Multivariate analysis of visceral pleural invasion in pleural-adherent non-small cell lung cancer with maximum diameter ≤30 mm

Objective To explore the risk factors for the development of visceral pleural invasion (VPI) in pleural-adherent non-small cell lung cancer (NSCLC) with maximum diameter ≤30 mm.

Methods A cross-sectional study design was employed and a retrospective analysis was conducted on the clinical, imaging and histopathological examination data of 115 patients with pleural-adherent NSCLC diagnosed by post-operative histopathological examination from January 2023 to September 2024 in Kunshan Hospital, Jiangsu University. The patients included 48 males and 67 females, aged 63(54, 69) years. Using the elastic fiber staining results as the "gold standard", the 115 pleural-adherent NSCLC patients were divided into VPI-positive and VPI-negative groups. Chi-square test (or Fisher exact probability method if the expected frequency is <1) or Mann-Whitney U test was used for comparisons between two groups. Binary Logistic regression was employed to identify independent risk factors for VPI in pleural-adherent NSCLC and construct a predictive model. The predictive performance of each independent risk factor and the model was evaluated using receiver operating characteristic curves.

Results Among the 115 patients with pleural-adherent NSCLC, 68 cases were allocated to the VPI-negative group and 47 cases were allocated to the VPI-positive group. Univariate analysis showed statistically significant differences in gender (χ²=6.028, P=0.014), age (Z=−2.636, P=0.008), pathological type (Fisher exact probability method, P=0.001), cell proliferating nuclear antigen Ki-67 (Z=−4.583, P<0.001), density type (χ²=15.517, P<0.001), bronchial changes (χ²=5.452, P=0.020), pleural indentation or thickening (χ²=7.315, P=0.007), maximum lesion diameter (Z=−3.169, P=0.002), maximum diameter of consolidation (Z=−4.533, P<0.001), consolidation-to-tumor ratio (Z=−3.577, P<0.001), lesion-pleural contact length (Z=−4.797, P<0.001), consolidation-pleural contact length (Z=−5.665, P<0.001), lesion-pleural contact feature (χ²=32.053, P<0.001) and adjacent slice feature (χ²=45.698, P<0.001). Multivariate Logistic regression analysis demonstrated that consolidation-pleural contact length (OR=1.087, 95%CI: 1.003–1.177), pleural indentation or thickening (OR=7.628, 95%CI: 1.561–37.279), lesion-pleural contact feature (type Ⅲ) (OR=7.060, 95%CI: 2.048–24.340) and adjacent slice features (type Ⅳ) (OR=10.653, 95%CI: 3.388–33.500) were independent risk factors for VPI in patients with pleural-adherent NSCLC (all P<0.05). The combined predictive model established on the basis of these four factors had favorable predictive performance, with an area under the curve of 0.919 (95%CI: 0.869–0.968), a sensitivity of 83.0% and a specificity of 88.2%.

Conclusions For patients with pleural-adherent NSCLC, the possibility of VPI is highly suggestive when preoperative CT showing that the lesion is adherent to the pleura in an obtuse angle, the presence of multiple (≥3) pleural tags on adjacent slices, pleural indentation or thickening, with the consolidation-pleural contact length being greater than 12.5 mm.