Trial of FOLFIRI + Zimberelimab + Domvanalimab vs FOLFIRI With a Hybrid Synthetic Control Arm in Second Line Treatment of Neuroendocrine Carcinoma of Gastro-enteropancreatic or Unknown Origin (REWENEC 01)

Background Neuroendocrine carcinomas (NECs) of gastro-entero-pancreatic (GEP) or unknown (UK) origin are rare and highly aggressive diseases. The recommended first-line (L1) treatment is platinum-etoposide combination therapy, which has a progression-free survival (PFS) of only 4-9 months and a median overall survival (OS) of approximately 12 months. All patients experience relapse, often rapidly after this first line of chemotherapy. The standard second-line (L2) chemotherapies recommended by ESMO, ENETS, and NCCN, FOLFIRI and FOLFOX, have modest efficacy with a PFS of 3 months and a median OS of 6 months. The BEVANEC study (PHRCK 2014, NCT02820857) reported no benefit of FOLFIRI + bevacizumab compared to FOLFIRI in a randomized phase II study that enrolled 150 patients in 26 centers over a period of 5 years in France.

To date, the most promising efficacy data for this highly aggressive cancer come from clinical trials of immune checkpoint inhibitors (ICIs) targeting the PD-1/PD-L1 checkpoint. For example, in France, the non-comparative phase II NIPINEC trial (NCT03591731) randomized patients to receive nivolumab +/- ipilimumab in L2/3 and achieved its primary evaluation criterion (ORR-8 weeks\>10%). Other trials in Europe and worldwide have also reported efficacy data in the context of single-arm studies.

Scientific Questions and Unmet Needs:

1. New therapeutic options/perspectives are necessary for patients with GEP/UK NECs given the limited overall survival. 2. Approximately 50% of patients experienced early progression under immunotherapy in the NIPINEC trial and other trials, which may be explained by the absence of chemotherapy combined with immunotherapy and/or the existence of resistance mechanisms. 3. In the GEP/UK NEC indication, the design of these immunotherapy trials has been non-comparative single-arm studies because the realization of randomized comparative trials is considered very difficult for these very rare cancers (incidence \<5/million).

Rationale for the REWENEC-01 Trial The DURIGAST PRODIGE 59 study, conducted by the FFCD, demonstrated the feasibility and safety of the FOLFIRI + double immune checkpoint inhibitor (anti-PD-1 and anti-CTLA4) combination, as well as for the combination Folfox-Domvanalimab-Zimberelimab (anti-PD-1 and anti-TIGIT). In a translational study of the immune phenotype in patients with NECs treated with the anti-PD1 pembrolizumab, an increase in TIGIT expression was observed after pembrolizumab treatment and higher TIGIT expression on T cells in the blood of patients with high Ki67 expression in their tumors. These data suggest that TIGIT is a potential complementary therapeutic target to PD-1/PD-L1 checkpoint inhibition in GEP/UK NECs. Domvanalimab has been developed as an anti-TIGIT monoclonal antibody and zimberelimab as an anti-PD-1.

Design and primary objective of the REWNEC-01 Trial The REWENEC-01 trial is a comparative phase II trial that will randomize GEP/UK NEC patients between an experimental arm FOLFIRI+Zimberelimab + Domvanalimab and a control arm FOLFIRI in L2. The FOLFIRI arm will be a "hybrid" synthetic control arm composed of patients from historical/external data from the FOLFIRI arm of BEVANEC and French retrospective studies RBNEC and CEPD, mixed with patients recruited prospectively during the trial and randomized to the control arm. The randomization ratio for patients included prospectively during the trial will be 4:1 (4 patients assigned to FOLFIRI+Zimberelimab + Domvanalimab for 1 patient assigned to FOLFIRI). The randomization algorithm will take into account "external" patients assigned progressively to the control arm to obtain a 1:1 ratio between the trial arms, with balanced distributions of stratification factors between the two arms.

With 77 patients to be included, this strategy will provide statistical power equivalent to that of a trial including 122 patients, sufficient to demonstrate an advantage in overall survival rate at 12 months from 32% to 50%.

The hypotheses related to efficacy criteria are formulated a priori, as recommended by the FDA guidance document on trials with synthetic/external control arms. The proof of concept has been reported at ESMO 2023. The primary judgment criterion will be the overall survival rate at 12 months because it is a strong and significant criterion for translating the clinical benefit of the Chemotherapy + Zimberelimab + Domvanalimab combination. The design with a hybrid synthetic control arm allows for the consideration of a randomized comparative study in a cancer as rare as neuroendocrine carcinoma.