Androgen receptor loss tracks with aggressive endometrial cancer in small study

The findings, published in Diagnostic Pathology, come from a team at Fayoum University Hospital in Egypt that analyzed archival tissue from 40 patients who had undergone total hysterectomies.

The study, led by Alaa saad Abd elhamid Mohamed and colleagues, used immunohistochemistry, a laboratory technique that applies antibodies to thin tissue sections to reveal where a specific protein is located. In this case, the researchers stained for the androgen receptor, a nuclear transcription factor that regulates responses to androgen hormones in female reproductive tissues. A tumor was scored as AR-positive only when more than 10 percent of its neoplastic cells showed distinct brown nuclear staining, detected with a standard polymer-based detection system and DAB chromogen. Benign prostatic hyperplasia tissue served as an external positive control in every staining batch, and negative controls omitted the primary antibody entirely.

The cohort consisted of 40 formalin-fixed, paraffin-embedded specimens: 32 cases (80 percent) of endometrioid endometrial carcinoma and 8 cases (20 percent) of serous carcinoma. Patients had a mean age of 60 years, with a range from 42 to 80. Two independent, board-certified pathologists who were blinded to clinical data established the histological diagnoses according to World Health Organization criteria, and tumors were staged and graded using International Federation of Gynecology and Obstetrics (FIGO) guidelines. For statistical purposes, staging was grouped into low-stage (Stages I and II) and high-stage (Stage III) disease, while grading was grouped into low-grade (Grades 1 and 2) and high-grade (Grade 3) categories.

The headline result was the stark subtype divide. AR expression was detected in 62.5 percent of endometrioid carcinomas but in none of the serous carcinomas, a difference that reached statistical significance at P = 0.002. However, the authors themselves urge caution here: their serous subgroup contained only eight cases, and they note that larger studies have reported considerably higher AR positivity rates in serous tumors, with some documenting expression in 70 to 92 percent of serous cases. The complete absence of AR in their own serous cohort, they write, is almost certainly a limitation of the very small sample size and should be interpreted with extreme caution.

Even more pronounced were the associations with grade and stage. AR positivity was found in 76 percent of low-grade tumors (19 of 25) but in only 6.7 percent of high-grade tumors (1 of 15), a difference with P < 0.001. Similarly, 62.1 percent of low-stage tumors (18 of 29) expressed the receptor, compared with just 18.2 percent of high-stage tumors (2 of 11), at P = 0.013. Patient age at diagnosis, by contrast, showed no statistically significant correlation with AR status (P = 0.206). Taken together, the data suggest that AR expression in this cohort clustered in well-differentiated, early-stage endometrioid disease.

These numbers fit into a broader literature that the authors review in their discussion. Previous investigations have reported overall AR positivity rates ranging widely, from about 20 percent to 85 percent, with recent studies landing near the 50 percent mark: Mahdi and colleagues found 51 percent positivity among 261 cases, Zadeh and colleagues 54 percent among 50 cases, and Masoodi and colleagues 46 percent among 50 cases. A large evaluation of more than 18,000 tumors by Viehweger and colleagues detected AR staining in 53 percent of 200 endometrial carcinoma specimens. The authors attribute the variability across institutions to methodological differences, including different antibody clones, scoring systems, and positivity thresholds such as a 1 percent versus a 10 percent cutoff.

The biological rationale for studying androgen signaling in the endometrium rests on the idea that AR activation may act as a brake on tumor growth. The authors cite in vitro and murine model evidence that dihydrotestosterone suppresses endometrial carcinoma proliferation, migration, and tumor growth through the AR pathway, and molecular data suggesting that AR signaling can competitively antagonize estrogen-driven transcription networks. Large-scale randomization analyses and observational cohorts have also indicated that elevated endogenous testosterone and bioavailable androgens influence the risk profile for developing both endometrioid and non-endometrioid tumors. In postmenopausal women, peripheral intracrine biosynthesis converts circulating precursors into bioactive androgens within the tumor microenvironment, making local androgen signaling potentially relevant precisely in the age group where endometrial cancer is most common.

The study has limitations that the authors lay out explicitly. Beyond the small sample of 40 cases and the underpowered serous subgroup, AR expression was evaluated as a binary variable based on a predefined threshold rather than a validated semi-quantitative scoring system such as the H-score, an approach that may obscure biologically meaningful gradients of receptor expression. The study also relied on traditional morphologic classification without TCGA molecular subgrouping into POLE-ultramutated, mismatch repair-deficient, p53-abnormal, and no-specific-molecular-profile categories, which limits the contemporary clinical relevance of the findings. Critical variables including patient BMI, adjuvant treatments, recurrence rates, and survival outcomes were not available, so no claims about prognostic value can be made from this cohort alone. Because p53 immunohistochemistry was not performed, the authors cannot exclude overlap between AR-positive cases and the subset of histologically low-grade endometrioid tumors that harbor p53 abnormalities and fall into a high-risk category.

What the study does establish is a hypothesis-generating association: in this Egyptian cohort, AR expression marked endometrioid tumors with favorable clinicopathological parameters, and its loss coincided with high grade and advanced stage. The authors suggest that integrating AR immunohistochemistry into routine pathology panels could lay the groundwork for future clinical trials of targeted antiandrogen therapies, while stressing that future research should prioritize large-scale, prospective, multi-center studies, standardized IHC scoring criteria, and multi-marker panels combining AR with estrogen and progesterone receptors to capture the full hormonal interplay in endometrial carcinoma. For now, the 50 percent figure and the sharp grade-related gradient join a growing, if inconsistent, body of evidence that the androgen receptor deserves a closer look in endometrial cancer biology.

Subject of Research: Androgen receptor expression in endometrial carcinoma and its association with histological subtype, grade and stage

Article Title: Immunohistochemical evaluation of androgen receptor expression in endometrial carcinoma and its clinicopathological significance

Article References: Mohamed, A. S. A. E., Hosni, H. A., Mahmoud, M. S., & Mohammed, A. A. (2026). Immunohistochemical evaluation of androgen receptor expression in endometrial carcinoma and its clinicopathological significance. Diagnostic Pathology, 21(1), Article 67. https://doi.org/10.1186/s13000-026-01848-x

Image Credits: AI Generated

DOI: 10.1186/s13000-026-01848-x

Keywords: androgen receptor, endometrial carcinoma, immunohistochemistry, endometrioid carcinoma, serous carcinoma, FIGO staging, tumor grade, biomarker, hormone signaling, retrospective study, Diagnostic Pathology, gynecological pathology