Search

OBJECTIVE To investigate treat-and-extend (T&E) regimens of anti-vascular endothelial growth factor (anti-VEGF) therapy for the treatment of macular oedema secondary to retinal vein occlusions (RVOs). METHODS Ovid MEDLINE, Ovid EMBASE and CENTRAL were searched on 25 February 2021. Randomized controlled trials, cohort studies, case-control studies and case series were included. The primary outcome was the change in Early Treatment Diabetic Retinopathy Score (ETDRS) letters from baseline. Conversions from Snellen to ETDRS letters were performed utilizing a published protocol. Secondary outcomes included improvement in retinal thickness from baseline, number of anti-VEGF injections and frequency of adverse events. Outcomes were examined at 12 and 24 months. Certainty of evidence was assessed utilizing GRADE (Grading of Recommendations Assessments, Development and Evaluations) guidelines. RESULTS Seven hundred eighty-six eyes from 16 studies were included. Meta-analysis demonstrated a mean improvement of 15.7 (95% CI: 13.3-18.0) ETDRS letters at 12 months. Central retinal thickness improved 269.7 μm (95% CI: 233.64-305.90) at 12 months. Injections were performed 8.1 (95% CI: 7.4-8.7) and 13.1 (95% CI: 9.4-16.8) times at 12 and 24 months respectively. Adverse events were infrequent across all studies. Grading of Recommendations Assessments, Development and Evaluations (GRADE) certainty of evidence was very low across all outcomes. CONCLUSIONS The results support the viability of T&E regimens for the treatment of macular oedema secondary to RVOs.

PURPOSE: The current process for radiotherapy treatment plan quality assurance relies on human inspection of treatment plans, which is time‐consuming, error prone and oft reliant on inconsistently applied professional judgments. A previous proof‐of‐principle paper describes the use of a Bayesian network (BN) to aid in this process. This work studied how such a BN could be expanded and trained to better represent clinical practice. METHODS: We obtained 51 540 unique radiotherapy cases including diagnostic, prescription, plan/beam, and therapy setup factors from a de‐identified Elekta oncology information system from the years 2010–2017 from a single institution. Using a knowledge base derived from clinical experience, factors were coordinated into a 29‐node, 40‐edge BN representing dependencies among the variables. Conditional probabilities were machine learned using expectation maximization module using all data except a subset of 500 patient cases withheld for testing. Different classes of errors that were obtained from incident learning systems were introduced to the testing set of cases which were withheld from the dataset used for building the BN. Different sizes of datasets were used to train the network. In addition, the BN was trained using data from different length epochs as well as different eras. Its performance under these different conditions was evaluated by means of Areas Under the receiver operating characteristic Curve (AUC). RESULTS: Our performance analysis found AUCs of 0.82, 0.85, 0.89, and 0.88 in networks trained with 2‐yr, 3‐yr 4‐yr and 5‐yr windows. With a 4‐yr sliding window, we found AUC reduction of 3% per year when moving the window back in time in 1‐yr steps. Compared to the 4‐yr window moved back by 4 yrs (2010–2013 vs 2014–2017), the largest component of overall reduction in AUC over time was from the loss of detection performance in plan/beam error types. CONCLUSIONS: The expanded BN method demonstrates the ability to detect classes of errors commonly encountered in radiotherapy planning. The results suggest that a 4‐yr training dataset optimizes the performance of the network in this institutional dataset, and that yearly updates are sufficient to capture the evolution of clinical practice and maintain fidelity.

The recent explosion in machine learning efforts in the quality assurance (QA) space has produced a variety of proofs-of-concept many with promising results. Expected outcomes of model implementation include improvements in planning time, plan quality, advanced dosimetric QA, predictive machine maintenance, increased safety checks, and developments key for new QA paradigms driven by adaptive planning. In this article, we outline several areas of research and discuss some of the unique challenges each area presents.

Purpose: The final state of the tumor at the end of a radiotherapy course is dependent on the doses given in each fraction during the treatment course. This study investigates the feasibility of using dynamic adaptive radiotherapy (DART) in treatinglungcancers assuming CBCT is available to observe midtreatment tumor states. DART adapts treatment plans using a dynamic programming technique to consider the expected changes of the tumor in the optimization process. Methods: DART is constructed using a stochastic control formalism framework. It minimizes the total expected number of tumor cells at the end of a treatment course, which is equivalent to maximizing tumor control probability, subject to the uncertainty inherent in the tumor response. This formulation allows for nonstationary dose distributions as well as nonstationary fractional doses as needed to achieve a series of optimal plans that are conformal to the tumor over time, i.e., spatiotemporally optimal plans. Sixteen phantom cases with various sizes and locations of tumors and organs-at-risk (OAR) were generated using in-house software. Each case was planned with DART and conventional IMRT prescribing 60 Gy in 30 fractions. The observations of the change in the tumor volume over a treatment course were simulated using a two-level cell population model. Monte Carlo simulations of the treatment course for each case were run to account for uncertainty in the tumor response. The same OAR dose constraints were applied for both methods. The frequency of replanning was varied between 1, 2, 5 (weekly), and 29 times (daily). The final average tumordose and OAR doses have been compared to quantify the potential dosimetric benefits of DART. Results: The average tumor max, min, mean, and D95 doses using DART relative to these using conventional IMRT were 124.0%–125.2%, 102.1%–114.7%, 113.7%–123.4%, and 102.0%–115.9% (range dependent on the frequency of replanning). The average relative maximum doses for the cord and esophagus, mean doses for the heart and lungs, and D05 for the unspecified tissue resulting 84%–102.4%, 99.8%–106.9%, 66.9%–85.6%, 58.2%–78.8%, and 85.2%–94.0%, respectively. Conclusions: It is feasible to apply DART to the treatment of NSCLC using CBCT to observe the midtreatment tumor state. Potential increases in the tumordose and reductions in the OAR dose, particularly for parallel OARs with mean or dose–volume constraints, could be achieved using DART compared to nonadaptive IMRT.

Purpose: To investigate the impact of using spatiotemporal optimization, i.e., intensity-modulated spatial optimization followed by fractionation schedule optimization, to select the patient-specific fractionation schedule that maximizes the tumor biologically equivalent dose (BED) under dose constraints for multiple organs-at-risk (OARs). Methods: Spatiotemporal optimization was applied to a variety of lung tumors in a phantom geometry using a range of tumor sizes and locations. The optimal fractionation schedule for a patient using the linear-quadratic cell survival model depends on the tumor and OAR sensitivity to fraction size (α/β), the effective tumor doubling time (Td ), and the size and location of tumor target relative to one or more OARs (dose distribution). The authors used a spatiotemporal optimization method to identify the optimal number of fractions N that maximizes the 3D tumor BED distribution for 16 lung phantom cases. The selection of the optimal fractionation schedule used equivalent (30-fraction) OAR constraints for the heart (D mean ≤ 45 Gy), lungs (D mean ≤ 20 Gy), cord (D max ≤ 45 Gy), esophagus (D max ≤ 63 Gy), and unspecified tissues (D 05 ≤ 60 Gy). To assess plan quality, the authors compared the minimum, mean, maximum, and D 95 of tumor BED, as well as the equivalent uniform dose (EUD) for optimized plans to conventional intensity-modulated radiation therapy plans prescribing 60 Gy in 30 fractions. A sensitivity analysis was performed to assess the effects of Td (3–100 days), tumor lag-time (Tk = 0–10 days), and the size of tumors on optimal fractionation schedule. Results: Using an α/β ratio of 10 Gy, the average values of tumor max, min, mean BED, and D 95 were up to 19%, 21%, 20%, and 19% larger than those from conventional prescription, depending on Td and Tk used. Tumor EUD was up to 17% larger than the conventional prescription. For fast proliferating tumors with Td less than 10 days, there was no significant increase in tumor BED but the treatment course could be shortened without a loss in tumor BED. The improvement in the tumor mean BED was more pronounced with smaller tumors (p-value = 0.08). Conclusions: Spatiotemporal optimization of patient plans has the potential to significantly improve local tumor control (larger BED/EUD) of patients with a favorable geometry, such as smaller tumors with larger distances between the tumor target and nearby OAR. In patients with a less favorable geometry and for fast growing tumors, plans optimized using spatiotemporal optimization and conventional (spatial-only) optimization are equivalent (negligible differences in tumor BED/EUD). However, spatiotemporal optimization yields shorter treatment courses than conventional spatial-only optimization. Personalized, spatiotemporal optimization of treatment schedules can increase patient convenience and help with the efficient allocation of clinical resources. Spatiotemporal optimization can also help identify a subset of patients that might benefit from nonconventional (large dose per fraction) treatments that are ineligible for the current practice of stereotactic body radiation therapy.

