The Optimal Therapy after Progression on Immune Checkpoint Inhibitors in MSI Metastatic Gastrointestinal Cancer Patients: A Multicenter Retrospective Cohort Study

SIMPLE SUMMARY: Programmed death 1 (PD1)/programmed death ligand-1 (PD-L1) inhibitor is the standard therapy for advanced microsatellite instability (MSI)/mismatch repair-deficient (dMMR) gastrointestinal cancers. However, the suitable therapy after the progression of anti-PD1/PD-L1 for MSI/dMMR gastrointestinal cancer patients was unknown, until now. Here, we conducted a retrospective study to evaluate the efficacy of anti-PD1/PD-L1 plus other drug therapy versus chemotherapy with or without targeted therapy for patients who had progressed on prior anti-PD1/PD-L1 monotherapy. Our study found that anti-PD1/PD-L1 plus other drug therapy had significantly improved the disease control rate, progression-free survival, and overall survival, along with a numerically higher objective response rate versus chemotherapy with or without targeted therapy. The promising findings of our retrospective study need to be further confirmed in prospective trials. ABSTRACT: Background: In microsatellite instability (MSI)/mismatch repair-deficient (dMMR) gastrointestinal cancers, the optimum therapy after the progression of immune checkpoint inhibitors (ICIs) is yet unknown. Here, we compared the efficacy of programmed death 1 (PD1)/programmed death ligand-1 (PD-L1) inhibitors plus other therapy and chemotherapy with or without targeted therapy in MSI/dMMR gastrointestinal cancer patients after progression on anti-PD1/PD-L1 monotherapy. Methods: We retrospectively recruited MSI/dMMR gastrointestinal cancer patients who had progressed on anti-PD1/PD-L1 monotherapy. Objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and PFS ratio (PFSr) were compared between patients who received anti-PD1/PD-L1 plus other therapy (ICI-plus group) and patients who received chemotherapy with or without targeted therapy (chemo-targeted group). Results: In total, 26 and 25 patients were recruited in the ICI-plus group and chemo-targeted group, respectively. Significantly better DCR (80.8% vs. 44.0%, p = 0.007), PFS (median PFS 6.9 months vs. 3.0 months, p = 0.001), OS (median OS NR vs. 14.1 months, p = 0.043), and PFSr (2.4 vs. 0.9, p = 0.021), along with a numerically higher ORR (23.1% vs. 12.0%, p = 0.503) were observed in the ICI-plus group compared with the chemo-targeted group. Multivariate analyses identified the therapy regimen as an important prognostic factor in gastrointestinal cancers. Conclusions: Compared to conventional chemotherapy with or without targeted therapy, continuing anti-PD1/PD-L1 in combination with other treatments showed better clinical outcomes in MSI/dMMR gastrointestinal cancer patients who progressed on PD1/PD-L1 blockade, which should be validated prospectively in clinical trials.