Mismatch Repair Deficiency in Ovarian Carcinoma: Frequency, Causes, and Consequences

Am J Surg Pathol. 2020 May;44(5):649-656. doi: 10.1097/PAS.0000000000001432.

Abstract

Mismatch repair deficiency (MMRD) is involved in the initiation of both hereditary and sporadic tumors. MMRD has been extensively studied in colorectal cancer and endometrial cancer, but not so in other tumors, such as ovarian carcinoma. We have determined the expression of mismatch repair proteins in a large cohort of 502 early-stage epithelial ovarian carcinoma entailing all the 5 main subtypes: high-grade serous carcinoma, endometrioid ovarian carcinoma (EOC), clear cell carcinoma (CCC), mucinous carcinoma, and low-grade serous carcinoma. We studied the association of MMRD with clinicopathologic and immunohistochemical features, including tumor-infiltrating lymphocytes in EOC, the histologic type in which MMRD is most frequent. In addition, MLH1 promoter methylation status and massive parallel sequencing were used to evaluate the proportion of sporadic and Lynch syndrome-associated tumors, and the most frequently mutated genes in MMRD EOCs. MMRD occurred only in endometriosis-associated histologic types, and it was much more frequent in EOC (18%) than in CCC (2%). The most frequent immunohistochemical pattern was loss of MLH1/PMS2, and in this group, 80% of the cases were sporadic and secondary to MLH1 promoter hypermethylation. The presence of somatic mutations in mismatch repair genes was the other mechanism of MMRD in sporadic tumors. In this series, the minimum estimated frequency of Lynch syndrome was 35% and it was due to germline mutations in MLH1, MSH2, and MSH6. ARID1A, PTEN, KTM2B, and PIK3CA were the most common mutated genes in this series. Interestingly, possible actionable mutations in ERRB2 were found in 5 tumors, but no TP53 mutations were detected. MMRD was associated with younger age and increased tumor-infiltrating lymphocytes. Universal screening in EOC and mixed EOC/CCC is recommended for the high frequency of MMRD detected; however, for CCC, additional clinical and pathologic criteria should be evaluated to help select cases for analysis.

Publication types

  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Age Factors
  • Biomarkers, Tumor / deficiency
  • Biomarkers, Tumor / genetics*
  • Carcinoma / genetics*
  • Carcinoma / mortality
  • Carcinoma / pathology
  • Carcinoma / therapy
  • Colorectal Neoplasms, Hereditary Nonpolyposis / genetics*
  • Colorectal Neoplasms, Hereditary Nonpolyposis / mortality
  • Colorectal Neoplasms, Hereditary Nonpolyposis / pathology
  • Colorectal Neoplasms, Hereditary Nonpolyposis / therapy
  • DNA Methylation
  • DNA Mismatch Repair*
  • DNA Mutational Analysis
  • DNA Repair Enzymes / deficiency
  • DNA Repair Enzymes / genetics*
  • Disease Progression
  • Female
  • Genetic Predisposition to Disease
  • High-Throughput Nucleotide Sequencing
  • Humans
  • Immunohistochemistry
  • Lymphocytes, Tumor-Infiltrating / pathology
  • Middle Aged
  • Mutation
  • Ovarian Neoplasms / genetics*
  • Ovarian Neoplasms / mortality
  • Ovarian Neoplasms / pathology
  • Ovarian Neoplasms / therapy
  • Phenotype
  • Progression-Free Survival
  • Prospective Studies
  • Registries
  • Risk Factors
  • Spain
  • Time Factors

Substances

  • Biomarkers, Tumor
  • DNA Repair Enzymes