Satisfied quantitative value can be acquired by short-time bone SPECT/CT using a whole-body cadmium–zinc–telluride gamma camera

The aim of this study was to evaluate the quantitative values of short-time scan (STS) of metastatic lesions compared with a standard scan (SS) when acquired by whole-body bone SPECT/CT with cadmium–zinc–telluride (CZT) detectors. We retrospectively reviewed 13 patients with bone metastases from prostate cancer, who underwent SPECT/CT performed on whole-body CZT gamma cameras. STSs were obtained using 75, 50, 25, 10, and 5% of the list-mode data for SS, respectively. Regions of interest (ROIs) were set on the increased uptake areas diagnosed as metastases. Intraclass correlation coefficients (ICCs) of standardized uptake values (SUVs) for the ROIs were calculated between the SS and each STS, and ICC ≥ 0.8 was set as a perfect correlation. Moreover, the repeatability coefficient (RC) was calculated, and RC ≤ 20% was defined as acceptable. A total of 152 metastatic lesions were included in the analysis. The ICCs between the SS vs. 75%-STS, 50%-STS, 25%-STS, 10%-STS, and 5%-STS were 0.999, 0.997, 0.994, 0.983, and 0.955, respectively. The RCs of the SS vs. 75%-STS, 50%-STS, 25%-STS, 10%-STS, and 5%-STS were 7.9, 12.4, 19.8, 30.8, and 41.3%, respectively. When evaluating the quality of CZT bone SPECT/CT acquired by a standard protocol, 25%-STS may provide adequate quantitative values.


Scientific Reports
| (2021) 11:24320 | https://doi.org/10.1038/s41598-021-03853-0 www.nature.com/scientificreports/ evaluating the treatment effects of 223 Ra-chloride, which is used as radionuclide therapy of bone metastases from prostate cancer 12 . It is also utilized for benign lesions, such as in the evaluation of the clinical stage of antiresorptive agent-related osteonecrosis of the jaw 13 , or in the evaluation of osteoblastic activity in the epiphyseal growth plates of children 14 . Although instability or vulnerability may be associated with SUVs for SPECT in comparison with those for PET, it has been found to be acceptable in most conditions after phantom-base analysis for quality assessment 9,15,16 , or in clinical-based assessments such as test-retest repeatability 17 . Therefore, images acquired in a shorter time might be acceptable for clinical use when the variance of the SUV is within a specified allowance. The present study aimed to evaluate the clinical feasibility of short acquisition times for the bone SPECT/ CT using whole-body CZT detectors.

Methods
Patients. This retrospective study was conducted in accordance with the guidelines of the Declaration of Helsinki. All experimental protocols in the present study were approved by Institutional Review Board of Saitama Medical University Hospital (No. 20158.01), and the need for written informed consent was waived due to the retrospective nature of the study. From the male patients with bone metastases from prostate cancer who underwent bone scans at our institution between December 2019 and December 2020, we selected 13 patients who met the following criteria: SPECT/CT was acquired using a fixed protocol as shown in the next subsection, and at least one metastatic lesion was confirmed based on the consensus of two experienced nuclear medicine physicians. We estimated 39 lesions to be the minimum required for an adequate analysis of intraclass correlation coefficient (ICC), based on the table from Shoukri et al. 18 , with the parameters of n = 2, α = 0.05, β = 0.20, ρ 0 = 0.6, and ρ 1 = 0.8, where n was observations per subject, α was the probability of type-I error, β was the probability of type-II error, ρ 0 was the ICC when the null hypothesis was true, and ρ 1 is the ICC when the alternative hypothesis was true.
Bone SPECT/CT protocol: standard scan. Initially, the patients received intravenous injections of 99m Tc-methylene diphosphonate ( 99m Tc-MDP). Although 740 MBq was set as the standard target dose, we measured the dose in the syringe before and after administration, and calculated the precise dose injected from these values. Approximately 3 h after injecting the tracer, the patients were asked to void, after which whole-body bone and the associated SPECT/CT images were acquired using a Discovery NM/CT 670 CZT scanner (GE Healthcare, Chicago, IL, USA).
SPECT images for the standard scan (SS) were acquired using the following parameters: a total of 60 projections of 20 s each over 360° in a non-circular orbit, step-and-shoot acquisition by dual-head CZT detectors, high-energy high-resolution collimator, and energy window of 140.5 keV ± 7.5%. It indicated that 10 min was the standard scan time per table position when we did not consider time for gamma-camera rotation. All SPECT images were reconstructed using the ordered subset expectation maximization method, with iteration 4 and subset 10, the matrix size was 128 × 128, and the voxel size was 4.42 × 4.42 × 4.42 mm. CT images were acquired using the following parameters: 120 kV and auto mA (noise index 35) using an ASiR reconstruction system (GE Healthcare), 512 × 512 matrix, 1.375 pitch, and 0.5 s rotation. Although this SPECT/CT system could acquire SPECT data during the rotation of the gamma camera, named as SwiftScan (GE Healthcare), these data were not used in the present study to clarify the analysis.
Bone SPECT/CT protocol: short-time scan. The SPECT/CT image data were acquired using list-mode, and the images of STSs were reconstructed using 75, 50, 25, 10, and 5% data of SS. The 5%-STS image indicates the shortest acquisition for the scanner, indicating 1 s per step of the gamma cameras.

