Fibroblast activation protein targeted therapy using [177Lu]FAPI-46 compared with [225Ac]FAPI-46 in a pancreatic cancer model

Purpose Fibroblast activation protein (FAP), which has high expression in cancer-associated fibroblasts of epithelial cancers, can be used as a theranostic target. Our previous study used 64Cu and 225Ac-labelled FAP inhibitors (FAPI-04) for a FAP-expressing pancreatic cancer xenograft imaging and therapy. However, the optimal therapeutic radionuclide for FAPI needs to be investigated further. In this study, we evaluated the therapeutic effects of beta-emitter (177Lu)-labelled FAPI-46 and alpha-emitter (225Ac)-labelled FAPI-46 in pancreatic cancer models. Methods PET scans (1 h post injection) were acquired in PANC-1 xenograft mice (n = 9) after the administration of [18F]FAPI-74 (12.4 ± 1.7 MBq) for the companion imaging. The biodistribution of [177Lu]FAPI-46 and [225Ac]FAPI-46 were evaluated in the xenograft model (total n = 12). For the determination of treatment effects, [177Lu]FAPI-46 and [225Ac]FAPI-46 were injected into PANC-1 xenograft mice at different doses: 3 MBq (n = 6), 10 MBq (n = 6), 30 MBq (n = 6), control (n = 4) for [177Lu]FAPI-46, and 3 kBq (n = 3), 10 kBq (n = 2), 30 kBq (n = 6), control (n = 7) for [225Ac]FAPI-46. Tumour sizes and body weights were followed. Results [18F]FAPI-74 showed rapid clearance by the kidneys and high accumulation in the tumour and intestine 1 h after administration. [177Lu]FAPI-46 and [225Ac]FAPI-46 also showed rapid clearance by the kidneys and relatively high accumulation in the tumour at 3 h. Both [177Lu]FAPI-46 and [225Ac]FAPI-46 showed tumour-suppressive effects, with a mild decrease in body weight. The treatment effects of [177Lu]FAPI-46 were relatively slow but lasted longer than those of [225Ac]FAPI-46. Conclusion This study suggested the possible application of FAPI radioligand therapy in FAP-expressing pancreatic cancer. Further evaluation is necessary to find the best radionuclide with shorter half-life, as well as the combination with therapies targeting tumour cells directly. Supplementary Information The online version contains supplementary material available at 10.1007/s00259-021-05554-2.


Introduction
The stroma, which comprises up to 90% of tumour mass, promotes tumour growth, migration, and progression. The fibroblast activation protein (FAP) is highly expressed in cancer-associated fibroblasts (CAFs) of the stroma of many epithelial cancers and is associated with poor prognosis [1][2][3]. In contrast, low FAP expression is found in normal tissues. Therefore, FAP is an excellent target for the imaging and therapy. FAP inhibitors (FAPI) are used for theranostics in oncology [4][5][6]. In previous studies, [ 68 Ga]-labelled FAPI positron emission tomography (PET)/computed tomography (CT) were proven to be effective in the clinical diagnostics of various cancers [7][8][9]. [ 99m Tc]-labelled FAPI derivatives were also synthesized successfully for single photon emission computed tomography imaging [6]. However, reports regarding the therapeutic applications of FAPI are relatively limited. Lindner et al. used [ 90 Y]FAPI-04 for the targeted therapy in a breast cancer patient resulting in pain reduction [10]. Our previous study labelled FAPI-04 with 225 Ac, an alpha particle emitter with a half-life of 10 days for its first decay, and investigated the therapeutic effects of [ 225 Ac] FAPI-04 in FAP-expressing human pancreatic cancer [11]. However, FAPI showed rapid excretion via the kidneys, and its biological half-life did not match the physical half-life of 225 Ac. Therefore, it is necessary to compare the therapeutic effects of FAPI with improved tumour retention, to investigate a better combination of its kinetics and physical decay. In this study, we used 177 Lu, a beta emitter with a half-life of 6.7 days, and 225 Ac to label FAPI-46 for the targeted therapy and [ 18

Preparation of [ 18 F]FAPI-74, [ 177 Lu], and [ 225 Ac] FAPI-46 solutions
The precursor molecules of FAPI-46 and FAPI-74 were obtained from the Heidelberg University based on a material transfer agreement for collaborative research. [ 18 F]FAPI-74 was produced following the methods of previous reports [12]. [  [ 225 Ac]labelled FAPI-46 was prepared as per the method provided in a previous paper [11]. A mixture of 30 µL of 1 mM FAPI-46, 100 µL of 0.2 M ammonium acetate, 100 µL of 7% sodium ascorbate, and 200 µL of 225 Ac solution (300 kBq) was obtained at 80℃ after 2 h. The solution was diluted with 0.9% saline, and was used for the animal studies without further purification. The chemical structure of [ 225 Ac]FAPI-46 is shown in Sup Fig. 1c.

