Douglas Rottmann

Reproducibility of scoring criteria for HER2 immunohistochemistry in endometrial serous carcinoma: a multi-institutional interobserver agreement study

Targeted anti-human epidermal growth factor receptor 2 (HER2) therapy has recently been proven to improve progression-free and overall survival of patients with advanced stage or recurrent endometrial serous carcinoma. To date, no specific pathology HER2 testing or scoring guidelines exist for endometrial cancer. However, based on evidence from the recent successful clinical trial and comprehensive pre-trial pathologic studies, a new set of HER2 scoring criteria have been proposed for endometrial serous carcinoma-distinct from the existing breast and gastric cancer-specific criteria. We present the first study assessing interobserver agreement of HER2 scores using the proposed serous endometrial cancer-specific scoring system. A digitally scanned set of 40 HER2-immunostained slides of endometrial serous carcinoma were sent to seven gynecologic pathologists, who independently assigned HER2 scores for each slide following a brief tutorial. Follow-up fluorescent in situ hybridization (FISH) for HER2 gene amplification was performed on cases with interobserver disagreement when a 2+ HER2 score was assigned by at least one observer. Complete agreement of HER2 scores among all 7 observers was achieved on 15 cases, and all but one case had an agreement by at least 4 observers. The overall agreement was 72.3% (kappa 0.60), 77.5% (kappa 0.65), and 83.3% (kappa 0.65), using four (0 to 3+ ), three (0/1+ , 2+ , 3+ ), or two (0/1+ , 2/3+ ) HER2 scoring categories, respectively. Based on the combination of HER2 immunostaining scores and FISH, the interobserver disagreement may have potentially resulted in a clinically significant difference in HER2 status only in three tumors. We conclude, that the proposed serous endometrial cancer-specific HER2 scoring criteria are reproducible among gynecologic pathologists with moderate to substantial interobserver agreement rates comparable to those of previously reported in breast and gastric carcinomas. Our findings significantly strengthen the foundation for establishing endometrial cancer-specific HER2 scoring guidelines in the future.

Minimal uterine serous carcinoma and endometrial polyp: a close clinicopathological relationship

Frequently involving an endometrial polyp, minimal uterine serous carcinoma (MUSC) represents the earliest recognizable forms of endometrial serous carcinoma. The aim of this study was to provide a comprehensive morphological and clinical outcome assessment of MUSC involving endometrial polyp. A total of 77 fully staged MUSCs involving endometrial polyp were identified, including 53 MUSCs confined to polyp and 24 nonpolyp confined tumors. Extrauterine disease was found in 17% (9/53) of polyp-confined MUSCs compared to 41.7% (10/24) of nonpolyp confined tumors (p = 0.02). Lymphovascular invasion was observed in 3.8% (2/53) of polyp-confined cases compared to 25% (6/24) of nonpolyp confined cases (p = 0.047). Lymph node metastasis was observed in 11.3% (6/53) of polyp-confined cases, compared to 29.2% (7/24) of nonpolyp confined cases (p = 0.058). Positive pelvic washing cytology was seen in 18.9% (10/53) of polyp-confined versus 37.5% (9/24) of nonpolyp confined tumors (p = 0.078). Overall, 58 of 77 (75.3%) patients had low tumor stage (57 stage I cases and 1 stage II case) and only two patients (3.5%) had a recurrence. In contrast, 19 of 77 (24.7%) patients had advanced stage (stage III or IV) disease and 17 (89.5%) patients had recurrence (p < 0.0001). Only one of 57 low-stage patient (1.7%) versus 11 of 19 high-stage patients (57.8%) died of the tumor (p < 0.0001). Five of 53 (9.4%) patients with polyp-confined MUSC and 7 of 24 (29.2%) patients with nonpolyp confined MUSC died of the disease (p = 0.03). In conclusion, while a small percentage of MUSCs exist without the involvement of an endometrial polyp, a close topographic relationship between MUSC and the endometrial polyp is confirmed in this largest series, supporting the theory that most if not all MUSCs arise in an EMP. Patients with MUSC without extrauterine spread have an excellent prognosis. Compared to patients with MUSC confined to an endometrial polyp, patients with MUSC extending to the background endometrium have a significantly higher risk for high-stage disease at presentation.