Search

Uterine endometrial stromal sarcomas (ESS) with YWHAE::NUTM2gene fusions are typically morphologically high-grade tumors composed of atypical round cells, sometimes associated with a low-grade fibromyxoid component; they are currently included in the category of high-grade ESS (HGESS). We report 5 morphologically pure low-grade endometrial stromal tumors harboring YWHAE::NUTM2fusions, including 1 endometrial stromal nodule (ESN) and 4 low-grade endometrial stromal sarcomas (LGESS), an association only occasionally reported previously. Patients ranged from 30 to 51 (mean=43) years and tumors from 4.5 to 7.5 cm (mean=5.7). All were stage I at diagnosis (confined to the uterus). Microscopically, the 4 LGESS showed extensive “tongue-like” invasion of the myometrium, and the ESN was entirely confined to the endometrium with no myometrial invasion. All tumors were composed entirely of morphologically uniform bland ovoid cells resembling proliferative endometrial stroma. A fibromyxoid component was seen in 1 LGESS and the ESN; in the LGESS, this was the sole component. Atypical round cells characteristic of YWHAE::NUTM2HGESS were not identified. Mitotic count ranged from <1 to 13 per 10 high-power fields (mean: 3). CD10 was positive in 2/4 (focal), estrogen receptor in 5/5 (focal=1; diffuse=4), progesterone receptor in 5/5 (focal=1; diffuse=4) and cyclin D1 was diffusely positive in 3/4. Follow-up was available in all 5 patients and ranged from 6 to 159 months (mean=72). Two patients with LGESS had recurrent disease at 15 and 155 months; 1 showed predominantly LGESS with rare round cells in the initial recurrence and pure HGESS in a subsequent recurrence, while the other patient’s recurrent tumor was predominantly HGESS (90%) in a background of focal fibromyxoid LGESS (10%). Both patients rapidly progressed and died of disease within 5 months of high-grade recurrence. We show that rare cases of morphologically pure low-grade endometrial stromal tumors, some but not all with a fibromyxoid component, harbor YWHAE::NUTM2fusions and may recur rapidly, with transformation to HGESS and aggressive behavior. Our findings suggest that at least a subset of YWHAE::NUTM2HGESS evolves from LGESS. We suggest that cyclin D1 and CD10 staining should be performed in all LGESS. Diffuse staining for cyclin D1 and/or negative or focal staining for CD10 should suggest the possibility of a YWHAE::NUTM2fusion, and appropriate molecular testing should be undertaken. Since no single morphological or immunohistochemical parameter is entirely sensitive for fusion status, we also suggest that testing for a YWHAE::NUTM2gene fusion should be considered in all cases of LGESS and, if a fusion is present, this should raise the possibility of subsequent high-grade transformation and aggressive behavior, even though such cases should still be categorized as LGESS. Although seemingly rare, ESN and LGESS with YWHAE::NUTM2fusions may be under-recognized due to a lack of routine fusion testing.

Sertoli-Leydig cell tumors (SLCTs) are uncommon ovarian sex cord-stromal neoplasms which are currently classified into well, moderately, and poorly differentiated and retiform types. Well-differentiated SLCT is the least common and typically occurs in pure form, whereas moderately and poorly differentiated and retiform types often comprise a morphologic spectrum with an admixture of all 3. DICER1pathogenic variants are very common in SLCTs but, as far as we are aware, have not been reported in well-differentiated neoplasms, although the number of cases studied is small due to the rarity of this neoplasm. We undertook DICER1molecular testing in a cohort of 18 well-differentiated SLCTs and show all these to be DICER1wild-type. None of the cases harbored the p.FOXL2C134W hotspot mutation. Based upon the DICER1molecular results, together with morphologic observations, we propose that well-differentiated SLCT is an unrelated neoplasm to the more common moderately/poorly differentiated and retiform SLCTs and is a fundamentally distinct and unrelated tumor type within the ovarian sex cord-stromal tumor family. The implications for tumor nomenclature and recommendations for future tumor classification are discussed within the context of tumors collectively known as SLCTs.

With the widespread application of next-generation sequencing, the genetic landscape of uterine mesenchymal neoplasms has been evolving rapidly to include several recently identified fusion genes. Although chromosomal rearrangements involving the 10q22 and 17q21.31 loci have been reported in occasional uterine leiomyomas decades ago, the corresponding KAT6B::KANSL1fusion has been only recently identified in 2 uterine tumors diagnosed as leiomyoma and leiomyosarcoma. We herein describe 13 uterine stromal neoplasms carrying a KAT6B::KANSL1(n=11) and KAT6A::KANSL1(n=2) fusion. Patient ages ranged from 33 to 81 years (median, 49 y). Tumor size was 2.6 to 23.5 cm (median, 8.2 cm). Nine tumors were myometrium-centered, and 3 had an intracavitary component. Original diagnoses were mostly low-grade endometrial stromal sarcoma (LG-ESS; 10 cases) with atypical features (limited CD10 expression, sex cord-like features, pericytic vasculature, and frequent myxoid changes). Treatment was hysterectomy±bilateral salpingo-oophorectomy (10), myomectomy (1), and curettage (2). Five patients were disease-free at 6 to 34 months, 3 (27%) died of disease at 2 to 47 months, and 3 were alive with disease at 2, 17, and 17 years. Histologically, most tumors showed variable overlap with LG-ESS, but they were generally well-circumscribed lacking the extensive permeative and angioinvasive growth typical of LG-ESS. They were composed of monotonous medium-sized oval and spindle cells arranged into diffuse sheets with prominent spiral-type arterioles and frequent pericytoma-like vascular pattern. Variable myxoid stromal changes were frequent. Mitotic activity ranged from 1 to >20 in 10 HPFs. Immunohistochemistry showed variable expression of CD10 (12/13), estrogen receptor (8/11), progesterone receptor (8/11), smooth muscle actin (9/11), desmin (4/12), h-caldesmon (2/10), calretinin (3/8), inhibin (1/7), WT1 (4/7), cyclin D1 (5/11; diffuse in only 1 case), and pankeratin (5/10). This series characterizes a KAT6B/A::KANSL1fusion-positive uterine stromal neoplasm within the morphologic spectrum of LG-ESS but with atypical features. The relationship of these neoplasms to genuine LG-ESS remains unclear. This molecular subtype of uterine endometrial stromal sarcoma has the potential for an unfavorable clinical course despite the absence of widely invasive growth; nevertheless, analysis of more cases is necessary to delineate the phenotypic spectrum and biological potential of this tumor.

