Patients with microsatellite instability–high or mismatch repair–deficient prostate cancer may be more likely to respond to treatment with checkpoint inhibitors compared with those who have high tumor mutational burden.
Andrew T. Lenis, MD
Patients with microsatellite instability (MSI)–high (MSI-H) or mismatch repair–deficient (dMMR) prostate cancer may be more likely to respond to treatment with checkpoint inhibitors compared with those who have high tumor mutational burden (TMB-H), according to retrospective findings presented at the 2021 AUA Annual Meeting.1
Of 21 patients with MSI-H/dMMR prostate cancer who had been treated with immunotherapy, most achieved a partial response (n = 9) or stable disease (SD; n = 10) as their best response. In the 7 patients with prostate cancer whose tumors were TMB-H, 5 patients had SD and 2 patients had disease progression.
“Many of these MSI-H/dMMR patients had prolonged and durable responses; on the other hand, no patient with a TMB-high tumor treated with a checkpoint blockade had a response,” lead study author Andrew T. Lenis, MD, a urology fellow at Memorial Sloan Kettering Cancer Center (MSK), said in a virtual presentation during the conference. “MSI-H/dMMR and TMB-H are rare molecular subgroups; approximately 40% of patients with MSI-H/dMMR prostate cancers respond to immune checkpoint blockade.”
Patients who have MSI-H or dMMR prostate cancer, including those with TMB-H tumors that are microsatellite stable (MSS), are eligible for treatment with the PD-1 inhibitor pembrolizumab (Keytruda). Although MSI-H/dMMR and TMB-H are related, they are of biologically distinct molecular subgroups, Lenis said, adding that both phenotypes lead to hypermutated states that allow for neoantigen targets for checkpoint inhibitor–enhanced immune responses.
“Few MSI-high/dMMR and TMB-high prostate cancers were included in the initial studies and the decision to use immune checkpoint blockade is thus guided by small case series,” Lenis explained. “Our objective was to genomically characterize MSI-high/dMMR and TMB-high prostate cancers and evaluate and compare the responses with immune checkpoint blockade.”
In the analysis, investigators retrospectively identified men with prostate cancer who underwent genomic profiling via MSK-IMPACT. To meet the criteria for genomically-identified MSI-H/dMMR, tumors needed to have a MSISensor score of 10 or greater, or a score that is “indeterminant” between 3 and 10 and harbor a deleterious alteration in MSH2, MSH6, MLH1, or PMS2. TMB-H tumors needed to have at least 10 mutations/megabase.
Investigators identified 2813 patients with prostate cancer who were enrolled on the prospective, phase 3 IMPACT trial (NCT00065442) from April 7, 2015 to February 18, 2021. Sixty-four patients (2.9%) were deemed to have MSI-H/dMMR tumors, 32 (1.4%) had TMB-H tumors, and 2146 (95.9%) had MSS/TMB-low (TMB-L) tumors. Patients with inadequate samples were excluded from the analysis.
Across all patients analyzed, the median age was 62 years (range, 56-68), most patients (83.7%) were White, and more than half (53.8%) were never smokers. Patients had either grade 1 (8.2%), 2 (14.1%), 3 (16.1%), 4 (18.7%), or 5 (34.9%) disease; 8.1% of patients’ tumor grades were not available. The prostate-specific antigen (PSA) level at diagnosis was less than 4 (11.2%), between 4 and 10 (39.1%), between 10 and 20 (16.4%), or greater than 20 (27.6%); 5.7% of PSA levels were not identified.
Furthermore, 65.9% of patients had stage N0M0 disease, followed by those with M1 (27.5%) and N1M0 (6.5%) stage disease.
Although no significant differences were observed in age and race between the 3 cohorts, patients with MSI-H or TMB-H tumors were more likely to present at grade 5 at 56.2% and 53.1%, respectively, compared with 34% of those with MSS disease. N1M0 and M1 stage disease were also more prevalent in MSI-H (9.4% and 34.4%, respectively) and TMB-H tumors (15.6% and 18.8%) vs in those with MSS/TMB-L tumors (6.2% and 27.4%).
Lenis added that patients with MSI-H/TMB-H tumors also had higher MSISensor scores and TMB scores.
When comparing responses between the MSI-H/dMMR and TMB-H/MSS subgroups, data showed that although the PSA50 responses appeared similar between the 2 groups, only patients in the MSI-H/dMMR cohort experienced a response to immune checkpoint blockade.
Additionally, the median radiographic progression-free survival was 41 months for those in the MSI-H/dMMR cohort compared with 7 months for patients with TMB-H disease (P <.01).
Lenis concluded that despite the limited sample size in the analysis, MSI-H/dMMR prostate cancers are more frequent responders, coupled with more durable responses, to immunotherapy compared with TMB-H prostate cancers.
“Genomic comparisons between responders and non-responders to immune checkpoint blockade is ongoing,” he added.