Forty-three cases of medulloblastoma were retro spectively evaluated. Patients who initially received irradiation to the entire central nervous system had a better survival than those receiving limited irradiation. Patient survival was improved with a dose of 4500 to 5500 rads to the posterior fossa, 3500 rads to the remainder of the brain, and 3000 rads to the spinal cord. There was a 27% complication incidence in 11 patients who received supplemental intrathecal radiogold and their survival has not been significantly better than those who received only external beam therapy to the spinal cord. Reirradiation of intracranial recurrences improved survival, while reirradiation of spinal core recurrences did little to improve survival. Proven intracranial recurrences were all anterior to the posterior fossa. We have concluded that the entire brain should receive 5000 rads, the spine should receive 3500 rads, and recurrences should be accurately localized and retreated by irradiation.
ORTY-THREE CASES OF MEDULLOBLASTOMA
F have been retrospectively evaluated. T h e cases include all patients treated at the University of Minnesota since 1942, with histologic proof of medulloblastoma.
T h e study was undertaken to evaluate the incidence, site, and prognosis in those patients who developed recurrences. Because a variety of radiotherapeu tic approaches have been employed over these years, the number of cases in each treatment group is small and statistically significant figures cannot be obtained.
Methods
Thirty-seven of the 43 cases received irradiation and were examined retrospectively for irradiation dose and method of therapy, recurrence pattern with respect to the primary ther-