Endometrioid Cancer Associated With Endometriosis: From the Seed and Soil Theory to Clinical Practice

Abstract

Endometriosis is a benign condition characterized by the presence of ectopic endometrial tissue. It is still debated whether endometriosis is a disease that can predispose to the pathogenesis of endometrial cancer outside the uterus. Deficiencies in mismatch repair (MMR) genes are a known risk factor for developing endometrioid cancer. Starting from two cases of patients with abnormal MMR endometrioid carcinoma of the uterus and synchronous endometrioid carcinoma in non-ovarian and ovarian endometriosis, we performed a somatic mutation profile and phylogenetic analysis of the lesions in order to identify if they were metastasis or primary de novo tumors. In the first case, we identified de novo activating mutations in PIK3CA and KRAS in endometrioid cancer lesions but not in endometriosis. Although the acquisition of a de novo mutation in ESR1 and a decrease in mutant allele fraction (MAF) from the endometrial tumor to the localizations in the endometriosis lesions, the clonal relationship was confirmed by the limited number of heteroplasmic mutations in D-loop mitochondrial DNA region. In the other case, the clonal behavior was demonstrated by the overlap of MAF at each site. Our data support the hypothesis of a retrograde dissemination of tumor cells, moving from the primary carcinoma in the endometrium to ectopic sites of endometriosis where localizations of tumor arise.

Keywords: endometrioid adenocarcinoma of the endometrium; endometriosis; mismatch repair (MMR) deficiency; tumor dissemination; uterine carcinoma.

Figure 1

Mutations on representative cancer-associated genes of case 1 (A) and case 2 (B). Each color represents a different sample. Triangles represent the mutant allele frequency (MAF), corresponding to those of the vertical axis.

Figure 2

Phylogenetic tree based on Multiple Alignment using Fast Fourier Transform (MAFFT) with the unweighted pair group method with arithmetic mean and the Jukes–Cantor substitution model. All the samples of case 1 are very close to each other, as we found a limited number of mutations in heteroplasmy, 4 of which in the sample of the peritoneum and 2 in endometriosis sample.

Figure 3

The “seed and soil” hypothesis: retrograde flow of endometrial cells already harboring cancer-associated mutation moving from the primary carcinoma in the endometrium to ectopic sites of endometriosis where tumor localizations arise.