IMRT TREATMENT PLANNING FOR ARTERIOVENOUS MALFORMATIONS: PATIENT STRATIFICATION AND DOSIMETRIC QUALITY ASSURANCE by Marcus Sonier B.Sc., University of Victoria, 2008 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Physics) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) January 2013 © Marcus Sonier, 2013 Abstract Stereotactic Radiosurgery is a treatment of choice for Arteriovenous Malformations (AVMs) in anatomical locations associated with a risk of surgical complications. However, SRS has a risk of toxicity due to radiation injury to brain tissue. Therefore, use of intensity-modulated radiotherapy (IMRT) has been advocated because, compared to 3D Conformal Radiotherapy (3DCRT), it leads to improved PTV conformity and Normal Tissue (NT) sparing. The aim of this study was: 1) to develop stratification rules for AVM patients based on benefits they receive from IMRT; 2) to assess optimized dose distributions against prospectively collected data for symptomatic radiation injury; 3) to test and benchmark IMRT QA procedures for patient applications with the iPlan system. Thirty-one AVM patients previously treated with 3DCRT were replanned using static gantry IMRT for BrainLab microMLC using the iPlan system, with the 3DCRT plans as a reference. First, PTV constraints were applied and the conformity of the prescription dose to the PTV was compared between the treatment techniques. Next, NT constraints were introduced into the IMRT plans at the 7 and 12Gy isodoses. These constraints were manipulated to achieve maximum NT sparing while maintaining PTV coverage. Then, NT volumes receiving 7 and 12Gy were compared between the plan types. Finally, ion chamber and film dose verification were performed to scrutinize the accuracy of the IMRT improvements and determine the clinical validity of each plan. Examination of conformity index, NT max dose, and 7 and 12Gy isodose volumes showed a separation of patients into those who did and did not benefit from IMRT for two plan types: PTV Only and OAR Low. For PTV Only, each subset of patients received improvements of 0.10-0.68, 4.0-12.3%, 0-7.072cc, and 0.5-4.496cc, respectively, while, for OAR Low, patients received improvements of 0.10-0.58, 0-6.5%, 1.0-7.952cc, and 0.5-3.704cc, respectively. The 12Gy vol- ume results translated to a decrease in the probability of symptomatic injury by 0.3-11.2% and 0.3-9.3% for PTV Only and OAR Low IMRT. In conclusion, this work indicates the potential for significant patient improvements when treating AVMs and provides rules to predict which patients are likely to benefit from IMRT. ii Preface • A portion of the work contained in each chapter has previously been submitted for publication with JACMP. The initial design of the research was proposed by the supervisors and evolved over time with the students contributions. All the research and data analysis contained within the submission as well as the writing of the manuscript was entirely completed by the student; whereas, the supervisors and co-authors approved the research results and reviewed the manuscript prior to submission, offering suggestions for improvements. At the time of submission of this thesis, the work submitted to JACMP was under review with acceptance pending. • This study was approved by the BC Cancer Agency ethics committee: BC Cancer Agency Research Ethics Board Number H09-00404. iii Table of Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii List of Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Arteriovenous Malformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Stereotactic Radiosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Dosimetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.4 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.1 Treatment Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.1.1 3D Conformal Radiotherapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.1.2 Intensity Modulated Radiotherapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.1.3 Plan Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2 Quality Assurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.2.1 Ion Chamber Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.2.2 Film Dosimetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.1 Patient Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.2 Patient Stratification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.3 IMRT Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.3.1 Ion Chamber Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.3.2 EBT3 Film Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 iv 4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 4.1 Outcomes from Patient Stratification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 4.2 Ion Chamber and Film Dosimetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.2.1 Ion Chamber Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.2.2 Film Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 5 Conclusion and Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 5.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 5.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 A. Patient DVHs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 v List of Tables 1.1 Spetzler-Martin AVM Grading Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.1 PTV Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.2 Ion Chamber Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.3 Film Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.1 Patient Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3.2 PTV Conformity Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 3.3 NT Max Dose Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.4 PTV Max Dose Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.5 7Gy Isodose Volume Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.6 12Gy Isodose Volume Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.7 PTV Conformity Comparisons between IMRT and 3DCRT . . . . . . . . . . . . . . . . . . . . . 37 3.8 NT Max Dose Comparisons between IMRT and 3DCRT . . . . . . . . . . . . . . . . . . . . . . 38 3.9 PTV Max Dose Comparisons between IMRT and 3DCRT . . . . . . . . . . . . . . . . . . . . . . 39 3.10 7Gy Isodose Volume Comparisons between IMRT and 3DCRT . . . . . . . . . . . . . . . . . . . 40 3.11 12Gy Isodose Volume Comparisons between IMRT and 3DCRT . . . . . . . . . . . . . . . . . . 41 3.12 PTV Only IMRT Stratification Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3.13 OAR Low IMRT Stratification Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.14 Ion Chamber Patient Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.15 Percent Difference between Ion Chamber Measurements and TPS . . . . . . . . . . . . . . . . . 44 4.1 IMRT Stratification of Spetzler-Martin Grading Scheme . . . . . . . . . . . . . . . . . . . . . . . 63 vi List of Figures 1.1 X-ray Angiography Image of an AVM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.2 Representations of AVMs by Grade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1.3 Sample 3DCRT plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.4 3DCRT Beam’s Eye View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.5 IMRT Dose Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.6 IMRT Fluence Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.1 BrainLAB m3 microMLC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.2 IMRT Optimization Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.3 OAR Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.4 Cube Phantom Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.5 CIRS Head Phantom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.6 Film Calibration Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.1 Patient Distribution of Conformity Improvements in PTV Only IMRT . . . . . . . . . . . . . . . 44 3.2 Patient Distribution of Conformity Improvements in OAR Low IMRT . . . . . . . . . . . . . . . 45 3.3 Patient Distribution of NT Max Dose Improvements in PTV Only IMRT . . . . . . . . . . . . . . 46 3.4 Patient Distribution of NT Max Dose Improvements in OAR Low IMRT . . . . . . . . . . . . . . 47 3.5 Patient Distribution of 7Gy Isodose Volume Improvements in PTV Only IMRT . . . . . . . . . . 48 3.6 Patient Distribution of 7Gy Isodose Volume Improvements in OAR Low IMRT . . . . . . . . . . 49 3.7 Patient Distribution of 12Gy Isodose Volume Improvements in PTV Only IMRT . . . . . . . . . . 50 3.8 Patient Distribution of 12Gy Isodose Volume Improvements in OAR Low IMRT . . . . . . . . . . 51 3.9 TPS Calculated Dose Distribution for Patient 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.10 Isodose Comparison Map for Patient 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.11 X Direction Dose Profile Cross Section for Patient 2 . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.12 Y Direction Dose Profile Cross Section for Patient 2 . . . . . . . . . . . . . . . . . . . . . . . . . 55 3.13 2D Dose Difference Fluence Map for Patient 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 A.1 DVH Comparisons for Patient 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 A.2 DVH Comparisons for Patient 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 A.3 DVH Comparisons for Patient 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 A.4 DVH Comparisons for Patient 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 A.5 DVH Comparisons for Patient 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 A.6 DVH Comparisons for Patient 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 vii A.7 DVH Comparisons for Patient 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 A.8 DVH Comparisons for Patient 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 A.9 DVH Comparisons for Patient 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 A.10 DVH Comparisons for Patient 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 A.11 DVH Comparisons for Patient 11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 A.12 DVH Comparisons for Patient 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 A.13 DVH Comparisons for Patient 13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 A.14 DVH Comparisons for Patient 14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 A.15 DVH Comparisons for Patient 15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 A.16 DVH Comparisons for Patient 16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 A.17 DVH Comparisons for Patient 17 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 A.18 DVH Comparisons for Patient 18 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 A.19 DVH Comparisons for Patient 19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 A.20 DVH Comparisons for Patient 20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 A.21 DVH Comparisons for Patient 21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 A.22 DVH Comparisons for Patient 22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 A.23 DVH Comparisons for Patient 23 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 A.24 DVH Comparisons for Patient 24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 A.25 DVH Comparisons for Patient 25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 A.26 DVH Comparisons for Patient 26 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 A.27 DVH Comparisons for Patient 27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 A.28 DVH Comparisons for Patient 28 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 A.29 DVH Comparisons for Patient 29 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 A.30 DVH Comparisons for Patient 30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 A.31 DVH Comparisons for Patient 31 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 viii List of Abbreviations 3DCRT 3D Conformal Radiotherapy AVM Arteriovenous Malformation CT Computed Tomography DRR Digitally Reconstructed Radiograph DVH Dose-Volume Histogram GEM Gas Electron Multiplier IMRT Intensity Modulated Radiotherapy Linac Linear Accelerator MLC Multi-Leaf Collimator MOSFET Metal Oxide Semiconductor-Field Effect Transistor MR Magnetic Resonance MU Monitor Unit NT Normal Tissue OAR Organ At Risk OSLD Optical Stimulated Luminescence Detector PTV Planning Target Volume QA Quality Assurance QUANTEC Quantitative Analysis of Normal Tissue Effects in the Clinic SRS Stereotactic Radiosurgery TLD Thermoluminescence Detector TPS Treatment Planning System ix Acknowledgements I would like to thank my research supervisors: Dr. Vitali Moiseenko and Dr. Ermias Gete, for their invaluable assistance throughout all aspects of my MSc degree. They devoted countless hours to ensuring my successful completion of the program and it will always be appreciated. I would also like to extend my appreciation to the Medical Physics staff of the BC Cancer Agency. In par- ticular, Brad Gill, Kurt Luchka, and Rosie Vellani for their advice with treatment planning techniques, Richard Lee for his help with the film processing procedure, and Vince Strgar for his guidance in the use and operation of Linacs. In addition, I would like to thank Dr. Chris Herbert and Dr. Michael McKenzie for their dedicated support throughout the project. Finally, I would like to thank my parents, brother, sisters, and extended family for their continued and endless support of my education. x 1 Introduction 1.1 Arteriovenous Malformations Arteriovenous Malformations (AVMs) are a relatively uncommon lesion in which arteriovenous shunting man- ifests within a complex nidus of vascular connections [1]. Due to this shunting, arterial blood drains directing into an abutting vein with no capillary bed separating the two. The result is a vein being required to withstand a pressure similar to what is found in the feeding artery: a characteristic not typical of veins within the human body. Consequently, AVMs run the risk of rupturing for two reasons: 1) the inability of the veins to endure the high blood pressure, and 2) degradations or deficienies within the malformations arterial or venous cell walls [1]. Furthermore, this condition has been found to present itself in multiple locations throughout the body; but, principally in the head and neck region where the outcome of a rupture could have potentially devastating results [2]. These possible regions include anywhere soft tissue is present: ears, brain, pancreas, small intestine, spine, eyelids, and many others [2-9]. However, the work presented here will focus specifically on the treatment of intracranial AVMs within various brain regions and, as such, alternate AVMs will not be discussed further. Figure 1.1 shows an x-ray angiography image of a confirmed diagnosis of a typical patient’s intracranial AVM with the complex network of malformed blood vessels clearly visible due to the use of contrast during imaging. The occurrence of intracranial AVMs within the population is noted to be within the range of 1.4-4.3% through the discovery of a palpable extent of the lesion during autopsy studies [10]. Despite this, patients symptomatically affected by intracranial AVMs fall into a much lower range of 0.001-0.52% of the population [11-14]. Afflicted patients can present with a selection of symptoms: seizures, headache, stroke-like symptoms, and bleeding that can manifest as mild to severe cases depending on the size and location of the AVM [15]. Unfortunately, regardless of whether or not a patient diagnosed with an AVM presents with symptoms, the risk of a rupture, resulting in a cerebral hemorrhage, is the same. This hemorrhage risk has been extensively studied in the literature and determined to amount to a probability of 0.9-4.0% per year for previously unruptured AVMs [1, 10, 16-21]. Additionally, by assuming this rate, the cumulative lifetime risk of a patient experiencing an AVM related rupture can be approximated by: Lifetime risk (percent)=105 – Patient’s age (years) (1) while the risk of a recurrent hemorrhage is slightly higher immediately following an initial rupture [22-23]. Finally, in the event that a rupture occurs, patient morbidity can be as high as 53-81% with the chances of mortality approximately equal to 10.0-17.6% [18, 24-26]. Intracranial AVMs can be treated using a variety of techniques including: surgery, embolization, or Stereotac- 1 tic Radiosurgery (SRS), any of which can be combined to obtain optimal results. The particular choice of treat- ment depends on the diagnosis grade assigned to a patient’s AVM defined within the system initially suggested by Spetzler and Martin [27]. By considering the size of the lesion, pattern of venous drainage, and neurological elo- quence of adjacent brain tissue, a patient’s AVM can be ranked from one to five to determine the most appropriate treatment strategy. Table 1.1 depicts the Spetzler-Martin grading scheme with the AVM grade equal to a tally of the points assigned per lesion characteristic while figure 1.2 connects each of these grades to pictorial AVMs that may be encountered clinically. If an AVM is deemed grade I or II then surgery is recommended due to the fact that the probability of complications is relatively low and the cure is immediate and permanent. Alternatively, if the AVM has a grade ≥III or is in an anatomical location associated with surgical complications then SRS is the treatment of choice, occasionally within a multidisciplinary approach on an individual case-by-case basis [10]. Additionally, embolization is typically performed alongside alternate treatments to sufficiently reduce surgical or SRS volumes and decrease treatment time requirements [10]. However, SRS solely remains the focus of this thesis; thus, alternate treatment modalities will not be discussed further. 