Fig. 1: TCGA-based analysis shows upregulated MZT2B expression correlating with adverse clinicopathological features and poor prognosis in NSCLC.

Transcriptomic analyses of The Cancer Genome Atlas (TCGA) datasets were undertaken to measure the differential expression of MZT2B in neoplastic lung tissues, specifically lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC), relative to their normal counterparts (A). Subsequent direct comparisons were carried out on matched tumor-normal tissue pairs to delineate MZT2B expression profiles within LUAD (B) and LUSC (C) tumor specimens juxtaposed with their adjacent normal pulmonary tissues. Concomitant analyses explored the association of MZT2B expression with a spectrum of clinicopathological attributes, encompassing pathological T stages (T1 versus T2, T3&T4) (D), the presence and extent of lymph node involvement (N0 versus N1, N2&N3) (E), metastatic dissemination to distant sites (M0 versus M1) (F), patient sex (G), age demographics (≤65 versus >65 years) (H), cumulative smoking exposure expressed in pack-years (<40 versus ≥40) (I), and current/former smoking habit versus never-smoker status (J) in NSCLC cohorts. Kaplan–Meier survival methodologies were employed to ascertain the correlative relationship between MZT2B expression levels and overall survival (OS) among NSCLC patients (K). The prognostic utility of MZT2B expression in NSCLC was further evaluated through receiver operating characteristic (ROC) curve analysis (L). HR hazard ratio, CI confidence interval, TPM transcripts per million, FPR, false positive rate, TPR true positive rate. The Kruskal–Wallis post hoc Dunn test was used in (A, D, E), the Wilcoxon signed rank test in (B, C). Mann–Whitney U test in (F–J), Cox regression analysis in (K). Asterisks signify levels of statistical significance: *P < 0.05, **P < 0.01, ***P < 0.001.