Preparation of the animals
The human pancreatic cell line, PANC-1, was obtained from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in RPMI1640 medium with l-glutamine and Phenol Red (FUJIFILM Wako Pure Chemical, Osaka, Japan), supplemented with 10% heat-inactivated fetal bovine serum and 1% penicillin-streptomycin.
Male nude mice (BALB/cSlc-nu/nu) were purchased from Japan SLC Inc. (Hamamatsu, Japan). Animals were housed under a 12-h light/12-h dark cycle and allowed free access to food and water. The mice were injected with PANC-1 cells (1 × 10 7 cells) in a mixture of phosphatebuffered saline and Matrigel (0.1 mL, 1:1; BD Biosciences, Franklin Lakes, NJ, USA). Tumour xenograft models were evaluated 3 weeks after transplantation of PANC-1 cells (tumour volume = 716 ± 304 mm 3 ). Euthanasia was performed in the following conditions: (1) when the animals had unbearable suffering, (2) when a significant decrease in activity or a marked decrease in food and water intake was observed, and (3) at the end of the observation period (up to 44 days for [ 177 Lu]FAPI-46 and 32 days for [ 225 Ac] FAPI-46). Euthanasia was performed by deep anesthesia using isoflurane inhalation.

[ 18 F]FAPI-74 PET imaging and analysis
PET images were acquired with a small animal PET scanner (Siemens Inveon PET/CT) 3 weeks after the implantation in PANC-1 xenograft mice (9 weeks old, body weight = 25.3 ± 1.2 g, n = 9). Under 2% isoflurane anesthesia, [ 18 F]FAPI-74 (12.4 ± 1.7 MBq) was injected in the tail vein. Dynamic PET scans (scan duration = 70 min, n = 2) were started simultaneously with the bolus injection. Static PET scans (scan duration = 10 min, n = 7) were performed 1 h after injection, followed by a CT scan. PET data were reconstructed into 2-min frames in the dynamic PET scan (2 min × 35 frames) and one frame in the static PET scan by three-dimensional ordered-subset expectation-maximization (16 subsets, 2 iterations), with attenuation and scatter correction. Regions of interest were drawn on the muscle, heart, lungs, liver, gallbladder, kidneys, intestine, and tumour. The mean standardized uptake values (SUVmean) and maximum standardized uptake values (SUVmax) were measured using PMOD (Version 4.0).

Immunohistochemistry and histological analysis
All mice were killed after [ 18 F]FAPI-74 PET imaging, and tumour xenografts were removed. Immunohistochemical staining was performed using anti-FAP alpha antibody (ab53066; Abcam, Cambridge, UK), and the Dako EnVision + System-HRP Labelled Polymer Anti-Rabbit (K4003) (DAKO Corp., Glostrup, Denmark). To evaluate toxicity, the kidneys were removed after the mice treated with [ 177 Lu] FAPI-46 and [ 225 Ac]FAPI-46 were sacrificed. The tissues were fixed in 10% neutral buffered formalin solution for paraffin blocks and stained with hematoxylin and eosin (H&E). Tumour blocks in all mice were also stained with H&E.

Statistical analysis
Data were expressed as the mean ± standard deviation. Comparisons among the four groups were performed using an unpaired t test in Microsoft Excel (version 2016) with Bonferroni correction, and p < 0.05 were considered statistically significant.

