Time point-independent tumor positivity of 68Ga-PSMA-PET/CT pre- and post-biopsy in high-risk prostate cancer

Objective Prostate-specific membrane antigen (PSMA)-PET/CT imaging has gained increasing clinical importance for the detection and staging of high-risk primary prostate cancer (PCa). However, it is unclear whether the routine practice of prostate biopsy obscures the image finding of PSMA-PET/CT. This study aimed to compare the tumor positivity rate of PSMA-PET/CT performed pre- (PSMA-PET/CTpre) and post-biopsy (PSMA-PET/CTpost) in high-risk PCa patients. Patients and methods We matched 58 PSMA-PET/CTpost with 58 PSMA-PET/CTpre studies for primary detection of high-risk PCa according to clinical characteristics. Three subgroups of PSMA-PET/CTpost were defined by the intervals after biopsy (≤ 1 week, 1 ~ 2 weeks, and 2 ~ 5 weeks). Tumor positivity rates were determined, and SUVmax of primary tumors were compared separately for the two main groups and the related subgroups. Malignant prostate tissues from 20 of these patients were examined by immunohistochemical analysis of PSMA. In addition, the values of PSMA-PET/CTpre and PSMA-PET/CTpost in assessing seminal vesicle invasion (SVI) were evaluated in patients who underwent radical prostatectomy. Results All the primary tumors were positive on PSMA-PET/CTpost and PSMA-PET/CTpre imaging, resulting in a patient-based positivity rates of 100% (58/58) in both groups. All examined IHC results (20/20) confirmed the high-level expression of PSMA. SUVmax of primary tumors did not differ between the two main groups (16.1, IQR 9.8–26.6 vs. 16.5, IQR 11.0–26.7, p > 0.05). Subgroup analysis of PSMA-PET/CTpost (≤ 1 week, 1 ~ 2 weeks, and 2 ~ 5 weeks) also showed no significant difference in tumor SUVmax (15.8, IQR 9.5–22.2; 17.8, IQR 9.8–29.2; and 15.4, IQR 10.1–30.3. p > 0.05). PSMA-PET/CTpost and PSMA-PET/CTpre exhibited similar value in SVI detection as well. Conclusions The tumor positivity rate was consistently high for PSMA-PET/CT pre- and post-biopsy. A prior biopsy does not seem to affect the tumor positivity rate of PSMA-PET/CT in high-risk PCa. Supplementary Information The online version contains supplementary material available at 10.1007/s12149-022-01732-w.


Introduction
Prostate cancer (PCa) is currently the most common neoplasm in males and the second leading cause of cancerrelated deaths for males in western countries [1]. Standard procedure for PCa diagnosis includes detection of elevated prostate-specific antigen (PSA) level and digital rectal examination (DRE) in patients [2].
Imaging techniques opens up a new horizontal for the diagnosis and management of males linked with risk of prostate cancer. Multiparametric magnetic resonance imaging (mpMRI), for example, proves to be a valuable tool for risk stratification, biopsy guidance, and local staging of clinically significant PCa [3,4]. The procedure of invasive prostate biopsy, however, may cause biopsy-related artifacts including hemorrhage and edema in the imaging results, and subsequently lead to over-or under-estimation of tumor burden. As a result, prostate MRI is usually performed prior to a biopsy and the use of MRI in PCa management within a short time after biopsy should be cautioned [5,6], especially in those with previous negative or inconclusive biopsy reports.
Prostate-specific membrane antigen (PSMA) is a transmembrane protein with a 100-to 1000-times higher expression in primary and metastatic prostate tumors than in normal tissues [7,8]. PSMA-targeted positron emission tomography (PET) is increasingly used to assess the recurrence of PCa as well as to localize primary disease and metastases [9][10][11][12][13][14][15][16][17]. The recent prospective PRIMARY trial [18] showed that the combination of PSMA-PET and mpMRI was superior to mpMRI alone in the diagnosis of clinically significant PCa. While PSMA-PET gains popularity as the primary diagnostic modality for PCa, it is also brought to attention that a pre-imaging biopsy may compromise the performance of PSMA-PET/CT or PSMA-PET/ MR, in a similar way to that of MRI. On the other hand, the imaging results of PSMA-PET/CT are reconstructed from the distribution of PSMA molecules rather than anatomical features, and therefore, may be unaffected by the biopsy-related artifacts. Up to date, it is unclear if a biopsy affects the tumor positivity rate of the ensuing PSMA-PET/ CT for primary PCa, or whether it is necessary to implicate a waiting-period between the biopsy and the following PSMA-PET/CT. Therefore, we herein present this retrospective, matchedpair analysis to assess the tumor positivity rate of pre-and post-biopsy 68 Ga-PSMA-PET/CT in patients with high-risk PCa.

