Deep learning-based automatic segmentation of cardiac substructures for lung cancers.

PURPOSE: Accurate and comprehensive segmentation of cardiac substructures is crucial for minimizing the risk of radiation-induced heart disease in lung cancer radiotherapy. We sought to develop and validate deep learning-based auto-segmentation models for cardiac substructures. MATERIALS AND METHODS: Nineteen cardiac substructures (whole heart, 4 heart chambers, 6 great vessels, 4 valves, and 4 coronary arteries) in 100 patients treated for non-small cell lung cancer were manually delineated by two radiation oncologists. The valves and coronary arteries were delineated as planning risk volumes. An nnU-Net auto-segmentation model was trained, validated, and tested on this dataset with a split ratio of 75:5:20. The auto-segmented contours were evaluated by comparing them with manually drawn contours in terms of Dice similarity coefficient (DSC) and dose metrics extracted from clinical plans. An independent dataset of 42 patients was used for subjective evaluation of the auto-segmentation model by 4 physicians. RESULTS: The average DSCs were 0.95 (+/- 0.01) for the whole heart, 0.91 (+/- 0.02) for 4 chambers, 0.86 (+/- 0.09) for 6 great vessels, 0.81 (+/- 0.09) for 4 valves, and 0.60 (+/- 0.14) for 4 coronary arteries. The average absolute errors in mean/max doses to all substructures were 1.04 (+/- 1.99) Gy and 2.20 (+/- 4.37) Gy. The subjective evaluation revealed that 94% of the auto-segmented contours were clinically acceptable. CONCLUSION: We demonstrated the effectiveness of our nnU-Net model for delineating cardiac substructures, including coronary arteries. Our results indicate that this model has promise for studies regarding radiation dose to cardiac substructures.

Actinium-225 Targeted Agents: Where Are We Now?

α-particle targeted radionuclide therapy has shown promise for optimal cancer management, an exciting new era for brachytherapy. Alpha-emitting nuclides can have significant advantages over gamma- and beta-emitters due to their high linear energy transfer (LET). While their limited path length results in more specific tumor 0kill with less damage to surrounding normal tissues, their high LET can produce substantially more lethal double strand DNA breaks per radiation track than beta particles. Over the last decade, the physical and chemical attributes of Actinium-225 (225Ac) including its half-life, decay schemes, path length, and straightforward chelation ability has peaked interest for brachytherapy agent development. However, this has been met with challenges including source availability, accurate modeling for standardized dosimetry for brachytherapy treatment planning, and laboratory space allocation in the hospital setting for on-demand radiopharmaceuticals production. Current evidence suggests that a simple empirical approach based on 225Ac administered radioactivity may lead to inconsistent outcomes and toxicity. In this review article, we highlight the recent advances in 225Ac source production, dosimetry modeling, and current clinical studies.

Modulators of radiation-induced cardiopulmonary toxicities for non-small cell lung cancer: Integrated cytokines, single nucleotide variants, and HBP systems imaging.

Radiation therapy (RT)-induced cardiopulmonary toxicities remain dose-limiting toxicities for patients receiving radiation dosages to the thorax, especially for lung cancer. Means of monitoring and predicting for those receiving RT or concurrent chemoradiation therapy before treatment begins in individual patients could benefit early intervention to prevent or minimize RT-induced side effects. Another aspect of an individual's susceptibility to the adverse effects of thoracic irradiation is the immune system as reflected by phenotypic factors (patterns of cytokine expressions), genotypic factors (single nucleotide variants SNVs; formerly single nucleotide polymorphisms [SNPs]), and aspects of quantitative cellular imaging. Levels of transcription, production, and functional activity of cytokines are often influenced by SNVs that affect coding regions in the promoter or regulatory regions of cytokine genes. SNVs can also lead to changes in the expression of the inflammatory cytokines, interferons, interleukins (IL-6, IL-17) and tumor necrosis factors (TNF-α) at the protein level. RT-induced cardiopulmonary toxicities could be quantified by the uptake of 18F-fluorodeoxyglucose (FDG), however, FDG is a sensitive but not specific biomarker in differential diagnosis between inflammation/infection and tumor recurrence. FDG is suitable for initial diagnosis of predisposed tissue injuries in non-small cell lung cancer (NSCLC). 99mTc-ethylenedicysteine-glucosamine (99mTc-EC-G) was able to measure tumor DNA proliferation and myocardial ischemia via hexosamine biosynthetic pathways (HBP). Thus, 99mTc-EC-G could be an alternative to FDG in the assessment of RT doses and select patients in HBP-directed targets for optimal outcomes. This article reviewed correlative analyses of pro-inflammatory cytokines, genotype SNVs, and cellular imaging to improve the diagnosis, prognosis, monitoring, and prediction of RT-induced cardiopulmonary toxicities in NSCLC.