Biodistribution and radiation dosimetry of [¹⁸F]-JK-PSMA-7 as a novel prostate-specific membrane antigen-specific ligand for PET/CT imaging of prostate cancer

Between December 2017 and March 2018, 10 patients were subjected to [¹⁸F]-JK-PSMA-7 PET/CT due to the progression of prostate cancer. Eight patients suffered from biochemical recurrence (BCR). Two patients had metastasized castration-resistant prostate cancer (mCRPC) and received androgen deprivation therapy (ADT).

PET/CT imaging was performed in accordance with the Institutional Review Board. All patients gave written informed consent to PET imaging and inclusion of their data in a retrospective analysis. All procedures were performed in compliance with the regulations of the local authorities responsible (District Administration of Cologne, Germany).

[¹⁸F]-JK-PSMA-7 was prepared in two steps according to the guidelines of a good manufacturing practice. First, 2,3,5,6-tetrafluorophenyl-6-([¹⁸F]fluoro)-4-methoxynicotinate ([¹⁸F]-Py-OMe-Tfp) was prepared by the reaction of [¹⁸F]fluoride with 2-methoxy-N,N,N-trimethyl-5-((2,3,5,6-tetrafluorophenoxy)carbonyl)pyridine-2-aminiumtrifluoro-methanesulfonate in a mixture of EtOH, MeCN, and tBuOH. The activated ester was purified by SPE purification and added to a freshly prepared ethanolic solution of (((S)-5-amino-1-carboxypentyl)carbamoyl)-l-glutamic acid (HO-Lys-C(O)-Glu-OH) and tetraethylammonium bicarbonate in EtOH. After coupling, the reaction mixture was purified by preparative HPLC. Radiochemical yield of [¹⁸F]-JK-PSMA-7 amounted to 30% (non-decay corrected). The obtained specific activity was 380 ± 16 GBq/μmol and the volumic activity 668 ± 38 MBq/ml. The [¹⁸F]-JK-PSMA-7 solution was administered to the patient by intravenous injection (mean 359.3 MBq, SD 17.1 MBq, range 329–384 MBq). More details about the synthesis and automated production of ¹⁸F-JK-PSMA-7 have recently been published [20].

All PET/CT scans were performed on a Siemens Biograph mCT (mCT 128 Flow Edge, Siemens, Knoxville, USA). In total, eight sequential whole-body PET scans were acquired from the base of the skull to mid-thigh. Patients were injected with 359 ± 17 MBq (4.25 ± 0.55 MBq/kg) of [¹⁸F]-JK-PSMA-7 by a slow intravenous push. The PET acquisition was subdivided into two blocks and the patients into two groups of five according to the following scheme: in patient group no. 1, the first acquisition block started at 20 min p.i. and was repeated four times every 10 min up to 50 min p.i. After a break of 60 min, the second acquisition block started at 110 min p.i and was again repeated four times every 10 min up to 140 min p.i. During an intermission, the patients were allowed to leave the table to void as needed. For each acquisition block, a low-dose non-enhanced CT (120 kV, mA modulation, pitch 1.2, slice thickness 5.0 mm) was performed for attenuation correction. A similar procedure was used in patient group no. 2, the only difference being that the first PET scan was acquired 80 min p.i. and the last one 200 min p.i. All emission data were corrected for attenuation, randoms, scatter, and decay. Reconstruction was conducted with an ordered subset expectation maximization (OSEM) algorithm with 4 iterations and 12 subsets and Gauss-filtered to a transaxial resolution of 5 mm at full-width at half-maximum (FWHM).

Radiation dosimetry was performed using the QDOSE dosimetry software suite (ABX-CRO, Dresden, Germany). All PET and CT data sets were automatically co-registered. Non-target organs such as the kidneys, liver, spleen, lungs, and salivary gland were segmented into volumes of interest (VOI). Time activity curves (TAC) were calculated for the segmented organs. Curve fitting and integration was applied to all TACs to obtain the cumulated activity and residence time. The calculation of cumulated activity was divided into the following three sections: between time 0 and the first measuring point, a linear increase of the TAC was assumed. All measuring points were integrated numerically using trapezoidal approximation. From the last measuring point to infinity, a mono-exponential function was fitted over the last four measuring points and integrated. Whereas the absorbed dose to the salivary glands was determined using the spherical model from OLINDA/EXM1.1 [21], the effective dose and absorbed partial and whole-body dose were calculated using the ICRP endorsed IDAC 1.0 package [22].

