Dataset Information

Clinical Translation of a Deep Learning Model of Radiation-Induced Lymphopenia for Esophageal Cancer.

ABSTRACT:

Purpose

Radiation-induced lymphopenia is a common immune toxicity that adversely impacts treatment outcomes. We report here our approach to translate a deep-learning (DL) model developed to predict severe lymphopenia risk among esophageal cancer into a strategy for incorporating the immune system as an organ-at-risk (iOAR) to mitigate the risk.

Materials and methods

We conducted "virtual clinical trials" utilizing retrospective data for 10 intensity-modulated radiation therapy (IMRT) and 10 passively-scattered proton therapy (PSPT) esophageal cancer patients. For each patient, additional treatment plans of the modality other than the original were created employing standard-of-care (SOC) dose constraints. Predicted values of absolute lymphocyte count (ALC) nadir for all plans were estimated using a previously-developed DL model. The model also yielded the relative magnitudes of contributions of iOARs dosimetric factors to ALC nadir, which were used to compute iOARs dose-volume constraints, which were incorporated into optimization criteria to produce "IMRT-enhanced" and "intensity-modulated proton therapy (IMPT)-enhanced" plans.

Results

Model-predicted ALC nadir for the original IMRT (IMRT-SOC) and PSPT plans agreed well with actual values. IMPT-SOC showed greater immune sparing vs IMRT and PSPT. The average mean body doses were 13.10 Gy vs 7.62 Gy for IMRT-SOC vs IMPT-SOC for patients treated with IMRT-SOC; and 8.08 Gy vs 6.68 Gy for PSPT vs IMPT-SOC for patients treated with PSPT. For IMRT patients, the average predicted ALC nadir of IMRT-SOC, IMRT-enhanced, IMPT-SOC, and IMPT-enhanced was 281, 327, 351, and 392 cells/µL, respectively. For PSPT patients, the average predicted ALC nadir of PSPT, IMPT-SOC, and IMPT-enhanced was 258, 316, and 350 cells/µL, respectively. Enhanced plans achieved higher predicted ALC nadir, with an average improvement of 40.8 cells/µL (20.6%).

Conclusion

The proposed DL model-guided strategy to incorporate the immune system as iOAR in IMRT and IMPT optimization has the potential for radiation-induced lymphopenia mitigation. A prospective clinical trial is planned.

SUBMITTER: Hu Z

PROVIDER: S-EPMC11369390 | biostudies-literature | 2024 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Clinical Translation of a Deep Learning Model of Radiation-Induced Lymphopenia for Esophageal Cancer.

Hu Zongsheng Z Mohan Radhe R Chu Yan Y Wang Xiaochun X van Rossum Peter S N PSN Chen Yiqing Y Grayson Madison E ME Gearhardt Angela G AG Grassberger Clemens C Zhi Degui D Hobbs Brian P BP Lin Steven H SH Cao Wenhua W

International journal of particle therapy 20240805

<h4>Purpose</h4>Radiation-induced lymphopenia is a common immune toxicity that adversely impacts treatment outcomes. We report here our approach to translate a deep-learning (DL) model developed to predict severe lymphopenia risk among esophageal cancer into a strategy for incorporating the immune system as an organ-at-risk (iOAR) to mitigate the risk.<h4>Materials and methods</h4>We conducted "virtual clinical trials" utilizing retrospective data for 10 intensity-modulated radiation therapy (IM ...[more]

PMID: 39228692

Similar Datasets

Project description:PurposeTo assess possible differences in radiation-induced lymphocyte depletion for esophageal cancer patients being treated with the following 3 treatment modalities: intensity-modulated radiation therapy (IMRT), passive scattering proton therapy (PSPT), and intensity-modulated proton therapy (IMPT).Methods and materialsWe used 2 prediction models to estimate lymphocyte depletion based on dose distributions. Model I used a piecewise linear relationship between lymphocyte survival and voxel-by-voxel dose. Model II assumes that lymphocytes deplete exponentially as a function of total delivered dose. The models can be fitted using the weekly absolute lymphocyte counts measurements collected throughout treatment. We randomly selected 45 esophageal cancer patients treated with IMRT, PSPT, or IMPT at our institution (15 per modality) to demonstrate the fitness of the 2 models. A different group of 10 esophageal cancer patients who had received PSPT were included in this study of in silico simulations of multiple modalities. One IMRT and one IMPT plan were created, using our standards of practice for each modality, as competing plans to the existing PSPT plan for each patient. We fitted the models by PSPT plans used in treatment and predicted absolute lymphocyte counts for IMRT and IMPT plans.ResultsModel validation on each modality group of patients showed good agreement between measured and predicted absolute lymphocyte counts nadirs with mean squared errors from 0.003 to 0.023 among the modalities and models. In the simulation study of IMRT and IMPT on the 10 PSPT patients, the average predicted absolute lymphocyte count (ALC) nadirs were 0.27, 0.35, and 0.37 K/μL after IMRT, PSPT, and IMPT treatments using Model I, respectively, and 0.14, 0.22, and 0.33 K/μL using Model II.ConclusionsProton plans carried a lower predicted risk of lymphopenia after the treatment course than did photon plans. Moreover, IMPT plans outperformed PSPT in terms of predicted lymphocyte preservation.

Dataset Information

Clinical Translation of a Deep Learning Model of Radiation-Induced Lymphopenia for Esophageal Cancer.

Purpose

Materials and methods

Results

Conclusion

Publications

Clinical Translation of a Deep Learning Model of Radiation-Induced Lymphopenia for Esophageal Cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets