An international, randomized, open-label, phase III trial of adjuvant nab-paclitaxel plus gemcitabine vs gemcitabine alone for surgically resected pancreatic adenocarcinoma (APACT): primary analysis and quality of life outcomes

Abstract

Introduction: Patients with pancreatic cancer (PC) have a poor prognosis. Even with resectable PC, the 5‐year postsurgical survival rate is only ≥20%. Furthermore, surgical resection is associated with postoperative morbidity and diminished quality of life (QoL). In metastatic PC, nab‐paclitaxel plus gemcitabine (nab‐P/Gem) demonstrated significantly longer overall survival (OS) than gemcitabine (Gem) alone. The phase III APACT trial aimed to assess the efficacy and safety of nab‐P/Gem vs Gem monotherapy in patients with surgically resected PC. Primary results and QoL outcomes are reported here. Methods: Patients with previously untreated (including neoadjuvant), histologically confirmed pancreatic adenocarcinoma with macroscopic complete resection (R0/R1) were eligible. Key eligibility criteria included ECOG PS 0/1 and CA19‐9 level < 100 U/ mL within 14 days of randomization. Treatment was to be started within 12 weeks postsurgery. Patients were randomized to nab‐P 125 mg/m2 plus Gem 1000 mg/m2, or Gem 1000 mg/m2 alone, for six 28‐day cycles. The primary endpoint was independentlyassessed disease‐free survival (DFS) and the only clinical data provided to reviewers were baseline data and scans. Secondary endpoints were OS and safety. QoL, an exploratory endpoint, was assessed using the EORTC QLQ‐C30 and QLQ‐PAN26 questionnaires. Clinically meaningful differences were defined by a minimal important difference (MID) of 10. Results: The study included 866 randomized pts. Median age was 64 years (range 34‐ 86); 60% had ECOG PS 0, 72% were LN+ and 76% had R0. Overall, 69% of pts completed 6 cycles of treatment (nab‐P/Gem: 66%, and Gem: 71%). Median follow‐up time was 38.5 months. For the primary endpoint, median DFS (n=439 events) was 19.4 mo for nab‐P/Gem vs 18.8 mo for Gem (HR 0.88; 95% CI, 0.729‐1.063; stratified log‐rank P= .1824). Investigator‐assessed DFS (n=571 events) was 16.6 mo for nab‐ P/Gem vs 13.7 mo for Gem (HR 0.82; 95% CI, 0.694‐0.965; nominal P=.0168). Interim OS (n=427 events) was 40.5 mo for nab‐P/Gem) vs 36.2 mo for Gem (HR 0.82; 95% CI, 0.680‐0.996; nominal P= .045). Grade≥3 AEs were reported in 86% vs 68% of pts in the nab‐P/Gemvs Gem arms. The most common grade≥3 hematologic and nonhematologic TEAEs reported were neutropenia (49% vs 43%) and fatigue (10% vs 3%). QoL scores generally improved throughout the study. Although the mean EORTC QLQ‐C30 global health status/QoL score was lower for nab ‐P/Gem compared with Gem through 48 weeks (69.97 vs 72.59 at week 48), differences were MID for Gem. Adjuvant nab‐P/Gem may be an option for those not eligible for adjuvant FOLFIRINOX; however, the impact of this regimen in the adjuvant setting may be clearer as survival data mature.