The palliative prognostic score and survival in patients with advanced solid tumors receiving chemotherapy

Purpose: To evaluate the accuracy of the Palliative Prognostic Score (PaP score) in selecting metastatic gastrointestinal or nonsmall-cell lung cancer patients candidate to palliative chemotherapy. Materials and methods: The PaP score was calculated in 173 patients with advanced, pretreated gastrointestinal or nonsmall-cell lung cancer before starting a further line of chemotherapy with palliative aim. Symptom distress score was calculated using the Edmonton Symptom Assessment System (ESAS) before every course of chemotherapy. Univariate analysis of survival was performed using the logrank test; multivariate analysis was performed using the Cox regression model. Symptom distress scores were compared using multivariate analysis of variance test for repeated measures, and overall symptom distress score was compared using analysis of variance test for repeated measures. Results: Overall median survival was 26 weeks; in PaP score class A it was 32 weeks, and in class B 8 weeks (p<0.0001). No patient was classified in class C. The two-class PaP score resulted in an independent prognostic factor (p=0.022), as well as Karnofsky performance status (p=0.002) and colorectal cancer (p=0.017). A trend towards worsening of symptom distress was observed in the entire population and in class A. The high number of missed data did not permit an adequate analysis in class B. Conclusions: The PaP score seems to discriminate patients who could benefit by palliative chemotherapy from those who could better benefit by supportive and palliative approach. However, the data are insufficient to validate the use of the PaP score in patients to be treated with palliative chemotherapy, and further trials should be planned to assess its ability to improve the quality of care in oncology and the appropriateness in the choice of palliative chemotherapy. © 2007 Springer-Verlag.