p16 is the most useful diagnostic marker for human papillomavirus (HPV)-associated anogenital lesions. In the cervix, the pattern of p16 immunoreactivity generally correlates with lesion severity. p16 expression in anal intraepithelial neoplasia (AIN) is far less studied. Whether such correlation holds true has to be determined. We correlated the degree and pattern of p16 immunohistochemistry (IHC) results with morphologic diagnoses of 1000 anal squamous and transitional zone biopsy specimens. Using the Lower Anogenital Squamous Terminology criteria, p16 IHC results were classified as block staining, partial staining, or negative. Among 150 samples without morphologic evidence of AIN, p16 was negative in 85% and partial staining in 15%. AIN 1 (n=400) revealed diverse results: 28% negative, 35% partial, and 37% block staining. Among AIN 2 (n=298), 89% were block, 9% partial staining, and 2% negative. AIN 3 (n=152) revealed block (95%) or partial staining (5%). For the detection of AIN 2/3, p16 block staining yielded 91% sensitivity, 73% specificity, 80% positive predictive value, 91% negative predictive value, and a Youden Index of 0.64. Combining block staining and partial staining slightly increased sensitivity (99%) and negative predictive value (98%), but significantly decreased specificity (43%), positive predictive value (59%) and Youden Index (0.42, P<0.001). As with the cervix, p16 immunoreactivity correlates with morphologic diagnoses of AIN. Block staining offers the optimal diagnostic value for AIN 2/3. Caution is required since AIN 1 frequently exhibits block staining; the prognostic value of p16 warrants further investigation.

FOXL2somatic mutation occurs in a high percentage of ovarian adult granulosa cell tumors and DICER1mutations in a high proportion of Sertoli-Leydig cell tumors. These mutations have only been studied in a limited number of juvenile granulosa cell tumors (JGCTs), and their occurrence and frequency in these neoplasms is controversial. We aimed to determine the frequency of FOXL2and DICER1mutations in a large cohort of 50 JGCTs, and to evaluate the prognostic impact of these mutations. A FOXL2hotspot mutation was found in 2/50 JGCTs. Review of these 2 cases reclassified them as adult granulosa cell tumors. Thus, FOXL2mutation was absent from our large cohort of JGCTs. DICER1mutations in the RNase IIIb domain were found in 4 cases. After review of the mutated cases, 1 was reclassified as a gynandroblastoma with a prominent JGCT component. Thus, DICER1mutations were detected in 3/47 (6%) of pathologically confirmed JGCTs. Our results show that FOXL2mutations are not present in JGCT, whereas a small percentage of these neoplasms exhibit DICER1mutations.

Mesonephric adenocarcinoma (MA) and mesonephric-like adenocarcinoma (MLA) are uncommon neoplasms of the gynecologic tract that have until recently been poorly understood. Although their morphologic, immunohistochemical, and molecular profiles have been recently defined, little is known about their clinical behavior. Small studies have demonstrated inconsistent findings and no large studies have examined the clinical behavior of these adenocarcinomas. In this multi-institutional study, representing the largest and most stringently defined cohort of cases to date, we examined the clinicopathologic features of 99 MAs and MLAs (30 MAs of the uterine cervix, 44 MLAs of the endometrium, and 25 MLAs of the ovary). Only tumors with characteristic mesonephric morphology and either immunohistochemical or molecular support were included. Our results demonstrate that the majority of mesonephric neoplasms presented at an advanced stage (II to IV) (15/25 [60%] MA of the cervix, 25/43 [58%] MLA of the endometrium, and 7/18 [39%] MLA of the ovary). The majority (46/89 [52%] overall, 12/24 [50%] MA of the cervix, 24/41 [59%] MLA of the endometrium, and 10/24 [42%] MLA of the ovary) developed recurrences, most commonly distant (9/12 [75%] MA of the cervix, 22/24 [92%] MLA of the endometrium, and 5/9 [56%] MLA of the ovary). The 5-year disease-specific survival was 74% (n=26) for MA of cervix, 72% (n=43) for MLA of endometrium, and 71% (n=23) for MLA of ovary. Our results confirm that mesonephric neoplasms are a clinically aggressive group of gynecologic carcinomas that typically present at an advanced stage, with a predilection for pulmonary recurrence.

Mullerian adenosarcoma is a biphasic neoplasm composed of benign or atypical Müllerian epithelium and a malignant mesenchymal component that is usually, but not always, of low grade. Focal architectural or cytologic atypia of the epithelial component resembling atypical hyperplasia may uncommonly be present and foci of adenocarcinoma have been rarely reported. Whether the coexistence of these 2 tumor components is a result of independent primaries (collision tumor), adenocarcinoma arising from the epithelial component of the adenosarcoma, an unusual form of carcinosarcoma or some other mechanism is uncertain. To establish the diagnostic criteria and clinical significance of the coexistence of adenocarcinoma in close association with Müllerian adenosarcoma, we conducted a multi-institutional study of these rare tumors. Twenty-six patients were identified with “mixed” adenosarcoma and adenocarcinoma; they ranged in age from 43 to 87 years (median: 66 y). Tumors occurred in the uterine corpus (n=22), ovary (n=2), and the pelvis (n=2). All but 6 had International Federation of Gynecology and Obstetrics (FIGO) stage I disease. All extrauterine tumors were associated with endometriosis. The tumor size ranged from 2 to 25 cm (median: 7.9 cm). The sarcomatous component was of low grade in 18 and high grade in 8 (the majority demonstrating rhabdomyoblastic differentiation); 9 had stromal overgrowth. Twenty-five carcinomas were endometrioid in type (23 FIGO grade 1; 3 FIGO grade 2) and 1 carcinoma was dedifferentiated with FIGO grade 1 endometrioid adenocarcinoma component; 33% of the uterine neoplasms were associated with adjacent endometrial hyperplasia. Next-generation sequencing in 2 tumors identified similar molecular abnormalities in the sarcomatous and carcinomatous components supporting a clonal relationship. Of 10 patients with available follow-up (median: 18 mo), 8 had no evidence of disease and 2 died of recurrent sarcoma at 7 and 8 months. Endometrioid adenocarcinomas that arise in close spatial association with Müllerian adenosarcoma appear to be clonally related to the sarcoma. Unlike carcinosarcomas, these tumors are usually early stage at presentation. The prognosis appears to be driven by the sarcomatous component. These tumors should be distinguished from carcinosarcomas, dedifferentiated endometrial carcinomas, and corded and hyalinized endometrioid carcinomas.

