Safety and efficacy of preoperative or postoperative chemotherapy for resectable pancreatic adenocarcinoma (PACT-15): a randomised, open-label, phase 2-3 trial.

BACKGROUND: Pancreatic ductal adenocarcinoma are known to metastasise early and a rationale exists for the investigation of preoperative chemotherapy in patients with resectable disease. We aimed to assess the role of combination chemotherapy in this setting in the PACT-15 trial.

METHODS: We did this randomised, open-label, phase 2-3 trial in ten hospitals in Italy. We report the phase 2 part here. Patients aged 18-75 years who were previously untreated for pancreatic ductal adenocarcinoma, with Karnofsky performance status of more than 60, and pathologically confirmed stage I-II resectable disease were enrolled. Patients were randomly assigned (1:1:1), with a minimisation algorithm that stratified treatment allocation by centre and concentrations of carbohydrate antigen 19-9 (CA19-9 ≤5 × upper limit of normal [ULN] vs >5 × ULN), to receive surgery followed by adjuvant gemcitabine 1000 mg/m2 on days 1, 8, 15 every 4 weeks for six cycles (arm A), surgery followed by six cycles of adjuvant PEXG (cisplatin 30 mg/m2 , epirubicin 30 mg/m2 , and gemcitabine 800 mg/m2 on days 1 and 15 every 4 weeks and capecitabine 1250 mg/m2 on days 1-28; arm B), or three cycles of PEXG before and three cycles after surgery (arm C). Patients and investigators who gave treatments or assessed outcomes were not masked to treatment allocation. The primary endpoint was the proportion of patients who were event-free at 1 year. The primary endpoint was analysed in the per-protocol population. Safety analysis was done for all patients receiving at least one dose of study treatment. The trial is registered with ClinicalTrials.gov, number NCT01150630.

FINDINGS: Between Oct 5, 2010, and May 30, 2015, 93 patients were randomly allocated to treatment. One centre was found to be non-compliant with the protocol, and all five patients at this centre were excluded from the study. Thus, 88 patients were included in the final study population: 26 in group A, 30 in group B, and 32 in group C. In the per-protocol population, six (23%, 95% CI 7-39) of 30 patients in group A were event-free at 1 year, as were 15 (50%, 32-68) of 30 in group B and 19 (66%, 49-83) of 29 in group C. The main grade 3 toxicities were neutropenia (five [28%] of 18 in group A, eight [38%] of 21 in group B, eight [28%] of 29 in group C before surgery, and ten [48%] of 21 in group C after surgery), anaemia (one [6%] in group A, four [19%] in group B, eight [28%] in group C before surgery, and five [24%] in group C after surgery), and fatigue (one [6%] in group A, three [14%] in group B, two [7%] in group C before surgery, and one [5%] in group C after surgery). The main grade 4 toxicity reported was neutropenia (two [11%] in group A, four [19%] in group B, none in group C). Febrile neutropenia was observed in one patient (3%) before surgery in group C. No treatment-related deaths were observed.

INTERPRETATION: Our results provide evidence of the efficacy of neoadjuvant chemotherapy in resectable pancreatic ductal adenocarcinoma. Since the trial began, the standard of care for adjuvant therapy has altered, and other chemotherapy regimens developed. Thus, we decided to not continue with the phase 3 part of the PACT-15. We are planning a phase 3 trial of this approach with different chemotherapy regimens.

FUNDING: PERLAVITA ONLUS and MyEverest ONLUS.