Background: The C-reactive protein/albumin (CRP/Alb) ratio has been reported as a prognostic factor of survival for patients with a variety of cancers. However, its prognostic impact for advanced cancer patients receiving palliative care remains presently unknown. Objective: The present study assessed the prognostic value of the CRP/Alb ratio, and compared this with that of the Glasgow Prognostic Score (GPS) and Palliative Prognostic Index (PPI) in a cohort of advanced cancer patients receiving palliative therapy. Methods: The medical records of 262 eligible patients who died of advanced cancer from February 1, 2013 to December 30, 2017 in the palliative care unit of the Fudan University Shanghai Cancer Center were retrospectively reviewed for the analysis. Results: The present results revealed that a CRP/Alb ratio ≥1.31 (hazard ratio [HR], 2.33 [1.78-3.05], p < 0.001) can predict poor prognosis through univariate analysis. In addition, the multivariate analysis revealed that CRP/Alb (HR, 2.09 [1.54-2.84], p < 0.001), GPS (HR, 1.81 [1.07-3.07], p < 0.001), and PPI (HR, 3.35 [2.25-4.99], p < 0.001) were all independent prognosis factors. To compare the discriminatory performance of the CRP/Alb ratio with that of other established prognostic indexes in palliative care settings, the c-statistics, integrated discriminatory improvement index, net reclassification index, and receiver operating characteristic curves were generated, and it was demonstrated that the CRP/Alb ratio (c-statistics, 0.64 [0.61-0.68]) was able to discriminate advanced cancer patients with different survivals, with analogous discriminatory ability as GPS (c-statistics, 0.63 [0.59-0.66]) and PPI (c-statistics, 0.64 [0.60-0.68]). Notably, the combination of multiple prognostic indexes exerted higher discriminatory ability, compared with any single predictive index (c-statistics, 0.69 [0.66-0.73], p < 0.001). Conclusions: The present study suggests that the CRP/Alb ratio is a promising prognostic factor in predicting cancer patient survival in palliative care settings. Incorporating both objective parameters and the subjective index may improve the prediction accuracy of prognosis.

	Year	Citations

About