Ultra-Low Count Quantitative SPECT for Alpha-Particle Therapies

点击翻译按钮获取中文摘要

PROJECT SUMMARY: The overall goal of this project is to develop and comprehensively validate ultra-low count quantitative SPECT (ULC-QSPECT) methods for alpha particle radiopharmaceutical therapies (αRPTs), including in a first-in-man trial in patients with bone metastatic castration-resistant prostate cancer. αRPTs, such as those based on Actinium-225, Thorium-227 and approved Radium-223 isotopes, are an emerging class of cytotoxic therapies for patients with disseminated metastatic disease using internally administered alpha-particle emitting agents. Despite the great potential of these therapies, current αRPT regimens are not personalized, with administered activity dependent merely on mass, likely leading to non-optimal therapy. To address this challenge, there is a crucial unmet need for methods to measure the isotope uptake, and hence the absorbed radiation dose with these therapies, both at sites of disease and in vital dose-limiting organs. SPECT provides a clinically translatable mechanism to achieve this goal. However, a key challenge to SPECT-based quantification is that the administered activities in αRPTs are orders of magnitude lower than a typical SPECT scan, leading to ultra-low detected count levels. Conventional approaches to quantification that reconstruct the isotope distribution and estimate the regional uptake from reconstructed images are erroneous at these low count levels. To address this issue, we put forwards a novel computational ULC-QSPECT framework for regional activity estimation from αRPTs. These methods quantify regional uptake directly from projection data skipping the reconstruction step, and at the same time, extract the maximal possible information from the acquired projection data. For this purpose, we propose novel methods that accurately model the physics of imaging αRPTs, including stray- radiation-related noise, use data from multiple-energy windows, incorporate scatter-window photons for quantification, and process data in list-mode format. Our extensive preliminary data show that the proposed methods result in nearly unbiased uptake and variances close to the theoretical limit. We propose to further develop and rigorously evaluate these methods. Our proposed evaluations include studies over multiple scanners with different detectors and different collimator configurations. Further, our goal is clinical translation of these methods. Towards this goal, we propose to clinically evaluate these methods for measuring activity concentrations at sites of uptake of [223Ra]RaCl2 in men with castrate resistant prostate cancer. These methods will enable quantification of activity at disease sites in the skeleton as well as clearance through the intestine. The approach has direct relevance to patients as it achieves noninvasive imaging of low-administered activity therapies. Further, this proposal has substantial potential impact to improve both safety and efficacy of drug development efforts in this rapidly evolving space. Further, the methods developed in this proposal will strengthen the clinical utility of SPECT in managing patients with these therapies.