Optimization of the Solvent and In Vivo Administration Route of Auranofin in a Syngeneic Non-Small Cell Lung Cancer and Glioblastoma Mouse Model

The antineoplastic activity of the thioredoxin reductase 1 (TrxR) inhibitor, auranofin (AF), has already been investigated in various cancer mouse models as a single drug, or in combination with other molecules. However, there are inconsistencies in the literature on the solvent, dose and administration route of AF treatment in vivo. Therefore, we investigated the solvent and administration route of AF in a syngeneic SB28 glioblastoma (GBM) C57BL/6J and a 344SQ non-small cell lung cancer 129S2/SvPasCrl (129) mouse model. Compared to daily intraperitoneal injections and subcutaneous delivery of AF via osmotic minipumps, oral gavage for 14 days was the most suitable administration route for high doses of AF (10–15 mg/kg) in both mouse models, showing no measurable weight loss or signs of toxicity. A solvent comprising 50% DMSO, 40% PEG300 and 10% ethanol improved the solubility of AF for oral administration in mice. In addition, we confirmed that AF was a potent TrxR inhibitor in SB28 GBM tumors at high doses. Taken together, our results and results in the literature indicate the therapeutic value of AF in several in vivo cancer models, and provide relevant information about AF’s optimal administration route and solvent in two syngeneic cancer mouse models.


Introduction
Auranofin (AF) is an orally available, lipophilic, organogold compound with a well-known safety profile. Its chemical name is [2,3,4,6-tetra-o-acetyl-L-thio-β-D-glycol-pyranoses-S-(triethylphosphine)-gold(I)] in which the triethyl group and tetraacetylthioglucose (TATG) stabilize the gold molecule [1]. Due to its chemistry, AF is considered a relatively stable compound that exists as a monomer in solution and is freely soluble in lipid membranes. Furthermore, it is likely to undergo ligand exchange reactions in the presence of competing thiols [2]. In 1985, AF was approved by the U.S. Food and Drug Administration (FDA) for the treatment of rheumatoid arthritis (RA) under the brand name Ridaura®. To temper RA disease progression, AF is administered as a 3 mg oral capsule once or twice daily, after which 25-30% of the administered gold is detected in the plasma mostly bound to cysteine-34 of albumin [1,3]. Over the past decades, there was a decline in the prescription of AF as an antirheumatic drug. Firstly, due to the adverse side effects associated with its long-term use, which were predominantly gastrointestinal problems. Secondly, more effective medication for RA patients became available, and AF was replaced by other disease-modifying antirheumatic drugs (DMARDs) or nonsteroidal anti-inflammatory drugs (NSAIDs) [4].
Recently, gold compounds have increasingly been investigated as potential anti-tumor agents based on their medical and chemical properties described in the literature [5]. Scientific studies in the mid-1980s called attention to AF's antiproliferative effects against cancer cells for the first time [6]. In the following years, AF has received wide attention as a multifunctional compound with anticancer and antibacterial properties, among others [4,7]. Nowadays, AF is mostly under investigation for oncological application through its main mechanism of action, the inhibition of the redox enzyme thioredoxin reductase 1 (TrxR).
To obtain preclinical evidence on AF as a potent anticancer agent, its antineoplastic activity has already been investigated in various in vitro and in vivo models [4,8]. We provide an overview of the in vivo studies that use AF for the treatment of cancer, including the solvent, dose, treatment schedule and administration route that were used (Table 1). However, we noticed that there were inconsistencies on administration parameters of AF in vivo. The in vivo dose of AF varied between 0.1 and 15 mg/kg and were mostly administered via intraperitoneal (i.p.) injections with varying ways of formulation. To a lesser extent, AF has also been administered per os (p.o.) via oral gavage or via vein injections. The most widely used model to investigate the anti-cancer activity of AF in vivo is the tumor xenograft model, whereby human tumor cells are transplanted into immunocompromised nude mice. The majority of studies using this tumor model demonstrated an inhibitory effect of AF on tumor growth. Next to its use as a monotherapy, AF has also been comprehensively tested in combination with other drugs to further enhance its anti-cancer activity in vivo (Table 1). Table 1 highlights the therapeutic value of AF in various in vivo cancer models, but with inconsistencies on several important parameters of in vivo AF administration. Therefore, we studied the optimal solvent to enhance the solubility of AF and three administration routes (i.p. injection, oral gavage and osmotic minipumps) to identify the most optimal method to reduce AF-mediated toxicity and increase its efficacy in a syngeneic SB28 glioblastoma (GBM) C57BL/6J mouse model and 344SQ non-small cell lung cancer (NSCLC) 129S2/SvPasCrl (129) mouse model. Our results provide information about the optimal administration route and in vivo solvent of AF in our syngeneic C57BL/6J and 129-mouse models; oral gavage with a solvent of 50% DMSO, 40% propylene glycol 300 (PEG300) and 10% ethanol were used, respectively. In addition, we wanted to highlight that the in vivo administration route of AF should be carefully considered based on the mouse model and tumor type. Together with the in vivo results in the literature (Table 1), this study highlights the therapeutic value of AF in several in vivo cancer models.