1.2 Stereotactic Radiosurgery Stereotactic Radiosurgery is the treatment of well-defined target structures using multiple beam arrangements to deliver high doses of radiation in a single, highly accurate session [28]. Furthermore, SRS is available in many forms: from the traditional 3D Conformal Radiotherapy (3DCRT) to the more complex treatments of Intensity Modulated Radiotherapy (IMRT) and Volumetric Modulated Arc Therapy [29]. In either case, despite the method of dose delivery differing significantly between techniques, the desired outcome of complete nidal obliteration remains unchanged. In fact, the probability of obtaining this ablative result has been investigated in the literature and is reported to fall into the range of 60-90% with a requirement of 2-3 years to manifest [10, 30-36]. Unfortu- nately, during this extensive period the chances of the AVM hemorrhaging still remain at the estimated 0.9-4.0% (described in Chapter 1.1). Additionally, post-treatment complications can arise and are a result of a SRS session irradiating Normal Tissue (NT), adjacent to an AVM, to a high dose. The chance of these complications occurring has been a primary focus in the literature and is reported to be in the range of 3-7% following a single fraction SRS treatment [10, 31, 34, 37-40]. More recently, investigators have focused on studying factors predictive of both complete AVM obliteration and SRS related complications. The conclusions from these studies have suggested that complete AVM oblitera- tion depends principally on the minimum dose to the Planning Target Volume (PTV), advised to be >18Gy [16, 36, 41-42]. Conversely, two distinct factors associated with complications from SRS were found to be strong 2 predictors of symptomatic injury: Vx, volume receiving x Gy or more, and conformity of the dose distribution encompassing the PTV [32, 34, 38, 40, 43-45]. In fact, several studies have presented dose-volume predictors for the incidence of radiation-induced brain injury. A selection of these studies are illustrated in a recent Quantitative Analysis of Normal Tissue Effects in the Clinic (QUANTEC) report which summarizes toxicity data from a se- lection of studies and demonstrates that V10 and V12 are particularly strong and consistently observed predictors, with V12 serving as a predictor of choice [45]. Similarly, a strong correlation between various Vx and subsequent brain injury have also been reported with lower dose cut-offs, e.g., V7, observed to be strong predictors as well [44]. SRS plans, therefore, must be balanced so that the risk of symptomatic brain injury will not outweigh the likelihood of nidus obliteration. The work conducted in this thesis focuses on two of the available SRS techniques: 3DCRT and IMRT. Each technique is similar in that the treatment planner must select the number of beams as well as their orientations (angle with respect to the patient); however, the method of treatment optimization and dose delivery is where they differ significantly. Comparatively, 3DCRT is the simpler treatment technique utilizing a forward-planning approach dependent upon: 1) the delineation of the target and critical structures, and 2) the dose prescription [46]. This procedure in formulating a patient plan is essentially a trial-and-error process in which the planner selects a Multi-Leaf Collimator (MLC) configuration for each beam. These configurations then allow for an irregularly shaped beam of uniform intensity to irradiate the PTV as a whole, imparting the prescribed dose. Once this process is completed for each beam orientation, the resultant dose distribution can be studied to determine if target volume coverage with the prescribed dose has been achieved with constraints on dose homogeneity met. If dose coverage of the 3D target volume is not obtained or if hotspots are present within this volume or adjacent normal tissue then the planner must reorient selected beams and/or alter individual MLC positions to produce a more acceptable dose distribution. This new dose result can again be investigated for target volume coverage and hotspots, repeating the process of beam alterations if necessary. Figure 1.3 shows 3DCRT beam orientations for a select patient illustrating the arrangement of multiple beams irradiating a single target volume at isocenter. By spreading the prescription dose out among multiple beams the dose to normal tissue is reduced while dose homogeneity, across the target volume, is improved provided the beams overlap minimally outside the PTV. Additionally, figure 1.4 shows the MLC configuration for a single beam, from the treatment plan depicted in figure 1.3, through the beam’s eye view. Here, each individual MLC can be adjusted to increase/decrease dose coverage by retracting/extending the appropriate MLC. Contrary to 3DCRT, IMRT is much more computationally expensive using an inverse-planning strategy. With this technique, non-uniform intensities are optimally assigned to minuscule subdivisions of beams, or beamlets. These beamlets are produced through the use of MLCs blocking out portions of the beam at different times, irradiating a subsection of the PTV. By irradiating individual components of the PTV for varying times, a custom- 3 design dose distribution is created with a sharp dose fall-off at the boundaries of the target volume [46]. Currently, there are two methods utilized to deliver a desired dose distribution to a target volume using the MLCs: step-and- shoot and sliding window IMRT [47-48]. In step-and-shoot IMRT, the MLCs move between stationary positions while the beam is turned off with the beam turning on for a predetermined number of Monitor Units (MUs), a measurement of machine output correlated to dose, at each position. Conversely, in sliding window IMRT, the beam remains on at all times while the MLCs move continuously to deliver a fluence map, to the PTV, calculated by the treatment planning system. In either case, a non-uniform dose is distributed across the PTV for a discrete beam, containing as many as hundreds of beamlets, and only by combining multiple beams is the desired dose coverage of the target volume achieved. Importantly, both IMRT methods produce treatment plans of comparable quality; however, sliding window IMRT contains an inherent limitation on the speed of MLC motion which may inhibit the delivery of complex fluence maps. In this thesis, step-and-shoot IMRT is considered due to the fact that potential MLC motion limitations are avoided and that dose distributions are more accurately modeled with the iPlan Treatment Planning System (TPS), producing a reduced uncertainty in the dose imparted to a patient or phantom, when compared to sliding window IMRT [49]. Generating an IMRT plan first requires constraints to be placed on the dose to the target volume and any critical structures. These constraints typically involve multiple points for a single structure such as: minimum, or prescription, and maximum dose to the PTV as well as maximum doses to various percentages of the total volume of a critical structure. Using these constraints, the system employs an iterative process to minimize or maximize a chosen cost function, depending on the function itself [50]. The system allows for the constraints to be violated at an additional cost to this function; therefore pushing the optimal plan to deliver the prescribed dose to the PTV while minimizing the dose to adjacent critical structures, the ideal result. Figure 1.5 shows a sample Dose-Volume Histogram (DVH) with constraints set at various dose and volume points on NT and the PTV, specified for IMRT optimization. Furthermore, figure 1.6 shows a single beam’s eye view from a sample patient plan with multiple IMRT beams irradiating a target volume. In this case, each beam contains its own individual fluence map, similar to what is observed in the figure, which illustrates the variation of intensity across the beam and is a result of the beamlets selected for treatment. The differences between 3DCRT and IMRT lead to various advantages and disadvantages for each technique that come from their distinct natures: where one focuses on producing a dose and investigating if the dose con- straints are met while the other employs an optimization procedure specifying the constraints first then examining the resultant dose distribution. The advantages of IMRT over 3DCRT begin with the use of an optimization pro- cedure and individual beamlets. With this, IMRT is able to produce dose distributions that form patterns much more complex than that of 3DCRT resulting in an improved dose conformity to the PTV, particularly in the case of concave target volumes [51-52]. Furthermore, any desired dose inhomogeneities can be imparted to the irradiated 4 region, referred to as dose painting, to reduce the dose in subsections of the PTV adjacent to or partially within a critical structure or increase the dose in boost regions declared by an oncologist [51-52]. Additionally, dose distributions can be created that are more conformal than those of 3DCRT with a sharper fall-off at the boundaries of the treatment volume, eliminating hotspots and increasing NT sparing [46]. This emphasizes the importance of accurate delineation of target and critical structures since errors could result in overdosing of NT or underdosing of the target volume. While there are many potential benefits to the use of IMRT over 3DCRT, there are also substantial disadvan- tages. Each of the major disadvantages comes from a single factor: increased time requirements [52]. In order to achieve accurate structure delineation there is the need for a clinician to spend more time ensuring the accuracy of critical structure contours. There is also an increase in the treatment time for each patient treated with fixed gantry IMRT due to the use of many beamlets at varying intensities rather than a single PTV encompassing beam at a uniform intensity, reducing patient throughput. In fact, the increased treatment time also results in an increased total body dose for the patient due to an increased amount of radiation, or beam-on time, needed for all the indi- vidual beamlets causing an escalation in the total leakage and scatter escaping from the Linear Accelerator (Linac) [52]. Notably, the use of MLC modulation and the extended beam-on times result in an increased workload, or strain, on a Linac and its MLCs requiring additional and more extensive system checks to be administered [53- 54]. Finally, individual patient treatment verification practices need more time to complete due to the increased complexity of the beam characteristics and dose delivery (to be discussed in detail in Chapter 1.3). Ultimately, the subtle differences between 3DCRT and IMRT can result in largely different treatment plans, with identical beam arrangements, for a specific patient which may, in some cases, drastically improve the thera- peutic outcome. However, before any patient can be treated with IMRT to receive the potential benefits, the TPS calculated doses and any Monte Carlo simulations must be meticulously verified. The process of this verification is referred to as Quality Assurance (QA) and uses stringently established dosimetric protocols to validate that the MUs, suggested by the TPS, deliver the required dose. 1.3 Dosimetry Patient specific QA is an important and necessary step when performing SRS and, especially, when utilizing multiple beam arrangements or complex treatment techniques such as IMRT. In the case of SRS, AVM patients are prescribed relatively high doses (as described in Chapter 1.2) that may produce acute or long-term complications if a discrepancy between the intended treatment volume and the clinically irradiated volume occurs. Thus, the treatment plan must be verified for adequate dose fall-off at the AVM boundaries along with ensuring that a sufficient dose is achieved for complete obliteration. Whereas in IMRT, due to the modulation of the beam 5 from the distribution of MUs throughout the step-and-shoot delivery, the resultant dose to an individual point is difficult to mathematically determine through a check of beam MUs [55]. Therefore, accurate dosimetric measurements are required to ensure the desired dose distribution is produced from the elaborate and non-uniform intensity of each beam. Moreover, when performing these verification measurements, alternative QA procedures can simultaneously be considered, including collision checks to ensure that each patient position for a select beam results in adequate clearance between the gantry and head frame immobilization/positioning equipment attached to the couch. To verify dose at various locations in a treatment plan and approve such a plan for clinical release, a wide vari- ety of dosimetric tools are available and employed for measuring and recording point doses as well as 2D and 3D dose distributions. In point dose measurements, a selection of available dosimeters include: ion chambers, silicon diodes, diamond detectors, Metal Oxide Semiconductor-Field Effect Transistors (MOSFETs), Optical Stimulated Luminescence Detectors (OSLDs), Thermoluminescence Detectors (TLDs), Calorimeters, and Alanine Detectors [56-59]. However, due to the availability, ease of use, high accuracy, and absolute dose results, the ion chamber is the sole point dose dosimeter employed in the examination of the dose to the isocenter in this thesis; thus, alter- nate point dose dosimeters will not be discussed further. Ion chambers are considered the gold standard and relied upon to validate isocenter doses; however, it is important to note that due to their finite volume, they essentially measure a mean dose encompassing the isocenter together with adjacent points rather than an infinitesimally small point dose [60]. Due to this, the dose measured by the ion chamber is rarely expected to coordinate precisely with the results given by the treatment planning system; therefore, measurements within 3% of the TPS are considered acceptable [61-62]. Consequently, this 3% threshold for the pass/fail of a patient plan includes the uncertainties in MLC leaf positions during patient treatments which has been found to intermittently contribute to slight dose discrepancies [61]. Finally, the calibration of an ion chamber before use is an important step that may involve numerous calibration factors and extensive evaluation in order to achieve precise results; however, if the dose per MU is explicitly known for a particular phantom set-up then a single factor relating electrometer reading to dose can be calculated and is all that is required to obtain accurate absolute dose results (procedure to be discussed in Chapter 2.2). In 2D dose distribution measurements, a selection of available dosimeters include: scintillators, amorphous silicon detectors, Gas Electron Multiplier (GEM) chambers, and film; any of which can also be used to verify point doses [56, 63-65]. In IMRT, it is desirable to compare the computed dose distribution in the plane of the isocenter with a measured dose distribution to ensure modulation of the beam produced the desired result; therefore, film was chosen for its unique properties and alternate 2D dosimeters will not be discussed further. The two intrinsic qualities of film that make it an ideal dosimeter when performing QA verification of stereotactic IMRT treatment plans are: 1) high-spatial resolution; and 2) large dose response range [61, 66-67]. Steep dose 6 gradients are characteristic of IMRT beams and, with the high-spatial resolution of film, these extreme dose gradients are accurately measured and reproduced. Furthermore, the large dose response range makes film ideal for examining SRS dose distributions that can range from low doses in NT on the order of 0.1-1Gy to high doses across the treatment volume up to approximately 25Gy. Unfortunately, absolute film dosimetry can be inaccurate and unreliable due to the dependency on the performance of the optical scanner in use and the meticulous attention to detail throughout the handling and calibration procedures; thus, film results within 5% of the computed dose are deemed acceptable [61, 66, 68-69]. In 3D dose distribution measurements, available dosimeters can also be used in 2D or point dose verification and include: polymer gels and PRESAGE® (Heuris Inc., Skillman, New Jersey, USA) [56, 70-72]. Polymer gels, in fact, have a number of practical issues that limit their usefulness in the clinic including: interference of oxygen with the polymerization procedure, ambient light degrading gel radiosensitivity, imaging sensitivity varying with temperature, and repeated skin contact causing nervous system disorders [71]. In the end, although 3D dosimetry is a valuable tool, neither polymer gels or PRESAGE® were available for the work presented in this thesis; con- sequently, 3D measurements of IMRT plans could not be performed. 