Gastric adenocarcinoma is the 4th most frequent cancer and the 2nd leading cause of cancer mortality. Most of the patients have metastatic, locally advanced or recurrent unresectable disease. So, systemic treatment remains an important issue especially since chemotherapy improves survival and quality of life (compared to best supportive care alone).

Second-line chemotherapy-based treatment improves overall survival (OS) as compared to best supportive care alone in patients with an acceptable general condition (performance status 0-2). Indeed, with docetaxel monotherapy there was a significant difference in overall survival for the chemotherapy arm with a median of 5.2 versus 3.6 months in best supportive care alone arm (HR=0.67, p=0.01). Irinotecan monotherapy also significantly improves overall survival compared to supportive care alone in a phase III study (4.0 versus 2.4 months; HR=0.48, 95%CI 0.25-0.92; p=0.012).

Based on a phase III trial FOLFIRI (5-FU plus irinotecan) is one most used regimen in second-line in European countries, especially in France. FFCD 0307 trial, a phase III comparing FOLFIRI-ECX (epirubicin-cisplatin-capecitabine) to the reverse sequence (ECX-FOLFIRI), showed that both sequences are possible.

Preliminary results in metastatic gastric cancer with anti-PD1 mAbs are highly promising. In a trial with pembrolizumab, only PD-L1 positive tumors were eligible to the treatment with a cut off at 1%. Thirty-nine patients were enrolled and 67% had received at least two prior chemotherapy regimens. The overall response rate was 22%. The median PFS and OS were 1.9 months and 11.4 months, respectively. KEYNOTE-059 Phase 2 multicohort study with pembrolizumab monotherapy in advanced gastric cancer treatment has been presented at ASCO 2017 meeting. Among 259 patients included in the trial response rate was 11.6%. OS was 5.6 months. Response rates were 15.5% in PDL1+ tumors versus 6.4% in PDL1- tumors and 57.1% in MSI tumors versus 9% in MSS tumors. Up until now, overlap between microsatellite instability and PD-L1 expression is unknown in gastric cancer. An anti-PD-L1 mAb (avelumab) was evaluated in a phase Ib expansion study (n=20, Japanese patients), with 15% of objective response rate and 11.9 weeks for progression-free survival. A second cohort with avelumab included 55 patients for maintenance therapy after first-line chemotherapy, with 7.3% of objective response rate and 14 weeks of PFS. Phase I/II CheckMate-032 evaluated nivolumab (anti-PD-1) ± ipilimumab (anti-CTLA4) at different doses in advanced gastric cancer (17). The overall response rate was between 8% to 24% and the median OS between 4.8 to 6.9 months according to treatment arm.

Others anti-PD1/anti-PD-L1/anti-CTLA4 mAbs are also currently under investigation in gastric cancer alone or in combination with chemotherapy. Nevertheless, up until now there is no published data concerning ICI plus chemotherapy in gastric cancer. The present randomized multicentric non-comparative phase II study aimed to assess the rate of patients alive and without progression at 4 months with advanced gastric or gastro-oesophageal junction (GEJ) adenocarcinoma, pre-treated with fluoropyrimidine + platinum +/- taxane, with two arms Folfiri plus durvalumab versus Folfiri plus durvalumab plus tremelimumab. Indeed, most patients in the French multicentric first-line GASTFOX trial (506 patients planned between 2017 and 2020) can be included in the second-line setting in the DURIGAST trial. Due to the lack of data concerning Folfiri plus durvalumab plus tremelimumab combination, a safety run-in phase will be performed at the beginning of the DURIGAST trial.

Gastric adenocarcinoma is the fourth most frequent cancer and the second leading cause of cancer mortality. Advanced gastric adenocarcinoma has a poor prognosis with short overall survival (ranging from 10% to 15% at 5-years) even after surgical complete resection and despite the progress in therapeutic approaches. Most of the patients have metastatic, locally advanced or recurrent unresectable disease. So, systemic treatment remains an important issue especially since chemotherapy improves survival and quality of life (compared to best supportive care alone). First-line chemotherapy depends on HER2 status, which also influenced overall survival (14 months for HER2 positive versus 10 months for HER2 negative tumors). In HER2 negative tumors standard first-line regimen is a doublet of fluoropyrimidine (5-fluorouracil or capecitabine) plus a platinum salt (cisplatin or oxaliplatin). 5-fluorouracil (5-FU) and capecitabine as also cisplatin and oxaliplatin have similar efficacy but different toxicities.

