Dostarlimab Plus Cobolimab Improves MPR Rate vs Dostarlimab in High-Risk Resectable Melanoma

High-Risk Resectable Melanoma

| Image credit: © Galina - stock.adobe.com

The addition of cobolimab to dostarlimab-gxly (Jemperli) was tolerable and led to efficacy improvements compared with dostarlimab monotherapy for the neoadjuvant treatment of patients with high-risk resectable melanoma, according to data from the phase 2 NEO-MEL-T trial (NCT04139902) presented during the 2025 ASCO Annual Meeting.

Findings from the primary analysis of NEO-MEL-T demonstrated that patients who received the combination (n = 27) achieved a major pathological response (MPR) rate of 55.6% and those who received dostarlimab alone (n = 30) experienced an MPR rate of 33.3%. The pathological complete response rates were 37% and 33.3%, respectively. Additionally, 18.5% and 3.7% of patients in the combination arm experienced a pathological major response or a pathological partial response, respectively.

“Despite the advances we’ve seen with PD-1 blockade, with or without CTLA-4 [inhibition], investigation into other strategies is warranted, as not all patients respond to our current standards [of care],” Meghan Mooradian, MD, said during the presentation. “We have strong preclinical and early clinical data supporting TIM-3 [inhibition] as a rational strategy. In our proof-of-concept, phase 2 randomized study, we found the addition of the TIM-3 inhibitor cobolimab to the PD-1 inhibitor dostarlimab to be tolerable with improved MPR and relapse-free survival [RFS] compared with neoadjuvant PD-1 monotherapy historical controls.”

Mooradian is a clinical oncologist at Massachusetts General Hospital, and an instructor at Harvard Medical School, both in Boston.

Examining the Study Design and Unpacking the Baseline Characteristics

NEO-MEL-T was a randomized, multicenter trial that enrolled patients with stage IIIB to IVA de novo or recurrent high-risk resectable cutaneous melanoma. Patients were required to be amenable to pretreatment biopsy and not have central nervous system disease. Prior treatment with anti–PD1, anti–CTLA-4, anti–LAG-3, or BRAF/MEK inhibitors was not permitted.

Patients received neoadjuvant dostarlimab or dostarlimab plus cobolimab. Dostarlimab was administered at a dose of 500 mg every 3 weeks for 6 weeks in both arms. Patients in the combination arm were also treated with 300 mg of cobolimab every 3 weeks for 6 weeks. Following therapeutic lymph node dissection (TLND), patients in both arms received 500 mg of adjuvant dostarlimab every 3 weeks for 4 cycles, followed by 1000 mg every 6 weeks for 6 cycles, over the course of 48 weeks.

The primary end point was the MPR rate in either arm compared with neoadjuvant anti–PD-1 therapy as the historical control. Secondary end points included overall survival (OS), distant metastasis-free survival (DMFS), overall response rate (ORR), RFS, and safety, including the delay or cancellation of surgery.

At baseline, the median ages in the monotherapy and combination arms were 65 years (range, 37-91) and 67 years (range, 31-91), respectively. Most patients in both arms were male (66.7% vs 66.7%), had lactate dehydrogenase levels at or above the upper limit of normal (63.3% vs 63.0%), and did not have in-transit disease (63.3% vs 74.1%). Molecular alterations present in both arms consisted of BRAF (26.7% vs 29.7%) and NRAS (16.7% vs 25.9%) mutations, as well as other (16.7% vs 18.5%) and unknown (40% vs 25.9%) alterations. Disease stages by American Joint Committee on Cancer 8th edition criteria included IIIB (46.7% vs 40.8%), IIIC (46.7% vs 48.1%), IIID (6.6% vs 0.0%), and IV-A (0.0% vs 11.1%).

Additional Secondary End Point and Safety Data

Additional findings from NEO-MEL-T revealed that the ORRs in the combination and monotherapy arms were 33.3% and 16.7%, respectively (1-sided P = .1187). At a median follow-up of 22 months (range, 0-55), the median RFS was not reached (NR) vs 48 months, respectively (log-rank P = .1919). The 1-year RFS rates were 87% and 82%, respectively. The median OS and DMFS were NR in both arms.

The 1-year RFS rate among patients who achieved an MPR was 100% in both arms. Among all patients with an MPR, the median RFS was NR compared with 48 months among all patients without an MPR (log-rank P = .0026). The median RFS was NR among patients with (n = 15) or without (n = 12) an MPR who received the combination. These values were NR and 48 months among patients in the monotherapy arm who did (n = 10) and did not (n = 20) achieve an MPR.

No patients in either the monotherapy or combination arms were unable to complete planned neoadjuvant treatment or undergo TLND. Patients in both arms were unable to complete adjuvant therapy due to toxicity (16.7% vs 14.8%) or disease progression (20.0% vs 11.1%); 26.7% and 22.2% of patients remained on adjuvant therapy at the data cutoff, respectively. Melanoma-specific events in both arms included distant metastasis (30.0% vs 14.8%) and death (6.7% vs 7.4%); no patients in either arm experienced regional relapse.

No treatment-related deaths were reported. Patients in the monotherapy and combination arms experienced treatment-emergent death at respective rates of 6.7% and 7.4%.

In terms of safety, common grade 1 or 2 treatment-related adverse effects (TRAEs) in the monotherapy arm included rash (43%), fatigue (37%), and arthralgias (30%). Grade 3 TRAEs consisted of colitis (3%), hyponatremia (3%), anemia (3%), and thrombocytopenia (3%).

In the combination arm, common grade 1 or 2 TRAEs included rash (44%), liver function abnormality (33%), and fatigue (30%). Grade 3 TRAEs consisted of pancreatic insufficiency (3%), diabetic ketoacidosis (3%), hyponatremia (3%), and creatine elevation (3%).

“Toxicities were well balanced between each arm,” Mooradian said in her conclusion. “There were no unusual toxicities, with the majority being grade 1 [or] 2. [There were] uncommon grade 3 events [totaling] approximately 12% in each arm, and, notably, [there were] no toxicities that prevented or delayed surgery.”

Disclosures: Mooradian reported consulting or advisory roles with AstraZeneca, Bristol Myers Squibb Foundation, Istari Oncology, Regeneron, and Xilio Therapeutics. She also has other relationships with Aptitude Health, Curio Science, Dava Oncology, and OncLive/MJH Life Sciences.

Reference

Mooradian M, Karunamurthy A, Wang H, et al. Randomized phase II study of neoadjuvant (neoadj) anti-PD-1 dostarlimab (D) vs. D + anti-TIM-3 cobolimab (C) in high-risk resectable melanoma (mel) (NEO-MEL-T): primary analysis. J Clin Oncol. 2025;43(suppl 17):LBA9504.doi:10.1200/JCO.2025.43.17_suppl.LBA9504