Real-World Retrospective Study Confirms Efficacy, Safety of Neoadjuvant ICIs in dMMR/MSI Nonmetastatic Colon Cancer

A retrospective observational study across 6 French cancer centers has confirmed the safety and efficacy of neoadjuvant immune checkpoint inhibitors (ICIs) such as pembrolizumab (Keytruda) or nivolumab (Opdivo) and ipilimumab (Yervoy) in patients with mismatch repair-deficient (dMMR)/microsatellite instability (MSI) nonmetastatic colon cancer. However, the study reported a major pathologic response (pMR) of 64% and a complete pathologic response (pCR) of 42%, which were lower than the rates seen in previous clinical trials. This discrepancy suggests that real-world outcomes may differ from those observed in highly selected patient populations.¹

Several studies have confirmed the efficacy of neoadjuvant immune checkpoint inhibitor (ICI) regimens, reporting pCR rates of 43% to 68%.^2,3 Lemaire C et al wrote in the paper that “previous studies were predominantly conducted in highly selected patient populations, typically involving younger patients with ECOG performance status 0 and often in single-center settings.”

Between 2019 and 2024, 32 eligible patients were treated with either pembrolizumab (n = 22, 69%) or a combination of nivolumab and ipilimumab (n = 10, 31%). The primary end point was pathological response, with secondary end points including tolerability, toxicity, recurrence-free survival, and overall survival.

In the monotherapy arm, patients received 2 main regimens: 400 mg every 6 weeks (n = 12) and 200 mg every 3 weeks (n = 9). In the combination arm, patients received 3 mg/kg of nivolumab and 1 mg/kg of ipilimumab for 1 cycle, followed by 1 cycle of nivolumab monotherapy 2 weeks later.

The primary end point was pathological response after neoadjuvant ICI treatment. Secondary end points included tolerability, toxicity, recurrence-free survival (RFS), and overall survival (OS).

Baseline Characteristics

Overall, baseline characteristics were similar between the 2 groups. The median age was 70 years (range, 51-76), 53% of patients were female, and the median tumor size was 71.5 mm (range, 42-79). Three patients had 2 synchronous tumor sites, resulting in a total of 35 tumors, 77% of which were in the proximal colon. Ten patients (31%) had Lynch syndrome.

The median time from diagnosis to the start of immunotherapy was 4.4 weeks (IQR, 3.8-5.6 weeks). The median number of cycles before surgery was 2 (IQR, 2-4 cycles). The median duration of immunotherapy was 12.3 weeks (IQR, 7.3-20.8 weeks). A total of 11 patients received adjuvant treatment (pembrolizumab = 5; chemotherapy = 6).

Safety

In the overall cohort, 19% of patients experienced grade 1/2 AEs, and 19% experienced grade 3 or greater AEs. Of the grade 3 or higher AEs, 14% were in the pembrolizumab arm and 30% were in the nivolumab/ipilimumab arm. Three patients discontinued treatment due to toxicity.

In this real-world setting, 64% of the 33 tumors analyzed achieved a major pathologic response, including 42% with a complete pathologic response. The pCR rate was notably higher in the combination arm (67%) compared to the monotherapy arm (33%).

In the current study, the pCR rate in the combination arm was 67%, which is consistent with previous studies, however, it was only 33% in the monotherapy arm. The investigators wrote that, “These discrepancies underline the importance of comparing results obtained in clinical trials with those obtained in real-world cohorts, and suggest that monotherapy may be less effective than combination therapy.” Further, grade 3 or greater toxicity was higher than in published studies suggesting the need for caution when interpreting efficacy and safety data from clinical trials conducted on selected patients.

According to the investigators, this study provides the first real-world evidence on preoperative immune checkpoint inhibitors (ICIs) in patients with MSI/dMMR nonmetastatic colon cancer, offering valuable insights into prescribing practices and patient outcomes.

The investigators reported lower pathologic response rates and more severe toxicities compared with previously published trials, highlighting the need for caution when applying this treatment in less-selected patient populations. These discrepancies may be attributed to several limitations of the study, including its retrospective design, small sample size, the absence of a central review of baseline CT scans, and the lack of standardized neoadjuvant protocols.

Nonetheless, the findings underscore the importance of developing more standardized patient selection criteria and preoperative treatment regimens to optimize the tolerability and efficacy of this emerging treatment.

REFERENCES:

1. Lemaire C, Boileve A, Manceau G, et al. Neoadjuvant immunotherapy for nonmetastatic dMMR/MSI colon cancer: a real-world retrospective AGEO study. ESMO Open. Published online July 31, 2025. doi:10.1016/j.esmoop.2025.105516

2. Chalabi M, Verschoor YL, Tan PB, et al. Neoadjuvant immunotherapy in locally advanced mismatch repair-deficient colon cancer. N Engl J Med. 2024;390(21):1949-1958.

3. Hu H, Kang L, Zhang J, et al. Neoadjuvant PD-1 blockade with toripalimab, with or without celecoxib, in mismatch repair-deficient or microsatellite instability-high, locally advanced, colorectal cancer (PICC): a single-centre, parallel-group, non-comparative, randomised, phase 2 trial. Lancet Gastroenterol Hepatol. 2022;7(1):38-48.