We evaluate a photon convolution-superposition algorithm used to model a fast neutron therapy beam in a commercial treatment planning system (TPS). The neutron beam modeled was the Clinical Neutron Therapy System (CNTS) fast neutron beam produced by 50 MeV protons on a Be target at our facility, and we implemented the Pinnacle3 dose calculation model for computing neutron doses. Measured neutron data were acquired by an IC30 ion chamber flowing 5 cc/min of tissue equivalent gas. Output factors and profile scans for open and wedged fields were measured according to the Pinnacle physics reference guide recommendations for photon beams in a Wellhofer water tank scanning system. Following the construction of a neutron beam model, computed doses were then generated using 100 monitor units (MUs) beams incident on a water-equivalent phantom for open and wedged square fields, as well as multileaf collimator (MLC)-shaped irregular fields. We compared Pinnacle dose profiles, central axis doses, and off-axis doses (in irregular fields) with 1) doses computed using the Prism treatment planning system, and 2) doses measured in a water phantom and having matching geometry to the computation setup. We found that the Pinnacle photon model may be used to model most of the important dosimetric features of the CNTS fast neutron beam. Pinnacle-calculated dose points among open and wedged square fields exhibit dose differences within 3.9 cGy of both Prism and measured doses along the central axis, and within 5 cGy difference of measurement in the penumbra region. Pinnacle dose point calculations using irregular treatment type fields showed a dose difference up to 9 cGy from measured dose points, although most points of comparison were below 5 cGy. Comparisons of dose points that were chosen from cases planned in both Pinnacle and Prism show an average dose difference less than 0.6%, except in certain fields which incorporate both wedges and heavy blocking of the central axis. All clinical cases planned in both Prism and Pinnacle were found to be comparable in terms of dose-volume histograms and spatial dose distribution following review by the treating clinicians. Variations were considered minor and within clinically acceptable limits by the treating clinicians. The Pinnacle TPS has sufficient computational modeling ability to adequately produce a viable neutron model for clinical use in treatment planning.

Purpose: To evaluate how a more flexible and thorough multiobjective search of feasible IMRT plans affects performance in IMRT optimization. Methods: A multiobjective evolutionary algorithm (MOEA) was used as a tool to investigate how expanding the search space to include a wider range of penalty functions affects the quality of the set of IMRT plans produced. The MOEA uses a population of IMRT plans to generate new IMRT plans through deterministic minimization of recombined penalty functions that are weighted sums of multiple, tissue-specific objective functions. The quality of the generated plans are judged by an independent set of nonconvex, clinically relevant decision criteria, and all dominated plans are eliminated. As this process repeats itself, better plans are produced so that the population of IMRT plans will approach the Pareto front. Three different approaches were used to explore the effects of expanding the search space. First, the evolutionary algorithm used genetic optimization principles to search by simultaneously optimizing both the weights and tissue-specific dose parameters in penalty functions. Second, penalty function parameters were individually optimized for each voxel in all organs at risk (OARs) in the MOEA. Finally, a heuristic voxel-specific improvement (VSI) algorithm that can be used on any IMRT plan was developed that incrementally improves voxel-specific penalty function parameters for all structures (OARs and targets). Different approaches were compared using the concept of domination comparison applied to the sets of plans obtained by multiobjective optimization. Results: MOEA optimizations that simultaneously searched both importance weights and dose parameters generated sets of IMRT plans that were superior to sets of plans produced when either type of parameter was fixed for four example prostate plans. The amount of improvement increased with greater overlap between OARs and targets. Allowing the MOEA to search for voxel-specific penalty functions improved results for simple cases with three structures but did not improve results for a more complex case with seven structures. For this modification, the amount of improvement increased with less overlap between OARs and targets. The voxel-specific improvement algorithm improved results for all cases, and its clinical relevance was demonstrated in a complex prostate and a very complex head and neck case. Conclusions: Using an evolutionary algorithm as a tool, it was found that allowing more flexibility in the search space enhanced performance. The two strategies of (a) varying the weights and reference doses in the objective function and (b) removing the constraint of equal penalties for all voxels in a structure both generated sets of plans that dominated sets of plans considered to be “Pareto optimal” within the conventional, more limited search space. When considering voxel-specific objectives, the very large search space can lead to convergence problems in the MOEA for complex cases, but this is not an issue for the VSI algorithm.

In common usage, historical distance refers to a position of detached observation made possible by the passage of time. Understood in these terms, distance has long been regarded as essential to modern historical practice, but this conception narrows the idea of distance and burdens it with a regulatory purpose. I argue that distance needs to be re-conceived in terms of the wider set of engagements that mediate our relations to the past, as well as the full spectrum of distance-positions from near to far. Re-imagined in these terms, distance sheds its prescriptiveness and becomes a valuable heuristic for examining the history of historical representation. When distance is studied in relation to the range of mediations entailed in historical representation, it becomes evident that the plasticities of distance/proximity are by no means limited to gradients of time; rather, temporality is bound up with other distances that come from our need to engage with the historical past as (simultaneously) a realm of making, of feeling, of doing, and of understanding. Thus for every historical work, we need to consider at least four basic dimensions of representation as they relate to the problem of mediating distance: 1. the genres, media, and vocabularies that shape the history's formal structures of representation; 2. the affective claims made by the historical account, including the emotional experiences it promises or withholds; 3. the work's implications for action, whether of a political or moral nature; and 4. the modes of understanding on which the history's intelligibility depends. These overlapping, but distinctive, distancesa-formal, affective, ideological, and conceptuala-provide an analytic framework for examining changing modes of historical representation.

Aim: The aim of this study was to investigate the incidence, risk factors and developmental outcomes of cerebellar haemorrhage in the extremely preterm infant.

