Terbium-161 for PSMA-targeted radionuclide therapy of prostate cancer

Purpose The prostate-specific membrane antigen (PSMA) has emerged as an interesting target for radionuclide therapy of metastasized castration-resistant prostate cancer (mCRPC). The aim of this study was to investigate 161Tb (T1/2 = 6.89 days; Eβ-uperscript>av = 154 keV) in combination with PSMA-617 as a potentially more effective therapeutic alternative to 177Lu-PSMA-617, due to the abundant co-emission of conversion and Auger electrons, resulting in an improved absorbed dose profile. Methods 161Tb was used for the radiolabeling of PSMA-617 at high specific activities up to 100 MBq/nmol. 161Tb-PSMA-617 was tested in vitro and in tumor-bearing mice to confirm equal properties, as previously determined for 177Lu-PSMA-617. The effects of 161Tb-PSMA-617 and 177Lu-PSMA-617 on cell viability (MTT assay) and survival (clonogenic assay) were compared in vitro using PSMA-positive PC-3 PIP tumor cells. 161Tb-PSMA-617 was further investigated in therapy studies using PC-3 PIP tumor-bearing mice. Results 161Tb-PSMA-617 and 177Lu-PSMA-617 displayed equal in-vitro properties and tissue distribution profiles in tumor-bearing mice. The viability and survival of PC-3 PIP tumor cells were more reduced when exposed to 161Tb-PSMA-617 as compared to the effect obtained with the same activities of 177Lu-PSMA-617 over the whole investigated concentration range. Treatment of mice with 161Tb-PSMA-617 (5.0 MBq/mouse and 10 MBq/mouse, respectively) resulted in an activity-dependent increase of the median survival (36 vs 65 days) compared to untreated control animals (19 days). Therapy studies to compare the effects of 161Tb-PSMA-617 and 177Lu-PSMA-617 indicated the anticipated superiority of 161Tb over 177Lu. Conclusion 161Tb-PSMA-617 showed superior in-vitro and in-vivo results as compared to 177Lu-PSMA-617, confirming theoretical dose calculations that indicate an additive therapeutic effect of conversion and Auger electrons in the case of 161Tb. These data warrant more preclinical research for in-depth investigations of the proposed concept, and present a basis for future clinical translation of 161Tb-PSMA-617 for the treatment of mCRPC. Electronic supplementary material The online version of this article (10.1007/s00259-019-04345-0) contains supplementary material, which is available to authorized users.


