SUMMARY: Colorectal cancer (CRC) is the third most common cancer diagnosed in both men and women in the United States. The American Cancer Society estimates that approximately 154,270 new cases of CRC will be diagnosed in the United States in 2025 and about 52,900 patients will die of the disease. The lifetime risk of developing CRC is about 1 in 23. The majority of CRC cases (about 75 %) are sporadic whereas the remaining 25 % of the patients have a family history of the disease. Only 5-6 % of patients with CRC with a family history background are due to inherited mutations in major CRC genes, while the rest are the result of accumulation of both genetic mutations and epigenetic modifications of several genes. Colorectal Cancer is a heterogeneous disease classified by its genetics, and even though the diagnosis of Colorectal Cancer in the US is dropping among people 65 years and older, the incidence has been rising in the younger age groups, with 12% of Colorectal Cancer cases diagnosed in people under age 50.
The DNA MisMatchRepair (MMR) system is responsible for molecular surveillance and works as an editing tool that identifies errors within the microsatellite regions of DNA and removes them. Defective MMR system leads to MSI (Micro Satellite Instability) and hypermutation, with the expression of tumor-specific neoantigens at the surface of cancer cells, triggering an enhanced antitumor immune response. MSI is therefore a hallmark of defective/deficient DNA MisMatchRepair (dMMR) system and occurs in 15% of all colorectal cancers. Defective MMR can be a sporadic or heritable event. Approximately 65% of the MSI high colon tumors are sporadic and when sporadic, the DNA MMR gene is MLH1. Defective MMR can manifest as a germline mutation occurring in MMR genes including MLH1, MSH2, MSH6 and PMS2. This produces Lynch Syndrome often called Hereditary Nonpolyposis Colorectal Carcinoma–HNPCC, an Autosomal Dominant disorder that is often associated with a high risk for Colorectal and Endometrial carcinoma, as well as several other malignancies including Ovary, Stomach, Small bowel, Hepatobiliary tract, Brain and Skin. MSI is a hallmark of Lynch Syndrome-associated cancers. MSI high tumors tend to have better outcomes and this has been attributed to the abundance of tumor infiltrating lymphocytes in these tumors from increase immunogenicity. These tumors therefore are susceptible to blockade with immune checkpoint inhibitors.
MSI testing is performed using a PCR or NGS based assay and MSI-High refers to instability at 2 or more of the 5 mononucleotide repeat markers and MSI-Low refers to instability at 1 of the 5 markers. Patients are considered Micro Satellite Stable (MSS) if no instability occurs. MSI-L and MSS are grouped together because MSI-L tumors are uncommon and behave similar to MSS tumors. Tumors considered MSI-H have deficiency of one or more of the DNA MMR genes. MMR gene deficiency can be detected by ImmunoHistoChemistry (IHC). NCCN Guidelines recommend MMR or MSI testing for all patients with a history of Colon or Rectal cancer. Unlike Colorectal and Endometrial cancer, where MSI-H/dMMR testing is routinely undertaken, the characterization of Lynch Syndrome across heterogeneous MSI-H/dMMR tumors is unknown.
Background
Checkpoint inhibitors have revolutionized the treatment landscape for MisMatch Repair–deficient (dMMR) metastatic solid tumors, offering durable responses across tumor types. This paradigm is now being explored in early-stage settings. Dostarlimab (JEMPERLI®) is a monoclonal antibody that binds to the PD-1 receptor and blocks its interaction with PD-L1 and PD-L2. Blocking the Immune checkpoint proteins unleashes the T cells, resulting in T cell proliferation, activation and a therapeutic response. Building on prior success in dMMR rectal cancer, this Phase 2, multicenter study investigated the feasibility of using neoadjuvant PD-1 blockade with Dostarlimab to achieve organ preservation in early-stage, surgically resectable dMMR solid tumors, potentially shifting the standard of care away from surgery and cytotoxic therapies.
Study Design and Patient Population
Conducted at Memorial Sloan Kettering Cancer Center, Hartford HealthCare, and Baptist Health Miami Cancer Institute, the study enrolled patients with newly diagnosed Stage I–III dMMR solid tumors, defined by loss of MLH1, PMS2, MSH2, or MSH6 expression on immunohistochemistry, that were amenable to curative-intent surgery. Two cohorts were formed:
- Cohort 1: Patients with locally advanced rectal cancer.
- Cohort 2: Patients with nonrectal dMMR solid tumors (including esophagogastric, colon, hepatobiliary, genitourinary, and gynecologic cancers).
All patients received Dostarlimab 500 mg IV every 3 weeks for 6 months (nine cycles). Clinical response was assessed within 8 weeks of completing therapy via tumor-specific imaging, endoscopy, and biopsy where applicable. Patients with residual disease were offered standard neoadjuvant therapy and surgery, while those achieving a clinical Complete Response (cCR) could opt for nonoperative management.