Purpose: To investigate how using different sets of decision criteria impacts the quality of intensity modulated radiation therapy(IMRT) plans obtained by multiobjective optimization. Methods: A multiobjective optimization evolutionary algorithm (MOEA) was used to produce sets of IMRT plans. The MOEA consisted of two interacting algorithms: (i) a deterministic inverse planning optimization of beamlet intensities that minimizes a weighted sum of quadratic penalty objectives to generate IMRT plans and (ii) an evolutionary algorithm that selects the superior IMRT plans using decision criteria and uses those plans to determine the new weightsand penalty objectives of each new plan. Plans resulting from the deterministic algorithm were evaluated by the evolutionary algorithm using a set of decision criteria for both targets and organs at risk (OARs). Decision criteria used included variation in the target dose distribution, mean dose, maximum dose, generalized equivalent uniform dose (gEUD), an equivalent uniform dose ( EU D α , β ) formula derived from the linear-quadratic survival model, and points on dose volume histograms (DVHs). In order to quantatively compare results from trials using different decision criteria, a neutral set of comparison metrics was used. For each set of decision criteria investigated, IMRT plans were calculated for four different cases: two simple prostate cases, one complex prostate Case, and one complex head and neck Case. Results: When smaller numbers of decision criteria, more descriptive decision criteria, or less anticorrelated decision criteria were used to characterize plan quality during multiobjective optimization,dose to OARs and target dose variation were reduced in the final population of plans. Mean OAR dose and gEUD (a = 4) decision criteria were comparable. Using maximum dose decision criteria for OARs near targets resulted in inferior populations that focused solely on low target variance at the expense of high OAR dose. Target dose range, ( D max - D min ) , decision criteria were found to be most effective for keeping targets uniform. Using target gEUD decision criteria resulted in much lower OAR doses but much higher target dose variation. EU D α , β based decision criteria focused on a region of plan space that was a compromise between target and OAR objectives. None of these target decision criteria dominated plans using other criteria, but only focused on approaching a different area of the Pareto front. Conclusions: The choice of decision criteria implemented in the MOEA had a significant impact on the region explored and the rate of convergence toward the Pareto front. When more decision criteria, anticorrelated decision criteria, or decision criteria with insufficient information were implemented, inferior populations are resulted. When more informative decision criteria were used, such as gEUD, EU D α , β , target dose range, and mean dose, MOEA optimizations focused on approaching different regions of the Pareto front, but did not dominate each other. Using simple OAR decision criteria and target EU D α , β decision criteria demonstrated the potential to generate IMRT plans that significantly reduce dose to OARs while achieving the same or better tumor control when clinical requirements on target dose variance can be met or relaxed.

Purpose: To identify the most informative methods for reporting results of treatment planning comparisons. Methods: Seven articles from the past year of International Journal of Radiation Oncology Biology Physics reported on comparisons of treatment plans for IMRT and IMAT. The articles were reviewed to identify methods of comparisons. Decision theoretical concepts were used to evaluate the study methods and highlight those that provide the most information. Results: None of the studies examined the correlation between objectives. Statistical comparisons provided some information but not enough to provide support for a robust decision analysis. Conclusions: The increased use of treatment planning studies to evaluate different methods in radiation therapy requires improved standards for designing the studies and reporting the results.

Purpose: Current inverse planning methods for IMRT are limited because they are not designed to explore the trade-offs between the competing objectives between the tumor and normal tissues. Our goal was to develop an efficient multiobjective optimization algorithm that was flexible enough to handle any form of objective function and that resulted in a set of Pareto optimal plans. Methods: We developed a hierarchical evolutionary multiobjective algorithm designed to quickly generate a diverse Pareto optimal set of IMRT plans that meet all clinical constraints and reflect the trade-offs in the plans. The top level of the hierarchical algorithm is a multiobjective evolutionary algorithm (MOEA). The genes of the individuals generated in the MOEA are the parameters that define the penalty function minimized during an accelerated deterministic IMRT optimization that represents the bottom level of the hierarchy. The MOEA incorporates clinical criteria to restrict the search space through protocol objectives and then uses Pareto optimality among the fitness objectives to select individuals. Results: Acceleration techniques implemented on both levels of the hierarchical algorithm resulted in short, practical runtimes for optimizations. The MOEA improvements were evaluated for example prostate cases with one target and two OARs. The modified MOEA dominated 11.3% of plans using a standard genetic algorithm package. By implementing domination advantage and protocol objectives, small diverse populations of clinically acceptable plans that were only dominated 0.2% by the Pareto front could be generated in a fraction of an hour. Conclusions: Our MOEA produces a diverse Pareto optimal set of plans that meet all dosimetric protocol criteria in a feasible amount of time. It optimizes not only beamlet intensities but also objective function parameters on a patient-specific basis.

Facing stiff competition, the challenge at Mediplan was straightforward—innovate or die. Partnering with an industrial design team from a local university, Kirsty Smart, Mark Phillips, and Mike Press tell how the company staked its future on developing a new nurse-call hand unit. Ultimately, however, this story is not about product design, but rather about design as it can transform and revitalize a moribund corporate culture.

Tumor boundary delineation using positron emission tomography (PET) is a promising tool for radiation therapy applications. In this study we quantify the uncertainties in tumor boundary delineation as a function of the reconstruction method, smoothing, and lesion size in head and neck cancer patients using FDG-PET images and evaluate the dosimetric impact on radiotherapy plans. FDG-PET images were acquired for eight patients with a GE Advance PET scanner. In addition, a 20 cm diameter cylindrical phantom with six FDG-filled spheres with volumes of 1.2 to 26.5 cm{sup 3} was imaged. PET emission scans were reconstructed with the OSEM and FBP algorithms with different smoothing parameters. PET-based tumor regions were delineated using an automatic contouring function set at progressively higher threshold contour levels and the resulting volumes were calculated. CT-based tumor volumes were also contoured by a physician on coregistered PET/CT patient images. The intensity value of the threshold contour level that returns 100% of the actual volume, I{sub V100}, was measured. We generated intensity-modulated radiotherapy (IMRT) plans for an example head and neck patient, treating 66 Gy to CT-based gross disease and 54 Gy to nodal regions at risk, followed by a boost to the FDG-PET-based tumor. The volumes ofmore » PET-based tumors are a sensitive function of threshold contour level for all patients and phantom datasets. A 5% change in threshold contour level can translate into a 200% increase in volume. Phantom data indicate that I{sub V100} can be set as a fraction, f, of the maximum measured uptake. Fractional threshold values in the cylindrical water phantom range from 0.23 to 0.51. Both the fractional threshold and the threshold-volume curve are dependent on lesion size, with lesions smaller than approximately 5 cm{sup 3} displaying a more pronounced sensitivity and larger fractional threshold values. The threshold-volume curves and fractional threshold values also depend on the reconstruction algorithm and smoothing filter with more smoothing requiring a higher fractional threshold contour level. The threshold contour level affects the tumor size, and therefore the ultimate boost dose that is achievable with IMRT. In an example head and neck IMRT plan, the D95 of the planning target volume decreased from 7770 to 7230 cGy for 42% vs 55% contour threshold levels. PET-based tumor volumes are strongly affected by the choice of threshold level. This can have a significant dosimetric impact. The appropriate threshold level depends on lesion size and image reconstruction parameters. These effects should be carefully considered when using PET contour and/or volume information for radiotherapy applications.« less

Characterization measurements and calculations were performed on a new medical seed developed by IsoRay Inc. in Richland, Washington, that utilizes the short-lived isotope 131Cs. This is the first medical seed to utilize 131Cs, and this model has recently received FDA 510(k) clearance. The objective of this work was to characterize the dosimetric properties of the new seed according to the AAPM Task Group 43 recommendations. Cesium-131 is a low-energy x-ray emitter, with the most prominent peaks in the 29 keV to 34 keV region. The intended application is brachytherapy for treating cancers in prostate, breast, head and neck, lung, and pancreas. The evaluations performed included air-kerma strength, radial dose function, anisotropy in phantom, half-life, energy spectra in phantom and in air, and internal activity. The results indicate the CS-1 seeds have a dose-rate constant of 0.911 cGy hr-1 U-1 in water, dose penetration characteristics similar to 125I and 103Pd, anisotropy function values on the order of 0.71 at short distances and small angles, and an average anisotropy factor of 0.964. The overall dosimetric characteristics are similar to 125I and 103Pd seeds with the exception of half-life, which is 9.7 days, as compared to 17 days for 103Pd and 60 daysmore » for 125I. The shorter half-life may offer significant advantages in biological effectiveness.« less