Mice
Male 129S2/SvPasCrl (129) mice, age 6-9 weeks, were obtained from Charles River. Female C57BL/6J mice, age 6-10 weeks were obtained from Jackson Laboratories. All animal care and experimental procedures were approved by the Ethics Committee of the University of Antwerp (2020-32 and 2020-20). Upon arrival, mice were given a 7-day adaptation period before being used in experiments to reduce stress levels. All mice were housed in a temperature-controlled environment with 12 h light/dark cycles and received food and water ad libitum. Mice were checked on a daily basis to inspect health and wellbeing.

Murine Cell Lines
Murine adenocarcinoma lung cancer cell line 344SQ derived from KrasLa1/+p53R172H∆G mice (subcutaneous (s.c.) metastasis) were kindly provided by Jonathan M. Kurie (University of Texas, MD Anderson Cancer Center, Houston, TX, USA). This cell line is syngeneic to the male 129S2/SvPasCrl mice. The 344SQ cells were cultured in RPMI cell culture medium supplemented with 10% FBS and 10 mM L-Glutamine. The murine glioblastoma cell line SB28 (provided by H. Okada, UCSF, San Francisco, CA, USA) is syngeneic to the female C57BL/6J mice. The SB28 cells were cultured in DMEM supplemented with 10% heatinactivated FBS, 1% HEPES and 1% GlutaMAX. Cell lines were maintained at 37 • C and 5% CO 2 . All cell lines were tested on a routine base for mycoplasma contamination.

Tumor Kinetics and Survival
Prior to injection, 344SQ and SB28 cells were harvested using TrypLE, washed three times with sterile PBS and put through a 70-µm cell strainer to assure single-cell suspension without any contaminants. Next, 129-mice were injected s.c. with 1 × 10 6 344SQ cells suspended in 100 µL sterile PBS in the left shaved flank, while C57BL/6J mice were injected at the same position with 1 × 10 6 SB28 cells in 100 µL sterile PBS. When tumors reached an average size of 40-50 mm 3 , mice were randomized based on tumor size and divided into different treatment groups. Tumor size was measured three times a week using a digital caliper. Tumor volume was calculated using the formula (length × width 2 )/2. Mice were sacrificed when a tumor size of 1500 mm 3 was reached.

In Vivo Administration of AF
AF was administered via three different routes of administration in SB28 and 344SQ tumor-bearing mice. First, AF was administered via daily i.p. injections over a period of 14 consecutive days. Second, AF was administered s.c. via osmotic minipumps (Alzet, CA, USA, type 1002). In practice, the osmotic minipumps were filled with either 100 µL AF or vehicle (solvent). The required AF concentration for these devices was based on the average weight per treatment group and calculated using the online tool provided by Alzet [47]. Mice were anaesthetized with isoflurane. The hair of the neck was shaved and a small incision was made between the ears. Using a hemostat, a s.c. pocket was made wide enough for an osmotic minipump. The pocket was flushed with saline and an osmotic minipump was inserted. Then, the incision was closed with two sterile surgical staples. The osmotic minipump provided a long-term delivery of 14 days after which it was removed while the mice were under anesthesia. Third, AF was administered daily via oral gavage using a 20G flexible feeding needle for a period of 14 days. Calculations of the required AF concentration in mg/kg were made for each individual mouse based on the individual body weight. The toxicity of AF was measured based on total body weight, behavior, the Mouse Grimace Scale (MGS) and post-mortem evaluation.

