Dostarlimab Demonstrates Durable Antitumor Activity in Mismatch Repair Proficient/dMMR Endometrial Cancer

Dostarlimab showcased durable antitumor activity in patients with advanced or recurrent DNA mismatch repair deficient (dMMR) and proficient (MMRp) endometrial cancer, with a notable disease control rate (DCR) and a promising safety profile, according to updated results from the phase 1 GARNET trial (NCT02715284) that were presented during the 2020 ESMO Virtual Congress.¹

Results showed that the investigational PD-1 antibody elicited an objective response rate (ORR) of 44.7% in patients with dMMR disease and 13.4% in those with MMRp disease. In the dMMR cohort (n = 103), 11 complete responses (CRs) were observed, along with 35 partial responses (PRs). Thirteen patients achieved stable disease (SD), while 39 patients experienced disease progression. In the MMRp cohort (n = 142), 3 patients had CRs, 16 had PRs, 31 achieved stable disease, and 77 patients experienced progressive disease.

The median duration of response (DOR) was not reached in the dMMR cohort (range, 2.63-28.09+) nor the MMRp cohort (range, 1.54+ to 30.36+). Additionally, DCRs with dostarlimab in the dMMR cohort and the MMRp cohorts were 57.3% (95% CI, 47.2-67.0) and 35.2% (95% CI, 27.4-43.7), respectively.

“Dostarlimab demonstrated durable antitumor activity in both dMMR and MMRp advanced and recurrent endometrial cancer,” Ana Oaknin, MD, PhD, lead study author and head of the Gynecologic Cancer Program at Vall d’Hebron Institute of Oncology, in Barcelona, Spain, said in a presentation during the congress. “The dMMR status determined by immunohistochemistry (IHC) was associated with a higher response rate. Dostarlimab also demonstrated a notable DCR in patients with MMRp endometrial cancer, a cohort that was comprised of a higher percentage of patients with Type II disease, which is historically associated with a worse prognosis.”

In the single-arm, phase 1 study, investigators set out to evaluate dostarlimab monotherapy in patients with advanced solid tumors. Part 2B of the trial was comprised of 5 expansion cohorts: dMMR/microsatellite instability–high (MSI-H) endometrial cancer (cohort A1), MMRp (cohort A2), non–small cell lung cancer (cohort E), dMMR/MSI-H non-endometrial cancer (cohort F), and platinum-resistant ovarian cancer without BRCA mutations (cohort G).²

At the 2020 ESMO Virtual Congress, Oaknin shared data from the 2 endometrial-cancer cohorts: dMMR and MMRp. Mismatch repair status was determined by IHC. In part 2B of the trial, the PD-1 inhibitor was given at the recommended dose determined from parts 1 and 2A of the trial: 500 mg intravenously (IV) every 3 weeks for 4 cycles followed by 1000 mg IV every 6 weeks until progressive disease.

To be eligible for participation, patients had to have progressed on or following treatment with a platinum-doublet therapy, they had to have received 2 or fewer prior lines of therapy for recurrent or advanced disease, and they had to have measurable disease at baseline. Additionally, patients could not have received previous PD-1/PD-L1 inhibitor therapy, and they had to undergo screening based on local MMR/MSI testing performed in a certified local laboratory and have confirmed MMR IHC results.

The primary end points of the trial were ORR and DOR.

Previous results from the GARNET trial were presented during the 2020 Society of Gynecologic Oncology Virtual Congress. In a cohort of 71 patients with endometrial carcinoma, dostarlimab elicited an ORR of 42% (95% CI, 31-55); this comprised a CR rate of 13% and a PR rate of 30%.3 The DCR with the agent had been 58%, and at a median follow-up of 11.2 months, the median DOR had not yet been reached.

At the data cutoff date of March 1, 2020, a total of 126 patients in the dMMR cohort and 145 patients in the MMRp cohort were enrolled to the trial and received treatment with dostarlimab; these patients comprise the safety population of the trial. To be eligible for inclusion in the efficacy population, participants had to have at least 6 months of follow-up time on the study with at least 1 measurable lesion at baseline. A total of 103 patients in the dMMR cohort and 142 in the MMRp cohort satisfied those criteria, according to Oaknin.

