Criteria for evaluating patients undergoing chemotherapy for malignant brain tumors

Abstract

✓ The authors describe their criteria for evaluating brain-tumor patients, and present a numerical rating scale devised to designate response to testing. They discuss the reliability of test combinations that permit accurate prediction of response or deterioration during therapy in their experience with 100 patients treated on the Chemotherapy Service at the Brain Tumor Research Center, University of California, San Francisco School of Medicine. Specifically, the paper summarizes the predictive value of the neurological examination, radionuclide scintiscan, computerized tomographic brain scan (CT scan), and electroencephalogram (EEG), in the determination of response (tumor regression) or deterioration (tumor growth) during brain-tumor chemotherapy and chemotherapy-radiotherapy. By retrospective analysis, the neurological examination, radionuclide scintiscan, and CT scan were of equal value individually as tests to predict response to therapy. However, the prognostic values of the neurological examination or the radionuclide scintiscan proved moderately superior to the CT scan in predicting deterioration during therapy. Under circumstances whereby a neurological examination, radionuclide scintiscan, and CT scan were all performed during the same testing session, and steroid dosage was carefully monitored, two of the three tests accurately predicted deterioration in 65% of patients, and response to therapy in 82% of patients. Two of the three tests correctly established deterioration in the remaining 35% of patients, and response in the remaining 18% of patients, when the two positively correlated tests had occurred within a mean period of 7 weeks.