Diagnostic accuracy of immunohistochemistry for mismatch repair proteins as surrogate of microsatellite instability molecular testing in endometrial cancer

Microsatellite instability (MSI) defines one of the four molecular groups of endometrial carcinoma identified by The Cancer Genome Atlas (TCGA). Immunohistochemistry for mismatch repair (MMR) proteins (MLH1, MSH2, MSH6, PMS2) has been proposed as a widely applicable technique to identify this group in the common practice. However, the diagnostic accuracy of such approach has never been calculated. We aimed to assess: 1) the diagnostic accuracy of MMR proteins immunohistochemistry as surrogate of MSI molecular testing in endometrial carcinoma; 2) whether a combination of only two MMR proteins may be used as a still cheaper test. A systematic review and meta-analysis of was performed by searching electronic databases from their inception to September 2019. All studies assessing endometrial carcinoma with both MMR proteins immunohistochemistry and MSI molecular testing were included. Diagnostic accuracy was assessed as sensitivity, specificity, positive and negative likelihood ratios (LR+, LR-), diagnostic odds ratio (DOR) and area under the curve (AUC) on SROC curves. A subgroup analysis was performed for a combination of only two MMR proteins (MLH1-MSH2 vs MSH6-PMS2). Ten studies with 3097 patients were included. Out of these, 1110 were suitable for the meta-analysis. Immunohistochemistry for all the four MMR proteins showed sensitivity = 0.96, specificity = 0.95, LR + =17.7, LR- = 0.05, DOR = 429.77, and high diagnostic accuracy (AUC = 0.988). The combination of MLH1 and MSH2 showed sensitivity = 0.88, specificity = 0.96, LR + =22.36, LR- = 0.15, DOR = 200.69, and high diagnostic accuracy (AUC = 0.9838). The combination of MSH6 and PMS2 showed the same results as the complete panel of four MMR proteins. In conclusion, MMR proteins immunohistochemistry is a highly accurate surrogate of MSI molecular testing in endometrial carcinoma. A combination of MSH6 and PMS2 may allow reducing the cost without decrease in the diagnostic accuracy.