Introduction
The prostate-specific membrane antigen (PSMA) is a cellsurface glycoprotein that is expressed in normal prostate tissue and overexpressed in prostate cancer [1,2]. There are indications that the expression level of PSMA correlates with the stage of the disease and the risk of disease progression [3,4]. PSMA is, therefore, an interesting target to use for radionuclide therapy of metastasized castration-resistant prostate cancer (mCRPC) [5][6][7][8]. The topic of PSMA targeting became popular with the development of small-molecule-based radioligands [9]. Initial compounds were designed for radioiodination, suitable for nuclear imaging, and the first to be used therapeutically in patients [10]. Subsequently, PSMA ligands were developed with a chelator to allow their use in combination with radiometals for both imaging and therapeutic purposes [5,8,11]. PSMA-617 and PSMA I&T, equipped with a DOTA and DOTAGA chelator, respectively, have been used for targeted radionuclide therapy of mCRPC in clinics [7,12,13]. For this purpose, they were mostly labeled with 177 Lu (T 1/2 = 6.65 d; Eβ av = 134 keV; Eγ = 113 keV, I = 6.17%, Eγ = 208 keV, I = 10.36%), which is currently the mostoften applied radiometal for therapeutic purposes in the clinics [14]. In specific cases, 225 Ac-PSMA-617 was employed for the treatment of patients at end-stage without further treatment options [15][16][17]. 225 Ac decays with a half-life of 10 days, emitting several αand β¯-particles while decaying via a sequence of radioactive daughter nuclides [18]. Although the results obtained with 225 Ac-PSMA-617 were impressive, undesired side effectsreferring to irreversible damage of salivary and lacrimal glandshave been reported [17]. The question arises, therefore, whether alternative radiometals could be used for targeted radionuclide therapy of mCRPC which would be potentially more powerful than the currently-employed 177 Lu, without causing additional sideeffects.
In this work, we investigated 161 Tb, a recently-introduced radiolanthanide for therapeutic applications [19]. 161 Tb decays with a half-life of 6.89 days to stable 161 Dy, while emitting β¯-particles (Eβ av = 154 keV) suitable for therapeutic purposes and γ-radiation (Eγ = 49 keV, I = 17.0%; Eγ = 75 keV, I = 10.2%) useful for SPECT imaging. In this regard, 161 Tb closely resembles 177 Lu, even though the emitted γradiation is of lower energy. 161 Tb also emits a substantial number of low-energy conversion and Auger electrons, which makes this radionuclide exceptionally interesting for the treatment of disseminated cancers with multiple metastases ranging from a single cell (diameter:~10 μm) to micro cell clusters (diameter: < 1 mm) [20]. Monte Carlo simulations performed by Hindié et al. to assess the dose delivered to 10-μm spheres revealed a 3.5-fold increased value when using 161 Tb as compared to 177 Lu [21]. In larger tumors (diameter > 10 mm), the emitted electron energy from 161 Tb and 177 Lu respectively is almost entirely absorbed, resulting in a 1.3-fold higher absorbed electron energy fraction per decay for 161 Tb (total electron emission of 197 keV/decay) compared to 177 Lu (147 keV/decay), making 161 Tb more potent than 177 Lu [21]. An additional advantage of 161 Tb over 177 Lu may be the existence of diagnostic counterparts, including 152 Tb (T 1/2 = 17.5 h; Eβ + av = 1140 keV, I = 20.3%) and 155 Tb (T 1/2 = 5.32 d; Eγ = 87 keV, 32.0%, 105 keV, I = 25.1%) for PET and SPECT imaging respectively, potentially enabling pretherapeutic dosimetry with chemically identical radiopharmaceuticals [22][23][24][25]. The results of theoretical calculations performed by Champion and co-workers also indicate that 161 Tb outperforms other clinically employed ( 177 Lu, 90 Y) and nonstandard therapeutic radionuclides ( 47 Sc, 67 Cu) with regard to the dose delivery to small lesions [21,26].
The production of 161 Tb via the 160 Gd(n,γ) 161 Gd → 161 Tb nuclear reaction was previously reported by Lehenberger et al. [19]. At the Paul Scherrer Institute (PSI), the method of processing Gd targets irradiated in high neutron flux reactors (RHF, Institut Laue-Langevin, Grenoble, France or SAFARI-1, Necsa, Pelindaba, South Africa) or at a spallation neutron source (SINQ, PSI, Switzerland) was implemented some years ago [22]. The chemical separation of 161 Tb from the target material has since been further developed and optimized at PSI.
The topic of the present study was to investigate 161 Tb with regard to its application for radionuclide therapy. 161 Tb was, therefore, used to label PSMA-617 to enable preclinical comparison with 177 Lu-PSMA-617. The in-vitro experiments and biodistribution studies in PC-3 PIP/flu tumor-bearing mice were performed to confirm equal chemical and pharmacokinetic properties of 161 Tb-PSMA-617 and 177 Lu-PSMA-617 respectively. Importantly, the effect of 161 Tb-PSMA-617 was compared to that obtained with 177 Lu-PSMA-617 by means of in-vitro cell viability and survival assays, and the therapeutic effect of 161 Tb-PSMA-617 was shown in vivo using tumorbearing mice.

