Integrated analysis of two phase 2 trials of nivolumab in metastatic rare solid cancer with dMMR or pMMR (MASTER KEY sub-studies: ROCK and ROLL trial).

e14603 Background: Immune checkpoint inhibitors play a critical role in mismatch repair deficient (dMMR) / microsatellite instability-high (MSI-H) cancers, but has yet to investigate the activity of nivolumab in a subset of rare cancers (incidence < 6 / 100,000/year), that are MMR proficient (pMMR). Methods: We conducted two single arm phase 2 trials to investigate the efficacy and safety of nivolumab monotherapy in patients with advanced rare tumors with either pMMR (jRCT2091220345) or dMMR (jRCT2091220344). dMMR was defined by the complete loss of at least one of the following proteins: MLH1, MSH2, MSH6, PMS2. pMMR was defined when all four proteins were retained. Primary endpoint was independently reviewed response rate (RR). Secondary endpoints included progression free survival (PFS), overall survival (OS), and adverse events (AE). The dMMR/MSI-H trial enrolled 5 to 15 patients, based on a Bayesian design with a threshold RR of 5% and an expected RR of 30%, requiring ≥2 responders to achieve a posterior probability of ≥95% that the true RR of nivolumab is at least 5%. The pMMR trial was designed to explore the clinical activity in patients with pMMR, but also to confirm the clinical significance of “dMMR/MSI-H” as a predictive marker. If the upper limit of 95% confidence interval of the objective RR in the pMMR trial is below 30%, it implies that dMMR/MSI-H serves as a predictive marker. To evaluate this hypothesis at one-sided significance level of 5% with 90% power, a sample size of 54 patients with pMMR is necessary, assuming that the expected RR of nivolumab to be 15%. Results: Between May 2018 and March 2021, 11 and 57 patients received nivolumab treatment in the dMMR/MSI-H trial and pMMR trial, respectively. 10 and 54 patients in the respective trials were included in the full analysis set. The objective RR was 60% (95% CI 26.2-87.8) and 5.6% (95% CI 1.2-15.4) in the dMMR/MSI-H trial and pMMR trial, respectively. Median PFS was 10.1 months (95% CI 0.9-NE), and median OS was not yet reached in the dMMR/MSI-H trial, while the median PFS was 2.5 months (95% CI 1.9-4.8), and median OS was 12.7 months (95% CI 8.1-18.5) in the pMMR trial. Histology of the three responders in the pMMR trial were adnexal carcinoma of skin (2/4, 50%), and soft tissue sarcoma (1/14, 7.1%). In dMMR/MSI-H patients with tumor mutation burden (TMB) data, 5 of 6 were TMB ≥ 10/Mb, all of whom showed response. In contrast, in the pMMR group, two of the 22 patients with TMB data had TMB > 10/Mb, but neither showed response. Treatment-related grade 3 or higher AEs were observed in 0 and 8 (14%) patients in the dMMR/MSI-H and pMMR trials, respectively. Conclusions: Clinical benefit of nivolumab in pMMR advanced rare cancer patients was limited in contrast to dMMR/MSI-H patients, confirming the clinical significance of dMMR/MSI-H as a predictive marker. Further predictive biomarker studies are needed for pMMR tumors. Clinical trial information: jRCT2091220345 , jRCT2091220344 .