increase in specificity while the sensitivity for AVA-NAbs (72%) vs. MxA-NAbs (53%) is greatly decreased. This almost two-fold increase in detection of samples negative by ELISA but positive by MxA was unexpected, particularly when ELISAs, by their nature should detect more antibodies than neutralizing assays. Upon closer investigation of this data subset, we found that only 36 of the 289 samples (12% of all samples tested by ELISA and MxA assay) were responsible for these ELISA negative but MxA assay positive samples. We then looked at the distribution of the titres for the MxA assay positive samples; comparing between the ELISA negative/MxA assay positive (discordant) and the ELISA positive/MxA assay positive (concordant) (Figure 2.9). The range of positive titres within the Berlex MxA assay data set included samples with MxA assay titres ranging from 20 NU/ml to 17,080 NU/ml. ELISA false negative samples were found to have NAb titres of <184 NU/ml, and 95% (34/36) often fall into a titre range below 66 NU/ml. A similar comparison of the ELISA results to the Berlex AVA data set showed that the range of positive titres was between 20 NU/ml and 9,428 NU/ml (Figure 2.10). The ELISA negative/AVA positive (discordant) samples (13 of 285) had NAb titres < 236 NU/ml, 95% of which fall below 192 NU/ml. The low NAb titres exhibited by these ELISA negative/NAb positive samples suggests that this is not a sensitivity problem for the ELISA but rather an indication that the cut-off point for positivity used by the neutralizing assays is too low. This suggests that the cut-off point could be increased to >66 NU/ml for the NAb assays. Historically, investigators used the limit of detection of the AVA (20 NU/ml) as a cutoff point, which was eventually carried over to the MxA assay. This emphasizes that criteria for positivity were not based on any clinical or diagnostic relevance. Having addressed the so called false negative ELISA samples, we then went on to look at the false positive samples. A high false positive rate in the ELISA would be expected. The concept that total binding antibodies would form earlier in treatment than NAbs is generally accepted, hence the ELISA should be read as positive in some cases where neutralizing assays are read 55 Figure 2.9: Titre Distribution of MxA Assay NAb Positive Samples Chart A Chart B > 500 NU/ml 500 450 400 | 350 3 5. 300 £ j= 250 CO < 200 | 150 100 50 0 • 66 NU/ml • 500 450 400 350 300 250 200 150 100 50 0 ELISA Negative Samples ELISA Positive Samples Figure 2.10: Titre Distribution of AVA NAb Positive Samples > 500 NU/ml Chart A Chart B i i i i i i i i i i i i ELISA Negative Samples ELISA Positive Samples Figures 2.9 and 2.10: Each point represents a sample with a positive NAb titre (>20NU7ml). "Chart A" in each figure shows the distribution of ELISA negative/ NAb positive (discordant) samples. "Chart B" in each figure shows the distribution for ELISA positive/NAb positive (concordant) samples. 56 as negative (see Figure 2.11). This was confirmed by a closer look at the "false positive" sample dataset; the samples positive for the ELISA but negative for NAbs. For the MxA assay 34 of 289 samples were positive by ELISA but negative by MxA assay. Of these 34 samples, 25 came from patients who at a later time point in the study tested positive by MxA assay. Of the remaining 9 samples which came from patients who never exhibited NAbs, 3 came form treated patients and 6 were from patients who had received placebo. In light of these results we assume that the remaining 9, which reflect a false positive rate of 5%, are probably the only true false positives. Similar results were found in the AVA/ELISA comparison dataset. Forty of 285 samples were detected "false positive" by ELISA. Of these 40 samples, 27 came from patients who eventually exhibited NAbs. Thirteen of the 27 samples came from NAb negative patients, 6 of which were from the placebo treated group, which reflect a false positive rate of only 6%. The purpose of the cross-sectional study involving patients from the QoL follow-up study was to see how many patients who had been antibody positive in the past, had reverted back to negative antibody status. What we saw was that antibodies disappeared in a substantial proportion of patients. Specifically, sero-reversion was seen in >80% of patients who were on long-term active treatment and had been antibody positive some time earlier in treatment. The significance of reversion becomes important if we assume that the presence of antibodies is associated with decreased treatment efficacy and consequently antibody negative patients would have better treatment outcomes. Although the sampling population here was small, we see agreement in all but 2 samples between the UBC ELISA and MxA assay. These 2 discordant samples can probably be explained as false positives. The sample which was UBC ELISA positive but MxA assay negative came from a patient who had reverted already once in the past and was now positive again; in fact a subsequent sample from this patient, when tested, showed that she had reverted, once more, to negative. The sample which was ELISA negative but MxA assay 57 Figure 2.11: Time curve of the development of Binding vs. NAbs \ Figure 2.11: An example of antibody development in a patient who has received long term IFN beta-1 b therapy. This time curve shows that binding antibodies, as detected by the UBC ELISA, develop 3-6 months prior to NAbs, as detected by the AVA and MxA Assay. 58 positive came from a patient who had received placebo during the course of the pivotal trial, and at the time of rollover onto active drug had completed only 3 months (May 93-Jul 93) of active drug treatment. Sustained sero-conversion (positive between the time of pivotal rial termination and QoL Study follow-up) was seen in 2 patients. These 2 patients are noteworthy due to the fact that both exhibited extremely high NAb titres throughout the course of the pivotal trial. The magnitude of the MxA assay titres seen in the entire Vancouver Cohort (52 patients) ranged from 20 NU/ml up to 17,080 NU/ml. The 2 above mentioned patients accounted for the highest titres seen in this cohort. The final NAb titre at trial termination for each of these 2 patients was: 7015 NU/ml and 5832 NU/ml. The highest recorded MxA assay titre recorded was 7015 NU/ml and 17, 080 NU/ml. If the titres from these 2 patients were excluded, the highest titre attained by any of the other patients within the cohort was significantly lower at < 705 NU/ml. Although the true significance of the high titres seen in these 2 patients is not known, it is quite possible that this apparent "high titre" subset of antibody positive patients is prone to sustained and continued positivity in the long-term. The clinical significance of these high titre antibodies and sustained positivity is as yet to be determined. Much of the data collected to date has suggested that the measurement of binding antibodies may be irrelevant. However, even with two different methods of NAb detection, clinical significance of antibodies in relation to treatment effect has not been clearly demonstrated. The data collected here, with respect to the ELISA false negative results indicate that there is a need to redetermine relevant cut-off points. The clinical significance of "sustained, high titre" antibodies versus "reverting, low titre" antibodies has not yet been investigated. It is possible that characterization of antibody positive patients into one of these two response categories may provide a clearer picture of antibody relevance. 59 Conclusions In an effort to gain a greater understanding of the role that antibodies play during IFN beta treatment it is important to have a standard technique for measuring these antibodies. Herein lies the problem of interpreting the effect of antibodies during treatment. Several different assay techniques have been developed, each claiming its own relevance based on the "subset" of antibodies it can detect (ie. binding vs. neutralizing antibodies); however the hindrance in comparing each of these assays lies not only in the differing detection methods but in the simple absence of a true gold standard. Despite the growing number of clinical trials and results reported on the occurrence of anti-IFN antibodies during IFN therapy, little work has been devoted to defining a gold standard for IFN antibody testing. The most obvious factors to influence antibody production are the type of IFN preparation, the route of injection and the duration of treatment. However, differences can still be due to the lack of standardization. Without standardization clinical significance can only be inferred. By 1983, the WHO had already made recommendations regarding the AVA as a standard method for NAb detection.[99] In an effort to overcome some of the limitations of the AVA, several other assays have been developed. In 1987, Hennes et al set out to evaluate 3 different methods in use for the detection of IFN alpha-2a antibodies.11241 Comparisons were made between the EIA, RIA and AVA to determine which assay was most sensitive. Results showed a 97% agreement between the EIA and RIA results, a 93% (458/492) agreement between the RIA and AVA and a 92% (1450/1580) agreement between the EIA and AVA. Both immunoassays had a 0% false negative rate when compared to the AVA, that is to say that no samples were ever found to be negative by immunoassay and positive to AVA, therefore indicating a sensitivity of 100% and specificity of 93% for RIA and 92% for EIA. There was an r= 0.83 correlation between quantitative results of EIA and AVA. 60 In a comparative study performed by Spiegel et al, [ 5 1 ] it was suggested that the IRMA/RIA assays had equivalent sensitivity to that of the AVA. However 18% of samples from one study'501 and 23% of samples from another study when assayed by IRMA/RIA had false negative results when compared to the AVA. In another study Itri et a l [ 4 3 ] found that EIA was far more sensitive than IRMA/RIA where IRMA/RIA exhibited a 48% false negative rate when compared with the EIA. During the IFN beta studies in MS, assays for the detection of antibodies focused on the ELISA, AVA and MxA assay. The exclusive development and use of the MxA induction assay by Berlex Laboratories has made independent comparison between these three methods impossible. Interest in the development of a new assay arose as a result of questionable sensitivity and specificity of the AVA assays for the detection of IFN activity. Reproducibility was another problem with the AVA, which was subject to significant inter and intra-assay variability. It was felt that the cytopathic effect was not an effective quantitative measure of the effect of IFN and anti-IFN antibodies in cellular bioassays; hence, the need to use an endpoint other than cytopathic effect. The relatively recent finding that human intracellular MxA protein was specifically induced by type I IFNs, lead to the development of MxA specific monoclonals and an immunoassay which could detect and quantify MxA protein.'1051 To create an assay which could measure and quantify IFN activity and NAb activity would minimize the limitations seen in the AVA. Having created this assay, Berlex set out to validate it.!100'1031 Much of the serum that had been collected during the pivotal trial was available for validation of this new assay. Only the first three years of study samples had been assayed by AVA. With these preliminary AVA results a cross-section of 35 patients with a total number of 309 samples were re-tested using ELISA, AVA and MxA assay and compared to the original AVA results. 61 The patients selected were known to cover a wide range of NAb titres. Samples were blinded prior to testing. "Good" correlation was found between the AVA and MxA assay with false positive rates being low for both assays. The MxA assay was identified as the more sensitive of the two assays. The MxA assay was also tested for specificity to antibody-mediated neutralization using a subtraction style comparison, normal MxA assay results were compared with the results of MxA assay performed on protein-A scrubbed sera, where all the IgG's were removed from the sera prior to testing. Only 7 out of 309 (2%) of samples exhibited neutralization activity which was not IgG-mediated, certainly an improvement over AVA assays. Comparison with the Berlex ELISA showed a very poor correlation between NAb and total binding antibodies. This poor correlation is most likely attributed to the lack of specificity in the ELISA used. Our experience suggests that the ELISA can be improved by reducing it's sensitivity.1961 In the report on NAbs by the IFNB MS study group, total binding antibodies, as measured by ELISA or Western blot were reported to develop in 97% of treated patients (the percentage in the placebo group was not reported).'781 Our comparisons show not only that an ELISA can be both sensitive enough and yet remain specific, but they also suggest new cutoff points for the NAb assays. The antiviral assays lead the way to establishing a cut-off point of 20 NU/ml, which was eventually adopted by the MxA assay. Criteria for sample positivity was based on the limitations of the assay system used at that time, which in early studies was found to be twice the lowest limit of detection of the antiviral assay. Since then, better techniques have been applied, thus significantly lowering the limit of detection in both the AVA and MxA assays, and yet the cutoff points have not been re-evaluated or modified. Although one can reliably detect antibodies by each of the assay methods available, no clinical correlation has as yet been attributed to the presence or absence of antibodies as measured by these assays using a cutoff point of 20 NU/ml. A cutoff point must be selected based on 62 meaningful levels of systemic antibodies, and not merely on the basis that they are detectable by very sensitive assay systems. Our comparisons suggest that the cutoff should be increased to at least 66 NU/ml. It is very probable that patients receiving IFN therapy may have a certain level of background antibodies which are inconsequential to treatment efficacy. To answer this question, it is essential that a study be conducted not only with the purpose of validating new cut-off points, but to definitively establish what influence NAbs may have on the effect of therapy. 