Materials and methods
Production and chemical separation of 161 Tb 161 Tb was produced as previously reported [22]. Enriched 160 Gd targets were irradiated over a period of 1-2 weeks at the SAFARI-1 reactor at Necsa, Pelindaba, South Africa, or at the RHF at Institut Laue-Langevin, Grenoble, France. In some cases, 3-week irradiations were performed at the spallation-induced neutron source SINQ, PSI, Switzerland. 161 Tb was chemically separated from the Gd target material and impurities by cation exchange chromatography, using an optimized method of the previously-published process (Supplementary Material) [19,22].

In-vivo studies
In-vivo experiments were approved by the local veterinarian department and conducted in accordance with the Swiss law of animal protection. Athymic nude BALB/c mice were obtained from Charles River Laboratories (Sulzfeld, Germany) at the age of 5-6 weeks. Mice were subcutaneously inoculated with PC-3 PIP tumor cells (6 × 10 6  The data were compared with those previously obtained for 177 Lu-PSMA-617 [27] and analyzed for significance using a one-way ANOVA with Tukey's multiple comparison post-test using GraphPad Prism software (version 7). A p-value of < 0.05 was considered statistically significant.

Dosimetry estimations
The mean specific absorbed doses (Gy/MBq) to the tumor xenografts and the kidneys were estimated for 161 Tb-PSMA-617 and 177 Lu-PSMA-617 (Supplementary material). The tissue distribution profile of 161 Tb-PSMA-617 was considered as equal to the previously-determined biodistribution data of 177 Lu-PSMA-617 [27,32]. The [% IA/g] values were converted to non-decay corrected values using the respective halflives of the radionuclides to obtain time-integrated activity to infinity. The mean absorbed energy per decay to cells in the cell viability study was calculated using Monte Carlo simulations with PENELOPE-2014 [33].

SPECT/CT imaging studies
In a separate study, SPECT/CT experiments were performed 12-14 days after tumor cell inoculation using a dedicated small-animal SPECT/CT camera (NanoSPECT/ CT TM , Mediso Medical Imaging Systems, Budapest, Hungary) as previously reported (Supplementary material) [34]. 161 Tb-PSMA-617 (~25 MBq/nmol) was diluted in saline for injection. Scans were acquired at 1 h, 4 h, and 24 h after injection of the radioligands (~25 MBq, 1 nmol, 100 μL). During the in-vivo scans, mice were anesthetized using a mixture of Isoflurane and oxygen.

Therapy study
Three groups of mice (n = 6) were injected with only saline, 161 Tb-PSMA-617 (5.0 MBq; 1 nmol/mouse), or 161 Tb-PSMA-617 (10 MBq; 1 nmol/mouse) at Day 0 of the therapy, 6 days after PC-3 PIP tumor cell inoculation ( Table 1). The mice were monitored by measuring body weights and the tumor sizes every other day over 12 weeks. Mice were euthanized when pre-defined endpoint criteria were reached, or when the study was terminated at Day 84 (Supplementary material). The relative body weight (RBW) was defined as [BW x / BW 0 ], where BW x is the body weight in gram at a given day x, and BW 0 the body weight in grams at Day 0. The tumor dimension was determined by measuring the longest tumor axis (L) and its perpendicular axis (W) with a digital caliper. The tumor volume (V) was calculated according to the eq. [V = 0.5 * (L * W 2 )]. The relative tumor volume (RTV) was defined as [TV x /TV 0 ], where TV x is the tumor volume in mm 3 at a given day x, and TV 0 the tumor volume in mm 3 at Day 0.