Adult granulosa cell tumor (AGCT) is a low-grade malignant neoplasm with a significant propensity for late recurrence and metastasis. Almost all AGCTs are composed of cells with bland nuclear features and even when these tumors recur or metastasize, the nuclear features are almost always low-grade. We report 5 cases of AGCT in patients aged 37 to 88 years composed of areas of typical AGCT with low-grade morphology admixed with areas of high-grade morphology, with marked nuclear atypia, often with bizarre multinucleate cells and high mitotic activity; this is the first reported series of high-grade transformation in AGCTs. The high-grade areas often morphologically closely resembled juvenile granulosa cell tumor with abundant eosinophilic cytoplasm, significant mitotic activity, and intermediate sized follicles. Four cases were FIGO stage IA at diagnosis and 1 was stage IIIC with omental involvement. FOXL2mutation analysis of both the morphologically low-grade and high-grade areas in 4 of 5 cases confirmed the presence of missense point mutation, c.402C>G, p.(Cys134Trp), providing conclusive evidence that the high-grade component represents transformation of typical AGCT rather than the coexistence of another sex cord-stromal tumor, such as juvenile granulosa cell tumor, which has been suggested for such neoplasms. In 3 of 4 cases where immunohistochemistry was undertaken, there was a striking difference between the p53 staining in the low-grade and high-grade components with wild-type staining in the former and diffuse mutation-type immunoreactivity in the latter, suggesting that TP53mutation is likely to play a role in high-grade transformation. TP53mutation analysis covering exons 4 to 10 was undertaken in 4 cases and TP53mutations were identified in the high-grade component of 2 of the cases. In 1 case, there was diffuse block-type p16 staining in the high-grade component. Follow-up in the 4 stage IA neoplasms revealed no evidence of tumor recurrence in 3 (6 to 9 mo follow-up) while the other patient developed mediastinal, peritoneal, and pulmonary metastasis 17 months after diagnosis. High-grade transformation is uncommon in AGCTs and given that one of our cases was advanced stage at diagnosis, another exhibited widespread metastasis within a short period and there have been occasional case reports of aggressive behavior in AGCTs with high-grade transformation, this event may herald an aggressive clinical course.

Micropapillary adenocarcinoma has been reported as an aggressive variant of adenocarcinoma in several organs, where it is associated with poor clinical outcome. This study reports the clinicopathologic features and outcomes of cervical adenocarcinomas with a micropapillary component (micropapillary cervical adenocarcinomas); this represents the largest reported study of these neoplasms. The study comprised 44 cervical adenocarcinomas of usual (human papillomavirus–related)-type (84%), mucinous, not otherwise specified (4.5%), gastric-type (4.5%), endometrioid (4.5%), and adenosquamous carcinoma (2%). The micropapillary component comprised >50% of the neoplasm in 34 cases (77%) (group 1), and 10% to 50% in 10 cases (23%) (group 2). Lymph node metastasis was present in 41 of 44 (93%) cases and typically the nodal tumor retained a prominent micropapillary morphology. Follow-up ranged from 7 to 123 months (mean, 65.9 mo). Seventeen of 44 (38.6%) patients had no evidence of disease on follow-up, 6/44 (13.6%) were alive with disease, and 21/44 (47.7%) died of disease. There were no survival differences between group 1 and group 2. On univariate analysis, lymph node metastasis (P=0.0015), lymphovascular space invasion (P=0.002), parametrial involvement (P=0.03), and depth of stromal invasion (P=0.045) were related to tumor recurrence. On multivariate analysis, lymph node metastasis (P=0.001), and extent of lymphovascular space invasion (P=0.027) were significant independent predictors of tumor recurrence. Our study shows that a micropapillary component in cervical adenocarcinoma may be associated with aggressive behavior and that a micropapillary architecture may occur within a variety of types of cervical adenocarcinoma.

Cervical gastric-type adenocarcinomas are aggressive non–human papillomavirus-related carcinomas with a propensity for extracervical spread, including unusual sites such as the omentum, peritoneum, and ovary. We report 7 cases of cervical gastric-type adenocarcinoma with fallopian tube involvement predominantly in the form of mucosal colonization without underlying invasion. As far as we are aware, this has not been previously described and this report adds to the literature regarding metastatic neoplasms, which may exhibit tubal mucosal involvement and mimic an in situ lesion at this site. In all cases, there was associated ovarian involvement and in 6 of 7 cases, there was endometrial colonization. We speculate that the fallopian tube (and ovarian) involvement is secondary to transuterine spread. Given the occasional occurrence of multifocal gastric-type glandular lesions (benign or malignant) involving different sites in the female genital tract, we discuss the distinction between synchronous independent and metastatic lesions.

Endometrial stromal sarcomas (ESSs) are mesenchymal uterine tumors characterized by recurrent genetic events, most commonly chromosomal rearrangements, that create oncogenic gene fusions. High-grade endometrial stromal sarcomas (HG-ESSs), as defined in the 2014 World Health Organization Classification, typically contain oncogenic YWHAE-NUTM2fusions; however, although not well characterized, there are tumors morphologically overlapping with HG-ESS that do not contain the YWHAE-NUTM2fusions. These fusions are also found in certain pediatric primitive sarcomas, including clear cell sarcoma of the kidney and soft tissue undifferentiated round cell sarcoma of infancy. A subset of these same pediatric sarcomas lack YWHAE-NUTM2fusions and instead have internal tandem duplications (ITDs) involving exon 15 of BCOR(BCORITD). We investigated the presence of BCORITD by targeted sequencing in a series of 31 uterine sarcomas, comprising 5 low-grade ESS, 13 uterine sarcomas diagnosed as HG-ESS, and 13 undifferentiated uterine sarcomas. BCORITD were present in 1 uterine sarcoma diagnosed as HG-ESS and 2 undifferentiated sarcomas with uniform nuclear features, all of which lacked any of the recurrent chromosome translocations known to occur in ESS. These 3 high-grade sarcomas with BCORITD affected young patients (average age, 24) and morphologically were composed of nonpleomorphic spindle cells admixed with epithelioid and round cell areas. Focal myxoid stroma was present in 2 cases. Mitotic activity was brisk, necrosis was present, and there was lymphovascular involvement in all cases. The 3 uterine sarcomas with BCORITD exhibited diffuse cyclin D1 immunohistochemical expression and there was diffuse BCOR expression in the 2 cases tested. Long-term follow-up in 2 patients revealed 1 to be tumor-free after 22 years and the other to die of disease after 8 years. In conclusion, BCORITD is an oncogenic alternative to YWHAE-NUTM2fusion in high-grade uterine sarcomas with uniform nuclear features. We propose that neoplasms with the morphology described and BCORITD be regarded as a unique subtype of high-grade uterine sarcoma, possibly within the family of endometrial stromal neoplasia.

