Objective: Discrepancies may exist between an original pathology report and formal pathology review, with subsequent implications for treatment. We conducted a study of pathology review in endometrial cancer from a population-based study to identify areas of discrepancy and effect on treatment.
Methods: This was a retrospective cohort study in Ontario, Canada from 1996 to 2000. We identified hysterectomy cases from patients with endometrial cancer that were subject to formal pathology review by a gynecologic pathologist at one of six tertiary care centers. Sarcomas and other rare histologic subtypes with fewer than five cases were excluded. We evaluated discrepancy between original pathology and review by demographics, stage, grade, and risk group. Four risk groups were defined: 1) low (stage I), 2) intermediate (stage I and II), 3) high-risk (stage I and II), and 4) advanced stage (all stage III and IV). Reclassification from one risk group to another upon pathology review represented a potential change in treatment. Factors associated with significant discrepancy were identified by a multivariable logistic regression model.
Results: Formal pathology review was available on 450 cases. There were no differences by age, year, or hospital type. The overall discrepancy rate was 42.7% (95% confidence interval 38.2-47.3%). The intermediate-risk group had the highest rate of reclassification into another group (33.1%). The most significant rates of discrepancy were associated with endometrioid grades 2 and 3 tumors and stage IIA disease (39.8%, 50.9%, and 79.6%, respectively).
Conclusion: There was significant discrepancy between original pathology and formal review in endometrial cancer, with implications for guidelines on pathology review at a population level.
Level of evidence: III