Abstract
Objectives. We explored the prognostic utility of the unique combination of C‐reactive‐protein‐to‐albumin ratio (CAR) and significant weight loss (WL > 5%) over the preceding 6 months, namely, the CARWL score, in stage IIIC non‐small‐cell lung cancer (NSCLC) patients who underwent concurrent chemoradiotherapy (CCRT). Methods. For each patient, the CAR was calculated using C‐reactive protein and albumin measurements obtained on the first day of CCRT: CAR = C‐reactive protein ÷ albumin. The availability of an ideal CAR cutoff that may categorize patients into two distinct progression‐free (PFS) and overall survival (OS) outcomes was explored by employing receiver operating characteristic (ROC) curve analysis. Patients were additionally divided into two groups based on their status of significant WL according to the well‐recognized Delphi criteria. Then, the CARWL score was created by combining all feasible combinations of the CAR and significant WL groupings. The potential links between pretreatment CARWL groups and the post‐CCRT OS and PFS outcomes were determined as the primary and secondary endpoints. Results. This retrospective cohort study comprised a total of 651 stage IIIC NSCLC patients. ROC curve analysis indicated that rounded 3.0 was the ideal CAR cutoff (area under the curve (AUC): 70.1%; sensitivity: 67.8%; specificity: 65.9%), which categorized the patients into CAR < 3.0 (N = 324) and CAR ≥ 3.0 (N = 327) groups. There were 308 (47.3%) and 343 (52.7%) patients without and with significant WL, respectively. The created CARWL groups were CARWL‐0: CAR < 3.0 and WL ≤ 5.0%; CARWL‐1: CAR < 3.0 and WL > 5.0%, or CAR ≥ 3.0 and WL ≤ 5.0%; and CARWL‐2: CAR > 3.0 and WL > 5.0%. The Kaplan–Meier curves showed that the PFS (14.2 vs. 11.4 vs. 7.5 months; P < 0.001) and OS (37.3 vs. 23.6 vs. 12.8 months; P < 0.001) durations were gradually and significantly lowered from the CARWL‐0 to CARWL‐2 groups. The CARWL score’s significant impacts on PFS and OS outcomes were found to be independent of the other variables in the multivariate analysis (P < 0.001, for each). Conclusions. Our findings indicate that the novel CARWL score, which accounts for pretreatment CAR and significant WL during the preceding 6 months, can reliably stratify newly diagnosed stage IIIC NSCLC patients into three groups with significantly different PFS and OS after definitive CCRT.