Interobserver size measurement variability in part-solid lung adenocarcinoma using pre-operative computed tomography

Measurement of solid size in early-stage lung adenocarcinoma by virtual 3D thin-section CT applied artificial intelligence

An artificial intelligence (AI) system that reconstructs virtual 3D thin-section CT (TSCT) images from conventional CT images by applying deep learning was developed. The aim of this study was to investigate whether virtual and real TSCT could measure the solid size of early-stage lung adenocarcinoma. The pair of original thin-CT and simulated thick-CT from the training data with TSCT images (thickness, 0.5-1.0 mm) of 2700 pulmonary nodules were used to train the thin-CT generator in the generative adversarial network (GAN) framework and develop a virtual TSCT AI system. For validation, CT images of 93 stage 0-I lung adenocarcinomas were collected, and virtual TSCTs were reconstructed from conventional 5-mm thick-CT images using the AI system. Two radiologists measured and compared the solid size of tumors on conventional CT and virtual and real TSCT. The agreement between the two observers showed an almost perfect agreement on the virtual TSCT for solid size measurements (intraclass correlation coefficient = 0.967, P < 0.001, respectively). The virtual TSCT had a significantly stronger correlation than that of conventional CT (P = 0.003 and P = 0.001, respectively). The degree of agreement between the clinical T stage determined by virtual TSCT and the clinical T stage determined by real TSCT was excellent in both observers (k = 0.882 and k = 0.881, respectively). The AI system developed in this study was able to measure the solid size of early-stage lung adenocarcinoma on virtual TSCT as well as on real TSCT.

Prognostic impact of deep learning-based quantification in clinical stage 0-I lung adenocarcinoma

OBJECTIVES

To evaluate the performance of automatic deep learning (DL) algorithm for size, mass, and volume measurements in predicting prognosis of lung adenocarcinoma (LUAD) and compared with manual measurements.

METHODS

A total of 542 patients with clinical stage 0-I peripheral LUAD and with preoperative CT data of 1-mm slice thickness were included. Maximal solid size on axial image (MSSA) was evaluated by two chest radiologists. MSSA, volume of solid component (SV), and mass of solid component (SM) were evaluated by DL. Consolidation-to-tumor ratios (CTRs) were calculated. For ground glass nodules (GGNs), solid parts were extracted with different density level thresholds. The prognosis prediction efficacy of DL was compared with that of manual measurements. Multivariate Cox proportional hazards model was used to find independent risk factors.

RESULTS

The prognosis prediction efficacy of T-staging (TS) measured by radiologists was inferior to that of DL. For GGNs, MSSA-based CTR measured by radiologists (RMSSA%) could not stratify RFS and OS risk, whereas measured by DL using 0HU (2D-AIMSSA0HU%) could by using different cutoffs. SM and SV measured by DL using 0 HU (AISM0HU% and AISV0HU%) could effectively stratify the survival risk regardless of different cutoffs and were superior to 2D-AIMSSA0HU%. AISM0HU% and AISV0HU% were independent risk factors.

CONCLUSION

DL algorithm can replace human for more accurate T-staging of LUAD. For GGNs, 2D-AIMSSA0HU% could predict prognosis rather than RMSSA%. The prediction efficacy of AISM0HU% and AISV0HU% was more accurate than of 2D-AIMSSA0HU% and both were independent risk factors.

CLINICAL RELEVANCE STATEMENT

Deep learning algorithm could replace human for size measurements and could better stratify prognosis than manual measurements in patients with lung adenocarcinoma.

KEY POINTS

• Deep learning (DL) algorithm could replace human for size measurements and could better stratify prognosis than manual measurements in patients with lung adenocarcinoma (LUAD). • For GGNs, maximal solid size on axial image (MSSA)-based consolidation-to-tumor ratio (CTR) measured by DL using 0 HU could stratify survival risk than that measured by radiologists. • The prediction efficacy of mass- and volume-based CTRs measured by DL using 0 HU was more accurate than of MSSA-based CTR and both were independent risk factors.

Is using a consolidation tumor ratio 0.5 as criterion feasible in daily practice? Evaluation of interobserver measurement variability of consolidation tumor ratio of lung cancer less than 3 cm in size

BACKGROUND

Consolidation tumor ratio (CTR) calculated as the ratio of the tumor consolidation diameter to the tumor maximum diameter on thin-section computed tomography (CT) of lung cancer has been reported as an important prognostic factor. It has also been used for treatment decision-making. This study aimed to investigate the interobserver variability of CTR measurements on preoperative CT and propose a clinically useful CTR-based classification criterion.

METHODS

We enrolled 119 patients who underwent surgery for suspected or diagnosed small-sized lung cancer (≤3.0 cm in diameter). Nine doctors reviewed preoperative CT scans to measure CTR. Interobserver variability of CTR measurements was evaluated using the coefficient of variation (CV) and Fleiss' κ. The prognostic effect of the CTR-based classification was assessed using the Kaplan-Meier method.

RESULTS

Interobserver variability of CTR measurement was the highest for tumors with the lowest CTR (CTR = 0); it decreased as CTR increased and reached a plateaued level of low variability (CV <0.5) at CTR of 0.5. We proposed a three-group classification based on the findings of CTR interobserver variability (CTR < 0.5, 0.5 ≤ CTR < 1, and CTR = 1). Interobserver agreement of the judgment of the CTR-based classification was excellent (Fleiss' κ = 0.81). The classification significantly stratified patient prognosis (p < 0.001, 5-year overall survival rates with CTR < 0.5, 0.5 ≤ CTR < 1, and CTR = 1 were 100, 88, and 73.8%, respectively).

CONCLUSIONS

CTR 0.5 is a clinically relevant and helpful cutoff for treatment decision-making in patients with early-stage lung cancer based on high interobserver agreement and good prognostic stratification.

Impact of a Ground-glass Opacity Component on c-Stage IA Lung Adenocarcinoma

Studies have shown that tumors with ground-glass opacity (GGO) components are associated with favorable outcomes. However, this view should be confirmed in an international cohort. We aimed to verify the impact of a GGO component on clinical (c)-stage IA lung adenocarcinoma and to describe the biological discrepancies between the part-solid and pure-solid groups. We evaluated 1333 cases of surgically resected c-stage IA lung adenocarcinomas, including 484 part-solid and 849 pure-solid tumors. Furthermore, we matched the solid size between the 2 groups and examined 470 patients. We compared the prognoses between the 2 groups before and after matching. The prognostic and biological differences were described before and after matching. Compared with the pure-solid group, the part-solid group was associated with favorable outcomes [5-year overall survival (OS) 99.4% vs 87.6%, P < 0.001; 5-year recurrence-free survival (RFS) 96.9% vs 82.2%, P < 0.001]. Similar results were obtained after matching (5-year OS 98.9% vs 92.2%, P = 0.012; 5-year RFS 95.0% vs 88.5%, P = 0.007). Multivariable analyses revealed that GGO component appearance was a factor of better OS and RFS. The part-solid tumor, regardless of the size of the solid component, had a similar outcome to the pure-solid tumor of c-stage T1a classification. Also, more epidermal growth factor receptor, human epidermal growth factor receptor-2 mutations, and receptor tyrosine kinase ROS-1-positive were observed in the part-solid group. In comparison, more wild types and Kirsten-Ras were observed in the pure-solid group. Adenocarcinomas with a GGO component were associated with superior outcomes. The GGO component should be considereda new clinical T descriptor. Early-stage lung adenocarcinomas with and without a GGO component may be 2 distinct tumor types.