Background/Objectives: No reliable noninvasive biomarkers are available to predict RT-induced cardiotoxicity. Because the pericardial sac is a fast responder to cardiac injury, we investigated whether RT-induced radiographic pericardial changes might serve as early imaging biomarkers for late cardiotoxicity. Methods: We performed a retrospective study of 476 patients (210 males, 266 females; median age, 69 years; median follow-up, 26.7 months) treated with chemo-RT for small cell and non-small cell lung cancers at one single institution from 2009 to 2020. The heart and its 4 mm outmost layer (representing the pericardial sac) were contoured on standard-of-care baseline CTs. Six-month post-RT follow-up CTs were deformably registered on the baseline CTs. Data were harmonized for the effect of contrast. We labeled voxels as Fat, Fluid, Heme, Fibrous, and Calcification using Hounsfield units (HUs). We studied pericardial HU-change histograms as well as volume change and voxel-based mass change in each tissue composition. Results: Pericardial HU-change histograms had skewed distributions with a mean that was significantly correlated with mean pericardial dose. Voxels within Fluid, Heme, and Fibrous had mass changes consistent with the dose. In Kaplan–Meier curves, Fibrous and Heme volume changes (translating into thickening and effusion), Fat mass change, mean doses to heart and pericardium, history of cardiac disease, and being male were significantly associated with shorter survival, whereas thickening and effusion were significantly associated with shorter time to a post-RT cardiovascular disease diagnosis. Conclusions: Pericardium composition distribution has dose-dependent changes detectable on standard-of-care CTs at around 6 months post-RT and may serve as surrogate markers for clinically relevant cardiotoxicity. The findings should be validated with additional research.

Background: Deep learning (DL) methods have shown strong performance in predicting dose distributions from patient anatomy and target structures; however, accurate dose prediction does not guarantee a deliverable plan after conversion into machine parameters and dose recalculation in the treatment planning system (TPS), motivating training objectives that explicitly couple prediction to deliverability. Lymphoma presents substantial clinical heterogeneity in target location, extent, and prescription/fractionation, so fast automated methods that produce deliverable plans are particularly valuable. Purpose: To investigate whether a deliverability‐aware training framework improves the end‐to‐end deliverability of DL‐generated intensity‐modulated radiotherapy (IMRT) plans for lymphoma. Methods: A retrospective single‐institution dataset of 619 plans (571 patients; 2009–2021) was used. Clinical volumetric modulated arc therapy (VMAT) plans were converted to standardized 15‐field coplanar sliding‐window IMRT reference plans in Eclipse using automated scripting. Two DL approaches were compared: a baseline dose‐only objective model (DL‐DO) and a multi‐objective model (DL‐MO) trained with dose supervision, fluence supervision, and a dose–fluence–dose cycle‐consistency constraint implemented via fixed inverse and forward networks. Predicted fluence maps were imported into the TPS, leaf‐sequenced, and recalculated with AcurosXB without further optimization; plan quality was evaluated on a held‐out test set ( n = 61) using planning target volume (PTV) D ₉₈ , D ₂ , and D _max with paired Wilcoxon testing and Holm correction. Results: DL‐MO produced significantly better machine‐deliverable plans than DL‐DO, improving coverage and reducing hotspots (median paired differences: +12.31 percentage points (pp) in D ₉₈ , −3.43 pp in D ₂ , and −3.18 pp in D _max ; Holm‐corrected p < 0.0001 for all). Mean ± SD PTV metrics were D ₉₈ = 90.10 ± 3.74% vs. 78.59 ± 5.92%, D ₂ = 106.28 ± 1.13% vs. 109.65 ± 2.40%, and D _max = 110.32 ± 2.20% vs. 113.27 ± 3.03% (DL‐MO vs. DL‐DO). DL‐MO approached the TPS reference IMRT plans ( D ₉₈ = 91.62 ± 3.45%) with residual hotspot elevation. Conclusions: Cycle‐consistency training improves the end‐to‐end deliverability of DL‐generated IMRT plans for heterogeneous lymphoma, yielding substantially better TPS‐recalculated, machine‐deliverable plans than conventional dose‐only supervision while maintaining sub‐second fluence‐map generation. However, statistically significant differences from the TPS reference remained for D ₉₈ , D ₂ , and D _max .

Background: Intensity‐modulated radiation therapy (IMRT) is widely used for lymphoma because a multileaf collimator (MLC) can sculpt dose around irregular targets. Planning, however, is time‐consuming and difficult to standardize. Fluence map prediction (FMP) using deep learning (DL) could potentially be used to automate planning, but it has not been evaluated in full‐body lymphoma, where prescriptions, organ‐at‐risk (OAR) sets, and anatomical presentations vary substantially. Here, we evaluate multiple DL configurations to determine whether they can be adapted to lymphoma and generate IMRT plans that are deliverable in a clinical treatment planning system (TPS). Purpose: To investigate the feasibility and limitations of DL–based dose and fluence prediction methods in the heterogeneous setting of lymphoma. Methods: A single‐institution lymphoma cohort (571 patients, 619 plans) was replanned from clinical volumetric modulated arc therapy (VMAT) to a consistent 15‐field IMRT reference using the Eclipse Scripting API (ESAPI), yielding 9285 beam‐level pairs. We implemented a two‐stage DL pipeline that first predicts beam‐wise dose from CT and planning target volume (PTV), then maps dose to fluence, and finally computes the corresponding dose in the TPS after leaf sequencing. We evaluated several architectural and training variants, from replication‐focused models to configurations with a clinical loss based on dose–volume histogram (DVH) aimed at moving beyond simple plan replication. For each configuration, we analysed how these choices influenced predicted dose realism and the quality of the resulting plans. Results: In Stage 1 (dose prediction), replication‐focused DL models were numerically inferior to the IMRT and VMAT references, while DVH‐driven variants yielded superior target DVHs. After fluence mapping and TPS sequencing, all configurations produced machine‐deliverable IMRT plans but with lower target coverage and higher hot spots than the references; fluence was largely preserved by MLC sequencing, indicating that current limitations mainly stem from the predicted dose distributions and the learned dose‐to‐fluence mapping. Conclusions: DL–based dose and fluence prediction can generate fast, machine‐deliverable IMRT plans for heterogeneous lymphoma that lie in the clinical ballpark, but they remain dosimetrically inferior to manual IMRT and VMAT, highlighting the current limitations of these methods for automated planning in this setting.