Application of artificial intelligence in quantifying lung deposition dose of black carbon in people with exposure to ambient combustion particles

BACKGROUND

Understanding lung deposition dose of black carbon is critical to fully reconcile epidemiological evidence of combustion particles induced health effects and inform the development of air quality metrics concerning black carbon. Macrophage carbon load (MaCL) is a novel cytology method that quantifies lung deposition dose of black carbon, however it has limited feasibility in large-scale epidemiological study due to the labor-intensive manual counting.

OBJECTIVE

To assess the association between MaCL and episodic elevation of combustion particles; to develop artificial intelligence based counting algorithm for MaCL assay.

METHODS

Sputum slides were collected during episodic elevation of ambient PM2.5 (n = 49, daily PM2.5 > 10 µg/m3 for over 2 weeks due to wildfire smoke intrusion in summer and local wood burning in winter) and low PM2.5 period (n = 39, 30-day average PM2.5 < 4 µg/m3) from the Lovelace Smokers cohort.

RESULTS

Over 98% individual carbon particles in macrophages had diameter <1 µm. MaCL levels scored manually were highly responsive to episodic elevation of ambient PM2.5 and also correlated with lung injury biomarker, plasma CC16. The association with CC16 became more robust when the assessment focused on macrophages with higher carbon load. A Machine-Learning algorithm for Engulfed cArbon Particles (MacLEAP) was developed based on the Mask Region-based Convolutional Neural Network. MacLEAP algorithm yielded excellent correlations with manual counting for number and area of the particles. The algorithm produced associations with ambient PM2.5 and plasma CC16 that were nearly identical in magnitude to those obtained through manual counting.

IMPACT STATEMENT

Understanding lung black carbon deposition is crucial for comprehending health effects of combustion particles. We developed "Machine-Learning algorithm for Engulfed cArbon Particles (MacLEAP)", the first artificial intelligence algorithm for quantifying airway macrophage black carbon. Our study bolstered the algorithm with more training images and its first use in air pollution epidemiology. We revealed macrophage carbon load as a sensitive biomarker for heightened ambient combustion particles due to wildfires and residential wood burning.

DNA-PKc deficiency drives pre-malignant transformation by reducing DNA repair capacity in concert with reprogramming the epigenome in human bronchial epithelial cells

The expression of DNA-dependent protein kinase catalytic subunit (DNA-PKc) is highly variable in smokers and reduced enzyme activity has been associated with risk for lung cancer. An in vitro model of lung pre-malignancy was used to evaluate the role of double-strand break DNA repair capacity in transformation of hTERT/CDK4 immortalized human bronchial epithelial cells (HBECs) and reprograming of the epigenome. Here we show that knockdown of DNA-PKc to levels simulating haploinsufficiency dramatically reduced DNA repair capacity following challenge with bleomycin and significantly increased transformation efficiency of HBEC lines exposed weekly for 12 weeks to this radiomimetic. Transformed HBEC lines with wild type or knockdown of DNA-PKc showed altered expression of more than 1,000 genes linked to major cell regulatory pathways involved in lung cancer. While lung cancer driver mutations were not detected in transformed clones, more than 300 genes that showed reduced expression associated with promoter methylation in transformed clones or predictive for methylation in malignant tumors were identified. These studies support reduced DNA repair capacity as a key factor in the initiation and clonal expansion of pre-neoplastic cells and double-strand break DNA damage as causal for epigenetic mediated silencing of many lung cancer-associated genes. The fact that DNA damage, repair, and epigenetic silencing of genes are causal for many other cancers that include colon and prostate extends the generalizability and impact of these findings.