Supplemental Digital Content is available in the text.Ovarian Sertoli-Leydig cell tumors (SLCTs) are uncommon sex cord-stromal tumors associated with both germ-line and somatic DICER1mutations, the frequency of which has varied widely in different studies (0% to 62.5%). The current World Health Organization Classification includes 3 histologic types of SLCTs (well-differentiated, moderately differentiated, and poorly differentiated); heterologous elements and/or retiform patterns may be present in moderately and poorly differentiated neoplasms. We investigated the frequency of DICER1mutations in a series of 38 ovarian tumors initially diagnosed as SLCTs, and explored whether identified mutations were associated with specific morphologic features. Specialist pathology review performed blinded to molecular results confirmed 34 tumors to be SLCTs (22 moderately differentiated, 8 poorly differentiated; 4 well-differentiated), while the remaining 4 neoplasms were considered not to represent SLCTs. Of the 34 cases diagnosed as SLCTs, 30 (88%) harbored ≥1 DICER1mutation. All 30 moderately differentiated/poorly differentiated SLCTs contained mutations, but we did not find deleterious DICER1mutations in the 4 well-differentiated SLCTs. Our study reports the highest DICER1mutation frequency to date in SLCTs, with 100% of moderately differentiated and poorly differentiated tumors being DICER1-mutated. This suggests that DICER1mutation may be a defining feature of these neoplasms. Although the number of cases is limited, well-differentiated SLCTs appear to be DICER1-independent. Moderately differentiated and poorly differentiated SLCT components often coexist with each other and form part of a spectrum, while well-differentiated SLCTs usually occur in pure form, suggesting that fundamentally, these represent 2 separate and independent tumor types with a different pathogenesis. We suggest that all patients with ovarian SLCTs undergo germ-line DICER1mutation testing.

In recent years, a number of benign and malignant cervical glandular lesions exhibiting gastric differentiation have been described but premalignant gastric-type lesions have not been well characterized. We report a series of 9 cases of a rare form of cervical adenocarcinoma in situ (AIS) distinguished by gastric and sometimes intestinal differentiation and lack of association with human papillomavirus (HPV) infection. The lesions occurred in women aged 25 to 73 years (mean age 51 y). All cases were located at (or just proximal to) the cervical transformation zone and there was extension to the lower uterine segment in 3 cases, 2 of which also involved the endometrium. In all cases, the normal cervical glandular architecture was largely preserved but in 5 cases there was a mild degree of increased intraglandular architectural complexity. The glandular epithelium ranged from almost purely gastric in type (4 cases) to mixed gastric and intestinal (5 cases), with varying proportions of intermixed goblet cells. In contrast to the basophilic cytoplasm of normal endocervical glands, the gastric-type epithelium was typically predominantly composed of cells with eosinophilic or pale pink cytoplasm, but conspicuous foamy or clear cell cytoplasm was present in some cases. Nuclear atypia was present in all cases but was considered low-grade in 8. High-grade features such as marked nuclear pleomorphism and hyperchromasia were evident in only 1 case. Mitotic activity and apoptotic bodies were present but both were noted to be less frequent than in usual type (HPV-related) AIS. Immunohistochemically, there was usually positive staining with CK 7 (7/7 cases) and MUC 6 (7/8 cases) and some cases were positive with CK 20 (3/7), CDX2 (5/9), PAX 8 (5/9) and CEA (2/6). Estrogen receptor and progesterone receptor were usually negative, although Estrogen receptor was positive in 3 of 9 cases. p16 was negative or exhibited mosaic-type staining (nonblock staining) in all cases and there was mutation-type p53 staining in 2 of 9 cases. HPV molecular testing was negative in all 4 cases tested. We believe this unusual subtype of AIS, which we term “gastric-type AIS (gAIS),” represents a precursor to gastric-type adenocarcinoma of the cervix and suggest that gAIS and so-called “atypical lobular endocervical glandular hyperplasia” are related entities within a spectrum of premalignant gastric-type lesions for which we propose the umbrella term gAIS. The malignant potential and optimal management of gAIS are currently unknown but in one of our cases a gastric-type adenocarcinoma developed 6 years after removal of a cervical polyp which contained gAIS. The introduction of HPV vaccination will result in a relative increase in incidence of premalignant and malignant cervical glandular lesions exhibiting gastric differentiation and these will not be detected by HPV-based screening programs.

Supplemental Digital Content is available in the text.Inflammatory myofibroblastic tumor (IMT) can occur in a number of anatomic sites, including the uterus. Like its soft tissue counterpart, uterine IMT frequently expresses ALK and harbors ALKgenetic rearrangements. The aim of this study is to fully characterize the genetic fusions that occur in uterine IMT. We studied 11 uterine IMTs with typical histology and 8 uterine myxoid smooth muscle tumors (5 leiomyomas, 1 smooth muscle tumor of uncertain malignant potential, and 2 leiomyosarcomas) in which the differential of IMT was considered, using a RNA-sequencing–based fusion assay to detect genetic fusions involving ALK, ROS1, RET, NTRK1/3, and other genes. ALK was expressed in 10 of 11 IMTs and 1 tumor initially categorized as a myxoid leiomyoma (granular cytoplasmic staining with paranuclear accentuation). Fusion transcripts involving ALKwere identified in 9 of 10 ALK immunopositive IMTs, with 3 harboring IGFBP5-ALK, 3 harboring THBS1-ALK, 2 harboring FN1-ALK, and 1 harboring TIMP3-ALK. Among the smooth muscle tumors, IGFBP5-ALKfusion transcript was identified in only 1 ALK immunopositive case. Further review revealed that although a diagnosis of IMT was considered for the ALK immunopositive myxoid leiomyoma, this diagnosis was not initially rendered only because fluorescence in situ hybridization analysis was interpreted as negative for ALKgenetic rearrangement; this case is best reclassified as an IMT. Notably, all the ALKfusions identified in our study included the transmembrane domain-encoding exon 19 of ALK. Our findings confirm the high frequency of ALKfusions in uterine IMT, with an enrichment of novel 5′ ALKfusion partners (IGFBP5, THBS1, and TIMP3) and exon 19-containing ALKfusion. Given that IGFBP5and FN1are both situated on the same chromosome as ALK, fluorescence in situ hybridization analysis for ALKrearrangement may not be reliable and a negative result should not exclude a diagnosis of uterine IMT if the histologic features and ALK immunostaining findings are supportive.

