Search

Preferentially expressed antigen in melanoma (PRAME) is a cancer testes antigen initially employed as a diagnostic marker for melanoma. Although negative in most normal tissues, its expression has been reported in benign endometrial glands. Additionally, PRAME expression has been identified in a growing list of solid and hematologic malignancies and is of interest as a predictive biomarker, as cancer vaccination strategies and adoptive T-cell transfer targeting this molecule are under clinical investigation; additionally, PRAME may identify candidates for retinoid therapy. However, expression of PRAME has not been well-studied in endometrial cancers. We herein evaluate PRAME expression in endometrial carcinomas to better characterize its limitations as a diagnostic melanoma marker as well as its potential as a predictive biomarker in endometrial carcinomas. PRAME expression was evaluated in 256 endometrioid (n=235) and serous (n=21) endometrial carcinomas via tissue microarray. In all, 89% (227/256) demonstrated some degree of nuclear PRAME expression, including 88% (207/235) of endometrioid carcinomas and 95% (20/21) of serous carcinomas. Diffuse (>50%) expression was observed in 70% (179/256) of all cases, including 69% (163/235) of endometrioid carcinomas and 76% (16/21) of serous carcinomas. There was no association between degree of expression and grade, mismatch repair protein status, or stage. The widespread expression of PRAME in endometrial carcinomas suggests this marker should not be interpreted as specific for melanoma in this context. However PRAME may have utility as a predictive biomarker in endometrial cancer, and expansion of testing of PRAME-based therapies to endometrioid and serous endometrial carcinomas may lead to new therapeutic options for these endometrial cancer subtypes., Competing Interests: The authors declare no conflict of interest., (Copyright © 2022 by the International Society of Gynecological Pathologists.)

Cervical carcinoma remains one of the most common cancers affecting women worldwide, despite effective screening programs being implemented in many countries for several decades. The International Collaboration on Cancer Reporting (ICCR) dataset for cervical carcinoma was first developed in 2017 with the aim of developing evidence-based standardized, consistent and comprehensive surgical pathology reports for resection specimens. This 4th edition update to the ICCR dataset on cervical cancer was undertaken to incorporate major changes based upon the updated International Federation of Obstetricians and Gynecologists (FIGO) staging for carcinoma of the cervix published in 2018 and the 5th Edition World Health Organization (WHO) Classification of Female Genital Tumors published in 2020 and other significant developments in pathologic aspects of cervical cancer. This updated dataset was developed by a panel of expert gynecological pathologists and an expert gynecological oncologist, with a period of open consultation. The revised dataset includes "core" and "noncore" elements to be reported; these are accompanied by detailed explanatory notes and references providing the rationale for the updates. Standardized reporting using datasets such as this helps facilitate consistency and accuracy, data collection across different sites and comparison of epidemiological and pathologic parameters for quality and research purposes., Competing Interests: The authors declare no conflict of interest., (Copyright © 2022 by the International Society of Gynecological Pathologists.)

Vulvar squamous cell carcinoma (vSCC), although rare, carries significant morbidity and a high rate of recurrence. Treatment options beyond surgical excision remain limited. Lymphocyte activation gene-3 (LAG-3) and its binding partner galectin-3 (GAL-3) are an immuno-inhibitory checkpoint pair that represent potential immunotherapy targets for the treatment of vSCC. This study examined the expression of LAG-3 and GAL-3 alongside programmed cell death ligand-1 expression in invasive SCC and vulvar intraepithelial neoplasia (VIN) by immunohistochemical analysis of formalin-fixed paraffin-embedded tissue. A total of 35 cases were selected for evaluation: 13 VIN3 [human papillomavirus (HPV)-associated VIN/usual-type VIN], 2 differentiated VIN (dVIN), 16 HPV-associated vSCC, and 4 dVIN-associated vSCC. LAG-3+ tumor-infiltrating lymphocytes were identified in 91% (32/35) of cases of vulvar squamous neoplasia. Tumor cells were positive for GAL-3 in 71% of the vulvar neoplasia cases. HPV-associated vSCC was more likely to demonstrate GAL-3 tumoral positivity when compared with dVIN-associated vSCC (24/29 vs. 1/6, P=0.004). We observed co-expression of all 3 immunomarkers in 40% (14/35) of cases evaluated. In light of these findings, use of immunomodulatory drugs that target the LAG-3/GAL-3 pathway may be potentially beneficial in vSCC and efficacy may be increased when combined with anti-programmed cell death ligand-1 therapy., Competing Interests: The authors declare no conflict of interest., (Copyright © 2021 by the International Society of Gynecological Pathologists.)

