Nivolumab Plus Ipilimumab Yields High Rate of pCR in the Neoadjuvant Setting for MSI/dMMR Oeso-Gastric Adenocarcinoma


Patients with resectable microsatellite instable/mismatch repair–deficient oeso-gastric junction adenocarcinoma experienced a high rate of pathologic complete response following treatment with neoadjuvant nivolumab and ipilimumab followed by adjuvant nivolumab.

Treatment with nivolumab (Opdivo) plus ipilimumab (Yervoy) in the neoadjuvant setting plus adjuvant nivolumab resulted in a high rate of a pathologic complete response (pCR) in patients with resectable microsatellite instable (MSI)/mismatch repair deficient (dMMR) oeso-gastric junction (OGJ) adenocarcinoma, according to findings from the the phase 2 GERCOR NEONIPIGA trial (NCT04006262).

Results, which were presented during the 2022 Gastrointestinal Cancers Symposium, showed a pCR rate of 59%, meeting the study’s primary end point. Additionally, 58.6% of 29 evaluable patients had tumor regression grade (TRG) 1 by Mandard classification, meaning that they experienced complete regression/fibrosis with no tumor cells.

Moreover, 13.8% had TRG 2, which was fibrosis with scattered tumor cells; 6.9% had TRG 3, which was fibrosis and tumor cells with a dominance of fibrosis; 13.8% had TRG 4, which was fibrosis and tumor cells with a dominance of tumor cells; and 6.9% had TRG 5, which was no evidence of regression.

Per the Becker grading system, 58.6% of patients had TRG 1a following treatment, which meant that they had complete tumor regression without residual tumor. Additionally, 13.8% of patients had TRG 1b, which was defined has having less than 10% of residual tumor per tumor bed; 6.9% had TRG 2, defined as having 10% to 50% of their residual tumor following treatment; and 21.7% were TRG 3, which meant that they had more than 50% of residual tumor cells.

“The primary objective [of] pCR rate was met, and neoadjuvant nivolumab and ipilimumab is feasible in patients with MSI/dMMR, resectable gastric/OGJ adenocarcinoma,” presenting author Thierry André, MD, of Sorbonne University, Department of Medical Oncology, Saint-Antoine Hospital, Paris, France, said in a presentation on the data. “NEONIPIGA raises the question of whether the surgery can be delayed or avoided for some patients with localized MSI/dMMR gastric/OGJ adenocarcinoma if immune checkpoint inhibitors are effective.”

Those with locally advanced, MSI/dMMR gastric/OGJ adenocarcinoma tend to have a better prognosis than those who have microsatellite stable, proficient mismatch repair disease. Questionable benefit is derived by those with MSI/dMMR disease with the use of perioperative chemotherapy comprised of fluoropyrimidines and platinum salt; in fact, this approach may result in decreased disease-free survival and overall survival, according to André.

Data have indicated that MSI/dMMR status may be used to predict the efficacy of immune checkpoint inhibitors. Investigators have hypothesized that the use of these agents in this population both prior to, and following, radical surgery may result in improved outcomes.

NEONIPIGIA enrolled patients with histologically proven, nonmetastatic gastric adenocarcinoma, or of the OGJ T2 to T4, NxM0, who were between the ages of 18 years and 75 years, had MSI and dMMR disease, an ECOG performance status of 0 or 1 (or a status of 0 if older than 70 years), an absolute neutrophil count of at least 1.5 x 109/L, a platelet count of at least 100 x 109/L, and hemoglobin of at least 9 g/dL. Patients needed to have acceptable renal and liver function, and they could not have received prior treatment for localized oeso-gastric cancer.

Study participants received intravenous (IV) nivolumab at 240 mg every 2 weeks in combination with ipilimumab at 1 mg/kg every 6 weeks as neoadjuvant treatment for a total of 6 cycles, equating to 12 weeks of treatment. Approximately 5 weeks later, patients underwent surgery. Then, they went on to receive adjuvant treatment with single-agent nivolumab, given at a dose of 480 mg every 4 weeks for 9 cycles, equating to 9 months.

Once treatment is completed, patients will be followed up every 2 months for 2 years and then every 6 months until 5 years from study inclusion.

Among the 32 patients included in the trial, the median age was 65 years (range, 40-84), 72% were male, 59% had an ECOG performance status of 0, 50% had gastric cancer, and 50% had GEJ cancer. Regarding histology, 75% had intestinal disease, 19% had diffuse disease, and 6% had missing information. All patients had dMMR positivity per immunohistochemistry; 81.25% had loss of MLH1 and/or PMS2, and 18.75% had loss of MSH2 and/or MSH6. All patients also had MSI disease. Nineteen percent of patients had Lynch syndrome.

