Dostarlimab Provides Survival Benefits in dMMR Endometrial Cancer and Other Solid Tumors

Single-agent dostarlimab-gxly (Jemperli) showcased durable antitumor activity in patients with mismatch repair–deficient (dMMR) endometrial cancer and other solid tumors, with landmark estimates at various time points underscoring stable progression-free survival (PFS) benefits in responders, according to data on key secondary end points of the phase 1 GARNET trial (NCT02715284).

Data from the third interim analysis were presented at the 2022 ESMO Congress. Findings showed that at a median follow-up of 27.7 months, dostarlimab resulted in a median PFS of 6.9 months (95% CI, 4.2-13.6) in 327 evaluable patients with dMMR solid tumors spanning 16 tumor types (n = 327). In patients with dMMR endometrial cancer (n = 141), the median PFS was 5.6 months (95% CI, 4.1-16.6) at a median follow-up of 27.6 months. In dMMR non–endometrial cancer solid tumors (n = 186), the median PFS was 7.0 months (95% CI, 3.6-16.5) at a median follow-up of 29.8 months.

In all patients, the 12-, 24-, and 36-month estimated probability of PFS was 45.8% (95% CI, 40.2%-51.2%), 40.6% (95% CI, 35.0%-46.1%), and 39.7% (95% CI, 33.9%-45.3%), respectively. Among those with endometrial cancer, these rates were 46.0% (95% CI, 37.4%-54.1%), 39.6% (95% CI, 31.2%-48.0%), and 39.6% (95% CI, 31.2%-48.0%), respectively. Those rates were 45.6% (95% CI, 38.2%-52.8%), 41.4% (95% CI, 34.0%-48.7%), and 39.6% (95% CI, 31.8%-47.4%), respectively, in those with other dMMR solid tumors.

The median overall survival (OS) was not yet reached (NR; 95% CI, 31.6-NR) in all patients; this was also true for those with endometrial cancer (95% CI, 25.7-NR) and those with other solid tumors (95% CI, 21.7-NR).

In all patients with dMMR solid tumors, the 12-, 24-, and 36-month estimated probability of survival rates were 70.6% (95% CI, 65.3%-75.3%), 58.4% (95% CI, 52.5%-63.9%), and 55.9% (95% CO, 49.7%-61.7%), respectively. Those rates were 72.9% (95% CI, 64.7%-79.5%), 59.9% (95% CI, 50.8%-67.8%), and 57.7% (95% CI, 48.5%-65.9%), respectively, in those with endometrial cancer. In those with non–endometrial cancer solid tumors, these rates were 68.7% (95% CI, 61.4%-75.0%), 57.3% (95% CI, 49.3%-64.5%), and 54.4% (95% CI, 45.9%-62.2%), respectively.

Although dMMR or microsatellite instability–high (MSI-H) can be observed across solid tumors, it is known that the frequency of these markers varies by disease type. For example, endometrial cancer is known to have a high prevalence of dMMR or MSI-H, with these markers observed in approximately 25% to 30% of patients; this is also true for colorectal cancer (CRC), with these markers noted in approximately 10% to 15% of patients.

Dostarlimab is an anti–PD-1 monoclonal antibody designed to block interaction with PD-L1/2 ligands. The agent is under investigation in the ongoing, multicenter, open-label, single-arm GARNET study, where it is being explored as a potential therapeutic option for patients with advanced or recurrent dMMR solid tumors.

Cohort A1 of the trial included patients with dMMR endometrial cancer who progressed on or after platinum-doublet chemotherapy. Cohort F featured patients with dMMR solid tumors other than endometrial cancer. To participate, those with CRC needed to have progressed on, or been intolerant to, fluoropyrimidine, oxaliplatin, and irinotecan. Those with platinum-resistant ovarian cancer were permitted to receive up to 1 line of systemic therapy after developing platinum resistance.

No patients were allowed to have prior exposure to anti–PD-1/PD-L1 therapy.

Participants were administered 500 mg of intravenous dostarlimab every 3 weeks for 4 cycles, then 1000 mg every 6 weeks thereafter. Treatment continued until disease progression, discontinuation, or withdrawal.

Among all enrolled patients, the median age was 63.0 years (range, 24-85), 71.9% were female, 63.0% were White, and 60.6% of patients had an ECOG performance status of 1. Additionally, 41.9% of patients received 1 prior line of systemic therapy, 36.1% received 2 prior lines, and 22% received at least 3 prior lines. Prior treatment included surgery for 85.3% of patients and radiotherapy for 42.8% of patients.

