Fibroblast activation protein-targeted radionuclide therapy: background, opportunities, and challenges of first (pre)clinical studies

Introduction Fibroblast activation protein (FAP) is highly overexpressed in stromal tissue of various cancers. While FAP has been recognized as a potential diagnostic or therapeutic cancer target for decades, the surge of radiolabeled FAP-targeting molecules has the potential to revolutionize its perspective. It is presently hypothesized that FAP targeted radioligand therapy (TRT) may become a novel treatment for various types of cancer. To date, several preclinical and case series have been reported on FAP TRT using varying compounds and showing effective and tolerant results in advanced cancer patients. Here, we review the current (pre)clinical data on FAP TRT and discuss its perspective towards broader clinical implementation. Methods A PubMed search was performed to identify all FAP tracers used for TRT. Both preclinical and clinical studies were included if they reported on dosimetry, treatment response or adverse events. The last search was performed on July 22 2022. In addition, a database search was performed on clinical trial registries (date 15th of July 2022) to search for prospective trials on FAP TRT. Results In total, 35 papers were identified that were related to FAP TRT. This resulted in the inclusion of the following tracers for review: FAPI-04, FAPI-46, FAP-2286, SA.FAP, ND-bisFAPI, PNT6555, TEFAPI-06/07, FAPI-C12/C16, and FSDD. Conclusion To date, data was reported on more than 100 patients that were treated with different FAP targeted radionuclide therapies such as [177Lu]Lu-FAPI-04, [90Y]Y-FAPI-46, [177Lu]Lu-FAP-2286, [177Lu]Lu-DOTA.SA.FAPI and [177Lu]Lu-DOTAGA.(SA.FAPi)2. In these studies, FAP targeted radionuclide therapy has resulted in objective responses in difficult to treat end stage cancer patients with manageable adverse events. Although no prospective data is yet available, these early data encourages further research. Supplementary Information The online version contains supplementary material available at 10.1007/s00259-023-06144-0.


Introduction
According to Global Cancer Statistics, there were roughly 19.3 million new cancer diagnoses and 10.0 million cancer deaths globally in 2020 [1]. Cancers develop in complex environments composed of tumor cells and the surrounding stroma. As early as 1889, the "seed and soil" idea highlighted the interdependent relevance of both elements [2]. The majority of diagnostic and therapeutic approaches have focused on tumor cells. However, the tumor microenvironment (TME) is receiving incremental attention.
The TME consists of immune cells, vasculature, extracellular matrix, and cancer-associated fibroblasts (CAF). Treatment failure can occur as a result of the development of an immunosuppressive TME that shields tumor cells from This article is part of the Topical Collection on Theragnostic.
Bastiaan M. Privé and Mohamed A. Boussihmad shared first author. therapeutic agents. In solid tumors, CAFs are one of the TMEs most prevalent components [3]. However, CAFs are heterogeneous cells, and various conditions have been shown to result in both tumor-promoting and tumor-suppressive activities [4]. They are able to remodel the ECM structure, thereby contributing to tumor initiation, neovascularization, and metastasis. This can either act as a physical barrier to prevent the infiltration of immune cells or as a structural scaffold for intercellular interaction between tumor cells and nontumor cells in the TME [5]. Adversely, CAFs can release a variety of chemokines and cytokines, including interleukin-6 (IL-6), CC-chemokine ligand 2, and transforming growth factor (TGF), in order to attract inhibitory immune cell subsets in the tumor stroma, hence promoting immune evasion [6]. Platelet-derived growth factor receptor-β (PDGFR-β), α-smooth muscle actin (α-SMA), and fibroblast activation protein (FAP) are only a few of the biological markers found in CAFs. FAP is a type II integral membrane glycoprotein from the serine protease family that has a role in fibrogenesis and ECM remodeling. Because of the wide distribution of FAP in numerous cancer types (e.g., sarcoma, prostate cancer, breast cancer, lung cancer, pancreatic cancer, head and neck cancer, colorectal cancer) the question arises whether FAP can be a novel therapeutic target in cancer (Fig. 1).
Despite the literature's inconsistent findings about the prognostic significance of FAP in malignancies, high expression of FAP has been demonstrated to be an independent poor prognostic predictor for outcomes in lung cancer, hepatocellular carcinoma, and colon cancer [7,8]. FAP overexpression has been shown to enhance tumor development in a mouse model which may afterwards be decreased by anti-FAP antibodies [9]. However, sibrotuzumab (a humanized version of the mouse anti-FAP antibody) was ineffective as a therapy for metastatic colorectal cancer in an early phase II study [10]. Yet, the field of nuclear medicine has recently seen an increase in interest of radiolabeled small molecules that target FAP. It has been hypothesized that positron emission tomography (PET) of radiolabeled FAP inhibitors (FAPI) with short-living positron emitters (e.g., 68 Ga/ 18 F) may replace 18 F-FDG in several major tumor types (e.g., breast cancer, pancreatic cancer, sarcoma) [3,11]. Moreover, due to the high tumor uptake of FAP tracers and low expression of FAP in healthy tissues, FAP is deemed as a promising target for FAP TRT. As response to TRT is a result of the cytotoxic effect of radiation and not to a direct inhibitory effect of a FAP-binding molecule, the earlier reports on anti-FAP antibodies (e.g., sibrotuzumab) does not apply using this approach. While the FAP-targeting agent ensures uptake in the tumor stroma, the cross-fire effect of the radiolabeled (alpha or beta emitting) radionuclides deliver tumoricidal doses to the ECM including the cancer cells. Currently, several clinical trials are investigating 177 Lu, 90 Y, and 225 Ac-labeled FAP-targeting tracers (e.g., NCT04939610, NCT04849247, ACTRN12621000935831), and several small case series have been published to date. Therefore, we aimed to review the presently reported results on therapeutic FAP ligands for TRT and provide an overview of the active studies. Finally, potential opportunities and challenges are discussed.

