Pathologic stratification of operable lung adenocarcinoma using radiomics features extracted from dual energy CT images

Abstract

Purpose: To evaluate the usefulness of surrogate biomarkers as predictors of histopathologic tumor grade and aggressiveness using radiomics data from dual-energy computed tomography (DECT), with the ultimate goal of accomplishing stratification of early-stage lung adenocarcinoma for optimal treatment.

Results: Pathologic grade was divided into grades 1, 2, and 3. Multinomial logistic regression analysis revealed i-uniformity and 97.5th percentile CT attenuation value as independent significant factors to stratify grade 2 or 3 from grade 1. The AUC value calculated from leave-one-out cross-validation procedure for discriminating grades 1, 2, and 3 was 0.9307 (95% CI: 0.8514-1), 0.8610 (95% CI: 0.7547-0.9672), and 0.8394 (95% CI: 0.7045-0.9743), respectively.

Materials and methods: A total of 80 patients with 91 clinically and radiologically suspected stage I or II lung adenocarcinoma were prospectively enrolled. All patients underwent DECT and F-18-fluorodeoxyglucose (FDG) positron emission tomography (PET)/CT, followed by surgery. Quantitative CT and PET imaging characteristics were evaluated using a radiomics approach. Significant features for a tumor aggressiveness prediction model were extracted and used to calculate diagnostic performance for predicting all pathologic grades.

Conclusions: Quantitative radiomics values from DECT imaging metrics can help predict pathologic aggressiveness of lung adenocarcinoma.

Keywords: dual energy CT; heterogeneity; lung adenocarcinoma; radiomics; texture analysis.

Figure 1. Receiver operating characteristic (ROC) curves for prediction of pathologic grade based on significant imaging parameters

The AUC calculated from leave-one-out CV for discriminating grade 1 from the other grades was 0.9307, which was the highest among three ROC curves. The AUC was 0.8610 for discriminating grade 2 and 0.8394 for discriminating grade 3.

Figure 2. Flow chart of the study population

ADC = adenocarcinoma.

Figure 3. Overview of dual-energy imaging

(A) Diagram of three-material decomposition of voxel used by dual-energy software. (a), (b), (c), and (d) are fixed points of CT attenuation values from two different energies for air, fat, soft tissue, and iodine. Intercept x or y along iodine axis represents iodine content of voxel on this two-energy plot. (x) is the degree of enhancement of ground glass opacity nodule, whereas (y) is the degree of enhancement of solid nodule. (B) Three types of data sets were generated from the DECT scanning: the 80 kV, 140 kV, and enhanced weighted-average images. The weighted-average images were generated by combining the 140-kV and 80-kV data sets with a weighting factor of 0.6 (60% of the information derived from the 80 kV image and 40% derived from the 140 kV image) and these were approximately 120 kV images. The virtual non-enhanced images and iodine-enhanced images were made by using the liver Virtual Non- Contrast (VNC) application mode of dedicated dual-energy postprocessing software (Syngo Dual Energy; Siemens Medical Solutions, Forchheim, Germany).

Figure 4. Lung adenocarcinoma in a 76-year-old woman

(A) Photomicrograph shows internal scar tissue (*), surrounding areas of acinar and papillary (#) adenocarcinoma patterns, and lepidic pattern (arrows) (hematoxylin-eosin stain; original magnification, x10). (B) Schematic of tumor components shows estimated percentages of grade 1 (yellow area, 5%), grade 2 (blue area, 60%), grade 3 (green area, 5%), and central fibrosis (red area, 15%).