1.4 Purpose For SRS to safely obliterate AVMs with a sufficiently high dose while sparing NT adequately to avoid com- plications, the dose prescribed to the AVM must be highly conformal with the dose distribution maintaining a steep gradient beyond the AVM; thus, minimizing the volume of NT irradiated to a high dose [38]. This objec- tive can be particularly difficult to achieve for complex-shaped AVMs whose surfaces rapidly vary from convex to concave. As a result, IMRT has been advocated as a means to improve plan conformity and thereby reduce the risk of complications [16]. IMRT implementation, however, is resource-intensive. In addition to planning, patient-specific QA is required, usually involving ion chamber measurements at the isocenter and a comparison of planned and delivered fluence maps through high resolution film dosimetry. Furthermore, since AVM patients are often treated using invasive frames, longer wait times between putting on a frame and delivering a treatment plan can also affect a patient’s comfort. The purpose of this work is then to find a subset of patients that will benefit from utilizing frameless IMRT. For these patients, IMRT plans will achieve NT sparing superior to the sparing possible with 3DCRT while not compromising PTV coverage. Importantly, to optimize operation of the clinic, patients who stand to benefit from IMRT should be identified based on the size and shape of their AVM prior to the treatment planning process. Therefore, the specific aims of this thesis are threefold: 1) to develop simple, but reliable, stratification rules for AVM patients based on the benefits they receive from IMRT compared to 3DCRT; 2) to assess optimized dose distributions against prospectively collected data for symptomatic radiation injury 7 following SRS; 3) to test and benchmark IMRT QA procedures for patient applications with the iPlan treatment planning system. Table 1.1: Spetzler-Martin AVM Grading Scheme The grading scheme originally developed by Spetzler and Martin is shown. The overall grade of the AVM is the sum of the points assigned for size, eloquence, and venous drainage. Reprinted with permission of AANS, Journal of Neurosurgery [27]. Graded Feature Points Assigned Size of AVM: Small (<3cm) 1 Medium (3-6cm) 2 Large (>6cm) 3 Eloquence of Adjacent Brain: Non-Eloquent 0 Eloquent 1 Pattern of Venous Drainage: Superficial Only 0 Deep 1 8 Figure 1.1: X-ray Angiography Image of an AVM A typical x-ray angirography image from a sample patient is shown. The vertical and horizontal lines are metallic rods within a frame, positioned around the patients head at the time of image acquisition, to allow for depth localization of the AVM. In this case, the application of a contrast agent illustrates the location of the AVM marked by the red ellipse. The complexity of this malformation can be understood from the immeasurable number of blood vessels within the small nidal network. 9 Figure 1.2: Representations of AVMs by Grade Possible AVMs from all potential grade combinations (in table 1.1) are shown. Reprinted with permission of AANS, Journal of Neurosurgery [27]. 10 Figure 1.3: Sample 3DCRT plan A typical 3DCRT plan is shown. Multiple beams are used to irradiate the PTV reducing the dose deposited in NT. 11 Figure 1.4: 3DCRT Beam’s Eye View The 3DCRT Beam’s Eye View of the highlighted beam in figure 1.3 is shown. Each individual MLC with select leaves retracted, to allow the beam to irradiate the PTV (pink), and others extended, to protect the OARs: brainstem and eye (green), can be seen. 12 Figure 1.5: IMRT Dose Constraints A DVH for a sample patient with PTV and NT constraints for an OAR Low plan is shown. The constraints on the PTV (pink) ensure 100% PTV coverage is achieved, with the prescribed isodose, while the OAR (green) constraints are positioned at points below the DVH to force the optimization procedure to push for maximum NT sparing. 13 Figure 1.6: IMRT Fluence Map An IMRT Beam’s Eye View fluence map for the same beam in figure 1.4 is shown. The intensity modulation across the beam, for this OAR Low IMRT plan, is represented with varying shades of gray. Darker shades show regions where few MUs are calculated while lighter regions receive a higher dose with more MUs delivered. 14 2 Methods 2.1 Treatment Planning Prior to treatment planning, Computed Tomography (CT), Magnetic Resonance (MR), and Angiogram image sets for thirty-one AVM patients previously treated at the BC Cancer Agency were obtained. These image sets were stored in the province wide database CAIS upon the completion of each radiotherapy treatment for the availability of patient follow-up and use in retrospective studies, such as this. Each image set was then transferred to the iPlan TPS for use with the BrainLab m3 microMLC software (BrainLAB AG, Heimstetten, Germany). The m3 was designed as a detachable mount from the gantry head with the specific purpose of producing highly conformal beams for the treatment of small and oddly shaped lesions, such as AVMs. It is made up of 52 tungsten leaves (26 on either side) that are individually motorized with widths increasing from 3mm at the isocenter to 5.5mm on the outer edges at the plane of the isocenter, producing higher resolution accuracy than the standard larger width MLCs contained within the gantry. Figure 2.1 shows the BrainLab m3 microMLC. In order to begin planning within the iPlan system, the CT and Angiogram images were localized by the TPS, through the identification of high density metallic rods within a localization box present at the time of scanning, and fused to the MR images. Fusion of the CT and MR image sets was done automatically by the iPlan system by shifting and rotating the reconstructed images so that bone structures present on both CT and MR scans were aligned. This alignment was then verified by a radiation oncologist specializing in SRS who then performed man- ual contouring of each patient’s AVM. Next, each AVM was grown, by applying a 1mm margin in 3-dimensions, in order to produce the PTV to which dose was prescribed. This AVM expansion was necessary so as to account for the uncertainty associated with patient set-up and alignment at the treatment unit which contains inherent tolerances of up to 1mm. At this point, the volume and diameter, defined as the largest dimension in any plane and not correlated with volume, of the PTV were recorded and treatment plans utilizing 3DCRT and static gantry IMRT were produced. Table 2.1 shows the PTV characteristics for all patients. 2.1.1 3D Conformal Radiotherapy In 3DCRT, plans previously constructed for patient treatment were consulted as a guideline for the positioning and number of beams used in irradiating AVMs at each anatomical location. With these guidelines, single isocen- ter treatment plans were produced with PTV coverage as the primary objective and Organ At Risk (OAR), or NT, sparing considered important. For each patient in the planning process, dose was prescribed by an oncologist specializing in SRS to the isocenter at the center of the PTV. MLC shapes were then manually adjusted to achieve PTV coverage, with the 15 80% isodose, with maximum NT sparing. This was achieved through the positioning of multiple fields such that beam overlap outside the PTV was minimized while meticulous manipulation of the MLCs ensured excess dose was removed from NT and PTV coverage was obtained. In essence, this optimization procedure for each patient plan mimicked the clinical process, fulfilling the requirements necessary for each plan to be suitable for clinical use. PTV coverage was then validated in the axial, coronal, and sagittal slice views on both CT and MR image sets and in the DVH view ensuring that the primary priority of PTV coverage was achieved. Upon the completion of a patient’s plan, the conformity index for the 80% isodose, defined as a ratio of the total volume (PTV + NT) to the PTV volume receiving the indicated dose, was recorded. Additionally, the volume of NT receiving 7 and 12Gy, and the maximum dose to the PTV and NT were recorded. 2.1.2 Intensity Modulated Radiotherapy In IMRT, the MLC settings applied in the 3DCRT treatments were reset to their initial settings and each plan was converted to an IMRT plan to maintain identical beam arrangements. The dose to the PTV margin, corresponding to the 80% isodose in the 3DCRT plans, was specified as a hard constraint to ensure PTV coverage. In contrast to prescribing the dose to the isocenter, as in 3DCRT, in IMRT PTV coverage with the desired isodose is set as an objective. This produces the most uniform dose distribution across the PTV and, thus, the most clinically acceptable plan. For accurate dose delivery to the PTV while maintaining maximum NT sparing, the dose optimization settings were selected as follows: Beamlet size max = 1mm (without forcefully aligning beamlets), Step-and-Shoot leaf sequencing, and Tongue-and-Groove Optimization for MLC leaf positioning. First of all, a beamlet size max of 1mm was selected so as to increase the number of beamlets across the PTV with maximum width of 1mm and, by doing so, allow the system to introduce increased beam modulation resulting in a dose distribution that is highly conformal to the shape of the PTV. The option to forcefully align beamlets was rejected as this would impose a 1mm width on each and every beamlet while also aligning them to the isocenter of the PTV. This would then cause the optimization procedure to utilize enough beamlets to cover the entire extent of the PTV but would not modulate their size at the PTV boundaries, instead creating an overhang at the PTVs edges and allowing a larger volume of NT to be irradiated. Next, Step-and-Shoot leaf sequencing was selected since, as described in Chapter 1.2, it is more accurately modeled by the TPS resulting in less of a discrepancy between planned and measured dose values. Finally, the Tongue-and-Groove Optimization was selected to restrict the PTV underdosing effect from tongue or groove beam attenuation, a well-known and accepted limitation in IMRT treatments. Studies in the literature have shown that if the tongue or groove of an MLC covers a portion of the PTV for any extended duration then significant underdosing by as much as 10-15% can occur [73]. Therefore, by having iPlan consider 16 this problem during the optimization procedure, individual leaves can be set such that pieces of the PTV do not remained covered by a tongue or groove for an extensive period of time, minimizing the underdosing effect. Figure 2.2 shows a screenshot of the IMRT optimization window, with the above settings indicated, from a sample patient plan. Once the IMRT settings were optimized to obtain maximum NT sparing, a sample plan was generated to de- termine the volume of NT enclosed by the 7Gy isodose. With this sample plan, multiple distance measurements, across a collection of image slices in all three planes, were performed from the center of the PTV to the 7Gy isodose. Each measurement was then noted and the largest distance plus 1mm was selected to be the radius of the NT sphere. With this radius recorded, a NT OAR sphere fully enclosing the 7Gy volume was produced with PTV removed. Figure 2.3 illustrates the OAR sphere encompassing the PTV, within a subtracted section, as well as notable isodoses ≥7Gy. This OAR sphere then allowed for NT hard constraints to be input into the IMRT optimization procedure with varying weighting as compared to PTV coverage. It was then with these weights that four variations of an IMRT treatment plan were produced: PTV Only, OAR Low, OAR Medium, and OAR High, with no initial consideration for NT constraints followed by increasing weighting of 33%, 66%, and 100%, respectively. In order to produce these four distinct plans, the NT constraints were applied at the dose-volumes of interest, 7 and 12Gy, and set at percentages of the total sphere volume such that the optimization procedure could not simultaneously fulfill the NT constraints while achieving PTV coverage. In fact, the position of the NT con- straints followed a trial-and-error process in order to achieve maximum NT sparing without pushing so hard on the NT constraints that PTV coverage was sacrificed. Importantly, in each case, before accepting the IMRT plan, PTV coverage was checked by viewing the DVH and ensuring that the requirement of≥99.9% of the PTV volume receiving the prescribed dose was achieved. Notably, when viewing isodose distributions across image slices of the PTV, coverage appeared to fall unacceptably below 100%; however, this is due to the system being unable to accurately display correct isodose distributions in regions of a high-dose gradient. Due to this, PTV coverage cannot be determined through visual inspection; thus, it was sufficient to investigate coverage through analysis of the DVH solely. Consequently, the variations between each IMRT plan manifested primarily in differences in the volumes of NT receiving 7 and 12Gy as well as major changes in PTV/NT hotspots and maximum dose. Finally, each of these values were recorded for the four plans and, similarly to the 3DCRT plans, the conformity of the dose prescribed to the PTV margin, for each plan, was also recorded. 2.1.3 Plan Comparisons Following the completion of all radiotherapy plans, the dosimetric parameters for each IMRT plan were recorded and compared directly to the 3DCRT results in order to determine the extent of improvements ob- 17 tained with IMRT. Each change in value was then calculated and patients were organized into preliminary groups depending on the benefits received from IMRT for each parameter. Next, using the physical dimensions of the PTV and the change in volume due to the 1mm expansion from the AVM nidus to the PTV to estimate surface area, various plots were constructed from the patient specific data. These plots depicted the improvements from the selected IMRT plan that showed the largest benefit for the majority of patients and were generated for all pos- sible combinations of PTV Volume/Surface Area (Vol./SA), PTV volume, and PTV diameter, for each dosimetric parameter, to find any visual distribution of benefits that clearly indicated potential stratification rules. This distri- bution was also found by shifting the dividing line for IMRT benefits between groups in a trial-and-error process until the majority of patients clustered together fell into the same beneficial range. Finally, all plots in which these rules could be identified were chosen for stratification criteria. 2.2 Quality Assurance As described in Chapter 1, IMRT plans require extensive QA procedures in order to verify that the MUs given, for a particular field, deliver the calculated dose. In order to determine if the dose within the iPlan TPS is actually what a patient is treated with, two independent dose measurements were completed on two separate phantoms. First, these measurements included the use of a Type CC01 Ion Chamber (IBA, Tennessee, USA), the gold standard for point dose verification in radiotherapy, and Precision Electrometer/Dosimeter Model 530 (Victoreen, Elimpex-Medizintechnik, Mödling, Austria) together with a cube phantom for isocenter dose verifi- cation. This cube phantom was made of a polystyrene material that reproduced the interactions of radiation in water and, subsequently, soft tissue and had dimensions 18.5x18.5x18.5cm3. Figure 2.4 shows the cube phantom with ion chamber inserted and connections extending to the electrometer which displays the charge collected in the sensitive volume of the ion chamber when irradiated. Second, Gafchromic EBT3 film (International Specialty Products, Wayne, NJ, USA), a high resolution 2D dose reconstruction tool, together with a CIRS Model 605 Ra- diosurgery Head Phantom (CIRS, Virginia, USA) was utilized for 2D dose plane verification. This head phantom was constructed to contain a mixture of materials: one that modeled radiation interactions in soft tissue (skin and brain) and another that accurately modeled bone inhomogeneities present in patients, and contained an insert for the positioning of dosimetric tools. Figure 2.5 depicts the CIRS head phantom with film insert as well as the stereotactic mask used for positioning. 2.2.1 Ion Chamber Verification To perform ion chamber measurements, ten PTV Only IMRT treatment plans were randomly selected: patients 2, 4, 8, 9, 11, 13, 14, 26, 27, and 31, and their respective MUs were imported to the Phantom Planning package in 18 iPlan directly onto the CT reconstruction of the cube phantom. The MUs from each beam were then renormalized, by 5 for patient 13 who had the largest PTV with the lowest prescription dose and by 10 for the other nine patients, to reduce the beam-on time of each plan. The dose to the isocenter was then recalculated and recorded. After completing this step for all patients, the plans were exported to the treatment unit. At this point, the cube phantom was set on the couch such that the geometric center of the cube, where the ion chamber is inserted, was located at the isocenter of the Linac. Brainlab’s m3 microMLC was then mounted onto the gantry head and initialzed using the computers at the treatment station. With the experimental setup complete, the ion chamber was calibrated for the current conditons within the surounding environment. For calibration, a 4.2x4.2cm2 field, 6MV beam was used to irradiate the ion chamber, enclosed within the phantom at isocenter, delivering 200MUs. This 200MUs translated directly to a known dose of 137.84cGy for the current setup and, when combined with the nC/200MU reading measured with the electrometer, a calibration factor for converting nC to cGy was obtained. This calculation for the calibration was repeated three to four times depending on the level of variation between electrometer readings and then the values were averaged to obtain a single calibration factor for the daily readings. Table 2.2 depicts the electrometer readings, averaged readings, and calculated calibration factors for each day that ion chambers measurements were conducted. Next, the electrometer reading for each beam within a patient plan was recorded upon delivery and the resultant dose was calculated by applying the calibration factor. The total dose at the isocenter of the patient plan was then determined by summing the dose delivered from each individual beam. Finally, this total dose was compared with the estimated dose from iPlan using the equation: iPlan Dose - Measured Dose iPlan Dose ∗100% (2) to determine if the 3% pass/fail criteria was achieved. 