Immune checkpoint blockade combined with targeted therapies to provide disease control in mismatch repair-proficient colorectal cancer with peritoneal metastases.

Abstract

133 Background: Metastatic colorectal adenocarcinoma (mCRC) with proficient mismatch repair (pMMR) is associated with poor response to cytotoxic chemotherapy in the refractory setting, with a median overall survival (OS) of 10.8 months. Recent immunotherapy (IO) regimens utilizing a combination of immune checkpoint blockade (ICB) and tyrosine kinase inhibitors (TKIs) that suppress tumor-associated macrophages (TAMs) have shown promise for extra-hepatic, but not hepatic metastases. However, few studies include patients with peritoneal metastases (PMs) and the response rate of PMs to these therapies is unknown. We sought to examine the disease control rates (DCR) and OS of patients treated with PM who were treated with ICB regimens. Methods: Patients with pMMR tumors treated with ICB and TKI after progression on 1st and 2nd-line therapy from 2015 to 2022 were identified. Date of start of therapy, end of therapy, last follow up and vital status were determined from the medical record. Radiographic disease response within each metastasis site at 6 months and at last follow-up on therapy was determined for each patient. Overall survival was estimated using the Kaplan-Meier method. Results: A total of 93 patients treated with an ICB and TKI after at least two lines of cytotoxic systemic therapy were identified. Five were excluded due to dMMR/MSI-H status and eight due to lack of follow-up. A total of 33 patients with PM, with or without other metastases, had a median follow-up on therapy of 6.1 months and median follow-up until death or censoring of 9.8 months. Median OS was 15.1 (9.7-29.5) months, and 18-month survival was 43% (95%CI: 24-62%). OS was not associated with either BRAF (HR 0.98, 0.38-2.6, p=0.98) or RAS (HR 1.50, 0.62- 3.63, p=0.37) mutation status. Partial response of PM was noted in 5 patients (15%), stable disease in 14 (42%) and progressive disease in 14 (42%), resulting in DCR of 57%. Disease control for all metastatic sites was noted in 9 (27%) of patients, and median OS was not reached at 2 years. Median OS for patients without peritoneal progression was 22.0 months [12.3-NR], and PM progression was associated with worse OS (HR 0.66, p=0.024). A total of 32 patients with any metastatic site were treated for at least 6 months. Median follow-up was 16.1 months for this cohort. DCR at 6 months for CRLM was 23% (3/13), PM was 58% (14/24, p=0.08 vs CRLM) and for LM was 62% (8/13, p=0.11 vs CRLM). Conclusions: Response to ICB and targeted therapy combinations can vary widely by metastatic site, with liver trending towards less response than peritoneum or lung. Disease control in the peritoneum can be achieved in the majority of patients, and median OS compares favorably to standard-of-care therapy in the refractory setting. Patients with peritoneal disease should be included in future trials of ICBs as treatment options are otherwise limited.