William T. Watkins

Clinical adequacy assessment of autocontours for prostate IMRT with meaningful endpoints

Purpose To determine if radiation treatment plans created based on autosegmented (AS) regions‐of‐interest (ROI)s are clinically equivalent to plans created based on manually segmented ROIs, where equivalence is evaluated using probabilistic dosimetric metrics and probabilistic biological endpoints for prostate IMRT. Method and materials Manually drawn contours and autosegmented ROIs were created for 167 CT image sets acquired from 19 prostate patients. Autosegmentation was performed utilizing Pinnacles Smart Probabilistic Image Contouring Engine. For each CT set, 78 Gy/39 fraction 7‐beam IMRT treatment plans with 1 cm CTV‐to‐PTV margins were created for each of the three contour scenarios; PMD using manually delineated (MD) ROIs, PAS using autosegmented ROIs, and PAM using autosegmented organ‐at‐risks (OAR)s and the manually drawn target. For each plan, 1000 virtual treatment simulations with different systematic errors for each simulation and a different random error for each fraction were performed. The statistical probability of achieving dose–volume metrics (coverage probability (CP)), expectation values for normal tissue complication probability (NTCP), and tumor control probability (TCP) metrics for all possible cross‐evaluation pairs of ROI types and planning scenarios were reported. In evaluation scenarios, the root mean square loss (RMSL) and maximum absolute loss (MAL) of coverage probability of dose–volume objectives, E[TCP], and E[NTCP] were compared with respect to the base plan created and evaluated with manually drawn contours. Results Femoral head dose objectives were satisfied in all situations, as well as the maximum dose objectives for all ROIs. Bladder metrics were within the clinical coverage tolerances except D35Gy for the autosegmented plan evaluated with the manual contours. Dosimetric indices for CTV and rectum could be highly compromised when the definition of the ROIs switched from manually delineated to autosegmented. Seventy‐two percent of CT image sets satisfied the worst‐case CP thresholds for all dosimetric objectives in all scenarios, the percentage dropped to 50% if biological indices were taken into account. Among evaluation scenarios, (MD,PAM) bore the highest resemblance to (MD,PMD) where 99% and 88% of cases met all CP thresholds for bladder and rectum, respectively. Conclusions When including daily setup variations in prostate IMRT, the dose–volume metric CP, and biological indices of ROIs were approximately equivalent for the plans created based on manually drawn targets and autosegmented OARs in 88% of cases. The accuracy of autosegmented prostates and rectums are impediment to attain statistically equivalent plans created based on manually drawn ROIs.

Detection of dose delivery variations on TomoTherapy using on-board detector based verification

A clinical case of delivery dose deviations on a TomoTherapy treatment was discovered during a patient specific treatment quality assurance (QA) verification. An in-house developed QA system, MCLogQA, for TomoTherapy has been implemented in our clinic for patient specific treatment QA. The MCLogQA system utilizes the log file and detector-based multileaf collimator (MLC) leaf opening time (LOT) to assess accuracy of treatment plan delivery. Recently, the MCLogQA system discovered  >10% dose deviation for a low dose/fraction treatment plan. To verify the adequacy of the MCLogQA result, a delivery quality assurance (DQA) plan was created and performed. The treatment plan was also transferred to a second TomoTherapy unit and planning system to investigate if the plan-delivery deviation was unit dependent. Further testing was carried out in phantom plans. MCLogQA showed MLC LOT was on average 2.4% higher than the planned LOT, resulting in 3.5% increase in mean dose, and 14% increase in dose to 1 cc volume of max dose in PTV. Independent DQA verification confirmed the MCLogQA result. For the transferred treatment plan delivery, the MCLogQA also showed an average increase of 6.6% in MLC LOT, resulting in increases in mean dose by 9.3% and dose to 1 cc volume of max dose in PTV by 16%. The inaccurate MLC LOT was a result of a poor latency model at very small LOT. Phantom testing confirmed low LOT will result in relatively large dosimetric variation, and detector-based MCLogQA will detect differences in planned and measured LOT. Accuracy in TomoTherapy treatment delivery can be susceptible to LOT uncertainty. Using MCLogQA for QA verification not only validates the treatment delivery, but also provides information on LOT variation and comprehensive dose distribution. This information can help decision making when large plan-delivery deviation occurs.

