# Digital Twin for Radiation-Induced Immune Suppression Estimation and Reduction

> **NIH NIH P01** · MASSACHUSETTS GENERAL HOSPITAL · 2024 · $250,000

## Abstract

This application is being submitted in response to the Notice of Special Interest (NOSI) identified as NOT-CA-24-015.
Radiation-induced lymphopenia (RIL) is a common immunotoxicity associated with radiotherapy (RT) or
chemoradiotherapy. Severe RIL (sRIL) has been shown to significantly impact patient survival and other outcomes. Most
recently, it has become evident that sRIL can also diminish the effectiveness of immunotherapy significantly, when used
in conjunction with RT. We have demonstrated that proton therapy, because of its smaller dose bath, can lead to
significant sparing of the immune system. Project 2 of the parent NCI P01 is exploring the dosimetric and clinical
determinants of RIL and correlation of RIL with patient outcomes, developing personalized machine learning RIL
prediction models, and investigating strategies to mitigate RIL for cancers of the esophagus, lung, brain, liver, and
pancreas. Our preliminary studies have suggested that the most effective RIL mitigation approach is the use of intensity-
modulated proton therapy (IMPT) optimized based on personalized RIL models. The purpose of the research proposed
in this supplement application is to develop a DTRO framework with the following objectives: (1) Incorporate the
response deduced from lymphocyte depletion rate after initial treatment fractions (up to 5) to dynamically boost the RIL
prediction accuracy for a given patient and adapt (reoptimize) the IMPT plan to minimize immune toxicity. (2) Validate
and dynamically update the personalized RIL prediction models based on continuous feedback from actual responses.
(3) Quantify uncertainties in model predictions and consider them when making treatment decisions. Specific Aims
include the following: (1) Deploy and evaluate a bi-directional feedback DTRO system between the physical and virtual
components of the Phase I clinical trial planned in Project 2 of the parent grant. (2) Amend the Phase I trial schema to
form a feedback loop to dynamically update the personalized models. (3) Conduct uncertainty quantification analyses to
establish confidence intervals around the model predictions and incorporate uncertainty estimates in decisions before
treatment start and again after adaptation and reoptimization based on the response to initial fractions. The proposed
trial will be implemented for a cohort of esophagus patients to investigate the potential of DTROs in general and the
clinical effectiveness of the RIL mitigation approach. Our research will utilize multiscale data, including patient clinical
characteristics, dosimetric parameters, adjuvant and concurrent chemotherapy details, and immune system biomarkers
for a large cohort of esophagus patients previously treated at our institution. Similar data will be collected for each
patient enrolled and treated prospectively in the DTRO-based trial. This research will be a collaborative effort among
radiation oncologists, radiation physicists and data scientists. Our ultimate obj...

## Key facts

- **NIH application ID:** 11056996
- **Project number:** 3P01CA261669-04S1
- **Recipient organization:** MASSACHUSETTS GENERAL HOSPITAL
- **Principal Investigator:** Theodore S Hong
- **Activity code:** P01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $250,000
- **Award type:** 3
- **Project period:** 2021-09-21 → 2026-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/11056996

## Citation

> US National Institutes of Health, RePORTER application 11056996, Digital Twin for Radiation-Induced Immune Suppression Estimation and Reduction (3P01CA261669-04S1). Retrieved via AI Analytics 2026-06-11 from https://api.ai-analytics.org/grant/nih/11056996. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*