TMB-H Status Linked With High Responses to Dostarlimab in Endometrial Cancer

Patients with endometrial cancer who had a high tumor mutational burden (TMB-H) experienced high responses to single-agent dostarlimab-gxly (Jemperli), irrespective of mismatch repair (MMR) or microsatellite stability (MSI) status, according to data from a post-hoc analysis of the phase 1 GARNET trial (NCT02715284) presented during the 2021 ESMO Congress.¹

Dostarlimab elicited an overall response rate (ORR) of 44.9% (95% CI, 34.8%-55.3%) in patients with TMB-H status (n = 44/98) vs 13.0% (95% CI, 8.1%-19.3%) in those with TMB-low (TMB-L) status (n = 20/154).

In those with TMB-H and mismatch repair deficient (dMMR)/microsatellite instability­–high (MSI-H) status (n = 39/87), the ORR with the agent was 44.8% (95% CI, 34.1%-55.9%) vs 23.1% (95% CI, 5.0%-53.8%) in those with dMMR/MSI-H and TMB-low status (n = 3/13). In those with TMB-H status and mismatch repair proficient (MMRp)/microsatellite stable (MSS) disease, the ORR with the agent was 45.5% (95% CI, 16.7%-76.6%) vs just 12.1% (95% CI, 7.2%-18.6%) in those with TMB-L status and MMRp/MSS disease (n = 17/141).

“Notably, the ORR [experienced by] patients with MMRp and TMB-H endometrial cancer was comparable to the ORR of those with TMB-H and dMMR/MSI-H [disease],” lead study author Ana Oaknin MD, principal investigator of Vall d'Hebron Institute of Oncology’s Gynecological Malignancies Group, noted in a poster presentation on the data. “TMB-H status in patients with MMRp endometrial cancer was not due to MSI-H or POLe-mutated status.”

Dostarlimab is a humanized PD-1 monoclonal antibody that blocks interaction with the PD-1 ligands, PD-L1, and PD-L2. In April 2021, the FDA granted an accelerated approval to the agent for use as a monotherapy in adult patients with dMMR recurrent or advanced endometrial cancer that has progressed on or after a platinum-containing regimen.² In the same month, the agent was given the green light in the European Union to be used as a single agent in patients with dMMR/MSI-H endometrial cancer that has progressed on or following a platinum-containing regimen.³

Both of the regulatory decisions were based on data from the single-arm phase 1 GARNET trial, which evaluated dostarlimab monotherapy in multiple tumor types, including 2 cohorts of patients with endometrial cancer.

TMB has been previously studied as a predictor of response to anti–PD-1 therapy, but data in endometrial cancer are limited. As such, in the analysis, investigators sought to examine the antitumor activity of dostarlimab in patients with dMMR/MSI-H or MMRp/MSS endometrial cancer by TMB status.

The GARNET study was conducted in multiple parts, with part 1 serving as the dose-finding portion, and part 2A comprised of a fixed-dose safety run-in. Part 2B of the study included expansion cohorts of patients with dMMR/MSI-H (cohort A1; n = 129) and MMRp/MSS endometrial cancer (cohort A2; n = 161).

To be eligible for enrollment on cohorts A1 and A2, patients must have progressed on or after platinum doublet therapy, have received 2 or fewer prior lines of treatment for recurrent or advanced disease, and had measurable disease at baseline. Patients also needed to be naïve to anti–PD-L1 agents.

Patients could be screened based on local MMR/MSI testing results through the use of immunohistochemistry (IHC), polymerase chain reaction, or next-generation sequencing (NGS) performed in a certified local laboratory. However, patient cohort assignment was based on MMR IHC results.

Study participants received 500 mg of intravenous (IV) dostarlimab every 3 weeks for 4 cycles, and then 1000 mg every 6 weeks thereafter, until disease progression.

The primary end points of the study were ORR and duration of response per RECIST v1.1 criteria and blinded independent central review (BICR).