Quality assurance tool for organ at risk delineation in radiation therapy using a parametric statistical approach

PURPOSE To develop a quality assurance (QA) tool that identifies inaccurate organ at risk (OAR) delineations. METHODS The QA tool computed volumetric features from prior OAR delineation data from 73 thoracic patients to construct a reference database. All volumetric features of the OAR delineation are computed in three-dimensional space. Volumetric features of a new OAR are compared with respect to those in the reference database to discern delineation outliers. A multicriteria outlier detection system warns users of specific delineation outliers based on combinations of deviant features. Fifteen independent experimental sets including automatic, propagated, and clinically approved manual delineation sets were used for verification. The verification OARs included manipulations to mimic common errors. Three experts reviewed the experimental sets to identify and classify errors, first without; and then 1 week after with the QA tool. RESULTS In the cohort of manual delineations with manual manipulations, the QA tool detected 94% of the mimicked errors. Overall, it detected 37% of the minor and 85% of the major errors. The QA tool improved reviewer error detection sensitivity from 61% to 68% for minor errors (P = 0.17), and from 78% to 87% for major errors (P = 0.02). CONCLUSIONS The QA tool assists users to detect potential delineation errors. QA tool integration into clinical procedures may reduce the frequency of inaccurate OAR delineation, and potentially improve safety and quality of radiation treatment planning.

Dose to mass for evaluation and optimization of lung cancer radiation therapy

PURPOSE To evaluate potential organ at risk dose-sparing by using dose-mass-histogram (DMH) objective functions compared with dose-volume-histogram (DVH) objective functions. METHODS Treatment plans were retrospectively optimized for 10 locally advanced non-small cell lung cancer patients based on DVH and DMH objectives. DMH-objectives were the same as DVH objectives, but with mass replacing volume. Plans were normalized to dose to 95% of the PTV volume (PTV-D95v) or mass (PTV-D95m). For a given optimized dose, DVH and DMH were intercompared to ascertain dose-to-volume vs. dose-to-mass differences. Additionally, the optimized doses were intercompared using DVH and DMH metrics to ascertain differences in optimized plans. Mean dose to volume, Dv‾, mean dose to mass, DM‾, and fluence maps were intercompared. RESULTS For a given dose distribution, DVH and DMH differ by >5% in heterogeneous structures. In homogeneous structures including heart and spinal cord, DVH and DMH are nearly equivalent. At fixed PTV-D95v, DMH-optimization did not significantly reduce dose to OARs but reduced PTV-Dv‾ by 0.20±0.2Gy (p=0.02) and PTV-DM‾ by 0.23±0.3Gy (p=0.02). Plans normalized to PTV-D95m also result in minor PTV dose reductions and esophageal dose sparing (Dv‾ reduced 0.45±0.5Gy, p=0.02 and DM‾ reduced 0.44±0.5Gy, p=0.02) compared to DVH-optimized plans. Optimized fluence map comparisons indicate that DMH optimization reduces dose in the periphery of lung PTVs. CONCLUSIONS DVH- and DMH-dose indices differ by >5% in lung and lung target volumes for fixed dose distributions, but optimizing DMH did not reduce dose to OARs. The primary difference observed in DVH- and DMH-optimized plans were variations in fluence to the periphery of lung target PTVs, where low density lung surrounds tumor.