MLH1/PMS2 loss due to epigenetic hypermethylation of the MLH1 promoter is the most common cause of mismatch repair deficiency in endometrial carcinoma, and typically provides reassurance against an associated germline mutation. To further characterize the genetic features of MLH1/PMS2-deficient endometrial cancers, the departmental database was searched for cases with dual MLH1/PMS2 loss and retained MSH2/6 expression which underwent MLH1 hypermethylation testing. Genetic testing results were obtained when available. One hundred seventeen endometrial cancers met inclusion criteria: 100 (85%) were MLH1-hypermethylated, 3 (3%) were low-level/borderline, 7 (6%) were nonmethylated, and 7 (6%) were insufficient for testing. Sixteen cases (12 MLH1-hypermethylated, 3 nonmethylated, and 1 insufficient for testing) underwent germline testing, 6 of which (37.5%) demonstrated germline variants of unknown significance (VUS) (MSH6, PMS2, POLD1, BRIP1, RAD51D, CHEK2) but no known deleterious mutations. Notably, however, the patients harboring the MSH6 and PMS2 germline VUS had clinical features concerning for Lynch syndrome. One nonmethylated, germline-normal case underwent somatic tumor testing, and demonstrated a somatic MLH1 mutation. In summary, MLH1-hypermethylation accounts for the vast majority of MLH1/PMS2-deficient cancers in a universally screened population, although MLH1 somatic and germline mutations can occur. Occasionally, patients with MLH1-hypermethlated tumors also bear germline VUS in other mismatch repair genes as well as genes implicated in other hereditary cancer syndromes, but their clinical relevance is unclear. Family and personal cancer histories must always be evaluated to determine the need for germline testing in women with loss of MLH1/PMS2, even in the setting of hypermethylation., Competing Interests: The authors declare no conflict of interest., (Copyright © 2021 by the International Society of Gynecological Pathologists.)

Uterine carcinosarcomas have few adjuvant treatment options. Programmed cell death ligand-1 (PD-L1) expression in these tumors may predict response to checkpoint inhibitor therapies. An increase in PD-L1 expression has been shown in endometrial carcinomas with mismatch repair (MMR) deficiencies; however, few studies have evaluated PD-L1 expression in uterine carcinosarcomas. We examined PD-L1 expression in 41 cases of uterine carcinosarcoma using combined positive scores (CPS) and tumor proportion scores (TPS), and correlated with MMR status, p53 expression, and epithelial histotype. In addition to confirming the diagnosis of carcinosarcoma, the epithelial components were stratified based on endometrioid versus serous histology. Thirty-three cases (80%) were positive for PD-L1, defined as a CPS score of ≥1 or a TPS score of ≥1%. Twelve cases (29%) showed high expression of PD-L1, defined as a CPS score of ≥10 or a TPS score of ≥10%. The majority of the morphologically adjudicated carcinosarcomas had a serous epithelial component (83%) rather than endometrioid (17%), which was reinforced by aberrant p53 staining predominantly within cases with serous morphology. The majority of carcinosarcomas showed at least focal PD-L1 expression, predominantly in tumor-associated immune cells. Carcinosarcomas with endometrioid morphology were significantly more likely to have high-level PD-L1 (5/7 vs. 7/34; P=0.015). MMR-deficient carcinosarcomas were also more likely to have high-level PD-L1 (2/3 vs. 10/28); however, this did not reach statistical significance (P=0.2) and overall MMR-deficiency was uncommon (3 cases, 7%). These findings suggest that PD-L1 may be additive to MMR testing as a predictive biomarker for checkpoint inhibitor vulnerability in carcinosarcomas., Competing Interests: The authors declare no conflict of interest., (Copyright © 2020 by the International Society of Gynecological Pathologists.)

