Search

Purpose Reducing margins during treatment planning to decrease dose to healthy organs surrounding the prostate can risk inadequate treatment of subclinical disease. This study aimed to investigate whether lack of dose to subclinical disease is associated with increased disease progression by using high-quality prostate radiation therapy clinical trial data to identify anatomically localized regions where dose variation is associated with prostate-specific antigen progression (PSAP). Methods and Materials Planned dose distributions for 683 patients of the Trans-Tasman Radiation Oncology Group 03.04 Randomized Androgen Deprivation and Radiotherapy (RADAR) trial were deformably registered onto a single exemplar computed tomography data set. These were divided into high-risk and intermediate-risk subgroups for analysis. Three independent voxel-based statistical tests, using permutation testing, Cox regression modeling, and least absolute shrinkage selection operator feature selection, were applied to identify regions where dose variation was associated with PSAP. Results from the intermediate-risk RADAR subgroup were externally validated by registering dose distributions from the RT01 (n = 388) and Conventional or Hypofractionated High Dose Intensity Modulated Radiotherapy for Prostate Cancer Trial (CHHiP) (n = 253) trials onto the same exemplar and repeating the tests on each of these data sets. Results Voxel-based Cox regression revealed regions where reduced dose was correlated with increased prostate-specific androgen progression. Reduced dose in regions associated with coverage at the posterior prostate, in the immediate periphery of the posterior prostate, and in regions corresponding to the posterior oblique beams or posterior lateral beam boundary, was associated with increased PSAP for RADAR and RT01 patients, but not for CHHiP patients. Reduced dose to the seminal vesicle region was also associated with increased PSAP for RADAR intermediate-risk patients. Conclusions Ensuring adequate dose coverage at the posterior prostate and immediately surrounding posterior region (including the seminal vesicles), where aggressive cancer spread may be occurring, may improve tumor control. It is recommended that particular care be taken when defining margins at the prostate posterior, acknowledging the trade-off between quality of life due to rectal dose and the preferences of clinicians and patients.

Hypofractionation of prostate cancer radiotherapy achieves tumour control at lower total radiation doses, however, increased rectal and bladder toxicities have been observed. To realise the radiobiological advantage of hypofractionation whilst minimising harm, the potential reduction in dose to organs at risk was investigated for biofocused radiotherapy. Patient-specific tumour location and cell density information were derived from multiparametric imaging. Uniform-dose plans and biologically-optimised plans were generated for a standard schedule (78 Gy/39 fractions) and hypofractionated schedules (60 Gy/20 fractions and 36.25 Gy/5 fractions). Results showed that biologically-optimised plans yielded statistically lower doses to the rectum and bladder compared to isoeffective uniform-dose plans for all fractionation schedules. A reduction in the number of fractions increased the target dose modulation required to achieve equal tumour control. On average, biologically-optimised, moderately-hypofractionated plans demonstrated 15.3% (p-value

Introduction This study quantified inter-observer contouring variations for multiple male pelvic structures, many of which are of emerging relevance for prostate cancer radiotherapy progression and toxicity response studies. Methods Five prostate cancer patient datasets (CT and T2-weighted MR) were distributed to 13 observers for contouring. CT structures contoured included the clinical target volume (CTV), seminal vesicles, rectum, colon, bowel bag, bladder and peri-rectal space (PRS). MR contours included CTV, trigone, membranous urethra, penile bulb, neurovascular bundle and multiple pelvic floor muscles. Contouring variations were assessed using the intraclass correlation coefficient (ICC), Dice similarity coefficient (DSC), and multiple additional metrics. Results Clinical target volume (CT and MR), bladder, rectum and PRS contours showed excellent inter-observer agreement (median ICC = 0.97; 0.99; 1.00; 0.95; 0.90, DSC = 0.83 ± 0.05; 0.88 ± 0.05; 0.93 ± 0.03; 0.81 ± 0.07; 0.80 ± 0.06, respectively). Seminal vesicle contours were more variable (ICC = 0.75, DSC = 0.73 ± 0.14), while colon and bowel bag contoured volumes were consistent (ICC = 0.97; 0.97), but displayed poor overlap (DSC = 0.58 ± 0.22; 0.67 ± 0.21). Smaller MR structures showed significant inter-observer variations, with poor overlap for trigone, membranous urethra, penile bulb, and left and right neurovascular bundles (DSC = 0.44 ± 0.22; 0.41 ± 0.21; 0.66 ± 0.21; 0.16 ± 0.17; 0.15 ± 0.15). Pelvic floor muscles recorded moderate to strong inter-observer agreement (ICC = 0.50-0.97), although large outlier variations were observed. Conclusions Inter-observer contouring variation was significant for multiple pelvic structures contoured on MR.

