3D computed tomography density of stage T1 lung cancer, which value is most stable for predicting its invasiveness?

Abstract

Background: Medical research indicates that computed tomography (CT) values are vital for diagnosing and predicting the invasiveness of pulmonary nodules. This study investigates which three-dimensional (3D) CT density measurement value is most stable for predicting the invasiveness of T1 stage lung adenocarcinoma and advantageous for preoperative planning.

Methods: A retrospective analysis was conducted on a total of 2,080 patients with pulmonary nodules of atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive carcinoma (MIA), and invasive adenocarcinoma (IAC) confirmed by surgery in six centers including Zhengzhou People's Hospital (between November 2017 and November 2023), Cancer Hospital Chinese Academy of Medical Sciences (between August 2023 and April 2024), The First Affiliated Hospital of Zhengzhou University (between June 2023 and December 2023), Huadong Hospital Affiliated to Fudan University (between May 2012 and September 2018), Beijing Liangxiang Hospital (between May 2017 and December 2023), and the Ninth People's Hospital of Zhengzhou (between November 2017 and January 2024). Clinical history and thin-layer chest CT were collected. Patients were classified into non-invasive (AAH, AIS, MIA; n=1,297) and invasive (IAC; n=783) groups based on pathology. Thin-layer CT images were analyzed using the Shukun artificial intelligence (AI) workstation, and subgroup analysis was performed according to the 3D maximum diameter (T1a: ≤10 mm, T1b: 10 mm < T1b ≤ 20 mm, T1c: 20 mm < T1c ≤ 30 mm). 3D CT density-related features were selected, receiver operating characteristic (ROC) curves were drawn, and the area under the curve (AUC) and 95% confidence interval (CI) were calculated. Further subgroup analysis was conducted on T1a, T1b, and T1c groups to determine the most stable CT value for predicting invasiveness.

Results: The study encompassed 2,080 nodules with a gender distribution of 33.17% male and 66.83% female, averaging 56.54±11.31 years. Nodule distribution: 728 (35.00%) in right upper lobe (RUL), 117 (5.62%) in right middle lobe (RML), 393 (18.89%) in right lower lobe (RLL), 531 (25.53%) in left upper lobe (LUL), and 311 (14.95%) in left lower lobe (LLL). The non-invasive group included 1,297 nodules, and the invasive group 783. The 90th percentile of CT values emerged as the most stable indicator, with an AUC of 0.863 (95% CI: 0.864-0.879), at a threshold of -241.5 Hounsfield unit (HU). Subgroup analysis confirmed the 90th percentile of CT values as the most stable predictor in all groups, with AUCs and 95% CIs of [0.830, 0.778-0.881], [0.843, 0.817-0.868], and [0.893, 0.850-0.936], respectively.

Conclusions: The 90th percentile of CT values reliably predicts invasiveness in stage T1 lung adenocarcinoma. This finding facilitates the clinical assessment of its invasiveness, enabling a better grasp of the timing of clinical intervention and the formulation of surgical plans.

Keywords: 90th percentile; Artificial intelligence (AI); invasiveness; lung adenocarcinoma; three-dimensional computed tomography values (3D CT values).

Figure 1

Flowchart of identifying the most stable CT density-related features for predicting the invasiveness of stage T1 lung adenocarcinoma. Hospital 1: Zhengzhou People’s Hospital. Hospital 2: Cancer Hospital of the Chinese Academy of Medical Sciences. Hospital 3: The First Affiliated Hospital of Zhengzhou University. Hospital 4: Huadong Hospital affiliated with Fudan University. Hospital 5: Beijing Liangxiang Hospital. Hospital 6: Zhengzhou Ninth People’s Hospital. AI, artificial intelligence; CI, confidence interval; CT, computed tomography; ROC, receiver operating characteristic.

Figure 2

A 69-year-old female with a ground-glass nodule (17 mm × 10 mm) at the upper right lobe junction on CT lung window image, with lobulated margins and vascular proximity, was diagnosed with invasive adenocarcinoma post-surgery. White arrow indicates the pulmonary nodule. (A) Transverse position. (B) Coronal plane. (C) Sagittal plane. (D) Pathological HE image of the nodule (×100). CT, computed tomography; HE, hematoxylin-eosin staining.

Figure 3

ROC curves of CT density-related features for non-invasive and invasive groups. The figure illustrates the ROC curves for various CT density-related features that distinguish between non-invasive and invasive groups. The AUC values indicate the predictive performance of each feature, with higher values suggesting better discrimination. The 90th percentile of CT values shows the highest AUC among the assessed features. AUC, area under the curve; CI, confidence interval; CT, computed tomography; ROC, receiver operating characteristic.

Figure 4

ROC curves for CT density-related features in the T1a subgroup analysis. CI, confidence interval; CT, computed tomography; ROC, receiver operating characteristic.

Figure 5

ROC curves for CT density-related features in the T1b subgroup analysis. CI, confidence interval; CT, computed tomography; ROC, receiver operating characteristic.

Figure 6

ROC curves for CT density-related features in the T1c subgroup analysis. CI, confidence interval; CT, computed tomography; ROC, receiver operating characteristic.