Ovarian high-grade serous carcinomas (HGSC) have shown lackluster responses to immunotherapies targeting the PD-1/PD-L1 axis, perhaps due to the coexistence of other mechanisms of immune evasion in this tumor type. Lymphocyte activation gene-3 (LAG-3) is another inhibitory immune checkpoint often expressed on tumor-associated lymphocytes which is targeted by drugs currently in clinical trials. Forty-eight HGSC with known germline BRCA mutation status were immunohistochemically stained for LAG-3, CD8, and FOXP3. Positive tumor-associated lymphocytes were enumerated and averaged over 10 high-power fields (HPF). PD-L1 immunostaining was also preformed and expression was evaluated on tumor cells and using the combined positive score (CPS). The average number of LAG-3-positve tumor-associated lymphocytes was 6/HPF (range: 0-25.6). Cytotoxic (CD8) T cells averaged 30/HPF (range: 0-168.9), and regulatory (FOXP3) cells averaged 6.6/HPF (range: 0-76.3). Tumoral PD-L1 expression of ≥1% was observed in 27% (13/48) of cases, with only 8% (4/48) showing >5% staining; 81% (39/48) cases had a CPS ≥1. LAG-3-positive lymphocytes and PD-L1 expression were positively correlated, even after controlling for the overall level of CD8 and FOX3P lymphocyte infiltration. Germline BRCA status was not significantly associated with LAG-3, CD8, FOXP3, or PD-L1 expression. These findings indicate that immunotherapies targeting LAG-3 may benefit some ovarian HGSC patients, particularly when used in conjunction with anti-PD-1/PD-L1 approaches. The typically limited expression of LAG-3 and PD-L1 suggests that immunotherapeutic response may be muted in most HGSC even with a combination approach.

LAG-3 is an immunosuppressive checkpoint molecule expressed on T cells. One of its ligands, GAL-3, can promote the progression of malignancy and has been identified on tumor cells. Both LAG-3 and GAL-3 are the targets of emerging immunotherapies, but have not been well-studied in endometrial carcinomas. LAG-3, CD3, and GAL-3 immunohistochemistry was performed on 75 endometrial cancers (25 nonmethylated mismatch repair-deficient, 25 MLH1-hypermethylated mismatch repair-deficient, and 25 mismatch repair-intact). LAG-3 and CD3 lymphocytes were averaged per high-power field. Tumoral GAL-3 expression was semiquantitatively scored. Tumor-infiltrating lymphocyte expression of LAG-3 and CD3 were positively correlated (Spearman ρ=0.521, P<0.001) and greater in mismatch repair-deficient compared with mismatch repair-intact tumors (LAG-3: P<0.001; CD3: P<0.001). The majority (64%) of endometrial carcinomas demonstrated ≥1% tumoral GAL-3 expression, with higher rates in mismatch repair-deficient versus intact tumors at the ≥1% (80% vs. 32%, P<0.001) and the ≥5% thresholds (52% vs. 16%, P=0.003). At the ≥5% threshold, nonmethylated mismatch repair-deficient cancers were more likely than intact tumors carcinomas to express GAL-3 (60% vs. 4/25 16%, P=0.003). LAG-3 lymphocytes were positively correlated with GAL-3 expression in nonmethylated mismatch repair-deficient endometrial carcinomas only (Spearman ρ=0.461, P=0.020). LAG-3 tumor-associated lymphocytes and GAL-3 neoplastic cells are common in endometrial carcinomas, particularly in nonmethylated mismatch repair-deficient cancers. This supports a role for immunotherapies targeting LAG-3 and/or GAL-3 in a subset of endometrial carcinomas, potentially in concert with other checkpoint inhibitors.