Background and purpose This study aimed to identify anatomically-localised regions where planned radiotherapy dose is associated with gastrointestinal toxicities in healthy tissues throughout the pelvic anatomy. Materials and methods Planned dose distributions for up to 657 patients of the Trans Tasman Radiation Oncology Group 03.04 RADAR trial were deformably registered onto a single exemplar computed tomography dataset. Voxel-based multiple comparison permutation dose difference testing, Cox regression modelling and LASSO feature selection were used to identify regions where dose-increase was associated with grade ≥2 rectal bleeding (RB) or tenesmus, according to the LENT/SOMA scale. This was externally validated by registering dose distributions from the RT01 (n = 388) and CHHiP (n = 241) trials onto the same exemplar and repeating the tests on each of these data sets, and on all three datasets combined. Results Voxel-based Cox regression and permutation dose difference testing revealed regions where increased dose was correlated with gastrointestinal toxicity. Grade ≥2 RB was associated with posteriorly extended lateral beams that manifested high doses (>55 Gy) in a small rectal volume adjacent to the clinical target volume. A correlation was found between grade ≥2 tenesmus and increased low-intermediate dose (∼25 Gy) at the posterior beam region, including the posterior rectum and perirectal fat space (PRFS). Conclusions The serial response of the rectum with respect to RB has been demonstrated in patients with posteriorly extended lateral beams. Similarly, the parallel response of the PRFS with respect to tenesmus has been demonstrated in patients treated with the posterior beam.

Linear scaling is used to convert raw computed tomography (CT) pixel values into Hounsfield units corresponding to different tissue values. Analysis of a benchmarking study, presented here, where the same CT scan was imported into and then exported from multiple radiotherapy treatment planning systems, found inconsistencies in HU scaling parameter values exported along with the images, particularly when images were exported using the Radiation Therapy Oncology Group format. Several methods of estimating conversion parameters, based on estimating pixel values corresponding to air and water within the image, for pelvic CT images from a large multi-centre trial were compared against original Digital Imaging and Communications in Medicine export parameters. In general using the mean of a sample region at the centroid of the bladder to estimate the value of water was more accurate than using the minimum or maximum or a single value at the centroid. Accuracy of methods of air estimation tested were dependent in part on features of the CT scanners and treatment planning systems, making it difficult to pick one method as superior that was independent of scanner and treatment planning system type. Based on the above analysis, methods for estimating air and water were selected for use in performing linear scaling of a set of pelvic CT images prior to their use in an interpatient image registration application. The selected methods were validated against a more recent and homogeneous dataset. Estimation error was found to be much lower within the validation set.

BACKGROUND AND PURPOSE To evaluate the impact of treatment planning and delivery factors on treatment outcome as measured by post-treatment disease progression. MATERIALS AND METHODS Accruing 813 external beam radiotherapy participants during 2003-2007, the RADAR trial collected a comprehensive range of clinical treatment factor data for each participant. Both the Fine and Gray competing risks modelling and the Kaplan-Meier (KM) analysis were undertaken to determine the impact of these factors on local-composite progression (LCP), with 709 participants available for analysis. RESULTS Participants with treatments involving 7 or more beams experienced significantly higher incidence of LCP, with a sub-hazard ratio (relative to 3-beam participants) of 3.056 (CI: 1.446-6.458, p

Traditionally rectal symptoms following pelvic/prostate radiotherapy are correlated to the dosimetry of the anorectum or a substructure of this. It has been suggested that the perirectal fat space (PRS) surrounding the rectum may also be relevant. This study considers the delineation and dosimetry of the PRS related to both rectal bleeding and control-related toxicity. Initially, a caseâcontrol cohort of 100 patients from the RADAR study were chosen based on presence/absence of rectal control-related toxicity. Automated contouring was developed to delineate the PRS. 79 of the 100 auto-segmentations were considered successful. Balanced caseâcontrol cohorts were defined from these cases. Atlas of Complication Incidence (ACI) were generated to relate the DVH of the PRS with specific rectal symptoms; rectal bleeding and control-related symptoms (LENT/SOM). ACI demonstrated that control-related symptoms were related to the dose distribution to the PRS which was confirmed with Wilcoxon rank sum test (pâ¯

Aims This study aimed to develop a framework for optimising prostate intensity-modulated radiotherapy (IMRT) based on patient-specific tumour biology, derived from multiparametric MRI (mpMRI). The framework included a probabilistic treatment planning technique in the effort to yield dose distributions with an improved expected treatment outcome compared with uniform-dose planning approaches. Methods IMRT plans were generated for five prostate cancer patients using two inverse planning methods: uniform-dose to the planning target volume and probabilistic biological optimisation for clinical target volume tumour control probability (TCP) maximisation. Patient-specific tumour location and clonogen density information were derived from mpMRI and geometric uncertainties were incorporated in the TCP calculation. Potential reduction in dose to sensitive structures was assessed by comparing dose metrics of uniform-dose plans with biologically-optimised plans of an equivalent level of expected tumour control. Results The planning study demonstrated biological optimisation has the potential to reduce expected normal tissue toxicity without sacrificing local control by shaping the dose distribution to the spatial distribution of tumour characteristics. On average, biologically-optimised plans achieved 38.6% (p-value: p-value: Conclusions It was concluded that varying the dose distribution within the prostate to take account for each patient’s clonogen distribution was feasible. Lower doses to normal structures compared to uniform-dose plans was possible whilst providing robust plans against geometric uncertainties. Further validation in a larger cohort is warranted along with considerations for adaptive therapy and limiting urethral dose.

