Prognostic scoring in adult astrocytic tumors using patient age, histopathological grade, and DNA histogram type

• Published in: Journal of neurosurgery Vol. 80; no. 5; p. 877
• Main Authors: Salmon, I, Dewitte, O, Pasteels, J L, Flament-Durand, J, Brotchi, J, Vereerstraeten, P, Kiss, R
• Format: Journal Article
• Language: English
• Published: United States 01.05.1994
• Subjects: Adolescent; Adult; Age Factors; Aged; Analysis of Variance; Astrocytoma - epidemiology; Astrocytoma - genetics; Astrocytoma - pathology; Astrocytoma - therapy; Brain Neoplasms - epidemiology; Brain Neoplasms - genetics; Brain Neoplasms - pathology; Brain Neoplasms - therapy; Child; DNA; Humans; Middle Aged; Ploidies; Prognosis
• ISSN: 0022-3085
• DOI: 10.3171/jns.1994.80.5.0877

High-grade astrocytic tumors constitute the most serious as well as the most common group of primary brain tumors. Although several prognostic factors have been proposed, little is known about the prognostic value of deoxyribonucleic acid (DNA) ploidy in adult astrocytic tumors. In a series of 146 adult patients, aged 16 to 82 years, the individual prognostic values of six variables were studied, namely: tumor histopathological grade, treatment, patient age, extent of tumor, ploidy level, and DNA histogram type. Cox's proportional hazard model was then applied to the data to ascertain which factors might independently determine patient survival. Univariate analyses revealed that histopathological grade, age, and DNA histogram type were very powerful prognostic factors. The statistical significance of the influence of adjuvant radiotherapy and chemotherapy was at a borderline level, and the two remaining variables (tumor extent and ploidy level) had no prognostic relevance. Multivariate analyses showed that age, histopathological grade, and DNA histogram type were independent, statistically significant prognostic factors. A prognostic score was calculated from Cox's polynomial function in which those factors were introduced. The best score corresponded to a patient aged 16 years with a hypertriploid low-grade astrocytoma, while the worst score corresponded to a patient aged 82 years with a diploid high-grade astrocytoma. The worst score:best score ratio revealed a risk 71 times higher for a bad prognosis. It is concluded that patient age, histopathological grade, and DNA histogram type are very powerful prognostic factors for adult astrocytic tumors. A prognostic score including those factors could be used to characterize astrocytic tumor aggressiveness presurgically on fine-needle aspirates, and to monitor the patient's postsurgical evolution to define the appropriate therapy.