Proliferation marker KI-S5 discriminates between favorable and adverse prognosis in advanced stages of neuroblastoma with and without MYCN amplification

Abstract

Background: The biologic behavior of neuroblastoma is notoriously variable, and even carefully elaborated prognostic models fail to predict the clinical course in a portion of cases. Because the proliferative activity is determined by the sum of all molecular imbalances that influence cell cycling, the authors investigated the potential prognostic relevance of the tumor growth fraction in neuroblastoma.

Methods: A retrospective analysis was conducted on a cohort of 161 neuroblastoma patients with a median follow-up period of 72.8 months. Tumors were classified according to Hughes typing and grading criteria. The proliferative index (PI) was assessed immunohistochemically on archival biopsy specimens using monoclonal antibody Ki-S5 (Ki-67), and the MYCN status was determined by means of Southern blot analysis.

Results: The PI, MYCN status, International Neuroblastoma Staging System (INSS) stage, International Neuroblastoma Pathology Classification grade, Hughes grade, and the patients' age at diagnosis were all found to be significant predictors of event free survival by univariate Kaplan-Meier analysis. However, the PI identified prognostically distinct subsets in higher tumor stages and Grade 2 and 3 neuroblastomas as well as tumors with unfavorable histology, and enabled risk stratification in tumors with and without MYCN amplification (P = 0.034 and 0.002, respectively). Multivariate Cox regression analysis selected INSS stage (relative risk [RR], 4.05; P < 0.0001) and the PI (RR, 2.49; P = 0.007) as the sole independent prognostic indicators, whereas MYCN entered the selection only after exclusion of the PI.

Conclusions: It emerges that the PI as a single factor has greater predictive power than the MYCN status. Proliferation measurements therefore might significantly improve the accuracy of current prognostic models for neuroblastoma.