Aim To investigate late gastrointestinal toxicity in a large pooled population of prostate cancer patients treated with radical radiotherapy. Normal tissue complication probability models were developed for late stool frequency and late rectal pain. Methods and materials Population included 1336 patients, 3-year minimum follow-up, treated with 66–80 Gy. Toxicity was scored with LENT-SOMA-scale. Two toxicity endpoints were considered: grade ⩾2 rectal pain and mean grade (average score during follow-up) in stool frequency >1. DVHs of anorectum were reduced to equivalent uniform dose (EUD). The best-value of the volume parameter n was determined through numerical optimization. Association between EUD/clinical factors and the endpoints was investigated by logistic analyses. Likelihood, Brier-score and calibration were used to evaluate models. External calibration was also carried out. Results 4% of patients (45/1122) reported mean stool frequency grade >1; grade ⩾2 rectal pain was present in the TROG 03.04 RADAR population only (21/677, 3.1%): for this endpoint, the analysis was limited to this population. Analysis of DVHs highlighted the importance of mid-range doses (30–50 Gy) for both endpoints. EUDs calculated with n = 1 (OR = 1.04) and n = 0.35 (OR = 1.06) were the most suitable dosimetric descriptors for stool frequency and rectal pain respectively. The final models included EUD and cardiovascular diseases (OR = 1.78) for stool frequency and EUD and presence of acute gastrointestinal toxicity (OR = 4.2) for rectal pain. Conclusion Best predictors of stool frequency and rectal pain are consistent with findings previously reported for late faecal incontinence, indicating an important role in optimization of mid-range dose region to minimize these symptoms highly impacting the quality-of-life of long surviving patients.

Background and purpose Most predictive models are not sufficiently validated for prospective use. We performed independent external validation of published predictive models for urinary dysfunctions following radiotherapy of the prostate. Materials/methods Multivariable models developed to predict atomised and generalised urinary symptoms, both acute and late, were considered for validation using a dataset representing 754 participants from the TROG 03.04-RADAR trial. Endpoints and features were harmonised to match the predictive models. The overall performance, calibration and discrimination were assessed. Results 14 models from four publications were validated. The discrimination of the predictive models in an independent external validation cohort, measured using the area under the receiver operating characteristic (ROC) curve, ranged from 0.473 to 0.695, generally lower than in internal validation. 4 models had ROC >0.6. Shrinkage was required for all predictive models' coefficients ranging from −0.309 (prediction probability was inverse to observed proportion) to 0.823. Predictive models which include baseline symptoms as a feature produced the highest discrimination. Two models produced a predicted probability of 0 and 1 for all patients. Conclusions Predictive models vary in performance and transferability illustrating the need for improvements in model development and reporting. Several models showed reasonable potential but efforts should be increased to improve performance. Baseline symptoms should always be considered as potential features for predictive models.

Purpose: Given the paucity of available data concerning radiotherapy-induced urinary toxicity, it is important to ensure derivation of the most robust models with superior predictive performance. This work explores multiple statistical-learning strategies for prediction of urinary symptoms following external beam radiotherapy of the prostate. Methods: The performance of logistic regression, elastic-net, support-vector machine, random forest, neural network, and multivariate adaptive regression splines (MARS) to predict urinary symptoms was analyzed using data from 754 participants accrued by TROG03.04-RADAR. Predictive features included dose-surface data, comorbidities, and medication-intake. Four symptoms were analyzed: dysuria, haematuria, incontinence, and frequency, each with three definitions (grade ≥ 1, grade ≥ 2 and longitudinal) with event rate between 2.3% and 76.1%. Repeated cross-validations producing matched models were implemented. A synthetic minority oversampling technique was utilized in endpoints with rare events. Parameter optimization was performed on the training data. Area under the receiver operating characteristic curve (AUROC) was used to compare performance using sample size to detect differences of ≥0.05 at the 95% confidence level. Results: Logistic regression, elastic-net, random forest, MARS, and support-vector machine were the highest-performing statistical-learning strategies in 3, 3, 3, 2, and 1 endpoints, respectively. Logistic regression, MARS, elastic-net, random forest, neural network, and support-vector machine were the best, or were not significantly worse than the best, in 7, 7, 5, 5, 3, and 1 endpoints. The best-performing statistical model was for dysuria grade ≥ 1 with AUROC ± standard deviation of 0.649 ± 0.074 using MARS. For longitudinal frequency and dysuria grade ≥ 1, all strategies produced AUROC>0.6 while all haematuria endpoints and longitudinal incontinence models produced AUROC