Of those who were enrolled to the trial, 43.1% had endometrial cancer, 32.1% had CRC , 6.4% had gastric and gastroesophageal junction cancer, 5.8% had small intestinal cancer, 3.4% had pancreatic carcinoma, 3.1% had biliary neoplasm, 2.1% had ovarian cancer, and 4% had a tumor classified as other.

At data cutoff, 232 of 341 total patients who were enrolled and received at least 1 dose of dostarlimab discontinued treatment due to disease progression (n = 131), adverse effects (AEs; n = 47), patient request (n = 9), clinical criteria (n = 28), or another unspecified reason (n = 7). A total of 109 patients remained on treatment.

Updated data indicated that in all patients with dMMR solid tumors, dostarlimab elicited an ORR of 44.0% (95% CI, 38.6%-49.6%), which comprised a complete response (CR) rate of 13.1%, a partial response (PR) rate of 30.9%, and a stable disease (SD) rate of 14.4%; 34.9% of patients experienced progressive disease (PD) and 6.7% were not evaluable. The disease control rate (DCR) was 58.4% (95% CI, 52.9%-63.8%).

Patients with endometrial cancer achieved an ORR of 45.4% (95% CI, 37.0%-54.0%) with dostarlimab. In these patients, the CR, PR, SD, and PD rates were 15.6%, 29.8%, 14.9%, and 36.2%, respectively; 3.5% of patients were not evaluable. In this group, the DCR was 60.3% (95% CI, 51.7%-63.8%). In patients with non–endometrial cancer solid tumors, dostarlimab induced an ORR of 43.0% (95% CI, 35.8%-50.5%). Here, the CR, PR, SD, and PD rates were 11.3%, 31.7%, 14.0%, and 33.9%, respectively; 6.7% of patients were not evaluable. The DCR for this group was 57.0% (95% CI, 49.5%-64.2%).

Additional data showed that the median duration of response (DOR) was not reached in the overall, endometrial cancer, and non–endometrial cancer populations.

In the overall population, the probability of maintaining a response at 6, 12, and 24 months was 95.7% (95% CI, 90.6%-98.0%), 92.4% (95% CI, 86.4%-95.9%), and 84.7% (95% CI, 76.7%-90.2%), respectively. Those rates were 96.7% (95% CI, 87.5%-99.2%), 93.1% (95% CI, 82.7%-97.4%), and 83.4% (95% CI, 70.3%-91.0%), respectively, in the endometrial cancer population, and 94.8% (95% CI, 86.7%-98.0%), 92.0% (95% CI, 83.0%-96.3%), and 86.3% (95% CI, 75.1%-92.8%), respectively, in the non–endometrial cancer population.

Regarding safety, 98.8% of all 341 evaluable patients experienced at least 1 treatment-emergent AE (TEAE) of any grade; 54.8% experienced at least 1 grade 3 or higher TEAE. Any-grade treatment-related AEs (TRAEs) were reported in 71.3% of patients; grade 3 or higher TRAEs occurred in 16.7% of patients.

Moreover, 7.3% of patients discontinued treatment due to TRAEs such as increased alanine aminotransferase (ALT; 1.5%) and pneumonitis (1.2%). TRAEs led to death in 2 patients; 1 patient with biliary neoplasm had hepatic ischemia, and 1 patient with CRC committed suicide.

Immune-related AEs of any grade were reported in 34.9% of patients, with grade 3 or higher IRAEs experienced by 11.4% of patients. Immune-related TRAEs of any grade occurred in 27.0% of patients; 8.8% of cases were grade 3 or higher. Ten percent of patients experienced serious TRAEs.

The most common any-grade TRAEs experienced in at least 10% of patients were asthenia (15.2%), diarrhea (15.0%), pruritis (13.2%), fatigue (12.0%), hypothyroidism (10.3%), and nausea (9.1%). The most frequently experienced grade 3 or higher TRAEs included anemia (2.6%), increased ALT (2.1%), and increased lipase (1.5%).

Immune-related TRAEs of any grade to occur in at least 2% of patients included hypothyroidism (6.2%), increased ALT (4.4%), arthralgia (3.2%), increase aspartate aminotransferase (2.3%), hyperthyroidism (2.3%), pneumonitis (2.3%), pruritis (2.3%), and rash (2.3%). The only grade 3 or higher immune-related TRAE to occur in at least 1% of patients was increased ALT (2.1%).

Reference

André T, Berton-Rigaud D, Curigliano G, et al. Progression-free survival (PFS) and overall survival (OS) in patients with mismatch repair deficient (dMMR) solid tumors treated with dostarlimab in the GARNET study. Ann Oncol. 2022;33(suppl 7):S799-S800. doi:10.1016/j.annonc.2022.07.677