Supplemental Digital Content is available in the text.The 2014 World Health Organization Classification of Tumors of Female Reproductive Organs endorsed the new category of seromucinous carcinoma, a neoplasm that exhibits morphologic and immunophenotypic overlap with other histotypes of ovarian carcinoma. The goal of this study was to determine whether seromucinous carcinoma was a distinct histotype by assessing its diagnostic reproducibility and comparing its molecular composition to the 5 major histotypes of ovarian carcinoma. Thirty-two tumors diagnosed as seromucinous carcinomas from 2 centers were studied. Eighteen cases were randomly selected for a review set comprising a total of 50 ovarian carcinomas of various histotypes. Morphologic histotype was independently assessed by 4 pathologists. For the 32 seromucinous carcinomas, a histotype-specific immunophenotype was assigned using a diagnostic immunohistochemical panel. Histotype-specific genotype was assigned using a combination of immunohistochemistry and targeted next-generation sequencing for somatic mutations, including genes recurrently mutated in ovarian carcinomas. There was low to modest agreement between pathologists with the reference diagnosis of seromucinous carcinoma, ranging from 39% to 56% for the 4 observers. The immunophenotype was not unique but overlapped predominantly with endometrioid and to a lesser extent with mucinous and low-grade serous carcinoma. Genomic and immunohistochemical alterations were detected in a number of target genes, including KRAS(70%), PIK3CA(37%), PTEN(19%), and ARID1A(16%); no CTNNB1mutations were identified. Nine cases (30%) harbored concurrent KRAS/PIK3CAmutations. An endometrioid genotype was assigned to 19 cases, a low-grade serous genotype to 9, and a mucinous genotype to 1 and 3 cases were uninformative. Integrating morphology, immunophenotype, and genotyping resulted in reclassifying the seromucinous carcinomas to endometrioid 23/32 (72%), low-grade serous 8/32 (25%), and mucinous 1/32 (3%). The morphologic diagnosis of seromucinous carcinomas is not very reliable and it does not exhibit a distinct immunophenotype or genotype. The molecular features overlap mostly with endometrioid and low-grade serous carcinomas. Our data suggest the category of seromucinous carcinoma be discontinued as ancillary molecular tests can assign cases to one of the major histotypes.

We report a series of 7 unusual and morphologically distinct cervical or upper vaginal lesions in women aged 41 to 70 years. The lesions involved the cervix in 3 cases, the upper vagina in 2, the cervix and vagina in 1, and in 1 case the site of origin could not be determined. The lesions had a consistent morphologic appearance with a surface “plaque-like” or “stuck-on” configuration apparent in those cases where surrounding normal tissues were present. Broad coalescing solid sheets and interconnecting trabeculae of cytologically bland cells with a rather “basaloid” appearance emanated from the surface and there were scattered squamous eddies. Other features included peripheral palisading and a stroma containing hyalinized basement membrane-like material. Immunohistochemically, the lesions were diffusely positive with p63, CK5/6, and 34βE12 and focally positive with CK7, but largely negative with CK20, EMA, CEA, and BerEP4. p16 was negative or exhibited nonblock-type immunoreactivity and GATA3 was negative or weakly positive. Molecular testing detected human papillomavirus type 42 in 3 of 7 cases, with no virus detected in the remaining 4 cases. Rarely, similar cases have been reported previously as inverted transitional papilloma of the cervix or vagina, but based on the morphology and immunophenotype we do not feel these represent transitional lesions. We suggest the term seborrheic keratosis-like lesions to designate this new and rare entity, which may be associated with low-risk human papillomavirus infection. Limited follow-up in a small number of cases suggests that these lesions follow a benign clinical course.

Adenocarcinomas exhibiting gastric differentiation represent a recently described and uncommon subtype of non–human papillomavirus (HPV)-related cervical adenocarcinoma. They comprise a spectrum from a well-differentiated variant (adenoma malignum/mucinous variant of minimal deviation adenocarcinoma) to a more poorly differentiated overtly malignant form, generally referred to as gastric-type adenocarcinoma. Rarely, such tumors have also been described as primary vaginal neoplasms. Gastric-type adenocarcinomas exhibit considerable morphologic overlap with adenocarcinomas originating outside the female genital tract, especially mucinous adenocarcinomas arising in the pancreas and biliary tract. Moreover, they often metastasize to unusual sites, such as the ovary and peritoneum/omentum, where they can be mistaken for metastatic adenocarcinomas from other, nongynecologic sites. There is little information regarding the immunophenotype of gastric-type adenocarcinomas, and knowledge of this is important to aid in the distinction from other adenocarcinomas. In this study, we undertook a detailed immunohistochemical analysis of a large series of cervical (n=45) and vaginal (n=2) gastric-type adenocarcinomas. Markers included were cytokeratin (CK)7, CK20, CDX2, carcinoembryonic antigen, CA125, CA19.9, p16, estrogen receptor, progesterone receptor, MUC6, PAX8, PAX2, p53, hepatocyte nuclear factor 1 beta, carbonic anhydrase IX, human epidermal receptor 2 (HER2), and mismatch repair (MMR) proteins. All markers were classified as negative, focal (<50% of tumor cells positive), or diffuse (≥50% tumor cells positive) except for p53 (classified as “wild-type” or “mutation-type”), HER2 (scored using the College of American Pathologists guidelines for gastric carcinomas), and MMR proteins (categorized as retained or lost). There was positive staining with CK7 (47/47—45 diffuse, 2 focal), MUC6 (17/21—6 diffuse, 11 focal), carcinoembryonic antigen (25/31—12 diffuse, 13 focal), carbonic anhydrase IX (20/24—8 diffuse, 12 focal), PAX8 (32/47—20 diffuse, 12 focal), CA125 (36/45—5 diffuse, 31 focal), CA19.9 (11/11—8 diffuse, 3 focal), hepatocyte nuclear factor 1 beta (13/14—12 diffuse, 1 focal), CDX2 (24/47—4 diffuse, 20 focal), CK20 (23/47—6 diffuse, 17 focal), and p16 (18/47—4 diffuse, 14 focal). Most cases were negative with estrogen receptor (29/31), progesterone receptor (10/11), PAX2 (18/19), and HER2 (25/26). p53 showed “wild-type” and “mutation-type” staining in 27 of 46 and 19 of 46 cases, respectively. MMR protein expression was retained in 19 of 20 cases with loss of MSH6 staining in 1 patient with Lynch syndrome. Molecular studies for HPV were undertaken in 2 tumors, which exhibited diffuse “block-type” immunoreactivity with p16, and both were negative. This is the first detailed immunohistochemical study of a large series of gastric-type adenocarcinomas of the lower female genital tract. Our results indicate immunophenotypic overlap with pancreaticobiliary adenocarcinomas but suggest that PAX8 immunoreactivity may be especially useful in distinguishing gastric-type adenocarcinomas from pancreaticobiliary and other nongynecologic adenocarcinomas, which are usually negative. Diffuse “block-type” p16 immunoreactivity in a cervical adenocarcinoma is not necessarily indicative of a high-risk HPV-associated tumor.

GATA3 is a transcription factor critical for embryogenesis, development, and cell differentiation. Recent studies have suggested that GATA3 is a sensitive and relatively specific biomarker for urothelial and breast carcinomas, with most Müllerian carcinomas being negative. We investigated GATA3 expression in mesonephric/Wolffian remnants and tumors in the female genital tract. A western blot was performed to assess specificity for the GATA3 antibody. GATA3 immunohistochemistry was performed on 59 formalin-fixed paraffin-embedded mesonephric samples, including 17 mesonephric remnants (MR; 11 cervical and 6 fallopian tube), 15 mesonephric hyperplasias, 21 mesonephric carcinomas, and 6 female adnexal tumors of probable Wolffian origin. Thirty conventional endocervical adenocarcinomas (ENDO-CA), 9 gastric-type cervical adenocarcinomas, and 165 endometrial adenocarcinomas (EM-CA) were also evaluated. GATA3 nuclear intensity and extent of staining was evaluated. The western blot revealed GATA3 expression in seminal vesicle and cell lines derived from breast and urothelial carcinomas, but not in other cell lines including ovarian, cervical, and endometrial cancers. All cervical MRs and mesonephric hyperplasias, 5/6 (83%) fallopian tube MRs, and 20/21 (95%) mesonephric carcinomas were GATA3 positive, although with great variability in both intensity (weak to strong) and extent (1+ to 3+) of staining. Only 1/6 (17%) female adnexal tumors of probable Wolffian origin showed weak multifocal staining. One of 30 (3%) usual-type ENDO-CAs and 3/165 EM-CAs exhibited weak-moderate GATA3 immunoreactivity; all gastric-type cervical adenocarcinomas were negative. GATA3 is a highly sensitive and specific marker for mesonephric lesions in the lower genital tract; however, its utility in the upper genital tract may be more limited. In addition, GATA3 can aid in distinguishing lower genital mesonephric lesions from usual-type and gastric-type ENDO-CAs and uterine EM-CAs.

Since our first description of the microcystic stromal tumor (MST) of the ovary, a rare and distinctive neoplasm with a definitional, usually striking microcystic pattern and a CD10+/vimentin+/inhibin−/calretinin−immunophenotype, 3 examples with β-catenin nuclear localization, and CTNNB1mutation have been reported. We undertook a detailed immunohistochemical study and molecular analysis of CTNNB1and FOXL2of 15 cases of MST to further characterize this neoplasm and establish its histogenesis. Diffuse nuclear staining for FOXL2, WT-1, cyclin D1, and β-catenin was present in all tumors tested, and 12/15 were positive for steroidogenic factor-1 (SF-1). Heterozygous missense point mutations in exon 3 of CTNNB1were detected in 8 of 14 cases, resulting in amino acid changes at codons 32, 34, 35, and 37. There was no correlation between CTNNB1exon 3 mutation status and tumor immunophenotype. All 14 cases tested showed wild-type FOXL2. Our study establishes that MST of the ovary exhibits a characteristic FOXL2/SF-1/WT-1/cyclin D1/nuclear β-catenin–positive immunohistochemical profile, which may be useful in diagnosis and in the exclusion of histologic mimics. The presence of diffuse nuclear FOXL2 and WT-1 immunostaining in all cases and SF-1 in most supports the classification of MST within the sex cord-stromal category. Aberrant nuclear β-catenin expression, detected in all MSTs, appears to be the result of stabilizing CTNNB1mutations in 57% of cases, providing further evidence that dysregulation of the Wnt/B-catenin pathway is involved in the tumorigenesis of MST and may involve activation of β-catenin with upregulation of cyclin D1.

Atypical polypoid adenomyoma (APA) is an uncommon uterine lesion that commonly recurs after local excision and is occasionally associated with or precedes the development of atypical hyperplasia or endometrioid adenocarcinoma. Despite the fact that about 230 cases have been reported in the literature, only 2 studies of 6 and of 7 cases have investigated the molecular aspects; as such, molecular alterations that occur in APA remain largely unknown. We undertook a comprehensive immunohistochemical and molecular analysis of 21 cases of APA in 17 patients (including 4 recurrent/persistent lesions). The analyzed genes were PTENand TP53(by fluorescence in situ hybridization) and KRAS, BRAF, EGFR, and NRAS(all by polymerase chain reaction). Immunohistochemical staining was performed for PTEN, p53, mTOR, β-catenin, HNF-1β, and GLUT1 and for the mismatch-repair proteins MLH-1, MSH-2, MSH-6, and PMS-2. In most cases, there was nuclear expression of β-catenin in squamous morules and expression of HNF-1β, mTOR, and GLUT1 in the glandular component. All cases exhibited “wild-type” expression of p53. A common finding was loss of PTEN expression (6/19 cases). In 1 of these cases, loss of PTEN expression was accompanied by PTENdeletion. Mutation of the KRASgene was found in 5/19 cases. Intact mismatch-repair protein expression was present in all cases, and TP53abnormalities or mutations of EGFR, NRAS, or BRAFgenes were not found. Given the association with atypical hyperplasia and endometrioid adenocarcinoma and the shared immunohistochemical and molecular features, we feel that, conceptually, APA is best regarded as analogous to a localized form of atypical hyperplasia.

