Improving prediction accuracy of spread through air spaces in clinical-stage T1N0 lung adenocarcinoma using computed tomography imaging models

Abstract

Objectives: To develop computed tomography (CT)-based models to increase the prediction accuracy of spread through air spaces (STAS) in clinical-stage T1N0 lung adenocarcinoma.

Methods: Three cohorts of patients with stage T1N0 lung adenocarcinoma (n = 1258) were analyzed retrospectively. Two models using radiomics and deep neural networks (DNNs) were established to predict the lung adenocarcinoma STAS status. For the radiomic models, features were extracted using PyRadiomics, and 10 features with nonzero coefficients were selected using least absolute shrinkage and selection operator regression to construct the models. For the DNN models, a 2-stage (supervised contrastive learning and fine-tuning) deep-learning model, MultiCL, was constructed using CT images and the STAS status as training data. The area under the curve (AUC) was used to verify the predictive ability of both model types for the STAS status.

Results: Among the radiomic models, the linear discriminant analysis model exhibited the best performance, with AUC values of 0.8944 (95% confidence interval [CI], 0.8241-0.9502) and 0.7796 (95% CI, 0.7089-0.8448) for predicting the STAS status on the test and external validation cohorts, respectively. Among the DNN models, MultiCL exhibited the best performance, with AUC values of 0.8434 (95% CI, 0.7580-0.9154) for the test cohort and 0.7686 (95% CI, 0.6991-0.8316) for the external validation cohort.

Conclusions: CT-based imaging models (radiomics and DNNs) can accurately identify the STAS status of clinical-stage T1N0 lung adenocarcinoma, potentially guiding surgical decision making and improving patient outcomes.

Keywords: clinical-stage T1N0 lung adenocarcinoma; deep neural network; radiomics; spread through air spaces.

Graphical abstract

(a) STAS(+); (b) STAS(−).

Figure 1

Recruitment flowchart for owed in this study.

Figure 2

Definition of STAS-positive in histologic examination: STAS-positive was defined as the presence of tumor cells in the lung space beyond the margin of the primary tumor (×10 objective lens, ×10 magnification). A, STAS-positive case (blue box, ×80 magnification) and B, STAS-negative.

Figure 3

A, Overall framework for constructing the radiomic model. B, Overall MultiCL model framework. Single-channel 3D images are used as an example. In the actual training process, multi-channel 3D images are used as the deep neural network input.

Figure 4

Kaplan-Meier curves (with 95% confidence interval) of relapse-free survival (RFS) and overall survival (OS). According to the Kaplan-Meier method and log-rank test in the survival analysis, for all patients in the primary cohort, the RFS rate was 98.7% and the OS rate was 99.7%. A, The 5-year RFS rate was statistically significantly higher for the STAS-negative group compared to the STAS-positive group (97.2% vs 91.4%; P < .001). B, The 5-year OS rate was statistically significantly higher for the STAS-negative group compared to the STAS-positive group (100% vs 94.7%; P < .001). For all patients in the validation cohort, the RFS rate was 96.0% and the OS rate was 97.8%. C, The 5-year RFS rate was statistically significantly higher for the STAS-negative group compared to the STAS-positive group (98.1% vs 84.8%; P < .001). D, The 5-year OS rate was statistically significantly higher for the STAS-negative group compared to the STAS-positive group (99.3% vs 88.7%; P < .001).

Figure 5

Area under the curve of (A) radiomics and (B) deep neural network DNN models for training, testing, and external validation cohorts. TPR, True-positive rate; FPR, false-positive rate.

Figure 5

Area under the curve of (A) radiomics and (B) deep neural network DNN models for training, testing, and external validation cohorts. TPR, True-positive rate; FPR, false-positive rate.

Figure 5

Area under the curve of (A) radiomics and (B) deep neural network DNN models for training, testing, and external validation cohorts. TPR, True-positive rate; FPR, false-positive rate.