Abstract
TS27
Introduction: 177Lu-PSMA therapy for metastatic castration-resistant prostate cancer is currently approved without the use of post-treatment imaging for localization and/or dosimetry (1). However, post-treatment imaging can be a useful diagnostic tool for evaluating injection quality and lesion assessment, and dosimetry imaging ensures that critical organ dose is minimized and below the threshold levels for adverse effects as patients receive repeat courses of treatment (2). The future of dosimetry imaging is promising as 177Lu-PSMA therapy becomes less of a last resort treatment option (3). Further, 177Lu SPECT/CT imaging is preferred over planar imaging for dosimetry calculation as SPECT/CT protocols possess superior accuracy of activity quantification (2,4). Prior to starting routine post-treatment imaging, facilities must establish imaging protocols that provide quality information. Because SPECT/CT imaging can be a time-consuming process, balancing patient comfort and workflow with image spatial resolution is essential. This research project aims to determine if it is possible to reduce SPECT/CT acquisition time for 177Lu imaging without degrading spatial resolution.
Methods: A National Electrical Manufacturers Association (NEMA) body phantom with the lung insert removed was loaded per the NEMA User’s Manual (5). The cavity was filled with 20.5 mCi 177Lu-PSMA in 10.5 L, and the spheres filled with a stock solution of 1.99 mCi of the tracer in 100 mL of water, resulting in a roughly 10:1 activity concentration between the spheres and cavity. The size of each sphere and its filled activity was recorded (Fig. 1). A 50 mL syringe was assayed in a Biodex Atomlab 500 dose calibrator before and after filling the spheres and cavity to determine their activity.The phantom was placed on a GE NM/CT 870 dual head CZT gamma camera equipped with Medium Energy High Resolution & Sensitivity (MEHRS) collimators and set at 174- and 208-keV with a 128x128 matrix and undergoing a SPECT/CT acquisition at various seconds per stop. A step-and-shoot technique was used to acquire a 360° rotation stopping every 3° for a total of 60 stops or 120 views. Acquisitions were made at 18, 15, 12, 10, 8, and 5 seconds per stop, and were reconstructed using OSEM iterative reconstruction. The facility’s standard of care imaging protocol for 177Lu-PSMA is 15 seconds per stop, using a 2-field of view SPECT/CT to image the chest, abdomen, and pelvis.Transverse slices displaying the spheres were compiled to evaluate the various acquisition times (Fig. 2). Each image was created using the same range of slices through the phantom. The images were evaluated semi-quantitatively to provide an assessment of spatial resolution (6).
Results: Upon visual inspection by the technologist, the smallest two of the six spheres (10 mm and 13 mm) were not resolved using any of the tested acquisition techniques. The largest two spheres (37 mm and 28 mm) were resolved in all six techniques. The 22 mm and 17 mm spheres were resolved in all images, but resolution was limited. There appeared to be less noise in the images with increasing stop time. There does not appear to be a discernable difference in spatial resolution when visually assessing the spheres between the different acquisition techniques.
Conclusions: Spatial resolution is a considerable factor when opting for shorter acquisition times that are easier on the patient and enhance workflow. While it appears that there is minimal visual difference between the varying acquisition times, the decrease in noise, size of the discernible spheres, and potential variation of dosimetry results are significant considerations when assessing time per stop.To further this research, evaluation of these reconstructions using dosimetry software is necessary to quantify the lowest time per stop possible without compromising quantitative dosimetry analysis. Additionally, dosimetry data can facilitate SPECT/CT spatial resolution research on patients receiving 177Lu-PSMA therapy imaging.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.