Placement of region of interest.
We used a workstation with Xeleris v9.0 (GE Healthcare) for region of interest (ROI) placement and SUV calculation. Based on the SS images, each 3-dimensional (3D) ROI was placed to cover the increased uptake interpreted as bone metastasis by the consensus of two board-certified nuclear medicine physicians. The maximum SUV (SUVmax) was calculated for each ROI. The reference ROI, which was healthy bone, was then set on the proximal area of the femur. To choose the reference area, the ROI was set to avoid metastatic lesions.

Statistics. First, the variance of SUVmax between the SS and each STS was evaluated based on the ICC.
Statistical analyses were performed using SPSS 27 (IBM, Armonk, NY, USA). For the calculation of the ICCs, the two-way random model, absolute agreement type, and single measurement data were used in SPSS 27. An ICC ≥ 0.8 was considered as an almost perfect correlation 19 . Second, we drew Bland-Altman plots between the SS and each STS. The repeatability coefficient (RC), reflecting 95% limits of repeatability for the relative difference between the two measurements on the Bland-Altman plot, was calculated to be 1.96 times the standard deviation of relative differences. For the evaluation of RC, ≤ 20% was defined as acceptable for the variation based on the results of previous studies concerning the test-retest repeatability of SUVmax on PET or SPECT images [20][21][22] . Third, we calculated the rate at which SUVmax changed by ≥ 5% and ≥ 10% for each STS compared with the SS. Finally, we calculated the contrast-to-noise ratio (CNR) using the following formula: The CNR of the SS was statistically compared with that of each STS using the Wilcoxon signed-rank test, and a p-value of < 0.05 was considered statistically significant.  Table 1, and maximum intensity projection SPECT images of a representative case are shown in Fig. 1.

Agreement analysis.
The ICC and RC results are summarized in Table 2. The ICC was 0.955, even in 5%-STS, and all STS images were found to have almost perfect agreement. The Bland-Altman plots are shown in Fig. 2. The RCs between the SS and 75%-STS, 50%-STS, and 25%-STS were < 20% and were therefore found to be acceptable, whereas those between the SS and 10%-STS and 5%-STS were > 20%, and were therefore found to be non-acceptable. The SUVmax changes are shown in Fig. 3. The rates of SUVmax changes ≥ 10% were 1.3% at 75%-STS, 11.8% at 50%-STS, 25.0% at 25%-STS, 59.7% at 10%-STS, and 67.1% at 5%-STS.