Seromucinous neoplasms are a new category of ovarian epithelial tumor in the revised World Health Organization Classification of Tumours of the Female Reproductive Organs. Borderline variants are well described, but there have been few reports of seromucinous carcinomas. We report the clinicopathologic features in 19 cases of ovarian seromucinous carcinoma in patients aged 16 to 79 years (mean 47). In 16 cases, the neoplasm was unilateral and in 3 cases bilateral. The tumors ranged in size from 1.8 to 18 cm (mean 10.5 cm). The tumors were stage I (n=15), stage II (n=1), and stage III (n=3). The histologic features were highly variable both within and between individual tumors. The majority of neoplasms (12 cases) exhibited a predominant papillary architecture with lesser components of glandular, microglandular, and solid growth. A predominant glandular architecture was present in 6 cases, whereas 1 had a predominantly solid growth. A characteristic feature was an admixture of cell types. Most of the tumors (15 cases) were mainly composed of endocervical-like mucinous cells, whereas in 4 cases there was predominant endometrioid differentiation. Other cell types, present in varying proportions, included hobnail cells, eosinophilic cells, squamous cells, clear cells, and signet-ring cells. An infiltrate of neutrophil polymorphs was a prominent feature in most cases. Most cases also exhibited areas of microglandular architecture with cytoplasmic clearing and intraluminal polymorphs, the features closely resembling cervical microglandular hyperplasia. Areas of stromal hyalinization, adenofibromatous growth, and psammoma bodies were present in a minority of cases. Endometriosis was identified in the same ovary in 10 cases, and in 10 there was a component of seromucinous borderline tumor. Thirteen, 5, and 1 tumor were of grades 1, 2, and 3, respectively (using the FIGO grading system for endometrioid adenocarcinomas of the uterine corpus). A synchronous uterine endometrioid adenocarcinoma was present in 1 case. Immunohistochemically, there was positive staining with CK7 (17/17 cases), estrogen receptor (16/16 cases), progesterone receptor (6/7 cases), CA125 (15/15 cases), PAX8 (8/8 cases), CEA (8/13 cases), CA19.9 (8/9 cases), and WT1 (2/13 cases). CK20 and CDX2 were negative in all cases tested (16 and 14, respectively). p53 showed “wild-type” staining in 4/4 cases, and p16 was focally positive in 5/5 cases. Follow-up information was available in 8 patients. Seven were alive with no evidence of disease (follow-up 3 to 74 mo), whereas 1 patient who initially presented with a stage IIB tumor died of disease at 192 months. Given the characteristic admixture of cell types and the overlapping morphologic features with low-grade serous, mucinous, and endometrioid neoplasms, the most appropriate categorization of seromucinous carcinomas is uncertain, but we believe they are best regarded as a distinct type of ovarian epithelial malignancy and are most similar to endometrioid adenocarcinomas. We recommend grading them in an analogous manner to ovarian endometrioid adenocarcinomas.

Supplemental Digital Content is available in the text.Loss of expression of the SMARCB1 (INI1/BAF47/SNF5) tumor-suppressor protein, originally identified in pediatric malignant rhabdoid tumors, has been noted in significant percentages of epithelioid sarcomas of classical and proximal-type and in myoepithelial carcinomas. Epithelioid sarcoma and myoepithelial carcinoma are very rare in the vulvar region, and few of these cases have been evaluated for SMARCB1 protein loss by immunohistochemistry (IHC) or for SMARCB1gene alterations by molecular genetic techniques. We studied the clinicopathologic, IHC, and molecular genetic features of 14 SMARCB1-deficient vulvar neoplasms. All available routinely stained sections were reexamined, and IHC analysis for wide-spectrum cytokeratins, high–molecular weight cytokeratins, epithelial membrane antigen, S100 protein, CD34, smooth muscle actin, desmin, and SMARCB1 was performed. Multiplex ligation-dependent probe amplification and DNA sequencing of the SMARCB1gene was performed on 12 cases with sufficient available tissue. The 14 vulvar tumors occurred in adult women (mean age 46 y, range 22 to 62 y) and measured 1.1 to 8.8 cm in size (mean 4.7 cm). Tumors were classified as classical-type epithelioid sarcoma (N=1), proximal-type epithelioid sarcoma (N=6), myoepithelial carcinoma (N=4), and “SMARCB1-deficient vulvar sarcoma, not otherwise specified” (N=3) on the basis of combined histopathologic and IHC findings. One myoepithelial carcinoma showed divergent rhabdomyoblastic differentiation. All tested cases showed partial or complete SMARCB1deletions (homozygous: 9 cases; heterozygous: 3 cases). One case with a heterozygous deletion also showed a c.528delC mutation in exon 5. Fluorescence in situ hybridization for EWSR1rearrangement was performed for 3 cases classified as myoepithelial carcinoma and was negative. Follow-up (13 patients, range 5 to 72 mo, mean 31 mo) data showed 3 patients dead of disease, 1 alive with unresectable metastatic disease, 1 alive with radiographic evidence of extensive lymph nodal disease, and 8 alive without disease. We conclude that SMARCB1-deficient vulvar neoplasms chiefly comprise epithelioid sarcoma and myoepithelial carcinoma, although some defy easy classification. No association was seen between clinical behavior and the type of SMARCB1alteration.

Most nonuterine high-grade serous carcinomas (HGSCs) in women with hereditary breast and ovarian cancer syndrome, due to germline BRCA12mutation, arise in the fimbria of the fallopian tube. However, the site of origin of sporadic HGSC, which is usually widely disseminated at presentation, is not well established. We sought to characterize cases of HGSC discovered incidentally in patients not known to be at high risk, in order to determine the site distribution and possible origin of sporadic HGSC. Incidental microscopic, non–mass-forming cases of serous tubal intraepithelial carcinoma or HGSC in salpingo-oophorectomy specimens in which the tubes and ovaries had been extensively examined were identified. No patients were known or suspected BRCA12 mutation carriers. Twenty-one cases were identified (mean age: 57 y). Surgery was for benign disease (n=15), uterine endometrioid adenocarcinoma or atypical hyperplasia (n=3), bladder carcinoma (n=1), or ovarian serous borderline tumor (n=2). In 16 of 21 cases, the lesion was confined to the fallopian tube (unilateral in 14 cases, bilateral in 2). There was serous tubal intraepithelial carcinoma in all cases and invasive HGSC into the underlying lamina propria in 8 of these 16 cases; the invasive focus measured 1.3 cm or less in every case. In the remaining 5 cases, there was fallopian tube mucosal and ovarian involvement; in 2 of these cases, there was also microscopic peritoneal involvement. Sporadic cases of nonuterine HGSC arise in the fallopian tube fimbria in a large majority of cases, providing further evidence for the tubal origin of these neoplasms.