Thioredoxin Reductase Activity Assay
After 14 days of AF treatment p.o., SB28 tumors were dissected and disrupted in lysis buffer using a tissue homogenizer (Qiagen, Hilden, Germany). Afterwards, protein lysates were used to measure TrxR activity using the Thioredoxin Reductase Colorimetric Assay kit (Cayman chemical, Ann Arbor, MI, USA) according to the manufacturer's instructions. Absorbance was recorded at 405 nm with a Spark®Cyto (Tecan, Männedorf, The Switzerland) during the initial 5 min of the reaction. TrxR activity was calculated using the formula provided by the protocol, whereby background measurements were subtracted from all values. An equal amount of protein was loaded for each condition as determined by the Pierce BCA protein kit (Thermo Scientific, Merelbeke, Belgium).

Statistics
Differences were considered to be statistically significant if p < 0.05. To analyze differences in tumor kinetics overtime, we used R [48] with afex [49] and emmeans [50] packages to perform mixed model ANOVA. To assess differences between the TrxR activity in the different treatment groups, a Mann-Whitney U test was performed using SPSS v27. Graphs were made using GraphPad v9 software.

Daily I.P. Injections with AF Induce Weight Loss and Gut-Related Cytotoxicity in 344SQ 129-Mouse Model
In the literature, i.p. injections with 10 mg/kg AF were most often used for the in vivo administration of AF (Table 1). Therefore, we tested the administration of 10 mg/kg AF via daily i.p. injections over a period of 14 consecutive days in the 344SQ tumor-bearing 129-mouse model. We observed a significant delay in tumor growth in AF-treated mice, as compared to vehicle treated controls ( Figure 1A). Moreover, after five days of treatment, mice showed clinical signs of cytotoxicity, discomfort and weight loss of approximately 20% during the treatment period ( Figure 1B). The MGS was used to assess post-procedural pain by evaluating changes in the facial expressions of mice [51]. After i.p. injections with AF, mice showed a moderate or marked appearance of 3 out of 5 MGS action units (orbital tightening, ear position and whisker change). Postmortem examination also revealed bloated and obstructed intestines ( Figure S1). This cytotoxic effect was not mouse straindependent, since C57BL/6J mice (without a tumor) showed the same behavioral changes of the MGS immediately after i.p. injection with 10 mg/kg AF. In addition, we also encountered some problems with the solubility of AF at doses of 10 mg/kg and above. Dilutions in both PBS and 4% (v/v) DMSO/10% (v/v) ethanol resulted in precipitations of the compound over time. For future experiments, other routes of AF administration and another solvent should be investigated to improve the solubility of AF.

Continuous Slow Release of AF Treatment in 344SQ-and SB28-Bearing Mice via an Osmotic Minipump
Based on the literature [52] and chemical properties of AF, a mixture of 50% DMSO, 40% PEG300 and 10% absolute ethanol was used to successfully prepare an intermediate AF stock solution of 2 mg/mL without precipitation. To administer AF chronically for 14 days, osmotic minipumps (Alzet, type 1002) were implanted s.c. in both SB28 and 344SQ tumor-bearing mice to deliver 2, 5, 10 and 15 mg/kg AF per day at the same constant rate. Compared to the toxic i.p. injections with AF, these devices are easy to use, provide constant drug plasma levels for 14 days and induced only minimal stress in the animals [52,53]. After 14 days of AF delivery via the minipumps, there was no delay in tumor growth and no increase in survival in both SB28 C57BL/6J and 344SQ 129-mice treated with high doses (10-15 mg/kg) or low doses (2-5 mg/kg) of AF, compared to vehicle-treated mice (Figure 2A-F). The weight of mice in both models remained constant during and after the 14 days of treatment with different doses of AF, and there was no observed toxicity based on behavior or the MGS (Figure 2G-H). However, we observed that s.c. delivery of high doses of AF (10-15 mg/kg) via the osmotic minipumps resulted in skin irritation and lesions at the position of the pump where AF was released ( Figure S2). As a result of these lesions, high doses of AF may not have been properly released, whereby no significant positive effect on tumor growth and survival was observed (Figure 2A-F). These skin lesions were not observed when lower doses of AF (2-5 mg/kg) or the vehicle were administered to the mice via the s.c. minipumps. In conclusion, the mixture of 50% DMSO, 40% PEG300 and 10% absolute ethanol is a good vehicle for dissolving AF at both low and high concentrations. However, neither i.p. injections nor s.c. delivery via osmotic minipumps were the ideal route of administration for AF.