Assessment of therapy study
The efficacy of the radionuclide therapy was assessed by the tumor growth delay (TGD x ), which was calculated as the time required for the tumor volume to increase x-fold over the initial volume at Day 0. The tumor growth delay index [TGDI x = TGD x (T)/TGD x (C)] was calculated as the TGD x ratio of treated mice (T) over control mice (C) for a 2-fold (x = 2, TGD 2 ) and 5-fold (x = 5, TGD 5 ) increase of the initial tumor volume. Statistical analysis was performed by a one-way ANOVA with Tukey's multiple comparison post-test using GraphPad Prism software (version 7). A value of p < 0.05 was considered statistically significant. The median survival was calculated by Kaplan-Meier curves using GraphPad Prism software (version 7).
Potential early side-effects related to the exposure to radiation were evaluated by the assessment of absolute and relative (to body and to brain) organ weights, selected clinical chemistry plasma parameters including creatinine (CRE), blood urea nitrogen (BUN), alkaline phosphatase (ALP), total bilirubin (TBIL), and albumin (ALB), as well as via histological analysis of bone marrow and salivary glands. The data were analyzed for statistical significance (Supplementary material).

Production of 161 Tb
No-carrier-added 161 Tb was produced at high activities of 6-20 GBq (end of irradiation) depending on the irradiation parameters (neutron flux, irradiation time, and mass of target material). The chemical separation resulted in a radionuclidically pure ( 160 Tb < 0.007%) product of high radiochemical purity comparable to commercial, no-carrieradded 177 Lu (Supplementary material, Table S1). 161 Tb was made available at a high-activity concentration (10-20 MBq/ μL) in Suprapur TM HCl (0.05 M) to be used directly for radiolabeling of PSMA-617.

Internalization studies
The

In-vitro tumor-cell viability and survival
The reduction of viability and survival of PC-3 PIP tumor cells after exposure to 161 Tb-PSMA-617 and 177 Lu-PSMA-617 correlated with the applied activity concentration. 161 Tb-PSMA-617 was significantly more effective in reducing the tumor-cell viability (determined by MTT assays) and survival (determined by clonogenic assays) as compared to 177 Lu-PSMA-617 when applied at activity concentrations in the range of 0.1-10 MBq/mL (p < 0.05) and 0.05-5.0 MBq/mL (p < 0.05) respectively ( Fig. 1a/b). Under the given experimental conditions, the mean absorbed energy to tumor cells in MTT assays was calculated to be 3.2-4.2-fold higher for 161 Tb than for 177 Lu. Lower values reflect the situation for cell monolayers, whereas the higher value refers to the "single-cell situation" which was more the case during the treatment, particularly in the setting of the clonogenic assay where the cell number per well was low. The viability of PSMA-negative PC-3 flu cells was not affected when the radioligands were applied at concentrations of up to 10 MBq/mL. Only a slight reduction that was equal for both radioligands (p > 0.05) was detected at the highest concentration (20 MBq/mL). The survival of PC-3 flu cells was, however, affected at radioligand concentrations of 1 MBq/mL and higher, with a tendency of a more pronounced effect from 161 Tb-PSMA-617 (p > 0.05) (Fig.  1c/d). The viability and survival of PC-3 PIP tumor cells exposed to 161 Tb-DPTA and 177 Lu-DTPA were not affected, and showed only a marginal reduction at higher radioligand concentration, which was equal for both radionuclide complexes (p > 0.05) (Fig. 1e/f).