63 Chapter 3: Continuing Studies "The Berlex-Schering AG Interferon Beta Antibody Study" Designing a Study to Correlate Antibodies with Defined Clinical Outcome Measure In an active effort to provide more conclusive information we have put together a protocol aimed at specifically answering some of the questions which have arisen around antibodies and their clinical significance. This study will provide a means by which we can clinically validate the cut-off point established by the UBC ELISA. This application of the ELISA cutoff point to reality will be based on the comparison of 3 data subsets. These data subsets will be divided not only between data collected from antibody positive vs. antibody negative patients but also between data collected during scheduled clinic visits at regular intervals vs. clinic visits triggered by changes in antibody positivity of the patient. The 3 subsets include data collected on all parameters during: a1 - regular scheduled visits by antibody negative patients a2 - regular scheduled visits by antibody positive patients b - triggered visits (by antibody positive patients) Analyses will look not only for differences between groups a1 and a 2 , but also for differences which might be seen between groups a 1 + 2 and b, to glean more specific information about the immediate changes that antibody positivity (as established by the UBC ELISA cut-off) might have on patient responsiveness to treatment. Investigational Plan Trial Objectives 64 Primary Objective: To study and assess the relationship between changes in antibody status (both the appearance and disappearance of antibodies) and clinical course of the disease, including injection site reactions and surrogate markers of IFN beta effect. Secondary Objective: To identify a marker which can pre-determine and differentiate between patients who will and those who will not develop antibodies. Trial Design This study has been designed to answer a number of questions which have arisen over the years concerning the effect that antibodies might have on IFN beta treatment. Under the MS Special Therapies Programme patients choosing to start Betaseron therapy and who reside within the lower mainland (and in some cases in or near Victoria) will be offered to participate. All patients who enter the study must be willing to comply with the MS Special Therapies Programme requirements as stipulated under the MS Clinic - B.C. Pharmacare agreement. To initiate treatment, patients must have attended an Education Session and have received confirmation of eligibility for treatment as determined by the Pharmacare MS Drug Advisory Panel. Eligibility is determined by neurological evaluation by an MS Clinic neurologist based on relevant medical information pertaining to the conditions and criteria set forth by the Pharmacare MS Drug Advisory Panel. These include the following: i) a clinically definite or laboratory supported definite diagnosis of MS ii) at least 2 attacks of MS during the previous 2 years (an attack being defined as the appearance of new symptoms or worsening of old symptoms, lasting at least 24 hours in 65 the absence of fever, and preceded by stability for at least 1 month) iii) ambulatory with or without aid (EDSS < 6.5) iv) absence of any concurrent illness (affecting compliance or reduced life expectancy) v) absence of active, severe depression vi) contraindicated in women who are (planning to become) pregnant or nursing Once found eligible the patient will return to the MS Clinic to receive self-injection training from a registered nurse and start treatment (day 1). Prior to any treatment initiation, the patient will be required to complete screening bloodwork (including CBC and differential, AST, ALT, ALP, GGT and Total Bili) as per the treatment specific protocol. After treatment initiation the patient will be in contact with the RN by telephone to monitor any injection or side effect problems, followed by another visit to the MS Clinic one month later (month 1) for the purpose of clinical assessment, including management of side effects, completion of Beck's depression inventory and bloodwork. Bloodwork will continue to be monitored every 3 months by the family physician. Any additional clinical assessment is at the discretion of the family physician. However, patients are required to return to the MS Clinic 1 year later (annual review) where they will be re-evaluated by their MS Clinic neurologist to determine whether or not treatment should be discontinued according to termination criteria as outlined by the Pharmacare MS Drug Advisory Panel. Criteria for the cessation of IFN beta therapy include the following: i) treatment with at least 3 courses of immunosuppression therapy during the past year of therapy ii) active, severe depression or suicidal ideation while on therapy iii) non-compliance iv) severe drug toxicity and/or adverse events 66 v) planned or unplanned pregnancy In addition to the Programme requirements as outlined above, patients interested in participating in this study must agree to more frequent lab testing, clinical evaluations by a neurologist and registered nurse, as well as completing questionnaires and injection site reaction forms. In addition to treatment specific bloodwork (every 3 months), blood samples will be drawn on a monthly basis to evaluate the patient's IFN beta antibody status, immune cell reactivity and IFN beta intracellular biological activity. Injection site reactions will be monitored and recorded on the diary cards provided to each patient on a monthly basis. Scheduled clinical evaluations will increase from once yearly to every 6 months. 3 additional un-scheduled clinical visits, monthly over the span of 3 months, will be triggered by sero-conversion to antibody positive, and sero-reversion to antibody negative. These visits will involve the same clinical and neurological evaluations as for the scheduled yearly visits. Bloodwork during these months will be arranged to coincide with each visit. Questionnaires must be completed for every clinic visit, both scheduled and un-scheduled. In addition to these times, questionnaires will be given/sent to each patient every other month for the first 6 months (months 2,4 and 6) and every three months thereafter (months 9,12,15,18, 21 and 24). All bloodwork requisitions, injection site reaction diary cards and questionnaires not provided during clinic visits will be mailed to the patient as required to meet the study requirements. 67 Trial Population and Sample Size Patients who are part of the Berlex Canada sponsored Special Therapies-Pharmacare Programme administered by the UBC MS Clinic will be offered to participate if they can commit to the trial procedures outlined below and sign and informed consent. Based on an estimate of 30-40% conversion of treated patients to antibody positivity and the need to obtain significance, sample size has been established at 120 patients as determined by sample size calculations performed by Dr. J. Petkau. This number may be revised on the basis of antibody positive cut-off points determined from scatter analysis of the data from the Berlex Pivotal Trial in consultation with Dr. J. Petkau's group, study statisticians. Duration of Trial Time permitted for patient enrollment is approximately 1 year (November 1997-1998)with the possibility of extension to one more year of enrollment (to November 1999) at the time of an annual review by the investigators. The duration of an individual patient's participation is for 2 years. Trial Methods and Procedures Screening Phase Trial entry requirements are equivalent to eligibility criteria as outlined by the MS Drug Advisory Panel. Patients satisfying the IFN beta treatment eligibility requirements will come to the UBC MS Clinic to initiate treatment as outlined by the MS Special Therapies Programme. A brief description of the study will have been provided during the Education Session attended by the patient. Prior to treatment initiation, a chart review of each patient to confirm that all baseline information has 68 been completed by an MS Clinic neurologist (either in clinic or in private office) containing recent neurological and physical evaluations, and are transcribed into each patient's study file, for baseline data. Treatment Phase On day 1 of treatment the patient will meet with the MS Special Therapies Programme nurse clinician to be instructed in self-injection techniques. At this time, the patient's willingness to participate in the study will be determined and an Informed Consent Form for the study signed. A copy of the Informed Consent Form will be given to the patient for reference regarding the purpose of the study, procedures required, patient rights and expectations. The nurse clinician (RN) will go over this form carefully and answer any questions before asking the patient to sign this Informed Consent Form. Prior to the administration of the injection, the appropriate bloodwork will be drawn to satisfy antibody testing and immunological response marker testing. The patient will then be given a diary card to document injection site reactions, recording the size of both redness and induration for one site each month; and any relapse symptomology or concomitant medications taken throughout the following month of treatment. A package of questionnaires containing instructions, a Modified Beck's Depression Inventory, a modified Fatigue Questionnaire and a modified MS Quality of Life Indices (MSQLI) questionnaire will be provided. The patient will be required to complete these questionnaires on the same day as treatment initiation to provide a baseline reference for later data analyses. The patient will be asked to return in one month to see the RN. On month 1 of treatment the patient will meet with the RN and return the diary card and questionnaires provided during the day 1 visit. In turn a new diary card will be given to the patient to be completed during the next month. The RN, as per MS Special Therapies Programme requirements will clinically assess the patient as to progress, treatment concerns, adverse 69 events and mood. The patient will be asked to go to the laboratory to complete both the necessary bloodwork as outlined for the MS Special Therapies Programme as well as the bloodwork necessary for this study. At the end of this session, the patient will be informed that a package will arrive by mail in the ensuing months containing the necessary information and resources (blood tubes, lab requisitions, questionnaires and diary card) required to complete each month's study requirements. Patients will receive packages by mail from month 2 through 5, month 7 through 11, month 13 through 17 and month 19 through 23. These packages will contain the necessary materials required to complete all trial procedures. An appointment will be made with one of the neurologists to see each patient at month 6 of treatment. During this scheduled appointment the patient will visit with both the neurologist and the RN to complete the necessary forms and evaluations for the study. The neurologist will complete the MS follow-up form and physical examination form as per the MS Clinic-COSTAR guidelines, as well as the study neurologists global evaluation and EDSS forms. The RN will then interview the patient to obtain the patient's responses for the patient global opinion, as well as to aid in the completion of the ambulation status, RN global opinion and relapse history forms. The patient will be asked to complete the standard package of questionnaires and will be asked to go to the lab to complete the necessary bloodwork. A similar appointment will be made for the annual review, with the addition that the neurologist and RN complete all Special Therapies Pharmacare Programme annual review criteria, including reassessment for eligibility to continue treatment as set by the MS Drug Advisory Panel and determine effectiveness of treatment. The trial procedures performed for month 18 will follow the same format as those made for 70 month 6 and the appointment format made for month 24 will be the same as that made for month 12 (annual review). Laboratory values, including ELISA antibody measurements will be followed closely on a month by month basis. At one time point during the study the patient will be asked to participate in genetic marker typing by completing a separate Informed Consent Form for the DNA testing and providing some additional blood. Participation in this part of the study is optional. DNA will then be isolated and saved from a whole blood sample and tested to ascertain the genotype. If family are available and agree to be tested, an Informed Consent Form will