Continuous, maintenance or intermittent first line (1L) therapy for advanced pancreatic cancer (aPC): MI-OPTS-1 study.

Abstract

731 Background: 1L therapy for patients (pts) with aPC includes FOLFIRINOX, gemcitabine/ nab-paclitaxel (GnP), or NALIRIFOX typically until progression or toxicity. Phase 2 randomized PANOPTIMOX trial reported no differences in progression-free survival (PFS) or overall survival (OS) in the 6-month (mo) FOLFIRINOX arm vs maintenance (5FULV) arm. Due to data paucity, we evaluated the impact of continuous therapy until progression (C), maintenance therapy (M) and chemotherapy break (B) in pts with 1L therapy in the Michigan Medicine Optimal Treatment Schedule (MI-OPTS-1) study. Methods: This IRB approved retrospective study included consecutive pts 18 years or older who received 1L GnP or FOLFIRINOX for aPC between 01/2014-12/2024 with all available data. Pt characteristics, chemotherapy and related toxicity, and survival data were obtained through electronic medical records review at predefined intervals of 4 mo and end of therapy. In cohort C, pts received therapy until progression, intolerance or toxicity; cohort M pts received maintenance therapy (5FULV +/- irinotecan after FOLFIRINOX, or gemcitabine after GnP); and cohort B pts received intermittent therapy or break(s) – all cohorts received at least 4 mo of therapy without progression. Pts on M or B could revert to prior therapy after progression per pt/physician decision. Covariates between pairwise cohorts were compared using Chi-square or t test. Survival probabilities were estimated using the product-limit method of Kaplan-Meier using SAS (v9.4, Cary, NC). Results: In 183 pts who met criteria, median (range) age was 66 (29-83) years, 83 (45.3%) were female, 166 (90.7%) were White, and 161 (88%) had ECOG 0-1. FOLFIRINOX was used in 89 pts (48.6%) and GnP in 94 pts (51.4%). Baseline characteristics were similar across cohorts with exception of more females in C than B (P = 0.041), higher albumin in M than B or C (P=0.018 and P=0.019), and higher FOLFIRINOX use in M than B or C (P<0.001). Median (95% CI) PFS was 7.8 mo (6.9-8.2) for cohort C, 7.7 mo (5.8-9.9) for cohort M and 9.5 mo (8.4-11.3) for cohort B. Median (95% CI) OS was 12.9 mo (11.1-14.9) for cohort C, 25.2 mo (17-34.8) for cohort M and 21.1 mo (16.9-25.3) for cohort B. At 4 mo, no differences were seen in rates of anemia, neutropenia or thrombocytopenia, but incidence of 2+ peripheral neuropathy was lower for pts in M and C compared to B (P=0.049 and P<0.0001). At the end of 1L chemotherapy, pts in cohort M had a better-preserved ECOG than C or B (P=0.039 and P=0.002) while grade 2 or higher anemia was higher for pts in B and C compared to M (P=0.029 and P=0.025). Conclusions: Pt/physician decision to offer break or maintenance may be due to differences in group practice or higher peripheral neuropathy at 4 mo in the chemotherapy break cohort rather than consideration in elderly or unfit pts. Pts who received chemotherapy break(s) had significantly improved median PFS and OS compared to continuous therapy.