AUTHOR=Zhang Teng , Zhang Chengxiu , Zhong Yan , Sun Yingli , Wang Haijie , Li Hai , Yang Guang , Zhu Quan , Yuan Mei 

TITLE=A radiomics nomogram for invasiveness prediction in lung adenocarcinoma manifesting as part-solid nodules with solid components smaller than 6 mm

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.900049

DOI=10.3389/fonc.2022.900049

ISSN=2234-943X

ABSTRACT=Objective：To investigate whether radiomics can help radiologists and thoracic surgeons accurately predict invasive adenocarcinoma (IAC) manifesting as part-solid nodules (PSNs) with solid components < 6 mm and provide a basis for rational clinical decision-making.

Materials and Methods: Totally, 1210 patients (mean age ± standard deviation: 54.28 ± 11.38 years; 374 male and 836 female) from our and another hospital with 1248 PSNs pathologically diagnosed with adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or IAC were enrolled in study. Among them 1050 cases from our hospital were randomly divided into a derivation set (n= 735) and an internal validation set (n= 315), 198 cases from another hospital were used for external validation. Each labeled nodule was segmented and 105 radiomics features were extracted. Least absolute shrinkage and selection operator (LASSO) was used to calculate Rad-score and build radiomics model. Multivariable logistic regression was conducted to identify the clinicoradiological predictors and establish clinical-radiographic model. Combined model and predictive nomogram were developed based on identified clinicoradiological independent predictors and Rad-score using multivariable logistic regression analysis. Predictive performance of three models were compared via receiver operating curve (ROC) analysis. Decision curve analysis (DCA) was performed on both the internal and external validation sets to evaluate clinical utility of the nomogram.

Results: Radiomics model showed superior predictive performance than clinical-radiographic model in both internal and external validation sets (Az values, 0.884 vs 0.810, p = 0.001; 0.924 vs 0.855, p < 0.001, respectively). Combined model showed comparable predictive performance to radiomics model (Az values, 0.887 vs 0.884, p = 0.398; 0.917 vs 0.924, p = 0.271, respectively). Clinical application value of the nomogram developed based on Rad-score, maximum diameter and lesion shape was confirmed and DCA demonstrated application of Rad-score would be beneficial for radiologists predicting invasive lesions.  

Conclusions: Radiomics has the potential as an independent diagnostic tool to predict invasiveness of PSNs with solid components < 6 mm.