Purpose: A standard tool for ensuring the quality of radiation therapy treatments is the initial physics plan review. However, little is known about its performance in practice. The goal of this study is to measure the effectiveness of physics plan review by introducing simulated errors into “mock” treatment plans and measuring the performance of plan review by physicists. Methods: We generated six mock treatment plans containing multiple errors. These errors were based on incident learning system data both within the department and internationally (SAFRON). These errors were scored for severity and frequency. Those with the highest scores were included in the simulations (13 errors total). Observer bias was minimized using a multiple co-correlated distractor approach. Eight physicists reviewed these plans for errors, with each physicist reviewing, on average, 3/6 plans. The confidence interval for the proportion of errors detected was computed using the Wilson score interval. Results: Simulated errors were detected in 65% of reviews [51–75%] (95% confidence interval [CI] in brackets). The following error scenarios had the highest detection rates: incorrect isocenter in DRRs/CBCT (91% [73–98%]) and a planned dose different from the prescribed dose (100% [61–100%]). Errors with low detection rates involved incorrect field parameters in record and verify system (38%, [18–61%]) and incorrect isocenter localization in planning system (29% [8–64%]). Though pre-treatment QA failure was reliably identified (100%), less than 20% of participants reported the error that caused the failure. Conclusion: This is one of the first quantitative studies of error detection. Although physics plan review is a key safety measure and can identify some errors with high fidelity, others errors are more challenging to detect. This data will guide future work on standardization and automation. Creating new checks or improving existing ones (i.e., via automation) will help in detecting those errors with low detection rates.

Irradiation of the cranio-spinal axis is often one of the treatment modalities of certain childhood cancers, e.g., medulloblastoma. In order to achieve a uniform dose to the spinal cord, missing tissue compensators are required. In the past, our practice was to fabricate compensators out of strips of lead. We report on the use of intensity modulated fields to achieve the desired compensation. Seven cases of pediatric cancer whose treatment involved irradiation of the cranio-spinal axis had compensators designed using a beam intensity modulation method rather than making mechanical compensators. The compensators only adjusted for missing tissue along the spinal axis. Comparisons between calculated and measured doses were made at depth in phantoms and on the surface of the patient. The intensity modulated fields were delivered using a step-and-shoot delivery on an Elekta SL20 accelerator equipped with multileaf collimator. The intensity-modulated compensators provided more flexibility in design than the physical compensator method. Finer intensity steps were achievable, more accurate dose distributions were able to be calculated, and adjustments during treatment, e.g., junction changes, were more easily implemented. Convolution/superposition dose calculations were within 3 of measurements. Intensity modulated fields are a practical and more efficient method of delivering uniform doses to the spine in pediatric cancer treatments. They provide many advantages over mechanical compensators with regard to time and flexibility. 2003 American College of Medical Physics.

Intensity modulated radiation therapy (IMRT) theoretically allows detailed tailoring of the dose distribution in tissue. The goal of this study was to determine if a method of dynamic IMRT could be used to deliver a high dose of radiation to a concave shaped target around the cervical spinal cord. Fifteen young adult dogs from our laboratory population were randomly divided into two groups. A radiation dose of 84 Gy in 4 Gy fractions was delivered with a conventional 4 field technique for Group A dogs, and with dynamic IMRT for Group B dogs to a "C-shaped" target close to the cervical spinal cord. Neurologic status, magnetic resonance imaging results and histopathologic changes were compared among dogs in the two groups. Group A dogs developed myelomalacia with a latency period of 65 +/- 9 days. Group B dogs did not have any histologic changes to the cervical spinal cord when euthanasia was performed 12 months after irradiation. The results demonstrate that this IMRT technique can be safely and precisely delivered to a patient in a clinical situation.

A comparison was made of different treatment plans to determine the effect on the three-dimensional dose distributions of varying the allowed parameters in linac-based stereotactic radiosurgery with circular collimators; these parameters are arc position, length, and weighting, and collimator diameter. For the class of eccentrically shaped target volumes that are not so irregular as to require several separate isocenters, it was found that superior dose distributions could be achieved by varying arc length, arc position, arc weighting, and collimator diameter. An analysis of the results achieved with an automated planning program indicates that, in general, the variables of arc position and arc length are of greater importance than collimator size or beam weighting. However, there are cases where varying these latter two parameters does result in markedly better dose distributions. A deeper investigation into the effects of multiple collimators on the dose distribution in the area of steepest gradient demonstrated that multiple collimator sizes do not significantly degrade the dose falloff, which is in fact mostly determined by the effects of intersecting arcs.

Purpose: Late toxicity from radiation to critical structures limits the possible dose in Radiation Therapy. Perfectly conformal treatment of a target is not realizable, so the clinician must accept a certain level of collateral radiation to nearby OARs. But how much? General guidelines exist for healthy tissue sparing which guide RT treatment planning, but are these guidelines good enough to create the optimal plan given the individualized patient anatomy? We propose a means to evaluate the planned dose level to an OAR using a multi-institutional data-store of previously treated patients, so a clinician might reconsider planning objectives. Methods: The tool is built on Oncospace, a federated data-store system, which consists of planning data import, web based analysis tools, and a database containing:1) DVHs: dose by percent volume delivered to each ROI for each patient previously treated and included in the database.2) Overlap Volume Histograms (OVHs): Anatomical measure defined as the percent volume of an ROI within a given distance to target structures.Clinicians know what OARs are important to spare. For any ROI, Oncospace knows for which patients’ anatomy that ROI was harder to plan in the past (the OVH is less). The planned dose should be close to the least dose of previous patients. The tool displays the dose those OARs were subjected to, and the clinician can make a determination about the planning objectives used.Multiple institutions contribute to the Oncospace Consortium, and their DVH and OVH data are combined and color coded in the output. Results: The Oncospace website provides a plan quality display tool which identifies harder to treat patients, and graphically displays the dose delivered to them for comparison with the proposed plan. Conclusion: The Oncospace Consortium manages a data-store of previously treated patients which can be used for quality checking new plans. Grant funding by Elekta.

Purpose: To investigate a spatiotemporally optimal radiotherapy prescription scheme and its potential benefit for glioblastoma (GBM) patients using the proliferation and invasion (PI) glioma model. Methods: Standard prescription for GBM was assumed to deliver 46Gy in 23 fractions to GTV1+2cm margin and additional 14Gy in 7 fractions to GTV2+2cm margin. We simulated the tumor proliferation and invasion in 2D according to the PI glioma model with a moving velocity of 0.029(slow-move), 0.079(average-move), and 0.13(fast-move) mm/day for GTV2 with a radius of 1 and 2cm. For each tumor, the margin around GTV1 and GTV2 was varied to 0–6 cm and 1–3 cm respectively. Total dose to GTV1 was constrained such that the equivalent uniform dose (EUD) to normal brain equals EUD with the standard prescription. A non-stationary dose policy, where the fractional dose varies, was investigated to estimate the temporal effect of the radiation dose. The efficacy of an optimal prescription scheme was evaluated by tumor cell-surviving fraction (SF), EUD, and the expected survival time. Results: Optimal prescription for the slow-move tumors was to use 3.0(small)-3.5(large) cm margins to GTV1, and 1.5cm margin to GTV2. For the average- and fast-move tumors, it was optimal to use 6.0cm margin for GTV1 suggesting that whole brain therapy is optimal, and then 1.5cm (average-move) and 1.5–3.0cm (fast-move, small-large) margins for GTV2. It was optimal to deliver the boost sequentially using a linearly decreasing fractional dose for all tumors. Optimal prescription led to 0.001–0.465% of the tumor SF resulted from using the standard prescription, and increased tumor EUD by 25.3–49.3% and the estimated survival time by 7.6–22.2 months. Conclusion: It is feasible to optimize a prescription scheme depending on the individual tumor characteristics. A personalized prescription scheme could potentially increase tumor EUD and the expected survival time significantly without increasing EUD to normal brain.

