No abstract

Random forest and elastic net logistic regression yield higher discriminative performance in (chemo)radiotherapy outcome and toxicity prediction than other studied classifiers. Thus, one of these two classifiers should be the first choice for investigators when building classification models or to benchmark one's own modeling results against. Our results also show that an informed preselection of classifiers based on existing datasets can improve discrimination over random selection.

Summary Background The NRG/RTOG 9413 study showed that whole pelvic radiotherapy (WPRT) plus neoadjuvant hormonal therapy (NHT) improved progression-free survival in patients with intermediate-risk or high-risk localised prostate cancer compared with prostate only radiotherapy (PORT) plus NHT, WPRT plus adjuvant hormonal therapy (AHT), and PORT plus AHT. We provide a long-term update after no years of follow-up of the primary endpoint (progression-free survival) and report on the late toxicities of treatment. Methods The trial was designed as a 2 × 2 factorial study with hormonal sequencing as one stratification factor and radiation field as the other factor and tested whether NHT improved progression-free survival versus AHT, and NHT plus WPRT versus NHT plus PORT. Eligible patients had histologically confirmed, clinically localised adenocarcinoma of the prostate, an estimated risk of lymph node involvement of more than 15% and a Karnofsky performance status of more than 70, with no age limitations. Patients were randomly assigned (1:1:1:1) by permuted block randomisation to receive either NHT 2 months before and during WPRT followed by a prostate boost to 70 Gy (NHT plus WPRT group), NHT 2 months before and during PORT to 70 Gy (NHT plus PORT group), WPRT followed by 4 months of AHT (WPRT plus AHT group), or PORT followed by 4 months of AHT (PORT plus AHT group). Hormonal therapy was combined androgen suppression, consisting of goserelin acetate 3·6 mg once a month subcutaneously or leuprolide acetate 7·5 mg once a month intramuscularly, and flutamide 250 mg twice a day orally for 4 months. Randomisation was stratified by T stage, Gleason Score, and prostate-specific antigen concentration. NHT was given 2 months before radiotherapy and was continued until radiotherapy completion; AHT was given at the completion of radiotherapy for 4 months. The primary endpoint progression-free survival was analysed by intention to treat. This study is registered with ClinicalTrials.gov, number NCT00769548. The trial has been terminated to additional follow-up collection and this is the final analysis for this trial. Findings Between April 1, 1995, and June 1, 1999, 1322 patients were enrolled from 53 centres and randomly assigned to the four treatment groups. With a median follow-up of 8·8 years (IQR 5·07–13·84) for all patients and 14·8 years (7·18–17·4) for living patients (n=346), progression-free survival across all timepoints continued to differ significantly across the four treatment groups (p=0·002). The 10-year estimates of progression-free survival were 28·4% (95% CI 23·3–33·6) in the NHT plus WPRT group, 23·5% (18·7–28·3) in the NHT plus PORT group, 19·4% (14·9–24·0) in the WPRT plus AHT group, and 30·2% (25·0–35·4) in the PORT plus AHT group. Bladder toxicity was the most common grade 3 or worse late toxicity, affecting 18 (6%) of 316 patients in the NHT plus WPRT group, 17 (5%) of 313 in the NHT plus PORT group, 22 (7%) of 317 in the WPRT plus AHT group, and 14 (4%) of 315 in the PORT plus AHT group. Late grad...

Purpose: Many planning methods for high dose rate (HDR) brachytherapy treatment planning require an iterative approach. A set of computational parameters are hypothesized that will give a dose plan that meets dosimetric criteria. A dose plan is computed using these parameters, and if any dosimetric criteria are not met, the process is iterated until a suitable dose plan is found. In this way, the dose distribution is controlled by abstract parameters. The purpose of this study is to improve HDR brachytherapy planning by developing a new approach that directly optimizes the dose distribution based on dosimetric criteria.Method: We develop Inverse Planning by Integer Program (IPIP), an optimization model for computing HDR brachytherapy dose plans and a fast heuristic for it. We used our heuristic to compute dose plans for 20 anonymized prostate cancer patient image data sets from our clinic database. Dosimetry was evaluated and compared to dosimetric criteria.Results: Dose plans computed from IPIP satisfied all given dosimetric criteria for the target and healthy tissue after a single iteration. The average target coverage was 95%. The average computation time for IPIP was 30.1 seconds on a Intel(R) Core T M 2 Duo CPU 1.67 GHz processor with 3 Gib RAM.Conclusion: IPIP is an HDR brachytherapy planning system that directly incorporates dosimetric criteria. We have demonstrated that IPIP has clinically acceptable performance for the prostate cases and dosimetric criteria used in this study, both in terms of dosimetry and runtime. Further study is required to determine if IPIP performs well for a more general group of patients and dosimetric criteria, including other cancer sites such as GYN.