Predictors of survival in patients with incurable cancer

Abstract

Background: The aim of the present study was to investigate predictors of survival in a cohort of patients with incurable cancer. This may highlight areas of care which can be addressed to optimise outcomes. Methods: Patient demographics, performance status (ECOG), inflammatory markers (modified Glasgow Performance Score (mGPS)), and nutritional parameters [BMI, % weight loss (WL)] were recorded. Baseline body composition were examined using computed tomography (CT) images. Sarcopenia and low muscle attenuation (MA) were defined using published cut-offs. Cancer cachexia (CC) was defined using the consensus definition (2011). Cox models were used to estimate mortality hazard ratios, adjusted for known prognostic covariates – age, sex & site. Results: In total, 1027 patients were recruited (51% male, median age 66 years). Gastrointestinal cancer was most common (40%) and metastatic disease was present in 87% of patients. In total, 86% were on active chemotherapy treatment. On multivariate analysis, primary site, ECOG-PS, WL and mGPS predicted survival. Lung and GI cancer patients had higher risk of death compared to other tumour groups [HR:1.769 (95% CI:1.305–2.398), p < 0.001 and HR:1.576 (95% CI:1.211–2.053), p = 0.001, respectively]. mGPS score of 1 or 2 were associated with increased risk of death versus mGPS of 0 [HR:1.399 (95% CI:1.024–1.910), p = 0.035 & HR:1.831 (95% CI:1.442–2.324), p < 0.001, respectively]. WL > 10% in the preceding 3 months was associated with reduced survival [HR:2.501 (95% CI:1.425–4.389), p = 0.001]. Sarcopenia, CC and low MA were significant predictors of survival on univariate analysis, but not on multivariate analysis. Conclusions: Independent predictors of survival in this cohort of advanced cancer patients were cancer site, ECOG, WL > 10% and mGPS, the latter two being important components of the cancer cachexia syndrome. Attenuation of weight loss and inflammation may improve outcomes for patients undergoing palliative treatment for cancer. Legal entity responsible for the study: The authors. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.