Patients
Fifty-eight patients (mean age 68.1 ± 7.8 years; range 50-81 years) with newly diagnosed high-risk PCa who underwent post-biopsy 68 Ga-PSMA-617 PET/CT (PSMA-PET/CT post ) at our institution from April 2018 to December 2020 were consecutively enrolled and retrospectively analyzed. Based on the European Association Urology guidelines, high-risk patients were defined as those with the presence of one or more of the following criteria: PSA concentration > 20 ng/mL, ISUP grade group ≥ 4, or clinical stage ≥ T2c [19]. Of these, 48 had undergone a transrectal ultrasonography (TRUS)-guided transrectal 12-core prostate biopsy, 1 had trans-perineal mpMRI targeted biopsy (8 + 1-core) and 9 outpatient received biopsy in other institutions with unknown approaches and number of needle cores. All patients were examined by 68 Ga-PSMA-PET/CT within 5 weeks after prostate biopsy (PSMA-PET/CT post ) and were subcategorized into three subgroups based on the interval between biopsy and PSMA-PET/CT: ≤ 1 week, 1 ~ 2 weeks, and 2 ~ 5 weeks. Patients with a history of electron-prostatectomy or being treated with anti-cancer therapy prior to PSMA-PET/CT scan were excluded. PSA values tested within 5 weeks before PSMA-PET/CT were identified. Available postbiopsy MRI data of these patients for pretreatment local staging were also reviewed to determine the presence or absence of hemorrhage. Of the 58 PSMA-PET/CT post patients, 28 underwent radical prostatectomy (RP) within 1 month after the PSMA-PET/CT scan.
Fifty-eight corresponding patients (mean age 69.7 ± 8.7 years; range 46-85 years) with pathology-confirmed PCa who had undergone 68 Ga-PSMA-PET/CT before prostate biopsy (PSMA-PET/CT pre ) during the same time period were consecutively identified in the institution's database on the basis of the following clinical parameters: ISUP grade group (2-3 vs. 4-5), pre-scan PSA values (4-20 ng/mL, > 20 ≤ 100 ng/mL vs. > 100 ng/mL and clinical Tumor stage (cT) (cT2c, cT3 vs. cT4). Characteristics of the matched-pair cohorts are summarized in Table 1. For PSMA-PET/CT pre patients, there were longer time intervals between biopsy, radical prostatectomy, and the post-biopsy PSMA-PET, which resulted in patient dropouts from the study. Of the 58 PSMA-PET/CT pre patients, 14 underwent RP within 1 month after the examination.
This study was approved by the Ethics Committee of the Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (No. TJ-IRB20190422). All reported investigations were conducted in accordance with the Helsinki Declaration and national regulations. Patient data were de-identified and processed per institutional ethics guidelines.

Imaging procedure
68 Ga-PSMA-617 was administered as an intravenous bolus at a dose of 89-200 MBq (mean 114.6 ± 30.9 MBq) and PET acquisition was started at a mean ± SD time of 60 ± 11 min after tracer injection on a dedicated PET/CT hybrid tomography (GE Discovery 690, General Electronic Healthcare, USA). Image acquisition was started with a non-enhanced helical CT scan using automatic mA-modulation and 120 kV. CT scans were reconstructed to a slice thickness of 3.75 mm. PET scans were acquired in 3D mode with an acquisition time of 120 ~ 180 s per bed position from the mid-thighs to the skull base. PET images were reconstructed with the built-in GE VUE point method.

PET image analysis and quantification
All PSMA-PET/CT post and PSMA-PET/CT pre imaging were evaluated by two board-certified nuclear medicine physicians with specific training in 68 Ga-PSMA-PET/CT in consensus. Readers were blinded to patients' history and histopathological results. The scan was considered positive when intraprostatic tracer uptake higher than liver was noticed as reported in the PROstate cancer Molecular Imaging Standardized Evaluation (PROMISE) study [20]. Tumor uptake patterns were classified as unilateral focal, bilateral multifocal, and increased diffuse uptake. Maximum standardized uptake value (SUVmax) of the suspected prostate tumors with the highest PSMA-ligand uptake was noted. Seminal vesicle invasion was defined by the presence of focal or diffuse PSMA-ligand accumulation above the background by visual analysis. PROMISE criteria were adopted for interpretation of PSMA uptake in lymph node and bone metastases [20]. Tumor positivity rates were determined separately for PSMA-PET/CT post and PSMA-PET/CT pre group. SUV max of primary tumors were compared in (1)

Pathological and immunohistochemical (IHC) analyses
A total of 42 (28 from PSMA-PET/CT post group and 14 from PSMA-PET/CT pre group) patients underwent radical prostatectomy with resection of the seminal vesicles. PSMA was stained with an anti-PSMA rabbit monoclonal antibody (EPR6253, ab133579, Abcam, 1:500 dilution) on a Leica Bond-Max auto-stainer. The intensity of staining (weak, moderate or intense) and the percentage of positively stained cells (focal, regional, or diffuse) were graded as reported in a previous study [21]. Cases categorized as intense diffuse, intense regional, or moderate diffuse were considered as overexpressing PSMA protein.

