Open AccessThe prognostic value of serum C‐reactive protein‐bound serum amyloid A in early‐stage lung cancer
Author(s) -
Zhang XueYan,
Zhang Ge,
Jiang Ying,
Liu Dan,
Li ManZhi,
Zhong Qian,
Zeng ShanQi,
Liu WanLi,
Zeng MuSheng
Publication year - 2015
Publication title -
cancer communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.119
H-Index - 53
ISSN - 2523-3548
DOI - 10.1186/s40880-015-0039-1
Subject(s) - lung cancer , medicine , serum amyloid a , c reactive protein , gastroenterology , prospective cohort study , hazard ratio , serum amyloid a protein , cancer , proportional hazards model , confidence interval , oncology , inflammation
Background Elevated levels of serum C‐reactive protein (CRP) have been reported to have prognostic significance in lung cancer patients. This study aimed to further identify CRP‐bound components as prognostic markers for lung cancer and validate their prognostic value. Methods CRP‐bound components obtained from the serum samples from lung cancer patients or healthy controls were analyzed by differential proteomics analysis. CRP‐bound serum amyloid A (CRP‐SAA) was evaluated by co‐immunoprecipitation (IP). Serum samples from two independent cohorts with lung cancer (retrospective cohort, 242 patients; prospective cohort, 222 patients) and healthy controls (159 subjects) were used to evaluate the prognostic value of CRP‐SAA by enzyme‐linked immunosorbent assay. Results CRP‐SAA was identified specifically in serum samples from lung cancer patients by proteomic analysis. CRP binding to SAA was confirmed by co‐IP in serum samples from lung cancer patients and cell culture media. The level of CRP‐SAA was significantly higher in patients than in healthy controls (0.37 ± 0.58 vs. 0.03 ± 0.04, P < 0.001). Elevated CRP‐SAA levels were significantly associated with severe clinical features of lung cancer. The elevation of CRP‐SAA was associated with lower survival rates for both the retrospective (hazard ration [HR] = 2.181, 95% confidence interval [CI] = 1.641–2.897, P < 0.001) and the prospective cohorts (HR = 2.744, 95% CI = 1.810–4.161, P < 0.001). Multivariate Cox analysis showed that CRP‐SAA was an independent prognostic marker for lung cancer. Remarkably, in stages I–II patients, only CRP‐SAA, not total SAA or CRP, showed significant association with overall survival in two cohorts. Moreover, univariate and multivariate Cox analyses also showed that only CRP‐SAA could be used as an independent prognostic marker for early‐stage lung cancer patients. Conclusion CRP‐SAA could be a better prognostic marker for lung cancer than total SAA or CRP, especially in early‐stage patients.