GATA binding protein 3 (GATA3) immunohistochemistry is primarily used as a marker of breast and urothelial differentiation, particularly in metastatic settings. In the gynecologic tract it also serves a robust marker for mesonephric and trophoblastic tumors. However, expression has also been described in more common malignancies of gynecologic tract including ovarian, endometrial, and cervical carcinomas. Data on the distribution of GATA3 expression in gynecologic malignancies is somewhat limited, particularly across different histologic subtypes of ovarian, endometrial, and cervical carcinomas. To assess the rates of GATA3 expression among common gynecologic cancers of various histologic types, 100 ovarian carcinomas, 64 endometrial carcinomas/atypical hyperplasias, 16 cervical squamous cell carcinomas (SCCs), and 14 endocervical adenocarcinomas were evaluated by immunohistochemistry for GATA3 positivity. Eight percent of endometrial carcinomas expressed GATA3, including 2 serous carcinomas, 1 carcinosarcoma, and 1 case of atypical hyperplasia. Six percent of ovarian carcinomas were GATA3-positive including 2 clear cell carcinomas, 2 mucinous adenocarcinomas, and 2 high-grade serous carcinomas. Thirty-eight percent of cervical SCCs showed weak to moderate staining in up to 50% of tumor cells. All endocervical adenocarcinomas were entirely negative for GATA3. In summary, GATA3 shows focal weak to moderate expression in a subset of endometrial and ovarian carcinomas. In contrast, usual-type endocervical adenocarcinomas are typically negative for GATA3, which can be helpful in differentiating them from mesonephric proliferations or carcinomas. A larger proportion of cervical SCCs express GATA3, therefore caution should be exercised when using this stain in the setting of a lower genitourinary carcinomas.

African American women with high-grade serous ovarian carcinoma have worse outcomes compared with women of European descent. Although the discrepancy is partially attributed to differences in access to care, the tumor immune microenvironment may also contribute. Expression of targetable immune regulatory molecules such as programmed cell death ligand-1 (PD-L1) and indoleamine 2,3 dioxygenase (IDO) is of particular interest as it may help guide therapy in this population. Using cases from the largest study of African American women with ovarian cancer, the African American Cancer Epidemiology Study, we characterized PD-L1 and IDO expression in 112 high-grade serous ovarian carcinomas. Immunohistochemistry for PD-L1, IDO, CD8, FOX3p, and CD68 was performed. PD-L1 and IDO were scored as the percentage of positive tumor cells and tumor-associated immune cells. CD8 and FOX3p counts were averaged across 10 high-power fields. Cox proportional hazards regression was used to evaluate the association between PD-L1 and IDO expression and survival. Tumor cells were positive for PD-L1 and IDO in 29% and 58% of cases, respectively. The majority showed <10% staining, and no cases exceeded 25% positivity. The majority of PD-L1-positive cases coexpressed IDO. PD-L1 and IDO expression was associated with higher CD8 and FOX3p counts (P<0.05). No association was observed between PD-L1 and IDO and survival. In summary, expression of PD-L1 and IDO is seen in a subset of high-grade serous ovarian carcinoma from African American women and is correlated with elevated lymphocyte infiltration. While PD-L1 and IDO co-expression suggests a role for dual immunotherapy, diffuse expression of PD-L1 and IDO is rare, invoking caution regarding the potential for immunotherapeutic response.