In patients whose tumor overexpresses the HER2 receptor adding trastuzumab to fluoropyrimidine/cisplatin regimen increased overall survival compared to chemotherapy alone. In HER2 negative tumors the addition of docetaxel to cisplatin/fluoropyrimidine regimen increased overall survival but its use remains limited in clinical practice because of its high toxicity. Preliminary results demonstrated a high efficacy with less toxicities of docetaxel-oxaliplatin-fluoropyrimidine combination, also called TFOX/FLOT regimen. Indeed, in France a large phase III trial comparing TFOX versus FOLFOX in first-line treatment of patients with advanced gastric or gastro-oesophageal junction adenocarcinoma is ongoing (GASTFOX, trial NCT03006432). Primary endpoint is progression-free survival (PFS) and 506 patients are planned between 2017 and 2020 (actually at the date of January 30, 2018, 65 patients are included).

Second-line chemotherapy improves overall survival (OS) as compared to best supportive care alone in patients with an acceptable general condition (performance status 0-2). Indeed, with docetaxel monotherapy there was a significant difference in overall survival for the chemotherapy arm with a median of 5.2 versus 3.6 months in best supportive care alone arm (HR=0.67, p=0.01). Weekly paclitaxel monotherapy is also used because of its good efficacy-toxicity ratio. Irinotecan monotherapy also significantly improves overall survival compared to supportive care alone in a phase III study (4.0 versus 2.4 months; HR=0.48, 95%CI 0.25-0.92; p=0.012). Recently ramucirumab monotherapy demonstrated its efficacy on overall survival in a randomized, placebo-controlled second-line metastatic study. In a randomized phase 3 trial ramucirumab also showed its efficacy in combination with paclitaxel versus paclitaxel monotherapy with a median overall survival of 9.6 versus 7.4 months, respectively (p=0.017; HR=0.81). However, the "amelioration du service medical rendu" (ASMR) assessed by the French "Haute Autorité de Santé" (HAS) consider an insufficient benefit to a reimbursement of ramucirumab in France. The HAS gave a moderate ASMR opinion (ASMR IV).

Docetaxel is more and more frequently used in first-line chemotherapy then in this setting taxane (alone or combined with others drugs) cannot be used as second-line regimen. Indeed, based on a phase III trial FOLFIRI (5-FU plus irinotecan) is one most used regimen in second-line in European countries, especially in France. FFCD 0307 trial, a phase III comparing FOLFIRI-ECX (epirubicin-cisplatin-capecitabine) to the reverse sequence (ECX-FOLFIRI), showed that both sequences are possible.

Human tumors tend to activate the immune system regulatory checkpoints as a means of escaping immunosurveillance. For instance, interaction between PD1 (Program Death 1) and PD-L1 (Program Death 1 ligand) will lead the activated T cell to a state of anergy. PD-L1 is up regulated on a wide range of cancers. Anti-PD1 and anti-PD-L1 monoclonal antibodies (mAbs), called immune checkpoint inhibitors (ICIs), have consequently been designed to restore T cell activity. Others ICIs are investigated, notably cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) inhibitors. CTLA-4 transmits an inhibitory signal to T cells to prevent early excessive T cell activation. CTLA4 blockade may stimulate a more robust antitumor response by sustaining activation and proliferation of T lymphocytes and may overcome immune suppression mediated by regulatory T cells. ICIs have been recently tested in many cancers with promising results, especially in tumors with microsatellite instability (MSI) and/or PD-L1 overexpression.

Preliminary results in metastatic gastric cancer with anti-PD1 mAbs are highly promising. In a trial with pembrolizumab, only PD-L1 positive tumors were eligible to the treatment with a cut off at 1%. Thirty-nine patients were enrolled and 67% had received at least two prior chemotherapy regimens. The overall response rate was 22%. The median PFS and OS were 1.9 months and 11.4 months, respectively. KEYNOTE-059 Phase 2 multicohort study with pembrolizumab monotherapy in advanced gastric cancer treatment has been presented at ASCO 2017 meeting. Among 259 patients included in the trial response rate was 11.6%. OS was 5.6 months. Response rates were 15.5% in PDL1+ tumors versus 6.4% in PDL1- tumors and 57.1% in MSI tumors versus 9% in MSS tumors. Up until now, overlap between microsatellite instability and PD-L1 expression is unknown in gastric cancer. An anti-PD-L1 mAb (avelumab) was evaluated in a phase Ib expansion study (n=20, Japanese patients), with 15% of objective response rate and 11.9 weeks for progression-free survival. A second cohort with avelumab included 55 patients for maintenance therapy after first-line chemotherapy, with 7.3% of objective response rate and 14 weeks of PFS. Phase I/II CheckMate-032 evaluated nivolumab (anti-PD-1) ± ipilimumab (anti-CTLA4) at different doses in advanced gastric cancer (17). The overall response rate was between 8% to 24% and the median OS between 4.8 to 6.9 months according to treatment arm.

