Validation of a Predictive Model for Survival in Patients With Advanced Cancer: Secondary Analysis of RTOG 9714

Background The objective of this study was to validate a simple predictive model for survival of patients with advanced cancer. Methods Previous studies with training and validation datasets developed a model predicting survival of patients referred for palliative radiotherapy using three readily available factors: primary cancer site, site of metastases and Karnofsky performance score (KPS). This predictive model was used in the current study, where each factor was assigned a value proportional to its prognostic weight and the sum of the weighted scores for each patient was survival prediction score (SPS). Patients were also classified according to their number of risk factors (NRF). Three risk groups were established. The Radiation Therapy and Oncology Group (RTOG) 9714 data was used to provide an additional external validation set comprised of patients treated among multiple institutions with appropriate statistical tests. Results The RTOG external validation set comprised of 908 patients treated at 66 different radiation facilities from 1998 to 2002. The SPS method classified all patients into the low-risk group. Based on the NRF, two distinct risk groups with significantly different survival estimates were identified. The ability to predict survival was similar to that of the training and previous validation datasets for both the SPS and NRF methods. Conclusions The three variable NRF model is preferred because of its relative simplicity.