Deficient mismatch repair (dMMR) and microsatellite instability-high (MSI-H) metastatic colorectal cancer (mCRC) represents a molecularly distinct subgroup that is highly sensitive to immune checkpoint inhibitors (ICIs). Anti-PD-1 antibodies, either alone or combined with anti-CTLA-4 agents, have become the standard of care for first-line treatment of these patients, with some achieving long-term remission that may represent a functional cure.
Despite this success, up to 50% of patients with dMMR/MSI-H mCRC develop resistance to ICIs. Resistance is classified as innate resistance (IR) - disease progression or death within the first six months - or acquired resistance (AR) - progression that occurs after an initial period of disease control. The biological drivers of both IR and AR remain poorly understood.
Mucinous differentiation is a common histopathological feature found in up to 50% of dMMR/MSI-H colorectal cancers. Preclinical evidence links mucinous histology to an immune-cold tumor microenvironment, including lower T-cell densities and reduced PD-L1 expression. However, its specific impact on resistance patterns in the metastatic, immunotherapy-treated setting had not been systematically evaluated in large cohorts.
Mucinous adenocarcinoma is defined by the presence of at least 50% extracellular mucin pools within the tumor tissue. This classification is made by pathologists reviewing resected specimens or biopsies, and the 50% threshold used in this study reflects the standard diagnostic definition applied at each participating institution.
In dMMR/MSI-H colorectal cancers specifically, extracellular mucus pools and MUC1 expression have been linked to lower T-cell infiltration compared to non-mucinous counterparts. Additionally, mucinous dMMR/MSI-H tumors show reduced PD-L1 expression, which may dampen immune recognition and limit the effectiveness of checkpoint blockade.
Prior studies in early-stage disease had not found mucinous histology to independently affect cancer-specific survival after surgical resection. However, the metastatic and immunotherapy context is fundamentally different, motivating this investigation into whether mucinous differentiation shapes treatment resistance in the checkpoint inhibitor era.
Study design: This was a retrospective multinational observational cohort study enrolling patients with dMMR/MSI-H mCRC who received anti-PD-1 monotherapy or anti-CTLA-4 plus anti-PD-1 combination therapy in any treatment line. Data were collected from multiple academic hospitals in Europe (Italy, France, Spain), the United States (MD Anderson Cancer Center), and Israel, spanning treatments from 2014 to 2024.
MMR and MSI status were assessed locally using immunohistochemistry, multiplex PCR, and/or next-generation sequencing. Mucinous histology was defined as a mucinous component of 50% or greater assessed locally by pathologists. Key baseline variables collected included age, sex, ECOG performance status, tumor sidedness, primary tumor resection status, RAS/BRAF mutation status, metastatic timing (synchronous versus metachronous), number and sites of metastases, and ICI regimen used.
Primary endpoints were progression-free survival (PFS) and overall survival (OS). Innate resistance was defined as progressive disease or death within 6 months of ICI initiation. Acquired resistance was defined as progression or death after 6 months. For patients without innate resistance, PFS-1 and OS-1 were calculated from the 6-month landmark to specifically capture acquired resistance dynamics. Statistical analyses used Kaplan-Meier estimation and multivariable Cox proportional hazards regression.
From 984 patients initially screened, 929 patients were included in the final analysis after excluding those with missing histology data or insufficient follow-up. Of these, 316 patients (34%) had mucinous tumors and 613 (66%) had non-mucinous tumors. Median follow-up was 44.7 months.
Patients with mucinous histology differed in several baseline features: they were more likely to have right-sided primary tumors (79.4% vs 67.5%), had higher rates of primary tumor resection (90.2% vs 76.8%), had fewer liver metastases (31.9% vs 39.0%), and had numerically higher RAS mutation rates. These differences reflect known biological characteristics of mucinous colorectal tumors.
Overall, patients with mucinous histology experienced inferior PFS (HR 1.48, 95% CI 1.22-1.78, p less than 0.0001) and a trend toward inferior OS (HR 1.25, 95% CI 0.99-1.57, p=0.053). During the first 6 months, the PFS and OS curves for mucinous and non-mucinous groups were nearly identical (6-month PFS: 71.5% vs 73.0%), with divergence appearing only after this early period - the hallmark of acquired rather than innate resistance.
RECIST tumor response data were available for 666 of 929 patients (72%). Patients with mucinous histology had a significantly lower objective response rate (ORR) of 48.0% compared to 65.1% in non-mucinous patients (odds ratio 0.50, 95% CI 0.36-0.69, p less than 0.001). This 17-percentage-point gap represents a clinically meaningful difference in immunotherapy activity.
The difference was even more pronounced for complete response (CR) rates: 17.3% in mucinous vs 30.9% in non-mucinous patients (OR 0.47, 95% CI 0.31-0.70, p less than 0.001). Complete responses are especially important in MSI-H mCRC because they may represent immune-mediated tumor eradication. The substantially lower CR rate in mucinous patients suggests a reduced capacity for deep, durable immune responses.