An experiment was performed at the Lawrence Berkeley Laboratory BEVALAC to measure the multiple Coulomb scattering of 650-MeV/A uranium nuclei in 0.19 radiation lengths of a Cu target. Differential distributions in the projected multiple scattering angle were measured in the vertical and horizontal planes using silicon position-sensitive detectors to determine particle trajectories before and after target scattering. The results were compared with the multiple Coulomb scattering theories of Fermi and Molière, and with a modification of the Fermi theory, using a Monte Carlo simulation. These theories were in excellent agreement with experiment at the 2 sigma level. The best quantitative agreement is obtained with the Gaussian distribution predicted by the modified Fermi theory.

Purpose: To compare plans for NSCLC optimized using Dynamic Adaptive Therapy (DAT) with conventional IMRT optimization. DAT adapts plans based on changes in the target volume by using dynamic programing techniques to consider expected changes into the optimization process. Information gathered during treatment, e.g. from CBCT, is incorporated into the optimization. Methods and materials: DAT is formulated using stochastic control formalism, which minimizes the total expected number of tumor cells at the end of a treatment course subject to uncertainty inherent in the tumor response and organs-at-risk (OAR) dose constraints. This formulation allows for non-stationary dose distribution as well as non-stationary fractional dose as needed to achieve a series of optimal plans that are conformal to tumor over time. Sixteen phantom cases with various sizes and locations of tumors, and OAR geometries were generated. Each case was planned with DAT and conventional IMRT (60Gy/30fx). Tumor volume change over time was obtained by using, daily MVCT-based, two-level cell population model. Monte Carlo simulations have been performed for each treatment course to account for uncertainty in tumor response. Same OAR dose constraints were applied for both methods. The frequency of plan modification was varied to 1, 2, 5 (weekly), and 29 (daily). The final average tumor dose and OAR doses have been compared to quantify the potential benefit of DAT. Results: The average tumor max, min, mean, and D95 resulted from DAT were 124.0–125.2%, 102.1–114.7%, 113.7–123.4%, and 102.0–115.9% (range dependent on the frequency of plan modification) of those from conventional IMRT. Cord max, esophagus max, lung mean, heart mean, and unspecified tissue D05 resulted from AT were 84–102.4%, 99.8–106.9%, 66.9–85.6%, 58.2–78.8%, and 85.2–94.0% of those from conventional IMRT. Conclusions: Significant tumor dose increase and OAR dose reduction, especially with parallel OAR with mean or dose-volume constraints, can be achieved using DAT.

Purpose: Investigate the feasibility of using spatio-temporal optimization, i.e., fractionation schedule optimization combined with IMRT to select the patient-specific fractionation schedule that best maximizes the tumor biologically equivalent dose (BED). Methods and materials: The optimal fractionation schedule for a patient depends on the tumor and organ at-risk (OAR) sensitivity to fraction size (i.e., α/β), the effective tumor doubling time, and the size and location of tumor target relative to one or more OAR. We used a novel spatio-temporal optimization method to identify the number of fractions N that maximizes the tumor BED using 3D BED distribution for sixteen lung phantom cases. The selection of the optimal fractionation schedule used equivalent 30-fraction OAR constraints for the heart (Dmean < 45 Gy), lungs (Dmean < 20 Gy), cord (Dmax < 45 Gy), esophagus (Dmax < 63 Gy), and unspecified tissues (D05 < 63 Gy). To asses plan quality, we considered the min, mean, max and D95 tumor BED and the equivalent uniform dose (EUD) for optimized plans to the corresponding BED values for a conventional IMRT plan (60 Gy in 30 fractions). A sensitivity analysis was performed to assess the effects on optimal fractionation schedule of the effective tumor doubling time (Td = 3–100 days) and lag-time for accelerated repopulation (Tk = 0 to 10 days) Results: For a tumor α/β = 10Gy, tumor max, min, mean, and D95 BED were about 20% larger than the BED for the conventional prescription. The EUD for the optimized fractionation schedule was up to 17% larger than conventional prescription. For fast proliferating tumor (Td < 10 days), there was no significant increase in tumor dose but the treatment course could be shortened, offering convenience for patients and efficient allocation of resources.

Purpose: Advancement in Radiation Oncology (RO) practice develops through evidence based medicine and clinical trial. Knowledge usable for treatment planning, decision support and research is contained in our clinical data, stored in an Oncospace database. This data store and the tools for populating and analyzing it are compatible with standard RO practice and are shared with collaborating institutions. The question is - what protocol for system development and data sharing within an Oncospace Consortium? We focus our example on the technology and data meaning necessary to share across the Consortium. Methods: Oncospace consists of a database schema, planning and outcome data import and web based analysis tools.1) Database: The Consortium implements a federated data store; each member collects and maintains its own data within an Oncospace schema. For privacy, PHI is contained within a single table, accessible to the database owner.2) Import: Spatial dose data from treatment plans (Pinnacle or DICOM) is imported via Oncolink. Treatment outcomes are imported from an OIS (MOSAIQ).3) Analysis: JHU has built a number of webpages to answer analysis questions. Oncospace data can also be analyzed via MATLAB or SAS queries.These materials are available to Consortium members, who contribute enhancements and improvements. Results: 1) The Oncospace Consortium now consists of RO centers at JHU, UVA, UW and the University of Toronto. These members have successfully installed and populated Oncospace databases with over 1000 patients collectively.2) Members contributing code and getting updates via SVN repository. Errors are reported and tracked via Redmine. Teleconferences include strategizing design and code reviews.3) Successfully remotely queried federated databases to combine multiple institutions’ DVH data for dose-toxicity analysis (see below – data combined from JHU and UW Oncospace). Conclusion: RO data sharing can and has been effected according to the Oncospace Consortium model: http://oncospace.radonc.jhmi.edu/. John Wong - SRA from Elekta; Todd McNutt - SRA from Elekta; Michael Bowers - funded by Elekta

