Accelerated fractionation radiation therapy for liver metastases: selection of an optimal patient population for the evaluation of late hepatic injury in RTOG studies.

The palliation of hepatic metastases represents a major therapeutic problem in oncology. The generally unfavorable prognosis of patients with liver metastasis may complicate the evaluation of the efficacy and toxicity of various therapeutic modalities. The Radiation Therapy Oncology Group (RTOG) is evaluating new accelerated fractionation schemes for hepatic irradiation. In designing this study it became necessary to identify a favorable subpopulation of patients with an expected median survival of 6 months in whom the late effects of treatment could be evaluated. Data from two RTOG liver metastases studies (7605 and 8003) were analyzed using multi-variate techniques. Cases with a serum bilirubin level of greater than 1.5 mg%, performance score of less than 50, and gastric or pancreatic primary carcinomas were initially excluded because of the adverse influence of these factors on survival. One hundred and ninety cases met the criteria for inclusion in this analysis. A loglinear model was used to identify the patient characteristics associated with a favorable prognosis for survival. These included a performance score of 80-100, colorectal carcinoma primary, and no extrahepatic metastases. A logistic regression equation was derived and confirmed that the probability of surviving at least 6 months depended on the effects of performance score, primary site and the presence or absence of extra-hepatic metastases. Patients with all three favorable factors constituted 30% of the evaluable study population and had a predicted survival of 50% at 6 months. A patient population with these characteristics will be used to study the late effects of accelerated fractionated radiation therapy on the liver.