Introduction
Positron emission tomography/computed tomography (PET/CT) is becoming a standard component of the diagnosis and staging in the field of oncology. Especially, 18flourine-fluorodeoxyglucose (18F-FDG) PET/CT is used to quantify radiopharmaceutical uptake and quantitatively determine treatment response in the evaluation of the metabolic response in various tumors (1,2,3). PET images are analyzed in clinical practice either qualitatively using visual comparison of metabolism in lesions with normal tissue or semi-quantitatively using standard uptake values (SUV). SUV is obtained as the tissue concentration (MBq/mL) divided by the injected activity (MBq/g) per body weight (BW). Factors affecting SUV have been discussed in many studies (4,5,6). Since fat contributes to BW but accumulates very little 18F-FDG in a starvation state, SUV is relatively increased in patients who are obese than in thinner ones. A study found that lean body mass (LBM) SUV (SUL) correction is a more suitable quantitative method than BW or body surface area for patients who are obese (7).
Prostate-specific membrane antigen (PSMA) is a highly expressed human transmembrane protein that is low or moderate in normal or hyperplastic prostate tissues and high in primary adenocarcinomas and distinguishes malignant lesions from benign lesions with high accuracy and positively correlates with the degree of expression, tumor aggression, metastatic disease, and recurrence (8,9,10,11). In the literature, studies have shown that the SUV values obtained from a prostate tumor are higher as the Gleason score (GS) and prostate-specific antigen (PSA) value increase. Gafita et al. (12) investigated whether SUL is a more appropriate quantitative method than SUV, which is normalized by BW in gallium-68 (68Ga)-PSMA 11 PET/CT. They found that correction with lean BW disrupts positive correlations between absolute SUV and BW and that SUL may be preferred over SUV for quantitative analysis in 68Ga-PSMA 11 PET (12).
This study aimed to investigate the relationship between the SUV and SUL parameters obtained from the prostate tumor according to Gleason grade (GG) groups, D’Amico risk groups, and presence of metastasis in 68Ga-PSMA PET/CT and to determine whether SUL is superior to SUV.
Materials and Methods
Patients
Patients with prostate adenocarcinoma who underwent 68Ga-PSMA PET/CT for staging at our center between February 2017 and October 2018 were evaluated retrospectively. GG groups of patients were obtained from prostatectomy material in patients undergoing prostatectomy and fine-needle biopsy results in other patients. D’Amico criteria was considered for risk stratification [low risk group (PSA <10 ng/mL and GS <7 and T1-T2a), intermediate-risk group (PSA 10-20 ng/mL or GS 7 or T2b-T2c), and high-risk group (PSA >20 ng/mL or GS 8-10 or T3-T4)] (13). Patients were divided into two groups according to their GG (≥3 vs <3).
PET Image Analysis
Patients signed the informed consent form, and radiation safety and imaging protocol were described. An average of 3.2 millicurie (mCi) 68Ga-PSMA Imaging and Therapy was injected intravenously. Low-dose CT was used for attenuation correction an hour after injection. PET images were obtained for 1.5 min in each bed position in the supine position from the vertex to the toe tip in Philips Gemini TF PET/CT (Eindhoven, Netherlands). Row action maximum likelihood algorithm was used for reconstruction.
Patients who had PSMA expression on 68Ga-PSMA PET/CT images, which could be differentiated from background activity and thought to be related to prostate adenocarcinoma metastasis (PSMA-RADs 4 and 5) were considered to have metastatic disease (14).
Weights and heights of the patients were measured before imaging. LBM was calculated with the formula developed by Janmahasatian (15,16).
LBM = (9.27×10³×BW)/(6.68×10³+216×BMI)
In the prostate gland, a region of interest was drawn on the area where PSMA expression was observed above background activity. Maximum SUV (SUVmax), SUVmean, and SUVpeak values were obtained from this area. SULmax, SULmean, and SULpeak values were calculated from SUVmax, SUVmean, and SUVpeak values using the LBM value obtained from the Janmahasatian formula.
SUL = SUV × LBM/BW
In addition, a 3-cm volume of interest (VOI) was drawn to the right liver lobe to determine liver background activity. Liver SUVmean and liver SULmean values were obtained from this VOI (17).
Statistical Analysis
The free version of the Statistical Package for the Social Sciences v. 26.0 was used for statistical analysis. The correlations among liver SUVmean, liver SULmean, and BW were evaluated by Spearman correlation analysis. A p<0.05 value was considered significant.
The difference in SUVmax, SULmax, SUVmean, SULmean, SUVpeak, and SULpeak values from the prostate tumor between low-moderate and high-risk groups was analyzed with the Mann-Whitney U test. In addition, the difference in the SUV and SUL values between GG groups, between PSA groups (≥10 and <10 ng/mL), and between D’Amico risk groups (low-moderate and high) was evaluated with the Mann-Whitney U test.
The potential of SUV and SUL parameters in distinguishing GG groups and risk groups was evaluated by the receiver operating characteristics analysis. The area under the curve (AUC) values were compared, and significant difference between them was evaluated.
Ethics Approval
Dokuz Eylül University Ethics Committee approval was obtained (decision no: 2020/18-37, date: 10.08.2020). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. All patients gave their informed consent before their inclusion in the study.
Results
The study included 79 patients with a mean age of 65±7 (range, 44-78) years and mean BW of 81.5±14.5 (range, 53-125) kg. The median PSA value was 16 (range, 0.02-527.00) ng/mL. A total of 13 patients had a history of radical prostatectomy.