Purpose: EPID dosimetry is an emerging method for treatment verification and QA. Given that the in-vivo EPID technique is in clinical use at some centers, we investigate the sensitivity and specificity for detecting different classes of errors. We assess the impact of these errors using dose volume histogram endpoints. Though data exist for EPID dosimetry performed pre-treatment, this is the first study quantifying its effectiveness when used during patient treatment (in-vivo). Methods: We analyzed 17 patients; EPID images of the exit dose were acquired and used to reconstruct the planar dose at isocenter. This dose was compared to the TPS dose using a 3%/3mm gamma criteria. To simulate errors, modifications were made to treatment plans using four possible classes of error: 1) patient misalignment, 2) changes in patient body habitus, 3) machine output changes and 4) MLC misalignments. Each error was applied with varying magnitudes. To assess the detectability of the error, the area under a ROC curve (AUC) was analyzed. The AUC was compared to changes in D99 of the PTV introduced by the simulated error. Results: For systematic changes in the MLC leaves, changes in the machine output and patient habitus, the AUC varied from 0.78–0.97 scaling with the magnitude of the error. The optimal gamma threshold as determined by the ROC curve varied between 84–92%. There was little diagnostic power in detecting random MLC leaf errors and patient shifts (AUC 0.52–0.74). Some errors with weak detectability had large changes in D99. Conclusion: These data demonstrate the ability of EPID-based in-vivo dosimetry in detecting variations in patient habitus and errors related to machine parameters such as systematic MLC misalignments and machine output changes. There was no correlation found between the detectability of the error using the gamma pass rate, ROC analysis and the impact on the dose volume histogram. Funded by grant R18HS022244 from AHRQ

Purpose: To develop a probabilistic model of radiotherapy plans using Bayesian networks that will detect potential errors in radiation delivery. Methods: Semi-structured interviews with medical physicists and other domain experts were employed to generate a set of layered nodes and arcs forming a Bayesian Network (BN) which encapsulates relevant radiotherapy concepts and their associated interdependencies. Concepts in the final network were limited to those whose parameters are represented in the institutional database at a level significant enough to develop mathematical distributions. The concept-relation knowledge base was constructed using the Web Ontology Language (OWL) and translated into Hugin Expert Bayes Network files via the the RHugin package in the R statistical programming language. A subset of de-identified data derived from a Mosaiq relational database representing 1937 unique prescription cases was processed and pre-screened for errors and then used by the Hugin implementation of the Estimation-Maximization (EM) algorithm for machine learning all parameter distributions. Individual networks were generated for each of several commonly treated anatomic regions identified by ICD-9 neoplasm categories including lung, brain, lymphoma, and female breast. Results: The resulting Bayesian networks represent a large part of the probabilistic knowledge inherent in treatment planning. By populating the networks entirely with data captured from a clinical oncology information management system over the course of several years of normal practice, we were able to create accurate probability tables with no additional time spent by experts or clinicians. These probabilistic descriptions of the treatment planning allow one to check if a treatment plan is within the normal scope of practice, given some initial set of clinical evidence and thereby detect for potential outliers to be flagged for further investigation. Conclusion: The networks developed here support the use of probabilistic models into clinical chart checking for improved detection of potential errors in RT plans.

Biomedical informatics encompasses a very large domain of knowledge and applications. This broad and loosely defined field can make it difficult to navigate. Physicists often are called upon to provide informatics services and/or to take part in projects involving principles of the field. The purpose of the presentations in this symposium is to help medical physicists gain some knowledge about the breadth of the field and how, in the current clinical and research environment, they can participate and contribute. Three talks have been designed to give an overview from the perspective of physicists and to provide a more in-depth discussion in two areas. One of the primary purposes, and the main subject of the first talk, is to help physicists achieve a perspective about the range of the topics and concepts that fall under the heading of "informatics". The approach is to de-mystify topics and jargon and to help physicists find resources in the field should they need them. The other talks explore two areas of biomedical informatics in more depth. The goal is to highlight two domains of intense current interest--databases and models--in enough depth into current approaches so that an adequate background for independent inquiry is achieved. These two areas will serve as good examples of how physicists, using informatics principles, can contribute to oncology practice and research. Learning Objectives: 1. To understand how the principles of biomedical informatics are used by medical physicists. 2. To put the relevant informatics concepts in perspective with regard to biomedicine in general. 3. To use clinical database design as an example of biomedical informatics. 4. To provide a solid background into the problems and issues of the design and use of data and databases in radiation oncology. 5. To use modeling in the service of decision support systems as an example of modeling methods and data use. 6. To provide a background into how uncertainty in our data and knowledge can be incorporated into modeling methods.

Purpose: To implement a common database structure and user-friendly web-browser based data collection tools across several medical institutions to better support evidence-based clinical decision making and comparative effectiveness research through shared outcomes data. Methods: A consortium of four academic medical centers agreed to implement a federated database, known as Oncospace. Initial implementation has addressed issues of differences between institutions in workflow and types and breadth of structured information captured. This requires coordination of data collection from departmental oncology information systems (OIS), treatment planning systems, and hospital electronic medical records in order to include as much as possible the multi-disciplinary clinical data associated with a patients care. Results: The original database schema was well-designed and required only minor changes to meet institution-specific data requirements. Mobile browser interfaces for data entry and review for both the OIS and the Oncospace database were tailored for the workflow of individual institutions. Federation of database queries--the ultimate goal of the project--was tested using artificial patient data. The tests serve as proof-of-principle that the system as a whole--from data collection and entry to providing responses to research queries of the federated database--was viable. The resolution of inter-institutional use of patient data for research is still notmore » completed. Conclusions: The migration from unstructured data mainly in the form of notes and documents to searchable, structured data is difficult. Making the transition requires cooperation of many groups within the department and can be greatly facilitated by using the structured data to improve clinical processes and workflow. The original database schema design is critical to providing enough flexibility for multi-institutional use to improve each institution s ability to study outcomes, determine best practices, and support research. The project has demonstrated the feasibility of deploying a federated database environment for research purposes to multiple institutions.« less

Multiattribute decision making begins with the premise that the solution to the problem at hand involves multiple goals, some of which are mutually antagonistic, and not all of which can be achieved in a single solution. Choosing a multiobjective optimization algorithm should be done in conjunction with the selection of a decision making approach. A vital component of the decision making method is to be explicit about the criteria, how they are to be evaluated and how they relate to the value structure of the decision maker. Much of the theory of multiattribute decision theory stems from economic theory. In particular, Pareto optimality was first formulated as the condition in which it is impossible to make one person better off without making someone else worse off. Another economic concept, utility, is used to quantify values on a relative scale. In problems not related to economics, it is usually the case that the criteria used to make a decision are difficult to formulate in terms than can be mathematically optimized. For example, in radiation therapy, surrogates for patient outcomes must be formulated in terms of physical dose deposition. Mathematical optimization algorithms can take several forms. Prior methods require the user to input some preferences before the optimization can proceed. Such methods can produce a single “optimal” solution. Interactive methods rely on feedback between the decision maker and the optimization algorithm. These methods can be the most efficient at searching but require considerable commitment from the decision maker. Posterior optimization methods produce a set of solutions that hopefully spans the trade‐offs of interest. The mathematical optimization is followed by a decision making process that selects one of them. Learning objectives: 1. To appreciate the relationship between Pareto optimality and decision making. 2. To understand the current state of decision making and how it relates to other approaches. 3. To understand how multiattribute decision making approaches in radiation therapy relate to methods used in other fields such as operations research. 4. To become acquainted with the basics of decision theory such as utility theory.