PURPOSE To demonstrate selection of a small representative subset of images from a pool of images comprising a potential atlas (PA) pelvic CT set to be used for autosegmentation of a separate target image set. The aim is to balance the need for the atlas set to represent anatomical diversity with the need to minimize resources required to create a high quality atlas set (such as multiobserver delineation), while retaining access to additional information available for the PA image set. METHODS Preprocessing was performed for image standardization, followed by image registration. Clustering was used to select the subset that provided the best coverage of a target dataset as measured by postregistration image intensity similarities. Tests for clustering robustness were performed including repeated clustering runs using different starting seeds and clustering repeatedly using 90% of the target dataset chosen randomly. Comparisons of coverage of a target set (comprising 711 pelvic CT images) were made for atlas sets of five images (chosen from a PA set of 39 pelvic CT and MR images) (a) at random (averaged over 50 random atlas selections), (b) based solely on image similarities within the PA set (representing prospective atlas development), (c) based on similarities within the PA set and between the PA and target dataset (representing retrospective atlas development). Comparisons were also made to coverage provided by the entire PA set of 39 images. RESULTS Exemplar selection was highly robust with exemplar selection results being unaffected by choice of starting seed with very occasional change to one of the exemplar choices when the target set was reduced. Coverage of the target set, as measured by best normalized cross-correlation similarity of target images to any exemplar image, provided by five well-selected atlas images (mean = 0.6497) was more similar to coverage provided by the entire PA set (mean = 0.6658) than randomly chosen atlas subsets (mean = 0.5977). This was true both of the mean values and the shape of the distributions. Retrospective selection of atlases (mean = 0.6497) provided a very small improvement over prospective atlas selection (mean = 0.6431). All differences were significant (P

Purpose To use a high-quality multicenter trial dataset to determine dose-volume effects for gastrointestinal (GI) toxicity following radiation therapy for prostate carcinoma. Influential dose-volume histogram regions were to be determined as functions of dose, anatomical location, toxicity, and clinical endpoint. Methods and Materials Planning datasets for 754 participants in the TROG 03.04 RADAR trial were available, with Late Effects of Normal Tissues (LENT) Subjective, Objective, Management, and Analytic (SOMA) toxicity assessment to a median of 72 months. A rank sum method was used to define dose-volume cut-points as near-continuous functions of dose to 3 GI anatomical regions, together with a comprehensive assessment of significance. Univariate and multivariate ordinal regression was used to assess the importance of cut-points at each dose. Results Dose ranges providing significant cut-points tended to be consistent with those showing significant univariate regression odds-ratios (representing the probability of a unitary increase in toxicity grade per percent relative volume). Ranges of significant cut-points for rectal bleeding validated previously published results. Separation of the lower GI anatomy into complete anorectum, rectum, and anal canal showed the impact of mid-low doses to the anal canal on urgency and tenesmus, completeness of evacuation and stool frequency, and mid-high doses to the anorectum on bleeding and stool frequency. Derived multivariate models emphasized the importance of the high-dose region of the anorectum and rectum for rectal bleeding and mid- to low-dose regions for diarrhea and urgency and tenesmus, and low-to-mid doses to the anal canal for stool frequency, diarrhea, evacuation, and bleeding. Conclusions Results confirm anatomical dependence of specific GI toxicities. They provide an atlas summarizing dose-histogram effects and derived constraints as functions of anatomical region, dose, toxicity, and endpoint for informing future radiation therapy planning.

Many similarity metrics exist for inter-observer contouring variation studies, however no correlation between metric choice and prostate cancer radiotherapy dosimetry has been explored. These correlations were investigated in this study. Two separate trials were undertaken, the first a thirty-five patient cohort with three observers, the second a five patient dataset with ten observers. Clinical and planning target volumes (CTV and PTV), rectum, and bladder were independently contoured by all observers in each trial. Structures were contoured on T2-weighted MRI and transferred onto CT following rigid registration for treatment planning in the first trial. Structures were contoured directly on CT in the second trial. STAPLE and majority voting volumes were generated as reference gold standard volumes for each structure for the two trials respectively. VMAT treatment plans (78 Gy to PTV) were simulated for observer and gold standard volumes, and dosimetry assessed using multiple radiobiological metrics. Correlations between contouring similarity metrics and dosimetry were calculated using Spearman's rank correlation coefficient. No correlations were observed between contouring similarity metrics and dosimetry for CTV within either trial. Volume similarity correlated most strongly with radiobiological metrics for PTV in both trials, including TCPPoisson (ρ = 0.57, 0.65), TCPLogit (ρ = 0.39, 0.62), and EUD (ρ = 0.43, 0.61) for each respective trial. Rectum and bladder metric correlations displayed no consistency for the two trials. PTV volume similarity was found to significantly correlate with rectum normal tissue complication probability (ρ = 0.33, 0.48). Minimal to no correlations with dosimetry were observed for overlap or boundary contouring metrics. Future inter-observer contouring variation studies for prostate cancer should incorporate volume similarity to provide additional insights into dosimetry during analysis.

