3132 - Pretherapy Radiobiological Dosimetry Using Self-Calibrating Quantitative Hybrid Imaging for Metastatic Thyroid Cancer Treatment with <sup>131</sup>I
Presenter(s)
J. Shin1, Y. Seo2, Y. Li2, D. Diwanji2, R. Flavell2, Y. S. Yeom3, S. Morato Rafet1, C. Kitahara1, and C. Lee4; 1National Cancer Institute/National Institutes of Health, Rockville, MD, 2University of California, San Francisco, San Francisco, CA, 3Department of Radiation Convergence Engineering, Yonsei University, Wonju, Gangwon-do, Korea, Republic of (South), 4National Cancer Institute, National Institutes of Health, Rockville, MD
Purpose/Objective(s): The lack of a standardized, evidence-based dose-response model in radiopharmaceutical therapy has hindered the implementation of radiobiological modeling into patient-specific dosimetry, despite its potential to guide activity administration. Meanwhile, biological responses, which are influenced not only by mean absorbed dose but also by its spatial and temporal distribution, necessitate radiobiological dosimetry that can reconstruct time-dependent dose rate distribution within a patient body. This study aimed to develop patient-specific radiobiological dosimetry using widely available pretherapy imaging and to investigate the effects of varying doses and dose rates on biological responses for metastatic thyroid cancer treatment using 131I.
Materials/Methods: A 4D particle source was modeled using hybrid SPECT/CT and three planar images, then simulated with an in-house planning dosimetry tool to generate a spatiotemporal dose distribution stored in a sparse matrix. Dose metrics for radiobiological effect modeling, such as the Lea-Catcheside factor, Biological Effective Dose (BED), and Equivalent Uniform BED (EUBED), were reconstructed from the spatiotemporal dose distribution as a function of 131I therapeutic administration. Lesion-specific radiobiological dose metrics were compared with physical absorbed dose (AD) to evaluate the impact of protracted dose rate and spatially heterogeneous dose distribution on treatment effectiveness.
Results: Radiobiological dosimetry was performed for a total of 15 lesions from five metastatic thyroid cancer patients. Lesion volumes ranged from 3.4 cm3 to 163.3 cm3. Lesion ADs ranged from 7.5 mGy/MBq to 477.2 mGy/MBq. Lesion dose heterogeneity, defined as the ratio of D10% to D90%, ranged from 1.8 to 12.5. EUBEDs were always lower than the volumetric mean AD, with ratios to AD ranging from 0.3 to 0.87 indicating the treatment effectiveness was greatly degraded compared to AD due to intratumoral dose heterogeneity. Finally, the treatment effectiveness exhibited strong negative correlations with mean pretherapy AD (-0.93, p<0.001) and dose heterogeneity (-0.83, p<0.001), but showed no significant correlation with lesion volume (-0.2, p=0.65). In addition, treatment effectiveness varied across the patients and among the lesions within each patient, with the largest variation observed as 0.17 to 0.9 and the smallest variation as 0.64 to 0.98.
Conclusion: A new radiobiological dosimetry method using self-calibrating quantitative hybrid imaging has enabled the modeling of biological responses influenced by spatially and temporally varying absorbed doses. Our analysis revealed that a protracted dose rate increased dose heterogeneity, resulting in reduced treatment effectiveness. Because the proposed radiobiological dosimetry method was developed with a cost-effective imaging technique, radiobiological modeling can be implemented as part of personalized dosimetry in most nuclear clinics.