Click here to

Session: Personalized Treatment via Outcome Modeling [Return to Session]

Applications of a CT-Mesh Hybrid Computational Phantom for Patient-Specific Risk Prediction of Radiation-Induced Second Cancer in Proton Therapy

E Kollitz1*, M Roew1, H Han2, M Pinto1, F Kamp3, C Kim2, M Schwarz4, C Belka5, W Newhauser6, G Dedes1, K Parodi1, (1) Ludwig Maximilian University of Munich, Garching B. Munich, DE, (2) Hanyang University, Seoul, KR, (3) University Hospital of Cologne, Cologne, DE, (4) University of Washington, Seattle, WA, (5) Hospital of the Ludwig Maximilian University of Munich, Munich, DE, (6) Louisiana State University, Baton Rouge, LA


TH-D-BRC-2 (Thursday, 7/14/2022) 11:00 AM - 12:00 PM [Eastern Time (GMT-4)]

Ballroom C

Purpose: Out-of-field dose can increase the risk of second cancer, but unknown anatomy outside the patient CT hinders patient-specific treatment considerations. There is evidence that CT-mesh hybrid phantoms, which combine partial-body CTs with mesh-type phantoms to approximate out-of-field anatomy, can better estimate organ dose relative to reference phantoms alone. However, whether this translates to a meaningful improvement in secondary cancer risk prediction is uncertain.

Methods: Head and neck pencil beam proton treatment plans were created for seven patients and simulated on the corresponding hybrids, scaled mesh-type reference computational phantoms (MRCPs), and ground truth whole-body CTs (WBCT). Equivalent dose throughout the body was calculated and input into 5 secondary cancer risk models for excess absolute and excess relative risk (EAR and ERR). The hybrid phantom was evaluated by comparing equivalent dose and risk predictions against the ground truth WBCT.

Results: The hybrid most frequently provides whole-body secondary cancer risk predictions closer to the ground truth when compared to a scaled MRCP. The relative ranking of the hybrid compared to the scaled MRCP was consistent across ERR, EAR, and all risk models. For all in-field organs, as might be expected, the hybrid was better than or equal to the scaled MRCP regarding both equivalent dose and risk prediction. In out-of-field organs, the hybrid’s equivalent dose prediction was superior to the scaled MRCP in 48% of cases, equivalent in 34% of cases, and inferior in 18% of cases. For risk assessment, the hybrid’s prediction was superior to the scaled MRCP in 51.8% of cases, equivalent in 28.6% of cases, and inferior in 19.6% of cases.

Conclusion: Despite not always resulting in the closest prediction, the whole-body risk predictions from the hybrid are an improvement over reference phantoms and could aid in risk-optimized treatment planning and individual risk assessment to minimize secondary cancer incidence.

Funding Support, Disclosures, and Conflict of Interest: This work was funded by the Deutsche Forschungsgemeinschaft as part of Research Training Group GRK 2274.


Protons, Phantoms, Risk


TH- External Beam- Particle/high LET therapy: Proton therapy – out of field dosimetry/risk analysis

Contact Email