MLH1 and MSH2 expression in endometrial cancer - microscopic and computer assessment of immunohistochemical method

We aimed to compare MLH1 and MSH2 protein expression and Mismatch repair (MMR) status with clinical data: diagnosis, grading, and staging. Moreover, we wanted to assess any correlation between two immunohistochemical assessments: classic microscopic and computer analysis performed by a calibrated program.

Our studies were conducted on 95 cases of endometrial cancer. For each, we performed H+E staining and immunohistochemistry (IHC) of two MMR status proteins: MLH1 and MSH2. Two independent researchers assessed IHC on a 0 to ++++ scale. We classified cases as MMR-deficient based on the microscopic assessment of the absence of expression of at least one protein. For computer analysis, we used Olympus cellSens software to measure the positive IHC reaction area in µm² in five fields of vision.

Despite using two different assessment methods, we did not identify any statistically significant relationship between diagnosis, grading, staging, MMR protein expression, and MMR status. However, we found a strong correlation between computer analysis and semi-quantitative microscopic assessment (r=0.59 for MLH1 and r=0.76 for MSH2; p<0.001 for both). Furthermore, we revealed that computer measurement of the expression area could be a good objective prediction test for microscopic analysis (p<0.001).

We did not find a relationship between MMR status and grading, staging, or diagnosis. However, we present a novel approach to immunohistochemical assessment using computer analysis. It allows us to carry out more objective and accurate studies with the IHC method.