Impact of centralization in primary retroperitoneal sarcoma treatment: analysis using hospital-based cancer registry data in Japan.

ABSTRACT

BACKGROUND: To elucidate the clinicopathological features, hospital-based care volume and prognoses associated with primary retroperitoneal sarcoma (PRS). METHODS: Clinical data on PRS cases, diagnosed from 2008 to 2009 (cohort A) and from 2012 to 2015 (cohort B), were obtained from the national hospital-based cancer registry in Japan. Since data on survival, 5 years after PRS diagnosis, were available only for cohort A, patient prognoses were analyzed in this group alone. RESULTS: The numbers of participating hospitals were 154 in cohort A and 537 in cohort B. In total, 380 and 2011 patients with PRS were identified in cohorts A and B, respectively. The incidence of PRS among all the registered urogenital malignancies was 0.52% (2391/462,866). Liposarcoma was the most commonly observed PRS subtype (55.8%), followed by leiomyosarcoma (19.0%). In cohort A, the 5-year overall survival (OS) was 40.4%. The 5-year OS associated with stage I (n = 107), stages II and III (n = 61), and stage IV (n = 59) disease were 59%, 39%, and 6%, respectively. Only two institutions treated over ten patients per year in each cohort. When institutions were divided by hospital care volume (8 hospitals with ≥ = 3 cases and 149 with < 3 cases/year), there were any statistic differences in the OS. CONCLUSIONS: We presented the distribution and prognoses associated with PRS using a real-world large cohort database. Centralization for PRS management was not established in Japan, while the prognosis did not significantly depend on the treatment volume of hospitals.