be provided and a blood sample obtained for genotyping. All DNA samples will be coded and securely stored unless the patient requests it to be destroyed. No other laboratory will have access to these samples. In the event that a patient becomes ELISA antibody positive (sero-conversion), or later becomes ELISA antibody negative (sero-reversion), the patient will be called into the clinic for a series of 3 follow-up un-scheduled visits spanning an interval of 3 months. Some of these visits may or may not coincide with the above scheduled visits. Those un-scheduled visits spanning the time of a regular scheduled visit (ie months 6, 12, 18 or 24) will be treated as a regular scheduled visit, adhering to the same visit format as outlined above. Those visits which do not coincide with a regular scheduled visit will follow the visit format as described for the month 6 visit. This includes a visit with the neurologist, the RN, completion of questionnaires and study bloodwork (as per the bloodwork protocol outlined earlier). Patients may at any time have access to all MS Clinic and other available resources, beyond the scope of this study. Design Rationale 71 Based on the results published from the various studies, the incidence of antibody formation in MS patients being treated with IFN beta ranges from 5% to 64%. Although these antibodies have been reported to disappear in the majority of patients, there is still some concern that the therapeutic effect of the IFN beta is compromised once antibodies appear; hence it is important to determine what the possible consequences may be to antibody development. Clinical Evaluations Clinical assessment of each patient prior to, and for the duration of treatment is necessary to document any changes in the course and progression of the disease, in the exacerbation frequency, and how these may affect overall quality of life. The clinical parameters which will be followed include general examination and neurological examination as per the routine MS Clinic COSTAR databased examinations which generate EDSS and Scripps scores. In addition, some interview based questionnaires will be used to determine global evaluation of disease activity from the perspectives of the patient, RN and neurologist; relapse frequency and severity; and ambulation status. Because each patient initiating treatment will have been previously evaluated by an MS Clinic neurologist, baseline symptoms and general clinical evaluation will be established retrospectively through chart review and interview during the first month on study. These clinical follow-ups will occur more frequently than required through the MS Special Therapies Pharmacare Programme; at screen, day 1, month 1 of treatment and every 6 months thereafter for the duration of the study (up to 2 years); and also when patients show changes in their antibody status (as outlined above). Injection Site Reactions Injection site reactions are to be monitored by the patient to document the role that skin reactions 72 may play in the efficacy of Betaseron as pertaining to the presence or absence of antibodies. Measuring the local skin reaction on a monthly basis will be used to determine if a correlation between skin reaction, effect of treatment, and antibody status exists. The skin reaction is to be measured and documented by the patient for one injection each month. Patients will be asked to measure and record systematically the vertical and horizontal diameter of a single injection site, designated in each case to be on the right thigh, for both redness and induration separately; and to follow this reaction site over the next 4 weeks recording these measurements 24 hours, 1 week, 2 weeks and 3 weeks after the injection date. Questionnaires Patients will be asked to complete 3 take-home questionnaires at specific times throughout the study. These questionnaires are designed to provide additional information from the patient's perspective regarding day to day general mental and physical health. It is hoped that they will supplement the clinical outcome measure provided by the clinical evaluations. The modified Beck's depression inventory scale has been incorporated into the general MS Special Therapies program as a result of information provided from the pivotal trials suggesting an effect of IFN beta therapy (and possibly of antibodies) on mood, namely depression. The fatigue questionnaire, which has been modified to incorporate questions from both a standard Fatigue Impact Scale as well as specific questions found within the MSQLI is provided to asses the effects of Betaseron (and hence antibodies) on the fatigue known to affect some MS patients. The third and final questionnaire is an adapted form of the MSQLI. The original MSQLI is a comprehensive outcomes assessment battery designed to test health-related quality of life measures in individuals with MS. { 1 2 5 ] The MSQLI is a compilation of many different questionnaires designed to provide a way of measuring various facets of quality of life in the MS patient. The diversity of the questionnaires making up the MSQLI allows the investigator/physician to 73 customize it according study requirements. Hence, for the purposes of this study certain modules/questionnaires that were found to be repetitive in the scope of the clinical evaluations and other questionnaires of this study were omitted. The modules selected for this modified MSQLI include, Motor/Activities Questionnaire, Bowel/Bladder Control Questionnaire, Sensory/Pain Inventory Questionnaire and a Perceived Deficits Questionnaire. Laboratory Evaluations Patients will be asked to provide blood samples on a monthly basis. These laboratory visits may take place either at the outpatient laboratory of the UBC Hospital or at either MDS Metro or BC Biomedical Laboratories in the lower mainland. Laboratory samples must be sent within 24 hours of collection to the Neuroimmunology laboratory located within UBC Hospital to accommodate the necessary lab evaluations. Laboratory evaluations may be divided into two categories: i) antibody testing and ii) surrogate marker testing. The need to test for antibodies is obvious, as it is the basis for the study. Several methods of antibody testing are available, none of which have been embraced as the "gold standard" of testing. As a result, the three most common methods for antibody testing have been selected for use in this study: i) ELISA, ii) viral cytopathic assay (AVA) and iii) MxA induction assay. A Preliminary comparison of each of these methods based on data gathered during the Berlex Pivotal Trial, has been performed, and the conclusion drawn, for purposes of this study, is that the ELISA provides sufficiently reliable results, with a rapid turnaround, to support its use as the primary antibody monitoring tool by which clinical encounters in this study are prompted. Global review of the discrepancies between the NAb assays and the ELISA are based on a higher sensitivity of the ELISA. Binding antibodies are known to appear earlier than NAbs and in a majority of cases, binding antibodies are then followed by the appearance of NAbs. A comparison of the three assays in this study should provide some insight into the problem of cut-74 off point for determination of significant ELISA, AVA and MxA induction assay positivity. The neutralization assays (the AVA and the MxA induction assay) will be used retrospectively at trial termination to test patients with ELISA confirmed antibody positivity. The serum collected from each patient on a monthly basis will be tested within 1 week of receipt by ELISA to screen for changes in antibody status. Remaining serum will be aliquoted and stored frozen until sent for NAb testing. The AVA will be developed in cooperation with Viridae Clinical Sciences, and once established will be used to test all ELISA positive samples for CPE. Serum will also be sent to ImmunoDiagnostic Laboratories for MxA induction assay testing. The reason for testing NAbs retrospectively comes from the long turn-around time of these assays (presently 4-6 weeks), which would be incompatible with the intended study protocol. Other laboratory evaluations include surrogate marker testing. Despite evidence that several surrogate markers are sensitive to changes in systemic IFN concentrations and hence for the detection of IFN activity, only two of these markers have shown a potential sensitivity toward detecting changes in the overall IFN activity as triggered by changes in antibody status. These two markers include the measurement of intralymphocytic MxA protein within peripheral blood lymphocytes (PBLs) and in vitro IgG secretion also from PBLs. PBLs are isolated from heparinized whole blood received at the Neuroimmunology Laboratory 24 to 36 hours after collection. The same protein measured in the MxA induction assay, in response to in vitro stimulation of a specific cell line and inhibition of the MxA protein induction by NAbs present in patient serum, is measured in PBLs from treated patients to determine intralymphocytic MxA protein concentrations.11051 It has been found that MxA protein in circulating lymphocytes of patients increases by a factor of 1 to 30 after repeated injections of IFN beta. [ 1 2 6 ] This effect is obscured slightly (by a relative increase of 1 to 10) during MS exacerbations and steroid or 75 immunosuppression therapies. Despite this apparent confounding factor, the sensitivity of measuring changes in intralymphocytic MxA levels is many times greater than Neopterin or 2-5A oligoadenylate synthetase. Preliminary data has also shown that antibodies do not appear to influence the MxA level unless the antibody titre is high, therefore showing promise in answering the question surrounding the determination of what antibody cut-off titre is relevant. PBLs which were saved and stored frozen for Intracellular MxA level measurement will be sent to Dr. von Wussow's Laboratory for testing. In vitro IgG secretion levels by lymphocytes stimulated using pokeweed mitogen prior to treatment initiation has been found to be higher in patients destined to develop NAbs. [ 1 0 7 ] The amount of IgG secretion is relatively stable and untreated patients can be categorized into two groups: high level (high responders -HR) or low level (low responders-LR). NAb positive patients were more frequently associated with the HR group than those patients who remained NAb negative. In addition, IgG secretion following injections of IFN beta has been found to decrease. Furthermore, this effect seems to be lost when antibodies develop. The purpose of measuring IgG secretion in this patient group is to further investigate the relationship that this marker may have with any of the other markers of IFN beta activity and efficacy observed within the scope of the study. As well as providing a better understanding of the relationship between antibody conversion and IgG secretion. IgG secretion studies will be performed at the Neuroimmunology Lab, UBC. Genetic Markers In addition to looking at how specific immunologic markers may be predictive of or affected by changes in patient antibody status, genetic markers specific toward identifying patients who may be genetically predisposed to antibody development will be investigated. It is known that the immune system is critically linked to genetic control, and that susceptibility to MS is, therefore also 76 under that genetic control. It is, therefore plausible that this same immunogenetic background which influences the autoimmune process and the activity of the disease, is also involved in triggering the formation of NAbs. Genetic markers to be studied include the MHC Class II, TCR and IgG genes; this will be carried out in Dr. S. Beall's Laboratory. Remarks NAbs refer to the biological relevance of a binding antibody but biological relevance may not always correlate with clinical relevance. The definition of Nabs, if taken in context of the detection method, refers specifically to the inhibition of the anti-viral effects of IFN. This measurement does not necessarily relate to the ability of IFNs to exert anti-proliterative and immunomodulating properties, which are probably important in MS; one would like to first define how these IFNs work in MS so as to follow-up the inhibition of this function. It is not evident that by measuring NAbs in the context of the AVA and MxA assay, we are measuring the antibodies which neutralize the function truly relevant to MS therapy. It is possible that the detection of total binding antibodies may provide a better reflection of an effect related to the reduction of the therapeutical effect on the course of the disease. It is unfortunate that so few studies have looked specifically at a reduced treatment effect by antibodies. In most cases, clinical significance was determined in a retrospective and cross-sectional fashion. Despite all the studies to date, more attention to study design is required to produce reliable and consistently conclusive and comparative evidence regarding the incidence and significance of these antibodies. Only standardization of assay techniques and interpretation methods, definition of relevant cutoffs and determination of cross-reactivity, will provide a means of obtaining this information. We hope that through this study design we will ultimately be able to answer many of these 77 questions, or in the least provide enough conclusive evidence to serve as a guide for the development and design of future studies. 