Pitfalls in MLH1 promoter methylation assessment, including POLE^mut/MLH1^meth endometrial adenocarcinoma

MLH1 promoter methylation status may serve as an important diagnostic, prognostic, and predictive biomarker in management of mismatch repair (MMR)-deficient cancers. A new commercial assay for detection of MLH1 promoter hyper-methylation (EntroGen), which interrogates regions C and D, was assessed for its performance characteristics. False positive results were obtained in 5/21 non-lesional cases, with signals limited to region C. Three were explained by overloaded reactions. Two unexplained cases were both muscle samples (2/6 muscle samples, one each of skeletal and smooth). The assay was 100 % concordant (52/52) for lesional samples with expected MLH1 promoter methylation status. These included two exceptional cases-one Lynch-associated, and one POLE-mutated endometrial carcinoma; thus expanding the spectrum of extreme cases, and demonstrate neither germline or somatic NGS results completely rule out MLH1 promoter methylation, and vice versa. The POLE^mut/MLH1^meth carcinoma was notable for molecular features in keeping with POLE dysfunction, accompanied by multiple, additional genetic lesions in the MMR pathway. Exploring the TCGA dataset, 1/8 cases of POLE (ultramutated) endometrial carcinoma was notable for MLH1 silencing. Comprehensive genomic profiling assay was informative, allowing for correlation of MLH1 methylation and POLE genotype results with tumor mutation burden and mutational signature. Taken together, our data highlight the need for integrated approach in endometrial carcinoma biomarker testing, integrating NGS and MLH1 promoter methylation status, the latter of which benefits from assessing both regions C and D. Finding of MLH1 promoter methylation does not rule out either Lynch syndrome or ultramutated (POLE) carcinoma.