2.2.2 Film Dosimetry Before performing film dosimetric measurements using the CIRS head phantom, a calibration curve for the conversion of optical density to Gy was created. This procedure utilized an EBT2 film calibration protocol, set out by Micke et al., as a basis in which nine pieces of film are uniformly irradiated with a known dose and matched with the subsequent optical density [74]. First, the calibration irradiations were done by placing eight pieces of 10x10cm2 film at 5cm beyond the Linac isocenter and another full piece of film (20x25cm2) on the floor of the treatment room, at 1.125m beyond the Linac isocenter. This final piece was film was placed on the floor in order to take advantage of beam divergence to create a field large enough to irradiate the film sheet as a whole. In each case, 5cm of solid water was located proximal to the film to ensure electronic equilibrium was 19 achieved before the beam reached the film and 15cm of solid water was located distal to the film to allow for maximum backscatter conditions. Table 2.3 shows the MUs used for each irradiation and the known dose at the films location associated with the MU values. It was important to extend the calibration film doses above the maximum dose that was expected for the patient plans in order to prevent any data points from falling outside of the curve during the processing procedure. Next, after letting the films sit for 48 hours in complete darkness, to allow for any changes in optical density to subside, the films were scanned using an Epson Expression 10000XL scanner (Epson, Markham, Ontario, Canada) in 48-bit color transition mode at 72dpi with the color correction option turned off. Three warm-up scans were also performed with no film present to allow for any electronic noise to subside. Each film was placed at the center of the scanner in order to avoid undesirable edge effects such as: uneven light trasmitted through the film resulting in small, but noticeable, deviations across the film, contributing error to the final dose values. This error would then affect the final dose measurements in such a way as to indicate a disagreement between the planned and delivered fluence maps when, in actuality, this was not the case. The calibration film scans were then saved to be used together with the patient measurements within in-house software, a GUI designated as FUGU, specifically developed for film processing from the recommendations of Micke et al. [74]. With the calibration films prepared, film dosimetric measurements of patient plans were ready to be per- formed. First, three patients: 2, 4, and 26, with varying PTV dimensions were selected from the ten patients in which isocenter dose verification was performed. Next, the CIRS head phantom was fitted with a frameless stereotactic mask to maintain consistent phantom positioning, reducing the possibility of rotations and translations between the treatment planning orientations and beam delivery at the treatment unit. Prior to treatment planning, the head phantom was CT scanned using stereotactic protocols: 1.25mm slice thickness beginning at the base of the skull and extending superiorly to the top of the head. These scans were exported from the CT reconstruction station to iPlan and localized using a similar method to that of the patients in iPlan. Each of the chosen patient plans were imported to Phantom Planning in iPlan directly onto the CT reconstruction of the CIRS head phantom. MU renormalization was again performed, this time by a factor of 4 for all patients, and the dose plane at the isocenter was recalculated and exported as an ASCII file with a dose measurement resolution of 1mm. After completing this step for all patients, the plans were exported to the treatment unit. Once at the treatment unit, the positioning of the phantom was accomplished using ExacTrac (BrainLAB AG, Heimstetten, Germany). This process involved the use of two x-rays at different angles to check the alignment of the phantom’s bone anatomy with that of the Digitally Reconstructed Radiograph (DRR) calculated from the CT reconstruction. The position- ing of the phantom was then double-checked via ExacTrac to verify that maximum accuracy was accomplished when taking the inherent uncertainty of the system into account. With the correct phantom orientation, Brainlab’s m3 microMLC was connected to the gantry head and initialized using the treatment station computers. Following 20 this, the three patient plans were delivered with the film being exchanged for a fresh piece between irradiations. Upon the completion of all patient plans, the patient films were scanned using the same process as the calibra- tion films and both were input together into FUGU which simultaneously used the calibration films to construct a calibration curve then converted the patient film measurements to an ASCII file with dose values. Within this process, FUGU automatically looks for the edges of each film piece then moves 2cm towards the center of the film taking an average reading of the resultant square field and relating the optical density to the known dose delivered. Then, once the average optical density of each film has been determined by FUGU, an optical density versus dose plot is created and fitted with a third order polynomial. Future doses can then be easily computed by simply using the curve to convert optical dentical to dose electronically. Figure 2.6 shows the calibration curve obtained from FUGU. Consequently, with two ASCII files: one representing the film measurements and the other representing the dose plane exported from iPlan, the OmniPro I’mRT software (IBA, Tennessee, USA) was used to validate the IMRT treatment plans. 21 Table 2.1: PTV Dimensions The individual patient PTV dimensions are shown. For each patient, the diameter of the PTV was measured by an oncologist immediately following contouring while the volume was estimated by the iPlan TPS. Surface Area was calculated by observing the difference in the change in volume from the AVM to the PTV caused by the 1mm expansion then dividing this value by the expansion (0.1cm). Patient Diameter (cm) Volume (cc) Surface Area (cm2) 1 2.47 6.824 26.17 2 2.87 5.359 22.99 3 2.19 5.471 21.18 4 0.99 0.655 3.67 5 2.55 5.292 20.82 6 2.54 2.642 11.03 7 1.91 1.875 9.11 8 3.27 10.252 29.75 9 1.95 2.666 12.02 10 1.73 2.018 9.11 11 3.04 8.744 38.88 12 1.67 1.532 7.77 13 5.17 26.72 61.36 14 1.68 1.818 7.96 15 2.59 5.433 20.76 16 1.39 1.054 5.75 17 2.9 2.955 16.12 18 1.87 2.1 11.72 19 2.02 2.274 10.56 20 1.52 1.251 6.37 21 2.66 4.739 17.93 22 2.52 5.046 18.82 23 2.48 6.937 32.51 24 3.11 9.04 32.77 25 1.78 2.535 9.64 26 2.89 11.046 38.09 27 1.41 1.232 5.85 28 0.73 0.334 2.12 29 1.93 2.823 13.14 30 1.23 2.264 11.31 31 1.19 1.095 5.55 Table 2.2: Ion Chamber Calibration The Ion Chamber calibration measurements are shown by date along with the calculated calibration factors. Date Reading (nC/200MU) Calibration Factor1 (cGy/nC)1 2 3 4 Average Aug. 16/12 0.4304 0.4308 0.4301 0.4310 0.4306 320.12 Aug. 27/12 0.4312 0.4316 0.4308 N/A 0.4312 319.66 Aug. 30/12 0.4312 0.4309 0.4306 0.4310 0.4309 319.86 Aug. 31/12 0.4326 0.4323 0.4320 0.4328 0.4324 318.79 1. This value was obtained via dividing 137.84cGy/200MUs by the Average Reading (nC/200MUs). 22 Table 2.3: Film Calibration The dose and MUs per film calibration piece is shown. Film Size (cm2) Dose (Gy) MUs 10x10 0 0 10x10 0.125 15 10x10 0.25 30 10x10 0.5 61 10x10 1 121 10x10 2 242 10x10 4 484 10x10 8 968 20x25 (Full Size) 8.82 4371 Figure 2.1: BrainLAB m3 microMLC The BrainLAB microMLC is shown. One side of the tungsten leaves can be seen within the window at the center where the beam exits the Linac (when in use and connected to the gantry head). 23 Figure 2.2: IMRT Optimization Window The IMRT optimization window depicting the ideal settings for dose delivery to the PTV with maximum NT sparing is shown. The dose calculation grid sizes for the PTV and OAR Sphere as well as the maximum dose rate and beam modulation selections can be seen. 24 Figure 2.3: OAR Sphere The OAR Sphere encompassing the PTV as well as the 7 (blue) and 12Gy (orange) isodoses are shown. Using simple boolean options available in iPlan, the PTV volume was removed from the center of the OAR. Figure 2.4: Cube Phantom Setup The cube phantom with ion chamber inserted and connected to the electrometer are shown. When performing measurements, an extension cord was used so that the electrometer rested outside the treatment room. 25 Figure 2.5: CIRS Head Phantom The CIRS phantom together with the frameless stereotactic mask is shown. The top portion of the head phantom has been removed to expose the film insert which housed the film, during treatment, at position 7 from the posterior side. 26 Figure 2.6: Film Calibration Curve The film calibration curve produced with the use of FUGU is shown. The blue, red, and green fits can be seen with exact values for each data point in the table to the left of the curve. Additionally, a screenshot of the film can be seen below the table depicting how only the central region of the film is considered when determining the average optical density. 27 3 Results 3.1 Patient Characteristics Patient treatment characteristics are shown in Table 3.1. It is important to note that a large variation in PTV shape is present in this study resulting in a wide range of surface area conditions that allows for varying levels of patient benefits from the application of IMRT. This is observed from the fact that the patient’s PTV diameters are approximately evenly spread out throughout the range while the PTV volumes are clustered around the lower end, as shown by the median falling in the center of the range and towards the lower limit, respectively. Additionally, the prescribed dose for each patient was decided upon by a radiation oncologist with the range resultant from the large variation in PTV volumes and anatomic locations causing concerns for symptomatic injury. These anatomic locations of the AVMs varied widely throughout the brain, occurring next to relatively deep and important struc- tures such as the brainstem, optic nerves, and thalamus as well as out towards the brain’s periphery: in the frontal, temporal, parietal, and occipital regions. The dosimetric parameters measured for each plan per patient are shown in tables 3.2-3.6. In order to obtain these values, DVHs depicting the dose to the PTV and to the OAR sphere were generated and inspected at the dose points of interest. These points consisted of: the prescription dose to the PTV, the maximum dose to the PTV and OAR sphere, and the V7and V12 for the OAR sphere. 3.2 Patient Stratification Tables 3.7-3.11 show the individual patient plan improvements in PTV conformity, NT Max Dose, PTV Max Dose, and 7 and 12Gy isodose volumes, respectively, for all IMRT plans as compared to 3DCRT. Each value is the difference between the recorded results in IMRT and 3DCRT with negative values depicting cases where the parameter in the respective IMRT plan was worse than in 3DCRT while a positive value represents an improvement in the parameter. Ultimately, the data contained within these tables is used to determine the color of each patient data point in the subsequent stratification plots. Figures 3.1 and 3.2 show the distribution of PTV conformity index improvements for PTV Only and OAR Low IMRT, respectively. In terms of PTV Only IMRT, stratification is readily observed by relating the ratio of PTV volume to surface area with PTV diameter resulting in patients with Vol./SA <0.25cm and PTV diameter <2.25cm receiving the largest benefit. While for OAR Low IMRT, patient stratification is again observed by relating the ratio of PTV volume to surface area with PTV diameter resulting in patients with Vol./SA <0.2cm and PTV diameter <2cm receiving the largest benefit. Patient specific improvements in conformity indices ranged from -0.06 to 0.68, -0.37 to 0.58, -0.37 to 0.64, and -0.37 to 0.58 with a median of 0.15, 0.05, 0.03, and 0.02 for 28 PTV Only, OAR Low, OAR Medium, and OAR High IMRT, respectively, when compared to 3DCRT results. Figures 3.3 and 3.4 show the arrangement of improvements in the maximum dose to NT as a function of PTV volume and diameter for PTV Only and OAR Low IMRT, respectively. For this dosimetric parameter, max dose to NT, decreases ranged from -0.5 to 12.3%, -13.9 to 6.5%, -14.0 to 6.3%, and -15.4 to 2.9% with a median of 4.8%, -3.8%, -4.5%, and -5.9% for PTV Only, OAR Low, OAR Medium, and OAR High IMRT, respectively, when compared to 3DCRT results. In this case, for PTV Only IMRT, patients can be separated into distinct groups with those whose PTV diameters that are greater than 2.5cm receiving large decreases in NT max dose. While, for OAR Low IMRT, patients with PTV volumes greater than 11cc received the largest decreases in NT max dose. Comparatively, for the max dose to the PTV, decreases ranged from -1.9 to 13.3%, -12.7 to 3.6%, -13.3 to 2.1%, and -13.4 to -0.6% with a median of 5.5%, -3.0%, -4.7%, and -6.9% for PTV Only, OAR Low, OAR Medium, and OAR High IMRT, respectively, when compared to 3DCRT results. Unfortunately, patient stratification for this parameter was not obvious by means of a plot; however, the trend is that patients with larger PTVs showed, on average, a greater decrease in max dose as well as hotspot (>103%) size with less patients benefiting from OAR Low plans when compared to PTV Only plans. Finally, it should be noted that, when considering the decreases in maximum dose to both the PTV and NT, the PTV Only IMRT plans consistently illustrated significant concurrent benefits for a higher number of patients when compared to 3DCRT results. Figures 3.5 and 3.6 show the distribution of patient plan improvements concerning the reduction in NT volume irradiated with 7Gy for PTV Only and OAR Low IMRT, respectively. In both plots, patient stratification is dependent on PTV volume and diameter with OAR Low IMRT providing more significant patient benefits, when compared to PTV Only IMRT, in all but one case: the largest PTV. The decrease in the volume enclosed by the 7Gy isodose ranged from -8.528 to 7.072cc, -3.880 to 7.952cc, -3.760 to 6.760cc, and -4.0 to 8.232cc with a median of -2.080cc, 0.216cc, 0.112cc, and -0.072cc for PTV Only, OAR Low, OAR Medium, and OAR High IMRT, respectively, when compared to 3DCRT results. Overall, for PTV Only IMRT, those whose PTV volumes are greater than approximately 10cc are the only patients to receive any notable benefit. While, for OAR Low IMRT, patients whose PTV volumes are >5cc combined with diameters >3cm received the largest NT sparing at the 7Gy isodose. Figures 3.7 and 3.8 show the distribution of patient plan improvements concerning the reduction in NT volume irradiated with 12Gy for PTV Only and OAR Low IMRT, respectively. In both plots, patient stratification is dependent on PTV volume and diameter with a large number of patients benefiting from both PTV Only and OAR Low plans. The decrease in volume enclosed by the 12Gy isodose ranged from -2.248 to 4.496cc, -1.448 to 3.704cc, -1.368 to 3.176cc, and -1.552 to 3.064cc with a median of 0.224cc, 0.560cc, 0.528cc, and 0.296cc for PTV Only, OAR Low, OAR Medium, and OAR High IMRT, respectively, when compared to 3DCRT results. Consequently, for PTV Only and OAR Low IMRT, patients whose PTV diameters are greater than 3cm and 29 2.9cm, respectively, appear to receive the largest NT sparing at the 12Gy isodose. Furthermore, in both plan types, patients containing PTV diameters ranging from 1 to 2cm received notable 12Gy sparing despite exhibiting PTVs of relatively small volumes. Alternate PTV Only and OAR Low plot axes for each dosimetric parameter, not shown in figures 3.1-3.8, were found to not reveal any potential stratification rules to determine for which patients IMRT should be the treatment technique of choice. Due to this, these plots have not been included as they do not contribute any insight into which patients benefit from IMRT. Additionally, when investigating OAR Medium and OAR High results, it was found that a small number of patients received significant benefits in PTV conformity, NT 7Gy volume reduction, and/or NT 12Gy volume reduction; however, the maximum dose to both the PTV and NT became prohibitively large in most of these cases. This resulted in plans that would not stand up to clinical scrutiny and; thus, would not be approved for patient treatment. As a result, these plan types are included in the data tables comparing IMRT to 3DCRT to show how each dosimetric parameter changes between plan types but are not considered for the development of patient stratification rules as they would reflect no clinical relevance. Tables 3.12 and 3.13 summarize the stratification rules and predicted improvements in dose distribution pa- rameters derived from the data in Tables 3.1 to 3.8 for PTV Only and OAR Low IMRT, respectively. These tables illustrate the PTV characteristics required for patient treatment improvements in a succinct manner. Moreover, following the rules in each table is the predicted benefits available from the application of IMRT, for patients that fall into these regions, suggesting a range (measured improvements acquired from this investigation) for each plan quality factor. 