In the post-hoc analysis, and TMB status was an exploratory biomarker determined using the Foundation One test. TMB-H status was defined as having 10 or more mutations/megabase, and TMB-L status was defined as having less than 10 mutations/megabase.

As of a data cutoff date of March 1, 2020, a total of 129 patients with dMMR/MSI-H endometrial cancer and 161 patients with MMRp/MSS disease were enrolled and received treatment with dostarlimab. These populations comprise the safety populations of cohorts A1 and A2, respectively.

The primary efficacy population included those with 24 weeks or more of follow-up time in the study, and who had at least 1 measurable lesion at baseline, per BICR. Overall, 105 patients with dMMR/MSI-H disease and 156 patients with MMRp/MSS disease were evaluable for the analysis.

The median age of patients was 64 years (range, 30-86). Most patients with dMMR/MSI-H status had FIGO stage I or II disease at primary diagnosis (54.3%); the majority of those in the MMRp/MSS cohort had FIGO stage III or IV disease (62.8%). In terms of histology, most patients in cohort A1 presented with endometrioid carcinoma grade 1/2 (67.6%), and most patients in cohort A2 presented with serous disease (37.8%). In both cohorts A1 and A2, the majority of patients had a least 1 prior line of therapy (62.9% and 46.2%, respectively) and had received prior radiation (70.5% and 60.9%, respectively).

Among those with dMMR/MSI-H endometrial cancer (n = 105), 82.9% (n = 87) had TMB-H status, 12.4% (n = 13) had TMB-L status, and 4.8% (n = 5) had TMB status that could not be determined. For those with MMRp/MSS disease (n = 156), 90.4% (n = 141) had TMB-L status, 7.0% (n = 11) had TMB-H status, and 2.6% (n = 4) had TMB status that could not be determined.

The median number of mutations/megabase in patients with dMMR/MSI-H endometrial cancer (n = 100) was 20.17 (range, 2.52-428.69), compared with 3.78 (range, 0-83.22) in those with MMRp/MSS disease (n = 152).

The data cutoff date for the prespecified interim analysis was March 1, 2020, and additional data showed that TMB-high status and dMMR/MSI-H status had substantial overlap in the patient populations, and TMB-H endometrial cancers and dMMR/MSI-H endometrial cancers experienced similar response rates with dostarlimab, at 44.9% and 44.8% (n = 47/105; 95% CI, 35.0%-54.8%), respectively.

Moreover, the ORR achieved with the agent in patients with MMRp disease and TMB-H status was notably comparable to that ORR experienced by those with dMMR/MSI-H and TMB-H status, at 45.5% and 44.8%, respectively.

“Of the 11 patients with MMRp/TMB-H endometrial cancer, all had MSI and POLe test results available,” Oaknin said. “One patient was MSI-H, according to Foundation Medicine NGS testing, and 1 patient had an intermediate MSI score; 9 were MSS. None of the 11 patients had a POLe exonuclease domain mutation identified.”

The study was not powered to assess antitumor activity by TMB status, and interpretation is limited by the small number of patients in each subgroup, Oaknin concluded.

References

1. Oaknin A, Gilbert L, Tinker AV, et al. Analysis of antitumor activity of dostarlimab by tumor mutational burden (TMB) in patients (pts) with endometrial cancer (EC) in the GARNET trial. Presented at: 2021 ESMO Congress; September 16-21, 2021; Virtual. Abstract 76P.
2. FDA grants accelerated approval to dostarlimab-gxly for dMMR endometrial cancer. News release. FDA. April 22, 2021. Accessed September 16, 2021. https://bit.ly/3Almwtk
3. European Commission approves GSK’s Jemperli (dostarlimab), the first anti-PD-1 therapy approved for recurrent or advanced endometrial cancer. News release. GlaxoSmithKline, April 23, 2021. Accessed September 16, 2021. https://bit.ly/32HHMdC