**Preclinical Incorporation Dosimetry of [18F]FACH—A Novel 18F-Labeled MCT1**/**MCT4 Lactate Transporter Inhibitor for Imaging Cancer Metabolism with PET**

**Bernhard Sattler 1,\*,**†**, Mathias Kranz 2,3,4,**†**, Barbara Wenzel 2, Nalin T. Jain 2, Rare¸s-Petru Moldovan 2, Magali Toussaint 2, Winnie Deuther-Conrad 2, Friedrich-Alexander Ludwig 2, Rodrigo Teodoro 2, Tatjana Sattler 5, Masoud Sadeghzadeh 2,**‡**, Osama Sabri 1,**‡ **and Peter Brust 2,**‡


Academic Editors: Anne Roivainen and Xiang-Guo Li Received: 28 March 2020; Accepted: 20 April 2020; Published: 26 April 2020

**Abstract:** Overexpression of monocarboxylate transporters (MCTs) has been shown for a variety of human cancers (e.g., colon, brain, breast, and kidney) and inhibition resulted in intracellular lactate accumulation, acidosis, and cell death. Thus, MCTs are promising targets to investigate tumor cancer metabolism with positron emission tomography (PET). Here, the organ doses (ODs) and the effective dose (ED) of the first 18F-labeled MCT1/MCT4 inhibitor were estimated in juvenile pigs. Whole-body dosimetry was performed in three piglets (age: ~6 weeks, weight: ~13–15 kg). The animals were anesthetized and subjected to sequential hybrid Positron Emission Tomography and Computed Tomography (PET/CT) up to 5 h after an intravenous (iv) injection of 156 ± 54 MBq [18F]FACH. All relevant organs were defined by volumes of interest. Exponential curves were fitted to the time–activity data. Time and mass scales were adapted to the human order of magnitude and the ODs calculated using the ICRP 89 adult male phantom with OLINDA 2.1. The ED was calculated using tissue weighting factors as published in Publication 103 of the International Commission of Radiation Protection (ICRP103). The highest organ dose was received by the urinary bladder (62.6 ± 28.9 μSv/MBq), followed by the gall bladder (50.4 ± 37.5 μSv/MBq) and the pancreas (30.5 ± 27.3 μSv/MBq). The highest contribution to the ED was by the urinary bladder (2.5 ± 1.1 μSv/MBq), followed by the red marrow (1.7 ± 0.3 μSv/MBq) and the stomach (1.3 ± 0.4 μSv/MBq). According to this preclinical analysis, the ED to humans is 12.4 μSv/MBq when applying the ICRP103 tissue weighting factors. Taking into account that preclinical dosimetry underestimates the dose to humans by up to 40%, the conversion factor applied for estimation of the ED to humans would rise to 20.6 μSv/MBq. In this case, the ED to humans upon an iv application of ~300 MBq [ 18F]FACH would be about 6.2 mSv. This risk assessment encourages the translation of [18F]FACH into clinical study phases and the further investigation of its potential as a clinical tool for cancer imaging with PET.

**Keywords:** preclinical radiopharmaceutical dosimetry; image-based internal dosimetry; OLINDA; MCT1/MCT4 lactate transporter inhibitor; [18F]FACH; radiation safety

#### **1. Introduction**

Aerobic glycolysis is a common feature of cancer physiology. Even under adequate oxygenation, cancer cells generate energy mainly via the cytosolic conversion of glucose into pyruvate and not via the downstream mitochondrial respiratory chain, known as the Warburg effect [1]. Due to the conversion of pyruvate to lactate, this phenomenon results in intracellularly high concentrations of lactate along with a decrease in the pH. Consequently, to avoid apoptosis caused by an acidic environment in the cytoplasm, cancer cells facilitate the proton-coupled efflux of pyruvate and lactate [2,3] by the transmembrane monocarboxylate transporters (MCTs). Upregulation of several MCTs was observed in different cancer types [4–8] and pharmacological inhibition of MCT1/MCT4 impairs cancer cell proliferation and tumor growth [9]. Increased expression of MCT1 and MCT4 in brain malignancies is assumed to be linked to the pathogenesis of glioblastoma in particular [8]. Accordingly, MCT1 and MCT4 are interesting targets for inhibitor-based molecular imaging and pharmacological treatment of this aggressive form of glioma.

