RT Journal Article
SR Electronic
T1 Performance evaluation of a personalized remote radiation tracking vest for at home monitoring of Lu-177 DOTATATE therapy.
JF Journal of Nuclear Medicine
JO J Nucl Med
FD Society of Nuclear Medicine
SP 139
OP 139
VO 61
IS supplement 1
A1 Robert Miyaoka
A1 Larry Pierce
A1 Robert Harrison
A1 Hubert Vesselle
YR 2020
UL http://jnm.snmjournals.org/content/61/supplement_1/139.abstract
AB 139Objectives: The overall goal of this work is to develop wearable, personalized remote radiation tracking (PRRT) technology to enable customization of Lu-177 DOTATATE therapy in a cost-effective and patient-friendly manner. Internal radiation therapies are optimized by treating to the dose limit of the patient’s main organ at risk (OAR). For Lu-177 DOTATATE, the OAR are kidneys, bone marrow, liver and spleen, with kidneys the main OAR in a vast majority of patients. We previously introduced the concept of a PRRT vest for at home radiation monitoring of Lu-177 DOTATATE. In this work, we seek to further characterize the vest’s performance investigating the impact radiation detector geometry and repositioning offsets between daily wearing of the vest have on quantitative estimation of Lu-177 DOTATATE washout kinetics. Methods: The PRRT vest utilizes a sparse arrangement of small scintillator-based detectors. Pinhole collimation is used to shape the cone of response of each detector. The detectors are then strategically placed within a vest to measure radioactivity from OAR (e.g., kidneys, liver, spleen), tumors and background. The patient will wear the PRRT vest for ~2 minutes a day for up to three weeks. Using Monte Carlo simulation and digital phantoms, we investigated how the detectors’ pinhole diameter (2, 3 or 4 mm), the collimator geometry (i.e., air gap between pinhole and GAGG detector), the number of measurements (i.e., over many days), and the daily detector repositioning offsets affect estimation performance. Random daily positioning offsets up to 17 mm from the prescribed detector/vest location were simulated. The testing phantom was a 27 cm diameter right circular cylinder with anthropomorphic objects representing liver, spleen, right and left kidneys and two tumors (2.5 cm and 1.5 cm diameter) placed in the liver. Initial activity concentration and washout rates for liver, spleen, right and left kidneys, 2.5 cm and 1.5 cm tumors, and background were 4, 21, 24, 27, 166, 81.5 and 2.5 µCi/cc, and 69, 72, 50, 54, 88, 93 and 90 hrs, respectively. Tests were run for vests with up to 15 sensors. Activity in each object was decayed appropriately. Poisson sampling was used for accurate noise modeling. Decay rates were estimated using 3-21 measurements over 7-21 days. Results: The average error of the estimated washout for all OAR for all testing conditions was less than 6% when there was no repositioning offsets between daily measurements. Using a 4 mm diameter pinhole collimator and acquiring measurements for 21 days mitigated the effects of random daily repositioning and kept the error in washout estimation less than 5% for all OAR and the largest tumor for a maximum daily repositioning offset of ±8.5 mm and less than 5% for the largest tumor and all OAR, except the spleen, for a maximum daily repositioning offset of ±17 mm. Our expectation is that most individuals will be able to reposition the vest within ±8.5 mm. Dosimetry errors for the smallest tumors was larger than 10% for some test cases; however, the average error for the kidneys (the main OAR) remained less than 3% even under the most challenging testing conditions. Conclusions: Taking measurements for 21 days led to the most accurate estimates of radiotracer washout, especially when random repositioning offsets were included in the study. The detectors with the 4 mm pinhole size provided the best overall performance.