Targeted radiotherapy with internally administered radiopharmaceuticals has recently experienced renewed interest due to identification of better tumor biomarkers and development of new targeting agents. In any form of radiotherapy, the fraction of surviving cells, healthy organ toxicity and tumor response depend primarily on the quantity of energy deposited. The relationship between absorbed dose and malignant tissue response or normal tissue toxicity is the theme of any radiotherapy approach. The following questions will be addressed by this proposal: 1. Can we improve the accuracy of radiation absorbed dose estimate to a specific patient relative to current, simpler phantom-based methods, yet such a method be efficient enough for routine clinical use? 2) Will accounting for the temporal and spatial dose-rate gradient and tissue-response heterogeneity by using known radiobiologic models improve the predictive value of treatment efficacy and toxicity to bone marrow and kidney? Using an ultra fast discrete ordinate method (DOM) with quantitative SPECT/CT voxel-based dose engine developed by the PI with previous NIH support, the following aims are proposed to address these questions: 1) Validate the accuracy of DOM-QSPECT/CT in phantom with measurements and Monte Carlo simulations, 2) Incorporate known radiobiological effect dose (BED) models in the DOM-QSPECT/CT dose estimate and generate BED-volume-histograms for kidney, 3) Compare retrospectively DOM-QSPECT/CT dose estimates to MIRD and Monte Carlo dose estimates for a cohort of patients treated using Sm-153 EDTMP in a phase II clinical trial, and 4) Correlate the clinical outcomes (bone pain index and toxicity level to bone and kidney) to the DOM-QSPECT/CT-BED dose estimates. The proposed work is innovative because it capitalizes on a novel method to perform radiation transport in human tissue, a method that we have demonstrated to be as accurate as Monte Carlo simulations yet far more efficient for complex external beam radiotherapy regimens (i.e., computational results can be obtained within minutes compared to many hours or days for Monte Carlo). In addition, we propose the incorporation of radiobiological modeling to assess the treatment plan using the biological effective dose concept. With respect to expected outcomes, the combination of the work proposed in the four specific aims is expected to create a new platform for radiopharmaceutical dosimetry that will have a positive impact on the therapy field of nuclear medicine, in a way that will fundamentally advance the field of patient-specific internal radionuclide therapy treatment planning.
Targeted radiotherapy with internally administered radiopharmaceuticals has recently experienced renewed interest due to identification of better tumor biomarkers and development of new targeting agents. Our long term goal is to develop targeted radionuclide therapy planning tools comparable to those used in radiation oncology practice (e.g., for external beam and brachytherapy), in order to aid in trial design and tumor response and toxicity prediction, as well as replace the current, inferior planar image/MIRD-based dosimetry methodology. We developed a deterministic radiation transport method that we plan to validate and test in a cohort of patients from a phase II clinical trial of a skeletal targeted therapy radiopharmaceutical in Breast Cancer Patients with Bone Only Metastases.