78 References 1 Pestka S, Langer JA. Interferons and their actions. Ann Rev Biochem 1987;56:727-77 2 Weiner HL, Hafler DA. Immunotherapy of Multiple Sclerosis. Ann Neurol 1988;23:211-22 3 Farkkila M, livanainen M. Interferon in Multiple Sclerosis. Clin Neuropharmacol 1988;11:120-9 4 Sibley WA, Bamford CR, Clark K. Clinical viral infections and Multiple Sclerosis. Lancet 1985;1:1313-5 5 Panitch HS, Bever CT, Katz E et al. Upper respiratory tract infections trigger attacks of Multiple Sclerosis in patients treated with interferon [abstract]. J Neuroimmunol 1992;35(Suppl 1):125 6 Booss J, Kim JH. Evidence for a viral etiology of Multiple Sclerosis. In Cook SD (Ed.) Handbook of Multiple Sclerosis, Marcel Dekker, New York 1990;41-61 7 Cook SD, Dowling PC. Multiple Sclerosis and viruses: an overview. Neurology 1980;30:80-91 8 Dhib-Jalbut S, McFarlin DE. Immunology of Multiple Sclerosis. Ann Allerg 1989;64:433-444 9 McFarlin DE, McFarland HF. Multiple Sclerosis. N Engl J Med 1982;307:1183-8 10 Merrill J, Jondal M, Seeley J et al. Decreased NK killing in patients with Multiple Sclerosis: an analysis at the level of the single effector cell in peripheral blood and cerebrospinal fluid in relation to the activity of the disease. Clin Exp Immunol 1982;47:419-30 11 Neighbour PA, Miller AE, Bloom BR. Interferon responses of leucocytes in multiple sclerosis. Neurology 1981;31:561-6 12 Neighbour PA, Grayzel Al, Miller AE. Endogenous and interferon-augmented natural killer cell activity of human peripheral blood mononuclear cells in vitro: studies of patients with MS, SLE and rheumatoid arthritis. Clin Exp Immunol 1982;49:11-21 13 Antel JP, Bania MB, Reder A et al. Activated suppressor cell dysfunction in progressive Multiple Sclerosis. J Immunol 1986;137:137-41 14 Hirsch RL. Defective autologous mixed lymphocyte reactivity in Multiple Sclerosis. Clin Exp Immunol 1986;64:107-113 15 Rose LM, Ginsberg AH, Rothstein TL et al. Selective loss of a subset of T helper cells in active Multiple Sclerosis. Proc NY Acad Sci USA 1985;82:7389-93 16 Selmaj K, Platerzyberk C, Rockett KA et al. Multiple Sclerosis: increased expression of interleukin-2 receptors on lymphocytes. Neurology 1986;36:1392-5 17 The IFNB Multiple Sclerosis Study Group. Interferon beta-1 b is effective in relapsing remitting Multiple Sclerosis. I Clinical results of a multicentre, randomized, double-blind, 79 placebo controlled trial. Neurology 1993;43:655-61 18 Isaacs A, Lindenmann J. Virus interference. I. The interferon. Proc R Soc Lond B 1957;147:258-67 19 Traugott U, Lebon P. Multiple sclerosis: involvement of interferons in lesion pathogenesis. Ann Neurol 1988;24:243-51 20 Traugott U, Lebon P. Demonstration of alpha, beta and gamma interferon in active chronic multiples sclerosis lesions. Ann NY Acad Sci 1988;540:309-11 21 Fierz W, Endler B, Reske K, et al. Astrocytes as antigen presenting cells. I. Induction of la antigen expression in astrocytes by T cells via immune interferon and its effect on antigen presentation. J Immunol 1985;134:3785-93 22 Ransohoff RM, Devajyothi C, Estes ML et al. Interferon-beta specifically inhibits interferon-induced class II major histocompatibility complex gene transcription in a human astrocytoma cell line. J Neuroimm 1991;33:103-12 23 Benzcur M, Pentranyl GG, Palffy G et al. Dysfunction of natural killer cells in multiple sclerosis: a possible pathogenetic factor. Clin Exp Immunol 1980;39:657-62 24 Panitch H. Interferons in multiple sclerosis. A review of the evidence. Drugs 1992:44:946-62 25 Panitch HS, Hirsch RL, Schindler J et al. Treatment of muliptle sclerosis with gamma interferon: exacerbations associated with activation of the immune system. Neurology 1987;37:1097-102 26 Banting FG, Franks WR, Gaitus S. Physiological studies in metrazol shock. VII. Anti-insulin activity of insulin-treated patients. Am J Psychiat 1938;95:562-6 27 Yalow RS, Berson SA. Immunologic aspects of insulin. Am J. Med 1961;31:882-91 28 Prout TE. The antigenicity of insulin: A review. J Chron Dis 1962;15:879-85 29 Kahn CR, Rosenthal AS. Immunologic reactions to insulin: Insulin allergy, insulin resistance, and the autoimmune insulin syndrome. Diabetes Care 1979;2:283-95 30 Moore WV, Leppert P. Role of aggregated human growth hormone (hGH) in development of antibodies to hGH. J Clin Endocrinol Metab 1980;51:691-97 31 Thorner MD, Reschke J, Chitwood J et al. Development of antibody to growth hormone-releasing factor. N Engl J Med 1985;312:994 32 Shapiro SS. Antibodies to blood coagulation factors. Clin Haematol 1979;8:207-14 33 Fireman P Fineberg SE, Galloway JA. Development of IgE antibodies to human (recombinant DNA), porcine, and bovine insulins in diabetic subjects. Diabetes Care 1982;5:119-25 34 Velcovsky HG, Federlin KF. Insulin-specific IgG and IgE antibody response in type I diabetic subjects exclusively treated with human insulin (recombinant DNA). Diabetes Care 1982;5:126-8 80 35 Vallbracht A, Treuner J, Flehmig B et al. Interferon-neutralizing antibodies in a patient treated with human fibroblast interferon. Nature 1981 ;289:496-7 36 Gutterman JU, Fine S, Quesada J et al. Recombinant leukocyte A interferon: pharmacokinetics, single-dose tolerance, and biologic effects in cancer patients. Ann Internal Med 1982;96:549-56 37 Quesada J, Rios A, Swanson D et al. Antitumor activity of recombinant-derived interferon alpha in metastatic renal cell carcinoma. J Clin Oncol 1985;11:1522-8 38 Figlin R, deKernion J, Mukamel E et al. Recombinant leukocyte A interferon (rlFN-alpha-2a) antibody development in advanced renal cell carcinoma (RCC). Proc Am Soc Clin Oncol 1986;5:222 39 Itri L, Campion M, Dennin N et al. Incidence and clinical significance of neutralizing antibodies in patients receiving recombinant interferon alpha-2a by intramuscular injection. Cancer 1987;59:668-74 40 Quesada J, Itri L and Gutterman J. Alpha interferons in hairy cell leukemia (HCL): A five year follow up in 100 patients [abstract]. J Interferon Res 1987;7:678 41 Gauci L. Management of cancer patients receiving interferon alpha-2a. Int J Cancer 1987;(Suppl 1):21-30 42 Steis RG, Smith JW II, Urba WJ et al. Resistance to recombinant interferon alpha-2a in hairy cell leukemia associated with neutralizing anti-interferon antibodies. N Engl J Med 1988;318:1409-13 43 Itri LM, Sherman MJ, Palleroni AV et al. Incidence and clinical significance of neutralizing antibodies in patients receiving recombinant interferon-alpha 2a. J Interferon Res 1989;9(Suppl 1):S9-15 44 Porres JC, Carreno V, Ruiz M et al. Interferon antibodies in patients with chronic HBV infection treated with recombinant interferon. J Hepatol 1989;8:351-7 45 Lok SF, Lai CL, Leung KY. Interferon antibodies may negate the antiviral effects of recombinant alpha-interferon treatment in patients with chronic hepatitis B virus infection. Hepatol 1990;12:1266-70 46 Thomas HC, Lok AS, Carreno V et al. Comparative study of three doses of interferon-alpha2a in chronic active hepatitis B. The International Hepatitis Trial Group. J Viral Hepatitis 1994;1:139-48 47 Giannelli G, Antonelli G, Fera G et al. Biological and clinical significance of neutralizing and binding antibodies to interferon-alpha (IFN-alpha) during therapy for chronic hepatitis C. Clin Exp Immunol 1994;97:4-9 48 Antonelli G, Giannelli G, Currenti M et al. Antibodies to interferon (IFN) in hepatitis C patients relapsing while continuing recombinant IFN-alpha 2 therapy. Clin Exp Immunol 1996;104:384-7 49 Craxi A, Di Marco V, Volpes R et al. Anti-alpha interferon antibodies after alpha interferon treatment in patients with chronic viral hepatitis. Hepatogastroenterol 1988;35:304-5 81 50 Spiegel RJ, Spicehandler JR, Jacobs SL et al. Low incidence of serum neutralizing factors in patients receiving recombinant alpha-2b interferon (Intron A). Am J Med 1986;80:223-228 51 Spiegel R. The alpha interferons: Clinical overview. Semin Oncol 1987;14:1-12 52 Leavitt R, Ratanantharathorn V, Ozer H et al. Alpha-2b interferon in the treatment of Hodgkin's disease and non-Hodgkin's lymphoma. Semin Oncol 1987;14:18-23 53 Von Wussow P, Freund M, Block B et al. Clinical significance of anti-interferon titres during interferon therapy. Lancet 1987;ii:635-6 54 Golomb H, Ratain M, Fefer A et al. Randomized study of the duration of treatment with interferon alpha-2b in patients with hairy cell leukemia. J Natl Cancer Inst 1988;80:369-73 55 Oberg K. Long-term treatment with alpha interferons of patients with malignant carcinoid tumors-the six year experience [abstract]. J Interferon Res 1988;8:S22 56 Freund M, Von Wussow P, Diedrich H et al. Recombinant human interferon (IFN) alpha-2b in chronic myelogenous leukemia: dose dependency of response and frequency of neutralizing anti-interferon antibodies. Brit J Haematol 1989;72:350-356 57 Week PK, Leventhal BG, Brand C et al. Detection and incidence of neutralizing antibodies to interferon-alpha-nl J Interferon Res 1989;9(Suppl 1):S37-43 58 Knobler RL, Panitch HS, Braheny SL et al. Systemic alpha-interferon therapy of Multiple Sclerosis. Neurology 1984;34:1273-9 59 Panitch HS. Systemic alpha-interferon in Multiple Sclerosis. Arch Neurol 1987;44:61-3 60 AUSTIMS Research Group. Interferon-alpha and transfer factor in the treatment of Multiple Sclerosis: a double-blind, placebo-controlled trial. J Neurol Neurosurg Psychiatry 1989;52:566-74 61 Camenga DL, Johnson KP, Alter M et al. Systemic recombinant alpha-2 interferon therapy in relapsing Multiple Sclerosis. Arch Neurol 1986;43:1239-46 62 Durelli L, Bongioanni MR, Cavallo R et al. Chronic systemic high-dose recombinant interferon alpha-2a reduces exacerbation rate, MRI signs of disease activity, and lymphocyte interferon gamma production in relapsing-remitting Multiple Sclerosis. Neurology 1994;44:406-13 63 Kastrukoff LF, Oger J, Hashimoto SA et al. Systemic lymphoblastoid interferon therapy in chronic progressive Multiple Sclerosis. I. Clinical and MRI evaluation. Neurology 1990;40:479-86 64 Kinnunen E, Timonen T, Pirttila T et al. Effects of recombinant alpha-2b interferon therapy in patients with progressive MS. Acta Neurol Scand 1993;87:457-60 65 Jacobs LD, Cookfair DL, Rudick RA et al. Intramuscular interferon beta-1a for disease progression in relapsing Multiple Sclerosis. Ann Neurol 1996;39:285-294. 66 Rudick RA, Simonian NA, Alam JA et al. Incidence and significance of neutralizing antibodies in interferon beta-1a in Multiple Sclerosis. Neurology 1998;50:1266-1272 82 67 Byhardt RW, Vaickus L, Witt PL et al. Recombinant human interferon-beta (rHulFN-beta) and radiation therapy for inoperable non-small cell lung cancer. J Interferon Cytokine Res 1996;16:891-902 68 Bornstein J, Pasal B, Zarfati D et al. Recombinant human interferon-beta for condylomata acuminata: a randomized, double-blind, placebo-controlled study of intralesional therapy. Int J STD and AIDS 1997;8:614-21 69 Dinsmore W, Jordan J, O'Mahony C et al. Recombinant human interferon-beta in the treatment of condylomata acuminata. Int J STD and AIDS 1997;8:622-8 70 Abdul-Ahad AK, Galazka AR, Revel M et al. Incidence of antibodies to interferon-beta in patients treated with recombinant human interferon-betala from mammalian cells. Cytokines, Cellul Molecul Ther 1997;3:27-32 71 Abdul-Ahad AK, Shah S, Galazka A. Treatment of Multiple Sclerosis with interferon beta-la. Neurology 1997;49:641 72 The PRISMS Study Group. Randomised, double-blind, placebo-controlled study of interferon beta 1a in relapsing-remitting Multiple Sclerosis. Lancet 1998;352:1498-1504 73 Antonelli G, Bagnato F, Pozzilli C, et al. Development of neutralizing antibodies in patients with relapsing-remitting multiple sclerosis treated with IFNpMa. J Interferon Cytokine Res 1998;18:345-350 74 Rinehart J, Malspeis L Young D et al. Phase l/ll trial of human recombinant beta-interferon serine in patients with renal cell carcinoma. Cancer Res 1986;46:5364-7 75 Konrad MW, Childs AL, Merigan TC et al. Assessment of the antigenic response in humans to a recombinant mutant interferon beta. J Clin Immunol 1987;7:365-75 76 Larocca AP, Leung SC, Marcus SG et al. Evaluation of neutralizing antibodies in patients treated with recombinant interferon-beta(ser). J Interferon Res 1989;9(Suppl 1):S51-60 77 Knobler RL, Greenstein Jl, Johnson KP et al. Systemic recombinant human interferon-beta treatment of relapsing-remitting Multiple Sclerosis: pilot study analysis and six-year follow-up. J Interferon Res 1993;13:333-40 78 The IFNB Multiple Sclerosis Study Group and the UBC MS/MRI Analysis Group. Neutralizing antibodies during treatment of Multiple Sclerosis with interferon beta-1 b: Experience during the first three years. Neurology 1996;47:889-94 79 Kivisakk P, Aim GV, Tian WZ et al. Neutralising and binding anti-interferon-beta-ab (IFN-beta-1b) antibodies during IFN-beta-1b treatment of Multiple Sclerosis. Multiple Sclerosis 1997;3:184-90 80 European Study Group on Interferon 3-1 b in secondary progressive MS. Placebo-controlled multicentre randomised trial of interferon B-1 b in treatment of secondary progressive multiple sclerosis. Lancet 1998;352:1491-7 81 Fierlbeck G, Schreiner T, Schaber B et al. Neutralizing interferon-beta antibodies in melanoma patients treated with recombinant and natural interferon beta. Cancer Immunol Immunother 1994;39:263-8 83 82 Niijima T, Umeda T, Kuriyama M et al. Intravesical treatment of bladder cancer with recombinant human interferon-beta: Intravesical GKT-beta Chemotherapy Research Group. Cancer Immunol Immunother 1989;30:81-5 83 Dummer R, Muller W, Nestle F et al. Formation of neutralizing antibodies against natural interferon-beta, but not against recombinant interferon-gamma during adjuvant therapy for high-risk malignant melanoma patients. Cancer 1991 ;67:2300-4 84 Dore-Duffy P, Donaldson JO, Rothman BL et al. Antinuclear antibodies in multiples sclerosis. Muliple Sclerosis 1982;39:504-6 85 Schuller E, Delasnerie N, Lebon P. DNA and RNA antibodies in serum and CSF of Multiple Sclerosis and subacute sclerosing panencephalitis patients. J Neurol Sci 1978;37:31-6 86 Hayes EC, Gollovin SD, Machamer CE et al. Measles-specific antibodies in sera and cerebrospinal fluids of patients with Multiple Sclerosis. Infect Immunity 1980;27:1033-7 87 Shorr J, Rostrom B, Link H. Antibodies to viral and non-viral antigens in subacute sclerosing panencephalitis and Multiple Sclerosis demonstrated by thin-layer polyacrylamide gel isoelectric focusing, antigen immunofixation and autoradiography. J Neurol Sci 1981;49:99-108 88 Salmi A, Reunanen M, llonen J et al. Intrathecal antibody synthesis to virus antigens in Multiple Sclerosis. Clin Exp Immunol 1983;52:241-9 89 Larsen PD, Bloomer LC, Bray PF. Epstein-Barr nuclear antigen and viral capsid antigen antibody titers in Multiple Sclerosis. Neurology 1985;35:435-8 90 Shirodaria PV, Haire M, Fleming E et al. Viral antibody titers. Comparison in patients with Multiple Sclerosis and rheumatoid arthritis. Arch Neurol 1987;44:1237-41 91 Oger J, Roos R, Antel JP. Immunology of Multiple Sclerosis. In "Neurological Clinics of North America". 