Primary and Exploratory Endpoints
- Cohort 1: Co-primary endpoints were Overall Response Rate and sustained cCR at 12 months post-treatment.
- Cohort 2: Exploratory analyses evaluated cCR rates, Recurrence-Free Survival (RFS), safety, and molecular correlates of response, including circulating tumor DNA (ctDNA).
Results
A total of 117 patients were analyzed:
- Cohort 1 (Rectal Cancer): All 49 patients who completed therapy achieved a cCR and declined surgery. At 12 months, 37 maintained a sustained cCR, meeting the efficacy threshold.
- Cohort 2 (Nonrectal Tumors): Of 54 patients, 35 achieved a cCR, with 33 choosing nonoperative management.
Across both cohorts:
- 103 patients completed Dostarlimab therapy.
- 84 (82%) achieved cCR.
- 82 patients (80%) avoided surgery.
- Two-year RFS: 92% (95% CI, 86–99).
- Median follow-up for recurrence: 20 months (range, 0–60.8).
- Safety: Most adverse events were grade 1–2 (60%), with 35% reporting no treatment-related events. No patient lost the opportunity for curative surgery due to disease progression.
Genomic and ctDNA Findings
- Germline dMMR variants were present in 44% of patients.
- Tumor-informed ctDNA testing tracked up to 50 tumor-specific mutations using a highly sensitive and specific assay.
- ctDNA clearance correlated strongly with cCR: all patients with a cCR showed complete ctDNA clearance by end of treatment.
- Persistently detectable ctDNA was associated with residual disease or eventual recurrence, reinforcing its value as a real-time, noninvasive biomarker for treatment response and residual disease monitoring.
Discussion
The findings underscore the transformative potential of neoadjuvant PD-1 blockade for early-stage dMMR cancers. Key takeaways include:
- Tumor-Agnostic Efficacy: Dostarlimab elicited robust responses across a variety of histologies, suggesting that dMMR status, rather than tumor origin, may drive sensitivity to immunotherapy.
- Organ Preservation: Surgery, and its associated morbidities, was avoided in the majority of patients, including those with rectal cancers where standard treatment often compromises fertility, continence, or other organ functions. Three women with rectal cancer treated in this trial successfully conceived and delivered children.
- Variable Responses by Histology: While responses were highest in rectal, colon, hepatobiliary, and urothelial cancers, lower cCR rates were observed in prostate and upper gastrointestinal tumors. This suggests underlying biological variability despite shared dMMR status.
- Monitoring Strategy: Integration of imaging, endoscopy, and ctDNA is critical. Liquid biopsy offered a reliable surrogate for tumor biopsy, particularly in inaccessible tumors, but caution is warranted as ctDNA alone may miss certain cases.
- Safety and Feasibility: The 6-month regimen was generally well tolerated, and no patient lost surgical eligibility due to disease progression. This supports the feasibility of prolonged neoadjuvant immunotherapy in appropriately selected patients.
Clinical Implications and Future Directions
This study lays the groundwork for a paradigm shift in the management of early-stage dMMR tumors. However, key questions remain:
- Long-Term Durability: While initial outcomes are promising, especially in rectal cancer, longer follow-up and additional data are necessary to confirm sustained benefit across nonrectal histologies.
- Histology-Specific Trials: Basket trials and single-arm studies may suffice for anatomically sensitive tumors (e.g., rectum, bladder), but randomized trials may still be appropriate in less morbidly resectable cancers (e.g., colon).
- Treatment Optimization: Determining the minimal effective duration of immunotherapy could reduce adverse events and cost. Median times to biopsy negativity (1.5 months) and imaging response (6.1 months) suggest a window for shortening therapy in responders.
- Shared Decision-Making: Given the potential for curative nonoperative management, multidisciplinary care teams must align on strategies and engage patients in informed decision-making, particularly where standard surgery entails long-term quality-of-life tradeoffs.
Conclusion
Neoadjuvant PD-1 blockade with Dostarlimab achieved clinical Complete Responses in a substantial majority of patients with early-stage dMMR tumors, offering a path to organ preservation without compromising curative potential. These results highlight the tumor-agnostic power of checkpoint inhibitors and present a compelling case for redefining the treatment of dMMR solid tumors. As follow-up data matures and histology-specific nuances are better understood, immunotherapy may become the new cornerstone of early-stage dMMR cancer management.
Nonoperative Management of Mismatch Repair–Deficient Tumors. Cercek A, Foote MB, Rousseau B, et al. N Engl J Med 2025;392:2297-2308.