Statistical analysis
Statistical analyses were performed using SPSS version 26.0 software (IBM Corp., Armonk, New York). Quantitative data are expressed as mean values ± standard deviations (SD) or medium ± interquartile range (IQR). Tumor positivity rates were determined separately for the two groups. The sensitivity, specificity and accuracy of PSMA-PET/CT for evaluating SVI were calculated using histopathology results of RP as the reference standard. For comparisons of the tumor SUVmax between groups and in subgroups, p values were calculated by the Mann-Whitney U test. The correlations of tumor SUVmax with the tPSA level, ISUP grade, and pT were analyzed using Spearman's correlation. p values < 0.05 were considered statistically significant.

Primary tumor and PSMA protein expression by IHC
All the primary tumors were positive on PSMA-PET/CT post and PSMA-PET/CT pre (both 100%, 58/58) imaging. Twenty prostatectomy specimens from these primary tumors were available for immunohistochemistry, all of which showed strong expression of PSMA.
Post-biopsy MR imaging for preoperative local staging was available for review in eight of the PSMA-PET/ CT post patients, and hemorrhages was observed in seven of them. In contrast to the intense PSMA-ligand uptake in prostate tumors, the uptake in hemorrhage regions was minimal. Representative images of 68 Ga-PSMA-617 scans are shown in Figs. 1-5: primary tumor at 5 days (Fig. 1) and 1 week (Fig. 2) after prostate biopsy, bilateral focal lesions in prostate at 3 weeks after biopsy (Fig. 3), single focal lesion in prostate at 1 month after biopsy (Fig. 4), and increased tumor uptake prior to biopsy (Fig. 5).

Correlation of tumor SUVmax in PSMA-PET/CT and other parameters
A slightly positive correlation was observed between tumor SUVmax and the pT stage (r = 0.316, p < 0.05) in patients who underwent RP. There was a similar positive correlation between the tumor SUVmax and tPSA value (r = 0.358, p < 0.01) in PSMA-PET/CT pre patients. However, no significant correlation between tumor SUVmax and pre-scan tPSA value was found in the PSMA-PET/CT post patients (r = − 0.012, p > 0.05) (Fig. 6). Neither was a correlation established between ISUP grade and the tumor SUVmax (r = 0.085, p > 0.05).