Recently, [18F]FLac, [18F]FP, and an 11C-labeled coumarin analog were developed as radiotracers to monitor MCT1 with positron emission tomography (PET). However, defluorination, nonspecific binding, and insufficient affinity hamper the applicability of these tracers [10,11]. Recently, with [18F]FACH, a fluorinated analog of α-cyano-4-hydroxycinnamic acid (α-CHC) [6] ((*E*)-2-cyano-3-{4-[(3-[18F]fluoropropyl)(propyl)amino]-2-methoxyphenyl}acrylic acid), a novel radiolabeled MCT1/4 inhibitor was developed by our group. The inhibitory potency of FACH (IC50, MCT1 = 11 nM; IC50, MCT4 = 6.5 nM) indicates the suitability of [18F]FACH for in vivo molecular imaging of MCT1/MCT4 with PET [12,13].

For the clinical translation of newly developed radiopharmaceuticals, a safety and tolerability assessment is mandatory. One integral part of the assessment is the preclinical radiation dose assessment to estimate the radiation exposure caused by systemic, i.e., intravenous (iv), application of the radiotracer. For the estimation of organ doses (ODs) and the effective dose (ED), the pharmacokinetics and tissue distribution of the tracer have to be determined. The activity concentration in the different organs and tissues is required to determine time-integrated activity concentrations (TIAC) and finally calculate the region-specific numbers of disintegration (NOD). As recently shown, our approach to perform respective PET imaging studies combined with CT or Magnet Resonance Imaging (MRI) in a small number of piglets is suitable to generate pharmacokinetic data for dosimetric calculations [14–17]. With this fully imaging-based approach [18], the total number of animals used for dosimetry studies can be reduced to a minimum and the acquired data can be re-used for further research questions. It is important to note that when using animals for dose estimation in humans, an interspecies scaling has to be applied [19]. However, as also shown by our previous studies [14–16], preclinical dosimetry underestimates the ED in humans by up to 40% independent of the species and size of the test animal, and the radiation dose assessment has to be adapted accordingly.

Herein, we present the results of a preclinical dosimetry study of the newly developed MCT1/MCT4-specific radioligand [18F]FACH performed in juvenile pigs with a clinical PET/CT system and report on the estimation of the ED in humans as required for approval of application of the new radiopharmaceutical in clinical studies.

#### **2. Results**

The radiation safety of the newly developed MCT1/MCT4-specific radioligand [18F]FACH was preclinically investigated. Following intravenous injection of 156 ± 54 MBq (0.63 ± 0.49 μg) of the radiopharmaceutical, no adverse effects were observed based on vital sign monitoring, and three piglets were subjected to sequential PET/CT up to 252 min post injection (p.i.). Afterwards, volumes of interest (VOIs) of the various organs were manually defined using the respective whole-body CT dataset for anatomical orientation. The estimated TIACs were extrapolated to the human entity (Equation (2)), followed by estimation of the organ doses with OLINDA 2.1 and calculation of the effective dose using tissue weighting factors as published in Publication 103 of the International Commission of Radiation Protection (ICRP103) [20]. Finally, it can be concluded that [18F]FACH is safe with respect to the radiation risk that is caused by its systemic application for PET studies. This supports and encourages the translation of this newly developed radiopharmaceutical to clinical study phases.

The TIAC courses in organs allow for determination of the fraction of administered activity, also referred to as percent of injected dose (%ID), in the particular region, followed by calculation of the NOD by integration over time applied at mono-, bi-, or tri-exponentially fitted TIACs. Figure 1 shows six examples of the exponential fits of the fractions of activity (%ID) in different organs over time (all scaled to human dimensions). The fits of all organs of the three investigated animals as well as the exact exponential equations of the fits can be found in the Supplementary Materials (Figures S1–S3).

**Figure 1.** Examples of the mono-, bi-, or tri-exponential fits, including exponential fit equations and parameters of fit goodness (R-squared and squared error) of the human scale time–activity data using the EXM-Module of OLINDA. All fits are presented in the Supplementary Materials (Figures S1–S3).