Oral Administration of AF Treatment in 344SQ-and SB28-Bearing Mice
As an anti-rheumatic drug, AF is formulated as a capsule and given orally to patients. To mimic its clinical administration, we changed the administration route of AF to daily oral gavage using a 20G flexible feeding needle for a period of 14 days.
After daily treatment with 10 or 15 mg/kg AF for 14 consecutive days via oral gavage, there was no delay in tumor growth in SB28 tumor-bearing C57BL/6J mice compared to the vehicle group ( Figure 3A). In the 344SQ 129-mice, results showed a significant decrease in tumor volume after 10 mg/kg AF treatment compared to vehicle ( Figure 3B). Nonetheless, there was no observed cytotoxicity based on body weight or behavior in either C57BL/6J or 129-mouse models ( Figure 3C,D) and there were no visual signs of local toxicity in the peritoneal cavity or intestines after postmortem dissection. Compared to i.p. injections and s.c. minipump delivery of AF, oral gavage was the best route of administration for AF in our SB28 tumor-bearing C57BL/6J and 344SQ tumor-bearing 129-mouse models.

AF Inhibits TrxR Activity in SB28 Tumors
The main mechanism of action of AF is the inhibition of redox enzyme TrxR. Protein lysates were isolated from disrupted SB28 tumors after 14 days of oral gavage treatment with high AF doses (10 and 15 mg/kg) to check TrxR activity. TrxR activity was significantly decreased in both the 10 mg/kg and 15 mg/kg AF-treated groups compared to the vehicle group ( Figure 4). In conclusion, oral administration of high doses of AF was able to inhibit TrxR activity in SB28 tumors after 14 days.