INTRODUCTION Quality assurance methods are incorporated into multicentre radiotherapy clinical trials for ensuring consistent application of trial protocol and quantifying treatment uncertainties. The study's purpose was to determine whether post-treatment disease progression is associated with measures of the quality of radiotherapy treatment. METHODS The TROG 03.04 RADAR trial tested the impact of androgen deprivation on prostate cancer patients receiving dose-escalated external beam radiation therapy. The trial incorporated a plan-review process and Level III dosimetric intercomparison at each centre, from which variables suggestive of treatment quality were collected. Kaplan-Meier statistics and Fine and Gray competing risk modelling were employed to test for associations between quality-related variables and the participant outcome local composite progression. RESULTS Increased 'dose-difference' at the prostatic apex and at the anterior rectal wall, between planned and measured dose, was associated with reduced progression. Participants whose treatment plans included clinical target volume (CTV) to planning target volume (PTV) margins exceeding protocol requirements also experienced reduced progression. Other quality-related variables, including total accrual from participating centres, measures of target coverage and other variations from protocol, were not significantly associated with progression. CONCLUSIONS This analysis has revealed the association of several treatment quality factors with disease progression. Increased dose and dose margin coverage in the prostate region can reduce disease progression. Extensive and rigorous monitoring has helped to maximise treatment quality, reducing the incidence of quality-indicator outliers, and thus reduce the chance of observing significant associations with progression rates.

Perirectal space segmentation in computed tomography images aids in quantifying radiation dose received by healthy tissues and toxicity during the course of radiation therapy treatment of the prostate. Radiation dose normalised by tissue volume facilitates predicting outcomes or possible harmful side effects of radiation therapy treatment. Manual segmentation of the perirectal space is time consuming and challenging in the presence of inter-patient anatomical variability and may suffer from inter- and intra-observer variabilities. However automatic or semi-automatic segmentation of the perirectal space in CT images is a challenging task due to inter patient anatomical variability, contrast variability and imaging artifacts. In the model presented here, a volume of interest is obtained in a multi-atlas based segmentation approach. Un-supervised learning in the volume of interest with a Gaussian-mixture-modeling based clustering approach is adopted to achieve a soft segmentation of the perirectal space. Probabilities from soft clustering are further refined by rigid registration of the multi-atlas mask in a probabilistic domain. A maximum a posteriori approach is adopted to achieve a binary segmentation from the refined probabilities. A mean volume similarity value of 97% and a mean surface difference of 3.06 ± 0.51 mm is achieved in a leave-one-patient-out validation framework with a subset of a clinical trial dataset. Qualitative results show a good approximation of the perirectal space volume compared to the ground truth.

Given the paucity of available data concerning radiotherapy-induced urinary toxicity, it is important to ensure derivation of the most robust models with superior predictive performance. This work explores multiple statistical-learning strategies for prediction of urinary symptoms following external beam radiotherapy of the prostate.The performance of logistic regression, elastic-net, support-vector machine, random forest, neural network, and multivariate adaptive regression splines (MARS) to predict urinary symptoms was analyzed using data from 754 participants accrued by TROG03.04-RADAR. Predictive features included dose-surface data, comorbidities, and medication-intake. Four symptoms were analyzed: dysuria, haematuria, incontinence, and frequency, each with three definitions (grade ≥ 1, grade ≥ 2 and longitudinal) with event rate between 2.3% and 76.1%. Repeated cross-validations producing matched models were implemented. A synthetic minority oversampling technique was utilized in endpoints with rare events. Parameter optimization was performed on the training data. Area under the receiver operating characteristic curve (AUROC) was used to compare performance using sample size to detect differences of ≥0.05 at the 95% confidence level.Logistic regression, elastic-net, random forest, MARS, and support-vector machine were the highest-performing statistical-learning strategies in 3, 3, 3, 2, and 1 endpoints, respectively. Logistic regression, MARS, elastic-net, random forest, neural network, and support-vector machine were the best, or were not significantly worse than the best, in 7, 7, 5, 5, 3, and 1 endpoints. The best-performing statistical model was for dysuria grade ≥ 1 with AUROC ± standard deviation of 0.649 ± 0.074 using MARS. For longitudinal frequency and dysuria grade ≥ 1, all strategies produced AUROC0.6 while all haematuria endpoints and longitudinal incontinence models produced AUROC0.6.Logistic regression and MARS were most likely to be the best-performing strategy for the prediction of urinary symptoms with elastic-net and random forest producing competitive results. The predictive power of the models was modest and endpoint-dependent. New features, including spatial dose maps, may be necessary to achieve better models.