In Figure 2, the distribution of activity in one representative animal at different times p.i. of [ 18F]FACH is presented, showing a high uptake in the liver, pancreas, small intestines, and spleen. The rapid uptake in the liver and later also in the gall bladder indicates hepatobiliary excretion. Early accumulation in the bone structures might be related to defluorination. At later time points, activity also accumulated in the gall bladder and the urinary bladder.

An example of manually delineated VOIs using the PET or CT information to extract the accumulated amount of activity in the respective organ is presented in Figure 3. All biodistribution data, including TIACs for all organs, are available in the Supplementary Materials (Tables S1–S4; Figure S1).

**Figure 2.** Whole body dynamic positron emission tomography (PET) images in coronal (upper row) and sagittal (lower row) views of pig 1 after intravenous (iv) application of 191 MBq [18F]FACH.

**Figure 3.** PET image with volumes of interest (VOIs) superimposed in (**a**) coronal, (**b**) sagittal, and (**c**) transversal views. The left panel, pig 1, obtained at 42 min post injection (p.i.), shows the organs except for the peripheral bone, whereas the right panel, pig 2, obtained at 56 min. p.i., shows all delineated organs. (The gray scale in the middle refers to the original quantitative PET data, not to the colored VOI structures).

Based on the biokinetic data extracted from the VOI analysis in piglets and the following extrapolation to the human entity (Equations (3) and (4)), the ODs and the ED were estimated for the adult male model. The resulting mean values obtained for 24 organs are presented in Table 1.


**Table 1.** Complete results of the dose assessment of the three investigated piglets using OLINDA 2.1 (mean value ± standard deviation (SD)).

OD = organ equivalent dose; ED contr. = effective dose contribution.

Here, the effective dose (*E*) represents the tissue-weighted sum of the equivalent doses in all specified tissues and organs of the body (Equation (1)), where *HT* is the equivalent dose in the respective tissue or organ, *T*, and *wT* is the tissue weighting factor [20].

$$E = \sum\_{T} w\_{T} \times H\_{T} \left| \frac{mSv}{MBq} \right| \tag{1}$$

The highest OD was received by the urinary bladder (62.6 ± 28.9 μSv/MBq), followed by the gall bladder (50.4 ± 37.5 μSv/MBq), the pancreas (30.5 ± 27.3 μSv/MBq), the kidneys (26.2 ± 2.2 μSv/MBq), the right colon (24.0 ± 7.4 μSv/MBq), and the liver (23.3 ± 13.6 μSv/MBq). When involving the tissue weighting factor *wT*, which weights the organ equivalent dose in a tissue or organ to represent the relative contribution of that tissue or organ to the total health risk resulting from uniform irradiation of the body, this ranking changes: the highest contribution to the ED is by the urinary bladder (2.5 ± 1.1 μSv/MBq), followed by the red marrow (1.7 ± 0.3 μSv/MBq), the stomach (1.3 ± 0.4 μSv/MBq), the right colon (1.2 ± 0.4 μSv/MBq), the lungs (0.9 ± 0.05 μSv/MBq), and the liver (0.1 ± 0.05 μSv/MBq). The ED was found to be 12.4 μSv/MBq. According to this preclinically obtained data, a standard injection in humans of 300 MBq [18F]FACH for PET in three-dimensional (3D) mode would result in an ED of about 3.7 mSv. As known from other studies [21], preclinical incorporation dose estimates underrate the ED to humans by up to 40% [14–16]. Taking this into account, the conversion factor to estimate the ED to humans undergoing a clinical PET investigation using [18F]FACH would rise to 20.6 μSv/MBq and finally result in an ED of about 6.2 mSv/300 MBq of administered [18F]FACH. In comparison to ED values determined for other clinically applied diagnostic radiotracers [21], this risk assessment encourages the translation of [18F]FACH into clinical study phases and the further investigation of its potential as a clinical tool for PET imaging of MCT1/MCT4 in different pathologies, including oncological diseases.