Discussion
With the current shift towards an early PSMA-PET/CT scan for the primary diagnosis and staging of PCa, it is imperative to understand the potential impact of biopsy on the imaging quality and diagnostic power of the ensuing PET. Previous studies have demonstrated that PSMA-PET/CT could yield remarkable results for the pre-surgery identification of primary high-risk PCa and metastases [22][23][24]. For example, Fendler et al. [25] reported that 68 Ga-PSMA-PET/CT performed minimally 1-week post-biopsy could present satisfactory diagnostic accuracy for the detection of primary lesions. Nevertheless, the time interval between biopsy and the following PET/CT has yet to be specified in most of the studies. Therefore, we postulated that an evaluation on the impact of biopsy timing upon the tumor positivity rate of PSMA-PET/CT would help fill such a knowledge gap.
Our results demonstrate the pre-and post-biopsy PSMA-PET/CT both detected 100% of primary high-risk PCa lesions, showing no significant difference in SUV max of the primary tumors. Subgroup analysis of patients undergoing PSMA-PET/CT post revealed that the time interval between biopsy and ensuing PET scan did not have a significant impact on the SUVmax values of tumors. Neither did this time interval have any effects on the performance of PET/CT in locating tumor nodules. Taken together, we have shown that biopsy is unlikely to compromise the diagnostic power of ensuing PSMA-PET/CT for primary or metastatic lesions, and any delay of the post-biopsy PET scan is unnecessary. Our results provide valuable guidelines for clinicians, as an example, in scheduling a timely PSMA-PET scan for a highrisk patient that has a negative or inconclusive biopsy report.
Hemorrhage is the most frequent complication in prostate biopsy, and has been detected in MRI scans on 72.2%, 57.1%, and 52% patients at < 4, 4 ~ 6, and > 6 weeks after the biopsy, respectively [26]. While there have been no reports on the impact of biopsy-induced hemorrhage on the detectability of PSMA-PET in PCa, our study confirmed that hemorrhage or blood products did not take up radiotracers in PSMA-PET. In comparison, there was an intense uptake of radiotracers in pathologically verified prostate tumors. Therefore, we concluded that in spite of biopsy-induced hemorrhage, PSMA-PET/CT was capable of detecting primary tumor lesions in patients with high-risk PCa, regardless of the time interval between biopsy and PET.
PSAM-PET/CT's ability to withstand the interference by a prior biopsy can be attributed to the following factors. First, the high PSMA expression in PCa enables a specific imaging of PSMA molecules in prostate tumors [7,8]. Preclinical studies have shown that prostate tumors exhibited highly homogenous and intense expression of PSMA [27,28]. The IHC results of patients in our study also validated that PSMA was homogenously overexpressed throughout the prostate tumors, which constituted the molecular pre-requisite for increased tumor uptake of PSMA ligands. Second, the blood products of hemorrhage did not take up PSMA ligands. Third, the extent of post-biopsy hemorrhage was less severe in prostate tumor than in peritumor normal tissues, which became even less significant as the tumor size increased and the percentage of normal tissues dropped [29]. Notably, 64% of patients with advanced and large tumors in both study groups were presented with diffuse intense tumor uptake on PSMA-PET imaging, enabling PSAM-PET/CT ability to offset the interference. Fourth, the biopsy-related hemorrhage in tumor foci may have resolved spontaneously more rapidly than in normal prostatic tissue [30]. Lastly, PSMA expression in inflammatory lesions was reported to be extremely rare [7].
In our results, the tumor positivity rates and SVI detection value for 68 Ga-PSMA-PET/CT post and 68 Ga-PSMA-PET/ CT pre are slightly higher than those reported in the previous studies [23,29,[31][32][33], probably because PSA screening for early detection of PCa has not been widely adopted and therefore more patients already had advanced-stage tumors at the time of PET scans [34].
In line with previous studies, the tumor SUVmax was found to be higher in locally advanced PCa (≥ pT3) than organ-confined tumor (pT2) (p < 0.05) in patients who underwent RP (Table 2). However, the radiotracer uptake of primary lesions in patients with metastatic PCa was not consistently higher than that in patients without metastasis. Mannweiler et al. reported that metastatic PCa exhibited significant intra-and inter-tumor heterogeneity [35]. Silver et al. showed that metastatic lesions tended to have higher PSMA expression than primary lesions, whereas nodule metastases had higher PSMA expression than bone metastases [7]. Therefore, the variability of the intraprostatic uptake of PSMA-ligand may be caused by the inherent heterogeneity of metastatic PCa. The intensity of intraprostatic tracer uptake, however, did not have any impact on patient management in these metastatic cases.
Moreover, a slightly positive correlation was found between tumor uptake of PSMA and the pT stage in patients who underwent RP as well as the tumor SUVmax and tPSA value in PSMA-PET/CT pre patients. These findings were consistent with those of previous studies [22,24,36]. However, no significant correlation between the tumor SUVmax and pre-scan tPSA value was found in the PSMA-PET/CT post patients. This could be attributed to the spurious transient elevation of serum PSA associated with mechanical manipulation of the prostate by biopsy or catheterization in the postbiopsy cohort [37]. In addition, we did not observe a significant correlation between the tumor SUVmax and the ISUP grade, which can be explained in part by the small number of patients with ISUP grade 2 and 3 and the potential disagreement of ISUP grade between biopsy and prostatectomy specimens might be the underlying reasons [38].
Our study has some limitations. First, this is a retrospective matched-pair comparison in high-risk PCa patients conducted in a single institute. Second, the limited sample size and distribution of cancer risk population may also cause a potential bias. Third, the needle biopsy of the prostate was performed by physicians from different institutions, and their technique or the number of needle cores varied. In addition, a direct comparison between PSMA-PET/CT post with post-biopsy MRI imaging was not available for the whole cohort, as MRI is not mandatory in the post-biopsy setting in our institution. Meanwhile, whole-mount step-section of pathologic specimens from prostatectomy was not performed as a routine analysis. Thus, PSMA-PET/CT for PCa diagnosis was not evaluated using lesion-based analysis. Therefore, further investigations using a head-to-head comparison of PSMA-PET performed before and after prostate biopsy in a greater spectrum of cancer risk population are warranted.

Conclusion
The tumor positivity rate was consistently high for PSMA-PET/CT pre-and post-biopsy. A prior biopsy does not seem to affect the tumor positivity rate of PSMA-PET/CT in highrisk PCa.