1983; 1(3): 655-679. 92 Paty DW, Li DKB, the UBC MS/MRI Study Group and the IFNB Multiple Sclerosis Study Group. Interferon beta-1b is effective in relapsing-remitting Multiple Sclerosis. II. MRI analysis results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology 1993;43:662-7 93 The IFNB Muliple Sclerosis Study Group and the University of Birtish Columbia MS/MRI Analysis Group. Interferon beta-1b in the treatment of muliple sclerosis: Final outcome of the randomized controlled trial. Neurology 1995;45:1277-85 94 Simon JH, Jacobs LD, Campion M et al. Magnetic resonance studies of intramuscular interferon beta-1a for relapsing Multiple Sclerosis. Ann Neurol 1998;43:79-87 95 Rice G, Lesaux J and Ebers G. The evolution of neutralizing antibodies in patients taking beta interferon 1b [abstract]. Muliple Sclersos 1997;3:344 96 Oger J, Paszner B, Paty DW. Comparison of ELISA with MxA assay for the detection of antibodies to interferon-beta and results of long-term follow-up. American Academy of Neurology [abstract]. Neurology 1998;50:A342 97 Rice G, Paszner B, Oger, et al. The evolution of neutralizing antibodies in patients treated 84 with interferon. Neurology 1999;52:1277-9 98 Ebers GC, Oger J, Paty DW et al. The MS PRISMS Study: Prevention of Relapses and Disability by Interferon Beta 1a Subcutaneously in Multiple Sclerosis [abstract]. Ann Neurol 1997;42:986 99 WHO Expert Committee on biological standardization. Thirty-fifth report. WHO Technical Report Series 725. Geneva:WHO Organization, 1985 100 Pungor Jr E, Files JG, Gabe JD et al. A novel bioassay for the determination of neutralizing antibodies to interferon beta-1b. J Interferon Cytokine Res 1998;18:1025-30 101 Pachner AR. Anticytokine antibodies in beta interferon-treated MS patients and the need for testing: Plight of the practicing neurologist. Neurology 1997;49:647-50 102 Protzman WP, Jacobs SL, Minnicozzi M et al. A radioimmunoligc technique to screen for antibodies to alpha-2 interferon. J Immunol Meth 1984;75:317-23 103 Nestaas E, Files JG, Nelson JW et al. Quantitation and characterization of Multiple Sclerosis patient antibodies to interferon-beta. In Reder AT (Ed.) Interferon therapy of Multiple Sclerosis, Marcel Dekker, New York 1997;523-30 104 Kawade Y. Quantitation of neutralization of interferon by antibody. Meth Enz 1986;119:558-73 105 Towbin H, Schmitz A, Jakschies D et al. A whole blood immunoassay for the interferon-inducible human Mx protein. J Interferon Res 1992;12:67-74 106 Moosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Immunol Methods 1983;65:55-63 107 Oger J, Vorobeychick G, Al-Fahim A et al. Neutralizing antibodies in Betaseron treated MS patients and in vitro immune function before treatment. American Academy of Neurology [abstract]. Neurology 1997;48:A80 108 Petkau J, White R. Neutralizing antibodies and the efficacy of interferon beta-1b in relapsing-remitting Multiple Sclerosis [abstract]. Multiple Sclerosis 1997;3:402 109 Arnason BGW. Neutralizing antibodies: are they an issue? The Int MS J 1997;4:40-42 110 Finkelman FD, Madden KB, Morris SC et al. Anti-cytokine antibodies as carrier proteins. Prolongation of in vivo effects of exogenous cytokines by injection of cytokine-anticytokine antibody complexes. J Immunol 1993;151:1235-44 111 Tindall RSA. Humoral immunity in myasthenia gravis: clinical correlations of anti-receptor antibody avidity and titer. Ann NY Acad Sci 1981;377:316-31 112 Kurtz AB, Nabarro JDN. Circulating insulin-binding antibodies. Diabetologia 1980; 19:329-34 113 Reeves WG, Allen BR, Tattersall RB. Insulin-induced lipoatrophy: evidence for an immune pathogenesis. Br Med J 1980;280:1500-3 114 Dixon K, Exon PD, Malins JM. Insulin antibodies and the control of diabetes. Quart J Med 85 1975144:543-53 115 Vaughan NJA, Matthews JA, Kurtz AB et al. The bioavailability of circulating antibody-bound insulin following insulin withdrawal in type I (insulin-dependent) diabetes. Diabetologia 1983;24:355-8 116 Dozio N, Scavini M, Beretta A et al.. Imaging of the buffering effect of insulin antibodies in the autoimmune hypoglycemic syndrome. J Clin Endocrin Metab 1998;83:643-8 117 Anderson JH, Blackard WG, Goldman J et al. Diabetes and hypoglycemia due to insulin antibodies. Am J Med 1978;64:868-73 118 Berson SA, Yalow RS. Quantitative aspects of the reaction between insulin and insulin-binding antibody. J Clin Invest 1959;38:1996-2016 119 Berson SA, Yalow RS, Bauman A et al. Insulin 13H metabolism in human subjects: demonstration of insulin binding globulin in the circulation of insulin-treated subjects. J Clin Invest 1956;35:170-90 120 Paty DW, Goodkin D, Thompson A et al. Guidelines for physicians with patients on IFN beta-1 b: the use of an assay for neutralizing antibodies (NAB). Neurology 1996;47:865-6 121 Engvall E, Perlmann P. Enzyme-linked immunoadsorbent assay (ELISA). Quantitative assay of immunoglobulin G. Immunochem 1971;8:871-4 122 Van Weemen BK, Schuurs A. Immunoassay using antigen enzyme conjugates. FEBS Lett. 1971;15:232-5 123 Voller A, Bartlett A, Bidwell DE. Enzyme immunoassay with special reference to ELISA technique. J Clin Pathol 1978;31:507-20 124 Hennes U, Jucker W, Fischer EA et al. The detection of antibodies to recombinant interferon alfa-2a in human serum. J Biol Stand 1987;15:231-44 125 Ritvo PG, Fisher JS, Miller D et al. Multiple Sclerosis Quality of Life Inventory: A user's manual. National Multiple Sclerosis Society, New York 1997 126 Jackshies D, Hochkeppel H, Horisberger M et al. Emergence and decay of the human MX-homolog in mononuclear cells of cancer patients during and after IFN-therapy. J Biol Response Modifiers 1990;9:305-12 86 APPENDIX A SUMMARY OF UBC ELISA ASSAY RESULTS 3= o J O ' - ^ ' - ^ ' - ' - N N T-(Mrr-T-T-OCM o o o o o o o o o o o o o o o o E w n o 2 S s § o Q O O P o P q o 0 P 0 P P 0 P P O) CO m o o o g cococoopcncpcnoi Q- Q S 5 0 U : < 0 7 E p S r ^ i ^ c D ^ c 1 ) ^ • f 06-X>O-CZ 68-PO-tO QO pejsnfpv 9 9 • ^ 0 . 0 0 0 0 0 0 0 0 0 0 ooococoontDcou) S o o o o o o o o o o 0 0 0 0 0 0 0 0 0 0 S n*00)0)NtM(J)Om g y o o o o o o o o o o S O o p p p p p o p o o COCOO)0)00)00 COCOCO0pCOCOCT)<J) 9 Q U _ 5 < Z S 7 •«-OCN(NJ*-0*-0 O "? E6-|nr-0Z .•PO-61 |ui|go M-AON-ZO 68-tejfl-ej • 88<jaa-90 CO (O CM CM T- T-;88W-8t 9 <? QO pejsnfpv 9= 2 S o 1 ^ 0 . 0 0 0 0 0 0 0 0 0 0 • co o o o o o o o o o o a > o 0 0 . 9 0 . o o o o o O » 9 9 9 9 0 0 0 0 0 9 a. Z I L S < Z 5 7 < Z 7 CO E6-|nr-9Z CO CO CN CM ' ( T O p s j s n f p v a. 4 g o C N C N C N C M C S I C N C M C N C N C N C N s C L o d d o d d o d d d d Q 0 * - 0 » - * - 0 ' - * - * - * - * -Jn OOOOOOOOOOO d d d d d d d d d d d -DO S O » 3 < C0C0COCO*-C0C\|^ 'T-*-CO OOOOOOOOOOO 0 0 0 0 0 0 0 0 0 0 0 o a n o i c i o i o o a o n w c o c y e o e o c b c n a J c D 0 ? 0 ? o n ™ 3 0 o jg a. 3 a> o »Q<-;Arir~oSjbgui ),0 OOCMOO*~OOOcMCN •S-LO nr-90 68-J<jW-t.Z DOIA»r» bU 1co co CN CM Q O p e » s n f p v 8 g ? 9 9 89 ^ f J - d d d o o o d d o o o Q •^rsji-OOOT-CNO -^T-co OOOOOOOOOOO OOOOOOOOOOO * O O T - T - O ^ O I C O O J I O O Q O • n Q i o c s m i o N r - s s i S i n j S o o o o o o o o o o o ^OOOOOOOOOOO 0)0>c3)0)a>oooooco C in C O) > J» >* 0> T< (J) >« fflS 3 3 O J J | D O Jj J '-i-0(M(\|i-^ONO(M CO CO CN CM 8 g •Q'O P9»snfpv 3= o £ 0 . 0 0 0 0 0 0 0 0 0 0 a i f S ' - c o o a i m o j s O'-T-WO'-rMCjT-T-OOOOOOOOOO d d d d d d d d d d g o lO U) O CN T - CN _ _ _ _ _ P O O O d d d d d d o d d d ^ _ O) h- o r-TJQ ID N i- r- _ Q> Q O O O O O Q c o c o o p t x j o o c o o a ) - I » 5 - i o $ I § a. z - ^ ^ ^ O a S - ^ E d jCNi^syS^oor - -co - — CO 1-r-T-OOcNOOO . 8 co a CO • T 06-oea-90 rt06|des-K) 06-unr-ZO \ I [ 1 f -/ 68-oea-lZ 68-PO-SO 68-|nr-90 68-Jdv-H 68-UBr-ZL -89^ 0N-62 CO CO CN CM O •Q'O peisnfpv CL it o WO^NNi-T-^-T-i-r-T-r-^O-OOOOOOOOOOO Q Cy^OT-r-OOOO^-O </) q o o o o o o o o o o o ' o o o o o o o o o o T3 Q S O o o o o o o o o p o d E CO (0 2 ^ ^ 7 9 9 5 2 9 7 7 :6runr-L0 :P$fU oferinr-6i. - 0 mn-zz et osa- YZ -WPO-SO -ep-rnr-90 s s s s s s s s s s s s o ' o ' o o o o o o o o c j i q j Q O p © i s n f p v O ^ o o o o o o o o o o j d o o o o d d d o d d ni00)OOCQ(0(Dtt r rOrNr tOrO'" OOOOOOOOOO o d o d d o d d d d * NrNCMCOCOOlt-tOOO^  ! " 9 9 9 9 9 9 9 9 9 9 < O) CJ> O Oi CJ> o 2 ? ? I ? ? ci <g r- CN -^ uS T - O CM O O o O) cp "7 O CO 32 C6-1PC-20 08-in '-01 Ofl-JBN-si. 9^ ? 68/des-Z0 68-unr-ZZ 69-JdV-90 CO CO OJ CM Q O p s i s n f p y g 8 § § ° <? 5 5 -1 a. Ot<OIOU)ir)ir)iflU]iflU)U)U) ^ icococococococococococp © o p p o o o o o o o o o ^ 0 . 0 0 0 0 0 0 0 0 0 0 0 Q I - O 0 I - O N ( M I N ( M T - 0 (/) O O O O O O O O O O O d d d d d d d d d d o E TJ Q 0> o P 9 P 5 P O T - O O O O 9 9 9 0 9 0 0 0 0 0 0 2>0>0)0)O0oO*~^ C0 2?* c 0)>ja ;= -i 71 >>o) 5 - 7 < Z L L < 2 9 4 - 5 < T - O J O O O C N ^ - ^ T - ^ - C O ( -s O Q O O Q Q p O O C O C O C N C N T - T - O O O o o d o o d o c D o o c p i Q ' O p e j s n f p v 69HAB|ft|-u J o N W C M r N C N W C N C M C M t M C M a U d d d d d d d o ' d d d c^Of-co^ -rorM'^ -cO'f-ro fOr-T-T-i-rOT-rOO'-rN O O O O O O O O O O O o d d d d d d d d d d 2 0)nOU)0)r-0^ -CM(Oirt UQNtOlOT-jT-UJOi-N^ Q S O O O O O O O O O O Q w 9 9 9 9 9 9 9 9 9 9 o o > ° > 2 S J 9 S o o * -cocoopcoc^cn^cocr) 9- => 7 ^ - CN CN Q ' O p e t s n f P V 0-cn go. CO CO CO 0 ) 0 0 o o ' o d o o o d d o ' S N N T - O M C M I D O N »-COt\l ' -OT-rCOn O O O O O O O O O O d o o d d d d d d d S O S o O C N T - C N r - O J - > - t O C N Q O O O O O O O O O dp'oooo'o 'o 'o 'o © ra^OcuM^o^nja Q. 5 7 9 0 5 7 0 ^ 5 7 CM 1 -00 p e t s n f p v i c o c o o o c o c o c o c o c o c o c o c o ; C O 0 O C p C O C O 0 O C O C O 0 O C O C O l o o o o o o o o o o o •^cN -^coootomcotNCNcn K O O O O O O O O O O O 0 0 0 0 0 0 0 0 0 0 0 * l f l ^ - O « 0 N 0 3 ( D 0 0 0 ) ' - N " D Q t N ' - T - T - T - o n O ' - O i n ( D Q O O O O O O I - O O O O » d d d d d d d o ' d d d m O O n O O ^ - T— -^ ^CNfo a 9 O T § 9 ? 9 P ° ? P 9 S 6 c o ^ ^ o S c o c o - A r i c W t : 0 0 0 0 0 0 0 m o m o i o o m ' ^ • ' ^ • C O C O C M C N ' ^ T -d d d d d d d d 00 p e j s n f p v 93 o Oti'-'-'-WWIOIflLnLO "O^ f^^ fN^ CMCNICNICN ^ f J - o d o o o o d o o Q NSJOONCMi.... (/) o o o o o o o o o ddddddddd * T - M I O C D « ) O C O ^ C O gyOOOOQOOOO t j d o ' o ' o b o o o b « 20JOOOQOO*-CO i o o in m m i o o o o> co > CO CO CN CM CN £ 0 . 0 0 0 0 0 0 0 0 Jjj O O O O O O O O dddddddd ~- o •«- r- i • a Q CO CM CN I 0) o ° ° ° « 0 6 d d c i 9 d 9 « 3 OJ 0 0 o g Q 00 Sb cn cn co a> © £ & & a. < O S 5 to a E ub CM iA co cA Q ^ - O T - -^ o CO 2 2 CM ° • r t6-qad-9S CN 0 0 p e i s n f p v ;06-d8S-et \ 06-<BW-gi. • +08-JBH-I.0 + 68-PO-ZI 08iB"V9l ! 8 8 ) O T -> 0 9 a. CO 1 3 ...Ecorococooocococococo 5 ) 0 . 0 0 0 0 0 0 0 0 0 0 to o D, D ^ t O O C N O t N O O O t N (/) q o o o o o o o o o d o d o o o o d o d "OCDWCgiDNi-WOtD i Q ^ o s T - i n t o o o o o J Q p C N < N i - ; p p 5 p O ^ -J o o o d o o o d o d In c n o i O o Q O ' — T~ Q c ^ c o c n c n o > 0 ) 0 ) 0 > c n o ) a. Z Q 5 - ? < z u _ 5 < z E cNT i -<N*-doS^ebco'^ ' « O C M O J C N C O C N J C N C N O T -/ / t6-A0N-fl |.6-6nv-80 |.6-/B|fl|-£3 06-AON-6Z 06-6nv-oe 06-unr-I.S 06-JBW-Z3 - • r 68-AON-ZO CO CO CM CM QO psisnfpv 3 OtiCNCNCMCMCMCMCMCMCMCM A ± C M C V C M C M C i n n C O C O C O O Q T - ^ - T - ^ - T - T - O O O O ^ D - O O O O O O O O O O Q T - 3 ( O n N r N « W T -O O O ^ - O O O O O O ^ c o T - c M O ) ^ c o c o i n i o i o • O O * - T - C \ I C O T 7 0 1 C O C O C O C \ J O Q O O C N T I - P O O O O O m o ' o d d ° d d d d d m O Q O O T - ^ - ^ - T - C N C O Q < n § c n c r > o > o ) 0 ) 0 ) C ] ) a > W C03(l)5n3cDO(l)C0 E r ig iAc6uS<ocod>(A4 n T— CNJT— O O O * - '— CM Z6-q9d-8l. t6-d8S-eo 4 t6-unr-90 QO psisnfpv a. A 95 ^ D . d d d d d d d d d d d 0 0 0 0 0 0 0 S 0 S 0 d d d d d d d d d d d CO ^ tCONSO(D'-CO(DtO^ © Q O p * _ ( N , _ 0 0 0 0 0 q to 9 9 0 0 0 0 0 0 0 0 9 o. O Q U . 5 < C O ™ < 7 0 0 E co<ioo<oul)c464t:<o4 0 0 peisnfpv " 1 8 3 3 3 3 3 3 3 3 3 3 Q - 0 0 0 0 0 0 0 0 0 0 c n o o ^ o o o s c o o o N U ) T - O O T - ^ 0 ^ 0 0 4 0 0 0 0 0 0 0 0 0 0 0 dddddddddd to 0 0 CO 0 3 O 0 0 9 9 9 9 0)oooo*-i-*-*-co 000)0)0)0)0)0)0)0101 z 5 - j » z 5 - i ( o a < Otf>4eotNCOuSricN^ -CO'-*-*-C\IC\ICN,-,-cO ;Tee-6nv-i.e • , f i-e-oea-zi. t6-d9S-Zl 16-unr-gj 08-das-El ; 06-JBW-st-• + ^ 6 8 H A O N - o e QO PBJsnfpv a. O +S CO CO CO CO CO CO § o o o o 3 1 O O I D Q S lO CN (0 o o o o o o _ w s o 0) Q CN CN ^~ to o d d « 2 5? co Q °? ? "? ® I £ 5-1 2 1 £ 0 . 