Results
The time-activity curve of the PANC-1 tumour and the normal organs on [ 18 F]FAPI-74 PET are shown in Fig. 1a. [ 18 F] FAPI-74 was cleared rapidly by the kidneys but washout from the tumour occurred slowly. A static PET image is shown in Fig. 1b. The SUVmean of static scans were 0.24 ± 0.04 in the tumour, 0.05 ± 0.01 in the muscle, 0.08 ± 0.01 in the heart, 0.14 ± 0.02 in the liver, 0.66 ± 0.15 in the gallbladder, 0.61 ± 0.48 in the intestine, and 0.39 ± 0.07 in the kidneys (Fig. 1c). The accumulation in the tumour was significantly higher than in most organs at 1 h post-injection. Immunohistochemical staining showed FAP expression in the stroma of PANC-1 xenografts (Fig. 2).
The MBq groups showed a slight decrease without statistical significance, compared to the controls (Fig. 4b).
The results after the administration of [ 225 Ac]FAPI-46 are shown in Fig. 5 and Sup Fig. 2b. The tumour growth was suppressed immediately after treatment in the 10 kBq and 30 kBq groups, while the tumour-suppressive effects in the 3 kBq group were very mild. The tumour size of the 30 kBq groups was significantly smaller than those in the control group on days 5-9 and day 25. The body weight in all the groups showed a decreasing trend in the first week while the 3 kBq and 10 kBq groups showed recovery after day 7.
H&E staining of the tumours and kidneys are shown in Fig. 6  We demonstrated the effectiveness of alpha therapy for FAP-expressing pancreatic cancer using [ 225 Ac]FAPI-04 in a previous study [11]. [ 225 Ac]FAPI-04 was thought to irradiate tumour cells by the alpha particles emitted from CAFs in the stroma. However, the alpha irradiation also has affects on CAFs, the primary site of accumulation, which are supporting tumour progression. Since beta particles have a more extended range in tissue compared to alpha particles, beta irradiation may reach tumour cells more homogeneously compared to alpha irradiation. Thus, we used FAPI-46 labelled with 177 Lu, a beta emitter, for PANC-1 xenograft mice in the present study. Previous studies reported a rapid internalization of [ 177 Lu]-labelled FAPI derivatives into HT-1080-FAP cells [4] and a high uptake in HT-1080-FAP tumour-bearing mice [10,13]. In the present study, we also found a relatively high accumulation of [ 177 Lu]FAPI-46 in PANC-1 xenografts, which is considered to target FAP mainly expressed in the stroma.
In the present study, we found that  immediately after administering a high dose of [ 225 Ac]FAPI-46, while regrowth began at day 12 with the same tumour growth speed as in the control group. However, in a previous study, 225 Ac showed a lower survival rate of cells compared to cells treated with 177 Lu [15], according to more fatal double-strand breaks (DSBs) induced by alpha particles [16,17]. Meanwhile, [ 225 Ac]PSMA-617 was effective in metastatic prostate cancer patients refractory to [ 177 Lu]PSMA-617 [18,19]. We speculated that the reason for the difference seen in our study was due to the fact that the target cells of [ 177 Lu] FAPI-46 and [ 225 Ac]FAPI-46 were CAFs in the stroma as opposed to tumour cells. Stroma cells can tolerate a more fatal environment than other cells and are more radioresistant [20,21]. However, the effects of alpha irradiation on tumour stromal cells remain to be clarified. Due to a heterogeneous distribution of the stroma and tumour cells causing a heterogeneous dose distribution, it might be difficult for alpha particles to reach the tumour cells sufficiently. The tumour cells irradiated by 225 Ac caused death due to DSBs, while the tumour cells without irradiation survived and recovered. In contrast, the tumour cells are more likely to be irradiated by beta emission from [ 177 Lu]FAPI-46 but with lower cell-killing properties. In the present study, the marginal superiority of [ 177 Lu]FAPI-46 was observed compared to [ 225 Ac]FAPI-46 and possibly due to the wide effective area by cross-fire effect of beta emission from 177 Lu accompanied with bystander effects as well as the inefficient energy transfer by alpha emission of 225 Ac from the stroma.