Purpose We assessed the association of the spatial distribution of dose to the bladder surface, described using dose-surface maps, with the risk of urinary dysfunction. Methods and Materials The bladder dose-surface maps of 754 participants from the TROG 03.04-RADAR trial were generated from the volumetric data by virtually cutting the bladder at the sagittal slice, intersecting the bladder center-of-mass through to the bladder posterior and projecting the dose information on a 2-dimensional plane. Pixelwise dose comparisons were performed between patients with and without symptoms (dysuria, hematuria, incontinence, and an International Prostate Symptom Score increase of ≥10 [ΔIPSS10]). The results with and without permutation-based multiple-comparison adjustments are reported. The pixelwise multivariate analysis findings (peak-event model for dysuria, hematuria, and ΔIPSS10; event-count model for incontinence), with adjustments for clinical factors, are also reported. Results The associations of the spatially specific dose measures to urinary dysfunction were dependent on the presence of specific symptoms. The doses received by the anteroinferior and, to lesser extent, posterosuperior surface of the bladder had the strongest relationship with the incidence of dysuria, hematuria, and ΔIPSS10, both with and without adjustment for clinical factors. For the doses to the posteroinferior region corresponding to the area of the trigone, the only symptom with significance was incontinence. Conclusions A spatially variable response of the bladder surface to the dose was found for symptoms of urinary dysfunction. Limiting the dose extending anteriorly might help reduce the risk of urinary dysfunction.

Purpose To develop a modelling procedure for late rectal bleeding (LRB) that goes beyond maximum likelihood model fitting of data. Thus, to create a metamodel for grade 2–3 (G23) and grade 3 (G3) LRB starting from literature evidence and validate it on a large population. Methods Metamodelling allows to highlight specific features of the model itself. Characteristics that predictive models for LRB have in common are: the global harmony of radiobiological parameters (volume effect parameter, n, steepness of the curve, k, and dose parameter associated to 50% of complication probability, D50) and specific clinical factors that frequently occur. Models including rectal Equivalent Uniform Dose (EUD) with/without patient-related dose-modifying factors (DMF) were retrieved by literature search. Dosimetric coefficients were resolved by weighted mean of published values, using their standard deviation as weight. Identified clinical features, expressed by DMF, were differently weighted taking into account the prevalence of the features and the size of the study. Finally, both factors were inserted in a modified logit-EUD model. Metamodel was validated on a pooled population (3DCRT/IMRT) of three international cohorts. Performance was assessed through calibration. Results Literature search identified rectal EUD, previous abdominal surgery, hormone therapy and use of cardiovascular drugs as relevant features: associated coefficients are presented in Table 1 . Validation cohort included 1591 pts with 240 (15%) LRBG23 and 98 (6.2%) LRBG3 pts. The calibration (Fig. 1) showed the following results: concerning LRBG23 the slope was equal to 0.22 (R2 = 0.38); the corresponding values for LRBG3 were 1.12 and 0.97. Conclusions A metamodel for prediction of LRB was derived from literature. LRBG3 model was successfully validated on a large population proving to be a valuable tool for predicting toxicity before RT. The model for LRBG23 predicted very well toxicity rate below 25% (which involved 87% of the population) while partially failing at higher probabilities.

61 Background: Predictive models for late rectal bleeding (LRB) after radiotherapy (RT) for prostate cancer were established by several studies, with good performance on development population (POP). Nonetheless, they were found to be unsatisfactory in their generalization to independent validation cohorts. Aim of the study is to build a metamodel for grade 2-3 (G23) and grade 3 (G3) LRB (LENT/SOMA) starting from literature evidence and to validate it on a new POP. Methods: The metamodel was constituted by clinical/dosimetric features and by their coefficients. Available predictive models for LRB were selected by literature search. Models including rectal Equivalent Uniform Dose (EUD) with/without patient-related dose-modifying factors were retrieved. The process allowed identification of “evidence-based” clinical/dosimetric features associated to LRB. Coeff. for identified features were resolved by weighted mean of published values, using their standard deviation as weight and then inserted in a sigmoid-shaped logit-EUD model. The resulting metamodel was validated on a pooled POP of radically treated pts (3DCRT & IMRT). Performance was assessed through calibration. Results: The search identified rectal EUD, previous abdominal surgery (SURG), hormone therapy (HT) and use of cardiovascular drugs (CARDIO) as relevant features (coeff. presented in table). Validation POP included 1591 pts with 240 (15%) LRBG23 and 98 (6.2%) LRBG3 pts. Calibration was very good: LRBG23 slope = 1.4 with R2 = 0.86; LRBG3 slope = 1.1 with R2 = 0.96. Conclusions: A metamodel for prediction of LRB has been developed from literature, including all currently available information on association between LRB and clinical/dosimetric factors. The model was successfully validated on a large POP, proving to be a valuable tool for predicting LRB before RT. [Table: see text]