0 0 0 0 0 0 0 0 0 0 0 0 cMcootooocoo icomoomco S o o o o o o o o o o o o o o o d o d d d d o b o S Q P P C N C N ^ O O O O O O O » p p d o o d ° 0 0 0 0 0 ooooo>a)o>coo)OOooro cocococb<ococoa>0)c7)0>o> •> •> A }scCL>J»>.0)ic!L OOCD™D<0Oa)CtJ3OO Z Z L L . 5 - l W Z I i . 5 < Z « T ^ r ^ < o * - t o i ^ c o O u S c i t A o o O T - O C N O O C N C N T - O O C N •r E6-des-82 QO pe;snfpv a _J B 292£ococococococococo OOCOCOOOOTCpCOCOOOCOCO p p o o p o o o o o o 0 0 0 0 0 0 0 0 0 0 0 i-JP)mi-(N(Nri-i-»-T-OOOOOOOOOOO d d d d o o o o d o d TJQcoofoSjr^o^-ScNo © Q O O ^ - p p o o o P q o to d d t i p P c i o ' o P o c ) c o o p a p o p c o a ) o ^ § a 5 £ I I* S1 g i k -3 * «NCNic^r id>-^ -coc f>^ ^"OCNOOOCN'-T-t- o co co CN a S § § • ( T O p s i s n f p v £ 0 . 0 0 0 0 0 0 0 o d d S0O(D(M(Ot-CN(M(DP) Q OI0NM03NN'*'-09 O O O O C O O t - O O O d d d o d d d d d d -JT>cMOOf-oeoi-T-o ? 9 0 0 0 0 0 0 0 0 0 SS 9! o> co o> o •D CO CO m CO O) Q T < 7 O -^ - "S S 6 CO O « CM tN £ 0 5 4 di .A CM CM co 7 CD co 10 m ao po;sn[pv 9 9 98 3= o ^ C t o o o o o o o o o o d Q C M O ' - e O O N O T - O ' - p ( / ) O O O O O O O O O O O O O O O O O O O O O O "-S'O'otMoinwinNojs 3 0 - c n , t _ C O t ~ ' ^ 0 0 0 0 0 I) 9 0 0 0 0 0 0 0 0 0 0 < o M 0>01fllO)0)OOOOr(0 m H I «i m «"i rt\ rn rr\ /-T*\ rr\ i-n W O l O l U I U I U U U O r P ) ffiQcococDcnCDojcncncn a ) Q - 3 o a r a r o 3 o ( D 3 H . « f 7 C 0 Q S 5 < Z L L < 00(Ncgr-0(V)CN|'-'-(0 E6-6nv-|.€ CO CO CN CN 1 -Q O p e j s n f p y Q -I 4 0 " £ < o t o t o t D t D t D t D £ D t D t D 5o^ WCOCOcOP)«(*)fO(0 ^ 0 . 0 0 0 0 0 0 0 0 0 0 t o ^ - o j o ) 0 ) i - o o i r n - ^ T - T - C N ^ - T - C O ^ - T - C N C N I 0 0 0 0 0 0 0 0 0 0 o o o d d d o o o d -a Q S O < 9 0 0 6 0 6 0 0 0 6 u i o ) o > c n c n o o g o ' -oocooo<x>coa)C>o>a)0) Qj ^ o o D ra 3 o ro u _ 5 - 7 C 0 Z i i . 5 < O - 7 T - T - O T - C N C N Q O C N C N 0 0 0 • Q ' O p e i s n f p y 12 CNCNICNJCNCMCMCNCNCNCN d o d d o o o o o d — OtO'-tOtNOO', - -o o o o o o o d o o _°2°2cOCOlOCOCNa>0>C£> O O P P O C M C M T - T - ^ - T - T -o j o > o ) C n o o o o < - T -c o t o t o ^ o i < « c n c n o > o ) s T 3 i " ? s < i ~ ? i : cM»f>S c -) , -oco l "4co 7 l6-JdV-8l. L6-uer-fr3 K>iJ'iWt»t:'.r>/\ CO CO CN OJ QO P8»snfpv 9 9 ^0-0000000000 Q (0 (DMDOSOrlfllOCO O O O O O O O O O O TJ Q S O X I < j O c n c D c o t - T j - c M T - r - c D 9 0 l f l ( O C M N O I N » - 0 9 d d d d d d o d d < ^ o > c n O ) O O o o o c o < 9 c ^ c o c o o > c j > a > c » c n c n QO peisnfpv 9 9 100 3= o 3 O o O < ^ Q . 0 0 0 0 0 0 0 0 d o S(OIC$mOrMCOO(D o o r - O ' - ^ T - g i - T -o o o o o o o o o o d d d d o d d d o d •o o S O O P - ^ T - - ^ T - 0 - < - 0 0 o O d d d d o o o d O O O O O T - ' - T - T - C N cno^cncncncncncnoio) 0™330«ra30™ u _ 5 - > < Z U _ 5 < Q 5 O i - O f i M ^ - T - O ' - ' -\ / / \ \ 16-080-21. I6:6nv-|.0 t.6-^m-9l t6-qsd-n 06tAON-ZZ 06-iBnv-OE •f-0 •--e6Hqed-80 CO CO CN CM 8 8 T- O "Q'O P8isnfpv uDT-tDcor^g>^p^^ococDTj-r--Tj-cN Q N ^ Q C N l T - O ' - O N ^ ^ M ' r N l ^ T - ^ 0) o o o o o o o o o o o o o o o o d d d d d d d d d d d d d d d d C OOU)ONOSC3P)(NU)(DOU)U)(D « LOCO(Olf)T-fOT-tNCNCMT-OCMO^ rO o o o o o o o o o o o o o o o o o E d d d d d d d d d d d o o ' o ' o ' o ' "O Q SO l l l l l l l l l l l l l i l i cn -i-f > L_- si c •=" CL -rf c - Q . - . cb 50 I a* 0 1 L_ .^ w, „, ?ls'i'i'iii5"|g'i 9- 3 < -> cn cn / / / Y / i6J 1-661 11 1 1681 0661 o i | | j J - 066M ; j 0661. i 0661 066| 0660 offf 6860 :6||J + 6361 W 6 'Jdv6 'qsd 9Z 'uep 6Z 'oea f 'POEZ 'des 14 'mr 'jdys i'uer 6 'AbN IZ 'poo\. '6nv IE 8 8 •Q'O peisnfpv -688^  8 8 8 9 9 a _ i cs 3= p ;r o 1 ' ' ; 1 1 ' "} "} ; l d d d d o d d d c i c i o ' oooiootomrooocDcooco Q CN*-O*-C0*-CNC0C0f0<-5J o o o o o o o o o o o ddddddddddd •a a ; a> o m CD CN O CO CO O CO CO 088 .009 o O o 088 .009 9 O d d d o 9 o o o o o CO CO O) CO CO ^•J-dlA^QtNCOgriA nN(Nr??)CMT-OoS • « - ^ - T - CO en en o) o) 9 A li J. *!; > S O O Q Q o w o io CO CO CN CN T -Q O p e » s n f p v O t J N N N N N n r o t o n r o t o iCOCOCOOCOCOCOCOCOCQCO ^ l o o d d d o d d d d d ( O O O r t r i n o c i n c M i -S o o o o o o o o o o o ddddddddddd •a a S O Sf^S^cMUieNCDm £ £ J £ J L T ) T J - O O C O C O 9 9 9 0 0 0 0 0 0 S 3 o ° d o ffi ^ N ^ " N » - 0 « 0 CO f 66-JWl-OZ 1 CO CO CN CN Q O p e ; s n f p v i N N N i n o i n i n i n i o i o ' S S S S S ^ o c o c o c o l O p q o o o o o o o . o o o o o o o o o o 5 0 J 1 - 0 0 0 0 0 0 0 o p o p p o o o o o d o o o o o d o o d -OO B O lf)r-C0r-fON(DC00)CM o o o o P p o o p o 6 0 0 0 9 9 9 9 9 9 O l O l O q O O ' - ' - r r -C O C O O c n C D O C O C O C ' l O ) I» O b C i . > il >« rr> > 00)23000)03 3 0 Z D 5 T C 0 Z U . 5 < Z O t N C N J ' - C M C M C N l T - O T -6-AON-n -j- |6-Bnv-80 I6-A0N-63 / / i I9S-0Z t 06-J8W-ZZ S S g S S g g g g g C O C O C N C N T - T - O O O - T -d o d o ' o o o o o p QO pe;snfpv Q X « , _ c o c o ( o n c o c o t O L . g o C N I C N C N J C N C N t N C N C N ~ Q - d d d d o d d d C D C N ^ - ^ C D ^ - C D C O T - C N ^ - d O C N C M C N O O O O O O O O d o d o 0 o d d S O C O O O C O f - C N C O •DQtoW'-ocgoT-O n T T T T T P P P » 9 9 9 9 9 9 9 9 < c o o i o o o ^ r i o c o c y o > c n a ) a>9 ( 7 > 66-iocm |.6-AB|AJ-20 06-PO«Z •f 06-unrrZO m o m o 10 ^ ^ n n 01 N b b b b b b d b b QO psisnfpv 3 Q. d d d d d d d d d d d d d d d d Q T - C M O O O T - T - T - ^ ~ C O O C O T - T - C N C N 0j o o o o o o o o o o o o o o o o d d d d d d d d d d d d d d d d •o D 3 O WT-cni -of f i rMmsorto i incMmcM CDS(DOON<PSN(D(DSW(DCB(DIO OOOOOOOOOOOOOOOO I | ' S ' * S | § & S" 8" S £ § ='f 51 E Z T L L < 5 T ( D 2 Q _ ' < - 5 N < O Z CO ( O ^ C N U ; ^ ' - C N ^ < I ) C M ' - ' - ^ C N C N Q ' O p e j s n f p v Q _l CO O t z o o o o o o o o o o o o M±0)0)0)0)CDO)0)CDO)C7)CnO) ® o 0 ° o o o o o o o o o o £ D _ d d d o o d d d d d d d Offii-COlOSineONNCMO Q i-0(NT-rr(OOrO«-N (/) o o o o o o o o o o o o dddddddddddd e n o i n ^ N T - i M N r m - T -tn d d d d d d d d d d d d o > c j ) o o o S o 0 , , - , - ' , - ' _ co«0)Oicnwo)cno)o>0)o) S 8 § « A. S'-i &« I &* O Q 7 5 < : 5 7 « 5 T W O ^rioOT^-CNJ-^r^-^tDiiSr^ NO'-0'-(M,"CM'-00'- 68-09Q-ZO O O O O O O Q O O O O Q O Q m o u 5 o i o o i o 5 i o o i o 5 i n 5 (DoiomvvnnNcjrroo d d d d d d o d d d d d d d • Q ' O p e j s n f P V ! C 0 C O ( O ( O ( O C O < O C O ( O ( O ; 0 0 0 0 0 0 0 0 0 0 i C N C N C N C N C N C M C M C N C N C N d d o d d d o d d d N ^ T - T - T - O O C M M O C M T - C N I ^ - C M O T - C N C M O o o o o o o o o o o dddddddddd ^ O ^ Q c N I ^ t O t O C D C O C O K H O O O o O O O O O O » 9 9 9 0 9 9 9 9 9 9 0 ™ « a : - > O u _ 5 2 0 ~ ? N ( ^ e o ^ 6 c O ( o i 6 ^ t 66-PO-OC f B8-PO-0I-§ 8 8 8 CO CO CN CN QO P©»snfpv gO^^^^^CNCNCNCN p l d d d o d d d d d <OOtOOO*(NCM Q COtDtOWtNCOJCMn (0 O O O O O O O O O 0 0 o 0 0 0 d 0 0 2 CM , •a o S 3 1 a> o q o 1 O T— O") CO h- CO LO CD CO 00 CO o o * 9 0 0 d d d d d d c n o ) r o 0 ^ c n c p o ) 0 > c T > C D O j O u l j ^ t M i i 1 ) ^ ^ 0 * - ° C M ^ - ^ - C M C M C M 96-^ B|ft|-g2 4 l.6-8nv-Z2 06-08Q-U CO CO CN 00 peisnfpv a .A 105 it o 3 © o o o o o o o o o o o o ^ D - o d o d d o d d o c i d Q C N ' - O j n O M Q O ' - l f i d ) 55 O O O O O O O O O O O ciciociocioocDdo •OOODT-T-fflNNCOOlCO S Q O O O O O O P ' I - C O ' * - ' * -i) 0 0 0 0 0 0 9 6 6 6 6 < O Q CO O) O O) • • A. ,„ 0 0 0 ) C J ) 0 ) C 3 ) O O (0 OOcO*-i-OCNCN*-*-C\l CO CO CM CN QO psisnfpv o X 106 Q tQ-dddddddddd S 8 5 8 d d d d o o o o o d Is- Is- co r-- in O O O *-o o o o o 0 ^ 0 0 5 0 0 0 0 0 0 0 8 0 9 0 9 9 9 9 9 9 9 oio>o)Oiaiooo cocococococncno) J I I T w 2 i i . 5 < CNOOOCOCNJT-O o o AON-1.0 I 6B-AON-0E 68-dos-ZO 68-'Jnr-80 O Q Q O m o to o CO CO CM CM •v^uBf-gz s s Q'O peisnfpv OtieMfMCMCNCMCNICMCMCMCMCM gO^ C^OCOCOCOCOfOCOCOCO ^dddddddddddd a CO •o a & O OOOOOOOOOOO b b b b o d d o d d d OCOCDLO^COCNCNCNIOC) ddddddddddd s C00)0>0)0>0>OQOOc0 c o c o ^ c o c p < P 9 9 0 * 0 , 9 S c s « § S - 8 « » i , i ; B o - 5 ! 5 - ? c o a S 5 < z O ^ScoocM4•^i•'-4•cDuS4, t 68-uer-eo Q'O peisnfpv 107 3 O " £ t f ? o o o o o o o o o o o i^OCNCslCNCMCNCNCNCMCNCNCN S o N ( 0 « n r o r o n n c o « o o n ^ Q - d d d d d d d d d d d d a j c o N u i c o g > 5 t o N i n o ) ( o Q O T - C N ^ C N O O O O O O O (/) o o o o o o o o o o o o d d d d d d d d d d d d J a : JO i 0) 5 o • m co io • i o o o l <D ^ (O O t S CO Is- CD IO O O O O 9 9 9 9 9 9 9 9 9 9 9 re O O O O T - O O C N T - O O C N co CO CO CN CN QO Psisnfpv o o o o o o o o o o o r- in •<- CM o o oo oo o oo o o o m IT) o w o ^ CO CM s CM o o d d d d d d o d d d d •a Q S O Q. E S o O C O f t ( M O T - t P d d o d d d d d o d CO o a> x S i-90 o 1 •U CD 3 > o 6 CO N M o o \ 96-des-ZO + 06-*ON-l.O 4 06-6nv-60 08-ABH-ZI. 06-qed-zz 68-AON-OE 68-*>s-zo •• 68-unr-80 •• 68-JBIAH2 o o o o o o o o o o o o o o o e«-«er-9z to o to o m to o to o to (DStowf^nnNNT-T-o b b b b b b b b b b b b b b QO psisnfpv 9 9 a x 5 o t u) in io u) u) IA n ui m u) £ m m m m m m m m m m ^ Q - d d d o d d d o d d n^ cMCMMncMNcom T - C N C M Q C J C O - ^ C M O ^ O O O O O O O O O O d d d d d o d d d d ' n O q o S o p o p o q i d d o o d d d o o o ( D C O r x i r o f f i C D O c J l O l O ) 5? 8 8 CO CO CM 8 8 Q'O psisnfpv Q " c S S S S S S S S S S £ 0 . 0 0 0 0 0 0 0 0 0 0 0. I— O CO CO CO I— CO O Q> CO g n o Q r p j r n & T -OOOOOOOOOO d d o o d o o d d o E •o • s o o o ^ o o g o o o o P p P p p p P p o o o d d p p ' d o d d d o c o c o ^ c o o o a o Q C O C O o p p C p C O c o f f i C o C p o a> a- n <u co a. a " 8 i^Lu<7coii.<:29Q cococo*-c6(p4o>*-d> ?r 06-=8Q-oz ; 06-POW » oe-inr-ei. T 68-<l8S-92 4.«8-JdV-Et / 68-q8d-8Z wqsd-91. 8 •Q'O psjsnfpv 109 :ioiommmmiomiom i C O O C O c O f O C O C O C O C O t O . o o d o o o o o o d inmo>cDcococN*-T-tN CN^^O i^-COOOCN o o o o o o o o o o 0 0 0 0 0 0 0 0 0 0 5 WO«)N*lfl(D(MT-0 © ^ 0 0 0 0 0 0 0 0 0 0 * ? 9 d ? 9 ' 9 9 ? 9 9 a. u-<7<0OS5O-?^ a O O O C N O O C N O O O CO •.16-JBW-iO 06-inr-SO CO CO CM CM 0 0 p e i s n f p y § S § o ^ ^ 9 9 9 Q X i t s c o c o c o c o c o o o c o c o c o ^ C c O C O C O C O C O C O C O C O C O r -o o o o o a o o q o r i L d d d d c i d d d d d N o o c o i n n o i c o r ) ^ Q O T - T - I - « - O Q * - * - C M M J O O O O O O O O O O d d d d d d d d d d • n r » £ - ? - * - o o o T - o g ^ e S o o O o o o o o p P S u 0 0 9 9 9 9 9 9 9 9 o? en g? o) g g s a 9 » 9 9 • I SI'S S j Q. E 7 < 9 7 5 E CM O CO CO CO Is" a S CM o o CM « -in O CO > A. 7, Q ? 1 o CM pi. . O CM I y«6-das-l2 06TAON-ZZ 06-PO-ZO 06-inr-Ol 06-Jdv-Zl 06-uep-ez 68-PO-CO / < 6 8 r 6 n v - 8 0 68*unr-0Z • ^ 68H<BW-Z0 o 8 ao P8)sn[pv 110 U + j i o w i o m i n i o i O L O t n i n gO<NC\JCNC>JCNCNCN|CNCMe\| p l l d d d d o ' c i d d d o ' _ (O^ tOCO-tt-CNOlh-Tt-CNI 5) OOOOOOOOOO OOOOOOOOOO •O Q SO i n ( M O U ) N C O C O t O O > O Q O O O O O O O O 9 0 9 9 9 9 9 9 9 9 0>0)OOoOO -^f-T-< 9 ? 1 : 7 w q 5 7 w cN?toocNigri64sSto er-81-CO CO CN CM a'OPSlsnfpv it 0 >*SOOOQOOOOOOOOO > 0 0 0 0 0 0 0 0 0 0 0 0 0 0 • 0 . 6 0 0 0 0 6 0 0 0 0 0 0 0 o o i o ^ - i o i o o i c o w ^ c n c M i -0 0 0 0 0 0 0 0 0 0 0 0 0 6 6 6 6 6 6 6 6 6 6 6 6 6 ^Qr-coo)toc5^ iococoo)coeN^ OQ'-(0«)mtin*iO(on(DMn « 6 6 6 6 6 6 6 6 6 6 6 6 6 a> co CL 5 i ™ « CM CO r-jOOOO^T- — i-CMCMCM i g c n m c p c n o j c n i j o i c n c i j o i S24lfS(OCooSco<Ocoh-*-c\l \ \ / \ / / Z6-JBW-IE S6-uer-Z0 16-PO-80 ie-inr-91 •f 16-Jdv-91 06-090-81 06-AON-90 { 06-d8S-9Z 06-Bnv-fl. 06-inr-E0 O O O O O O O O O O ao ps;snfpv Q X 5f 111 3 ! o r T : T : r ; T : , ' r £ 0 - 0 0 0 0 0 0 0 CO CO CO O CD CN Q O O O T - O C S J Q (/) o o o o o o o o d o d d d d £ SO JO CD o> £ ^ CO S ' - Q o g wo P q P p P q o P d ° d P i " i d d 0> Q O O O GO O) O) O) O) 9 7 *f 7 ? N (D (N r- ^ " N O CO T - T - T - I 4 d CN N Q O p s i s n f p v • 06-AON-lO - 0 6 - u n f H Z • 06-Jdv-ZI. - 06-uer-fH !'fl8^ P0-ZI. 3 S - . a Q - o o o o o o o o o d d d (/) p p p o o ^ o o o o o o d d d d d d d d d d d d f^ ioroifleocococoiocNOU)^  pQT-^ CDCMtDCMtO^ mcOiOJT © Q O O C N C N I T - O O O O O O P o) d d d o d d d d d d d 9 3 ' T? < Q O P e j s n f p y 112 ~ ( L o o d 6 6 o d o o o 6 _ (Otooi«3«)^p)nT-s5 (/) OOOOOOOOOOO OOOOOOOOOOO • O Q O - ^ O C O O O O ° T - ^ - T -o g o o o o o o o O o o o w 9 9 9 9 9 d d 9 ' d o ' 9 s ta ooooo^-^-*-T-rgco Q O)O>cnO)0^0^CJ)O>O)C>C7) a L L S - ^ C O Z U . 5 < O U _ C O ^ ^^^^ob^cAcoo^-co - 0 6 * ^ 1 - 0 -T r c o c o r j c g - i - f - o o o T -d d d d d d 0 0 0 0 9 9 • Q ' O p e i s n f p v Q X (J+-OOQOOOOOOOOOO g o o o o o o o o o o o o o o j U d d d d d d d d d d d d d N©(OtM(Or>JW«-C))OCDO)N Q OQ'-'-(D(N(N'-'-M(OTrt-5) OOOOOOOOOOOf-O d d d d d d d d d d d d d « O I O C M C O C O L _ _ , „ _ _ . _ . . cD^oco^asenr--r--o -^r,--h-d d d d d d d d d d d d d OOOQoOOO*-*-'*-'* O)0)0)O)O)CJ>O)O)0)0)O>( c - i JK > * C O > O _ 6 JA >*6>; (00)™iSD3(llO(DiS3l 7 U . 5 5 7 < W Q L L 5 < ; fNuSr^dri^^cAcococsir O T - C M ^ C M O ^ O C M O J C M C \ / / —t 1 1 1 p. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CO L O * T CO CN O d d d d d o d ' Q ' O p e i s n f p v Z6-uer-Z0 1.6-AON-ZO t6-qsd-8Z 06-380-90 06-des-Sl 06-6ny-Z0 t 06-unr-|.2 06-<EW-01 06-JBIAI-ZZ t 06-q8d-9l. 06-uep-zo 113 APPENDIX B SUMMARY LIST OF ALL ASSAY RESULTS FOR ALL PATIENT SAMPLES PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 1-PL 25/10/1988 1 1 0 0 1-PL 06/12/1988 43 0 0 0 1-PL 17/1/1989 85 0 0 0 1-PL 28/2/1989 127 0 0 1-PL 11/4/1989 169 0 0 0 1-PL 23/5/1989 211 0 0 0 0 1-PL 04/7/1989 253 0 0 0 1-PL 26/9/1989 337 0 0 0 1-PL 19/12/1989 421 0 0 0 1-PL 13/3/1990 505 0 0 0 0 1-PL 05/6/1990 589 0 0 0 0 1-PL 28/8/1990 673 1 0 1-PL 09/10/1990 715 0 0 0 0 1-PL 15/1/1991 813 0 0 0 1-PL 02/4/1991 890 0 0 0 1-PL 17/9/1991 1058 0 0 0 1-PL 17/12/1991 1149 0 0 1-PL 03/3/1992 1226 0 0 1-PL 12/5/1992 1296 0 0 1-PL 11/8/1992 1387 0 1-PL 10/11/1992 1478 0 1-PL 04/2/1993 1564 0 1-PL 11/5/1993 1660 0 2-PL 01/11/1988 1 0 0 0 2-PL 13/12/1988 43 0 0 0 2-PL 24/1/1989 85 0 0 0 2-PL 07/3/1989 127 0 0 0 2-PL 18/4/1989 169 0 0 0 2-PL 30/5/1989 211 0 0 0 2-PL 11/7/1989 253 0 0 0 2-PL 03/10/1989 337 0 0 2-PL 20/3/1990 505 0 0 0 2-PL 12/6/1990 589 0 0 2-PL 11/9/1990 680 1 0 0 2-PL 23/10/1990 722 0 0 2-PL 19/2/1991 841 0 0 0 2-PL 09/5/1991 920 0 0 0 2-PL 09/7/1991 981 0 0 0 2-PL 03/10/1991 1067 0 0 0 2-PL 12/12/1991 1137 0 0 2-PL 17/3/1992 1233 0 0 2-PL 02/7/1992 1340 0 0 2-PL 01/9/1992 1401 0 2-PL 01/12/1992 1492 0 2-PL 25/2/1993 1578 0 2-PL 06/5/1993 1648 0 3-PL 17/11/1988 1 0 0 0 3-PL 03/1/1989 48 0 0 0 0 ' . ' = Sample Not Assayed 115 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 3-PL 07/2/1989 83 0 0 .0 3-PL 21/3/1989 125 0 0 3-PL 04/5/1989 169 0 0 0 3-PL 06/6/1989 202 0 0 0 0 3-PL 25/7/1989 251 0 3-PL 17/10/1989 335 0 0 0 0 3-PL 09/1/1990 419 0 0 0 3-PL 03/4/1990 503 0 0 0 0 3-PL 03/7/1990 594 0 0 0 0 3-PL 25/9/1990 678 1 0 0 0 3-PL 06/11/1990 720 0 0 0 3-PL 29/1/1991 804 0 0 0 3-PL 30/4/1991 895 0 0 0 3-PL 16/7/1991 972 0 0 0 3-PL 08/10/1991 1056 0 0 0 3-PL 16/1/1992 1156 0 4-PL 29/11/1988 1 0 0 0 4-PL 21/2/1989 85 0 0 0 0 4-PL 06/4/1989 129 0 0 0 4-PL 16/5/1989 169 0 0 0 0 4-PL 29/6/1989 213 0 0 0 4-PL 10/8/1989 255 0 0 0 0 4-PL 07/11/1989 344 0 0 0 0 4-PL 23/1/1990 421 0 0 0 4-PL 17/4/1990 505 0 0 0 4-PL 10/7/1990 589 0 0 0 4-PL 02/10/1990 673 0 0 0 4-PL 20/11/1990 722 0 0 0 0 4-PL 12/2/1991 806 1 0 0 4-PL 09/5/1991 892 0 0 0 4-PL 30/7/1991 974 0 0 0 4-PL 22/10/1991 1058 0 0 0 4-PL 14/1/1992 1142 0 0 4-PL 07/4/1992 1226 0 0 4-PL 14/7/1992 1324 1 0 4-PL 22/9/1992 1394 0 4-PL 15/12/1992 1478 0 4-PL 25/3/1993 1578 0 4-PL 18/5/1993 1632 0 5-PL 06/12/1988 1 0 0 0 5-PL 12/1/1989 38 0 0 0 5-PL 28/2/1989 85 0 0 0 5-PL 11/4/1989 127 0 0 0 5-PL 25/5/1989 171 0 5-PL 06/7/1989 213 0 0 0 5-PL 22/8/1989 260 0 5-PL 09/11/1989 339 0 0 0 5-PL 01/2/1990 423 0 0 0 ' . ' = Sample Not Assayed 116 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 5-PL 24/4/1990 505 0 0 0 5-PL 17/7/1990 589 0 0 0 5-PL 09/10/1990 673 0 0 0 5-PL 06/12/1990 731 0 0 0 0 5-PL 05/3/1991 820 0 0 0 5-PL 23/5/1991 899 0 0 0 5-PL 13/8/1991 981 0 5-PL 15/8/1991 983 0 0 0 5-PL 12/11/1991 1072 0 0 0 5-PL 06/2/1992 1158 0 0 5-PL 14/5/1992 1256 0 5-PL 13/8/1992 1347 0 5-PL 15/10/1992 1410 0 5-PL 14/1/1993 1501 0 5-PL 01/4/1993 1578 0 5-PL 18/5/1993 1625 0 6-PL 31/1/1989 1 0 0 6-PL 16/3/1989 45 0 0 0 0 6-PL 27/4/1989 87 1 0 0 6-PL 06/6/1989 127 0 0 0 6-PL 20/7/1989 171 0 6-PL 29/8/1989 211 1 0 0 0 6-PL 12/10/1989 255 1 0 0 6-PL 09/1/1990 344 0 0 6-PL 29/3/1990 423 0 0 6-PL 21/6/1990 507 0 0 0 6-PL 13/9/1990 591 0 0 0 6-PL 06/12/1990 675 1 0 0 6-PL 24/1/1991 724 0 0 0 6-PL 18/4/1991 808 1 0 6-PL 11/7/1991 892 0 0 0 6-PL 03/10/1991 976 1 0 0 6-PL 02/1/1992 1067 1 0 0 6-PL 26/3/1992 1151 0 6-PL 02/7/1992 1249 0 0 6-PL 17/9/1992 1326 0 6-PL 03/12/1992 1403 0 6-PL 25/2/1993 1487 0 6-PL 25/5/1993 1576 0 0 7-PL 21/2/1989 1 0 0 7-PL 06/4/1989 45 0 0 0 0 7-PL 18/5/1989 87 0 0 0 7-PL 27/6/1989 127 0 0 0 0 7-PL 29/6/1989 129 0 7-PL 15/8/1989 176 0 0 0 7-PL 21/9/1989 213 0 0 0 0 7-PL 07/11/1989 260 0 0 0 7-PL 23/1/1990 337 0 0 0 ' . ' = Sample Not Assayed 117 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 7-PL 17/4/1990 421 0 0 0 7-PL 10/7/1990 505 0 0 0 0 7-PL 02/10/1990 589 0 0 0 0 7-PL 15/1/1991 694 0 0 0 0 7-PL 12/2/1991 722 0 0 0 7-PL 07/5/1991 806 0 0 0 7-PL 30/7/1991 890 0 0 0 7-PL 22/10/1991 974 1 0 0 7-PL 14/1/1992 1058 0 0 0 7-PL 07/4/1992 1142 0 0 7-PL 07/7/1992 1233 0 7-PL 22/9/1992 1310 0 7-PL 17/12/1992 1396 0 7-PL 09/3/1993 1478 0 7-PL 11/5/1993 1541 0 0 8-PL 28/2/1989 1 0 0 0 0 8-PL 13/4/1989 45 0 0 0 0 8-PL 25/5/1989 87 0 0 0 8-PL 06/7/1989 129 0 0 0 0 8-PL 17/8/1989 171 0 8-PL 05/10/1989 220 0 0 0 0 8-PL 09/11/1989 255 0 0 0 8-PL 08/2/1990 346 0 0 0 8-PL 26/4/1990 423 0 0 0 8-PL 19/7/1990 507 0 0 0 0 8-PL 11/10/1990 591 0 0 0 0 8-PL 10/1/1991 682 0 0 0 0 8-PL 21/3/1991 752 0 0 0 8-PL 23/5/1991 815 0 0 0 8-PL 15/8/1991 899 0 0 0 8-PL 07/11/1991 983 0 0 0 8-PL 30/1/1992 1067 0 0 0 8-PL 23/4/1992 1151 0 0 8-PL 16/7/1992 1235 0 0 8-PL 08/10/1992 1319 0 8-PL 14/1/1993 1417 0 8-PL 25/3/1993 1487 0 8-PL 01/6/1993 1555 0 0 9-PL 18/4/1989 1 0 0 0 0 9-PL 01/6/1989 45 0 0 0 0 9-PL 18/7/1989 92 0 0 0 9-PL 24/8/1989 129 0 0 9-PL 03/10/1989 169 0 0 0 9-PL 23/11/1989 220 0 0 0 0 9-PL 09/1/1990 267 0 0 0 9-PL 22/3/1990 339 0 0 0 9-PL 14/6/1990 423 0 0 0 9-PL 06/9/1990 507 1 0 0 ' . ' = Sample Not Assayed 118 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 9-PL 22/11/1990 584 0 0 0 9-PL 21/2/1991 675 0 0 0 9-PL 11/4/1991 724 0 0 0 0 9-PL 18/7/1991 822 0 0 0 9-PL 26/9/1991 892 0 0 0 9-PL 12/12/1991 969 0 0 0 9-PL 12/3/1992 1060 0 0 0 9-PL 04/6/1992 1144 0 0 9-PL 03/9/1992 1235 0 9-PL 19/11/1992 1312 0 9-PL 11/2/1993 1396 0 9-PL 06/5/1993 1480 0 10-PL 23/5/1989 1 0 0 0 10-PL 06/7/1989 45 0 0 0 10-PL 17/8/1989 87 0 0 0 0 10-PL 21/9/1989 122 0 10-PL 09/11/1989 171 0 0 0 0 10-PL 21/12/1989 213 0 0 0 10-PL 03/1/1990 226 0 10-PL 01/2/1990 255 0 0 0 0 10-PL 26/4/1990 339 0 0 0 0 10-PL 19/7/1990 423 0 0 0 1 10-PL 18/10/1990 514 0 0 0 0 10-PL 03/11/1990 530 0 10-PL 03/1/1991 591 0 0 10-PL 21/3/1991 668 1 0 0 10-PL 16/5/1991 724 0 0 0 0 10-PL 15/8/1991 815 0 0 0 10-PL 31/10/1991 892 0 0 0 10-PL 23/1/1992 976 1 0 0 10-PL 16/4/1992 1060 0 0 0 10-PL 09/7/1992 1144 0 0 11-PL 13/3/1990 1 0 0 0 0 11-PL 24/4/1990 43 0 0 0 0 11-PL 05/6/1990 85 0 0 0 11-PL 31/7/1990 141 0 0 0 0 11-PL 28/8/1990 169 0 0 0 11-PL 09/10/1990 211 0 0 0 0 11-PL 20/11/1990 253 0 0 0 11-PL 12/2/1991 337 1 0 0 11-PL 07/5/1991 421 0 0 11-PL 30/7/1991 505 0 0 0 11-PL 29/10/1991 596 1 0 0 11-PL 21/1/1992 680 1 0 0 11-PL 24/3/1992 743 0 0 11-PL 16/6/1992 827 0 0 11-PL 08/9/1992 911 0 11-PL 17/11/1992 981 0 ' . ' = Sample Not Assayed 119 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 11-PL 09/2/1993 1065 0 11-PL 11/5/1993 1156 0 12-PL 29/5/1990 1 0 0 0 0 12-PL 10/7/1990 43 0 0 0 0 12-PL 21/8/1990 85 0 0 0 12-PL 02/10/1990 127 0 0 0 0 12-PL 13/11/1990 169 0 0 0 12-PL 18/12/1990 204 0 0 0 0 12-PL 05/2/1991 253 1 0 0 12-PL 30/4/1991 337 0 0 0 12-PL 23/7/1991 421 0 0 0 0 12-PL 22/10/1991 512 0 0 0 0 12-PL 07/1/1992 589 0 0 0 0 12-PL 22/1/1992 604 0 0 0 12-PL 31/3/1992 673 0 0 12-PL 19/5/1992 722 0 0 0 12-PL 11/8/1992 806 0 12-PL 03/11/1992 890 0 12-PL 09/2/1993 988 0 12-PL 20/4/1993 1058 0 12-PL 25/5/1993 1093 0 13-PL 29/8/1989 1 0 0 0 13-PL 12/10/1989 45 0 0 0 0 13-PL 23/11/1989 87 0 0 0 13-PL 04/1/1990 129 0 0 13-PL 01/3/1990 185 0 0 0 0 13-PL 29/3/1990 213 0 13-PL 15/5/1990 260 0 0 0 0 13-PL 31/7/1990 337 0 0 0 13-PL 23/10/1990 421 0 0 13-PL 15/1/1991 505 1 0 0 13-PL 02/4/1991 582 0 0 0 13-PL 16/7/1991 687 0 0 14-PL 29/8/1989 1 0 0 0 0 14-PL 10/10/1989 43 0 0 0 0 14-PL 21/11/1989 85 0 0 0 14-PL 02/1/1990 127 0 0 0 14-PL 27/2/1990 183 0 0 0 14-PL 03/4/1990 218 0 0 0 0 14-PL 15/5/1990 260 0 0 0 14-PL 31/7/1990 337 0 0 0 14-PL 23/10/1990 421 0 0 0 14-PL 15/1/1991 505 1 0 0 14-PL 09/4/1991 589 0 0 14-PL 25/6/1991 666 0 0 0 14-PL 20/8/1991 722 0 0 0 14-PL 07/11/1991 801 0 ' . ' = Sample Not Assayed 120 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 15-PL 22/5/1990 1 1 0 0 0 15-PL 03/7/1990 43 0 0 0 0 15-PL 21/8/1990 92 0 0 0 15-PL 25/9/1990 127 0 0 0 15-PL 08/11/1990 171 0 0 0 15-PL 18/12/1990 211 0 0 0 15-PL 31/1/1991 255 1 0 0 15-PL 30/4/1991 344 0 0 15-PL 16/7/1991 421 0 0 0 15-PL 22/10/1991 519 1 0 0 15-PL 07/1/1992 596 0 0 0 15-PL 24/3/1992 673 0 0 15-PL 26/5/1992 736 0 0 15-PL 04/8/1992 806 0 15-PL 27/10/1992 890 0 15-PL 20/1/1993 975 0 15-PL 13/4/1993 1058 0 15-PL 25/5/1993 1100 0 0 16-PL 14/11/1989 1 0 0 0 16-PL 21/12/1989 38 0 0 0 16-PL 08/2/1990 87 0 0 0 16-PL 22/3/1990 129 0 0 0 16-PL 26/4/1990 164 0 16-PL 21/6/1990 220 0 0 0 0 16-PL 26/7/1990 255 0 0 0 16-PL 18/10/1990 339 0 0 0 16-PL 10/1/1991 423 0 0 0 16-PL 04/4/1991 507 0 0 0 16-PL 27/6/1991 591 0 0 0 16-PL 16/9/1991 672 0 16-PL 19/9/1991 675 0 0 0 16-PL 14/11/1991 731 0 0 0 0 16-PL 06/2/1992 815 0 0 0 16-PL 30/4/1992 899 0 0 0 16-PL 23/7/1992 983 0 0 16-PL 15/10/1992 1067 0 16-PL 14/1/1993 1158 0 16-PL 01/4/1993 1235 0 16-PL 01/6/1993 1296 0 17-PL 19/12/1989 1 0 0 0 17-PL 01/2/1990 45 0 0 0 17-PL 15/3/1990 87 0 0 0 0 17-PL 26/4/1990 129 0 0 0 17-PL 14/6/1990 178 0 0 0 0 17-PL 19/7/1990 213 0 0 0 17-PL 13/9/1990 269 0 0 0 0 17-PL 22/11/1990 339 0 0 0 0 ' . ' = Sample Not Assayed 121 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 17-PL 14/2/1991 423 0 0 0 17-PL 02/5/1991 500 0 0 0 17-PL 01/8/1991 591 0 0 0 17-PL 07/11/1991 689 0 0 0 17-PL 12/12/1991 724 0 0 0 0 17-PL 05/3/1992 808 1 0 0 17-PL 04/6/1992 899 0 0 0 17-PL 13/8/1992 969 • 0 17-PL 12/11/1992 1060 18-LD 25/10/1988 1 0 0 0 18-LD 06/12/1988 43 0 0 0 0 18-LD 17/1/1989 85 1 0 0 18-LD 28/2/1989 127 0 0 0 18-LD 11/4/1989 169 1 0 0 18-LD 23/5/1989 211 1 0 49 0 18-LD 04/7/1989 253 1 0 53 18-LD 26/9/1989 337 1 0 50 0 18-LD 19/12/1989 421 1 104 104 18-LD 13/3/1990 505 1 176 85 18-LD 05/6/1990 589 416 31 18-LD 28/8/1990 673 0 57 18-LD 23/10/1990 729 0 41 0 18-LD 17/1/1991 815 19-LD 15/11/1988 1 0 0 0 0 19-LD 03/1/1989 50 1 0 0 19-LD 07/2/1989 85 1 0 0 1 19-LD 21/3/1989 127 0 1 19-LD 02/5/1989 169 1 0 0 19-LD 06/6/1989 204 1 0 0 0 19-LD 25/7/1989 253 0 19-LD 17/10/1989 337 1 0 0 19-LD 16/1/1990 428 0 0 19-LD 03/4/1990 505 1 0 0 19-LD 03/7/1990 596 1 0 0 19-LD 25/9/1990 680 1 0 0 19-LD 06/11/1990 722 0 0 0 19-LD 29/1/1991 806 1 0 0 19-LD 30/4/1991 897 1 0 25 19-LD 16/7/1991 974 0 0 19-LD 08/10/1991 1058 1 0 0 19-LD 07/1/1992 1149 1 20-LD 29/11/1988 1 0 1 20-LD 10/1/1989 43 1 0 0 1 20-LD 21/2/1989 85 1 0 0 20-LD 04/4/1989 127 37 35 1 20-LD 09/5/1989 162 1 117 82 20-LD 30/5/1989 183 1 416 21 ' . ' = Sample Not Assayed 122 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 20-LD 03/10/1989 309 0 0 20-LD 19/12/1989 386 20-LD 13/2/1990 442 > 21-LD 20/12/1988 1 0 0 0 21-LD 31/1/1989 43 1 0 0 21-LD 14/3/1989 85 1 0 0 21-LD 25/4/1989 127 1 0 0 21-LD 08/6/1989 171 1 0 0 21-LD 20/7/1989 213 0 21-LD 31/8/1989 255 1 0 34 21-LD 23/11/1989 339 1 0 0 21-LD 13/2/1990 421 1 0 20 21-LD 10/5/1990 507 1 0 33 21-LD 02/8/1990 591 1 0 22 21-LD 13/12/1990 724 0 0 21-LD 07/3/1991 808 1 0 0 21-LD 30/5/1991 892 1 0 0 21-LD 22/8/1991 976 1 0 0 21-LD 07/11/1991 1053 1 0 0 21-LD 06/2/1992 1144 1 22-LD 17/1/1989 1 0 0 1 22-LD 02/3/1989 45 1 0 0 1 22-LD 13/4/1989 87 1 0 0 22-LD 23/5/1989 127 1 0 24 1 22-LD 04/7/1989 . 169 1 0 21 22-LD 17/8/1989 213 23 0 22-LD 26/9/1989 253 1 0 44 22-LD 21/12/1989 339 0 57 22-LD 15/3/1990 423 1 0 75 22-LD 07/6/1990 507 1 0 84 22-LD 06/9/1990 598 1 0 160 22-LD 29/11/1990 682 0 114 22-LD 10/1/1991 724 0 184 0 22-LD 27/6/1991 892 0 122 22-LD 19/9/1991 976 1 250 86 22-LD 12/12/1991 1060 1 1500 196 22-LD 05/3/1992 1144 1 23-LD 31/1/1989 1 0 23-LD 14/3/1989 43 1 0 0 23-LD 25/4/1989 85 1 0 23-LD 06/6/1989 127 1 0 0 1 23-LD 18/7/1989 169 0 23-LD 12/9/1989 225 1 0 0 1 23-LD 12/10/1989 255 0 0 23-LD 09/1/1990 344 0 0 23-LD 29/3/1990 423 1 0 23-LD 21/6/1990 507 1 0 0 ' . ' = Sample Not Assayed 123 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 23-LD 13/9/1990 591 1 0 23-LD 13/12/1990 682 0 23-LD 24/1/1991 724 1 0 0 23-LD 18/4/1991 808 1 0 23-LD 25/7/1991 906 1 0 47 23-LD 17/10/1991 990 1 0 28 24-LD 21/2/1989 1 0 0 0 24-LD 06/4/1989 45 1 0 0 24-LD 18/5/1989 87 1 0 0 24-LD 29/6/1989 129 1 125 26 24-LD 10/8/1989 171 0 48 24-LD 21/9/1989 213 1 0 97 24-LD 07/11/1989 260 208 144 24-LD 25/1/1990 339 166 202 24-LD 19/4/1990 423 1165 202 24-LD 12/7/1990 507 1331 142 24-LD 02/10/1990 589 469 57 24-LD 17/1/1991 696 1083 87 24-LD 14/2/1991 724 118 0 0 24-LD 07/5/1991 806 0 33 24-LD 13/8/1991 904 167 28 24-LD 22/10/1991 974 147 25 24-LD 14/1/1992 1058 221 29 24-LD 07/4/1992 1142 69 24-LD 07/7/1992 1233 0 24-LD 22/9/1992 1310 0 24-LD 15/12/1992 1394 29 25-LD 07/3/1989 1 0 0 0 25-LD 18/4/1989 43 1 0 0 0 25-LD 30/5/1989 85 1 0 0 25-LD 11/7/1989 127 1 0 0 25-LD 22/8/1989 169 0 28 25-LD 03/10/1989 211 1 88 52 25-LD 16/11/1989 255 1 62 36 25-LD 06/2/1990 337 104 117 25-LD 26/4/1990 416 250 153 25-LD 26/7/1990 507 832 222 25-LD 18/10/1990 591 117 281 25-LD 03/1/1991 668 235 320 25-LD 07/3/1991 731 0 217 25-LD 30/5/1991 815 0 165 25-LD 27/8/1991 904 147 232 25-LD 05/12/1991 1004 192 212 25-LD 05/3/1992 1095 471 392 25-LD 04/6/1992 1186 98 25-LD 23/7/1992 1235 87 25-LD 08/10/1992 1312 57 25-LD 17/12/1992 1382 60 ' . ' = Sample Not Assayed 124 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 25-LD 01/4/1993 1487 62 25-LD 18/5/1993 1534 47 0 26-LD 25/4/1989 1 0 0 0 0 26-LD 08/6/1989 45 0 0 0 26-LD 20/7/1989 87 1 0 0 0 26-LD 29/8/1989 127 0 0 26-LD 10/10/1989 169 1 0 0 1 26-LD 28/11/1989 218 1 31 25 26-LD 11/1/1990 262 52 48 1 26-LD 29/3/1990 339 146 60 26-LD 21/6/1990 423 250 51 26-LD 13/9/1990 507 0 51 26-LD 06/12/1990 591 0 55 26-LD 28/2/1991 675 0 31 26-LD 18/4/1991 724 37 63 0 26-LD 09/7/1991 806 27-LD 18/7/1989 1 0 0 0 0 27-LD 31/8/1989 45 0 0 0 0 27-LD 10/10/1989 85 1 0 0 0 27-LD 21/11/1989 127 0 0 25 0 27-LD 09/1/1990 176 0 0 0 0 27-LD 27/2/1990 225 0 0 0 27-LD 05/4/1990 262 0 0 0 0 27-LD 12/6/1990 330 0 0 0 0 27-LD 11/9/1990 421 0 0 0 0 27-LD 04/12/1990 505 0 0 0 0 27-LD 26/2/1991 589 0 0 0 0 27-LD 21/5/1991 673 0 0 0 27-LD 09/7/1991 722 0 0 0 0 27-LD 01/10/1991 806 0 0 0 27-LD 17/12/1991 883 0 0 0 27-LD 17/3/1992 974 0 0 0 27-LD 09/6/1992 1058 0 0 0 28-LD 13/3/1990 1 1 0 0 0 28-LD 26/4/1990 45 1 0 0 28-LD 