PATH-09. SJMB12 CLINICAL TRIAL: DISCREPANCY BETWEEN LOCAL AND CENTRAL PATHOLOGY IN ASSESSING ANAPLASTIC MEDULLOBLASTOMA – REPORT FROM A SINGLE SITE EXPERIENCE

Abstract INTRODUCTION SJMB12 is a phase 2 clinical trial led by the St. Jude Children’s Research Hospital (St. Jude) that enrolls patients with medulloblastoma based on their biological subgroup. The large cell/anaplastic (LCA) histologic variant has been identified as an important independent risk factor associated with poor outcome. However, the histologic criteria for LCA is subjective, making the distinction between anaplastic and non-anaplastic medulloblastoma difficult in some cases. METHODS Pathological central review was performed at St. Jude. For all patients enrolled in the study to date, concordance was assessed between the initial and central review diagnosis and histologic variant calls made at the Royal Children’s Hospital Melbourne (RCH) and at St. Jude, respectively. RESULTS Since the SJMB12 clinical trial opened locally in 2014, 34 patients were enrolled, and 31 were eligible for this retrospective study. A total of 12 (39%) cases with discordance were identified. The most frequent disagreement was between the designation of LCA (10 cases, 32%). In five cases the tumour was not designated as LCA variant locally. In five cases the initial designation of LCA was refuted centrally. Overall, this led to a change of treatment stratum for four patients (13%). CONCLUSION A high discordance rate exists between neuropathologists in the designation of LCA variant. Differences in interpretation of the subjective histologic criteria and inconsistencies in the material submitted for central review contributed to the discordance. Incorporation of more objective histologic criteria and implementation of unbiased diagnostic tools may improve the generalisability of future risk stratification.