Tumor volume may be a better indicator than diameter for occult lymph node metastasis in patients with clinical stage-1 lung cancer.
Authors
Affiliations (3)
Affiliations (3)
- Department of Thoracic Surgery, Gazi University School of Medicine, 06560, Besevler, Ankara, Turkey. [email protected].
- Department of Thoracic Surgery, Ankara Yildirim Beyazit University School of Medicine, 06800, Bilkent, Ankara, Turkey.
- Department of Thoracic Surgery, Gazi University School of Medicine, 06560, Besevler, Ankara, Turkey.
Abstract
Occult lymph node metastasis (OLM) remains a major clinical challenge in patients with lung cancer despite comprehensive preoperative staging efforts. Although several predictors of OLM, including tumor diameter, metabolic activity, molecular characteristics, and radiomics, have been extensively investigated, the predictive role of tumor volume remains poorly defined. This study aimed to evaluate whether tumor volume, quantified semi-automatically using artificial intelligence (AI)-assisted Synapse 3D software, could predict OLM in patients with clinically early-stage lung cancer. Patients who underwent surgical resection for clinically staged cT1b-cN0M0 lung cancer between 2018 and 2024 were included. Tumor volume was quantified using Synapse 3D software. Associations between tumor volume and OLM were assessed using univariable and multivariable logistic regression analyses. Because tumor diameter and tumor volume may introduce multicollinearity in multivariable models, separate receiver operating characteristic (ROC) curves were generated, and the corresponding areas under the curve (AUCs) were compared using the DeLong test. A total of 248 patients were included, of whom 59 (23.8%) were female. The mean age was 65.4 years. OLM were identified in 10.8%, 4.8%, and 2.0% of patients at the N1, N2a, and N2b nodal stations, respectively. Multivariable logistic regression identified younger age (p = 0.004), greater tumor volume (p = 0.01), and histopathological subtypes other than adenocarcinoma and squamous cell carcinoma as independent predictors of OLM. Tumor diameter, extent of resection, surgical approach, and PET-CT metabolic activity were not significantly associated with OLM. ROC analysis demonstrated that tumor volume had significantly greater discriminatory ability for predicting OLM than tumor diameter (AUC: 69% vs. 58%; DeLong test, p = 0.004). AI-assisted tumor volumetry is an independent predictor of occult lymph node metastasis in patients with clinically early-stage lung cancer.