The biodistribution of [¹⁸F]-JK-PSMA-7 was quantified by maximum (max) and mean SUV values for all sequentially acquired PET images. The standardized uptake value (SUV) was defined by a circular region drawn around an area with focally increased uptake and automatically adapted to a three-dimensional VOI by software (syngo.via VB20A, Siemens, Erlangen, Germany). The kidneys, liver, spleen, lungs and salivary glands were evaluated with a VOI of 2–3 cm in diameter placed inside the organ parenchyma. PSMA-positive lesions were analyzed separately.

Blood radioactivity concentration was measured in the lumen of the left ventricle of the heart for all sequentially acquired PET images. Quantification was performed for the total blood volume, which was estimated by size, weight, and hematocrit [23]. For this purpose, mean values of the blood radioactivity concentration from both patient groups were determined.

All PET/CT images were reviewed and analyzed on a syngo.via workstation (syngo.via VB20A Software, Siemens, Erlangen, Germany) by two experienced nuclear medicine physicians (both with 20 years’ experience in PET scan reading, certificate for CT reading). The visual assessment of [¹⁸F]-JK-PSMA-7 was considered to be positive if there was focal uptake above the mediastinal blood pool or the liver value. Spherical volumes of interest (VOIs) were manually drawn around areas with focally increased uptake. Quantitative assessment of the uptake in lesions was performed by analysis of maximum (max) and peak SUV values. SUV_max is defined as the hottest voxel within a volume of interest. SUV_peak computes the mean SUV within a 1-cm³ sphere positioned within a VOI so as to maximize that mean. In doing so, voxel super-sampling is performed where the dimensions are halved until they are less than or equal to 0.5 mm on each axis [24]. Tumor-to-background ratios (TBRs) were calculated based on the following equation [25]:

The software package SPSS Statistics 25 (IBM, Armonk, USA) was used for statistical analysis. Dosimetric differences between [¹⁸F]-JK-PSMA-7 and other common ¹⁸F-labeled PSMA-ligands ([¹⁸F]-DCFBC [6], [¹⁸F]-DCFPyL [7], [¹⁸F]-PSMA-1007 [10]) were compared by a Mann-Whitney test. A p value of < 0.05 was considered statistically significant.

The patients were on average 69 years old (range 52–76 years), with an average weight of 86 kg (range 74–100 kg). The average PSA value was 15.6 μg/L (range 0.51–130 μg/L). Nine patients had undergone radical prostatectomy, and one patient had external beam radiation therapy for primary treatment. Two patients received androgen deprivation therapy (ADT) within the last 6 months prior to the examination. Detailed patient characteristics are summarized in Table 1.

Physiologic radiotracer accumulation was observed in the salivary and lacrimal glands, liver, spleen, and intestines, in a pattern resembling the distribution known from other PSMA tracers with excretion via urinary and biliary pathways. SUV_max and SUV_mean as well as their change over time in standard organs are presented in Fig. 1. Between 1 h and 3 h p.i., the uptake in the liver demonstrated an increase in SUV_max and SUV_mean of around 69%, whereas the uptake in the other standard organs showed a decrease in SUV_max and SUV_mean, amounting to − 37% for the kidneys, − 41% for the spleen, − 45% for the lungs, and − 14% for the salivary glands (mean values).

As can be seen in Fig. 2, fast excretion via the blood is evident. From the exponential fit of the TAC, an average half-life of the blood activity concentration of 74 min can be calculated. The whole blood pool contained in mean 13%, 8%, and 4% of the injected activity at 1 h, 2 h, and 3 h p.i., respectively. These initial clinical data are in good agreement with the preclinical data collected in rats [20].

