Neoadjuvant FOLFIRINOX versus upfront surgery for resectable pancreatic head cancer (NORPACT-1): a multicentre, randomised, phase 2 trial

Summary

Background

In patients undergoing resection for pancreatic cancer, adjuvant modified fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) improves overall survival compared with alternative chemotherapy regimens. We aimed to compare the efficacy and safety of neoadjuvant FOLFIRINOX with the standard strategy of upfront surgery in patients with resectable pancreatic ductal adenocarcinoma.

Methods

NORPACT-1 was a multicentre, randomised, phase 2 trial done in 12 hospitals in Denmark, Finland, Norway, and Sweden. Eligible patients were aged 18 years or older, with a WHO performance status of 0 or 1, and had a resectable tumour of the pancreatic head radiologically strongly suspected to be pancreatic adenocarcinoma. Participants were randomly assigned (3:2 before October, 2018, and 1:1 after) to the neoadjuvant FOLFIRINOX group or upfront surgery group. Patients in the neoadjuvant FOLFIRINOX group received four neoadjuvant cycles of FOLFIRINOX (oxaliplatin 85 mg/m², irinotecan 180 mg/m², leucovorin 400 mg/m², and fluorouracil 400 mg/m² bolus then 2400 mg/m² over 46 h on day 1 of each 14-day cycle), followed by surgery and adjuvant chemotherapy. Patients in the upfront surgery group underwent surgery and then received adjuvant chemotherapy. Initially, adjuvant chemotherapy was gemcitabine plus capecitabine (gemcitabine 1000 mg/m² over 30 min on days 1, 8, and 15 of each 28-day cycle and capecitabine 830 mg/m² twice daily for 3 weeks with 1 week of rest in each 28-day cycle; four cycles in the neoadjuvant FOLFIRINOX group, six cycles in the upfront surgery group). A protocol amendment was subsequently made to permit use of adjuvant modified FOLFIRINOX (oxaliplatin 85 mg/m², irinotecan 150 mg/m², leucovorin 400 mg/m², and fluorouracil 2400 mg/m² over 46 h on day 1 of each 14-day cycle; eight cycles in the neoadjuvant FOLFIRINOX group, 12 cycles in the upfront surgery group). Randomisation was performed with a computerised algorithm that stratified for each participating centre and used a concealed block size of two to six. Patients, investigators, and study team members were not masked to treatment allocation. The primary endpoint was overall survival at 18 months. Analyses were done in the intention-to-treat (ITT) and per-protocol populations. Safety was assessed in all patients who were randomly assigned and received at least one cycle of neoadjuvant or adjuvant therapy. This trial is registered with ClinicalTrials.gov, NCT02919787, and EudraCT, 2015-001635-21, and is ongoing.

Findings

Between Feb 8, 2017, and April 21, 2021, 77 patients were randomly assigned to receive neoadjuvant FOLFIRINOX and 63 to undergo upfront surgery. All patients were included in the ITT analysis. For the per-protocol analysis, 17 (22%) patients were excluded from the neoadjuvant FOLFIRINOX group (ten did not receive neoadjuvant therapy, four did not have pancreatic ductal adenocarcinoma, and three received another neoadjuvant regimen), and eight (13%) were excluded from the upfront surgery group (seven did not have pancreatic ductal adenocarcinoma and one did not undergo surgical exploration). 61 (79%) of 77 patients in the neoadjuvant FOLFIRINOX group received neoadjuvant therapy. The proportion of patients alive at 18 months by ITT was 60% (95% CI 49–71) in the neoadjuvant FOLFIRINOX group versus 73% (62–84) in the upfront surgery group (p=0·032), and median overall survival by ITT was 25·1 months (95% CI 17·2–34·9) versus 38·5 months (27·6–not reached; hazard ratio [HR] 1·52 [95% CI 1·00–2·33], log-rank p=0·050). The proportion of patients alive at 18 months in per-protocol analysis was 57% (95% CI 46–67) in the neoadjuvant FOLFIRINOX group versus 70% (55–83) in the upfront surgery group (p=0·14), and median overall survival in per-protocol population was 23·0 months (95% CI 16·2–34·9) versus 34·4 months (19·4–not reached; HR 1·46 [95% CI 0·99–2·17], log-rank p=0·058). In the safety population, 42 (58%) of 73 patients in the neoadjuvant FOLFIRINOX group and 19 (40%) of 47 patients in the upfront surgery group had at least one grade 3 or worse adverse event. 63 (82%) of 77 patients in the neoadjuvant group and 56 (89%) of 63 patients in the upfront surgery group had resection (p=0·24). One sudden death of unknown cause and one COVID-19-related death occurred after the first cycle of neoadjuvant FOLFIRINOX. Adjuvant chemotherapy was initiated in 51 (86%) of 59 patients with resected pancreatic ductal adenocarcinoma in the neoadjuvant FOLFIRINOX group and 44 (90%) of 49 patients with resected pancreatic ductal adenocarcinoma in the upfront surgery group (p=0·56). Adjuvant modified FOLFIRINOX was given to 13 (25%) patients in the neoadjuvant FOLFIRINOX group and 19 (43%) patients in the upfront surgery group. During adjuvant chemotherapy, neutropenia (11 [22%] patients in the neoadjuvant FOLFIRINOX group and five [11%] in the upfront surgery group) was the most common grade 3 or worse adverse event.

Interpretation

This phase 2 trial did not show a survival benefit from neoadjuvant FOLFIRINOX in resectable pancreatic ductal adenocarcinoma compared with upfront surgery. Implementation of neoadjuvant FOLFIRINOX was challenging. Future trials on treatment sequencing in resectable pancreatic ductal adenocarcinoma should be biomarker driven.

Funding

Norwegian Cancer Society, South Eastern Norwegian Health Authority, The Sjöberg Foundation, and Helsinki University Hospital Research Grants.