Seaward, Samantha A

4500 Background: Trastuzumab is a monoclonal antibody against human epidermal growth factor receptor 2 (HER2). The primary objective of RTOG 1010 was to determine if trastuzumab increases disease-free survival (DFS) when combined with trimodality treatment for patients with HER2 overexpressing esophageal adenocarcinoma. Methods: This open label, randomized phase III trial included patients with newly diagnosed stage T1N1-2, T2-3N0-2 adenocarcinoma of the esophagus involving the mid, distal, or esophagogastric junction and up to 5cm of the stomach. All patients received chemotherapy (C) of paclitaxel, 50mg/m2 and carboplatin AUC = 2, weekly for 6 weeks, with radiation (XRT: 3D-CRT or IMRT, 50.4 Gy in 28 fractions) followed by surgery. Patients were randomized 1:1 to receive weekly trastuzumab 4mg/kg week 1 then 2mg/kg/weekly x 5 during CXRT then 6 mg/kg for 1 dose prior to surgery and 6mg/kg every 3 weeks for 13 treatments after surgery. HER2 status was determined by IHC and gene amplification by FISH. With a 2-sided alpha of 0.05, 162 DFS events provide 90% power to detect a signal for an increase in median DFS from 15 to 25 months. DFS and overall survival (OS) were estimated by the Kaplan-Meier method. and arms were compared using the log rank test. The Cox proportional hazards model was used to analyze treatment effect. Results: 571 patients were entered for assessment of HER2 expression, 203 HER2+ patients randomized. The median follow-up for alive patients is 5.0 years. The estimated 2, 3, and 4-year DFS (95% CI) for the CXRT +trastuzumab arm were 41.8% (31.8%, 51.7%), 34.3% (24.7%, 43.9%), and 33.1% (23.6%, 42.7%), respectively, and for the CXRT arm were 40.0% (30.0%, 49.9%), 33.4% (23.8%, 43.0%), and 30.1% (20.7%, 39.4%), respectively; log-rank p = 0.85. The median DFS time is 19.6 months (13.5-26.2) for the CXRT +trastuzumab arm compared to 14.2 months (10.5-23.0) for the CXRT arm. The hazard ratio (95% CI) comparing the DFS of CXRT+trastuzumab arm to the CXRT arm was 0.97 (0.69, 1.36). The median OS time was 38.5 months (26.2-70.4) for the CXRT+trastuzumab arm compared to 38.9 months (29.0-64.5) for the CXRT arm, hazard ratio (95% CI): 1.01 (0.69, 1.47). There was no statistically significant increase in treatment-related toxicities with the addition of trastuzumab including no increase in cardiac events. Conclusions: The addition of trastuzumab to trimodality treatment did not improve DFS for patients with HER2 overexpressing esophageal adenocarcinoma. Supported by NCI grants U10CA180868, UG1CA189867, U10CA180822 and Genentech. Clinical trial information: NCT01196390 .

55 Background: To compare the predictive ability of health utility mapping algorithms derived using cross-sectional and longitudinal data specific to the Expanded Prostate Cancer Index Composite (EPIC). Methods: This mapping study utilized data from an international, multicenter, randomized controlled trial of patients with low-risk prostate cancer conducted by NRG Oncology (NCT00331773). Health-related quality-of-life (HRQoL) data were collected using EPIC, and health utilities were obtained using EuroQOL-5D (EQ5D) at baseline and 6, 12 and 24 months post-intervention. Data were split into an estimation sample (70%) and a validation sample (30%). Ordinary Least Squares (OLS) regression models were estimated using baseline cross-sectional data as well as pooled data from all assessment periods. Random effects (RE) specifications that explicitly model the longitudinal nature of the data were also estimated. Candidate models were selected based on root mean square error (RMSE). Results: A total of 196 (147) patients in the estimation sample had complete EQ5D and EPIC domain (subdomain) data at all time points. OLS models using combined data outperformed the counter-part RE models as well as OLS models using baseline data in the five-fold cross-validation. Addition of covariates to the models resulted in improved predictive ability. In the external validation, when only EPIC domain/ subdomain data are available, the OLS model using combined data predicted EQ5D utilities better than the counterpart RE model and OLS model using baseline data (RMSE=0.121108 & 0.111345). OLS model using baseline data outperformed other model types for algorithms with EPIC domains and demographics (RMSE=0.121757), while RE models outperformed the other two model types for algorithms with EPIC subdomains and demographic data, (0.112782) and for algorithms with EPIC domains/ subdomains, demographics, and clinical covariates (RMSE=0.123589 & 0.163093). Conclusions: While algorithms using pooled data outperformed other model types in internal validation, RE models showed better predictive ability in external validation for algorithms with covariates. Clinical trial information: NCT00331773.

Purpose: NRG/RTOG 0848 was designed to determine whether adjuvant radiation with fluoropyrimidine sensitization improved survival following gemcitabine-based adjuvant chemotherapy for patients with resected pancreatic head adenocarcinoma. In step 1 of this protocol, patients were randomized to adjuvant gemcitabine versus the combination of gemcitabine and erlotinib. This manuscript reports the final analysis of these step 1 data. Methods: Eligibility—within 10 weeks of curative intent pancreaticoduodenectomy with postoperative CA19-9<180. Gemcitabine arm—6 cycles of gemcitabine. Gemcitabine+erlotinib arm—gemcitabine and erlotinib 100 mg/d. Two hundred deaths provided 90% power (1-sided α=0.15) to detect the hypothesized OS signal (hazard ratio=0.72) in favor of the arm 2. Results: From November 17, 2009 to February 28, 2014, 163 patients were randomized and evaluable for arm 1 and 159 for arm 2. Median age was 63 (39 to 86) years. CA19-9 ≤90 in 93%. Arm 1: 32 patients (20%) grade 4 and 2 (1%) grade 5 adverse events; arm 2, 27 (17%) grade 4 and 3 (2%) grade 5. GI adverse events, arm 1: 22% grade ≥3 and arm 2: 28%, ( P =0.22). The median follow-up (surviving patients) was 42.5 months (min-max: <1 to 75). With 203 deaths, the median and 3-year OS (95% confidence interval) are 29.9 months (21.7, 33.4) and 39% (30, 45) for arm 1 and 28.1 months (20.7, 30.9) and 39% (31, 47) for arm 2 (log-rank P =0.62). Hazard ratio (95% confidence interval) comparing OS of arm 2 to arm 1 is 1.04 (0.79, 1.38). Conclusions: The addition of adjuvant erlotinib to gemcitabine did not provide a signal for increased OS in this trial.

4007 Background: NRG/RTOG 0848 is a 2-step study designed to determine whether erlotinib (E) added to gemcitabine (G) (randomized Ph II) &/or adjuvant radiation with concurrent 5-FU or capecitabine following 6 months of systemic chemotherapy (Ph III), improve survival in patients (pts) with resected pancreatic head adenocarcinoma. The erlotinib results are reported here. Methods: Eligible pts include those with resected pancreatic head adenocarcinoma, pathologic stage T1-T3, N0-1, M0; PS 0-1, & CA19-9 ≤ 180 IU/L. Pts in Arms 1 & 2 received G 1 gm/m2 weekly for 3 weeks in a 28-day cycle for 6 cycles. Pts in Arm 2 also received E 100 mg/day. The primary hypothesis for the E portion was that G+E would increase overall survival (OS) compared to G alone. With a 1-sided alpha of 0.15, 200 OS events provide 80%/90% power to detect a signal for an increase in median OS from 22 to 28.8/30.6 months (mos). OS was estimated by the Kaplan-Meier method & arms compared using the log rank test. The Cox proportional hazards model was used to analyze treatment effect. Results: 336 pts were randomized from 11/17/2009 to 2/28/2014, with 163 pts evaluable for G and 159 for G+E. Median age was 63 years (39-86). Most pts had pathologic T3 disease (78%) & CA19-9 ≤ 90 (93%). There are 32 pts (20%) with grade 4 adverse events (AEs) & 2 pts (1%) with grade 5 AEs on G and 27 (17%) & 3 (2%) on G+E arm, respectively. There are fewer grade ≥ 3 GI AEs on the G arm (22%) as compared to the G+E arm (28%), and 110 (69.2%) & 93 (59.6%) pts received at least 85% of planned G dose for the G & G+E arms, respectively. 58% of E pts received at least 85% of planned E dose. The median follow-up for alive pts is 42.5 mos (min-max: < 1-75). With 203 deaths, median & 3-yr OS (95% CI) are 29.9 mos (21.7-33.4) & 39% (30, 45) for G and 28.1 mos (20.7-30.9) & 39% (31, 47) for G+E; log-rank p = 0.62. The hazard ratio (95% CI) comparing OS of G+E to G is 1.04 (0.79- 1.38). Conclusions: The addition of adjuvant E to G did not provide a signal for increased OS in pts with resected pancreatic head cancer compared to G alone. Accrual to the trial is continuing to answer the Ph III radiation question. Clinical trial information: NCT01013649.

1 Background: To determine whether the efficacy of a hypofractionated (H) schedule is no worse than a conventional (C) schedule in men with low-risk prostate cancer. Methods: From April 2006 to December 2009, one thousand one hundred fifteen men with low-risk prostate cancer (clinical stage T1-2a, Gleason ≤ 6, PSA < 10) were randomly assigned 1:1 to a conventional (C) schedule (73.8 Gy in 41 fractions over 8.2 weeks) or to a hypofractionated (H) schedule (70 Gy in 28 fractions over 5.6 weeks). The trial was designed to establish with 90% power and alpha = 0.05 that (H) results in 5-year disease-free survival (DFS) that is not lower than (C) by more than 7% (hazard ratio (HR) < 1.52). Secondary endpoints include freedom from biochemical recurrence (FFBR) and overall survival. At the third planned interim analysis (July 2015), the NRG Oncology Data Monitoring Committee recommended that the results of the trial be reported. Results: One thousand one hundred and one protocol eligible men were randomized: 547 to C and 554 to H. Median follow-up is 5.9 years. Baseline characteristics are not different according to treatment arm. At the time of analysis 185 DFS events have been observed; 99 in the C arm and 86 in the H arm. The estimated 7-year disease-free survival is 75.6% (95% CI 70.3, 80.1) in the C arm and 81.8% (77.5, 85.3) in the H arm. The DFS HR (C/H) is 0.85 (0.64, 1.14). Comparison of biochemical recurrence (HR = 0.77, (0.51, 1.17)) and overall survival (HR = 0.95, (0.65, 1.41)) also met protocol non-inferiority criteria. Grade ≥ 3 GI toxicity is 3.0% (C) vs. 4.6% (H), Relative risk (RR) for H vs. C 1.53, (95% CI 0.86, 2.83); grade ≥ 3 GU toxicity is 4.5% (C) vs. 6.4% (H), RR = 1.43 (0.86,2.37). Conclusions: In men with low-risk prostate cancer, 70 Gy in 28 fractions over 5.6 weeks is non-inferior to 73.8 Gy in 41 fractions over 8.2 weeks. Clinical trial information: NCT00331773.