Discussion
Due to rising costs, high risk of failure and slow clinical translation of new drug discovery and development, there is an increasing interest in repurposing well-known and well-characterized licensed non-cancer drugs to the oncology domain, as underscored by the Repurposing Drugs in Oncology project [54][55][56]. Hence, the compound AF is receiving increasing interest as an object of repurposing strategies in cancer due to its inhibitory function against TrxR. The therapeutic efficacy of AF against cancer and its relative safety profile in RA patients emphasize the potential of AF as an attractive drug for further clinical investigation.
Aside from multiple in vitro studies, AF is also increasingly tested in various in vivo cancer models to predict its safety, toxicity and efficacy. However, as summarized in Table 1, there is a lack of consistency regarding the dosage, solvent and administration route for AF treatment in mice. Therefore, the goal of this study was to test different delivery methods and solvents for AF treatment in GBM and NSCLC mouse models.
In the literature, the in vivo dose of AF varies between 0.1 and 15 mg/kg and is mostly administered via i.p. injections with different treatment schedules and solvents (Table 1). In a preliminary in vivo experiment with 344SQ tumor-bearing 129-mice, we observed a significant delay in tumor growth after treatment with 10 mg/kg AF via daily i.p. injections for 14 days, compared to vehicle-treated mice. Similarly, several other in vivo studies demonstrated an inhibitory effect of AF monotherapy on tumor growth after i.p. injections in different cancer mouse models [9][10][11]13,[15][16][17][18][19][20][21]24,26,30,35,40] (Table 1). In CLL and BCP-ALL xenograft mouse models, i.p. AF injections caused a reduction in the leukemia cell burden and human blasts [9,12]. The reduction in tumor volume was related to AFmediated induction of apoptotic tumor cell death, as measured by TUNEL assay [22,26] or caspase-3 cleavage [10,21,30], and to a decreased proliferation, as measured by Ki67 staining [11,21,22] in different in vivo models (Table 1). Additionally, a limited number of studies reported that AF treatment via i.p. injections significantly improved the survival of mice compared to mice that received no treatment [9,12]. In studies investigating AF in combination with other clinically applicable compounds in vivo, i.p. injections of AF monotherapy ranging from 0.1 to 10 mg/kg often did not exhibit significant suppression of tumor growth in distinct cancer mouse models [25,28,33,37,38,42,45].
In the present study, we noticed severe overall toxicity in 129-mice after i.p. AF injections, since they lost on average 20% of their body weight. Moreover, 129-mice that were injected i.p. with AF showed visual signs of pain based on the MGS and had bloated and obstructed intestines compared to vehicle-treated mice, as observed in postmortem examination. This effect was not mouse strain dependent, since the C57BL/6J mouse showed the same visual signs of pain and behavior changes after i.p. injections with 10 mg/kg AF. Similarly, the most common adverse effect in about 50% of RA patients treated with AF are gastrointestinal problems such as loose stools, abdominal cramping and watery diarrhea during early months of administration [4,57]. These side effects are controlled by reducing the dosage, or temporary or permanent withdrawal of AF in these patients [57]. However, other studies using i.p. injections for AF delivery in mice did not report on this type of toxicity and even indicate absences of weight loss or blood count anomalies (Table 1). Therefore, we believe it is important that the administration route of AF is carefully considered based on the type of mouse model. Additionally, the prediction of drug cytotoxicity remains a major goal in drug development and the route of drugs to the clinic. It is very important that drugs are nontoxic with minimized side-effects for the patient. However, clinical trials with anticancer drugs often fail due to safety reasons and unmanageable toxicity [56,58]. It is critical at each stage of drug development to consider safety as a primary concern, even if it is not a primary objective [56]. Acute and chronic toxicity of drug candidates, which mimic the clinical dose regimen, are always examined in animal models. The accumulation of drugs in vital organs or blood cells is one the major factors for toxicity [58]. In both our mouse models, we experienced immediate acute toxicity after i.p. injections with AF. Therefore, we choose not to continue with this type of administration in both 129 and C57BL/6 models. We wanted to find an administration route for AF that showed treatment efficacy in vivo without inducing cytotoxic side-effects.
During the formulation of an intermediate stock solution of AF for i.p. injections, we experienced solubility problems when dissolving AF at a concentration of 2 mg/mL, for administration of 10 mg/kg AF or higher to the mice. Both PBS and 4% DMSO/10% ethanol resulted in precipitations of AF over time, even though these solvents were used in other studies (Table 1). In the literature, there is no uniformity on the appropriate solvent for formulation of AF in vivo (Table 1). In the present study, a vehicle composed of 50% DMSO, 40% PEG300 and 10% ethanol was found to improve the solubility of AF and to be nontoxic for mice, based on weight observations and postmortem checks of vital organs. This vehicle has already been shown to be compatible and safe to use for the in vivo delivery of compounds via osmotic minipumps or oral gavage, even with this high ratio of DMSO [52]. In the human setting, the risk of DMSO-mediated toxicity is bypassed since AF, as an anti-rheumatic drug or in clinical trials, is given to patients in oral capsules (3 mg).
Since daily i.p. injections with AF appeared to be toxic in our 344SQ tumor-bearing 129-mouse model, we opted for two different delivery methods. Firstly, we investigated the continuous and slow delivery of low and high doses of AF using osmotic minipumps implanted s.c. in SB28 and 344SQ tumor-bearing mice. We opted for this administration route since these small infusion pumps can provide accurate and continuous dosing of AF to mice [52]. They form a convenient and reliable alternative to the frequent i.p. injections by maximizing therapeutic efficacy and reducing adverse effects. However, high doses of AF (10 and 15 mg/kg) released s.c. via the minipump system induced skin ulceration and rupture at the position of the pump where AF is released. Due to these skin lesions, s.c. administration of AF in the mice was not reliable and no proper absorption at the delivery site was achieved; therefore, AF could not induce the desired effect on tumor growth or survival. AF also induces skin irritations and rash in 20% of RA patients within the first year of treatment [4]. Lower dose concentration of AF did not induce these skin lesions, but was not powerful enough to significantly affect tumor growth. Therefore, we are convinced that higher doses of AF are necessary to induce a strong anticancer effect as a monotherapy in vivo. Secondly, to mimic clinical administration of AF to RA patients, we tested the delivery of AF via oral gavage in our GBM and NSCLC mouse models. High doses of AF p.o. were well-tolerated in both models without any weight loss or visual signs of toxicity. A study by Abutaleb et al. reported that AF was stable after exposure to simulated gastric pH and was not affected by the enzymes of gastric fluids [59]. We showed a significant decrease in tumor volume after 10 mg/kg AF treatment in the 344SQ tumorbearing 129-mice, but not in SB28 tumor-bearing C57BL/6J mice. In a TP53-mutated diffuse large B-cell lymphoma (DLBCL) PDX model, treatment with 50 mg/kg AF via oral gavage for 21 consecutive days also significantly inhibited tumor growth, without any body weight differences [14]. