EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Establishment of a prediction model for malignant probability of pulmonary nodules in an adenocarcinoma-predominant cohort

Fengru Li, Xiaobo Yu, Kangqiao Xu, Guoqing Zhang

PeerJ · 18.06.2026 · 14 · e21353

DOI: 10.7717/peerj.21353 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ, online erschienen am 18.06.2026. Verfasst von Fengru Li und 3 weiteren Autor:innen. Beteiligte Einrichtung: Jiading Branch of Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. Hauptthemen laut MEDLINE: Lung Neoplasms, Multiple Pulmonary Nodules, Adenocarcinoma of Lung, Solitary Pulmonary Nodule. Die Arbeit nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jiangqiao Hospital Technology Innovation Basic Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

This study aimed to identify associated factors for malignancy in pulmonary nodules and establish a risk prediction model to provide a scientific basis for surgical decision-making.

Methods

Patients with pulmonary nodules who underwent surgical resection were retrospectively enrolled for this study. Demographic data, tumor marker levels, computed tomography (CT) features, and pathological results were collected. Univariate and multivariate analyses were performed to identify independent associated factors and construct a predictive model. The model’s performance was evaluated using the receiver operating characteristic (ROC) curve.

Results

A total of 568 patients were included, comprising 146 benign and 422 malignant nodules. The vast majority of malignant nodules were lung adenocarcinomas ( n = 406). Multivariate analysis showed that female gender (odds ratio (OR) 1.729; 95% confidence interval (CI) [1.114–2.685], P = 0.015), lobulation sign (OR: 2.250, 95% CI [1.301–3.890], P = 0.004), burr (OR: 2.965, 95% CI [1.740–5.053], P < 0.001), the vascular bundle sign (OR: 6.721, 95% CI [1.971–22.922], P = 0.002), the nature of nidus mixed ground glass nodules (OR: 2.627 95% CI [1.050–6.572], P = 0.039) and the nature of nidus solid pulmonary nodules (OR: 0.169 95% CI [0.102–0.279], P < 0.001) were independently associated with malignancy. The area under the ROC curve was 0.81 ( P < 0.05, 95% CI [0.767–0.849]), with a sensitivity of 91.0% and specificity of 47.3%.

Conclusion

In this adenocarcinoma-predominant cohort, female sex was a significant predictor of malignancy, suggesting that non-smoking women may benefit from increased attention in lung cancer screening strategies. Further validation in diverse populations is warranted.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Lung NeoplasmsMultiple Pulmonary NodulesAdenocarcinoma of LungSolitary Pulmonary NoduleAdenocarcinomaHumansFemaleMaleRetrospective StudiesMiddle AgedAgedTomography, X-Ray ComputedROC CurveRisk Factors

Stichwörter der Autor:innen

Clinical feathersDiagnostic valueLung adenocarcinomaPrediction modelPulmonary nodules

Autor:innen und Institutionen

  • Jiading Branch of Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.

Förderung

  • Jiangqiao Hospital Technology Innovation Basic Project

Zitieren

APA

Li, F., Yu, X., Xu, K., & Zhang, G. (2026). Establishment of a prediction model for malignant probability of pulmonary nodules in an adenocarcinoma-predominant cohort. PeerJ, 14, e21353. https://doi.org/10.7717/peerj.21353

Vancouver

Li F, Yu X, Xu K, Zhang G. Establishment of a prediction model for malignant probability of pulmonary nodules in an adenocarcinoma-predominant cohort. PeerJ. 2026;14:e21353. doi:10.7717/peerj.21353

BibTeX
@article{Li2026,
  author = {Li, Fengru and Yu, Xiaobo and Xu, Kangqiao and Zhang, Guoqing},
  title = {Establishment of a prediction model for malignant probability of pulmonary nodules in an adenocarcinoma-predominant cohort},
  journal = {PeerJ},
  year = {2026},
  volume = {14},
  pages = {e21353},
  doi = {10.7717/peerj.21353}
}
Kostenlos starten →