The demographic and baseline characteristics between the 2 cohorts proved to be similar, with the most notable difference related to histology. In the dMMR cohort, the most common histology was endometroid carcinoma type I at 68.0% (n = 70) and 23.2% (n = 33) in the MMRp cohort. The majority of participants in the latter cohort, or 76.8% (n = 109), had type II endometrial carcinoma compared with 31.1% (n = 32) in the dMMR cohort. Of the 109 MMRp patients, 54 had serous histology.

Additionally, 63.1% (n = 65) received 1 prior line of therapy in the dMMR cohort and 45.8% (n = 6%) in the MMRp cohort; 26.2% (n = 27) and 43.7% (n = 62), respectively, received 2 prior lines; and 10.7% (n = 11) and 10.6% (n = 15), respectively, received 3 or more prior lines of treatment.

Additional results from the trial showed that the Kaplain Meier–estimated probability of patients remaining in response in the dMMR and MMRp cohorts at 6 months was 97.8% and 83.0%, respectively. At 12 months, these estimated rates were 90.6% and 61.3%, respectively, and at 18 months, they were 79.2% and 61.3%, respectively.

The median follow-up time in the dMMR cohort was 16.3 months, while the follow-up time in the MMRp cohort was 11.5 months. In the dMMR cohort, 89% of patients (n = 41/46) continued to respond to treatment at the time of data cutoff; this was also true for 63.2% (n = 12/19) of those in the MMRp cohort. “The duration of response among responders to dostarlimab was quite long and the majority of those who responded remained in response at the time of data cutoff,” noted Oaknin.

Regarding safety, any-grade treatment-emergent adverse effects (TEAEs) occurred in 95.2% (n = 120) of patients in the dMMR cohort (n = 126) and 100% of those in the MMRp cohort (n = 145). Grade 3 or higher TEAEs were observed in 48.4% (n = 61) and 55.9% (n = 81) of those in the dMMR and MMRp cohorts, respectively.

Any-grade toxicities determined to be related to study treatment were reported in 63.5% (n = 80) of those in the dMMR cohort and 71.7% (n = 104) in the MMRp cohort. Grade 3 or higher treatment-related AEs (TRAEs) occurred in 13.5% (n = 17) and 19.3% (n = 28) of those in the dMMR and MMRp cohorts, respectively. Serious toxicities related to treatment were reported in 9.5% (n = 12) and 9.0% (n = 13) of patients, respectively.

The 3 most commonly experienced TRAEs of any grade included fatigue (13.5% in dMMR and 20.7% in MMRp), diarrhea (15.9% and 13.1%, respectively), and nausea (12.7% and 14.5%). Grade 3 or higher TRAEs included anemia (4.0% and 1.4%), alanine aminotransferase (ALT; 1.6% and 1.4%), and diarrhea (1.6% and 1.4%).

The top 3 any-grade immune-related TRAEs included hypothyroidism (5.6% and 7.6%), diarrhea (4.8% and 3.4%), and increased aspartate aminotransferase (AST; 1.6% and 2.8%). Grade 3 or higher immune-related TRAEs included increased ALT (1.6% and 1.4%), diarrhea (1.6% and 1.4%), and increased amylase (0.8% and 1.4%).

Four percent of patients (n = 5) in the dMMR cohort experienced TRAEs that resulted in discontinuation of treatment and 6.9% (n = 10) of patients in the MMRp cohort. These toxicities included increased ALT (0.8% in dMMR and 1.4% in MMRp), increased AST (0.8% and 0.7%, respectively), and increased transaminases (1.6% and 0%). Notably, no deaths related to study treatment were reported.

“No new safety signals were detected and only 5% of patients discontinued dostarlimab due to TRAEs,” concluded Oaknin. “Finally, I must highlight that these cohorts are the largest prospective evaluation of a PD-L1/PD-1 therapy in endometrial carcinoma to date and follow-up is ongoing.”

References

1. Oaknin A, Gilbert L, Tinker AV, et al. Safety and antitumor activity of dostarlimab in patients with advanced or recurrent DNA mismatch repair deficient (dMMR) or proficient (MMRp) endometrial cancer: results from GARNET. Presented at: 2020 ESMO Virtual Congress; September 19-21, 2020; Virtual. Abstract LBA36.
2. GSK presents new data from the GARNET study demonstrating potential of dostarlimab to treat a subset of women with recurrent or advanced endometrial cancer. News release. GlaxoSmithKline plc. April 23, 2020. Accessed September 19, 2020. https://bit.ly/33N1C7n.