Thirty-eight patients with a mean age of 53.2 years (19 to 75 years of age), who were receiving doxorubicin (D) for malignant disease, were studied in order to determine the relationship between functional and morphologic myocardial changes at different dose levels. Serial patient evaluations included physical examination, chest x-ray, electrocardiogram (ECG), endomyocardial biopsy (EMB), and rest-exercise gated nuclear angiography (GNA), at doses of D ranging from 144 to 954 mg/m2 (mean, 426 mg/m2). Physical examination, chest x-ray, and ECG proved to be insensitive predictors of D cardiotoxicity. Correlation of GNA and EMB in 31 patient evaluations, exclusive of known heart disease, did not reveal any false-positive angiograms, and all abnormal GNAs were associated with abnormal biopsies. Use of stress GNA uncovered six abnormal ventricles which could have been missed with a rest GNA alone. It has been suggested that: (1) GNA is a reliable monitor of D therapy; (2) an exercise study should be performed when the rest ejection fraction is normal, but is unnecessary when the rest EF is abnormal; (3) all patients with a resting ejection fraction of less than 45%, exclusive of other cardiac disease, should have D discontinued; and (4) endomyocardial biopsy is useful in assessing D cardiotoxicity in patients with other possible causes of an abnormal GNA.