Elevated Coronary Artery Calcium Quantified by a Validated Deep Learning Model From Lung Cancer Radiotherapy Planning Scans Predicts Mortality

Abstract

PURPOSE Coronary artery calcium (CAC) quantified on computed tomography (CT) scans is a robust predictor of atherosclerotic coronary disease; however, the feasibility and relevance of quantitating CAC from lung cancer radiotherapy planning CT scans is unknown. We used a previously validated deep learning (DL) model to assess whether CAC is a predictor of all-cause mortality and major adverse cardiac events (MACEs). METHODS Retrospective analysis of non–contrast-enhanced radiotherapy planning CT scans from 428 patients with locally advanced lung cancer is performed. The DL-CAC algorithm was previously trained on 1,636 cardiac-gated CT scans and tested on four clinical trial cohorts. Plaques ≥ 1 cubic millimeter were measured to generate an Agatston-like DL-CAC score and grouped as DL-CAC = 0 (very low risk) and DL-CAC ≥ 1 (elevated risk). Cox and Fine and Gray regressions were adjusted for lung cancer and cardiovascular factors. RESULTS The median follow-up was 18.1 months. The majority (61.4%) had a DL-CAC ≥ 1. There was an increased risk of all-cause mortality with DL-CAC ≥ 1 versus DL-CAC = 0 (adjusted hazard ratio, 1.51; 95% CI, 1.01 to 2.26; P = .04), with 2-year estimates of 56.2% versus 45.4%, respectively. There was a trend toward increased risk of major adverse cardiac events with DL-CAC ≥ 1 versus DL-CAC = 0 (hazard ratio, 1.80; 95% CI, 0.87 to 3.74; P = .11), with 2-year estimates of 7.3% versus 1.2%, respectively. CONCLUSION In this proof-of-concept study, CAC was effectively measured from routinely acquired radiotherapy planning CT scans using an automated model. Elevated CAC, as predicted by the DL model, was associated with an increased risk of mortality, suggesting a potential benefit for automated cardiac risk screening before cancer therapy begins.

Early Cardio-Oncology Intervention in Thoracic Radiotherapy: Results of the Cardiac Aggressive Risk Mitigation (CARMA) Pilot Study

“…Furthermore, our study demonstrated how readily available CV risk markers, such as the presence of CAC on RT planning CTs and/or LAD radiation dose exposure, can capture additional at-risk patients for CV events. Baseline CAC on RT planning CT scans has been associated with increased risk of major cardiac events in patients with breast 27,28 and lung cancer 9,[29][30][31] , yet CAC assessment prior to initiation of cancer treatment is not routinely performed. Similarly, evidence suggests that cardiac substructure radiation dose exposure, particularly to the left coronary arteries, is a better predictor of MACE compared to whole heart dose (i.e., mean heart dose) and warrants inclusion into the global CV risk assessment of patients receiving cardiotoxic cancer therapies 6, [10][11][12][13] We further characterized the prevalence of CHIP at baseline and after completion of RT.…”

Predicting Adverse Cardiac Events After Radiotherapy for Locally Advanced Non–Small Cell Lung Cancer

“…Given the strong association between CAC and cardiotoxicity, it is necessary to establish standardized approaches to compute CAC score. Although computationally intensive voxel-calculated scoring approaches have been used, 13 , 17 , 21 visual CAC scoring methods, such as the method of Chiles et al 17 we validated, have generated comparable risk assessments. We observed a high correlation between the visual scoring method and Agatston scoring ( Supplemental Figure 2 ).…”

“…CAC scores were tabulated using a modified Chiles method, 17 evaluating the initial treatment simulation CT study for all patients. 21 Although the initial intention was to use automated Agatston scoring, a significant portion of images were obtained with intravenous contrast, rendering automated Agatston methods suboptimal for scoring. Thus, CAC scoring was performed through visual analysis of individual coronary vessels ( Figure 1B ).…”

Coronary artery calcium score on standard of care oncologic CT scans for the prediction of adverse cardiovascular events in patients with non-small cell lung cancer treated with concurrent chemoradiotherapy

“…CAC measured on non-gated CT correlated well with the traditional gated CAC protocol acquired by a multi-detector CT (MDCT) in a study of 163 healthy participants who underwent both imaging studies on the same day (Spearman correlation coefficient 0.83, P < 0.001) (15). In a recent retrospective analysis of 428 patients with locally advanced lung cancer, a deep learning method was used to generate an Agatston-like CAC score and showed that CAC ≥1 was associated with increased risk mortality and a trend toward increased risk of major adverse cardiac events (23). Our study included patients with NSCLC stage II-IV treated with concurrent chemoradiation therapy, including photon or proton RT.…”