Purpose: To use a Markov Decision Process (MDP) to model the sequential treatment of recurrent brain metastases. The purpose of the model is to calculate the optimal set of treatment decisions given the state of the patient. Improved systemic cancer control has resulted in growing numbers of patients who undergo multiple treatments for brain metastases. Given this situation, it is important to determine the optimal treatment decision at any point in time given the condition of the patient and the possibility of future treatments. Methods: Sequential stochastic decisions can be optimized with respect to a given reward using dynamic programming methods. A model of patients with brain metastases was constructed using a definition of state, a set of actions (whole brain RT, surgery, stereotactic radiosurgery, waiting), transition probabilities between states dependent on a given action, and a reward based on quality of life. Transition probabilities were obtained from the literature and clinical data. The problem was solved for both an infinite horizon, resulting in a stationary policy, and a fixed horizon, resulting in optimal decisions for each decision epoch. Results: The model was developed and validated by comparison with clinical judgment. Selection of adequate state variables and suitable rewards was made by comparing optimal policies with standard clinical situations. After the development phase, the model was used to study the role that possible subsequent treatments have on optimal decisions. The effects of knowledge regarding expected lifespan on optimal policies was investigated. In addition, comparisons of stationary policies with finite horizon policies were made. Conclusion: A first attempt at applying MDPs to radiation therapy decision making focussed on improving treatment strategies for metastatic braincancer. An important application of this model will be to determine the extent to which improvements in our knowledge of a patient's prognosis will alter recommended therapies.

Purpose: To quantify the uncertainties in tumor boundary delineation in head and neck and lungcancer patients using FDG‐PET and FMISO‐PET images and evaluate the dosimetric impact on radiotherapy plans. Method and Materials: We contoured FDG‐PET and FMISO‐PET based tumor volumes using images from the GE Advance PET scanner. We use the autocontour function of the ADAC/Pinnacle radiation planning system to delineate tumor boundaries at successively higher PET signal levels. CT‐based tumor volumes are also contoured by a physician on coregistered PET/CT images. We generated intensity‐modulated radiotherapy(IMRT) plans for head and neck patients treating 66 Gy to CT‐based gross disease and 54 Gy to nodal regions at risk, followed by a boost to the PET‐based tumor.Results: The volumes of PET‐based tumors are a sensitive function of threshold intensity level for all patients. For FDG‐based volumes, a 10% decrease in threshold translates into an approximately 200% increase in volume. Lesions smaller than approximately 8 cm3 display a more pronounced threshold‐volume dependence. The threshold‐volume dependence in FMISO scans is significantly more sensitive than in FDG scans. Lungcancer patients show a similar trend to head and neck patients with a possible overall shift in sensitivity. IMRT planning results on head and neck patients show that the boost dose limit of FDG‐based volumes depends on the threshold level chosen for contouring. In one patient the D95 of the planning target volume decreased from 7770 cGy to 7230 cGy when the contour level changed from 42% to 55%. Conclusion: PET‐based tumor volumes are strongly affected by the choice of threshold level which has a direct dosimetric impact. Further validation and refinement of delineation methods, including a determination of the proper threshold level, should reduce PET‐related delineation uncertainties for radiotherapy applications.

Purpose: To generate adaptive, biologically optimized, patient‐specific IMRT plans with the potential to reduce normal tissue complications and increase treatment efficacy in the treatment of glioblastoma. Methods: A proliferation‐invasion radiation therapy (PIRT) mathematical model of glioblastoma characterizes patient‐specific tumor evolution and response to radiotherapy. An iterative dialog between the PIRT model and a multi‐objective evolutionary algorithm (MOEA) for IMRT plan generation results in adaptive, patient‐specific plans that can be optimized to clinical goals subject to defined restrictions. We performed simulations in a simplified geometry utilizing both the standard‐of‐care and optimized plans for a cohort of 11 patients exhibiting a wide range of tumor growth kinetics and compared the results. Results: The spatially non‐uniform, patient‐specific optimized plans reduced equivalent uniform dose (EUD) to healthy brain tissue (39 – 82%) and increased therapeutic ratio (the ratio of tumor EUD to normal tissue EUD) (50 – 265%). The model‐driven virtual evaluation of cancer treatment response (VECTR) score, a metric of treatment impact on survival, increased for all but one patient (8 – 181%). Both the normal tissue EUD and therapeutic ratio were linearly correlated with patient‐specific PIRT model parameters, indicating increased benefits or patients with more diffuse tumors. These results were robust to uncertainty in measured tumor radius of ±.5 mm and a 20% variation in the linear quadratic radiobiology parameter α/β. Conclusion: This analysis suggests that we can improve upon the standard‐of‐care radiation therapy with adaptive, individualized plans generated with a patient‐specific mathematical model of glioblastoma in combination with a MOEA for IMRT optimization. This work demonstrates a possible improvement of patient outcomes and lays the groundwork for further 3D anatomically accurate simulations that further optimize treatment and spare eloquent brain.

Purpose:Medical physicists are often asked to evaluate or choose appropriatetechnology for clinical applications. These are multidimensionalproblems that also suffer from different degrees of uncertainty in thevariables. Probabilistic decision models are a robust andmathematically correct means of handling these issues. The principlesof constructing such models are presented along with practicalexamples in the areas of IGRT,IMRT and proton therapy.Methods: Influence diagrams are used to model the variables and theiruncertainties and include action and reward variables. Influencediagrams are directed acyclic graphs that use Bayesian probabilitycalculus to propagate probabilities and to update prior probabilitiesin the presence of evidence. An influence diagram was used to modelthe question of whether braintumors are better treated with x‐rayIMRT or proton therapy, with or without CT‐guided localization. Datafor the conditional probabilities of the model were obtained from theliterature and included models of TCP, NTCP and induction of secondmalignancies, as well as data on the probability density functions forinterfraction patient motion. Dosimetric data were obtained using theCMS treatment planning system. Results: Several different tumor types and sites were studied. The critical variables in the model were identified andstudied using analyses of evidence, parameters and value ofinformation. The impact of imaging was significant, regardless of theradiation type. The models used in determining some of theconditional probabilities parameters also played an important role inranking alternatives. Conclusions: Although such choices are difficult, physicists mustproceed with the best data at hand. Without a rigorous framework onwhich to build a model of the process, decisions are likely to be based onunstated assumptions and incorrect inference. The example ofcomparing irradiation modalities for braintumors shows the power ofinfluence diagrams in this critical context.

Purpose: To evaluate how more flexibility in the search space affects performance in multiobjective IMRT optimization. Method: Conventional inverse planning combines multiple objectives using a weighted sum. Typical structure‐specific objectives are characterized by a specific reference dose. Optimization with variable weights explores a region of feasible plan space and yields the Pareto front for that region. Fixed references doses and structure‐specific objectives can limit the region explored. Effects of varying weights, reference doses and segmentation of structures were explored. A multiobjective evolutionary algorithm (MOEA) was used to optimize plans for an example prostate case. Both the objective weights and reference doses were optimized in the MOEA and results were compared to optimizing only one parameter. In a second experiment, each voxel throughout structures were allowed different weights and reference doses. Plans were evaluated using decision objectives (OAR mean dose and target variance) which were distinct from the optimized parameters. Results: By optimizing both weighting and reference doses that define the penalty function for each structure, a set of IMRT plans was found that was superior to plans generated by only optimizing weights or by only using optimizing reference dose. Furthermore, when the search space was expanded so that every voxel in every structure was allowed its own reference dose and weight, plans were found that were superior to the best set of plans generated using structure‐specific parameters. Conclusion: Algorithms that don't allow all input parameters to be optimized or consider a voxel‐specific search space will not approach the best possible set of plans given the physical limitations of IMRT treatment when judged by clinically relevant and possibly non‐convex metrics. The set of plans generated using a limited search space can potentially be improved in all decision criteria by increasing the scope of the search space.