Others anti-PD1/anti-PD-L1/anti-CTLA4 mAbs are also currently under investigation in gastric cancer alone or in combination with chemotherapy. Nevertheless, up until now there is no published data concerning ICI plus chemotherapy in gastric cancer. Finally, immunogenic cell death induced by chemotherapy may enhance efficacy of ICIs. Durvalumab (MEDI4736) is a human monoclonal antibody directed against PD-L1 in development for the treatment of many cancers. A phase I study included 16 patients with advanced gastric cancer and the objective response rate was 25%. Tremelimumab is a fully human monoclonal antibody against CTLA-4. Durvalumab plus tremelimumab combination showed a manageable tolerability profile, with antitumour activity irrespective of PD-L1 status in non-small cell lung cancer (NSCLC). Durvalumab alone or combined with tremelimumab is evaluated in phase III studies in NSCLC (e.g NEPTUNE and MYSTIC), small cell lung cancer (CASPIAN), hepatocellular carcinoma (HIMALAYA), bladder cancer (DANUBE) and head and neck cancer (EAGLE and KESTREL).

Concerning safety of anti-PD1 plus anti-CTLA4 combination, in the randomized phase I/II CheckMate-032 study, that included 160 patients, there was no unexpected toxicity signal. Grade 3 and 4 treatment-related adverse events were 17%, 47%, and 27%, respectively. These rates of grade 3 and 4 treatment-related adverse events are those usually found with the anti-PD1 plus anti-CTLA4 combination in other tumors, observed approximately in 40% of patients. Up until now, there is no published data concerning combination of ICIs plus irinotecan. Nevertheless, in all trial combining chemotherapy plus anti-PD1 and/or anti-CTLA4 chemotherapy drugs were used at full-dose (5FU, oxaliplatin, cisplatin…). An Italian trial just started and combined full-dose FOLFOXIRI (5-FU 3200 mg/m2 plus irinotecan 165 mg/m2 and oxaliplatin 85 mg/m2) with bevacizumab (5 mg/kg) and atezolizumab (anti-PD-L1, 840 mg) in metastatic colorectal cancers as first-line treatment. FOLFOXIRI is a triplet chemotherapy more "toxic" than FOLFIRI doublet chemotherapy and this trial is a randomized phase II (FOLFOXIRI plus bevacizumab and atezolizumab versus FOLFOXIRI plus bevacizumab). There is, however, a preliminary safety phase in 6 patients, once they have all received at least 2 cycles of treatment, the latter being administered at full dose (AtezoTRIBE trial, NCT03721653).

The present randomized multicentric non-comparative phase II study aimed to assess the rate of patients alive and without progression at 4 months with advanced gastric or gastro-oesophageal junction (GEJ) adenocarcinoma, pre-treated with fluoropyrimidine + platinum +/- taxane, with two arms Folfiri plus durvalumab versus Folfiri plus durvalumab plus tremelimumab. Indeed, most patients in the French multicentric first-line GASTFOX trial (506 patients planned between 2017 and 2020) can be included in the second-line setting in the DURIGAST trial. Due to the lack of data concerning Folfiri plus durvalumab plus tremelimumab combination, a safety run-in phase will be performed at the beginning of the DURIGAST trial.

Inclusion Criteria:

• Age ≥ 18 years.
• Body weight > 30kg.
• Histologically proven advanced-stage unresectable adenocarcinoma of the stomach or the GEJ (Siewert II or III).
• Known MSS/MSI status or tumor tissue available (frozen or paraffin-embedded, primary tumors or metastases) in order to allow determination of MSS/MSI status. The investigator needs to ensure that tumor tissues will be sent after patient randomization.
• Failure to platinium-based 1st line therapy with or without trastuzumab, or early recurrent disease after surgery with neo-adjuvant and/or adjuvant platinium-based chemotherapy (within 6 months of the end of chemotherapy) or progression during neo-adjuvant and/or adjuvant platinium-based chemotherapy.
• Eligible for a second-line treatment with irinotecan and 5-FU.
• Measurable or non-measurable lesion according to the Response Evaluation Criteria in Solid Tumors (RECIST 1.1).
• Eastern Cooperative Oncology Group (ECOG) performance status 0-1.
• Adequate organ function: ANC ≥ 1.5 x 109/L, haemoglobin ≥ 9 g/dL, platelets ≥ 100 x 109/L, AST/ALT ≤ 3 x ULN (≤ 5 x ULN in case of liver metastase(s)), GGT ≤ 3 x ULN (≤ 5 x ULN in case of liver metastase(s)), bilirubin ≤ 1.5 x ULN, creatinin clearance > 40 mL/min (MDRD).
• Evidence of post-menopausal status or negative urinary or serum pregnancy test for female pre-menopausal patients.
• Man and woman who childbearing potential agrees to use two methods (one for the patient and one for the partner) of medically acceptable forms of contraception during the study and for 6 months after the last treatment intake.
• Patient is able to understand, sign, and date the written informed consent form at the screening visit prior to any protocol-specific procedures performed.

Exclusion Criteria:

• - Concurrent enrolment in another clinical study - unless it is an observational study or during the follow-up period of an interventional study.
• Receipt of the last dose of anticancer therapy ≤ 2 weeks prior to the first dose of study drug.
• Radiotherapy within 4 weeks prior to the first dose of treatment.
• History of chronic inflammatory bowel disease (IBD).
• Current or prior bowel obstruction within 28 days before the first dose of study drugs.
• Any unresolved significant toxicity NCI CTCAE v4.0 ≥ grade 2 from previous anticancer therapy.
• Concurrent use of hormonal therapy for non-cancer-related conditions is acceptable
• Major surgical procedure (e.g. exploratory laparoscopy is not considered as a major surgical procedure) within 28 days prior to the first dose of treatment.
• Prior allogeneic bone marrow transplantation or prior solid organ transplantation.
• Active or prior documented autoimmune or inflammatory disorders (patients with alopecia, vitiligo, controlled hypo or hyperthyroidism, any chronic skin condition not requiring immunosuppressant therapy are eligible). Patients without active disease in the last 5 years may be included.
• Uncontrolled intercurrent illness, including but not limited to, ongoing or active infection, symptomatic congestive heart failure, uncontrolled hypertension, unstable angina pectoris, cardiac arrhythmia, interstitial lung disease, serious chronic gastrointestinal conditions associated with diarrhea, or psychiatric illness/social situations that would limit compliance with study requirement, substantially increase risk of incurring AEs or compromise the ability of the patient to give written informed consent.
• Severe cardiac disorders within 6 months.
• Severe liver dysfunction
• History of idiopathic pulmonary fibrosis, drug-induced pneumonitis, organizing pneumonia, or evidence of active pneumonitis on screening chest CT-scan.
• History of leptomeningeal carcinomatosis. Patients whose brain metastases have been treated may participate provided they show radiographic stability. In addition, any neurologic symptoms that developed either as a result of the brain metastases or their treatment must have resolved or be stable either, without the use of steroids, or are stable on a steroid dose of ≤10mg/day of prednisone or its equivalent for at least 14 days prior to the start of treatment
• Positive test for HIV, active hepatitis B or hepatitis C, active tuberculosis.
• History of active primary immunodeficiency
• Current or prior use of immunosuppressive medication within 14 days before the first dose of study drugs (excepted: intranasal, inhaled, topical steroids or local steroid injection -at physiologic dose does not exceed 10 mg/day of prednisone or its equivalent - steroids as premedication for hypersensitivity reactions).
• Receipt of live attenuated vaccine within 30 days prior to the first dose of treatment
• Known allergy or hypersensitivity to any of the study drugs or any of the study drug excipients. In order to check all the contraindications of each drugs, please refer to the updated versions of the SmPCs presented in Appendix 9.
• Current or prior use of St. John's Wort within 14 days before the first dose of study drugs (St. John's Wort is not allowed during participation in the trial).
• Treatment with sorivudine or analogs (brivudine).
• Treatment with phenytoin or analogs.
• Prior treatment with irinotecan, anti-PD1, anti PD-L1, anti-CLTA4 or other immunotherapy for cancer treatment
• Known Uridine Diphosphate Glucuronyltransferase (UGT1A1) or Dihydropyrimidine Dehydrogenase (DPD) enzyme deficiencies.
• Active infection requiring intravenous antibiotics at the time of Day 1 of Cycle 1.
• Other malignancy within 5 years prior to study enrolment, except for localized cancer in situ, basal or squamous cell skin cancer.
• Pregnant or breastfeeding female patient.