Biodistribution studies and dose estimation
Time-dependent biodistribution of 161 Tb-PSMA-617 was assessed in PC-3 PIP/flu tumor-bearing mice and compared to the data previously obtained with 177 Lu-PSMA-617 (Supplementary material, Table S2) [27,32]. The uptake of 161 Tb-PSMA-617 in PC-3 PIP tumor xenografts reached a maximum at 4 h p.i. (49 ± 5.5% IA/g) and decreased slowly over time (22 ± 4.3% IA/g at 96 h p.i.). Accumulation of 161 Tb-PSMA-617 in PC-3 flu tumors and other non-targeted organs was in the range of blood activity levels or below at any evaluated time point. The radioligand was cleared via the kidneys over the first few hours after injection (9.6 ± 1.3% IA/ g; 1 h p.i. 2.9 ± 0.14% IA/g; 4 h p.i.). These results confirmed that the tissue distribution profile of 161 Tb-PSMA-617 was equal (p > 0.05) to the data previously published for 177 Lu-PSMA-617 (Fig. 2) [27,32].
For the absorbed dose estimations, the absorbed fractions of the assumed spherical tumors (80 mm 3 Fig. 4). The tumor response in mice that received 10 MBq 161 Tb-PSMA-617 was highly variable among the six mice, ranging from similar effects to those observed after injection of 5.0 MBq 161 Tb-PSMA-617 to complete tumor remission (Fig. 5). The median survival time of mice treated with 161 Tb-PSMA-617 was 36 days, which was clearly longer than the median survival of the control mice (19 days). The application of 10 MBq 161 Tb-PSMA-617 increased the median survival of mice to 65 days. In two of the six cases of this group, the PC-3 PIP tumors disappeared entirely, so that the mice survived over 12 weeks without any signs of tumor regrowth (Figs. 4 and 5).

Monitoring of mice during therapy
In the group of mice that received 10 MBq 161 Tb-PSMA-617, the body weight was slightly higher than in the other two groups at therapy start. While control mice and mice that received 5.0 MBq 161 Tb-PSMA-617 experienced body weight loss over time, the body weight of mice that received 10 MBq 161 Tb-PSMA-617 remained stable (Supplementary material, Table S3). In line with this result, the average absolute organ mass, calculated for kidney, liver and spleen of these mice, were also higher compared to those recorded in mice from the two other groups. The same held true for these organ masses calculated relative-to-body mass and relative-to-brain mass (Supplementary material, Table S3/S4). This indicates that exposure to 161 Tb-PSMA-617 at 10 MBq per mouse mitigated the detrimental effects on the general health status observed in the other groups, which were probably caused by the rapidly growing tumors.
Evaluation of selected clinical chemistry parameters of renal and hepatic function (CRE, BUN ALP, TBIL, ALB) and the histological analysis of the bone marrow and salivary glands revealed no meaningful difference between the different groups (Supplementary material, Tables S5/S6).  Table S8).