Discussion
To evaluate the clinical applicability of the quantitative values acquired from STS SPECT/CT images, we utilized list-mode data to simulate short-time acquisition of whole-body SPECT/CT bone scans performed in patients with bone metastases from prostate cancer. The resulting ICCs between the SS and each STS showed almost perfect agreement, even between the SS and 5%-STS (1/20th scan time). ICC is one of the most frequently used indicators to evaluate reproducibility. Although there is no specific cutoff, an ICC of ≥ 0.8 19 or ≥ 0.75 23 is generally considered to be an almost perfect correlation, based on the definition of the kappa agreement value. From this point of view, an ICC of 0.955, even between SS and 5%-STS, showed extremely high reproducibility. We also evaluated RC, another absolute indicator for reproducibility, to reinforce the statistical result because some studies recommended using both relative and absolute indicators 24,25 . Although no optimal cutoff was identified on the RC, we set 20% as the cutoff for the present study, based on previous studies that evaluated the reliabilities of SUVs [20][21][22] . As a result, we found that 25%-STS may be reliable at providing quantitative values. In the 25%-STS images, the SUV changed by ≥ 10% in 25.0% of the lesions. Based on the results of the previous studies regarding repeatability, changes in the SUV may be acceptable at ≤ 10%. As  www.nature.com/scientificreports/ the incidence of the SUV change dramatically increases in the 10%-STS, the 25%-STS may be the shortest with which to maintain the reliability of quantitative values. Contrarily, background noise gradually increased as the scan time decreased. Statistically significant differences were observed in the CNR between the SS and STS, even in 75%-STS, which may lead to false-positive results in the interpretation of SPECT/CT bone scans when using a shorter acquisition time. There may be a discrepancy in the results of ICC or RC analysis. Generally, uptake on bone scans in bone metastases is considered quite high 11 compared with that in other types of metastatic or bone disorders 26 . Such a high quantitative value, therefore, may have been unaffected by the decrease in scan time and the subsequent noise on the images. These results are quite similar to those from the STS of a planar bone scan 8 . As a result, we feel that 25%-STS provides appropriate quantitative values for whole-body SPECT/CT scans for bone metastases from prostate cancer. Reducing scan time enables patients with pain to shorten the restrain. Instead of scan time, our results can apply for reducing tracer dose. For patients who need multiple follow-up scans, the benefits of low-dose scans must be significant. However, short acquisition time results in the reduction of image quality. The shorter the image acquisition time was, the more SUVmax changed, and the lower CNR was. On the other hand, more acquisition time, like > 75%, may be required for the initial diagnosis, even in the case of bone metastases from prostate cancer, or other types of bone metastasis or bone disorders that have lower uptakes. Therefore, we should use proper acquisition time or tracer dose for what is essential for the patient and how much the image quality is required. In addition, we should note that the median tracer dose we used in the present study was 740 MBq, which is relatively high when we consider 555 MBq as the standard dose. In that case, more acquisition time may be required for the appropriate quantification. Further studies are needed to evaluate the reliability of STS in different patient groups, including other malignant tumors or bone disorders.
Our proposed methods for short-time acquisition may be helpful to develop dynamic quantitative SPECT/ CT imaging, so far a domain of PET. Tracers with more complex kinetics, such as 99m Tc-sestamibi or even new radiopharmaceuticals, could be evaluated. In addition to 99m Tc labeled tracer, the development of new collimators of wide-energy high-resolution or medium-energy high-resolution can acquire images of other radionuclides, including 123 I or 177 Lu, that can apply for the dosimetry of radionuclide therapy 27 .
One limitation of the present study was that we did not evaluate the diagnostic capabilities of these scans, including sensitivity and specificity. Although we chose metastatic lesions based on the consensus of experienced nuclear medicine doctors, the results were not confirmed by pathological methods. Therefore, we can only discuss the quantitative values as the results of the present study.
In conclusion, regarding bone SPECT/CT images acquired using a standard protocol of 740 MBq tracer dose and 10 min/bed position, we found that 25%-STS might be acceptable for the clinical evaluation of the quantitative values of bone whole-body SPECT/CT images acquired using CZT detectors, specifically for bone metastases from prostate cancer.