In this study, 7 (9%) patients had GG 1, 21 (27%) had GG 2, 12 (15%) had GG 3, 15 (19%) had GG 4, and 24 (30%) had GG 5. In 68Ga-PSMA PET/CT, 41 (52%) patients did not have metastasis, while 38 patients had PSMA expression suggesting metastasis. Moreover, 33 (41.8%) patients had lymph node metastasis. Five of these patients had PSMA expression in cervical/mediastinal lymph nodes, 16 in abdominal lymph nodes, and 33 in pelvic lymph nodes. In addition, 19 (24.1%) patients had PSMA expression suggesting bone metastasis, whereas 2 (2.5%) patients had PSMA expression suggesting pulmonary metastasis. In the study group, no patient had liver metastasis. According to the D’Amico risk groups, two patients had low risk, 18 (22.8%) patients had moderate risk, and 59 (74.7%) patients had high-risk.
The mean ± standard deviation and median (range) values from the prostate tumor were calculated as follows: SUVmax, 11.5±9.3 and 8.0 (3.0-49.8); SULmax value, 8.4±6.1 and 6.1 (2.2-34.4); SUVpeak value, 8.6±7.2 and 5.8 (2.4-37.6); SULpeak value, 6.3±4.5 and 4.5 (1.7-25.9); SUVmean value, 5.0±2.6 and 4.2 (1.9-13.8); and SULmean value, 3.7±3.3 and 3.2 (1.3-9.4).
A significant difference was found in the prostate SUVmax, SULmax, SUVpeak, SULpeak, SUVmean, and SULmean values between patients with GG ≥3 and <3 (Table 1). However, when comparing AUC values of SUV and SUL parameters in distinguishing GG groups, no SUV/SUL parameters were superior to the other (Table 2).
No significant difference was observed in any SUV and SUL parameters between patients with GS 3+4 and 4+3.
Prostate SUVmax, SULmax, SUVpeak, SULpeak, SUVmean, and SULmean values were significantly higher in the high-risk group than in the other D’Amico risk groups (Table 1). However, no SUV/SUL parameter was superior to others in distinguishing risk groups (Table 2).
While all SUV and SUL parameters were higher in patients with a PSA value ≥10 ng/mL than in those with <10 ng/mL (Table 1), no significant difference was found in the discrimination power of any SUV and SUL parameters (Table 2).
No significant difference was found in any SUV and SUL parameters in patient groups with and without metastasis.
The mean liver SUVmean value was calculated as 4.0±1.1 (1.8-7.6), and the liver SULmean value was 3.0±1.4 (1.4-5.5). The liver SUVmean and SULmean values did not correlate with BW (p=0.387 and 0.132, respectively).
Discussion
In this study, a significant correlation was found with the BW of neither SUVmean nor SULmean obtained from the liver. Gafita et al. (12) reported that SUL can be preferred over SUV in 68Ga-PSMA 11 PET/CT. In this study, while the liver SUVmean value showed a significant correlation with BW, no significant correlation was found in the SULmean value. Despite the few studies on this subject, similar to our study, Li et al. (2) evaluated the 18F-DCFPyL uptake but could not detect a significant correlation between liver SUVmean and SULmean values and BW; as a result, they suggested using the SUV.
Several studies have also shown that PSMA SUV data are successful in differentiating GG groups and risk groups (18,19,20,21). Likewise, we were able to obtain significant differences in SUV parameters between these groups. However, our study is the first to evaluate the relationship between the success of SUL and SUV parameters obtained from 68Ga-PSMA PET in differentiating GG groups and risk groups. In our study, SUL parameters obtained from 68Ga-PSMA PET were not superior to SUV parameters in distinguishing GG groups and in distinguishing risk groups. Moreover, the studies that quantified uptake with body-weight-corrected SUV and LBM-corrected SUV have shown that the repeatability coefficient of SULmax and SUVmax within the same patient in a test-retest setting is comparable (22,23,24). These results show that SUL and SUV parameters are not superior to each other in 68Ga-PSMA PET in clinical practice.
Prostate SUV and SUL parameters were significantly higher in patients with GG ≥3 than in GG ≤2, and this is similar to the findings in the literature. Onal et al. (25) reported that SUVmax values obtained from primary tumors in 191 patients were significantly higher in patients with GS >7.
In their study, Uprimny et al. (18) did not find a significant difference in SUVmax values between GG 2 and 3 tumors. Similarly, Ergül et al. (19) did not found a significant difference in SUVmax values between grade 2 and 3 tumors. In addition, in our study, neither SUVmax values nor other SUV and SUL values were detected differently in tumors with GG 2 and 3.
In a previous study, Ergül et al. (19) analyzed 78 patients and found a significant difference in SUVmax values of prostate tumors with and without metastasis. However, Liu et al. (20) did not observe a significant difference in SUVmax in patients with and without metastasis. In the present study, similar to the study of Liu et al. (20), no significant difference was found in any SUV and SUL parameters between these groups.
Study Limitations
First, the retrospective design limits the generalizability of the results. Second, histopathological correlation is not technically and ethically possible from all foci considered metastasis. Finally, prostatectomy could not be applied to all patients, and the GS of some patients could only be obtained from the biopsy sample.
Conclusion
In this study, the superiority of SUL values obtained from 68Ga-PSMA PET to SUV was not determined. We think that both SUV and SUL parameters can be used for quantitative analysis in 68Ga-PSMA PET.
Ethics
Ethics Committee Approval: Dokuz Eylül University Ethics Committee approval was obtained (decision no: 2020/18-37, date: 10.08.2020).
Informed Consent: Retrospective cross sectional study.
Peer-review: Externally peer-reviewed.
Authorship Contributions
Surgical and Medical Practices: A.A., Concept: A.A., G.Ç.K., Design: A.A., G.Ç.K., Data Collection or Processing: A.A., Analysis or Interpretation: A.A., Literature Search: A.A., G.Ç.K., Writing: A.A., G.Ç.K.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study has received no financial support.