Primary intestinal-type glandular lesions of the vagina are rare. We report a series of 14 lesions, including 1 intestinal-type polyp without neoplastic features, 3 adenomas (2 with high-grade dysplasia), and 10 adenocarcinomas. Patients ranged in age from 20 to 86 years (mean 60 y) and presented with vaginal bleeding or a mass. No history of diethylstilbestrol exposure, adenosis, or endometriosis was elicited in any patient. The lesions were mostly polypoid, small (0.8 to 2.0 cm), and located in the posterior (6 cases) and lower (7 cases) vagina. One carcinoma metastasized to a para-aortic lymph node; the others were confined to the vagina. The neoplasms exhibited histologic features identical to those seen in primary large intestinal tumors, including variable numbers of goblet cells and in 1 case neuroendocrine cells. Five of the adenocarcinomas contained areas consistent with a precursor adenoma. In 3 cases, a benign urothelium-lined duct was adjacent to the lesion, and in 2 patients benign intestinal-type epithelium was present; no other potential benign precursor lesions were seen. Immunohistochemical analysis was performed on 6 cases; the tumors were positive for CDX-2 (66), CK20 (56), CEA (55), CK7 (46), and CA-125 (24) and were negative for ER (06) and p16 (02). Clinical outcome data were available in 3 patients with adenocarcinomas; 1 died of disease in <1 year, and 2 were alive with no evidence of disease at 2 and 7 years. The pertinent literature is reviewed, and the potential origin and differential diagnosis of these lesions are discussed.

Ovarian cellular fibromas are uncommon neoplasms, which may result in considerable diagnostic confusion with diffuse adult granulosa cell tumor. This is an important distinction, as the former usually exhibits benign behavior, whereas the latter is a low-grade malignant neoplasm capable of recurrence and metastasis. FOXL2mutation (402C→G) has been demonstrated in >95 of ovarian adult granulosa cell tumors, only rarely in other ovarian sex cord-stromal neoplasms, and never in ovarian fibromas. In this study, we evaluated a series of ovarian cellular fibromas or mitotically active cellular fibromas (n=22), 3 with minor sex cord elements, for FOXL2mutation. These were mostly received in consultation, often with a differential diagnosis of diffuse adult granulosa cell tumor. Immunohistochemically, 10 of 10 cases tested exhibited nuclear staining with FOXL2. FOXL2(402C→G) mutation was not demonstrated in any of the 22 cellular or mitotically active cellular fibromas. Three additional neoplasms composed of cellular nodules of epithelioid cells in a background fibrous stroma, raising the possibility of adult granulosa cell tumor with a prominent fibrothecomatous component, were also tested; 2 of these were mutation negative, and 1 contained a FOXL2mutation. FOXL2mutation analysis is a useful adjunct in distinguishing between diffuse adult granulosa cell tumor (mutation present) and cellular fibroma (mutation absent). Mutation testing should be considered in problematic cases, as this will provide prognostic information for the patient.

Intravascular leiomyomatosis (IVL) is characterized by the presence of smooth muscle in venous and lymphatic spaces within the myometrium. Although the intravascular component usually consists solely of typical smooth muscle or variants of smooth muscle differentiation, we report 5 cases in which the intravascular component also included endometrioid glandular and stromal elements. We propose the term “intravenous adenomyomatosis” to describe this unusual variant of IVL. The mean age of the patients in this series was 50.2 years, slightly older than that of patients with conventional IVL. In addition to intravenous adenomyomatosis, both adenomyosis and leiomyomas were identified in all of our cases, supporting the hypothesis that the intravascular smooth muscle component in IVL is derived from associated myometrial pathology rather than from vessel walls. In our series, intravenous adenomyomatosis had a similar benign clinical behavior to most cases of IVL with no metastatic or recurrent disease identified at follow-up in 4 cases for which follow-up information was available. The main differential diagnoses are adenomyosis with vascular involvement, low-grade endometrial stromal sarcoma (ESS), including ESS with smooth muscle and glandular differentiation, and adenosarcoma with lymphovascular invasion. The possibility of intravenous adenomyomatosis should be borne in mind when considering these diagnoses, particularly ESS and adenosarcoma, which have different implications for patient management and prognosis.

In its classical form, embryonal rhabdomyosarcoma (ERMS, botryoid type) is a vaginal neoplasm occurring in infants and young girls and is often not considered in the differential diagnosis of uterine corpus and cervical spindle cell tumors in adult women. Clinicopathologic and immunohistochemical features of 25 cases of ERMS in women 20 years of age or older were analyzed. Patient age ranged from 20 to 89 years (mean, 44.4 y; median, 46 y), with 8 patients aged 20 to 39 years, 14 patients aged 40 to 59 years, and 3 patients older than 60 years of age. Tumors originated in the cervix in 20 cases and in the uterine corpus in 5. They were characterized by an edematous hypocellular spindle cell proliferation, typically with cellular condensation beneath epithelial surfaces (cambium layer), in which tightly packed hypercellular foci were scattered. Neoplastic cells had hyperchromatic nuclei and minimal cytoplasm, usually with delicate cytoplasmic processes. Occasionally, elongated or globular cells with eosinophilic cytoplasm (rhabdomyoblasts) were evident, but cytoplasmic cross-striations were only rarely identified. Apoptotic bodies and mitotic figures were usually identified in the hypercellular foci. Hemorrhage was common, often making recognition of the hypercellular foci difficult. Desmin and myogenin were coexpressed in 22 of 23 (95.6) tumors evaluated. Proliferative activity, as assessed by Ki-67 expression, was notably elevated in all tumors evaluated, typically concentrated in the hypercellular foci. Estrogen and progesterone receptors were expressed focally in only 3 of 12 (25) and 1 of 8 (12.5) tumors evaluated, respectively. Follow-up was available in 7 cases. Five patients were alive without evidence of disease with follow-up of 3 to 8 years, and 1 patient was alive with disease at 5 months. One patient died at 5 months with pulmonary nodules, but it was not determined whether this was due to metastatic ERMS or the patient’s known ductal breast carcinoma. ERMS has a broader clinical profile than classically expected and should be considered in the differential diagnosis of a uterine corpus or cervical spindle cell tumor, regardless of patient age. Recognition can be rendered difficult by the hypocellular background, which can suggest a benign polyp or low-grade tumor, and hemorrhage, which can obscure the characteristic hypercellular foci. Identification of hypercellular foci in which mitotic activity and apoptotic bodies are found, desmin and myogenin are coexpressed, proliferative activity is notably elevated, and hormone receptor expression is usually absent is very useful for establishing the diagnosis.