Pathology review for paediatric non-Hodgkin's lymphoma patients in Japan; a report from the Japan association of childhood leukaemia study (JACLS)

A central pathology review system with an immunophenotyping laboratory was established in Japan to support the clinical trial, the Japan Association of Childhood Leukaemia Study (JACLS) NHL-98, for patients with paediatric non-Hodgkin's lymphoma (NHL). Pathology samples from 155 clinically-suspected NHL cases were evaluated centrally initially using the Revised European-American Lymphoma (REAL) classification in a rapid review (within 2 weeks after surgery/biopsy) and then later at the consensus review (once a year). The samples were subsequently re-classified according to the new World Health Organisation (WHO) classification. After the pathology review, 96 (62%) patients were eligible for the study, and 58 of them (60%) had extra-nodal primaries. These NHL cases included B-cell lymphomas (precursor B-cell, 11; Burkitt, 18; diffuse large B-cell, 18; not otherwise specified, 3) and T/Natural Killer (NK)-cell lymphomas (precursor T-cell, 23; anaplastic large cell, 20; others, 3). There was excellent concordance in making the diagnoses (95/96, 99%) and typing (93/96, 97%) of NHL between the rapid and consensus reviews. Five cases, initially diagnosed as diffuse large B-cell lymphoma by the review, were re-classified as Burkitt lymphoma according to the immunocytochemical criteria by the WHO classification. A total of 59 (38%) cases were excluded from the study: they were Hodgkin lymphoma (7), leukaemias (11), reactive lymphoid hyperplasia (20), necrotizing lymphadenitis (7), no consensus diagnosis (1), insufficient materials (2), and others (11). This is the first report of the central pathology review from the paediatric NHL group study in Japan. Because various diseases, either neoplastic or reactive, mimicked NHL, clinically and histopathologically, the central pathology review system was critical and essential for patient enrollment and protocol assignment in our clinical trial. Through the two-step review system, highly reliable data were generated to support this study.