Correlates of survival and the Daumas-Duport grading system for astrocytomas.

In order to examine the correlation between prognosis and the histological features of nuclear atypia, mitosis, endothelial proliferation, and necrosis in supratentorial adult astrocytomas, the authors reviewed 251 such cases treated at the Massachusetts General Hospital between 1972 and 1980. One point was given for the presence of each feature. The total number of features was translated into a grade as follows: none of the four features = Grade 1 (one patient), one feature = Grade 2 (36 patients), two features = Grade 3 (33 patients), and three or four features = Grade 4 (181 patients). The period of survival was significantly associated with grade, the presence or absence of each of the four histological features, patient's age, type of operation, radiation therapy, and extent of tumor (log rank, p less than 0.05). The variables associated with grade were age (p less than 0.001) and radiation therapy (p less than 0.02). After adjustment for these variables using a Cox proportional-hazards model, the difference in overall survival time between patients in Grades 2 and 3 was not statistically significant. When comparable groups of patients were examined in terms of age or receipt of radiation therapy, the median survival times differed markedly. Patients 50 years of age or less had a median survival time of 68 months (Grade 2 tumors), 29 months (Grade 3 tumors), and 13 months (Grade 4 tumors). Patients over 50 years of age had a median survival time of 6 months (Grade 2 and 4 tumors) and 9 months (Grade 3 tumors). Those patients who had received radiation therapy had a median survival time of 68 months (Grade 2 tumors), 21 months (Grade 3 tumors), and 11 months (Grade 4 tumors). Those patients who did not receive radiation therapy had a median survival time of 1 month (Grade 2 tumors) and 2 months (Grade 3 and 4 tumors); over half of these patients died within 2 months of surgery. This grading system, originally proposed by Daumas-Duport, et al., is simple, objective, and reproducible, and correlates well with survival times. The authors recommend that astrocytomas be graded on a scale of 1 to 4, with Grade 1 reserved for the rare adult supratentorial astrocytoma with none of the four histological features.