Consolidation Tumor Ratio Combined With Pathological Features Could Predict Status of Lymph Nodes of Early-Stage Lung Adenocarcinoma

Abstract

IntroductionStage IA lung adenocarcinoma manifested as part-solid nodules (PSNs), has attracted immense attention owing to its unique characteristics and the definition of its invasiveness remains unclear. We sought to develop a nomogram for predicting the status of lymph nodes of this kind of nodules.MethodsA total of 2,504 patients between September 2018 to October 2020 with part-solid nodules in our center were reviewed. Their histopathological features were extracted from paraffin sections, whereas frozen sections were reviewed to confirm the consistency of frozen sections and paraffin sections. Univariate and multivariate logistic regression analyses and Akaike information criterion (AIC) variable selection were performed to assess the risk factors of lymph node metastasis and construct the nomogram. The nomogram was subjected to bootstrap internal validation and external validation. The concordance index (C-index) was applied to evaluate the predictive accuracy and discriminative ability.ResultsWe enrolled 215 and 161 eligible patients in the training cohort and validation cohort, respectively. The sensitivity between frozen and paraffin sections on the presence of micropapillary/solid subtype was 78.4%. Multivariable analysis demonstrated that MVI, the presence of micropapillary/solid subtype, and CTR >0.61 were independently associated with lymph node metastasis (p < 0.01). Five risk factors were integrated into the nomogram. The nomogram demonstrated good accuracy in estimating the risk of lymph node metastasis, with a C-index of 0.945 (95% CI: 0.916–0.974) in the training cohort and a C-index of 0.975 (95% CI: 0.954–0.995) in the validation cohort. The model’s calibration was excellent in both cohorts.ConclusionThe nomogram established showed excellent discrimination and calibration and could predict the status of lymph nodes for patients with ≤3 cm PSNs. Also, this prediction model has the prediction potential before the end of surgery.