Abstract
PURPOSE: Patients with non-small cell lung cancer (NSCLC) complicated by malignant pleural effusion (MPE) face a dismal prognosis. Existing biomarkers (eg, VEGF, CEA) show limited sensitivity, while nutritional indices (eg, PNI) are emerging as prognostic factors. This study aimed to develop a novel nomogram integrating lipid metabolism and nutritional indices to predict survival in NSCLC-MPE patients. METHODS: Multicenter retrospective cohort study enrolling patients with confirmed NSCLC combined with MPE who underwent thoracentesis from 2018 to 2024 from each of two centers. Univariate, multifactorial Cox regression analysis was used to identify five key clinical variables, and a nomogram model was developed. The predictive accuracy of the model was evaluated by calculating the area under the curve of the work characteristics of the recipients. RESULTS: A total of 250 patients with NSCLC combined with MPE were analyzed in this study, 195 in the training group and 55 in the validation group. The multifactorial COX test showed an interaction between ECOG PS, pleural lactate dehydrogenase (LDH), T stage, low/high-density lipoprotein cholesterol concentration ratio (LHR), and prognostic nutritional index (PNI). At 1, 2, and 3 years, the area under the curve (AUC) values were 0.899, 0.808, and 0.748 for the training set and 0.899, 0.798, and 0.669 for the validation set, respectively. CONCLUSION: MPE carries a poor prognosis for NSCLC patients, and the clinical prediction model we constructed shows good promise in predicting OS in this patient, which can assist direct the selection of optimal treatment strategies.