3.3 IMRT Verification 3.3.1 Ion Chamber Results Table 3.14 shows the electrometer reading values and the resultant calculated dose for each beam per patient. Additionally, the individual beam doses, as suggested by iPlan, are shown for comparison. Table 3.15 then shows the cumulative isocenter dose from the ion chamber measurements and TPS calculations as well as the percent difference between the two values for each of the ten patients examined. Furthermore, the pass/fail results are shown, indicating whether or not the percent difference is less than or equal to 3%. In all cases, the ion chambers measurements received a passing outcome signifying that the MUs estimated within iPlan can accurately, within systematic uncertainty, deliver the required dose. 30 3.3.2 EBT3 Film Results Figures 3.9-3.13 show the OmniPro I’mRT comparisons between the iPlan TPS exported dose and the EBT3 film 2D dose plane, in the coronal plane, for a sample patient from this study, specifically patient 2. First, fig- ure 3.9 depicts the 2D dose map exported from iPlan that was used as the standard to which film measurements were compared. Following this, in figure 3.10 isodose line comparisons between the TPS and film dose maps are shown. Next, figures 3.11 and 3.12 show cross sectional dose distributions in the x and y directions intersecting at the isocenter of the treatment plan. Finally, figure 3.13 shows the calculated dose difference fluence map obtained from a direct comparison between each dose point in the patient’s film and TPS results. The resolution of the fluence difference map is based on the resolution of the film which, despite exporting the iPlan calculations with a 1mm dose resolution, has the higher resolution of the two. Table 3.1: Patient Characteristics A summary of patient data concerning lesion characteristics and prescribed treatment dose. Number of Patients 31 PTV Diameter (cm) 0.99-5.38 (2.71)1 PTV Volume (cc) 0.334-26.720 (2.666) Prescription Dose (Gy) 12-25 (20) 1. The range across all patients is shown while the median is shown in brackets. 31 Table 3.2: PTV Conformity Measurements The PTV Conformity measurements for all plan types are shown. In each case, the conformity index was determined by viewing the plan specific DVH and reading the value from the prescription isodose position, 80% of the total dose. Each value represents the ratio of total volume receiving this dose to the PTV volume resulting in lower numbers describing cases where less NT is being irradiated with the prescription dose. A conformity index of 1 is when the prescription dose is perfectly tailored to the shape of the PTV volume and is the lowest clinically acceptable value. Patient 3DCRT PTV Only IMRT OAR Low IMRT OAR Medium IMRT OAR High IMRT 1 1.58 1.47 1.53 1.53 1.61 2 1.99 1.61 1.66 1.65 1.66 3 1.54 1.47 1.53 1.55 1.55 4 2.17 1.49 1.59 1.59 1.59 5 1.63 1.58 1.59 1.60 1.61 6 1.75 1.56 1.85 1.85 1.85 7 1.65 1.51 1.61 1.65 1.65 8 1.66 1.44 1.57 1.62 1.61 9 1.6 1.65 1.74 1.74 1.74 10 2.05 1.70 1.63 1.64 1.63 11 1.94 1.80 1.90 1.90 1.89 12 1.58 1.29 1.63 1.63 1.63 13 1.54 1.38 1.51 1.51 1.57 14 1.99 1.48 1.52 1.52 1.54 15 1.61 1.60 1.68 1.68 1.70 16 1.85 1.32 1.39 1.39 1.51 17 2.1 1.56 1.80 1.80 1.80 18 1.67 1.55 1.79 1.79 1.79 19 1.48 1.37 1.37 1.37 1.34 20 1.46 1.26 1.31 1.36 1.36 21 1.78 1.63 1.72 1.70 1.73 22 1.49 1.48 1.57 1.60 1.58 23 1.78 1.45 1.78 1.83 1.79 24 1.62 1.59 1.54 1.59 1.59 25 1.43 1.35 1.38 1.40 1.46 26 1.6 1.56 1.44 1.47 1.49 27 1.62 1.68 1.99 1.99 1.99 28 1.38 1.19 1.38 1.38 1.38 29 1.57 1.45 1.61 1.61 1.61 30 1.82 1.54 1.76 1.79 1.79 31 2.11 1.53 1.53 1.47 1.59 32 Table 3.3: NT Max Dose Measurements The NT Max Dose measurements for all plan types are shown. This value was obtained directly from examining the DVH for the OAR sphere. Patient 3DCRT PTV Only IMRT OAR Low IMRT OAR Medium IMRT OAR High IMRT 1 100.1% 91.7% 99.7% 100.9% 105.8% 2 100.1% 99.8% 105.7% 106.0% 105.9% 3 95.2% 91.1% 102.2% 104.0% 104.4% 4 98.2% 95.9% 108.3% 107.8% 107.9% 5 101.7% 94.4% 100.2% 101.0% 101.3% 6 99.9% 89.7% 102.6% 102.2% 102.8% 7 92.5% 91.7% 100.6% 103.7% 107.9% 8 100.9% 93.3% 102.2% 105.3% 105.5% 9 98.6% 94.3% 102.9% 103.9% 103.7% 10 97.2% 94.7% 99.5% 99.6% 100.0% 11 102.0% 95.0% 103.7% 106.2% 105.0% 12 95.5% 91.4% 104.3% 105.4% 108.0% 13 106.2% 96.1% 99.7% 99.9% 103.3% 14 97.9% 94.1% 101.7% 103.6% 106.9% 15 98.3% 98.8% 107.3% 106.8% 107.2% 16 98.8% 91.0% 103.5% 103.3% 105.7% 17 99.8% 93.9% 105.0% 105.3% 104.8% 18 96.5% 93.3% 105.1% 105.5% 105.8% 19 98.8% 95.6% 98.8% 99.2% 98.9% 20 97.6% 93.3% 101.5% 101.7% 104.6% 21 98.8% 93.9% 101.2% 102.9% 105.9% 22 97.8% 90.8% 100.0% 102.0% 103.7% 23 102.1% 89.8% 101.3% 102.5% 103.6% 24 101.8% 93.0% 104.6% 110.2% 112.8% 25 93.3% 88.5% 95.2% 97.2% 103.1% 26 102.5% 94.0% 97.8% 101.6% 104.9% 27 94.4% 93.4% 108.3% 108.4% 108.7% 28 93.6% 92.4% 106.7% 106.7% 106.1% 29 95.7% 88.0% 100.1% 100.4% 100.3% 30 93.3% 92.4% 104.9% 105.1% 104.9% 31 98.1% 92.6% 98.0% 98.9% 101.7% 33 Table 3.4: PTV Max Dose Measurements The PTV Max Dose measurements for all plan types are shown. This value was obtained directly from examining the DVH for the PTV where a maximum dose of ≤ 103% is clinically acceptable; therefore, plans above this dose would not be used for patient treatment. Patient 3DCRT PTV Only IMRT OAR Low IMRT OAR Medium IMRT OAR High IMRT 1 101.3% 96.4% 102.7% 104.4% 109.6% 2 101.2% 103.1% 109.2% 109.9% 109.8% 3 103.3% 94.4% 103.5% 108.0% 108.8% 4 100.5% 98.7% 110.7% 110.2% 110.3% 5 103.6% 100.1% 103.8% 104.1% 104.2% 6 100.6% 94.6% 105.2% 105.0% 105.2% 7 100.3% 95.9% 106.3% 110.2% 113.7% 8 103.1% 97.5% 103.7% 106.5% 108.2% 9 101.6% 98.4% 107.0% 105.8% 106.3% 10 100.9% 96.5% 103.0% 103.3% 103.4% 11 103.0% 99.7% 104.2% 106.4% 108.3% 12 100.3% 94.8% 108.0% 110.5% 112.9% 13 107.8% 94.5% 106.8% 105.7% 111.2% 14 100.7% 96.4% 105.4% 108.4% 109.4% 15 102.3% 99.2% 109.3% 108.9% 109.4% 16 100.7% 94.8% 105.9% 107.1% 110.7% 17 100.4% 98.4% 108.9% 109.4% 109.6% 18 100.7% 99.6% 109.3% 109.4% 109.9% 19 100.2% 94.7% 101.5% 102.0% 101.8% 20 100.3% 93.3% 103.3% 105.0% 106.3% 21 102.6% 97.9% 104.5% 106.2% 110.0% 22 100.9% 93.6% 103.5% 106.4% 109.1% 23 103.0% 92.7% 102.2% 104.3% 105.6% 24 103.8% 95.5% 106.0% 114.0% 114.3% 25 100.9% 95.2% 98.7% 101.3% 107.8% 26 104.5% 98.3% 100.9% 105.7% 107.9% 27 100.3% 98.5% 113.0% 113.6% 113.4% 28 100.5% 95.2% 107.6% 107.6% 107.4% 29 100.6% 93.8% 103.4% 103.2% 103.4% 30 102.7% 96.2% 108.2% 107.6% 107.9% 31 100.4% 94.2% 104.6% 105.2% 107.5% 34 Table 3.5: 7Gy Isodose Volume Measurements The 7Gy isodose volume measurements in terms of cc (cm3) for all plan types are shown. This value was obtained directly from examining the DVH for the OAR sphere. Patient 3DCRT PTV Only IMRT OAR Low IMRT OAR Medium IMRT OAR High IMRT 1 59.264 58.712 54.504 54.432 56.464 2 31.104 30.936 29.456 29.576 29.752 3 31.360 37.024 31.344 32.344 31.824 4 6.336 5.816 5.328 5.296 5.352 5 46.656 53.728 47.888 47.832 47.976 6 22.784 24.920 23.112 22.960 23.144 7 13.312 15.232 13.392 13.888 13.928 8 57.280 51.328 49.328 50.520 49.048 9 20.544 24.768 22.304 22.232 22.176 10 22.912 24.992 22.032 22.008 22.000 11 67.776 72.984 66.752 66.720 64.984 12 8.576 8.632 8.176 8.448 8.344 13 71.360 64.288 68.128 68.792 71.120 14 12.544 12.766 11.414 11.362 11.333 15 28.800 34.064 32.680 32.560 32.800 16 8.384 8.488 7.816 7.840 7.792 17 22.208 30.736 24.032 24.128 24.128 18 20.608 23.616 22.128 22.152 22.168 19 15.872 19.304 16.776 16.992 16.528 20 9.088 10.064 8.872 8.976 8.984 21 36.544 39.640 35.248 35.400 35.808 22 29.760 33.984 30.776 31.176 30.856 23 47.808 52.280 48.608 49.488 48.448 24 42.688 46.296 39.096 40.848 40.512 25 16.576 18.224 15.592 16.008 16.648 26 41.792 45.136 37.720 39.360 40.144 27 14.080 15.944 15.592 15.656 15.608 28 3.328 3.628 3.453 3.476 3.458 29 16.192 17.424 17.096 17.104 17.040 30 13.504 15.824 14.616 14.576 14.552 31 7.616 7.032 6.288 6.240 6.216 35 Table 3.6: 12Gy Isodose Volume Measurements The 12Gy isodose volume measurements in terms of cc (cm3) for all plan types are shown. This value was obtained directly from examining the DVH for the OAR sphere. Patient 3DCRT PTV Only IMRT OAR Low IMRT OAR Medium IMRT OAR High IMRT 1 21.248 22.840 21.272 21.280 22.336 2 11.904 9.320 9.376 9.392 9.464 3 12.288 12.480 11.264 11.496 11.216 4 2.432 1.896 1.872 1.872 1.872 5 17.792 20.040 18.192 18.208 18.248 6 8.832 8.016 8.240 8.128 8.192 7 4.928 4.992 4.736 4.856 4.888 8 19.136 17.296 17.112 17.872 17.208 9 7.872 8.264 7.944 7.928 7.912 10 9.408 9.264 8.064 8.064 8.080 11 25.856 27.440 25.336 25.328 24.656 12 2.944 2.456 2.736 2.816 2.784 13 13.440 8.944 12.536 12.680 13.768 14 5.376 4.429 4.124 4.074 4.086 15 10.304 11.672 11.752 11.672 11.856 16 3.584 2.672 2.592 2.656 2.624 17 9.280 8.208 7.824 7.864 7.848 18 8.128 8.944 8.680 8.672 8.728 19 5.952 6.672 6.136 6.144 6.000 20 3.392 3.248 3.176 3.208 3.200 21 14.080 13.584 12.592 12.608 12.664 22 11.264 11.608 11.096 11.256 11.184 23 19.456 18.856 18.512 18.904 18.528 24 15.936 14.560 12.392 12.992 13.208 25 6.336 6.112 5.744 5.880 6.040 26 16.256 15.104 12.552 13.080 13.192 27 5.248 5.640 5.912 5.928 5.912 28 1.472 1.137 1.213 1.231 1.219 29 5.952 5.736 6.048 6.080 6.072 30 4.608 4.840 5.008 4.968 4.976 31 3.072 2.104 1.968 1.936 1.968 36 Table 3.7: PTV Conformity Comparisons between IMRT and 3DCRT The conformity index difference between 3DCRT and each of the IMRT plans is shown. A positive/negative value denotes that the conformity index in the respective IMRT plan has improved/worsened over its initial value in 3DCRT. Note that a decrease in conformity index is a favourable result as long as the final value remains greater than or equal to 1. Patient 3DCRT - PTV Only 3DCRT - OAR Low 3DCRT - OAR Medium 3DCRT - OAR High 1 0.11 0.05 0.05 -0.03 2 0.38 0.33 0.34 0.33 3 0.07 0.01 -0.01 -0.01 4 0.68 0.58 0.58 0.58 5 0.05 0.04 0.03 0.02 6 0.19 -0.10 -0.10 -0.10 7 0.14 0.04 0 0 8 0.22 0.09 0.04 0.05 9 -0.05 -0.14 -0.14 -0.14 10 0.35 0.42 0.41 0.42 11 0.14 0.04 0.04 0.05 12 0.29 -0.05 -0.05 -0.05 13 0.16 0.03 0.03 -0.03 14 0.51 0.47 0.47 0.45 15 0.01 -0.07 -0.07 -0.09 16 0.53 0.46 0.46 0.34 17 0.54 0.30 0.30 0.30 18 0.12 -0.12 -0.12 -0.12 19 0.11 0.11 0.11 0.14 20 0.20 0.15 0.10 0.10 21 0.15 0.06 0.08 0.05 22 0.01 -0.08 -0.11 -0.09 23 0.33 0 -0.05 -0.01 24 0.03 0.08 0.03 0.03 25 0.08 0.05 0.03 -0.03 26 0.04 0.16 0.13 0.11 27 -0.06 -0.37 -0.37 -0.37 28 0.19 0 0 0 29 0.12 -0.04 -0.04 -0.04 30 0.28 0.06 0.03 0.03 31 0.58 0.58 0.64 0.52 37 Table 3.8: NT Max Dose Comparisons between IMRT and 3DCRT The NT Max Dose difference between 3DCRT and each of the IMRT plans is shown. A positive/negative value denotes that the NT maximum dose in the respective IMRT plan has improved/worsened from its initial value in 3DCRT. Patient 3DCRT - PTV Only 3DCRT - OAR Low 3DCRT - OAR Medium 3DCRT - OAR High 1 8.4% 0.4% -0.8% -5.7% 2 0.3% -5.6% -5.9% -5.8% 3 4.1% -7.0% -8.8% -9.2% 4 2.3% -10.1% -9.6% -9.7% 5 7.3% 1.5% 0.7% 0.4% 6 10.2% -2.7% -2.3% -2.9% 7 0.8% -8.1% -11.2% -15.4% 8 7.6% -1.3% -4.4% -4.6% 9 4.3% -4.3% -5.3% -5.1% 10 2.5% -2.3% -2.4% -2.8% 11 7.0% -1.7% -4.2% -3.0% 12 4.1% -8.8% -9.9% -12.5% 13 10.1% 6.5% 6.3% 2.9% 14 3.8% -3.8% -5.7% -9.0% 15 -0.5% -9.0% -8.5% -8.9% 16 7.8% -4.7% -4.5% -6.9% 17 5.9% -5.2% -5.5% -5.0% 18 3.2% -8.6% -9.0% -9.3% 19 3.2% 0.0% -0.4% -0.1% 20 4.3% -3.9% -4.1% -7.0% 21 4.9% -2.4% -4.1% -7.1% 22 7.0% -2.2% -4.2% -5.9% 23 12.3% 0.8% -0.4% -1.5% 24 8.8% -2.8% -8.4% -11.0% 25 4.8% -1.9% -3.9% -9.8% 26 8.5% 4.7% 0.9% -2.4% 27 1.0% -13.9% -14.0% -14.3% 28 1.2% -13.1% -13.1% -12.5% 29 7.7% -4.4% -4.7% -4.6% 30 0.9% -11.6% -11.8% -11.6% 31 5.5% 0.1% -0.8% -3.6% 38 Table 3.9: PTV Max Dose Comparisons between IMRT and 3DCRT The PTV Max Dose difference between 3DCRT and each of the IMRT plans is shown. A positive/negative value denotes that the PTV maximum dose in the respective IMRT plan has improved/worsened from its initial value in 3DCRT. Patient 3DCRT - PTV Only 3DCRT - OAR Low 3DCRT - OAR Medium 3DCRT - OAR High 1 4.9% -1.4% -3.1% -8.3% 2 -1.9% -8.0% -8.7% -8.6% 3 8.9% -0.2% -4.7% -5.5% 4 1.8% -10.2% -9.7% -9.8% 5 3.5% -0.2% -0.5% -0.6% 6 6.0% -4.6% -4.4% -4.6% 7 4.4% -6.0% -9.9% -13.4% 8 5.6% -0.6% -3.4% -5.1% 9 3.2% -5.4% -4.2% -4.7% 10 4.4% -2.1% -2.4% -2.5% 11 3.3% -1.2% -3.4% -5.3% 12 5.5% -7.7% -10.2% -12.6% 13 13.3% 1.0% 2.1% -3.4% 14 4.3% -4.7% -7.7% -8.7% 15 3.1% -7.0% -6.6% -7.1% 16 5.9% -5.2% -6.4% -10.0% 17 2.0% -8.5% -9.0% -9.2% 18 1.1% -8.6% -8.7% -9.2% 19 5.5% -1.3% -1.8% -1.6% 20 7.0% -3.0% -4.7% -6.0% 21 4.7% -1.9% -3.6% -7.4% 22 7.3% -2.6% -5.5% -8.2% 23 10.3% 0.8% -1.3% -2.6% 24 8.3% -2.2% -10.2% -10.5% 25 5.7% 2.2% -0.4% -6.9% 26 6.2% 3.6% -1.2% -3.4% 27 1.8% -12.7% -13.3% -13.1% 28 5.3% -7.1% -7.1% -6.9% 29 6.8% -2.8% -2.6% -2.8% 30 6.5% -5.5% -4.9% -5.2% 31 6.2% -4.2% -4.8% -7.1% 39 Table 3.10: 7Gy Isodose Volume Comparisons between IMRT and 3DCRT The 7Gy isodose volume difference between 3DCRT and each of the IMRT plans is shown. A positive/negative value denotes that the 7Gy isodose volume in the respective IMRT plan has improved/worsened from its initial value in 3DCRT. Patient 3DCRT - PTV Only 3DCRT - OAR Low 3DCRT - OAR Medium 3DCRT - OAR High 1 0.552 4.760 4.832 2.800 2 0.168 1.648 1.528 1.352 3 -5.664 0.016 -0.984 -0.464 4 0.520 1.008 1.040 0.984 5 -7.072 -1.232 -1.176 -1.320 6 -2.136 -0.328 -0.176 -0.360 7 -1.920 -0.080 -0.576 -0.616 8 5.952 7.952 6.760 8.232 9 -4.224 -1.760 -1.688 -1.632 10 -2.080 0.880 0.904 0.912 11 -5.208 1.024 1.056 2.792 12 -0.056 0.400 0.128 0.232 13 7.072 3.232 2.568 0.240 14 -0.222 1.130 1.182 1.211 15 -5.264 -3.880 -3.760 -4.000 16 -0.104 0.568 0.544 0.592 17 -8.528 -1.824 -1.920 -1.920 18 -3.008 -1.520 -1.544 -1.560 19 -3.432 -0.904 -1.120 -0.656 20 -0.976 0.216 0.112 0.104 21 -3.096 1.296 1.144 0.736 22 -4.224 -1.016 -1.416 -1.096 23 -4.472 -0.800 -1.680 -0.640 24 -3.608 3.592 1.840 2.176 25 -1.648 0.984 0.568 -0.072 26 -3.344 4.072 2.432 1.648 27 -1.864 -1.512 -1.576 -1.528 28 -0.300 -0.125 -0.148 -0.130 29 -1.232 -0.904 -0.912 -0.848 30 -2.320 -1.112 -1.072 -1.048 31 0.584 1.328 1.376 1.400 40 Table 3.11: 12Gy Isodose Volume Comparisons between IMRT and 3DCRT The 12Gy isodose volume difference between 3DCRT and each of the IMRT plans is shown. A positive/negative value denotes that the 12Gy isodose volume in the respective IMRT plan has improved/worsened from its initial value in 3DCRT. Patient 3DCRT - PTV Only 3DCRT - OAR Low 3DCRT - OAR Medium 3DCRT - OAR High 1 -1.592 -0.024 -0.032 -1.088 2 2.584 2.528 2.512 2.440 3 -0.192 1.024 0.792 1.072 4 0.536 0.560 0.560 0.560 5 -2.248 -0.400 -0.416 -0.456 6 0.816 0.592 0.704 0.640 7 -0.064 0.192 0.072 0.040 8 1.840 2.024 1.264 1.928 9 -0.392 -0.072 -0.056 -0.040 10 0.144 1.344 1.344 1.328 11 -1.584 0.520 0.528 1.200 12 0.488 0.208 0.128 0.160 13 4.496 0.904 0.760 -0.328 14 0.947 1.252 1.302 1.290 15 -1.368 -1.448 -1.368 -1.552 16 0.912 0.992 0.928 0.960 17 1.072 1.456 1.416 1.432 18 -0.816 -0.552 -0.544 -0.600 19 -0.720 -0.184 -0.192 -0.048 20 0.144 0.216 0.184 0.192 21 0.496 1.488 1.472 1.416 22 -0.344 0.168 0.008 0.080 23 0.600 0.944 0.552 0.928 24 1.376 3.544 2.944 2.728 25 0.224 0.592 0.456 0.296 26 1.152 3.704 3.176 3.064 27 -0.392 -0.664 -0.680 -0.664 28 0.335 0.259 0.241 0.253 29 0.216 -0.096 -0.128 -0.120 30 -0.232 -0.400 -0.360 -0.368 31 0.968 1.104 1.136 1.104 Table 3.12: PTV Only IMRT Stratification Rules The stratification rules for PTV Only IMRT are shown. Additionally, the predicted benefits for any patient that falls within the range of the respective dosimetric parameter are presented. Notably, these expected improvements are actual benefits experienced by patients in the course of this thesis. Factors of Potential Benefit with PTV Only IMRT PTV Conformity NT Max Dose 7Gy Vol. 12Gy Vol. PTV Diameter <2.25cm >2.5cm N/A >3.0cm Stratification Volume N/A N/A >10cc N/A Rules Vol./SA <0.25cm N/A N/A N/A Predicted Improvements 0.10-0.68 4.0-12.3% 0-7.072cc 0.5-4.496cc 41 Table 3.13: OAR Low IMRT Stratification Rules The stratification rules for OAR Low IMRT are shown. Additionally, the predicted benefits for any patient that falls within the range of the respective dosimetric parameter are presented. Notably, these expected improvements are actual benefits experienced by patients in the course of this thesis. Factors of Potential Benefit with OAR Low IMRT PTV Conformity NT Max Dose 7Gy Vol. 12Gy Vol. PTV Diameter <2.0cm N/A >3.0cm >2.9cm Stratification Volume N/A >11cc >5.0cc N/A Rules Vol./SA <0.20cm N/A N/A N/A Predicted Improvements 0.10-0.58 0-6.5% 1.0-7.952cc 0.5-3.704cc 42 Table 3.14: Ion Chamber Patient Measurements The individual isocenter doses per patient per beam are shown. TPS(cGy) denotes the iPlan isocenter dose in cGy, IC(nC) is the isocenter reading displayed by the electrometer upon irradiation of the ion chamber, and IC(cGy) is the ion chamber isocenter dose obtained by applying the correct calibration factor from table 2.7. Patient (Date) IMRT Beam Number1 2 3 4 5 6 7 8 9 10 11 12 2 (Aug. 27/12) TPS (cGy) 23 25 21 21 25 22 22 IC (nC) 0.0712 0.0771 0.0631 0.0629 0.0776 0.0710 0.0685 IC (cGy) 22.760 24.646 20.171 20.107 24.806 22.696 21.897 4 (Aug. 31/12) TPS (cGy) 14 16 17 13 15 15 14 14 17 15 15 14 IC (nC) 0.0419 0.0477 0.0535 0.0384 0.0463 0.0462 0.0412 0.0464 0.0533 0.0483 0.0421 0.0429 IC (cGy) 13.357 15.206 17.055 12.242 14.760 14.728 13.134 14.792 16.992 15.398 13.421 13.676 8 (Aug. 30/12) TPS (cGy) 20 21 18 19 17 19 21 21 IC (nC) 0.0599 0.0635 0.0551 0.0574 0.0532 0.0590 0.0674 0.0646 IC (cGy) 19.160 20.311 17.624 18.360 17.017 18.872 21.559 20.663 9 (Aug. 27/12) TPS (cGy) 20 21 21 22 22 21 23 20 19 IC (nC) 0.0627 0.0656 0.0679 0.0693 0.0694 0.0642 0.0697 0.0628 0.0586 IC (cGy) 20.043 20.970 21.705 22.152 22.184 20.522 22.280 20.075 18.732 11 (Aug. 27/12) TPS (cGy) 23 24 22 22 23 19 20 18 IC (nC) 0.0717 0.0762 0.0677 0.0668 0.0711 0.0571 0.0603 0.0559 IC (cGy) 22.920 24.358 21.641 21.353 22.728 18.253 19.275 17.869 13 (Aug. 30/12) TPS (cGy) 34 25 30 32 30 31 28 IC (nC) 0.1041 0.0785 0.0925 0.1006 0.0933 0.0954 0.0848 IC (cGy) 33.297 25.109 29.587 32.178 29.843 30.515 27.124 14 (Aug. 31/12) TPS (cGy) 20 21 19 19 18 19 17 18 17 IC (nC) 0.0630 0.0649 0.0632 0.0607 0.0573 0.0592 0.0501 0.0575 0.0527 IC (cGy) 20.084 20.689 20.148 19.351 18.267 18.872 15.971 18.330 16.800 26 (Aug. 16/12) TPS (cGy) 18 21 16 17 20 18 17 IC (nC) 0.0552 0.0651 0.0478 0.0534 0.0618 0.0547 0.0525 IC (cGy) 17.671 20.840 15.302 17.094 19.783 17.511 16.806 27 (Aug. 16/12) TPS (cGy) 17 18 19 19 16 17 19 19 17 19 19 16 IC (nC) 0.0522 0.0557 0.0605 0.0569 0.0493 0.0512 0.0594 0.0592 0.0542 0.0596 0.0596 0.0527 IC (cGy) 16.710 17.831 19.367 18.215 15.782 16.390 19.015 18.951 17.351 19.079 19.079 16.870 31 (Aug. 30/12) TPS (cGy) 17 18 18 15 17 18 17 14 16 IC (nC) 0.0537 0.0582 0.0571 0.0473 0.0548 0.0563 0.0531 0.0448 0.0507 IC (cGy) 17.176 18.616 18.264 15.129 17.528 18.008 16.985 14.330 16.21743 Table 3.15: Percent Difference between Ion Chamber Measurements and