#### **3. Discussion**

Due to metabolic reprogramming, highly proliferating cancer cells utilize large amounts of glucose and convert the glucose carbon mainly to lactate to support their anabolic requirements [22–24]. Given the availability of suitable radiolabeled agents, non-invasive molecular imaging by PET enables the investigation of cancer-related alterations in the metabolism of cells. For example, the enhanced uptake of glucose in cancer cells can be monitored by [18F]fluoro-deoxyglucose ([18F]FDG), a substrate of the glucose transporters (GLUT) overexpressed in cancer cells that has been clinically used for more than three decades for diagnosis, staging, and treatment monitoring in oncology [25,26]. Monocarboxylate transporters (MCTs), which mediate the proton-coupled transport of small carboxylic acids such as lactate, the end product of aerobic glycolysis, are also known to be overexpressed in different cancers [27,28]. Accordingly, 11C- or 18F-labeled substrates of MCTs such as [18F]lactate and [11C]pyruvate have already been developed and evaluated regarding their potential to image the expression of MCTs in tumors using PET [10,29,30]. Besides, highly affine and small molecule inhibitors of MCTs designed for targeted cancer therapies also bear the potential for the development of respective imaging agents. In this context, our group has recently reported on the development and evaluation of [18F]FACH as a new MCT-targeting imaging agent possessing high inhibitory potency towards MCT1 and MCT4 [12,13]. For further assessment of the suitability of this radiopharmaceutical in clinical settings, a preclinical dosimetry study is necessary to estimate the doses delivered to humans to ensure the safe usage of [18F]FACH.

According to the data obtained in this study, the overall dose estimate and vital signs monitoring confirm the radiation safety and tolerability of [18F]FACH and support the translation of [18F]FACH to further clinical study phases.

As found for other low-molecular-weight 18F-labeled tracers, which were investigated by our group in recent years (Table 2), the organs involved in the renal as well as the hepatobiliary excretion received the highest ODs. Due to the high initial uptake of activity in the liver, although decreasing over time, the surrounding tissues, such as lung, pancreas, and kidney tissues, are exposed to comparatively high doses and thus belong to the organs contributing mainly to the ED. Furthermore, the increasing concentration in urine results in the high OD and subsequently ED of the urinary bladder wall. Notably, besides being excretory organs, the liver and kidneys are also target organs due to a high expression of MCT1/MCT4 [31–35].


**Table 2.** Dosimetry results of different PET radiopharmaceuticals preclinically estimated using piglet biokinetic data.

A limitation of this preclinical study is that the piglets did not void during the entire imaging phase, which results in a simplification of the data analysis. As the voiding bladder model implemented in the OLINDA software cannot be applied, the dose estimation for the wall of the urinary bladder is purely imaging based and uses data based on a continuously filling bladder over the imaging time (see Figure 1). Therefore, to reduce the dose to this dose-limiting organ in humans, participants of clinical studies should be instructed to void the bladder before and immediately after each imaging session.

Another limitation of this study is related to the increasing uptake of activity in the bones. Anatomically, high amounts of the red marrow are found not only in the backbone, sternum, and pelvis but also in the epiphyseal plates of the peripheral bone of juvenile piglets, because bone growth is not yet complete. Hence, target-specific accumulation of activity due to physiological expression of MCTs in erythrocytes produced by the red bone marrow [36,37] may contribute to comparatively high ODs and finally explain the high ED contribution of bone marrow. In addition, accumulation of [ 18F]fluoride is indicated by the radio-chromatographic analysis of plasma samples obtained during the PET imaging study (data are not shown).

Furthermore, a limitation of preclinical dosimetry studies is that the animals usually have to be anesthetized. Depending on the target of the tracer under investigation, anesthesia will have an influence on biodistribution and biokinetics. In this study, however, the anesthesia should have had a rather secondary influence. First and foremost, the anesthesia as described does not affect MCT1/MCT4 transporter inhibitors. It does act in the central nervous system (mainly the brain). With our protocol, we maintained a rather shallow anesthesia so that the animals respired spontaneously. Thus, the alterations of the biodistribution and biokinetics of [18F]FACH by this anesthesia should have had a rather small influence on the dosimetry result. We expect a slight general deceleration of the metabolism, particularly that of the gastro-intestinal tract, just as while sleeping very deeply. However, animal anesthesia is generally necessary to render preclinical incorporation dosimetry possible. Although it is a well-known limitation of preclinical dosimetry studies, it does not invalidate their results. Moreover, alongside the other limitations of preclinical dosimetry described, it will be part of the comparison once clinical data have been acquired and be involved in the (up)scaling of preclinical ED results for the assessment of the ED to humans by a particular radiotracer.

#### **4. Materials and Methods**