Search strategy
The selection of FAP ligand types for this review was based on a combination of electronic library searches. First, the different FAP ligands used for TRT were analyzed. Second, literature was searched by universal Pub-Med searches (see Supplementary Table 1)  For preclinical studies, data was reported on the percentage of injected dose per gram (%ID/g), dosimetry, and treatment response. Retrospective as well as prospective data of clinical studies were included if they reported on either dosimetry, treatment response, or adverse events. FAP imaging studies were excluded from this report.
A database search was performed on clinical trial registries such as clinicaltrials.gov (date 15 of July 2022) to search for active prospective trials using the terms: fibroblast activation protein, cancer-associated fibroblasts, FAP, and FAPI.

mAbF19
The first FAP TRT study was published in 1994. In this study, Welt et al. described the toxicity, imaging and biodistribution characteristics of 131 I (beta decay, t 1/2 8 days) labeled with the monoclonal antibody F19 ([ 131 I]I-mAbF19) [12].  [12,13]. Therefore, FAP TRT was rested, until the last few years when novel small-molecules targeting FAP were developed.

FAPI-04
Preclinical studies The first study on FAP inhibitors (FAPI) was published in 2018 by Lindner and colleagues. In this study, BALB/c nu/ nu mice were inoculated with HT-1080 cells transfected with FAP. When the tumor had grown to approximately 1 cm 3 , several different [ 177 Lu]Lu-labeled FAPIs were injected via the tail vein. In this study, FAPI-04 was the best candidate for TRT as it showed high tumor accumulation, slower excretion, and no significant increase in background activity. While [ 177 Lu]Lu-FAPI-13 showed higher tumor uptake leading to higher doses, the retention of the tracer in the blood was also higher. It was hypothesized that this may lead to higher bone marrow toxicity during treatment. Therefore, FAPI-04 was chosen as the preferred theranostic tracer despite having a rather short tumor retention time. The authors suggest using FAPI-04 with a radionuclide that follows this shorter retention time (e.g., 153 SM, beta decay, and t 1/2 of 1.92 days or 90 Y, beta decay, and t 1/2 of 2.66 days).

Preclinical studies
In recent years, the group of the University of Heidelberg (Germany) studied several FAPI compounds in the search for a FAP tracer with the best properties for TRT. FAPI-46 proved to be most suitable as a theranostic agent due to a longer tumor retention (compared with FAPI-04) while having a similar uptake in healthy tissues [19].
The therapeutic response of [ 177

Clinical data
The first clinical data in humans on FAPI-46 was published in 2020. This study reported on a patient with metastasized ovarian cancer and a patient with pancreatic cancer that both received 6 GBq [ 90 Y]Y-FAPI-46 as a last-line treatment. Significant tumor uptake was seen, with low uptake in healthy organs. No follow-up data was reported  [24]. Eighteen patients (ovarian cancer (n = 2), sarcoma (n = 1), colon cancer (n = 3), breast cancer (n = 5), pancreatic cancer (n = 2), prostate cancer (n = 1), cervical cancer (n = 1), round-cell tumor (n = 1), lung cancer (n = 1), and anaplastic thyroid cancer (n = 1)) were treated with doses of 1.  [28]. No therapyrelated AEs were observed. Eight weeks after TRT, imaging demonstrated a mixed response. The patient had an OS of 3 months.
More recently, a larger case series was published using [ 90 Y]Y-FAPI-46 and reported on its safety, dosimetry, and viability [29]. Between June 2020 and March 2021, nine patients underwent treatment for pancreatic cancer (n = 3) or metastatic soft-tissue or bone sarcoma (n = 6). All patients received a median activity of 3.