TPS The isocenter dose measurements performed with the Ion Chamber (IC) and calculated in the TPS are shown. The percent difference then determined from these two values is also displayed with the pass/fail result. Patient (Date) IC (cGy) TPS (cGy) Percent Difference (%) Pass/Fail 2 (Aug. 27/12) 1.571 1.59 1.21 Pass 4 (Aug. 31/12) 1.748 1.79 2.37 Pass 8 (Aug. 30/12) 1.536 1.56 1.56 Pass 9 (Aug. 27/12) 1.887 1.89 0.20 Pass 11 (Aug. 27/12) 1.684 1.71 1.52 Pass 13 (Aug. 30/12) 2.077 2.10 1.10 Pass 14 (Aug. 31/12) 1.685 1.68 0.30 Pass 26 (Aug. 16/12) 1.250 1.27 1.57 Pass 27 (Aug. 16/12) 2.146 2.15 0.17 Pass 31 (Aug. 30/12) 1.523 1.50 1.50 Pass Figure 3.1: Patient Distribution of Conformity Improvements in PTV Only IMRT The distribution of patient plan improvements regarding PTV conformity when comparing PTV Only IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting Vol./SA vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0.2 0.25 0.3 0.35 0.4 0.45 0.5 PTV Diameter (cm) PT V Vo lu m e/ Su rfa ce A re a (cm ) PTV Conformity Index Improvements in PTV Only IMRT Worsening of Conformity Improvement in Conformity by 0−0.1 Improvement in Conformity by 0.1−0.2 Improvement in Conformity by >0.2 44 Figure 3.2: Patient Distribution of Conformity Improvements in OAR Low IMRT The distribution of patient plan improvements regarding PTV conformity when comparing OAR Low IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting Vol./SA vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0.2 0.25 0.3 0.35 0.4 0.45 0.5 PTV Diameter (cm) PT V Vo lu m e/ Su rfa ce A re a (cm ) PTV Conformity Index Improvements in OAR Low IMRT Worsening of Conformity Improvement in Conformity by 0−0.1 Improvement in Conformity by 0.1−0.2 Improvement in Conformity by >0.2 45 Figure 3.3: Patient Distribution of NT Max Dose Improvements in PTV Only IMRT The distribution of patient plan improvements regarding the max dose to NT when comparing PTV Only IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting PTV Volume vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 PTV Diameter (cm) PT V Vo lu m e (cc ) Normal Tissue Max Dose Improvements in PTV Only IMRT Worsening of Max Dose Improvement in Max Dose by 0−4% Improvement in Max Dose by 4−8% Improvement in Max Dose by >8% 46 Figure 3.4: Patient Distribution of NT Max Dose Improvements in OAR Low IMRT The distribution of patient plan improvements regarding the max dose to NT when comparing OAR Low IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting PTV Volume vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 PTV Diameter (cm) PT V Vo lu m e (cc ) Normal Tissue Max Dose Improvements in OAR Low IMRT Worsening of Max Dose Improvement in Max Dose by 0−4% Improvement in Max Dose by 4−8% Improvement in Max Dose by >8% 47 Figure 3.5: Patient Distribution of 7Gy Isodose Volume Improvements in PTV Only IMRT The distribution of patient plan improvements regarding the volume of NT receiving 7Gy when comparing PTV Only IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting PTV Volume vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 PTV Diameter (cm) PT V Vo lu m e (cc ) Normal Tissue 7Gy Volume Reductions in PTV Only IMRT Worsening of 7Gy Volume Improvement in 7Gy Volume by 0−0.75cc Improvement in 7Gy Volume by 0.75−1.5cc Improvement in 7Gy Volume by >1.5cc 48 Figure 3.6: Patient Distribution of 7Gy Isodose Volume Improvements in OAR Low IMRT The distribution of patient plan improvements regarding the volume of NT receiving 7Gy when comparing OAR Low IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting PTV Volume vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 PTV Diameter (cm) PT V Vo lu m e (cc ) Normal Tissue 7Gy Volume Reductions in OAR Low IMRT Worsening of 7Gy Volume Improvement in 7Gy Volume by 0−0.75cc Improvement in 7Gy Volume by 0.75−1.5cc Improvement in 7Gy Volume by >1.5cc 49 Figure 3.7: Patient Distribution of 12Gy Isodose Volume Improvements in PTV Only IMRT The distribution of patient plan improvements regarding the volume of NT receiving 12Gy when comparing PTV Only IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting PTV Volume vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 PTV Diameter (cm) PT V Vo lu m e (cc ) Normal Tissue 12Gy Volume Reductions in PTV Only IMRT Worsening of 12Gy Volume Improvement in 12Gy Volume by 0−0.5cc Improvement in 12Gy Volume by 0.5−1.5cc Improvement in 12Gy Volume by >1.5cc 50 Figure 3.8: Patient Distribution of 12Gy Isodose Volume Improvements in OAR Low IMRT The distribution of patient plan improvements regarding the volume of NT receiving 12Gy when comparing OAR Low IMRT to 3DCRT is shown. Patient stratification into groups, depending on the benefit received from IMRT, is readily observed when plotting PTV Volume vs. PTV Diameter while alternate plots do not clearly indicate a stratified subset of patients with improved treatment parameters. 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 PTV Diameter (cm) PT V Vo lu m e (cc ) Normal Tissue 12Gy Volume Reductions in OAR Low IMRT Worsening of 12Gy Volume Improvement in 12Gy Volume by 0−0.5cc Improvement in 12Gy Volume by 0.5−1.5cc Improvement in 12Gy Volume by >1.5cc 51 Figure 3.9: TPS Calculated Dose Distribution for Patient 2 The 2D dose plane exported at the isocenter of the PTV for patient 2 is shown (coronal plane). The colors represent the 40 (light blue), 50 (yellow), 80 (light red), and 100% (dark red) isodoses. 52 Figure 3.10: Isodose Comparison Map for Patient 2 The isodose comparison map for patient 2 is shown (coronal plane). The solid lines depict the film measurement results while the dotted lines depict the TPS calculated isodoses. The colors represent the 40 (blue), 50 (green), 80 (red), and 100% (black) isodoses. 53 Figure 3.11: X Direction Dose Profile Cross Section for Patient 2 The x direction cross section of the dose profile across the PTV for patient 2 is shown (coronal plane). The red line depicts the film measurement results while the green line depicts the TPS calculated profile. 54 Figure 3.12: Y Direction Dose Profile Cross Section for Patient 2 The y direction cross section of the dose profile across the PTV for patient 2 is shown (coronal plane). The red line depicts the film measurement results while the green line depicts the TPS calculated profile. 55 Figure 3.13: 2D Dose Difference Fluence Map for Patient 2 The fluence map depicting the difference in relative doses between the film measurements and TPS calculations for patient 2 is shown (coronal plane). The legend on the right hand side shows each color associated with calculated dose differences. Notably, the red regions at the edges of the film denote the localization marks. 56 4 Discussion 4.1 Outcomes from Patient Stratification The first intent of this study was to stratify patients into groups dependent upon the treatment plan improve- ments from the use of IMRT over 3DCRT. This stratification came in four different factors of improvement: improved conformity index, reduction in NT max dose, and decreases in the 7 and 12Gy isodose volumes. Con- sequently, each plan quality factor resulted in different stratification rules suggesting that the PTV characteristics themselves determine the particular region of benefit available from IMRT. Another intent of this study was to then relate the degree of benefit received from IMRT to predictive factors of symptomatic injury following SRS and determine the resultant decrease in the probability of complications arising post-irradiation. The conformity index, in this study, was measured at the prescription dose to the PTV, the 80% isodose. The importance of this isodose fully encompassing the PTV has been expressed in a study produced by Herbert et al. where it was concluded that the dose to the PTV margin that is >20Gy is strongly associated with nidus obliteration [41]. Unfortunately, this isodose frequently overlaps with NT, imparting a relatively high dose, and because incidence of symptomatic brain injury increases as NT volume receiving high doses increases, plan conformity is of the utmost importance. This result has also been validated in a study by Friedman et al. who found that improved PTV conformity correlates with a reduced incidence of complications [38]. Thus, by improving the conformity index by as much as 0.10-0.68 in PTV Only IMRT or 0.10-0.58 in OAR Low IMRT, significant NT sparing at the >20Gy isodose can be accomplished to reduce the probability of SRS induced complications. Furthermore, this level of improvement in treatment plans was found to occur more so in patients with a large PTV surface area compared to their volume (small Vol./SA ratio) and small PTV diameters, for both plan types; thus, providing the first column of stratification rules observed in both tables 3.12 and 3.13. Next, the maximum dose delivered to NT and the PTV should be considered. Although there is a lack of studies in the literature investigating the correlation of NT/PTV max dose with incidences of complications, it is important to not overdose either structure. As stated previously, the PTV is the AVM with an added margin; however, the AVM itself is NT despite the fact that it is being treated with SRS. As discussed in Chapter 1.1, it is simply a network of blood vessels that bypass a capillary bed resulting in arterial blood draining directly into veins unable to withstand the pressure. Therefore, it is desirable to minimize hotspots by maintaining the dose to the PTV as close to, albeit above, the prescribed dose. The PTV Only IMRT plans produced in this study provide this result by not only reducing the max dose to both structures but by decreasing the overall volume of hotspots and distributing them throughout the PTV. As a result, a single large volume hotspot is avoided within a patient and a reduced risk of symptomatic brain injury is accomplished. Furthermore, while stratification rules for PTV 57 Only plans regarding the PTV max dose are not given, those for NT max dose are provided in the second column of table 3.12. Unfortunately, OAR Low IMRT plans only provided a marginal benefit in NT max dose reduction for a relatively small subset of patients while not providing any benefit whatsoever to the reduction in PTV max dose. Despite this undesirable result, some patients did in fact benefit from this plan type and, while stratification rules for PTV max dose are non-existent due to negligible patient benefit, those for the reduction in NT max dose provide the second column of table 3.13. Finally, the 12Gy isodose volume has been investigated by numerous groups and determined to be one of the most significant volumes for predicting incidences of complications. In fact, volumes receiving doses of at least 7Gy were shown to be associated with a risk of brain injury and a strong correlation between various Vx values and brain injury have been reported [44, 75]. Therefore, the 7 and 12Gy isodose volumes were chosen to constrain during IMRT planning which allowed the inclusive volumes to benefit as well. Essentially, it was anticipated that the 7 and 12Gy isodose volumes were in close enough proximity that when minimizing each individual dose-volume through IMRT constraints, the dose-volumes between 7 and 12Gy would be reduced as well. Consequently, this was the case for all patients in which OAR Low IMRT provided improved results. Due to this, the resultant stratification rules acquired from this procedure were quite similar and are indicated in the third and fourth columns of Table 3.13. Alternatively, when investigating PTV Only IMRT, it was found that a large number of patients received significant NT sparing at the 12Gy isodose while only two of the largest PTVs received notable benefits at the 7Gy isodose. Consequently, this result logically follows from the weighting of constraints used in the optimization procedure and from the nature of IMRT itself. During the optimization of PTV Only plans, no value is placed on the NT constraints causing them to be completely ignored by the system, which focuses only on delivering the prescribed dose to the PTV. In delivering this dose, as discussed in Chapter 1.2, IMRT produces a highly conformal dose distribution with a sharp dose fall off at the PTV boundaries inherently moderating higher doses and contracting these isodoses in towards the PTV while allowing lower doses to remain uninhibited. As a result, isodoses in the neighbourhood of 12Gy irradiate a lower volume overall while isodoses around 7Gy are unaffected or increase in volume. Accordingly, there are few patients who benefit from PTV Only IMRT at the 7Gy isodose; nevertheless, stratification rules for the reduction of both isodose volumes are supplied in the last two columns of table 3.12. In a report by QUANTEC, radiation dose-volume effects in the brain were reviewed and complication rates based on the 12Gy isodose volume were analyzed [45]. The investigators of this study concluded that brain toxicity increases rapidly when NT is irradiated with >12Gy above volumes of 5-10cc. It is, therefore, desirable to reduce the volume encompassed by the 12Gy isodose by as much as possible. Consequently, IMRT achieves such a result with the stratified patients in tables 3.12 and 3.13 receiving significant NT sparing effects. In fact, the predicted benefits for both isodose volumes in the PTV Only and OAR Low plans, respectively, translate 58 to considerable decreases in the probability of radiation-induced brain injury post-SRS when interpreting the results from the QUANTEC study [45]. In this publication, probabilities for radiation necrosis as a function of volume irradiated are plotted for a multitude of studies concerning different isodose volumes: 10Gy, 12Gy, and the treatment volume. However, only two data sets relating symptomatic injury to the 12Gy isodose volume are of interest: the Korytko 2006 and the Flickinger 1997 results. Each data set represents substantially different variations in radiation necrosis with the 12Gy isodose volume that, when simplified through averaging the slopes of the linear portions of the plot, amount to a 0.6%/cc and 2.5%/cc increase in incidence of complications for the Flickinger and Korytko studies, respectively. Finally, comparing this with the PTV Only 12Gy isodose volume decreases of 0.5-4.496cc (i.e. multiplying 0.6%/cc and 2.5%/cc with the lower and upper limits on the V12 benefits, respectively), the probability of symptomatic radiation injury in the stratified subset of patients may be decreased by 0.3-11.2%. While, for the OAR Low 12Gy isodose volume decreases of 0.5-3.704cc, the probability of symptomatic injury in the stratified subset of patients may be decreased by 0.3-9.3%. Ultimately, with the application of IMRT, patient benefits can be predicted on a case-by-case basis with the specific IMRT plan type of choice dependent upon the PTV characteristics themselves. Categorically, if a patient has a PTV with a volume to surface area ratio (Vol./SA) less than 0.25 (small volume and/or large surface area) and diameter less than 2.25cm, then PTV Only IMRT would be the treatment technique of choice due to high PTV conformity results with the added potential benefit of minor improvements in the NT max dose and 12Gy isodose volume. Unfortunately, in this case the patient may be required to endure a larger 7Gy isodose volume when compared to 3DCRT. Alternatively, if a patient has a PTV with a diameter greater than 3cm, then OAR Low IMRT would be the treatment technique of choice due to the effective and consistent reductions applied to the 7 and 12Gy isodose volumes, when compared to 3DCRT. In this case, however, it is likely that the NT max dose will be higher than in PTV Only IMRT; thus, it is recommended that this dosimetric parameter be immediately verified upon plan completion to ensure that clinical standards are maintained. If the max dose to NT is then found to exceed clinical acceptability, PTV Only IMRT would be recommended so as to decrease this dose and obtain minor PTV conformity benefits with moderate 12Gy isodose volume reductions over 3DCRT. Additionally, if a patient has a PTV adjacent to critical structures, then PTV Only IMRT would be recommended so as to reduce the probability of a high dose being deposited within this structure. With this plan type, improved PTV conformity would restrict the prescription isodose (>18Gy) more to the PTV, while simultaneously reducing both the NT max dose as well as the 12Gy isodose volume, reducing the likelihood of OAR specific complications. Consequently, by identifying diameter, volume, surface area, and location of a PTV, treatment planners can readily select the most optimal treatment technique, efficiently utilizing clinical resources with a minimal increase on staff burden. Finally, with the acquired knowledge of how IMRT techniques affect PTVs of varying size and shape, the recom- mendations presented in this work can be expanded upon to apply to the Spetzler-Martin AVM grading scheme 59 found in table 1.1. Table 4.1 shows how Spetzler-Martin grades translate to distinct SRS treatment techniques. 4.2 Ion Chamber and Film Dosimetry The purpose of performing ion chamber and film dosimetry in this work was to mimic the clinical procedure when treating patients with IMRT. With this method, IMRT treatment protocols for AVMs could be developed and consented upon for clinical use with patients. The stratification rules then attained in Chapter 4.1 would become relevant for direct patient care and allow IMRT techniques to be available for the improvement of patient outcomes following SRS. Also, with careful compliance of dosimetry verification standards, clinical commissioning of the CIRS head phantom would be concurrently achieved. In this case, the CIRS phantom would then become available for the clinical verification of any and all intracranial stereotactic treatment plans. 4.2.1 Ion Chamber Measurements For all ten patients in which ion chamber measurements were performed, the percent difference between the isocenter dose as measured with the ion chamber and calculated with iPlan ranged from 0.17-2.37% with an average measurement deviation of 1.15% from the TPS calculation. Additionally, by examining both tables 3.14 and 3.15, it can be seen that ion chamber measurements varied between higher and lower doses than that predicted by the TPS. This indicated that variations between the two dose values was due to systematic uncertainty, a well understood technological limitation, rather than any sort of equipment malfunction. As a result, every treatment plan that was investigated passed the requirement of the the ion chamber measurements and TPS calculation being within 3% of one another. Thus, the MUs calculated in iPlan accurately correspond to the desired dose essential for the treatment of the patient’s AVMs. However, these favorable results alone are not enough to consent to patient treatment. The added requirement of film dosimetry must be accurately verified as well, using the information from the ion chamber results. 4.2.2 Film Measurements Before discussing the results of the film measurements it is important to understand the context in which this work was performed and to recognize the benefits from such measurements regardless of the results. The IMRT measurements performed here were implemented as a preliminary study into the commissioning of the CIRS head phantom and EBT3 film. As such, there is an inherent weakness in any results obtained as the number of patients chosen for film verification are insufficient to acquire meaningful conclusions or validate clinical applications of iPlan’s IMRT software. Despite this, by stringently following the film dosimetry protocols, the first steps toward 60 releasing IMRT treatments for AVMs were taken and the shortcomings requiring attention have been identified. For each patient in which film dosimetric measurements were performed, isodose line comparisons, dose distributions along the x and y directions, and dose difference fluence maps were investigated to determine the agreement between the iPlan TPS and EBT3 dose maps. First, the isodose line comparisons were selected so as to show the analogous shapes of the 40, 50, 80, and 100% isodoses exported from the TPS and measured with EBT3 film. Comparing these various isodoses provided a qualitative estimate of the agreement between the two distributions. Next, the dose distributions in the x and y directions were selected to compare the dose differences between individual points along with the overall shape of each pair of cross sections within the dose region of interest, henceforth referred to as the PTV (by importing patient plans to a verification phantom, the physical PTV is lost leaving only a region of interest that was previously associated with the PTV). In these cases, the planar dose results from the film and TPS were renormalized such that they retained an equal dose value at the isocenter, the point of intersection between the x and y cross sections. Notably, this renormalization of relative doses was permitted because the absolute doses at this point were previously verified to be within uncertainty of one another by way of the ion chamber measurements. Finally, the dose difference fluence maps were generated so as to compare the relative dose measurements at each and every point on the film. Significantly, in the plots showing dose difference map comparisons, select regions on the outer edge of the film show vast disagreement with the TPS results. This is simply due to the film localization technique in which the patient orientations were written directly on the film in black ink. This resulted in light being unable to penetrate the ink regions during the scanning process which caused FUGU to interpret the points as a high optical density and, therefore, translated them to high dose values. Similarly, pieces towards the edges of the film, where cutting occured, resulted in the layers of the film slightly separating from one another which also affected the optical density readings. Consequently, the differences noted in these regions can be ignored when verifying the accuracy of the IMRT plans since they are an artifact of record keeping rather than the film dosimetry procedure. Unfortunately, for all three patients examined in this study, the film measurements and TPS calculations did not agree with one another by inspection of the plots considered in conjunction with the 5% pass/fail threshold for point dose values. For each patient, multiple points across the PTV and NT exceeded the 5% limit on dose difference, disagreeing by as much as 15%. Additionally, observation of isodose line agreements and cross sectional dose profiles were collectively unacceptable, ruling out the possibility of detector errors in the scanner or regions where damage was incurred on the film. Following is the data analysis for a sample patient (patient 2) whose results were representative of what was observed in the data of the other two patients. The comparison between isodose lines for patient 2, in figure 3.10, shows a poor agreement between the film and TPS results. In particular, the 100% film isodose line rarely overlaps directly with the predicted TPS isodose and also contains a larger region in the center where the dose falls below 100%. Furthermore, although the 50 and 61 80% isodoses have improved agreement over the 100% isodose, in each case, there are multiple positions where the film results deviate largely from what the TPS predicts. Additionally, the 40% line routinely falls within where the TPS suggests it should be except towards the lower end where convex portions are observed. If this was indeed the case in patient treatment then it would simply mean that NT is receiving a lower dose; however, when considering film dosimetry, it means that the measurement results are not accurate enough to approve this treatment plan for patient use. From figures 3.11 and 3.12, it can be seen how the cross sectional dose profile through the isocenter of the film measurements compares with the TPS results. In figure 3.11, the dose profile between the two distributions accurately agrees within the center of the PTV but begins to have a discrepancy out towards the edges where the dose fall off region occurs. Figure 3.12 shows similar results in the very center of the PTV and in the dose fall off regions; however, at the edges of the PTV, two horns of increased dose are visible on the film measurements depositing as much as 9-13% more dose at these points. The dose fall off regions show a steeper gradient than those calculated by the TPS, of which, would be beneficial to patients regarding NT sparing; however, the horns in the y plane create a large disagreement between the two plan types, reinforcing the result that this plan is not suitable for clinical use. Similarly, the dose difference fluence map, in figure 3.13, displays regions where the recorded values differ by >10% further supporting the unsatisfactory measurements obtained for this patient. By following the film dosimetry protocol put forth by Micke et al., it should be possible to obtain film mea- surements that are nearly identical to what is predicted by the TPS; however, these ideal results were not what was observed in this study and, because of this, an inquiry in the reasons for this disgreement is warranted [74]. There are a multitude of reasons why such unfavorable results were obtained, each of which was encountered and considered throughout the film procedure. The first possible explanation is the Tongue and Groove effect in IMRT treatments, previously mentioned in Chapter 2.1.2. This effect has been extensively studied in the literature and was initially found to cause dose variations of up to 10-15% between dosimetrically measured and TPS results; however, with further investigation using radiochromic film, it was suggested that this effect may be worse with the discrepancy increasing to as much as 15-28% [73]. Additionally, while iPlan optimized the IMRT treatments to minimize this effect, individual MLCs in select beams were found to protrude beyond neighbouring MLCs and cover portions of the PTV with a tongue or groove, intensifying this effect. Thus, it is possible that this effect contributed a significant source of error to the EBT3 results. A second explanation is the uncertainty in the alignment of the film with the isocenter when positioning the CIRS head phantom in the frameless SRS system. It is possible that a combination of systematic uncertainties led to the film plane, within the head phantom, being positioned at a point that was not located at or tilted with respect to the isocenter of the Linac. The tolerances of the ExacTrac software allow for a variation of 1mm and a 1orotation, in all directions, to be considered accu- rately aligned while the film insert, within the CIRS head phantom, also contains small, but noticeable, air gaps 62 that allow the inserts to shift. Additionally, when forming the frameless mold for the CIRS phantom, the room lasers were used in an attempt to align the film plane such that it was perfectly parallel with the floor and; thus, perpendicular to the direction of propagation of the beam. This, however, added extra uncertainty to the film positioning as it required both the room lasers and the couch, where the mold was formed (a separate location from the Linac), to be accurately leveled. Another explanation comes from utlizing the Epson Expression scanner. The Epson Expression scanner was newly introduced to both film analysis and the BC Cancer Agency clinic at the time this work was performed and, because of this, there were few studies in the literature detailing the effects that this scanner may have on film analysis as well as a lack of a procedure to follow when examining EBT3 film. In fact, the film analysis performed in this thesis was the first time that EBT3 film was utilized in this clinic and the scanner had not yet been formally commissioned for use. A final explanation is that the FUGU software could not use the uniformly irradiated full piece of film to perform a nonuniformity correction from each patient measurement which was simply due to a coding/memory allocation problem within the software itself. Regardless of the possible sources of error that may affect the outcome of the film verification measurements, it was beyond the scope of this thesis to quantify the possible uncertainties in film positioning, perform the commissioning of the Epson scanner and/or investigate the FUGU code for the purpose of reprogramming it. Table 4.1: IMRT Stratification of Spetzler-Martin Grading Scheme The IMRT plan type per Spetzler-Martin grade combination is shown. The key aspects of the AVM grade that determine the corresponding IMRT recommendation are size and eloquence. If an AVM is situated in an eloquent part of the brain or is of a small size then PTV Only IMRT is recommended; however, if the AVM is of a large size and not eloquently located then OAR Low IMRT is recommended. AVM Grade Size Eloquence Venous Drainage IMRT Recommendation 1 1 0 0 PTV Only 2 1 1 0 PTV Only 1 0 1 PTV Only 2 0 0 OAR Low 3 1 1 1 PTV Only 2 1 0 PTV Only 2 0 1 OAR Low 3 0 0 OAR Low 4 2 1 1 PTV Only 3 1 0 OAR Low 3 0 1 OAR Low 5 3 1 1 OAR Low 63 5 Conclusion and Future Work 5.1 Conclusion The stratification of patients based on comparisons of dosimetric characteristics from 3DCRT and IMRT plans provides an insight into potential protocols available for the treatment of AVMs. By observing the magnitude of improvements IMRT provides in the parameters: PTV conformity, NT max dose, 7Gy isodose volume, and 12Gy isodose volume, over 3DCRT and implementing the use of this treatment technique in the clinic, patients can receive a more targeted SRS session depending on the characteristics of their specific PTV. In particular, the discrete dosimetric parameter for which IMRT could provide a notable benefit depends primarily on the size and shape of the PTV in question. Essentially, patients with PTVs that are a relatively small volume will receive the greatest benefit in the conformity of the prescription isodose sparing NT, immediately adjacent to the PTV, of doses on the order of >18Gy. Furthermore, the measure of this conformity improvement varies between IMRT plan types: PTV Only, OAR Low, OAR Medium, and OAR high, with PTV Only providing the utmost improvement for a larger proportion of patients. Alternatively, patients with large volume PTVs receive the greatest benefit in reducing the max dose to NT, the 7Gy isodose volume, and the 12Gy isodose volume; however, the extent of these improvements does not remain consistent between plan types. In general, as the weighting on the NT constraints increases with changing IMRT plan, the max dose to both NT and the PTV increases and eventually reaches a point beyond clinical acceptability. While the NT max dose in OAR Low plans can be at an acceptable level, patients do not receive remarkable reductions to this dose unless being treated with PTV Only IMRT. On the other hand, OAR Low IMRT provided considerable reductions to both the 7 and 12Gy isodose volumes for PTVs of this type while PTV Only IMRT only showed improvements in the 12Gy volume. Besides simply noting the largest improvement in a single dosimetric parameter, it is important to determine which plan type provides the most overall NT sparing when the location of the PTV in the brain comes into question. The PTV Only plans provided a benefit to PTV conformity as well as moderately reduced the NT max dose and the 12Gy isodose volume. Conversely, the OAR Low plans reduced the 12Gy isodose volume by a similar amount across all patients and reduced the 7Gy isodose volume significantly while, in many cases, increasing the NT max dose. Thus, for PTVs adjacent to critical structures where it is important to minimize the dose these structures receive, PTV Only IMRT is the treatment technique of choice due to its ability to restrict the prescription dose to the PTV, reduce NT max dose considerably, and moderate the 12Gy isodose volume. Meanwhile, the dosimetric QA performed with the ion chamber and EBT3 film provided contradictory results. On one hand, the ion chamber measurements verified that the isocenter dose calculated in the iPlan TPS matched the measured results within uncertainty at the isocenter. This suggested that the MUs calculated in the TPS 64 accurately deliver the required absolute dose at the center of the PTV. While, on the other hand, the film results contained a large amount of discrepancies between the TPS and 2D measurements. These discrepancies do not allow for IMRT to be deemed clinically acceptable for the stereotactic treatment of AVMs, requiring additional measurements before patients may receive the potential benefits offered by IMRT. Ultimately, the work presented in this thesis indicates the potential for significant improvements to patient outcomes when receiving a SRS treatment for an AVM. Furthermore, the particular dosimetric parameter that receives the greatest benefits from IMRT is dependent only on PTV volume, diameter, surface area with these three characteristics also determining the corresponding magnitude of improvement. Therefore, patient benefits from IMRT treatments over 3DCRT can be readily predicted on a case-by-case basis, from a simple examination of patient CT scans, providing a treatment protocol to reduce the risk of symptomatic injury in the clinic. 5.2 Future Work Ongoing examinations of dosimetric characteristics in stereotactic IMRT treatments may provide improve- ments in multiple intracranial treatment strategies. IMRT is a comprehensive treatment modality that has the capability of improving patient outcomes for not only AVM treatments but also lesions that are located in anatom- ical locations associated with complications. In particular, by utilizing IMRT in the treatment of a variety of lesion types, improved sparing of white matter tracks and critical structures such as the brainstem and optic nerves can be achieved. However, before this can be accomplished the clinical commissioning of the IMRT software and all the equipment utilized throughout the verification process must undergo intense scrutiny. Specifically, the processes used for film dosimetry must be meticulously tested and reproducibly documented. Scanning procedures for all scanners available in the clinic need to be carefully examined in order to quantify the effects they may have on the film measurements and determine the steps required to minimize any uncertainties that may be introduced. Monte Carlo modeling is another widely used tool in dosimetric verification that would be beneficial to include in the commissioning of IMRT treatments. By applying Monte Carlo calculations to the specific iPlan IMRT technique used in this thesis, increasingly accurate dose estimations, improved over the pencil beam algorithms currently used, can be acquired. It would be essential that this code precisely reproduce the rapidly changing dose rate while correctly imitating the step-and-shoot technique with randomly sized beamlets. Additionally, in stereotactic treatments there is a loss of lateral electronic equilibrium due to the small field sizes required for accurate dose delivery; therefore, Monte Carlo modeling must also take this into account. All in all, with a systematically developed Monte Carlo simulation, the dose that patients receive can be accurately predicted, providing a gold standard for which ion chamber and film measurements can be compared against and improving 65 the quality of care that patients experience. 66 References [1] Fleetwood IG, Steinberg GK. Arteriovenous malformations. Lancet. 