Artificial intelligence for detection of microsatellite instability in colorectal cancer-a multicentric analysis of a pre-screening tool for clinical application.

BACKGROUND: Microsatellite instability (MSI)/mismatch repair deficiency (dMMR) is a key genetic feature which should be tested in every patient with colorectal cancer (CRC) according to medical guidelines. Artificial intelligence (AI) methods can detect MSI/dMMR directly in routine pathology slides, but the test performance has not been systematically investigated with predefined test thresholds. METHOD: We trained and validated AI-based MSI/dMMR detectors and evaluated predefined performance metrics using nine patient cohorts of 8343 patients across different countries and ethnicities. RESULTS: Classifiers achieved clinical-grade performance, yielding an area under the receiver operating curve (AUROC) of up to 0.96 without using any manual annotations. Subsequently, we show that the AI system can be applied as a rule-out test: by using cohort-specific thresholds, on average 52.73% of tumors in each surgical cohort [total number of MSI/dMMR = 1020, microsatellite stable (MSS)/ proficient mismatch repair (pMMR) = 7323 patients] could be identified as MSS/pMMR with a fixed sensitivity at 95%. In an additional cohort of N = 1530 (MSI/dMMR = 211, MSS/pMMR = 1319) endoscopy biopsy samples, the system achieved an AUROC of 0.89, and the cohort-specific threshold ruled out 44.12% of tumors with a fixed sensitivity at 95%. As a more robust alternative to cohort-specific thresholds, we showed that with a fixed threshold of 0.25 for all the cohorts, we can rule-out 25.51% in surgical specimens and 6.10% in biopsies. INTERPRETATION: When applied in a clinical setting, this means that the AI system can rule out MSI/dMMR in a quarter (with global thresholds) or half of all CRC patients (with local fine-tuning), thereby reducing cost and turnaround time for molecular profiling.

Lifestyle factors and risk of sporadic colorectal cancer by microsatellite instability status: a systematic review and meta-analyses.

Introduction: The association of lifestyle factors with molecular pathological subtypes of colorectal cancer (CRC), such as microsatellite instability (MSI), could provide further knowledge about the colorectal carcinogenic process. The aim of this review was to evaluate possible associations between lifestyle factors and risk of sporadic CRC by MSI status. Methods: PubMed and Web of Science were searched for studies investigating the association between alcohol, body mass index, dietary fiber, hormone replacement therapy (HRT), non-steroidal anti-inflammatory drugs, physical activity, red meat, smoking, or statin use, with MSI-high (MSI-H) and microsatellite stable (MSS) CRC. Meta-analyses were carried out to calculate summary relative risks (sRR). Results: Overall, 31 studies reporting on the association between lifestyle factors and CRC according to MSI status were included in this review. Ever smoking was associated with MSI-H (sRR = 1.62; 95% CI: 1.40-1.88) and MSS/MSI-low CRC (sRR = 1.10; 95% CI: 1.01-1.20), but the association was significantly stronger for MSI-H CRC. The use of HRT was associated with a 20% decrease (sRR = 0.80; 95% CI: 0.73-0.89) in the risk of MSS CRC, but was not associated with MSI-H CRC. An increase in body mass index per 5 kg/m2 was equally associated with MSS and MSI-H CRC (sRR = 1.22, in both cases), but was statistically significant for MSS CRC only (95% CI: 1.11-1.34 and 0.94-1.58, respectively). Limited evidence for associations between other lifestyle factors and CRC by MSI status exists. Conclusions: Lifestyle factors, such as HRT and smoking are differentially associated with the risk of MSI-H and MSS CRC. Further research on associations of lifestyle factors and CRC subtypes is necessary to provide a better understanding of the CRC disease pathway.