In the present study, the therapeutic effects of [ 177 Lu]FAPI-46 and [ 225 Ac]FAPI-46 were rather limited, with some of the tumour-suppressive effects being not significant compared to the control group. When we compare these therapeutic effects with previous reports using other compounds with a similar administered dose, such as [ 177 Lu]DOTATATE for neuroendocrine tumour xenografts, and [ 177 Lu]PSMA-617 for prostate cancer, the anti-tumour effects were inferior in our study [22][23][24]. A characteristic of FAPI is its quick distribution, but retention in the tumour was inferior to that of other compounds. Lindner et al. developed the FAPI compounds with improved retention, and FAPI-46 shows a better retention compared to FAPI-04 [13]. However, dramatic improvement of retention is not an easy task and the biological half-life of FAPI is short, compared to the long physical half-life of 177 Lu and 225 Ac. Thus, the possible strategy is to improve treatment effect is injecting high radioactivity with shorter half-life radionuclide. Radionuclides with a shorter half-life, such as 188 Re (half-life = 17.0 h) or 211 At (halflife = 7.2 h) that reaches tumour with high radioactivity at an early time of administrations, maybe optimal for FAPI therapy by increasing the local dose. Another alpha therapy targeting cancer-specific LAT1, which also showed fast clearance through urine, chose short-half-life radionuclide 211 At for labelling [25].
Although the procedure for labelling FAPI with 211 At has not been established yet, [ 188 Re]-labelled FAPI was synthesized successfully recently and administrated clinically [6].  [11]. Considering about the potential side effects, we set the maximum administration activity of [ 225 Ac]FAPI-46 as 30 kBq per mouse in the present study.
[ 18 F]FAPI-74 showed a high uptake in the joints, and similar uptakes in the joints and bone were reported in the use of [ 18 F] FGlc-FAPI in FAP-expressing xenograft models [5]. They also reported that possible specific binding in the joints and bones by blocking experiments although they are reported to be low in [ 68 Ga]FAPI-04 PET. [ 177 Lu]FAPI-46 and [ 225 Ac]FAPI-46 also showed relatively high uptake in the bone. In the present study, we assumed that these uptakes in mice may be due to the binding of radio-labelled FAPIs to the protein in the murine synovial fluid in the joints, since no high uptake was found in human joints [12].
In the present study, [ 225 Ac]FAPI-46 showed high accumulation in the liver, whereas the uptake of [ 177 Lu]FAPI-46 in the liver was low. A previous study also reported an increased accumulation of [ 225 Ac]DOTATOC in the liver compared to [ 177 Lu] DOTATOC [26]. The difference was thought to be due to the distribution of free 225 Ac since a high uptake of released 225 Ac in the liver was found in mice [27], suggesting better in vivo stability of [ 177 Lu]FAPI-46.
A slight decrease of body weight in control group was found, and it may due to the stress caused by the change of housing conditions. Both mice in [ 177 Lu]FAPI-46 and [ 225 Ac]FAPI-46 groups also showed a decrease of body weight, which is a sign of worry for potential clinical application. Extended single-dose toxicity study in normal mice should be performed in the future study to evaluate the possible side effects [28]. Renal   This study had several limitations. First, we used only one cell line, PANC-1, for the evaluation. However, stroma formation may be different from the tumour stroma in the patients. Therefore, patient-derived xenograft (PDX) models or other non-PDX models with different FAP expression should be used in future work for the better clinical translation. Second, the sample size of [ 225 Ac]FAPI-46 was insufficient because of the limited supply of 225 Ac. Third, we did not determine the maximum tolerated dose (MTD) of [ 177 Lu]FAPI-46 and [ 225 Ac]FAPI-46 in PANC-1 model, which helps to determine therapeutic window in the future clinical applications. We need to perform the toxicity study using normal mice to evaluate the possible side effects to determine the MTD in mice [28].

Conclusion
This study revealed therapeutic effects of [ 177 Lu]FAPI-46 and [ 225 Ac]FAPI-46 in PANC-1 xenografts, while the impact of [ 177 Lu]FAPI-46 appeared slow but lasted longer. Beta therapy and alpha therapy targeting FAP can be a potential treatment for pancreatic cancers and needs further evaluation to find the best combination of fast FAP kinetics and physical decay of the radionuclide as well as the combination with therapies targeting tumour cells.

Declarations
Ethics approval All experiments were performed in compliance with the guidelines of the Institute of Experimental Animal Sciences. The protocol was approved by the Animal Care and Use Committee of the Osaka University Graduate School of Medicine.

Consent for publication Not applicable.
Competing interests TL, UH, CK, FLG have a patent application for FAPI-ligands. TL, UH, CK, FLG also hold shares of a consultancy for iTheranostics.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.