Discussion
In this study, 161 Tb was investigated as a potential alternative to 177 Lu to be used in combination with PSMA-targeting ligands. The production of no-carrier-added 161 Tb has been developed to a quality that is comparable to that of no-carrier-added 177 Lu, enabling efficient radiolabeling of biomolecules under the same experimental conditions. Attempts to label PSMA-617 with 161 Tb at specific activities up to 100 MBq/nmol resulted in radiochemically pure 161 Tb-PSMA-617 (> 98%). The radiolytic degradation of 161 Tb-PSMA-617 was similar to 177 Lu-PSMA-617, indicating that the emitted conversion and Auger electrons did not play a critical role with regard to the radioligand's stability.
In agreement with previously-performed studies that compared 161 Tb-and 177 Lu-folate conjugates [29], the in-vitro properties of 161 Tb-PSMA-617 and 177 Lu-PSMA-617 were largely the same. This included the n-octanol/PBS distribution coefficient and cell uptake and internalization in PSMA-positive and PSMA-negative tumor cells. It was also confirmed that the pharmacokinetics of 161 Tb-PSMA-617 was equal to 177 Lu-PSMA-617, resulting in the same biodistribution profiles as expected (Fig. 2). It is likely that these findings can be extrapolated to  Lu became obvious from in-vitro data where the exposure to 161 Tb-PSMA-617 reduced the viability and survival of PC-3 PIP tumor cells in an activity-dependent manner. In agreement with dosimetric calculations, 161 Tb-PSMA-617 was up to 3-fold more effective than 177 Lu-PSMA-617 in vitro. This difference in efficacy of 161 Tb-PSMA-617 and 177 Lu-PSMA-617 was not observed when using PSMA-negative PC-3 flu cells or when PC-3 PIP cells were exposed to the DTPAcomplexes of the two radionuclides. These findings confirmed that the observed advantage of using 161 Tb-PSMA-617 over 177 Lu-PSMA-617 is dependent on PSMA binding and internalization. The in-vitro findings also corroborated previous invitro findings, where 161 Tb-folate was more effective in reducing KB tumor cell viability than 177 Lu-folate [29].
The treatment of PC-3 PIP tumor-bearing mice with 5.0 MBq and 10 MBq 161 Tb-PSMA-617, respectively, showed an activity-dependent tumor growth inhibition and prolonged survival of mice. When 161 Tb-PSMA-617 was applied at 10 MBq, the tumor xenografts disappeared entirely in two out of six mice, which were still alive at study-end after 12 weeks. As no signs of undesired side-effects were detectable, higher activities may be used to treat the tumors more effectively. The tumor growth inhibition and median survival (TGDI 2 = 4.2 ± 1.2; 36 days; Table 2) of mice that received 5.0 MBq 161 Tb-PSMA-617 indicated better therapy response that that achieved in previously-reported results obtained with 5.0 MBq 177 Lu-PSMA-617 (TGDI 2 = 2.1 ± 0.3, median survival: 32 days [34]). Individual mice treated with 5.0 MBq 161 Tb-PSMA-617 revealed a heterogeneous response pattern, where the last mouse reached the endpoint at Day 66. In contrast, the use of 177 Lu-PSMA-617 therapy resulted in the last mouse to be euthanized at Day 40 [34].
In additional experiments, we simulated the situation of tumor cells in vivo that have not yet grown to a tissue, in order to investigate whether the radioligands delayed the formation of solid tumors (Supplementary material). At that time, a vascularized tissue was not yet developed, and the measurable "swelling" could presumably be ascribed to the formation of a tumor cell cluster. When applied at activities of 2.5 MBq, 5.0 MBq, or 10 MBq, the effect of 161 Tb-PSMA-617 was enhanced when compared to that of 177 Lu-PSMA-617, and re-growth of already disappeared tumors was less frequent when using 161 Tb-PSMA-617 (Supplementary material; Fig.  S4; Table S8). These results confirmed the anticipated improved effect of 161 Tb over 177 Lu also at the level of single cancer cells or cancer cell clusters in vivo.
In line with these results, the dosimetry analysis revealed that 161 Tb has a 1.4-fold higher energy deposition in established tumors compared to 177 Lu. This ratio increases to about 4-fold for small cell clusters and single cells. Together with the biological results obtained in this study, the dosimetry confirms that 161 Tb may be better suited than 177 Lu for sterilizing small cell clusters in advanced metastatic prostate cancer with radiolabeled PSMA ligands.
To date, it remains unclear to what extent the design of the targeting ligand could contribute to fully exploiting the decay properties of 161 Tb. It has been stated in literature that nuclear localization is necessary to obtain effective Auger electron therapy [35][36][37][38]. In the case of 161 Tb, the additional effect is, however, given predominantly by the emission of conversion electrons of an energy and tissue range comparable to β¯particles of lowest energy. Hence, even when neglecting Auger electrons, the absorbed dose of 161 Tb is still superior to that of 177 Lu due to more emitted electrons per decay. It remains to be investigated whether PSMA ligands, comprising a nuclear localizing signal for effective delivery of the radionuclide to the cell nucleus, would improve the effect of 161 Tb further by also making full use of the emitted Auger electrons. More sophisticated ligand designs and more clinically-relevant mouse models for testing the effects will be the topic of future preclinical studies to obtain answers to these open questions.

161
Tb was used for the first time with a PSMA ligand, which demonstrated better results than 177 Lu-PSMA-617 in vitro and in vivo. Based on these findings, the postulated superiority of 161 Tb over 177 Lu was corroborated. Our preclinical research activities will be continued to further investigate 161 Tb, as we intend to translate it to clinics and provide prostate cancer patients with an optimized treatment option in the near future.