Background: In recent years, the role of pre-treatment C-reactive protein/albumin ratio (CAR) in prognosis of esophageal cancer (EC) has been investigated by several studies. This meta-analysis aimed to provide a more accurate and objective assessment of the prognostic value of pre-treatment CAR in EC.
Methods: Studies assessing the role of pre-treatment CAR in prognosis of EC were searched from PubMed, Embase and the Cochrane Library (last update by April 16, 2019). The hazard ratios (HRs) of CAR and the corresponding 95% CIs for overall survival (OS) or cancer-specific survival (CSS) in EC were extracted for pooled analysis.
Results: A total of eight observational studies including 2255 patients were collected. The pooled analysis showed that high CAR was related to worse OS in EC (pooled HR = 1.81; 95% CI = 1.40-2.35; P < 0.001). Subgroup analyses showed that the negative correlation between the CAR and OS was consistently demonstrated in subgroups stratified by country, pathological type, and cut-off value (P < 0.05). However, there was no relation between CAR and OS in subgroup of patients receiving neoadjuvant chemotherapy at a proportion of 100% (HR = 1.15, 95% CI = 0.56-2.69; P = 0.715). In addition, high CAR was also related to worse CSS in EC (pooled HR = 2.61; 95% CI = 1.67-4.06; P < 0.001).
Conclusions: High pre-treatment CAR was an adverse prognostic factor for EC patients. More large-sample clinical trials are still needed to verify the prognostic value of pre-treatment CAR in EC.
Keywords: C-reactive protein/albumin ratio; Esophageal cancer; Meta-analysis; Prognosis.