MMR-MSI TESTING PATTERNS IN A REAL-WORLD CANADIAN SETTING: INSIGHTS FROM AN AI-ENABLED ANALYSIS

Author(s)

François-Xavier Houde, PharmD, MSc1, Ruth Moulson, MPH2, Christopher Pettengell, MD2, Si Li, PharmD, MBA, MS3, Niloufar Zarghami, PhD1, Peter Kaskel, MBA, MD4, Christine Brezden-Masley, PhD, MD5.
1Merck Canada Inc., Kirkland, Montreal, QC, Canada, 2Healwell AI, Toronto, ON, Canada, 3Merck, Rahway, NJ, USA, 4MSD SHARP& DOHME GmbH, Munich, Germany, 5Mount Sinai Hospital, Toronto, ON, Canada.
OBJECTIVES: Mismatch repair deficiency (dMMR) and microsatellite instability-high (MSI-H) predict response to immunotherapy, but Canadian real-world data on prevalence, testing, and treatment pathways across solid tumours are limited.
METHODS: Retrospective observational study using the validated DARWEN™ AI engine to extract structured anonymized data from EHRs at Mount Sinai Hospital and Sunnybrook Health Sciences Centre (Toronto). We describe demographics, MMR/MSI testing timing and modality, dMMR-MSI-H prevalence, and systemic treatment patterns in pre-specified solid tumours, selected based on funding by Cancer Care Ontario/Ontario Health in 2024.
RESULTS: Among 4,606 eligible patients, 43.7% underwent MMR-MSI testing between 2019-2023. Testing varied widely by tumour type, ranging from 76% in endometrial (864 on 1,137) vs. 0% in less common tumours such as small bowel (0 on 64). Both overall testing and timely testing (≤30 days from diagnosis) improved over time, with 56.1% of colorectal cancer patients being tested <30 days in 2019 compared to 75.3% by 2023. Among patients with conclusive MMR-MSI test results (n=1,791), dMMR/MSI-H prevalence was 17.6%, higher in early-stage vs. late-stage disease (20.8% vs. 12.3%). Prevalence was highest in endometrial cancer (28.6%) and colorectal cancer (10.7%), regardless of stage. After a confirmed dMMR-MSI-H result, 18.0% of colorectal and 24.0% of endometrial cancer patients received systemic therapy (not including surgery or radiotherapy) during follow-up; chemotherapy was the most common therapy.
CONCLUSIONS: This study suggests that routine MMR-MSI testing increases over time, although variation exist across tumour types. Biomarker-driven therapy capture may be improved by standardized reflex testing. Findings of this study are limited by the retrospective design and possible incomplete capture of testing performed outside participating institutions. Because rare mutations occur infrequently, studies aimed at understanding their clinical significance require large patient cohorts to achieve reliable results. The integration of AI-enabled real-world data facilitates the execution of such studies and may identify and mitigate care gaps.

Conference/Value in Health Info

2026-11, ISPOR Europe 2026, Vienna, Austria

Value in Health, Volume 29, Issue 12S

Code

MSR28

Topic

Health Service Delivery & Process of Care, Methodological & Statistical Research, Real World Data & Information Systems

Topic Subcategory

Artificial Intelligence, Machine Learning, Predictive Analytics

Disease

Oncology, Personalized & Precision Medicine, Rare & Orphan Diseases

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×