Childhood Cancer Data Initiative Annual Symposium (Abstract Registration): Submission #77

Submission information
Submission Number: 77
Submission ID: 192594
Submission UUID: 7a4125cc-f57c-47aa-8d26-f1e490540392

Created: Thu, 08/27/2026 - 14:20
Completed: Thu, 08/27/2026 - 14:26
Changed: Thu, 08/27/2026 - 14:26

Remote IP address: 10.208.24.244
Submitted by: Anonymous
Language: English

Is draft: No
Abstract Submission for Poster Presentation
-------------------------------------------
Abstract Title:: Radiation dosimetry for the first large-scale systemic comparison of the risk of second cancers in children treated with proton versus photon therapy
Abstract::
The Pediatric Proton/Photon Therapy Comparison (PPTC) cohort study is the first large-scale study comparing the risk of second cancers in children treated with proton versus photon radiotherapy. The study has collected treatment records for more than 10,000 pediatric cancer patients from 17 hospitals. Because the cohort data are both large and distributed across multiple databases, we have deployed a cloud-based system to streamline data collection, monitoring, validation, and transfer to a high-performance computing (HPC) cluster. Data for ~7,500 patients have been collected to date in an industry-standard medical imaging format (DICOM), including radiation field parameters, planning computed tomography (CT) images with clinician-delineated anatomical structures, and dose distributions from the treatment planning system (TPS). Another critical component is the development of scalable methods for estimating individualized, organ-level radiation dose for epidemiological dose-response analyses. Planning CT scans typically cover only the treatment region, often omitting organs of interest for late effects research. To address this, we developed a method to extend partial-body CT images using a library of surrogate anatomies. In addition, due to variability and inconsistency in organ delineation and naming, we utilize a deep learning–based automatic segmentation tool to standardize organ delineation. Finally, we address limitations of TPS dose estimates by performing advanced Monte Carlo radiation transport simulations of both modalities. These simulations integrate patient data and detailed physics modeling and are efficiently executed on the NIH HPC cluster. This poster presents our efforts to implement a state-of-the-art dosimetry platform—from data collection to individualized dose calculations.

Authors::
1. First Name: Jungwook
   Last Name: Shin
   Degree(s): PhD
   Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
2. First Name: Matthew
   Middle Initial: M
   Last Name: Mille
   Degree(s): PhD
   Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
3. First Name: Caroline
   Last Name: Esposito
   Degree(s): BA
   Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
4. First Name: Todd
   Middle Initial: M
   Last Name: Gibson
   Degree(s): PhD
   Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
5. First Name: Keith
   Middle Initial: T
   Last Name: Griffin
   Degree(s): PhD
   Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
6. First Name: Jae Won
   Last Name: Jung
   Degree(s): PhD
   Organization: Department of Radiation Oncology, East Carolina University, Greenville, NC
7. First Name: Choonik
   Last Name: Lee
   Degree(s): PhD
   Organization: Department of Radiation Oncology, University of Michigan, Ann Arbor, MI
8. First Name: Sergio
   Last Name: Morató
   Degree(s): PhD
   Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
9. First Name: Torunn
   Middle Initial: I
   Last Name: Yock
   Degree(s): MD
   Organization: Massachusetts General Hospital and Harvard Medical School, Boston, MA
10. First Name: Amy
    Last Name: Berrington de González
    Degree(s): DPhil
    Organization: Division of Genetics and Epidemiology, Institute for Cancer Research, London, UK
11. First Name: Choonsik
    Last Name: Lee
    Degree(s): PhD
    Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
12. First Name: Cari
    Middle Initial: M
    Last Name: Kitahara
    Degree(s): PhD
    Organization: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD

Presenting Author:: Matthew M. Mille
Institution:: Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD
Email Address:: matthew.mille@nih.gov