Purpose: To assess if the abundance of apoptotic tumor cells is an independent prognostic factor in primitive neuroectodermal tumors (PNET) of the central nervous system.
Patients and methods: Formalin-fixed paraffin-embedded tumor tissue sections from 78 clinically well-characterized children with PNET were evaluated by terminal deoxytransferase-mediated deoxyuridinie-5'-triphosphate (dUTP) nick-end labeling (TUNEL). Apoptotic indices (AI) were determined by counting TUNEL-positive tumor cells either in the highest staining region (AI hot spot) or in at least 15 randomly chosen fields (AI random). The AI hot spot and AI random were then correlated with clinical variables and survival outcome.
Results: AI hot spot (median 0.56%; range 0%-6.54%) and AI random (median 0.30%; range 0%-3.21%) showed considerable intertumor variability. Moreover, 53% of the evaluated PNET showed a more than two-fold difference between AI hot spot and AI random, showing important intratumoral variability of the abundance of apoptotic cells in a subset of the evaluated PNET. No significant associations were found between AI hot spot and AI random with clinical variables or survival outcome.
Conclusion: The apoptotic index does not predict survival outcome and is not specifically associated with clinical variables of prognostic significance in childhood PNET.