07/6/1990 87 1 0 0 0 28-LD 19/7/1990 129 1 0 0 28-LD 30/8/1990 171 0 0 0 0 28-LD 11/10/1990 213 0 0 28-LD 22/11/1990 255 0 0 0 28-LD 14/2/1991 339 1 0 0 0 28-LD 16/5/1991 430 1 0 0 0 28-LD 01/8/1991 507 1 0 29 0 28-LD 31/10/1991 598 1 0 0 28-LD 23/1/1992 682 1 52 36 28-LD 12/3/1992 731 0 0 0 28-LD 16/6/1992 827 0 0 ' . ' = Sample Not Assayed 125 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 29-LD 27/3/1990 1 0 0 0 0 29-LD 15/5/1990 50 1 0 0 0 29-LD 19/6/1990 85 1 0 0 29-LD 31/7/1990 127 1 0 0 0 29-LD 11/9/1990 169 1 0 24 29-LD 23/10/1990 211 62 52 1 29-LD 04/12/1990 253 73 106 29-LD 26/2/1991 337 33 136 29-LD 21/5/1991 421 73 50 29-LD 20/8/1991 512 125 56 0 29-LD 05/11/1991 589 74 43 0 29-LD 28/1/1992 673 156 48 0 29-LD 24/3/1992 729 0 86 29-LD 26/5/1992 792 0 29-LD 16/6/1992 813 88 59 29-LD 08/9/1992 897 26 29-LD 01/12/1992 981 44 29-LD 09/3/1993 1079 55 29-LD 01/6/1993 1163 41 0 30-LD 28/11/1989 1 0 0 0 0 30-LD 20/2/1990 85 0 0 0 0 30-LD 03/4/1990 127 0 0 0 30-LD 15/5/1990 169 1 0 0 30-LD 03/7/1990 218 1 0 0 30-LD 14/8/1990 260 1 0 0 0 30-LD 30/10/1990 337 62 55 0 30-LD 22/1/1991 421 81 53 0 30-LD 16/4/1991 505 176 54 0 30-LD 09/7/1991 589 192 66 0 30-LD 01/10/1991 673 26 43 0 30-LD 19/11/1991 722 22 45 30-LD 18/2/1992 813 147 0 30-LD 28/4/1992 883 0 0 30-LD 28/7/1992 974 0 30-LD 03/11/1992 1072 0 30-LD 12/1/1993 1142 0 30-LD 06/4/1993 1226 0 30-LD 25/5/1993 1275 0 0 31-LD 10/10/1989 1 0 0 23 31-LD 21/11/1989 43 0 0 0 0 31-LD 02/1/1990 85 0 0 31-LD 27/3/1990 169 0 0 0 0 31-LD 08/5/1990 211 1 0 0 31-LD 22/5/1990 225 0 31-LD 21/6/1990 255 1 0 0 0 31-LD 11/9/1990 337 0 0 0 31-LD 04/12/1990 421 0 0 ' . ' = Sample Not Assayed 126 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 31-LD 12/3/1991 519 1 0 0 31-LD 02/5/1991 570 1 0 0 0 32-LD 14/11/1989 1 0 0 0 32-LD 21/12/1989 38 0 0 0 32-LD 08/2/1990 87 1 0 0 32-LD 22/3/1990 129 1 0 21 0 32-LD 26/4/1990 164 0 32-LD 14/6/1990 213 0 0 0 0 32-LD 26/7/1990 255 0 0 0 32-LD 18/10/1990 339 1 0 0 32-LD 10/1/1991 423 0 0 0 32-LD 04/4/1991 507 1 0 0 32-LD 27/6/1991 591 0 0 0 32-LD 19/9/1991 675 0 0 0 32-LD 14/11/1991 731 1 0 0 0 32-LD 20/2/1992 829 1 0 0 32-LD 30/4/1992 899 0 0 32-LD 23/7/1992 983 0 32-LD 15/10/1992 1067 0 32-LD 14/1/1993 1158 0 32-LD 01/4/1993 1235 0 32-LD 01/6/1993 1296 0 33-LD 24/10/1989 1 0 0 0 0 33-LD 07/12/1989 45 1 0 0 1 33-LD 03/1/1990 72 555 33-LD 18/1/1990 87 0 26 1 33-LD 01/3/1990 129 '57 50 1 33-LD 12/4/1990 171 291 143 1 33-LD 24/5/1990 213 832 480 1 33-LD 12/7/1990 262 0 175 1 33-LD 27/9/1990 339 416 362 1 33-LD 03/1/1991 437 1061 705 33-LD 14/3/1991 507 333 337 ' 1 33-LD 06/6/1991 591 589 200 1 33-LD 05/9/1991 682 1667 342 1 33-LD 17/10/1991 724 236 332 1 33-LD 16/1/1992 815 471 325 33-LD 02/4/1992 892 265 187 33-LD 23/6/1992 974 295 160 33-LD 17/9/1992 1060 111 33-LD 17/12/1992 1151 128 33-LD 04/3/1993 1228 119 33-LD 27/5/1993 1312 122 34-LD 30/11/1989 1 0 0 0 0 34-LD 18/1/1990 50 0 0 0 0 34-LD 22/2/1990 85 0 0 0 0 34-LD I 05/4/1990 127 0 0 0 0 ' . ' = Sample Not Assayed 127 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 34-LD 17/5/1990 169 0 0 0 0 34-LD 14/6/1990 197 1 0 0 0 34-LD 09/8/1990 253 0 0 0 34-LD 01/11/1990 337 0 0 0 0 34-LD 24/1/1991 421 0 0 0 0 34-LD 18/4/1991 505 1 0 43 0 34-LD 11/7/1991 589 0 0 0 0 34-LD 03/10/1991 673 0 0 0 0 34-LD 28/11/1991 729 0 0 0 0 34-LD 20/2/1992 813 1 0 0 34-LD 14/5/1992 897 0 0 0 34-LD 06/8/1992 981 0 34-LD 12/11/1992 1079 0 34-LD 19/1/1993 1147 0 34-LD 15/4/1993 1233 0 34-LD 08/6/1993 1287 0 35-LD 27/2/1990 1 0 0 0 35-LD 10/4/1990 43 1 0 0 0 35-LD 22/5/1990 85 1 0 0 35-LD 03/7/1990 127 1 0 0 35-LD 14/8/1990 169 1 0 0 35-LD 25/9/1990 211 1 0 0 35-LD 06/11/1990 253 22 23 35-LD 15/1/1991 323 44 44 35-LD 16/4/1991 414 147 50 35-LD 16/7/1991 505 104 54 35-LD 27/8/1991 547 236 66 0 35-LD 08/10/1991 589 36-HD 01/11/1988 1 0 0 36-HD 13/12/1988 43 0 0 0 36-HD 24/1/1989 85 0 0 0 0 36-HD 07/3/1989 127 1 0 0 36-HD 20/4/1989 171 1 0 0 0 36-HD 30/5/1989 211 1 0 0 36-HD 11/7/1989 253 1 0 23 0 36-HD 10/10/1989 344 1 0 0 0 36-HD 20/3/1990 505 1 0 0 36-HD 12/6/1990 589 1 0 0 36-HD 04/9/1990 673 1 0 21 36-HD 30/10/1990 729 0 0 0 36-HD 22/1/1991 813 1 0 0 36-HD 16/4/1991 897 1 0 0 36-HD 09/7/1991 981 0 0 36-HD 01/10/1991 1065 1 0 0 36-HD 17/12/1991 1142 1 0 36-HD 17/3/1992 1233 0 36-HD 09/6/1992 1317 1 0 36-HD j 01/9/1992 | 1401 1 1 f \ 1 _ ft. • I A < 0 ' . ' = Sample Not Assayed 128 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 36-HD 25/11/1992 1486 0 36-HD 23/2/1993 1576 0 36-HD 11/5/1993 1653 0 37-HD 08/11/1988 1 0 0 0 37-HD 20/12/1988 43 1 0 0 0 37-HD 31/1/1989 85 1 0 0 37-HD 14/3/1989 127 1 0 0 1 37-HD 21/3/1989 134 0 37-HD 20/4/1989 164 37-HD 25/5/1989 199 1 37-HD 22/6/1989 227 1 0 38-HD 22/11/1988 1 0 0 0 38-HD 03/1/1989 43 0 0 38-HD 14/2/1989 85 1 0 0 38-HD 30/3/1989 129 0 25 0 38-HD 09/5/1989 169 1 52 27 38-HD 15/6/1989 206 1 0 0 0 38-HD 03/8/1989 255 0 0 38-HD 26/10/1989 339 1 0 0 38-HD 12/4/1990 507 0 30 38-HD 12/7/1990 598 0 0 38-HD 27/9/1990 675 0 0 38-HD 15/11/1990 724 0 0 38-HD 07/2/1991 808 0 0 38-HD 02/5/1991 892 0 0 38-HD 25/7/1991 976 0 0 38-HD 10/10/1991 1053 0 0 38-HD 09/1/1992 1144 0 38-HD 02/4/1992 1228 0 38-HD 09/7/1992 1326 0 38-HD 17/9/1992 1396 0 38-HD 18/2/1993 1550 0 39-HD 22/11/1988 1 0 0 0 39-HD 03/1/1989 43 1 0 0 39-HD 14/2/1989 85 1 0 0 39-HD 30/3/1989 129 37 63 39-HD 11/5/1989 171 1 0 48 39-HD 22/6/1989 213 1 264 47 39-HD 03/8/1989 255 0 77 39-HD 26/10/1989 339 333 137 39-HD 18/1/1990 423 208 124 39-HD 12/4/1990 507 499 128 39-HD 05/7/1990 591 333 97 39-HD 27/9/1990 675 0 330 39-HD 15/11/1990 724 235 273 39-HD 21/2/1991 822 21 104 39-HD 02/5/1991 892 1291 506 ' . ' = Sample Not Assayed 129 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 39-HD 25/7/1991 976 1 625 468 39-HD 31/10/1991 1074 707 374 39-HD 09/1/1992 1144 40-HD 07/2/1989 1 0 0 0 0 40-HD 14/3/1989 36 0 0 40-HD 21/3/1989 43 0 40-HD 04/5/1989 87 1 0 0 40-HD 08/6/1989 122 1 0 0 0 40-HD 27/7/1989 171 0 40-HD 07/9/1989 213 1 0 0 0 40-HD 19/10/1989 255 0 0 40-HD 11/1/1990 339 0 0 40-HD 05/4/1990 423 1 0 0 40-HD 05/7/1990 514 1 0 0 40-HD 20/9/1990 591 1 0 46 40-HD 11/12/1990 673 0 0 40-HD 05/2/1991 729 0 0 0 0 40-HD 25/4/1991 808 1 0 0 40-HD 18/7/1991 892 0 0 40-HD 10/10/1991 976 1 0 0 40-HD 09/1/1992 1067 0 0 0 40-HD 19/3/1992 1137 0 40-HD 11/6/1992 1221 0 0 40-HD 24/9/1992 1326 0 40-HD 03/12/1992 1396 0 40-HD 25/2/1993 1480 0 40-HD 01/6/1993 1576 0 41-HD 07/2/1989 1 0 0 0 41-HD 21/3/1989 43 1 0 0 0 41-HD 04/5/1989 87 1 0 0 41-HD 08/6/1989 122 1 0 0 41-HD 27/7/1989 171 0 0 41-HD 07/9/1989 213 1 0 0 41-HD 19/10/1989 255 0 0 41-HD 16/1/1990 344 0 0 41-HD 05/4/1990 423 104 43 41-HD 05/7/1990 514 416 137 41-HD 27/9/1990 598 416 179 41-HD 13/12/1990 675 235 98 41-HD 31/1/1991 724 0 0 41-HD 25/4/1991 808 821 94 41-HD 11/7/1991 885 0 115 41-HD 17/10/1991 983 295 134 41-HD 16/1/1992 1074 265 128 41-HD 02/4/1992 1151 121 41-HD 11/6/1992 1221 29 41-HD 24/9/1992 1326 30 41-HD 03/12/1992 1396 31 ' . ' = Sample Not Assayed 130 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 41-HD 25/2/1993 1480 41-HD 27/5/1993 1571 29 42-HD 07/2/1989 -7 0 0 42-HD 28/3/1989 43 1 0 0 42-HD 09/5/1989 85 1 0 0 0 42-HD 20/6/1989 127 1 0 0 42-HD 03/8/1989 171 1 0 0 0 42-HD 14/9/1989 213 1 0 22 0 42-HD 26/10/1989 255 1 0 0 42-HD 18/1/1990 339 0 0 42-HD 12/4/1990 423 1 0 0 42-HD 05/7/1990 507 1 0 0 42-HD 27/9/1990 591 1 0 0 42-HD 20/12/1990 675 0 0 42-HD 07/2/1991 724 1 0 0 0 42-HD 02/5/1991 808 1 0 0 42-HD 25/7/1991 892 1 0 0 0 42-HD 17/10/1991 976 0 0 0 42-HD 16/1/1992 1067 0 0 0 42-HD 02/4/1992 1144 0 49 42-HD 11/6/1992 1214 0 0 42-HD 17/9/1992 1312 0 42-HD 15/12/1992 1401 0 42-HD 04/3/1993 1480 0 42-HD 27/5/1993 1564 0 43-HD 28/2/1989 1 0 0 0 0 43-HD 13/4/1989 45 1 0 0 43-HD 25/5/1989 87 1 0 0 43-HD 04/7/1989 127 1 0 0 0 43-HD 15/8/1989 169 0 0 43-HD 26/9/1989 211 1 0 0 0 43-HD 07/11/1989 253 1 0 0 43-HD 06/2/1990 344 1 0 32 0 43-HD 24/4/1990 421 1 0 42 0 43-HD 19/7/1990 507 1 0 48 0 43-HD 11/10/1990 591 0 27 0 43-HD 20/12/1990 661 0 33 0 43-HD 09/5/1991 801 1 44-HD 14/3/1989 1 0 0 0 0 44-HD 27/4/1989 45 1 0 0 0 44-HD 08/6/1989 87 1 0 0 44-HD 18/7/1989 127 0 23 0 44-HD 29/8/1989 169 1 0 41 44-HD 02/11/1989 234 26 0 44-HD 21/11/1989 253 1 0 30 44-HD 20/2/1990 344 1 0 46 44-HD 17/5/1990 430 1 0 24 ' . ' = Sample Not Assayed 131 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 44-HD 14/8/1990 519 1 0 32 0 44-HD 30/10/1990 596 0 0 0 44-HD 15/1/1991 673 0 0 23 0 44-HD 05/3/1991 722 1 0 29 44-HD 28/5/1991 806 1 0 21 44-HD 20/8/1991 890 0 0 0 44-HD 12/11/1991 974 1 0 0 44-HD 04/2/1992 1058 1 31 21 44-HD 12/5/1992 1156 0 44-HD 11/8/1992 1247 0 44-HD 27/10/1992 1324 0 44-HD 12/1/1993 1401 0 44-HD 06/4/1993 1485 0 44-HD 08/6/1993 1548 0 0 45-HD 27/6/1989 1 0 0 0 0 45-HD 08/8/1989 43 1 0 0 0 45-HD 19/9/1989 85 1 0 0 45-HD 31/10/1989 127 1 0 0 45-HD 12/12/1989 169 1 0 0 45-HD 23/1/1990 211 0 0 45-HD 06/3/1990 253 1 0 0 45-HD 29/5/1990 337 1 0 0 45-HD 21/8/1990 421 1 0 0 45-HD 13/11/1990 505 0 0 45-HD 27/11/1990 519 0 0 0 45-HD 05/2/1991 589 1 46-HD 20/2/1990 1 0 0 0 46-HD 05/4/1990 45 1 0 55 0 46-HD 15/5/1990 85 1 0 0 46-HD 05/7/1990 136 1 0 0 0 46-HD 16/8/1990 178 1 0 0 46-HD 20/9/1990 213 1 0 93 0 46-HD 01/11/1990 255 0 22 46-HD 24/1/1991 339 1 0 51 46-HD 18/4/1991 423 1 66 37 46-HD 03/10/1991 591 81 55 0 46-HD 09/1/1992 689 118 0 0 46-HD 20/2/1992 731 236 21 46-HD 14/5/1992 815 42 36 46-HD 06/8/1992 899 24 46-HD 29/10/1992 983 21 46-HD 21/1/1993 1067 29 46-HD 15/4/1993 1151 0 46-HD 01/6/1993 1198 0 0 47-HD 22/5/1990 1 1 0 0 1 47-HD 03/7/1990 43 1 0 0 1 47-HD 14/8/1990 85 1 0 0 1 ' . ' = Sample Not Assayed 132 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 47-HD 25/9/1990 127 0 0 0 1 47-HD 06/11/1990 169 0 0 1 47-HD 18/12/1990 211 0 41 1 47-HD 29/1/1991 253 1 42 58 1 47-HD 16/4/1991 330 1 471 113 1 47-HD 16/7/1991 421 943 90 1 47-HD 08/10/1991 505 1768 1171 1 47-HD 07/1/1992 596 589 1085 1 47-HD 31/3/1992 680 943 1177 1 47-HD 12/5/1992 722 3333 1394 1 47-HD 04/8/1992 806 2496 47-HD 27/10/1992 890 6136 47-HD 19/1/1993 974 6810 47-HD 13/4/1993 1058 6324 47-HD 01/6/1993 1107 7015 48-HD 12/9/1989 1 0 0 0 48-HD 24/10/1989 43 1 0 0 0 48-HD 05/12/1989 85 1 0 0 48-HD 18/1/1990 129 0 0 0 48-HD 01/3/1990 171 1 0 0 48-HD 12/4/1990 213 1 0 0 0 48-HD 24/5/1990 255 1 0 0 48-HD 14/8/1990 337 1 0 0 48-HD 08/11/1990 423 0 0 48-HD 29/1/1991 505 1 0 0 48-HD 25/4/1991 591 1 0 41 48-HD 18/7/1991 675 1 0 0 48-HD 05/9/1991 724 1 0 0 0 48-HD 05/12/1991 815 1 0 0 48-HD 20/2/1992 892 1 0 0 48-HD 19/5/1992 981 0 0 48-HD 06/8/1992 1060 0 48-HD 29/10/1992 1144 0 48-HD 21/1/1993 1228 0 48-HD 22/4/1993 1319 0 48-HD 25/5/1993 1352 0 49-HD 03/10/1989 1 0 0 0 49-HD 14/11/1989 43 1 0 0 0 49-HD 21/12/1989 80 0 0 49-HD 06/2/1990 127 1 0 23 1 49-HD 20/3/1990 169 1 0 23 49-HD 24/4/1990 204 1 42 43 1 49-HD 05/6/1990 246 1 208 61 49-HD 28/8/1990 330 0 53 49-HD 18/12/1990 442 0 47 49-HD 19/2/1991 505 0 36 49-HD 21/5/1991 596 0 25 49-HD 06/8/1991 673 0 0 ' . ' = Sample Not Assayed 133 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 49-HD 24/9/1991 722 0 0 0 49-HD 17/12/1991 806 37 0 49-HD 17/3/1992 897 0 0 49-HD 09/6/1992 981 0 0 49-HD 08/9/1992 1072 0 49-HD 01/12/1992 1156 0 49-HD 23/2/1993 1240 0 49-HD 18/5/1993 1324 0 50-HD 17/10/1989 1 0 0 0 0 50-HD 28/11/1989 43 1 0 0 50-HD 16/1/1990 92 0 0 0 50-HD 22/2/1990 129 1 0 53 50-HD 12/4/1990 178 1 0 0 0 50-HD 17/5/1990 213 1 0 0 50-HD 21/6/1990 248 1 0 0 0 50-HD 11/9/1990 330 0 0 0 50-HD 13/12/1990 423 0 0 50-HD 19/3/1991 519 1 0 0 50-HD 30/5/1991 591 0 51-HD 02/1/1989 -365 0 51-HD 02/1/1990 1 0 0 0 0 51-HD 15/2/1990 45 1 0 40 1 51-HD 27/3/1990 85 1 0 0 1 51-HD 10/5/1990 129 1 125 90 1 51-HD 21/6/1990 171 1 208 44 1 51-HD 02/8/1990 213 1 83 50 1 51-HD 13/9/1990 255 208 129 1 51-HD 06/12/1990 339 191 187 1 51-HD 28/2/1991 423 416 625 1 51-HD 23/5/1991 507 9428 2080 1 51-HD 22/8/1991 598 1885 10405 1 51-HD 07/11/1991 675 3771 3251 1 51-HD 02/1/1992 731 471 17080 1 51-HD 26/3/1992 815 707 5843 51-HD 11/6/1992 892 4714 1686 51-HD 09/10/1992 1012 1414 51-HD 03/12/1992 1067 3593 51-HD 18/2/1993 1144 12252 51-HD 06/5/1993 1221 5832 52-HD 06/2/1990 1 0 0 0 52-HD 22/3/1990 45 0 0 0 0 52-HD 10/5/1990 94 1 0 0 52-HD 14/6/1990 129 1 0 0 0 52-HD 26/7/1990 171 1 0 0 52-HD 06/9/1990 213 1 0 0 0 52-HD 18/10/1990 255 0 0 52-HD 17/1/1991 346 0 0 0 ' . ' = Sample Not Assayed 134 PATIENT ID SAMPLE DATE TREATMENT DAY Berlex ELISA (0=Ab-/1=Ab+) AVA (NU/ml) MxA Assay (NU/ml) UBC ELISA (0=Ab-/1=Ab+) 52-HD 11/4/1991 430 1 0 0 52-HD 27/6/1991 507 0 0 0 52-HD 19/9/1991 591 0 0 0 52-HD 12/12/1991 675 1 0 0 52-HD 06/2/1992 731 1 0 0 0 52-HD 05/3/1992 759 1 0 0 52-HD 11/6/1992 857 135 APPENDIX C SUMMARY LIST OF PER PA TIENT ANTIBODY STATUS FOR ALL ASSAYS 136 Patient ID Berlex ELISA AVA MxA assay UBC ELISA 1 -PL + - - -2 -PL + ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^  3 -PL + - -4 -PL + 5 -PL - -6 -PL + 7 -PL + - -8 -PL flPHHHHMHB 9 -PL + - - -10 -PL 11 -PL + _ + 12 -PL + T 13 -PL + - _ _ 14 -PL + p H p P H l 0 H H H H 15 -PL + _ + 16 -PL + 17 -PL + - -18 -LD + + + 19 -LD + - - + 20 -LD + + + + 21 -LD + - + + 22 -LD + + + + 23 -LD + - + + 24 -LD + + + + 25 -LD + + + + 26 -LD ^ ^ ^ ^ ^ ^ ^ ^ ^ + 27 -LD + _ _ _ 28 -LD 29 -LD + + + + 30 -LD + ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ 31 -LD + _ _ _ 32 -LD 33 -LD + + + + 34 -LD 35 -LD + + + + 36 -HD + 37 -HD + _ + 38 -HD 39 -HD + + + + 40 -HD 41 -HD + + + + 42 -HD 43 -HD + _ + _ 44 -HD ^ ^ ^ ^ ^ ^ ^ ^ ^ 45 -HD + - - -46 -HD 47 -HD + + + + 48 -HD 49 -HD + + + + 50 -HD 51 -HD + + + + 52 -HD i. i in i.. - - -= antibody negative 137 

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.831.1-0088986/manifest

Comment

Related Items