Dostarlimab Yields High cCR Rates and Organ Preservation in Early-Stage dMMR Tumors

Solid tumor | Image Credit:

© Sebastian Kaulitzki – stock.adobe.com

Neoadjuvant treatment with the PD-1 inhibitor dostarlimab-gxly (Jemperli) resulted in high rates of clinical complete response (cCR) and enabled organ preservation in patients with early-stage mismatch repair–deficient (dMMR) solid tumors, according to findings from a phase 2, single-arm study (NCT04165772) published in the New England Journal of Medicine.

Among 103 patients who completed treatment across the 2 cohorts, 82% (95% CI, 72%-88%) achieved a cCR and 80% (95% CI, 70%-87%) proceeded with nonoperative management. In cohort 1, all 49 patients with dMMR rectal cancer who completed dostarlimab therapy experienced a cCR and elected to forgo surgery. Of these, 37 maintained a sustained cCR at 12 months, meeting the prespecified criterion for efficacy. In cohort 2, which included patients with nonrectal dMMR solid tumors, 35 of 54 patients achieved a cCR, and 33 opted for nonoperative management.

At a median follow-up of 20.0 months (range, 0–60.8), the 2-year recurrence-free survival (RFS) rate across the full study population (n = 117) was 92% (95% CI, 86%–99%). No patients experienced loss of surgical eligibility during or following treatment.

"When the patients went to surgery, almost all of them had a significant decrease in tumor burden. It’s not that there was no efficacy—it’s just that they were not able to achieve a complete clinical response," lead study author Andrea Cercek, MD, explained in an interview with OncLive®. "It does appear to be somewhat tumor site dependent, and I [believe] that’s where further work is certainly needed to determine whether those tumor types simply require a longer duration of therapy, or potentially combination immunotherapy, or even combination immunotherapy and chemotherapy, to truly maximize those responses."

Cercek is the co-director of the Center for Young Onset Colorectal and Gastrointestinal Cancers as well as the section head for colorectal cancer at Memorial Sloan Kettering Cancer Center in New York, New York.

Phase 2 Trial Design and Patient Enrollment Criteria

The nonrandomized, single-arm, phase 2 study evaluated the potential for nonoperative management in patients with early-stage, dMMR solid tumors who were eligible for curative-intent surgery. Patients were screened at Memorial Sloan Kettering Cancer Center, Hartford HealthCare, and Baptist Health Miami Cancer Institute. Eligibility required a newly diagnosed stage I, II, or III solid tumor that was amenable to surgical resection and confirmed to be dMMR, as determined by loss of expression of MLH1, MSH2, MSH6, or PMS2 on immunohistochemical staining.

Eligible patients had either rectal or nonrectal dMMR tumors and received 500 mg of neoadjuvant dostarlimab administered intravenously every 3 weeks for 6 months. Clinical response assessments were performed up to 8 weeks after completion of dostarlimab therapy using tumor-specific imaging and endoscopy, when feasible. All feasible assessments were conducted. Patients with residual disease were offered standard neoadjuvant therapy followed by surgery. Those who achieved a cCR—defined as no evidence of residual disease on imaging and on endoscopy when feasible—were permitted to pursue nonoperative management without further therapy or surgery.

The study was conducted under a Simon’s two-stage minimax design for the rectal cancer cohort (cohort 1). If 4 or fewer patients achieved a sustained cCR for at least 12 months after completing dostarlimab, the study would be stopped for futility. If 13 or more achieved a sustained cCR, the trial would be stopped for efficacy. Given the high rate of sustained cCR observed in stage 1, the trial was discontinued early for efficacy, and stage 2 was not initiated.

Cohort 1 evaluated 2 primary end points: overall response and sustained cCR at 12 months following dostarlimab therapy, with or without chemoradiotherapy. Patients who underwent surgery and achieved a pathological complete response (pCR) were also included in the end point analysis.

Cohort 2 included patients with dMMR nonrectal solid tumors and was evaluated using exploratory end points, including RFS, safety, and durability of response. Tumor genomic profiling was performed via whole-exome sequencing on matched tumor and normal blood samples to assess somatic mutations and microsatellite instability. Circulating tumor DNA (ctDNA) was assessed using a tumor-informed assay that monitored up to 50 tumor-specific somatic mutations.

All statistical analyses were performed using R software version 4.3.2. Confidence intervals were not adjusted for multiplicity and were not used as a basis for formal hypothesis testing.

Baseline Patient Characteristics

A total of 124 patients were enrolled in the phase 2 study. Seven patients were excluded from the final analysis: 3 patients were found to have metastatic disease or second primary cancers prior to initiating treatment and subsequently withdrew from the study, 2 patients discontinued therapy after a single dose of dostarlimab, 1 patient discontinued after 4 doses, and 1 patient was lost to follow-up.

At the data cutoff, 103 patients had completed neoadjuvant dostarlimab therapy and 14 remained on treatment. In cohort 1, which included patients with dMMR rectal cancer, 49 patients completed treatment, with 48 completing all 9 planned cycles of dostarlimab. One patient in this cohort discontinued treatment after 8 cycles due to an asthma exacerbation that was unrelated to study treatment.

In cohort 2, 54 patients completed treatment. Of these, 49 completed the full 9 cycles of dostarlimab. Three patients discontinued therapy before achieving a cCR due to concerns regarding lack of treatment response, and 2 discontinued after achieving cCR due to treatment-related adverse effects (AEs) as assessed by the investigator.

Safety Analysis

Any-grade AEs were reported in 65% of patients who received at least 1 dose of dostarlimab. The majority of these effects were mild to moderate in severity, with reversible grade 1 or 2 AEs occurring in 60% of patients. The most frequently observed grade 1 or 2 AEs included fatigue (23%), rash or dermatitis (21%), and pruritus (19%).

Grade 3 AEs were uncommon and occurred in only 1 patient each; these included diabetes, lung infection, hypothyroidism, encephalitis, and neutropenia. A single case of grade 4 febrile neutropenia was reported.

"Sixty-five percent of patients experienced low-grade AEs, which were predominantly dermatologic—such as rashes, itching—and some fatigue. [Approximately] 13% of patients developed hypothyroidism. There isn’t much that can be done preventatively in terms of autoimmune toxicities; this has been well described in many prior immunotherapy studies. It’s mostly a matter of monitoring symptoms and intervening if necessary, typically with steroids or stronger medications. Thankfully, in our study, we saw very little of that, " Cercek explained.

Reference

Cercek A, Foote MB, Rousseau B, et al. Nonoperative management of mismatch repair–deficient tumors. N Engl J Med. Published online April 27, 2025. doi:10.1056/nejmoa2404512