Distribution and case-fatality ratios by cell-type for ovarian carcinomas: a 22-year series of 562 patients with uniform current histological classification.

BACKGROUND: Ovarian carcinoma is comprised of several different cell types reflecting different clinicopathologic features. Pathologic criteria for distinguishing cell types have evolved, and therefore non-contemporary literature on ovarian cancer may have limited current relevance. A new dualistic model of pathogenesis that distinguishes type I (endometrioid, mucinous, clear cell and low grade serous carcinomas) from type II (high grade serous carcinomas and carcinosarcomas) tumors has become widely accepted.

METHODS: A cohort of 562 patients with invasive ovarian carcinoma from a large community hospital practice was reviewed. Cell type, FIGO stage, mortality and interpathologist diagnostic reproducibility were analyzed.

RESULTS: Advanced stage ovarian carcinomas were type II in 86% of cases while low stage tumors were most often type I. Only 1.7% of type II tumors were confirmed to be stage I with comprehensive surgical staging. Type II tumors accounted for 85% of deaths, and clear cell carcinomas, 5% of deaths. Cell type-specific case-fatality ratios for type II tumors were 62% and 79% for high grade serous carcinoma and carcinosarcoma, respectively. For type I tumors, case-fatality ratios were 38%, 36%, 27% and 13% for low grade serous, clear cell, endometrioid and mucinous carcinomas, respectively. The kappa value for diagnostic reproducibility among 3 gynecologic pathologists was 0.83.

CONCLUSIONS: Current diagnostic criteria confirm that high grade serous carcinoma and carcinosarcoma account for the vast majority (85%) of ovarian cancer deaths. Cell type designation is highly reproducible among gynecologic pathologists. Type II tumors are rarely stage I (<2%) when comprehensively staged by a gynecologic oncologist.