Abstract
BACKGROUND: Complete blood cell count (CBC)-derived inflammatory biomarkers are widely used as prognostic parameters for various malignancies, but the best predictive biomarker for early-stage non-small-cell lung cancer (NSCLC) is unclear. We retrospectively analyzed early-stage NSCLC patients to investigate predictive effects of preoperative CBC-derived inflammatory biomarkers. PATIENTS AND METHODS: We selected 311 consecutive patients with pathological stage IA NSCLC surgically resected from April 2006 to December 2012. Univariate and multivariate Cox proportional analyses of recurrence-free survival (RFS) were used to test the preoperative systemic immune inflammation index (SII), neutrophil-lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), and monocyte-lymphocyte ratio (MLR). RESULTS: Preoperative high MLR levels were significantly associated with patient sex, smoking status, and postoperative recurrence (p <0.0001, p = 0.0307, and p = 0.0146, respectively), and preoperative high SII levels were significantly correlated with postoperative recurrence (p = 0.0458). Neither NLR nor PLR were associated with any related factors. Only preoperative MLR levels (p = 0.0269) were identified as an independent predictor of shorter RFS. The relative risk (RR) for preoperative high MLR level versus low level patients was 2.259 (95% confidence interval [CI]: 1.094-5.000). Five-year RFS rates in patients with preoperatively high MLR levels were significantly lower than in those with low MLR levels (82.21% vs. 92.05%, p = 0.0062). In subgroup analysis by tumor size and MLR level, the high MLR level subgroup with tumors >2 cm had significantly shorter RFS than other subgroups (p = 0.0289). CONCLUSIONS: The preoperative MLR level is the optimal predictor of recurrence in patients with pathological stage IA NSCLC.