OBJECTIVES: MSI-H/ dMMR is an actionable biomarker in the treatment of several tumors. Given data paucity, the study assessed the frequency of and testing patterns for MSI-H/dMMR across various solid tumors in a real-world community oncology setting in the US.

METHODS: A retrospective, electronic health record-based study was conducted among patients ≥18 years diagnosed with stage IV tumors (endometrial, colorectal, cervical, ovarian, pancreatic, prostate, renal, gastric, esophageal, breast, cholangiocarcinoma, small intestine, brain, and soft tissue sarcoma) between 01-Jan-2010 and 29-Feb-2020 and receiving care in the US Oncology Network. Testing details and demographic information were abstracted from structured fields of electronic health records.

RESULTS: A total of 8,899 stage IV solid tumor patients were reported as tested for MSI or MMR across 14 tumor types. The proportion (95% CI) with MSI-H and/or dMMR was 15.0% (14.3-15.8) for all tumor types and ranged from 3.2% (0.1-16.7) for soft tissue sarcomas to 28.9% (23.6-34.8) for endometrial cancers. Overall, 7,694 and 5,536 patients were tested for MSI and MMR with 8.5% (7.9-9.2) and 16.6% (15.7-17.7) reporting with MSI-H and dMMR, respectively. Median (range) age of MSI-H/dMMR patients was 65.8 (21.8-90+) years with 53.3% female and 71.9% white. MSI or MMR testing rates increased over time: 4.2% (2010-2016) vs. 17.7% (2017-2020).

CONCLUSIONS: Frequency of MSI-H/dMMR varied across tumors assessed with endometrial cancer having the highest proportion. Increases in MSI or MMR testing signifies uptake of MSI/MMR status evaluation/consideration among community oncologists, likely prompted by the 2017 approval of pembrolizumab for MSI-H/dMMR solid tumors; however, testing is still less than optimal. Though measuring the same biologic phenomenon, the higher proportion of patients tested for dMMR versus MSI-H suggests divergent testing practices by tumor type and institution, and the need for standardization. Data sources and testing practices should be considered when interpreting the results.