Clinical outcomes of helical conformal versus nonconformal palliative radiation therapy for axial skeletal metastases

PURPOSE Palliative radiation therapy (RT) for bone metastases has traditionally been delivered with conventional, nonconformal RT (NCRT). Conformal RT (CRT) is potentially more complex and expensive than NCRT, but may reduce normal tissue dose and subsequently toxicity. In this retrospective analysis, we compared CRT with NCRT to investigate the association between conformality and toxicity. METHODS AND MATERIALS A retrospective analysis of patients receiving palliative RT for axial skeletal bone metastases from 2012 to 2014 was conducted. Patient and treatment characteristics were obtained including dosimetric variables, acute toxicity, and subjective pain during treatment and in the acute posttreatment period (≤60 days after completion). Statistical analyses included t tests, χ2 tests, and multivariate logistic regression. RESULTS A total of 179 patients and 254 bone metastases were identified (142 CRT, 112 NCRT). The CRT and NCRT groups were well matched for baseline characteristics (number of fractions, field size, treatment sites, and concurrent chemotherapy). In multivariate logistic regression models, technique (CRT vs NCRT) was not associated with development of acute toxicity. Regarding toxicity, Eastern Cooperative Oncology Group performance status and total dose were significantly associated with a higher rate of acute toxicity during RT (odds ratios, 0.649 and 1.129 and P = .027 and .044, respectively), and only a higher number of vertebral bodies in the treatment field was significantly associated with acute toxicity post-treatment (odds ratios, 1.219, P = .028). CRT was associated with improvement in bone pain during and posttreatment (P = .049 and .045, respectively). CONCLUSIONS Our results demonstrate no difference in acute toxicity following palliative RT with CRT compared with NCRT for painful bone metastases; however, treatment volume did predict for increased toxicity. Larger studies may further elucidate the value of CRT including the impact of dose escalation for bone metastases and differences in patient reported outcomes between RT techniques.

WE-AB-209-12: Quasi Constrained Multi-Criteria Optimization for Automated Radiation Therapy Treatment Planning

PURPOSE To introduce quasi-constrained Multi-Criteria Optimization (qcMCO) for unsupervised radiation therapy optimization which generates alternative patient-specific plans emphasizing dosimetric tradeoffs and conformance to clinical constraints for multiple delivery techniques. METHODS For N Organs At Risk (OARs) and M delivery techniques, qcMCO generates M(N+1) alternative treatment plans per patient. Objective weight variations for OARs and targets are used to generate alternative qcMCO plans. For 30 locally advanced lung cancer patients, qcMCO plans were generated for dosimetric tradeoffs to four OARs: each lung, heart, and esophagus (N=4) and 4 delivery techniques (simple 4-field arrangements, 9-field coplanar IMRT, 27-field non-coplanar IMRT, and non-coplanar Arc IMRT). Quasi-constrained objectives included target prescription isodose to 95% (PTV-D95), maximum PTV dose (PTV-Dmax)< 110% of prescription, and spinal cord Dmax<45 Gy. The algorithms ability to meet these constraints while simultaneously revealing dosimetric tradeoffs was investigated. Statistically significant dosimetric tradeoffs were defined such that the coefficient of determination between dosimetric indices which varied by at least 5 Gy between different plans was >0.8. RESULTS The qcMCO plans varied mean dose by >5 Gy to ipsilateral lung for 24/30 patients, contralateral lung for 29/30 patients, esophagus for 29/30 patients, and heart for 19/30 patients. In the 600 plans computed without human interaction, average PTV-D95=67.4±3.3 Gy, PTV-Dmax=79.2±5.3 Gy, and spinal cord Dmax was >45 Gy in 93 plans (>50 Gy in 2/600 plans). Statistically significant dosimetric tradeoffs were evident in 19/30 plans, including multiple tradeoffs of at least 5 Gy between multiple OARs in 7/30 cases. The most common statistically significant tradeoff was increasing PTV-Dmax to reduce OAR dose (15/30 patients). CONCLUSION The qcMCO method can conform to quasi-constrained objectives while revealing significant variations in OAR doses including mean dose reductions >5 Gy. Clinical implementation will facilitate patient-specific decision making based on achievable dosimetry as opposed to accept/reject models based on population derived objectives.

