2140 - Personalized Simulation of Radiation-Induced Immune Suppression for Head-and-neck Cancer across Various Radiotherapy Regimens
Presenter(s)
J. Li1,2, R. D. Mali1, G. N. Gan1, C. E. Lominska3, K. Guida3, A. Juloori4, M. W. R. Chen1, W. Li1, J. Setianegara1, C. Wang1, Y. Lin1, Q. Li5, W. Chen2, and H. Gao1; 1Department of Radiation Oncology, University of Kansas Medical Center, Kansas City, KS, 2Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, Gansu, China, 3Department of Radiation Oncology, The University of Kansas Medical Center, Kansas City, KS, 4Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL, 5Institute of Modern Physics, Chinese Academy of Sciences, Lanzhou, China
Purpose/Objective(s): Radiation-induced lymphopenia (RIL) is a critical form of immunosuppression associated with survival outcomes in radiotherapy and radioimmunotherapy. This study aims to develop a patient-specific modeling approach to simulate lymphocyte dynamics during and after radiotherapy and to evaluate the lymphocyte-sparing effects of different RT treatment regimens.
Materials/Methods: Seventeen HNC patients who received proton or photon therapy were included. The Hematological Dose (HEDOS) model was utilized to estimate radiation exposure to circulating lymphocytes, incorporating auto-segmented carotid arteries and jugular veins. To capture the lymphocyte kinetics, a personalized mathematical model was developed to represent lymphocyte depletion and recovery, considering radiation effects on both circulating lymphocytes and lymph nodes. Patient-specific parameters were derived by fitting the model to weekly absolute lymphocyte counts (ALC) collected before, during, and after RT. The impact of four RT regimens—conventional fractionation (CONV), hypofractionation (HYPO), stereotactic body radiotherapy (SBRT), and FLASH—was analyzed in terms of lymphocyte preservation.
Results: Proton therapy led to a 17.1% reduction in severe (grade 3–4) RIL compared to photons. The mean radiation dose to circulating lymphocytes was lower for protons (0.73 ± 0.23 GyE) than for photons (1.17 ± 0.27 Gy). The patient-specific model effectively captured three distinct ALC patterns: plateau phase, normal recovery, and incomplete recovery, with a mean squared error (MSE) of 0.026 ± 0.024 between predicted and observed values. In FLASH RT, fewer than 20% of circulating lymphocytes received doses above 0.1 Gy. Three months post-RT, ALC levels were 10.1%, 26.3%, and 36.7% higher for HYPO, SBRT, and FLASH, respectively, compared to CONV. By one year post-RT, most patients demonstrated at least 70% ALC recovery across all regimens.
Conclusion: A personalized modeling approach was developed to characterize lymphocyte kinetics during and after RT for HNC patients. This framework enables comparison of different RT regimens in minimizing lymphocyte depletion and aiding immune recovery.
Abstract 2140 - Table 1| Metrics | CONV | HYPO | SBRT | FLASH |
| Lymphocyte mean dose (Gy) | 0.14±0.05 | 0.26±0.09 | 0.58±0.19 | 0.58±0.19 |
| Lymphocyte max dose (Gy) | 0.84±0.11 | 1.14±0.25 | 1.84±0.49 | 8.86±7.91 |
| Lymphocyte volume received > 0.1Gy (%) | 56.44±19.67 | 87.59±17.00 | 99.54±1.65 | 16.90±2.90 |
| ALC at nadir (% of baseline) | 35.31±15.49 | 41.61±14.50 | 59.41±12.19 | 71.51±10.01 |
| ALC after 3 months after RT (% of baseline) | 47.16±16.82 | 57.19±13.38 | 73.45±10.54 | 83.84±7.91 |
| ALC at 1 year after RT (% of baseline) | 87.32±13.41 | 91.19±9.51 | 95.28±5.55 | 97.48±3.33 |