Efficacy and safety of neoadjuvant immune checkpoint inhibitors in patients with localized mismatch repair deficient colorectal cancer: A systematic review.

Abstract

149 Background: Clinical data demonstrating remarkable anti-tumor activity of immune checkpoint inhibitors (ICI) in patients with mismatch repair deficient (dMMR) metastatic colorectal cancer (CRC) and preclinical data suggesting enhanced efficacy of ICIs in the neoadjuvant setting than in the adjuvant setting led to numerous studies with neoadjuvant ICI in various tumor types including in dMMR CRC. We performed a systematic literature review of the published studies reporting the efficacy and safety of neoadjuvant ICIs in patients with localized dMMR CRC. Methods: Medline and Embase were searched for eligible case reports, case series, and clinical trial reports. Eligibility criteria included: 1. patients aged 18 years or older, 2. localized dMMR CRC, and 3. received neoadjuvant therapy primarily with ICI. Patients receiving radiation or chemotherapy along with ICI were excluded. The systematic review was conducted according to the PRISMA harms guidelines. Results: The database search yielded 143 citations which were reviewed by 3 authors (UG, KV, and SC), and 14 citations fulfilled all criteria for inclusion. The analysis included 173 patients with colon cancer (CC) and 19 with rectal cancer (RC). Among the patients with CC, 41 (24 %) received single-agent programmed death 1 (PD-1) blocker, and 132 (76 %) received dual blockade with an anti-PD-1 agent plus an anti-cytotoxic T-lymphocyte antigen 4 (CTLA-4) agent. Resection was performed in 171/173 (99%) patients, and 117/171 (69%) achieved complete pathological response (pCR). After median follow-up ranging between 8 and 15 months, none of the patients experienced cancer relapse. Among patients with RC, 16/19 (84%) received a PD-1 blocker, and 3 (16%) received dual PD-1 plus CTLA-4 blocker. Complete clinical response (cCR) was reported in 14/19 (74%) patients, and of the 5 patients who underwent surgery, 4 (80%) achieved pCR. All patients remained progression-free after median follow-up ranging from 6 to 12 months. Cancer progression or death on neoadjuvant ICI was not reported in any studies with CC or RC patients, and only two studies reported grade 3+ toxicity in 3% of CC patients. Conclusions: Neoadjuvant therapy with ICIs results in a remarkably high rate of clinical and pathological tumor regression with negligible safety concerns in patients with dMMR localized CRC. Well-designed clinical trials with long-term follow-up are needed to evaluate the organ-sparing potential of neoadjuvant ICI therapy in patients with dMMR CRC.