Purpose: To determine if a photon convolution/superposition algorithm could be used to model a fast neutron therapy beam in a commercial treatment planning system. Methods: The beam to be modeled was the Clinical Neutron Therapy System (CNTS) fast neutron beam produced by 50 MeV protons on a Be target at the University of Washington(UW) Medical Center. The dose calculation model was that implemented in Pinnacle, 3 (Philips Medical Systems). Measured neutron dose data were acquired with an IC30 ion chamber with tissue equivalent gas. The neutron beam was modeled using the auto‐modeling tools available in the Pinnacle system for photon beams. Doses were then computed using a 100 MU beam incident on a water equivalent phantom for open and wedged square fields as well as MLC shaped irregular fields. Pinnacle generated profiles and central axis dose points were compared to two sets of doses: 1) doses computed with the UW PRISMtreatment planning system in the same geometry as the Pinnacle setup and 2) doses measured in a water tank. Results: The Pinnacle photonmodel incorporates most of the important dosimetric features of the neutron beam. Computed doses compared well to both the Prism TPS and measured data. We found that calculated dose points among open and wedged square fields were within 2% of both Prism and measured doses along the central axis, and within 5% of measurement in the penumbra region. Dose point calculations using irregular treatment type fields were within 3% of measured dose points. Conclusion: The Pinnacle TPS has sufficient computational modeling ability to adequately produce a viable neutronmodel for potential use in clinical treatment planning. Further testing of irregular fields must be performed and results analyzed prior to implementation in the clinic.

Purpose: A patient‐specific method to optimize IMRTtreatment of incurable disease by maximizing tumorcell death and minimizing dose to normal tissue is proposed. Methods: A multiobjective evolutionary algorithm was used to optimize IMRT dose distributions on an example patient throughout treatment using a published patient‐specific 4D mathematical model of glioblastoma proliferation and invasion [Rockne 2010] to calculate the distribution of diffusely invaded tumorcells, hypoxia, and radiosensitivity and recalculate distributions after each week of treatment. Each optimized IMRT plan was designed to maximize EUD for tumorcells while holding constant the EUD for normal tissue. Dose distributions were scaled such that the EUD delivered to the normal tissue for each fraction was equal to the EUD delivered from one fraction using the current clinical protocol of 1.8 Gy to the frank tumor with a 2.5 cm margin. Results: The mathematical model predicted that the total volume of tumor was reduced by 41.0% after one week and 72.5% after two weeks of treatment using the proposed protocol. The current clinical protocol was shown to reduce volume by only 5.5% after one week and 11.2% after two weeks for the same EUD delivered to normal tissue. After 7 weeks of treatment using the clinical protocol and 250% higher normal tissue EUD, the tumor burden was only reduced by 40.4%. Conclusions: A patient‐specific mathematical model showed that a 4D treatment plan that maximizes tumorcell killing while sparing normal tissue reduces the tumor volume more in one week than in seven weeks using current clinical protocol for the same EUD delivered to normal tissue. This approach has the potential to optimize and tailor IMRTtreatment planning to individual patient's disease through an iterative dialog between a mathematical model for disease and response to therapy with objective‐based treatment optimization.

Purpose: To evaluate the clinical effectiveness of a multiobjective evolutionary algorithm (MOEA) for IMRT plan generation.Methods: A MOEA was developed that generates a set of IMRT plans that approximates the clinical Pareto front in a time comparable to current commercial inverse planning systems. The stochastic nature of the algorithm permits the use of any objective function regardless of convexity. Selected plans generated by the MOEA were compared with those generated clinically using a commercial planning system (Pinnacle 8.0, Philips Electronics, N.V.). Cases of head & neck cancer and prostate cancer were planned for IMRT on both systems. MOEA plans were evaluated by comparing their performance in meeting and exceeding objectives used in generating plans using conventional methods. Characteristics of the MOEA‐generated solutions were compared using a range of commonly used objective functions. Results: Results are classified into three groups: (1) establishing the ability of the MOEA to approximate the clinical Pareto front and optimizing speed, (2) evaluating objective functions, and(3) comparing MOEA plans with current clinical IMRT methods. The effect of modifications and different objective functions on the algorithm were judged by assessing the fraction of plans generated with one algorithm that Pareto dominated those of another. Results of Aim (3) showed that plans selected from the MOEA performed better than the commercial algorithm. Conclusion:IMRT planning is inherently multiobjective and treatment planning decisions should be made using multiobjective systems. We describe an evolutionary algorithm that provides a set of plans that consistently contain multiple plans superior to those achieved using a conventional optimization algorithm and meets clinical requirements for speed and performance. Using mean dose objective for OARs and range objective for targets demonstrated better performance than using EUD. This work was supported by NIH RO1 CA112505.

Introduction: We propose a mathematical framework to optimize the beamlet intensity spatially and temporally that is adaptive to a patient's radiobiological response to radiotherapy as observed by advanced functional imaging or biomarkers that are soon to be available or as predicted by mathematical model. Methods and materials: Stochastic control approach was used to model and solve for optimal dose per voxel for any specific time period, i.e. fraction. Tumor cell density and BED for OAR were used to represent the system state, and beamlet intensity was the control variable. The final tumor cell density at the end of the treatment course was minimized by choosing the optimal sequence of beamlet intensities based on the observed or predicted patient's radiobiological response to radiotherapy. BED for OAR was constrained not to exceed its tolerance level and alpha/beta parameters for the tumor were varied stochastically. Results: For the special case of our stochastic control formulation where the stochasticity is ignored and additional information of the patient's response to radiotherapy is unknown, the constant dose per voxel for any given period is found to be optimal, which agrees with the current practice. Simulations revealed that there was almost 70% decrease in tumor cell density at the end of the treatment when the beamlet intensity was optimally adaptive to patient's response to radiotherapy compared to a constant dose per voxel as in current clinical practice. Conclusion: Our approach demonstrated the ability to simultaneously balance spatial and temporal aspects of the optimization problem in treatment planning from a biological perspective and adapt to the uncertainty in biological response to radiation over the treatment course. This is a first step in designing the individualized treatment plan that is adaptive biologically conformai radiotherapy and that exploits recent advances in functional imaging and/or mathematical modeling of tumorradiation response.

Purpose: To develop a probabilistic model of outcomes of radiation therapy which includes both dosimetric and non‐dosimetric predictors, and includes a decision‐making component to quantify the balance between disease cure and radiation‐induced side‐effects. This model was implemented to assess IMRTtreatment plans for individual patients for head and neck cancer.Materials and Methods: Physicians have available many resources that may not be easily reconcilable to predict patient outcomes. Dosimetric indicators, such as the EUD and NTCP are probabilistic in nature, without explicit representation of the underlying biology. Clinical trials focus on patient and disease characteristics, such as disease location, T‐stage, nodal involvement, Karnofsky performance status, and often include one treatment variable, such as DVH‐cutpoints or chemotherapy regimes. Newly recognized factors, such as HPV positivity, may affect outcome, however, without definitive clinical data, integrating such factors into clinical decision‐making is not straightforward. Finally, experience‐driven beliefs affect treatment choices and may vary between physicians. We combine all of the aforementioned resources using a Bayesian network in order to make an outcome prediction for each IMRT plan. Outcome predictions highlight the stark trade‐off between preventing recurrent disease that generally has a fatal prognosis and preventing radiation‐induced side effects that range from xerostemia to blindness to paralysis. We use a MarkovModel to compute a quality‐adjusted life expectancy using patient preferences for health states. Results: Probabilities of local and distant control matched published values well, as did life expectancies. The trade‐offs between quality of life and quantity of life are explored. Sensitivity analysis highlighted physician beliefs that affected treatment choices. Conclusions:Modeling of radiation therapy has grown progressively more sophisticated. We present a method by which probabilities and expected values of clinically relevant outcomes, based on a range of variables, are calculated.