SU-G-BRC-15: The Potential Clinical Significance of Dose Mapping Error for Intra- Fraction Dose Mapping for Lung Cancer Patients

PURPOSE To evaluate the dose-mapping error (DME) inherent to conventional dose-mapping algorithms as a function of dose-matrix resolution. METHODS As DME has been reported to be greatest where dose-gradients overlap tissue-density gradients, non-clinical 66 Gy IMRT plans were generated for 11 lung patients with the target edge defined as the maximum 3D density gradient on the 0% (end of inhale) breathing phase. Post-optimization, Beams were copied to 9 breathing phases. Monte Carlo dose computed (with 2*2*2 mm3 resolution) on all 10 breathing phases was deformably mapped to phase 0% using the Monte Carlo energy-transfer method with congruent mass-mapping (EMCM); an externally implemented tri-linear interpolation method with voxel sub-division; Pinnacles internal (tri-linear) method; and a post-processing energy-mass voxel-warping method (dTransform). All methods used the same base displacement-vector-field (or its pseudo-inverse as appropriate) for the dose mapping. Mapping was also performed at 4*4*4 mm3 by merging adjacent dose voxels. RESULTS Using EMCM as the reference standard, no clinically significant (>1 Gy) DMEs were found for the mean lung dose (MLD), lung V20Gy, or esophagus dose-volume indices, although MLD and V20Gy were statistically different (2*2*2 mm3 ). Pinnacle-to-EMCM target D98% DMEs of 4.4 and 1.2 Gy were observed (2*2*2 mm3 ). However dTransform, which like EMCM conserves integral dose, had DME >1 Gy for one case. The root mean square RMS of the DME for the tri-linear-to- EMCM methods was lower for the smaller voxel volume for the tumor 4D-D98%, lung V20Gy, and cord D1%. CONCLUSION When tissue gradients overlap with dose gradients, organs-at-risk DME was statistically significant but not clinically significant. Target-D98%-DME was deemed clinically significant for 2/11 patients (2*2*2 mm3 ). Since tri-linear RMS-DME between EMCM and tri-linear was reduced at 2*2*2 mm3 , use of this resolution is recommended for dose mapping. Interpolative dose methods are sufficiently accurate for the majority of cases. J.V. Siebers receives funding support from Varian Medical Systems.

SU-C-BRB-05: Determining the Adequacy of Auto-Contouring Via Probabilistic Assessment of Ensuing Treatment Plan Metrics in Comparison with Manual Contours

PURPOSE To determine if auto-contour and manual-contour-based plans differ when evaluated with respect to probabilistic coverage metrics and biological model endpoints for prostate IMRT. METHODS Manual and auto-contours were created for 149 CT image sets acquired from 16 unique prostate patients. A single physician manually contoured all images. Auto-contouring was completed utilizing Pinnacles Smart Probabilistic Image Contouring Engine (SPICE). For each CT, three different 78 Gy/39 fraction 7-beam IMRT plans are created; PD with drawn ROIs, PAS with auto-contoured ROIs, and PM with auto-contoured OARs with the manually drawn target. For each plan, 1000 virtual treatment simulations with different sampled systematic errors for each simulation and a different sampled random error for each fraction were performed using our in-house GPU-accelerated robustness analyzer tool which reports the statistical probability of achieving dose-volume metrics, NTCP, TCP, and the probability of achieving the optimization criteria for both auto-contoured (AS) and manually drawn (D) ROIs. Metrics are reported for all possible cross-evaluation pairs of ROI types (AS,D) and planning scenarios (PD,PAS,PM). Bhattacharyya coefficient (BC) is calculated to measure the PDF similarities for the dose-volume metric, NTCP, TCP, and objectives with respect to the manually drawn contour evaluated on base plan (D-PD). RESULTS We observe high BC values (BC≥0.94) for all OAR objectives. BC values of max dose objective on CTV also signify high resemblance (BC≥0.93) between the distributions. On the other hand, BC values for CTVs D95 and Dmin objectives are small for AS-PM, AS-PD. NTCP distributions are similar across all evaluation pairs, while TCP distributions of AS-PM, AS-PD sustain variations up to %6 compared to other evaluated pairs. CONCLUSION No significant probabilistic differences are observed in the metrics when auto-contoured OARs are used. The prostate auto-contour needs improvement to achieve clinically equivalent plans.

