Tumor Type and Substage Predict Survival in Stage I and II Ovarian Carcinoma: Insights and Implications

Abstract

Tumor grade has been the most accepted prognostic indicator of disease-specific survival among women with stage I and II ovarian carcinomas. Many investigators have included stage IB, grade 1 at diagnosis as low-risk disease and have suggested that women in this category should not receive adjuvant therapy. However, grading assignment appears to be unreliable because of problems with reproducibility and lack of consideration of biological differences between the different tumor types. Some investigators believe that classification based on tumor typing using new histopathological criteria was more reproducible than grade assignment and would more accurately reflect biological differences. This population-based cohort study evaluated the effectiveness of updated tumor type to predict outcome for patients with low-stage ovarian carcinoma at low risk of death. Between 1984 and 2003, 1326 women with a diagnosis of stage I–II ovarian carcinoma were referred to the British Columbia Cancer Agency. Of these, 652 (49%) were available for full slide review. Type and grade were assessed using contemporary histopathological criteria. Tumor types investigated accounting for 98% of ovarian carcinomas included high-grade serous, endometrioid, clear cell, mucinous, and low-grade serous. Of the 630 cases confirmed as ovarian carcinomas, 25 were rare types and were excluded, leaving 605 cases available for the final analysis. Outcomes among different subtypes were assessed using univariate models and recursive partitioning analysis. An excellent outcome (low risk of death) was defined as a disease-specific survival at 10 years (DSS10y) ≥95%. There was an excellent outcome in 77 cases with ovarian carcinomas of endometrioid and mucinous type, stage IA or IB. The outcome for high-grade serous carcinomas was poor (DSS10y = 57%); no subset could be identified with an excellent outcome. Although no subsets of clear cell carcinomas had an excellent outcome, a small group of patients with stage IA or IB tumors had a more favorable outcome (DSS10y = 87%) compared to those with stage IC–II (DSS10y = 66%). These findings indicate that tumor type is superior to grade in identifying the risk of death from ovarian carcinoma among women with stage I/II disease. Classification of all stage IA or IB endometrioid and mucinous ovarian carcinomas at time of presentation as low-risk would allow the proportion of patients for which adjuvant treatment could be avoided to be increased from 6.5% to 12.2%.