Endometrial carcinoma (ECA) is frequently hormonally driven and can be treated with endocrine-based therapy, yet hormone receptor status is not routinely assessed. In particular, little is known about the significance of androgen receptor (AR) in ECA. Androgen has antiproliferative effects in the healthy endometrium and could serve a similar role to progesterone in curbing the progression of estrogen-dependent neoplasia. There may also be a subset of ECA that benefits from androgen antagonistic therapy. We herein investigate AR expression across ECA subtypes and compare its expression to estrogen receptor (ER) and progesterone receptor (PR). Immunohistochemical staining for AR, ER, and PR was performed on an endometrial tissue microarray containing 50 ECA with a variety of morphologic subtypes as well as 20 benign and 9 atypical hyperplastic endometria. AR was expressed by 54% (27/50) of ECA including 60% of low grade endometrioid carcinomas, 70% high grade endometrioid carcinomas, 70% serous carcinomas, 50% carcinosarcomas, and 20% clear cell carcinomas. High AR expression was chiefly restricted to a subset of serous carcinomas (50%). AR expression occurred most often in concert with ER staining, although 5 high grade cancers (1 serous carcinoma, 4 carcinosarcomas) showed AR expression in the absence of ER. In summary, AR positivity is seen in over half of ECA in our study, including the majority of low grade endometrioid carcinomas, high grade endometrioid carcinomas, and serous carcinomas. High level expression is seen in half of serous carcinomas and a subset of serous carcinomas and carcinosarcomas show some degree of AR staining in the absence of ER, suggesting a possible role for androgen inhibition in treatment of these cases.

Defects in the DNA mismatch-repair system are identified in ∼25%-30% of endometrial carcinomas (ECs). Whereas some ECs are due to germline Lynch syndrome (LS)-associated mutations, the majority demonstrate sporadic MLH1 promoter hypermethylation (MLH1hm). MLH1hm characterizes a unique subset of colorectal cancers with a relatively poor prognosis; however, the morphology and behavior of sporadically methylated ECs (SMECs) are less well understood. We herein review the clinicopathologic features of 34 SMECs diagnosed at The University of Virginia and compare them with LS-associated and Lynch-like endometrial cancers. Most SMECs were centered in the uterine fundus (74%) and displayed Grade 1 (35%) or Grade 2 (41%) endometrioid morphology. Mucinous differentiation was present in 50%, and 65% showed tumor-infiltrating lymphocytes. Cases with undifferentiated components or sarcomatous differentiation were relatively rare (9%). Most SMECs presented at stage 1 or 2 (91%). When compared with LS and Lynch-like endometrial cancers, SMECs occurred ∼8 yr later and were significantly more likely to show conventional endometrioid morphology (P=0.04), mucinous differentiation (P=0.002), and tumor-infiltrating lymphocytes (P=0.002). These findings suggest that SMECs constitute a clinicopathologically distinct subset of mismatch-repair-deficient/high-level microsatellite instability tumors. Future studies are needed to further understand the implications of MLH1hm on the prognosis and treatment of endometrial cancer.

It is well documented that endometrial and ovarian carcinoma are associated with Lynch syndrome (LS), but the association, if any, between endocervical carcinoma and LS has not been fully evaluated. The relationship between endocervical carcinoma and LS is particularly relevant, given the apparent affinity of LS-associated endometrial carcinomas for the lower uterine segment and the attendant difficulties in determining tumor origin at this site. In this study, we examined mismatch repair (MMR) protein expression (MLH1, MSH2, MSH6, and PMS2) in 60 endocervical adenocarcinomas, including variants (minimal deviation adenocarcinoma, mesonephric adenocarcinoma, adenosquamous carcinoma, clear cell carcinoma) and a series of well-characterized lower-uterine segment carcinomas of known endocervical or endometrial origin (n=41). Two of the lower uterine segment tumors occurred in risk-reducing hysterectomy specimens from known LS patients. All endocervical adenocarcinomas including variants and lower uterine segment endocervical tumors (1 from a known LS patient) were proficient in all 4 MMR proteins. In contrast, 2/20 (10%) lower uterine segment endometrial cancers were deficient in at least 1 MMR (1 from a known LS patient). These data provide evidence that, unlike endometrial and ovarian adenocarcinoma, there is no association between LS and endocervical carcinoma. MMR testing is prudent in lower uterine segment tumors in women with possible LS, especially those for which definitive site of origin cannot be determined.