SU-E-T-472: Improvement of IMRT QA Passing Rate by Correcting Angular Dependence of MatriXX

Purpose: Multi-channel planar detector arrays utilized for IMRT-QA, such as the MatriXX, exhibit an incident-beam angular dependent response which can Result in false-positive gamma-based QA results, especially for helical tomotherapy plans which encompass the full range of beam angles. Although MatriXX can use with gantry angle sensor to provide automatically angular correction, this sensor does not work with tomotherapy. The purpose of the study is to reduce IMRT-QA false-positives by correcting for the MatriXX angular dependence. Methods: MatriXX angular dependence was characterized by comparing multiple fixed-angle irradiation measurements with corresponding TPS computed doses. For 81 Tomo-helical IMRT-QA measurements, two different correction schemes were tested: (1) A Monte-Carlo dose engine was used to compute MatriXX signal based on the angular-response curve. The computed signal was then compared with measurement. (2) Uncorrected computed signal was compared with measurements uniformly scaled to account for the average angular dependence. Three scaling factor (+2%, +2.5%, +3%) were tested. Results: The MatriXX response is 8% less than predicted for a PA beam even when the couch is fully accounted for. Without angular correction, only 67% of the cases pass the >90% points γ<1 (3%, 3mm). After full angular correction, 96% of the cases pass the criteria. Of three scaling factors, +2% gave the highest passing rate (89%), which is still less than the full angular correction method. With a stricter γ(2%,3mm) criteria, the full angular correction method was still able to achieve the 90% passing rate while the scaling method only gives 53% passing rate. Conclusion: Correction for the MatriXX angular dependence reduced the false-positives rate of our IMRT-QA process. It is necessary to correct for the angular dependence to achieve the IMRT passing criteria specified in TG129.

TH‐AB‐304‐08: Computational Modeling of Risk of Radiation Pneumonitis

Purpose: Current normal tissue complication probability models estimate incidence of radiation pneumonitis (RP) in the treatment of lung cancer based on single parameters, e.g. mean lung dose or V20, and therefore do not capture specific variations contained in dose volume histograms (DVHs). We therefore developed an RP-risk model which utilizes the entire DVH reasoning that such a model would be more robust for predicating radiation-induced lung disease (RILD). Methods: A logistic regression model was developed to predict RP-risk on each point of the cumulative DVH utilizing >1200 patient outcomes summarized in the Quantitative Analysis of Normal Tissue Effects in the Clinic (QUANTEC) lung study. Stepwise regression including linear, quadratic, and mixed terms in dose and volume was applied to the single dose-volume parameter-versus-outcome data. The model computes a relative importance at each dose-volume point and these values were linearly combined to estimate an overall RP-risk. A set of 80 DVHs were used to validate the algorithm prediction by comparison with an independent predictor of RP-risk using mean lung dose. Results: The logistic regression coefficients were all statistically significant (p 70%) of the model variability occurs at doses <30 Gy for DVH levels ranging from 30%-100%. The model is most sensitive to DVH variations in the V5-V30 range. Conclusion: The model’s predictive power is comparable to estimates from QUANTEC which use mean dose as input. Since the model conveys the relative importance of each dose-volume level it has the potential to better predict RILD than currently available models. In order to determine if our hypothesis is correct we will compare the model to established methods including complete DVH data and clinical outcomes. This abstract was supported by the George Amorino Pilot Grants in Radiation Oncology from the Department of Radiation Oncology, Univeristy of Virginia.