2002;359:863-873. [2] Madana J, Yolmo D, Saxena SK, Gopalakrishnan S, Nath AK. Giant congenital auricular arteriovenous mal- formation. Auris Nasus Larynx. 2010;37:511-514. [3] Halim AX, Singh V, Johnston SC, et al. Characteristics of brain arteriovenous malformations with coexisting aneurysms: a comparison of two referral centers. Stroke. 2002;33:675-679. [4] Ogawa H, Itoh S, Mori Y, Suzuki K, Ota T, Naganawa S. Arteriovenous malformation of the pancreas: assess- ment of clinical and multislice CT features. Abdom Imaging. 2009;34(6):743-752. [5] Rizvi AZ, Kaufman JA, Smith P, Silen ML. Solitary arteriovenous malformation of the small intestine. J Am Coll Surg. 2005;200(5):808-809. [6] Krings T, Mull M, Gilsbach JM, Thron A. Spinal vascular malformations. Eur Radiol. 2005;15(2):267-278. [7] Shaver J. Eyelid arteriovenous malformation treated with embolization leading to a branch retinal artery oc- clusion. Optometry. 2011;82(12):744-750. [8] Bandi G, Bianco FJ, Dhabuwala CB. Recurrent scrotal arteriovenous malformation. J Urol. 2004;171:1628. [9] Fukuda Y, Murata Y, Umehara I, et al. Perfusion and blood pool scintigraphy for diagnosing soft-tissue arteriovenous malformations. Clin Nucl Med. 1999;24(4):232-234. [10] Ogilvy CS, Stieg PE, Awad I, et al. Recommendations for the management of intracranial arteriovenous mal- formations: a statement for healthcare professionals from a special writing group of the stroke council, american stroke association. Circulation. 2001;103:2644-2657. [11] ApSimon HT, Reef H, Phadke RV, Popovic EA. A population-based study of brain arteriovenous malforma- tion: long-term treatment outcomes. Stroke. 2002;33(12):2794-2800. [12] Perret G, Nishioka H. Report on the cooperative study of intracranial aneurysms and subarachnoid hem- orrhage. Section VI. Arteriovenous malformations: an analysis of 545 cases of cranio-cerebral arteriovenous malformations and fistulae reported to the cooperative study. J Neurosurg. 1966;25(4):467-490. [13] Monteiro JM, Rosas MJ, Correia AP, Vaz AR. Migraine and intracranial vascular malformations. Headache. 1993;33(10):563-565. [14] Brown RD, Wiebers DO, Torner JC, O’Fallon WM. Incidence and prevalence of intracranial vascular mal- formations in Olmsted County, Minnesota, 1965 to 1992. Neurology. 1996;46(4):949-952. [15] The Toronto Brain Vascular Malformation Study Group. Brain arteriovenous malformations (AVM). http://b- rainavm.uhnres.utoronto.ca/malformations/brain_avm_index.htm. Last visited Oct. 30,2012. [16] De Salles AAF and Gorgulho AA. Radiosurgery for arteriovenous malformations. In: De Salles AAF, Gorgulho AA, Agazaryan N, et al., editors. Shaped beam radiosurgery: state of the art. Germany: Springer 67 Berlin-Heidelberg;2011:77-88. [17] Brown RD, Wiebers DO, Forbes G, et al. The natural history of unruptured intracranial arteriovenous mal- formations. J Neurosurg. 1988;68:352–357. [18] Graf CJ, Perret GE, Torner JC. Bleeding from cerebral arteriovenous malformations as part of their natural history. J Neurosurg. 1983;58:331–337. [19] Crawford PM, West CR, Chadwick DW, Shaw MDM. Arteriovenous malformations of the brain: natural history in unoperated patients. J Neurol Neurosurg Psychiatry. 1986;49:1–10. [20] Ondra SL, Troupp H, George ED, Schwab K. The natural history of symptomatic arteriovenous malforma- tions of the brain: a 24-year follow-up assessment. J Neurosurg. 1990;73:387–391. [21] Mast H, Young WL, Koennecke HC, et al. Risk of spontaneous haemorrhage after diagnosis of cerebral arteriovenous malformation. Lancet. 1997;350:1065–1068. [22] Kondziolka D, McLaughlin MR, Kestle JR. Simple risk predictions for arteriovenous malformation hemor- rhage. Neurosurgery. 1995;37(5):851–855. [23] Brown RD. Simple risk predictions for arteriovenous malformation hemorrhage. Neurosurgery. 2000;46(4): 1024. Letter. [24] Perret GE, Nishioka H. Arteriovenous malformations: an analysis of 545 cases of cranio-cerebral arteri- ovenous malformations and fistulae reported to the cooperative study. In: Sahs AL, Perret GE, Locksley HB, Nishioka H, eds. Intracranial aneurysms and subarachnoid hemorrhage: a cooperative study. Philadelphia: JB Lippincott;1969:200–222. [25] Wilkins RH. Natural history of intracranial vascular malformations: a review. Neurosurgery. 1985;16(3): 421–430. [26] Brown RD, Wiebers DO, Torner JC, O’Fallon WM. Frequency of intracranial hemorrhage as a present- ing symptom and subtype analysis: a population-based study of intracranial vascular malformations in Olmsted Country, Minnesota. J Neurosurg. 1996;85:29–32. [27] Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg. 1986;65: 476-483. [28] Adler JR Jr., Colombo F, Heilbrun MP, Winston K. Toward an expanded view of radiosurgery. Neurosurgery. 2004;55:1374-1376. [29] Otto, K. Volumetric modulated arc therapy: IMRT in a single gantry arc. Med Phys. 2008;35(1):310-317. [30] Bambakidis NC, Selman WR. Stereotactic radiosurgery for arteriovenous malformations. J Neurosurg. 2012;116:3-6. [31] Klopfenstein JD, Spetzler RF. Cerebral arteriovenous malformations: when is surgery indicated? Acta Neu- rochir (Wien). 2005;147:693-695. 68 [32] Flickinger JC, Kondziolka D, Lunsford LD, et al. Development of a model to predict permanent symptomatic postradiosurgery injury for arteriovenous malformation patients. Int J Radiat Oncol Biol Phys. 2000;46(5):1143- 1148. [33] Steiner L, Lindquist C, Cail W, Karlsson B, Steiner M. Microsurgery and radiosurgery in brain arteriovenous malformations. J Neurosurg. 1993;79:647-652. [34] Lunsford LD, Kondziolka D, Flickinger JC, et al. Stereotactic radiosurgery for arteriovenous malformations of the brain. J Neurosurg. 1991;75:512-524. [35] Blamek S, Tarnawski R, Miszcyk L. Linac-based stereotactic radiosurgery for brain arteriovenous malforma- tions. Clin Oncol. 2011;23:525-531. [36] Flickinger JC, Pollock BE, Kondziolka D, Lunsford LD. A dose-response analysis of arteriovenous malfor- mation obliteration after radiosurgery. Int J Radiat Oncol Biol Phys. 1996;36(4):873-879. [37] Federico C, Franco P, Giorgio C, Leopoldo C, Giampaolo DL, Paolo F. Linear accelerator radiosurgery of cerebral arteriovenous malformations: an update. Neurosurgery. 1994;34(1):14-21. [38] Friedman WA, Bova FJ, Bollampally S, Bradshaw P. Analysis of factors predictive of success or complica- tions in arteriovenous malformation radiosurgery. Neurosurgery. 2003;52:296-308. [39] SkjØth-Rasmussen J, Roed H, Ohlhues L, Jespersen B, Juhler M. Complications following linear accel- erator based stereotactic radiation for cerebral arteriovenous malformations. Int J Radiat Oncol Biol Phys. 2010;77(2):542-547. [40] Miyawaki L, Dowd C, Wara W, et al. Five year results of linac radiosurgery for arteriovenous malformations: outcome for large AVMs. Int J Radiat Oncol Biol Phys. 1999;44(5):1089-1106. [41] Herbert C, Moiseenko V, McKenzie M, et al. Factors predictive of obliteration after arteriovenous malforma- tion radiosurgery. Can J Neurol Sci. 2011;38:845-850. [42] Zabel A, Milker-Zabel S, Huber P, Schulz-Ertner D, Schlegel W, Debus J. Treatment outcome after linac- based radiosurgery in cerebral arteriovenous malformations: retrospective analysis of factors affecting oblitera- tion. Radiother Oncol. 2005;77:105-110. [43] Flickinger JC, Kondziolka D, Maitz AH, Lunsford LD. Analysis of neurological sequelae from radiosurgery of arteriovenous malformations: how location affects outcome. Int J Radiat Oncol Biol Phys. 1998;40(2):273- 278. [44] Herbert C, Moiseenko V, McKenzie M, et al. Factors predictive of symptomatic radiation injury after linear accelerator-based stereotactic radiosurgery for intracerebral arteriovenous malformations. Int J Radiat Oncol Biol Phys. 2012;83(3):872-877. [45] Lawrence YR, Li XA, Naqa IE, et al. Radiation dose-volume effects in the brain. Int J Radiat Oncol Biol Phys. 2010;76(Suppl. 3):S20-S27. 69 [46] Laborda AZ. 3D conformal radiotherapy and intensity-modulated radiotherapy: towards dose optimization. Rev Oncol. 2002;4(2):103-107. [47] Fraass BA, Kessler ML, McShan DL, et al. Optimization and clinical use of multisegment intensity- modulated radiation therapy for high-dose conformal therapy. Semin Radiat Oncol. 1999;9(1):60-77. [48] Spirou SV, Chui CS. Generation of arbitrary intensity profiles by dynamic jaws or multileaf collimators. Med Phys. 1994;21:1031-1041. [49] BrainLAB Inc. (private communication). [50] Intensity Modulated Radiation Therapy Collaborative Working Group. Intensity-modulated radiotherapy: current status and issues of interest. Int J Radiat Oncol Biol Phys. 2001;51(4):880-914. [51] Leibel SA, Fuks Z, Zelefsky MJ, et al. Intensity-modulated radiotherapy. Cancer J. 2002;8(2):164-176. [52] Taylor A, Powell MEB. Intensity-modulated radiotherapy-what is it? Cancer Imaging. 2004;4:68-73. [53] Mechalakos JG, St. Germain J, Burman CM. Results of a one year survey of output for linear accelerators using IMRT and non-IMRT techniques. J Appl Clin Med Phys. 2004;5(1):64-72. [54] Klein EE, Hanley J, Bayouth J, et al. Task group 142 report: quality assurance of medical accelerators. Med Phys. 2009;36(9):4197-4212. [55] Palta JR, Liu C, Li JG. Quality assurance of intensity-modulated radiation therapy. Int J Radiat Oncol Biol Phys. 2008;71(Suppl. 1):S108-S112. [56] Karger CP, Jäkel O, Palmans H, Kanai T. Dosimetry for ion beam radiotherapy. Phys Med Biol. 2010;55:R193- R234. [57] Dieterich S, Sherouse GW. Experimental comparison of seven commercial dosimetry diodes for measurement of stereotactic radiosurgery cone factors. Med Phys. 2011;38(7):4166-4173. [58] Yukihara EG, Mardirossian G, Mirzasadeghi M, Guduru S, Ahmad S. Evaluation of Al2O3:C optically stimulated luminescence (OSL) dosimeters for passive dosimetry of high-energy photon and electron beams in radiotherapy. Med Phys. 2008;35(1):260-269. [59] Ross CK, Klassen NV. Water calorimetry for radiation dosimetry. Phys Med Biol. 1996;41:1-29. [60] Kinhikar RA, Upreti R, Sharma S, Tambe CM, Deshpande DD. Intensity modulated radiotherapy dosimetry with ion chambers, TLD, MOSFET and EDR2 film. Australas Phys Eng Sci Med. 2007;30(1):25-32. [61] Fenoglietto P, Laliberté B, Aillères N, Riou O, Dubois JB, Azria D. Eight years of IMRT quality assurance with ionization chambers and film dosimetry: experience of the Montpellier comprehensive cancer center. Radiat Oncol. 2011;6(85):1-11. [62] Papatheodorou S, Rosenwald JC, Zefkili S, Murillo MC, Drouard J, Gaboriaud G. Dose calculation and verification of intensity modulation generated by dynamic multileaf collimators. Med Phys. 2000;27(5):960-971. 70 [63] Aland T, Kairn T, Kenny J. Evaluation of a gafchromic EBT2 film dosimetry system for radiotherapy quality assurance. Australas Phys Eng Sci Med. 2011;34:251-260. [64] Zeidan OA, Stephenson SAL, Meeks SL, et al. Characterization and use of EBT radiochromic film for IMRT dose verification. Med Phys. 2006;33(11):4064-4072. [65] International Specialty Products. Gafchromic® EBT2 self-developing film for radiotherapy dosimetry (Feb. 2009). http://online1.ispcorp.com/_layouts/Gafchromic/content/products/ebt2/pdfs/GAFCHROMICEBT2Techni- calBrief-Rev1.pdf (Accessible from http://gafchromic.com). Last visited Oct. 30, 2012. [66] Niroomand-Rad A, Blackwell CR, Coursey BM, et al. Radiochromic film dosimetry: recommendations of AAPM radiation therapy committee task group 55. Med Phys. 1998;25(11):2093-2115. [67] Bucciolini M, Buonamici FB, Casati M. Verification of IMRT fields by film dosimetry. Med Phys. 2004;31(1): 161-168. [68] Lynch BD, Kozelka J, Ranade MK, Li JG, Simon WE, Dempsey JF. Important considerations for ra- diochromic film dosimetry with flatbed CCD scanners and EBT GAFCHROMIC® film. Med Phys. 2006;33(12): 4551-4556. [69] International Specialty Products. Gafchromic® EBT2 scan handling guide (Feb. 2009). http://online1.ispcor- p.com/_layouts/Gafchromic/content/products/ebt2/pdfs/GAFCHROMICEBT2ScanHandlingGuide.pdf (Accessi- ble from http://gafchromic.com). Last visited Oct. 30, 2012. [70] Baldock C, De Deene Y, Doran S, et al. Polymer gel dosimetry. Phys Med Biol. 2010;55:R1-R63. [71] McJury M, Oldham M, Cosgrove VP, et al. Radiation dosimetry using polymer gels: methods and applica- tions. Br J Radiol. 2000;73:919-929. [72] Guo P, Adamovics J, Oldham M. A practical three-dimensional dosimetry system for radiation therapy. Med Phys. 2006;33(10):3962-3972. [73] Deng J, Pawlicki T, Chen Y, Li J, Jiang SB, Ma C-M. The MLC tongue-and-groove effect on IMRT dose distributions. Phys Med Biol. 2001;46:1039-1060. [74] Micke A, Lewis DF, Yu X. Multichannel film dosimetry with nonuniformity correction. Med Phys. 2011;38(5): 2523-2534. [75] Levegrün S, Hof H, Essig M, Schlegel W, Debus J. Radiation-induced changes of brain tissue after radio- surgery in patients with arteriovenous malformations: correlation with dose distribution parameters. Int J Radiat Oncol Biol Phys. 2004;59(3):796-808. 71 Appendix A. Patient DVHs Figure A.1: DVH Comparisons for Patient 1 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 1. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 72 Figure A.2: DVH Comparisons for Patient 2 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 2. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 73 Figure A.3: DVH Comparisons for Patient 3 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 3. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 74 Figure A.4: DVH Comparisons for Patient 4 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 4. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 75 Figure A.5: DVH Comparisons for Patient 5 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 5. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 76 Figure A.6: DVH Comparisons for Patient 6 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 6. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 77 Figure A.7: DVH Comparisons for Patient 7 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 7. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 78 Figure A.8: DVH Comparisons for Patient 8 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 8. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 79 Figure A.9: DVH Comparisons for Patient 9 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 9. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 80 Figure A.10: DVH Comparisons for Patient 10 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 10. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 81 Figure A.11: DVH Comparisons for Patient 11 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 11. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 82 Figure A.12: DVH Comparisons for Patient 12 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 12. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 83 Figure A.13: DVH Comparisons for Patient 13 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 13. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 84 Figure A.14: DVH Comparisons for Patient 14 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 14. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 85 Figure A.15: DVH Comparisons for Patient 15 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 15. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 86 Figure A.16: DVH Comparisons for Patient 16 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 16. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 87 Figure A.17: DVH Comparisons for Patient 17 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 17. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 88 Figure A.18: DVH Comparisons for Patient 18 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 18. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 89 Figure A.19: DVH Comparisons for Patient 19 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 19. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 90 Figure A.20: DVH Comparisons for Patient 20 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 20. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 91 Figure A.21: DVH Comparisons for Patient 21 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 21. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 92 Figure A.22: DVH Comparisons for Patient 22 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 22. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 93 Figure A.23: DVH Comparisons for Patient 23 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 23. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 94 Figure A.24: DVH Comparisons for Patient 24 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 24. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 95 Figure A.25: DVH Comparisons for Patient 25 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 25. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 96 Figure A.26: DVH Comparisons for Patient 26 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 26. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 97 Figure A.27: DVH Comparisons for Patient 27 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 27. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 98 Figure A.28: DVH Comparisons for Patient 28 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 28. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 99 Figure A.29: DVH Comparisons for Patient 29 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 29. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 100 Figure A.30: DVH Comparisons for Patient 30 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 30. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 101 Figure A.31: DVH Comparisons for Patient 31 (a) PTV Only and (b) OAR Low IMRT vs 3DCRT for the PTV (pink) and NT OAR Sphere (green) are shown for patient 31. The solid lines depict the IMRT results for the respective plan while the dashed line depicts the 3DCRT plan. (a) (b) 102