Preclinical trials
The first preclinical data of FAP-2286 was published in 2020 by Zboralski et al. [31]. Following intravenous injection, in vivo biodistribution in mice showed renal excretion of FAP-2286, little uptake in normal tissues, and fast absorption of 111 In and [ 177

Preclinical data
In 2020, Moon et al. presented the first data on DOTA. SA.FAPi. This tracer was labelled with 68 Ga and 177 Lu with high radiochemical yield (> 97%) [35].

Clinical data
In September 2019, a 31-year-old woman was found to have advanced metastatic breast cancer (ER-PR-HER2neu+) [36]. On a compassionate use basis, the patient was given a single cycle of 3.2 GBq [ 177 [37]. No proper treatment response data was presented in this study. However, patients that received [ 177 2 seemed safe, appears to be effective, and may provide a new therapeutic option for patients with aggressive radioiodine-refractory thyroid cancer who failed on all conventional therapy. A short while later, a case report emerged which described a 56-year-old man with aggressive medullary thyroid carcinoma, behaving clinically like anaplastic thyroid cancer. He was treated with one cycle of 1.65 GBq [ 177

ND-bisFAPI
There is one study on ND-bisFAPI [40]. The objective of this work was to create a bivalent FAP ligand that might be used for TRT as well as diagnostic PET imaging. In vitro studies of competitive binding to FAP, cellular internalization, and efflux characteristics were determined using FAP-positive A549-FAP cells and a murine tumor model. ND-bisFAPI demonstrated selective uptake, a high internalized proportion, and a delayed cellular efflux in A549-FAP cells. At 24, 72, 120, and 168 h, biodistribution analyses revealed that [ 177

PNT6555
PNT6555 is a novel FAP-targeting ligand. There is one conference abstract which described preclinical data of PNT6555 [41]. PNT6555 was labeled with 68 Ga and 177 Lu which showed potent activity in FAP inhibition assays using human, mouse, and rat sources of FAP. Direct organ analysis revealed little [ 177 Lu]Lu-PNT6555 accumulation and retention in healthy tissues while revealing significant tumor retention up to 168 h (> 10 %ID/g). Importantly, [ 177

EB-FAPI
FAP-targeted molecular imaging radiotracers have shown promising results; however, rapid clearance and inadequate tumor retention seem to be a recurring limitation. In order to overcome these limitations, FAPI tracers with albuminbinding moieties were developed. A group conjugated Evans Blue to the FAPI complex [42,43]. Several EB-FAPI-Bn were synthesized, which were based on FAPI-02, and labeled with 177  Both of these studies started in June 2022. The first study (NCT05400967) will be performed in patients with advanced cancer and will investigate dosimetry of [ 177 Lu] Lu-EB-FAPI following an injection of 1.11 GBq. Moreover, they will also assess the safety and therapeutic response. In a second study NCT05410821, 20 radioactive iodine refractory thyroid cancer patients will be treated with up to 3 cycles of 1.11, 2.22, or 3.33 GBq [ 177 Lu]Lu-DOTA-EB-FAPI. The aim of this phase II study is to assess safety, tolerability, and overall response rate.

FSDD
FSDD are FAP tracers with an albumin-binding moiety. In this study, FSDD 0 I, FSDD 1 I, and FSDD 3 I, were labeled with 68 Ga and 177 Lu and tested in human hepatocellular carcinoma patient-derived xenografts (HCC-PDXs) [46]. PET imaging showed that [ 68 Ga]Ga-FSDD 0 I had a better blood retention and tumor uptake compared with [ 68 Ga]Ga-FAPI-04, [ 68 Ga]Ga-FSDD 1 I, and [ 68 Ga]Ga-FSDD 3 I, and was therefore selected for further studying. For biodistribution studies, FSDD 0 I was labeled with 177 Lu. At 1-h post injection, [ 177 Lu]Lu-FSDD0I had high accumulated in the tumor (14.44 ± 1.174 %ID/g), bone (12.11 ± 2.83 %ID/g), and blood (10.30 ± 2.03 %ID/g). The authors postulate that the high bone uptake may be related to the physiological expression of FAP in murine osteoblasts and bone marrow stem cells. At the latest time point (48-h post injection), the ID%/g in tumor was ~ 5. The kidney uptake at 1-and 48-h post injection was ~ 5 ID%/g and ~ 1 ID%/g, respectively.
Because of this, the authors will select FSDD01 for their following studies. Hitherto, no prospective study is registered.

