Multi-agent Induction Chemotherapy Followed by Chemoradiation is Associated With Improved Survival Compared to Chemotherapy Alone in Locally Advanced Pancreatic Cancer

Abstract

Locally advanced pancreatic cancer (LAPC) remains a deadly disease. The role of radiotherapy is uncertain after multiple randomized clinical trials (RCT) have yielded mixed results. We used the National Cancer Database (NCDB) to determine whether chemoradiation (CRT) yields a survival benefit compared to chemotherapy alone (CT). Patients with non-metastatic, unresectable LAPC diagnosed from 2004-2014 were identified in the NCDB. Patients treated with CT were compared with patients who received CRT using chi-square analysis. Univariate and multivariate Cox regression were used to compare demographic, clinical, and treatment characteristics predictive of survival. Propensity score matching and shared frailty analysis were done. Subgroup analyses were undertaken to examine patients who underwent pancreatectomy and cohorts of patients who underwent different CT or CRT regimens. We identified 8,689 patients with unresectable LAPC. CRT was associated with improved survival (MS 13.5mos) compared to CT (MS 10.6mos) on MVA (HR 0.80, p<0.001). Induction chemotherapy prior to CRT (HR 0.67, p<0.001) and multi-agent chemotherapy (HR 0.72, p<0.001) were also independent predictors of survival. On subgroup analysis, CRT patients who underwent multi-agent induction chemotherapy had superior median survival and pancreatectomy rate (MS 17.5mos, HR 0.70, p<0.001, 10% pancreatectomy rate) compared to multi-agent CT alone (MS 12.4mos, 3.3%). Conversely, survival for patients with multi-agent CT alone was superior to CRT given concurrently with single-agent CT (MS 11.2 mos, p<0.001). Patients able to undergo pancreatectomy experienced improved survival (MS 22 mos vs. 10.6 mos, HR 0.39, p<0.001). Maximizing systemic chemotherapy prior to chemoradiation improves survival and pancreatectomy rates relative to chemotherapy alone in LAPC. This data supports continued analysis of CRT in properly selected patients after maximized systemic therapy.Tabled 1Abstract 2464; Table Survival Times and Pancreatectomy Rates Compared Across Treatment GroupsTreatment GroupNumber of patientsMedian Survival (mos)MVA HRP valueOverall Survival % at 1,2,5 years, 95% CIPancreatectomy RateSingle Agent Chemo only1,9209.01.41 (1.32-1.51)<0.0011: 36.9 (34.7-39.1)2: 11.7 (10.3-13.2)5: 3.8 (3.0-4.7)0.5%Multi Agent Chemo only2,26612.4ReferenceReference1: 54.6 (52.5-56.6)2: 21.4 (19.7-23.2)5: 8.5 (7.4-9.8)3.3%Chemoradiation: Single Agent, Concurrent1,71211.21.18 (1.10-1.26)<0.0011: 47.0 (44.6-49.3)2: 14.6 (12.9-16.3)5: 4.0 (3.1-5.0)3.8%Chemoradiation: Single Agent, Induction35014.50.87 (0.76-0.98)0.021: 64.7 (59.4-69.5)2: 22.6 (18.3-27.2)5: 6.3 (4.0-9.2)4.9%Chemoradiation: Multi Agent, Concurrent73013.60.98 (0.89-1.07)0.661: 60.2 (56.5-63.7)2: 17.9 (15.2-20.8)5: 4.4 (3.1-6.1)5.3%Chemoradiation: Multi Agent, Induction90917.50.70 (0.64-0.76)<0.0011: 78.9 (76.1-81.5)2: 32.9 (29.8-36.0)5: 10.7 (8.7-12.9)9.8% Open table in a new tab