Validation of testicular germ cell tumor staging in nationwide cancer registries.

Abstract Introduction Nationwide cancer registries such as the National Cancer Database and Surveillance, Epidemiology, and End Results rely on accurate data from tumor registries to formulate hypotheses and report outcomes and treatment patterns. We evaluated the accuracy of our institutional registry for testicular germ cell tumors by comparing data abstracted by urologists with data abstracted by registry. Methods We performed a retrospective review of patients receiving initial diagnosis and treatment for germ cell tumors at our hospital system from 2005 to 2016. We compared coding for American Joint Committee on Cancer TNMS staging, overall composite stage, and first-line treatment between urologists and tumor registry at the time of diagnosis. Results Paired staging from registry and urologist was available for 80 patients. T, N, M, and S-staging were accurate for 90%, 81%, 94%, and 54% of records, respectively. Composite staging and first-line treatment were concordant for 39% and 90% of patients, respectively. A separate review of 33 Stage IS patients per registry for composite staging revealed 15% concordance. Conclusion Our institutional tumor registry had substantial inconsistencies in accurately staging N stage, S stage, and thus, composite stage for testicular cancer. An educational intervention to improve abstraction by registry led to increased concordance. Assuming similar discrepancies may exist at other institutions and for other cancer types, caution should be used when interpreting staging data in nationwide cancer registries. This sheds light on the need for improved clarification of staging guidelines, dynamic institutional internal auditing, and training reform within cancer registries.