Preoperative identification of clinicopathological prognostic factors for relapse-free survival in clinical N1 non-small cell lung cancer: a retrospective single center-based study

BACKGROUND: Given the difficulty in preoperatively diagnosing lymph node metastasis, patients with Stage I–III non-small cell lung cancer (NSCLC) are likely to be included in the clinical N1 (cN1) group. However, better treatment options might be selected through further stratification. This study aimed to identify preoperative clinicopathological prognostic and stratification factors for patients with cN1 NSCLC. METHODS: This retrospective study evaluated 60 patients who were diagnosed with NSCLC during 2004–2014. Clinical nodal status had been evaluated using routine chest computed tomography (CT) and/or positron emission tomography (PET). To avoid biasing the fluorodeoxyglucose uptake values based on inter-institution or inter-model differences, we used only two PET systems (one PET system and one PET/CT system). Relapse-free survival (RFS) and overall survival (OS) were the primary study outcomes. The maximum standardized uptake value (SUVmax) was calculated for each tumor and categorized as low or high based on the median value. Patient sex, age, histology, tumor size, and tumor markers were also assessed. RESULTS: Poor OS was associated with older age (P = 0.0159) and high SUVmax values (P = 0.0142). Poor RFS was associated with positive carcinoembryonic antigen (CEA) expression (P = 0.0035) and high SUVmax values (P = 0.015). Multivariate analyses confirmed that poor OS was independently predicted by older age (hazard ratio [HR] = 2.751, confidence interval [CI]: 1.300–5.822; P = 0.0081) and high SUVmax values (HR = 5.121, 95% CI: 1.759–14.910; P = 0.0027). Furthermore, poor RFS was independently predicted by positive CEA expression (HR = 2.376, 95% CI: 1.056–5.348; P = 0.0366) and high SUVmax values (HR = 2.789, 95% CI: 1.042–7.458; P = 0.0410). The primary tumor’s SUVmax value was also an independent prognostic factor for both OS and RFS. CONCLUSIONS: For patients with cN1 NSCLC, preoperative prognosis and stratification might be performed based on CEA expression, age, and the primary tumor’s SUVmax value. To enhance the prognostic value of the primary tumor’s SUVmax value, minimizing bias between facilities and models could lead to a more accurate prognostication.