The Malignant Potential of Ovarian Steroid Cell Tumors Revisited: A Multi-institutional Clinicopathologic Analysis of 115 Cases.

Steroid cell tumors (SCTs) of the ovary are rare and understudied, and as such, uncertainties remain about their malignant potential, as well as clinicopathologic predictors of patient outcome. Based on a multi-institutional cohort of cases, we present findings from the largest study of SCT reported to date. Clinicopathologic data were documented on 115 cases of SCT that were assembled from 17 institutions. The median patient age was 55 years (range: 9 to 84). When measured, preoperative androgen levels were elevated in 84.2% (48/57) of patients. A total of 111 (96.5%) cases were classified as stage I (103 stage IA; 2 stage IB; 6 stage IC). The stage distribution for the remaining 4 patients was as follows: stage II (n = 1), III (n = 3; 1 IIIA, 1 IIIB, 1 IIIC). The median tumor size was 3 cm (range: 0.2 to 22). Cytologic atypia, microscopic tumor necrosis, microscopic tumor hemorrhage, and a mitotic index of >1 mitotic figure/10 high-power fields were present in 52% (60/115), 9.6% (11/115), 37% (43/115), and 19% (22/115) of cases, respectively. Of 115 patients, 7 (6.1%) recurred postexcision, 4 (3.5%) ultimately died of disease, and 10 (8.7%) either recurred, died of disease, or were advanced stage at presentation. The median duration to recurrence postresection was 33 months (range: 23 to 180). Four of the 7 recurrences were stage IA at baseline. Tumor size >4 cm, International Federation of Gynecology and Obstetrics (FIGO) stage ≥IB, tumor necrosis, and tumor hemorrhage were each significantly associated with reduced recurrence-free survival in log-rank tests and univariable Cox models, with age older than 65 years being of marginal significance (hazard ratio [HR]: 5.4, 95% CI: 1.0-30.0, P = 0.05). Multivariable analyses suggested that FIGO stage ≥IB (HR: 27.5, 95% CI: 2.6-290.5), and age older than >65 years (HR: 21.8, 95% CI: 1.6-303.9) were the only parameters that were independently associated with recurrence. Cross-section analyses showed that tumor necrosis, tumor hemorrhage, and larger tumor size were significantly associated with a FIGO stage ≥IB status, which bolstered the conclusion that they are not independent predictors of recurrence. In summary, <10% of SCTs are clinically malignant, a substantially lower frequency than has previously been reported in the literature. Clinicopathologic predictors of patient outcomes that are prospectively applicable in practice could not be definitively established. Recurrences may occur many years (up to 15 y in this study) after primary resection, even in stage IA cases.

GATA3 immunohistochemistry as a diagnostic adjunct for differentiated vulvar intraepithelial neoplasia: utility and limitations.

Herein, the authors evaluate the diagnostic utility and limitations of GATA3 immunohistochemistry for the distinction of differentiated vulvar intraepithelial neoplasia (dVIN) from its potential mimics. Immunohistochemical studies for GATA3, p53, and p16 were performed on 124 pathologic vulvar tissues, inclusive of dVIN (n = 21), vulvar aberrant maturation (n = 10), high-grade squamous intraepithelial lesion (HSIL) (n = 44), and 49 non-neoplastic vulvar dermatoses of various types. GATA3 expression was scored using a modification of previously proposed criteria: pattern 0 (no significant loss of basal layer staining, >75% staining), pattern 1 (25-75% staining), and pattern 2 (<25% staining). With the exception of lichen sclerosus, 8% of which showed pattern 1 or 2 staining, all other non-neoplastic lesions showed pattern 0 expression. Aberrant GATA3 expression (i.e., patterns 1 or 2) was present in 90% of dVIN cases (2 [9.5%], 3 [14.3%], 16 [76.2%] with patterns 0, 1, and 2 respectively), 90% of vulvar aberrant maturation cases (1 [10%],7 [70%], 2 [20%] with patterns 0, 1, and 2 respectively), and 15.9% of HSIL cases (84.1% pattern 0; 2.3% pattern 1; 13.6% pattern 2). All HSIL cases were p16 positive, including the 7 pattern 1 and 2 cases. All cases of dVIN-like HSIL were pattern 0, and all (n = 2) cases of HSIL-like (basaloid) dVIN were pattern 2 (both of the latter cases displayed complete absence of epidermal staining). Only 1 dVIN case was both pattern 0 and p53-wild-type. We conclude that GATA3 is useful for the distinction of dVIN from non-neoplastic dermatoses and from HSIL, but is best used as part of a panel that includes p53 and p16 to mitigate its limitations.

Periductal Stromal Tumor of the Breast: Case Report and Literature Review.

Periductal stromal hyperplasia is an exceedingly rare biphasic breast tumor with benign ductal elements and spindle-cell stromal proliferation lacking a phyllodes architecture. It is unclear whether this tumor belongs to the spectrum of phyllodes tumor or is a separate entity. We present a 57-year-old woman with a 3.0×2.3×1.0 cm lesion in the left breast. Sections of the excision specimen showed nodular hypercellular spindle cell proliferation with ductal centric arrangement and infiltration into the adipose tissue. The spindle cells showed bland nuclei, no visible mitoses, no leaflike growth pattern, and low Ki-67(<1%). Spindle cells were positive for CD34, and negative for SMA, myosin, CD117, and beta-catenin. Presentation of this case helps in further defining the clinical, histopathologic, management and outcome of this tumor.