Inter-and intra-observer reliability in histologic evaluation of necrosis rate induced by neo-adjuvant chemotherapy for osteosarcoma

Abstract

Purpose: Tumor necrosis rate following neo-adjuvant chemotherapy is one of the most important prognostic factors for patients with osteosarcoma, and it also provides the basis for selection of postoperative adjuvant chemotherapy. However, reported necrosis rates for the same tumors can vary among pathologists, complicating decision making for further treatment. Methods: Ten H&E stained pathology slides from 10 osteosarcoma patients treated with neo-adjuvant chemotherapy were randomly selected. Six expert pathologists were assigned to analyze the slides for tumor necrosis rate at four time points, with an interval of 3 weeks. Intraclass and interclass correlation coefficients (IntraCC & InterCC) and 95% confidence intervals (CI) were calculated. Results: The overall InterCC among the 6 observers was 0.652 (95% CI 0.294-0.820) for tumor necrosis rate (range: 0-100%), suggesting good reliability. IntraCCs were 0.799, 0.788, 0.867, 0.935, 0.962, and 0.947 respectively. The interCC among higher careers and lower careers were 0.603 (95% CI.: 0.387-0.843) and 0.696 (95% C. I.: 0.487-0.919), respectively, which was not significantly different. Major differences in tumor necrosis estimation were due to interpretation of areas with isolated atypical cells in fibrotic stroma. Conclusion: Low inter-observer and relatively high intra-observer reliability were observed in the histologic evaluation of necrosis rate after neo-adjuvant chemotherapy. The interCCs between the higher career and lower career groups did not vary. These findings suggest that a valid measurement protocol for tumor necrosis rate evaluation after chemotherapy is required to improve the clinical relevance of the quantification of response to neo-adjuvant therapy.