Deep Learning for Grading Endometrial Cancer.

Endometrial cancer is the fourth most common cancer in women in the United States; the lifetime risk for developing this disease is approximately 2.8%. Precise histologic evaluation and molecular classification of endometrial cancer are important for effective patient management and determining the best treatment modalities. This study introduces EndoNet, which uses convolutional neural networks for extracting histologic features and a vision transformer for aggregating these features and classifying slides based on their visual characteristics into high- and low-grade cases. The model was trained on 929 digitized hematoxylin and eosin-stained whole-slide images of endometrial cancer from hysterectomy cases at Dartmouth-Health. It classifies these slides into low-grade (endometrioid grades 1 and 2) and high-grade (endometrioid carcinoma International Federation of Gynecology and Obstetrics grade 3, uterine serous carcinoma, or carcinosarcoma) categories. EndoNet was evaluated on an internal test set of 110 patients and an external test set of 100 patients from The Cancer Genome Atlas public database. The model achieved a weighted average F1 score of 0.91 (95% CI, 0.86 to 0.95) and an area under the curve of 0.95 (95% CI, 0.89 to 0.99) on the internal test, and 0.86 (95% CI, 0.80 to 0.94) for F1 score and 0.86 (95% CI, 0.75 to 0.93) for area under the curve on the external test. Pending further validation, EndoNet has the potential to support pathologists without the need of manual annotations in classifying the grades of gynecologic pathology tumors.

Recurrent pembrolizumab-induced mechanical small bowel obstruction in a patient with metastatic cervical cancer.

•Adverse events of pembrolizumab have been documented, but more severe gastrointestinal effects are not as well described.•We report a case of a patient with cervical cancer treated with pembrolizumab who developed small bowel obstruction (SBO)•Histological analysis and gastrointestinal workup points to pembrolizumab as likely cause of SBO.

A Mimicker of Differentiated Vulvar Intraepithelial Neoplasia: Reactive Atypia From Noncompliance With Lichen Sclerosus Therapy.

ABSTRACT: Differentiated vulvar intraepithelial neoplasia (d-VIN) is an HPV-independent precursor to vulvar squamous cell carcinoma. The histology of d-VIN lesions is difficult to differentiate from that of non-neoplastic epithelial disorders, especially lichen sclerosus (LS). The authors present a case of LS, where relying on histopathology alone could have led to misdiagnosis. The patient was a 17-year-old female patient with clinical features of vulvar dermatitis and LS for 2 years. She was counseled to apply clobetasol 0.05% to the affected area daily but reported no improvement after 6 months. A biopsy of the right labia majora revealed histologic findings typical of d-VIN and near-contiguous p53 expression. These features are characteristic of d-VIN. However, d-VIN is exceedingly rare in young patients. The case was reviewed by 6 dermatopathologists and gynecologic pathologists, who observed that the degree of inflammation would be unusual postclobetasol therapy and could be due to noncompliance. A review of the patient's chart revealed that she "does not always remember to apply" clobetasol. The patient's clinician confirmed that there were compliance issues, and the follow-up biopsy was negative for d-VIN. The case was signed out as LS, with a note describing the above, and to rebiopsy if concern persisted. The authors conjecture that inflammatory infiltrates in the biopsied area caused reactive atypia due to lack of adherence to treatment. Although the patient's age helped rule out d-VIN, similar cases in elderly patients may be occurring. Pathologists must be aware that reactive forms of untreated LS can mimic d-VIN, to avoid misdiagnosis.

Correlation of histopathologic findings with clinical predictors of disease recurrence and progression to vulvar carcinoma in patients with differentiated vulvar intraepithelial neoplasia (dVIN).

Objective: To evaluate predictors of recurrence and the risk of progression to carcinoma in patients with dVIN. Methods: 36 self-identified White patients with dVIN from 2011 to 2022 were identified. Demographics, treatment and clinical course were abstracted. Histopathologic features and IHC stains were reviewed by 2 subspecialty pathologists. Standard statistical analyses were applied. Results: Median cohort age was 70 years (range 39-91). Median follow-up was 29.5 months (range 1-123). All patients were Caucasian. 67% had lichen sclerosus (LS) adjacent to dVIN. 56% of patients had recurrent dVIN a median of 11 months from diagnosis. 14 patients had invasive squamous cell carcinoma of the vulva (SCCV) during the study period: 9 (25%) with synchronous dVIN, 5 (14%) developed SCCV after a median of 21.5 months (range 8-57). Patients treated with surgery were more likely to have recurrent/persistent dVIN (p = 0.04) and synchronous or progression to SCCV (p = 0.02) than patients treated with topical therapy. Excluding 9 women with synchronous SCCV, no initial treatment (observation, topical therapy, surgery) was superior at preventing recurrent/ progressive disease in isolated dVIN. Mutation-type p53 expression was identified in 18 (64%) and aberrant GATA3 staining/expression in 20 (56%) of cases. Aberrant GATA3 expression was associated with a higher frequency of synchronous/progression to SCCV (p < 0.05). Conclusion: dVIN has an aggressive clinical course in white patients with a high risk of recurrence/persistence and synchronous/progression to SCCV despite treatment. Close surveillance with a low threshold for additional biopsies is warranted. P53 and GATA3 IHC stains may be useful markers of disease outcome.