In contrast, disease control rates (DCR) were similar between groups (79.8% mucinous vs 81.8% non-mucinous; OR 0.88, p=0.540). This finding indicates that mucinous tumors can achieve initial stabilization at comparable rates, but are less capable of achieving deeper responses. This pattern - lower ORR but similar DCR - is consistent with disease persistence rather than outright early progression.
A defining observation of this study was the distinctive timing of the survival curve separation: curves for mucinous and non-mucinous groups were essentially superimposable during the first six months, then progressively diverged. This pattern is the hallmark of acquired resistance - the tumor initially responds comparably but eventually escapes immune control - rather than innate resistance, which would manifest as early treatment failure.
Among the 674 patients without innate resistance, 182 patients (27%) developed acquired resistance. Of these, 41.4% had mucinous histology versus only 19.9% with non-mucinous histology. Mucinous patients had significantly inferior PFS-1 from the 6-month landmark (3-year PFS-1: 63.6% vs 81.1%; HR 2.06, 95% CI 1.54-2.75, p less than 0.001).
In multivariable analysis adjusting for other clinical factors, mucinous histology remained independently associated with shorter PFS-1 (HR 2.10, 95% CI 1.56-2.82, p less than 0.001). The association with OS-1 was numerically inferior but did not reach statistical significance (HR 1.36, 95% CI 0.91-2.02, p=0.130), possibly due to the impact of subsequent therapies after progression.
When outcomes were analyzed by both histology and ICI regimen, patients with mucinous tumors receiving anti-PD-1 monotherapy had the worst outcomes of any subgroup: 3-year PFS and OS rates of only 39.7% and 56.1%, respectively. This compares unfavorably to patients in the other three groups (non-mucinous on monotherapy, and either histotype on dual blockade).
Dual anti-CTLA-4/PD-1 blockade - such as nivolumab plus ipilimumab - was associated with longer PFS and OS in both mucinous and non-mucinous patients. Importantly, in the mucinous group, dual blockade appeared to at least partially counteract the adverse resistance phenotype, with the 6-month landmark analysis confirming that this benefit was sustained over time.
The hypothesis for why dual blockade may help in mucinous disease relates to its ability to rescue immune-depleted tumor microenvironments. Exploratory analyses from the CheckMate 142 trial suggested that ipilimumab addition may partially offset the adverse effect of an immune-cold microenvironment, which is characteristic of mucinous tumors. The recently reported CheckMate-8HW trial further demonstrated superiority of ipilimumab plus nivolumab over nivolumab monotherapy in dMMR/MSI-H mCRC overall.
The immune-suppressive tumor microenvironment of mucinous tumors offers a mechanistic explanation for the observed resistance. Extracellular mucus pools physically impede T-cell infiltration, while MUC1 overexpression further suppresses immune activity. In dMMR colon tumors specifically, mucinous differentiation is associated with lower T-cell densities and reduced PD-L1 expression - both features that limit checkpoint inhibitor efficacy.
The observed pattern of acquired rather than innate resistance aligns with these microenvironmental mechanisms. Mucinous tumors are not fundamentally refractory to initial immune activation - hence the similar 6-month PFS rates and DCR values - but they lack the capacity to sustain the immune response needed for durable tumor eradication.
Response assessment poses particular challenges in mucinous disease. Residual mucinous deposits on CT or MRI imaging can represent either acellular mucin (treatment response with no viable tumor) or mucin containing residual cancer cells - and these are often indistinguishable by conventional imaging. Functional imaging (PET) and liquid biopsy assays tracking circulating tumor DNA are promising tools to clarify true treatment response in these patients.
Mucinous histology should be incorporated into clinical decision-making for patients with dMMR/MSI-H mCRC starting immunotherapy. For patients with mucinous tumors, anti-PD-1 monotherapy is likely insufficient, and dual PD-1/CTLA-4 blockade should be preferred where clinically feasible, even when weighing toxicity concerns in less fit patients.
Patients with mucinous histology who initially respond to ICIs require closer monitoring for acquired resistance, since relapses can occur well beyond the first year of treatment. Multidisciplinary evaluation of residual disease for locoregional ablative strategies is also warranted, as these patients have a substantial risk of harboring residual viable tumor even after apparent imaging-based complete responses.
The dampened immunotherapy sensitivity of mucinous MSI-H tumors should encourage investigation of novel treatment strategies: therapeutic cancer vaccines, next-generation checkpoint inhibitors, and adoptive T-cell therapies are all rational candidates. Mucinous histology should also be considered explicitly in clinical trials evaluating treatment de-escalation or discontinuation strategies for dMMR/MSI-H mCRC, as these patients carry the highest risk of late relapse.