The effective dose from [¹⁸F]-JK-PSMA-7 for the whole body was calculated to be 1.09E−02 mGy/MBq. The highest radiation dose was observed in the kidneys (mean 1.76E−01 mGy/MBq), followed by liver (mean 7.61E−02 mGy/MBq), salivary glands (mean 4.75E−02 mGy/MBq), spleen (mean 1.89E−02 mGy/MBq), and lungs (1.10E-2 mGy/MBq). The individual results for each patient’s whole body and partial organ absorbed doses are presented in Table 2. For comparison, Table 3 presents the corresponding absorbed doses and residence times for other ¹⁸F-labeled PSMA-ligands (PSMA-1007, DCFPyL, DCFBC). As in this work, the QDOSE dosimetry software suite was used for [¹⁸F]-PSMA-1007. Other software packages were used for [¹⁸F]-DCFBC and [¹⁸F]-DCFPyL, but radiation dosimetry was performed in a similar way.

There was no statistically significant difference between the residence times of [¹⁸F]-JK-PSMA-7 and [¹⁸F]-PSMA-1007 for the examined organs (p > 0.05 for all organs). The comparison between [¹⁸F]-JK-PSMA-7 and [18F]-DCFPyL showed a statistically significant increase for residence times in the liver (47%, p = 0.034) and the salivary glands (40%, p = 0.003) for [¹⁸F]-JK-PSMA-7. For the kidneys (p > 0.05) and the spleen (p > 0.05), on the other hand, no significant difference in residence times was found. In the comparison between [¹⁸F]-JK-PSMA-7 and [¹⁸F]-DCFBC, a statistically significant increase was observed for the residence times in the kidneys (17%, p = 0.001) and the liver (28%, p = 0.002) for [¹⁸F]-JK-PSMA-7. There was no statistically significant difference for the residence time in the spleen and the lungs (p > 0.05).

Six out of ten patients were scored as PSMA-positive. Readings were based on the interpretation of the latest PET-scans, performed 140 min and 200 min, respectively. A total of 18 suspicious lesions were analyzed, which included six bone lesions, nine lymph nodes, and three local lesions within the prostate fossa. Three of these six patients (patients no. 1, 2, and 4) belonged to patient group no. 1 (acquisition protocol 20 min–140 min p.i.) and the other three (patients no. 5, 6, and 8) to patient group no. 2 (acquisition protocol 80 min–200 min p.i.).

All PSMA-positive lesions showed an increased uptake over time, although significant differences emerged between the two patient groups. In patient group no. 1, the SUV_max and SUV_peak values increase slowly over time and reach an almost stable plateau 100 min after injection. In patient group no. 2, on the other hand, the SUV_max and SUV_peak values continue to rise significantly. These differences in the time courses of SUV_max and SUV_peak are reflected in corresponding values for the TBR. Thus, between 100 and 140 min p.i., the mean TBR increased by 50%, whereas between 140 and 200 min p.i., the mean TBR increased by 78%. Figure 3 shows the time course of the SUV_max, SUV_peak, and TBR in PSMA-positive lesions. In addition, Fig. 4 shows the images of two patients, one from patient group no. 1 and one from patient group no 2.

All patients tolerated the examination well. No drug-related pharmacological effects or physiologic responses were observed. None of the patients reported any adverse events or side-effects at the time, when the PET-results were discussed with the patient or some weeks later, when the therapeutic consequences were discussed by telephone counseling.

Five out of the 6 patients with PSMA-positive lesions showed a decreased PSA level after radiotherapy of the PSMA-positive lymph nodes or after salvage radiotherapy of the PSMA-positive lesions within the prostate fossa. One patient had multiple PSMA-positive lymph nodes and skeletal lesions, which were also visible on the CT scan.

Two out of the 4 patients with a PSMA-negative PET result had a stable PSA level after 6 to 8 months without any therapy. One patient received salvage radiotherapy and the PSA level decreased. One patient developed osteosclerotic bone metastases after 9 months, then visible on the CT scan, but again PSMA-negative with [⁶⁸Ga]-PSMA-11.