All 32 patients received neoadjuvant treatment; 2 received 2 cycles, 2 received 3 cycles, 1 received 5 cycles, and 27 received 6 cycles. Three patients did not go on to receive surgery; 1 had metastatic, progressive disease following cycle 6 of treatment and 2 refused.

Of the 29 patients who underwent surgery, 3.5% had a total oesogastrectomy, 24% had a total gastrectomy, 31% had 4/5 gastrectomy, 38% had a Lewis-Santy procedure, and 3.5% had a pancreaticoduodenectomy.

Twenty-five patients went on to receive adjuvant nivolumab; 10 of these patients were still receiving treatment at the time of data cutoff. However, 15 were not, and this is because they received the full 9 cycles of nivolumab (n = 8), they experienced a serious adverse effect (AE; n = 3), or investigator decision (n = 4).

Regarding ypT stage, 19 patients had ypT0, 1 had ypT1a, 2 patients had ypT1b, 2 patients had ypT2, 5 patients had ypT3, and 3 had unknown status. In terms of ypN stage, 23 patients had ypN0, 6 had ypN1, and 3 had unknown status.

Additional data from the trial showed that with a median follow-up of 12 months (95% CI, 7.8-14.2), 2 patients experienced events. One person died 3 days following surgery and 1 patient experienced metastatic progressive disease following 6 cycles of treatment; the latter patient did not go on to receive surgery. A total of 31 patients were still alive at data cutoff, and 30 had not yet relapsed.

Regarding safety, 84% of patients experienced an any-grade treatment-related AE (TRAE) with neoadjuvant treatment; 25% of patients experienced a grade 3 or 4 TRAE. Sixteen percent of patients experienced a grade 3 or 4 TRAE that resulted in discontinuation.

The most common any-grade AEs included diarrhea (7%), colitis/ileitis (6%), fatigue (28%), pruritus (25%), pyrexia (25%), hepatitis (9%), adrenal insufficiency (3%), vomiting (6%), nausea (6%), rash (12%), hypothyroidism (6%), hyperthyroidism (22%), decreased appetite (9%), pancreatitis (3%), and other (28%).

The most common grade 3 or 4 AEs were diarrhea (3%), colitis/ileitis (6%), adrenal insufficiency (3%), vomiting (3%), decreased appetite (6%), and other (6%).

Moreover, 58.5% of 29 patients experienced a per- and/or post-operative complication; 22 patients had postoperative general complications. Complications included fistula (n = 6), pancreatitis (n = 3), ileus (n = 2), pneumonia (n = 2), atrial fibrillation (n = 2), death (n = 1), and other (n = 6).

“No new safety concerns [were observed], and surgical complications were as expected with these types of surgeries,” André concluded. “Ninety-four percent of patients are free of events with 12 months of follow-up.”


André T, Tougeron D, Piessen G, et al. Neoadjuvant nivolumab plus ipilimumab and adjuvant nivolumab in patients (pts) with localized microsatellite instability-high (MSI)/mismatch repair deficient (dMMR) oeso-gastric adenocarcinoma (OGA): the GERCOR NEONIPIGA phase II study. J Clin Oncol. 2022;40(suppl 4):244. doi:10.1200/JCO.2022.40.4_suppl.244

Related Videos
Administering immunotherapy following surgery in patients with early-stage triple-negative breast cancer does not appear to be as effective as beginning with neoadjuvant treatment, says Peter Schmid, MD, PhD.
Heather A. Parsons, MD, MPH, says she is excited about HER2 as a biomarker and investigating the biology of metastatic breast cancer with a minimally invasive liquid biopsy platform.
Cretostimogene grenadenorepvec’s efficacy compares favorably with the current nonsurgical standards of care in high-risk, Bacillus Calmette Guerin–unresponsive non-muscle invasive bladder cancer.
Artificial intelligence models may be “seamlessly incorporated” into clinical workflow in the management of prostate cancer, says Eric Li, MD.
Robust genetic testing guidelines in the prostate cancer space must be supported by strong clinical research before they can be properly implemented, says William J. Catalona, MD.
Treatment with tisotumab vedotin may be a standard of care in second- or third-line recurrent or metastatic cervical cancer, says Brian Slomovitz, MD, MS, FACOG.
Future analyses will look at durvalumab/olaparib for endometrial cancer populations with TP53 and POLE alterations, as well as those with estrogen receptor and progesterone receptor positivity.
Related Content