Quantitation of Lu¹⁷⁷ using General Purpose Whole-body Solid State CZT and Standard SPECT/CT Cameras

Objectives: Parameters for accurate quantitation of lutetium-177 (Lu¹⁷⁷) whole body SPECT studies were determined for a solid state CZT (cadmium zinc telluride) camera (CZTC) and a sodium iodide (NaI) based device (NaIC).

Methods: A NEMA IEC phantom emulating human torso dimensions was loaded with a “lung” insert and 6 hot spheres (10 to 37 mm diameter) with a 12:1 target-to-background ratio of Lu¹⁷⁷-PSMA and background concentrations ranging from 15 to 110 kBq/mL. The phantom was imaged on two SPECT/CT cameras which differed primarily in detector design: CZT vs. NaI detectors. For CZTC a rectangular hole low energy high resolution (LEHR) collimator was used for 4 acquisition protocols: (1) LE113TEW - low energy mode (LE), using the 113 keV peak and triple energy window scatter correction (TEW); (2) HE113208TEW - high energy mode (HE), using the 113 and 208 keV peaks with TEW; (3) HE208TEW - HE with the 208 keV peak and TEW; (4) HE208DEW - HE with the 208 keV peak and dual energy window scatter correction (DEW). For NaIC the 4 acquisition protocols were: (1) LEHR113TEW using a LEHR collimator and 113 keV peak with TEW; (2) MEGP113208TEW - using a medium energy general purpose collimator (MEGP), 113 and 208 peaks, and TEW; (3) MEGP208TEW - MEGP with the 208 keV peak and TEW; (4) MEGP208DEW - MEGP with the 208 keV peak and DEW. Acquisition durations were 30 minutes. The assessed parameters were: count rate linearity, contrast, peak standardized uptake values (SUV_peak), mean squared error (MSE) of recovery coefficients (Wilcoxon test, p<0.05), residual error of the lung insert, and background quantitation error (Mann-Whitney test, p<0.05).

Results: LE113TEW results on CZTC compared to LEHR113TEW on NaIC were not significantly different for contrast (3 largest spheres: 100.1 ± 25.8 % vs. 93.3 ± 22.3 %, respectively; mean ± std. dev.), SUV_peak (3 largest spheres: 11.5 ± 2.7 g/mL vs. 11.8 ± 2.6 g/mL), MSE (0.26 ± 0.24 vs. 0.27 ± 0.25), lung insert residual error (58.5 ± 9.7 % vs. 53.4 ± 5.3 %), and background quantitation error (─5.0 ± 3.0 % vs. ─1.7 ± 2.4 %). MSE of LE113TEW did not differ significantly from the best NaIC method (MEGP208DEW: 0.23 ± 0.25). Background quantitation error was not significantly different among all methods for both cameras (< 9%). MEGP208DEW gave a lower lung residual error (29.5 ± 8.8 %) than LEHR113TEW or any CZTC method, but not significantly different from MEGP113208TEW (42.6 ± 7.1 %) or MEGP208TEW (30.9 ± 5.8 %). All count rates were linear with the incident count rate (R² > 0.9996). Dead time loss was <3.2 % for MEGP208DEW on NaIC while LE113TEW on CZTC showed no dead time loss.

Conclusion: Overall, Lu¹⁷⁷ quantitation was equivalent on general purpose whole-body solid state CZT and standard SPECT/CT cameras. The high energy mode on CZTC and the double peak acquisitions on either camera did not improve significantly the quantitative results. MEGP collimation on NaIC provided less artefactual uptake in the “lung” insert than obtainable with LEHR on CZTC. CZTC showed no measurable dead time loss during Lu¹⁷⁷ imaging. Research Support: N/A