Background To assess the impact of incremental modifications of treatment planning and delivery technique, as well as patient anatomical factors, on late gastrointestinal toxicity using data from the TROG 03.04 RADAR prostate radiotherapy trial. Methods The RADAR trial accrued 813 external beam radiotherapy participants during 2003–2008 from 23 centres. Following review and archive to a query-able database, digital treatment plans and data describing treatment technique for 754 patients were available for analysis. Treatment demographics, together with anatomical features, were assessed using uni- and multivariate regression models against late gastrointestinal toxicity at 18-, 36- and 54-month follow-up. Regression analyses were reviewed in the context of dose-volume data for the rectum and anal canal. Results A multivariate analysis at 36-month follow-up shows that patients planned using a more rigorous dose calculation algorithm (DCA) was associated with a lower risk of stool frequency (OR: 0.435, CI: 0.242–0.783, corrected p = 0.04). Patients using laxative as a method of bowel preparation had higher risk of having increased stool frequency compared to patients with no dietary intervention (OR: 3.639, CI: 1.502–8.818, corrected p = 0.04). Despite higher risks of toxicities, the anorectum, anal canal and rectum dose-volume histograms (DVH) indicate patients using laxative had unremarkably different planned dose distributions. Patients planned with a more rigorous DCA had lower median DVH values between EQD23 = 15 Gy and EQD23 = 35 Gy. Planning target volume (PTV), conformity index, rectal width and prescription dose were not significant when adjusted for false discovery rate. Number of beams, beam energy, treatment beam definition, positioning orientation, rectum-PTV separation, rectal length and mean cross sectional area did not affect the risk of toxicities. Conclusions The RADAR study dataset has allowed an assessment of technical modifications on gastrointestinal toxicity. A number of interesting associations were subsequently found and some factors, previously hypothesised to influence toxicity, did not demonstrate any significant impact. We recommend trial registries be encouraged to record technical modifications introduced during the trial in order for more powerful evidence to be gathered regarding the impact of the interventions. Electronic supplementary material The online version of this article (doi:10.1186/s13014-014-0282-7) contains supplementary material, which is available to authorized users.

Dose constraints based on histograms provide a convenient and widely-used method for informing and guiding radiotherapy treatment planning. Methods of derivation of such constraints are often poorly described. Two non-parametric methods for derivation of constraints are described and investigated in the context of determination of dose-specific cut-points-values of the free parameter (e.g., percentage volume of the irradiated organ) which best reflect resulting changes in complication incidence. A method based on receiver operating characteristic (ROC) analysis and one based on a maximally-selected standardized rank sum are described and compared using rectal toxicity data from a prostate radiotherapy trial. Multiple test corrections are applied using a free step-down resampling algorithm, which accounts for the large number of tests undertaken to search for optimal cut-points and the inherent correlation between dose-histogram points. Both methods provide consistent significant cut-point values, with the rank sum method displaying some sensitivity to the underlying data. The ROC method is simple to implement and can utilize a complication atlas, though an advantage of the rank sum method is the ability to incorporate all complication grades without the need for grade dichotomization.

Multicentre radiotherapy clinical trials can incorporate quality assurance (QA) procedures for ensuring consistent application of the trial protocol in the planning, delivery and reporting of participant treatments. Subsequently detected variations from trial protocol have previously been shown to reduce treatment efficacy, although little has been shown for toxicity rates. The purpose of this study was to investigate the association of QA measures and protocol variations on toxicity incidence in the context of a prostate radiotherapy trial.Using QA records from the TROG 03.04 RADAR trial, the impact of variations on gastrointestinal (GI) and genito-urinary (GU) toxicities was investigated.Protocol variation rates were lower than reported in previous studies, and showed little correlation with GI toxicity outcomes. Variations classified as 'major' showed a non-significant trend for increased toxicity relative to those classified as 'minor'. Results from a Level III phantom-based dosimetry study showed some correlation with GI toxicity, whereas ranking on a set-up accuracy study did not impact on toxicity. Toxicity in general increased with the number of participants accrued per centre, at odds with previous reports relating to disease progression, with a potential link to increases in low-mid-range rectal doses in the cohort from higher-accruing centres. No QA-related variables correlated with GU toxicities.Besides non-significant trends, minimal association was observed between QA variables and toxicity rates for the RADAR trial. The intention of the trial's QA programme was to reduce treatment variations and minimise toxicity in the context of a relevantly advanced treatment approach.

Accurate perirectal fat segmentation in CT images aids in estimating radiation dose delivered to the region of fat around the rectum during radiation therapy treatment of prostate cancer. Such a process is important in determining the resulting toxicity of the neighboring tissues. However automatic or semi-automatic segmentation of the perirectal fat in CT images is a challenging task due to inter patient anatomical variability, contrast variability and imaging artifacts. We propose a combined schema of multi-atlas and multi parametric Gaussian mixture modeling for perirectal fat segmentation in CT images. Multi-atlas based soft segmentation and multi parametric Gaussian mixture modeling aids in identifying the volume of interest VOI. Thereafter expectation maximization EM based soft clustering of the intensities of the VOI refined with positional probabilities of the perirectal fat provides the segmentation of the perirectal fat. The proposed method achieves a mean sensitivity value of 0.88±0.07 and a mean specificity value of 0.998±0.001 with 5 patient datasets in a leave-one-patient-out validation framework. Qualitative results show a good approximation of the perirectal fat volume compared to the ground truth.

