The long-term postsurgical prognosis of patients with pineoblastoma

BACKGROUND: For this report, the authors comprehensively summarized the existing literature on patients with pineoblastoma and identified the variables and treatments that had an impact patient on outcomes. METHODS: A comprehensive search identified 109 studies that collectively described the outcomes of patients with pineoblastoma. Individual patient data were classified based on treatment and were subjected to univariate comparisons. Cox regression analysis included comparisons of survival outcomes controlling for age, extent of resection, and treatment group, and between-group survival comparisons were performed using the Kendall tau (rank correlation) statistic. RESULTS: Two hundred ninety-nine patients met inclusion criteria. The overall survival rate was 54% (175 of 299 patients) at a mean follow-up of 31 ± 1.9 months (range, 1-159 months). The analyses demonstrated a markedly worse prognosis for children aged ≤5 years compared with older patients (5-year survival rate: 15% for children aged ≥5 years vs 57% for children aged ≥5 years; log-rank P <.00001). In addition, a graded increase in survival was observed with increasing degrees of resection (5-year survival rate: 84% for patients who underwent gross total resection vs 53% for patients who underwent subtotal resection vs 29% for patients who underwent debulking; log-rank P <.0001). Multivariate analysis indicated that not achieving gross total resection markedly worsened patient survival (subtotal resection: hazard ratio, 6.47; 95% confidence interval, 2.3-19; P =.001. debulking: hazard ratio, 9.27; 95% confidence interval, 3.2-27; P <.0001). CONCLUSIONS: The current findings emphasize the importance of aggressive surgical resection in the treatment of pineoblastoma. In addition, the authors conclude that clinical trials should not mix young patients with older patients or patients who undergo subtotal resection with patients who undergo gross total resection, because such heterogeneity may alter the variability of responses to treatment and reduce the likelihood of success.