Using gene expression in patients with endometrial intraepithelial neoplasia to assess the risk of cancer

Abstract

Patients diagnosed with an endometrial cancer precursor lesion on biopsy may be found to have endometrial cancer at the time of subsequent surgery. The current study seeks to identify patients with endometrial intraepithelial neoplasia (EIN) on biopsy that may be harboring an occult carcinoma. Immunohistochemical stains for gene loss of expression (LOE) for 6 genes, PTEN, ARID1A, MSH6, MSH2, MLH1, and PMS2, were performed on 113 biopsy specimens with EIN. For the 95 patients with follow-up histology, 40 patients had cancer, 41 had EIN, and 14 had normal endometrium. PTEN LOE was found frequently in both EIN and endometrial cancer, and therefore had low positive predictive value. All specimens with ARID1A, MSH6, MSH2, MLH1, or PMS2 LOE on biopsy were subsequently found to have cancer. LOE of any gene was associated with modest sensitivity (0.78) in identifying patients with endometrial cancer who had EIN on biopsy. Further investigation is warranted to determine if gene LOE is a useful clinical tool when evaluating patients with EIN on biopsy.

Diagnostic and prognostic value of ARID1A in endometrial hyperplasia: a novel marker of occult cancer

“…At the end of the study selection process, seven studies were finally included in the systematic review (Fig. ).…”

“…For the ‘flow and timing’ domain, two studies were considered at unclear risk of bias (loss of patients unexplained or unclear ) and 5 at low risk.…”

Preoperative detection of occult endometrial malignancies in endometrial hyperplasia to improve primary surgical therapy: A scoping review of the literature

“…The results of the scoping review were reported according to the Preferred Reporting Items for Systematic reviews and Meta-criteria and were included in our review. We included 24 primary studies, [9][10][11][13][14][15][16][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36] three meta-analyses, [17][18][19] and one poster presentation. 12 The characteristics of the 28 included studies are shown in Table 2.…”

“…12 The characteristics of the 28 included studies are shown in Table 2. Five studies were multicentric, 17-19, 27,35 five studies were from Turkey, 11,15,20,23,25 four studies were from the USA, 12,13,21,22 two studies were from the UK, 14,16 two studies were from Canada 10,34 and there was one study from each of Australia, 9 South Korea, 24 France, 26 Belgium, 28 Italy, 29 Denmark, 30 Greece, 31 Norway, 32 Taiwan, 33 and China. 36 These studies included the data of 7409 patients and recorded 2377 concurrent EC cases (32.1%).…”

“…Three studies analyzed loss of phosphatase and tensin homolog (PTEN) expression for prediction of coexistent EC. 21,28,31 Only Robbe et al, 28 in a primary study of 39 patients, could establish loss of PTEN expression as a statistically significant predictor of coexistent EC with an OR of 0.67 (95% CI 0.51-0.87).…”