82 Background: To develop a model for grade 3 (G3) late rectal bleeding (LRB) after radical radiotherapy (RT) for prostate cancer, in a pooled population from 2 large prospective trials: Airopros0102 (Fellin RO2014) and TROG 03.04 RADAR (Ebert IJROBP2015). Methods: Both trials included patients (pts) treated with 3DCRT at 66-80Gy, conventional fractionation. Planning data were available for all pts, G3 LRB was prospectively scored using the SOMA/LENT, with a minimum follow-up of 3 years. Rectal dose-volume histograms were reduced to Equivalent Uniform Dose (EUD) calculated with volume parameter n derived by 3 studies: n=0.06 (Rancati RO2004), n=0.05 (Rancati RO2011) and n=0.018 (Defraene IJROBP2011). EUD was inserted into multivariable logistic regression (MVL) together with clinical and treatment features. Irradiation of seminal vesicles (SV), irradiation of pelvic nodes, hormonal therapy, hypertension, previous abdominal surgery (SURG), use of anticoagulants, diabetes, cardiovascular diseases and presence of acute toxicity were considered as potential dose-modifying factors. Goodness of fit was evaluated with Hosmer Lemeshow test (HL), calibration through calibration slope and AUC was used for discrimination power. Results: 1,337 pts were available: 708 RADAR and 669 Airopros. G3 LRB was scored in 95 pts (7.1%): 62 RADAR and 33 Airopros. EUD calculated with n=0.05 was the best dosimetric predictor for G3 LRB. A 4-variable MVL model was fitted including EUD (OR=1.07 p=0.16), SV (OR=4.75 p

Plan review systems often allow dose volume histogram (DVH) recalculation as part of a quality assurance process for trials. A review of the algorithms provided by a number of systems indicated that they are often very similar. One notable point of variation between implementations is in the location and frequency of dose sampling. This study explored the impact such variations can have on DVH based plan evaluation metrics (Normal Tissue Complication Probability (NTCP), min, mean and max dose), for a plan with small structures placed over areas of high dose gradient. Dose grids considered were exported from the original planning system at a range of resolutions. We found that for the CT based resolutions used in all but one plan review systems (CT and CT with guaranteed minimum number of sampling voxels in the x and y direction) results were very similar and changed in a similar manner with changes in the dose grid resolution despite the extreme conditions. Differences became noticeable however when resolution was increased in the axial (z) direction. Evaluation metrics also varied differently with changing dose grid for CT based resolutions compared to dose grid based resolutions. This suggests that if DVHs are being compared between systems that use a different basis for selecting sampling resolution it may become important to confirm that a similar resolution was used during calculation.

Large multicentre radiotherapy trials incorporating assessment of multiple outcomes at multiple timepoints can generate extensive datasets. We have investigated graphical techniques for presentation of this data and the associated underlying dose-volume response information, necessary for guiding statistical analyses and translating outcomes to future patient treatments. A relational database was used to archive reviewed plan data for patients accrued to the TROG 03.04 RADAR trial. Viewing software was used to clean and enhance the data. Scripts were developed to export arbitrary dose-histogram data which was combined with clinical toxicity data with a median follow-up of 72 months. Graphical representations of dose-volume response developed include prevalence atlasing, univariate logistic regression and dose-volume-point odds ratios, and continuous cut-point derivation via ROC analysis. These representations indicate variable association of toxicities across structures and time-points.

In Canada, community and policy leaders have issued urgent calls to collect, analyze, and mobilize disaggregated data to inform equity-oriented initiatives aimed at addressing systemic racism and gender inequity, as well as other social inequities. This essay presents critical reflections from a national Roundtable discussion regarding how meaningful community engagement within academia–community–government research collaborations offers the opportunity to harness disaggregated data and advanced analytics to centre and address the priorities of equity-deserving and sovereignty-seeking groups. Participants emphasized four key priorities: (1) Building equitable and engaged partnerships that centre community-driven priorities and address structural barriers to community engagement; (2) Co-creating ethical data governance policies and infrastructure to support community data ownership and access; (3) Stimulating innovation and pursuing community involvement to create contextualized, advanced analyses and effective visualizations of disaggregated data; and (4) Building the capacity of all partners to effectively contribute to partnership goals. Capacity building was viewed as a bridge across a diversity of lived and professional expertise, enabling intersectoral research teams to collaborate in culturally safe and respectful ways. Beyond identifying key structural barriers impeding the promise of disaggregated data, we present practical opportunities for innovation in community-engaged scholarship to address social justice challenges in Canada.