Discussion
This review provides an overview of the currently reported therapeutic tracers that target FAP for TRT. FAP is overexpressed by cancer-associated fibroblasts of several tumor types [47][48][49]. While the first outcomes of prospective trials of FAP TRT are still awaited, preclinical studies and selected case series showed encouraging data that FAP TRT may become a treatment option for different cancer types. Moreover, in current studies, FAP TRT seemed well tolerated, which corroborates with FDA-approved TRT in patients with neuro-endocrine tumors ([ 177 [30,33]. In these end-stage patients, there is an unmet need for novel anti-cancer therapies. However, the present studies had a retrospective study design with potential selection bias to more fit patients. Moreover, most studies had suboptimal protocols (e.g., dosimetry from whole-body scans using minimal time points, non-randomized, no systematic reporting on AE or outcomes) with short follow-up which may have skewed the outcomes. Importantly, while it is anticipated that most of these malignancies are sensitive to radiation as external beam radiotherapy generally result in a response, it is unknown whether alfa or beta radiation will also result in tumoricidal effects. Therefore, the present reports only suggest that FAP TRT may have efficacy in advanced cancer patients, and results of the first prospective studies are warranted.
To date, the reported side effects show a favorable toxicity profile with limited and manageable high-grade adverse events. The most important reported adverse events are related to bone marrow toxicity (anemia, leukopenia, and reduced number of platelets). This is in concordance to known side effects of [ 177 Lu]Lu-PSMA and [ 177 Lu]Lu-DOTATATE. At present, it is not feasible to determine which FAP inhibitor has the best toxicity profile, because of the small cohorts and lack of prospective data with systematic reporting on AEs. Yet, the present real-life data suggest that toxicity is low which supports further studying.
Although studies mostly relied on whole-body scans with limited number scanning of time points, there is dosimetry data of FAP TRT for tumors and the organs at risk (e.g., bone marrow and kidneys). The absorbed dose to the tumor ranges from 0.62 ± 0.55 Gy/GBq, 2.81 ± 1.25 Gy/Gbq, 3.0 ± 2.7 Gy/GBq, and 6.70 (IQR: 3.40-49) Gy/GBq, for [ 177  showing uptake up to 168-h post injection. However, this also seemed to increase the dose to the organs at risk. Other investigators tried to overcome the issue of short tumor retention by labelling it (e.g., FAPI-46) with 90 Y, which has a shorter T 1/2 of 2.7 days vs 6.6 days of 177 Lu. Therefore, the physical half-life of the isotope better matched the biological half-life of the tracer. Future studies will tell whether FAP tracers with prolonged retention time or FAP tracers with faster pharmacokinetics and shorter living radionuclides end up in better results. Moreover, we need to identify the malignancies that have good expression of FAP and are sensitive to alfa or beta radiation. Hence, most tumors show FAP expression in the TME, while some cancers actually express FAP on their cellular membrane (e.g., sarcoma, certain ovarian, and pancreatic cancers) [53]. Therefore, basket studies including different malignancies are awaited. On the other hand, this broad applicability is also its main opportunity with its potential use in a whole array of cancers. It is also postulated that FAP TRT is particularly useful in combination with other anti-cancer treatments. Here, FAP TRT could increase the permeability of the TME resulting in higher efficacy of concomitant chemo-and immunotherapies. All in all, we now await the first prospective phase I/II data of FAP TRT which are anticipated in 2023.

Conclusion
The long circulation time of FAP-targeting antibodies and poor tumor retention time of FAP-targeting small molecules used to be a matter of concern. However, several teams have developed methods to overcome this limitation by labeling small molecules that target FAP to radionuclides with shorter half-lives. Moreover, there is incremental reporting of novel tracers that show tumor uptake up to 168-h post injection. To date, data was reported on more than 100 patients that were treated with different FAP-targeted radionuclide therapies such as [ 177 2 . In these case series, FAP-targeted radionuclide therapy has resulted in objective responses in difficult to treat end-stage cancer patients with manageable adverse events. Although no prospective study data is yet available, the early data encourages further studying. For now, we await the first prospective phase I/II data of FAP TRT.
Author contribution All authors were involved in writing and reviewing of the manuscript. BP and MB were involved in the study design, paper analysis, and manuscript design.
Data availability Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Competing interests The authors declare no competing interests.
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/.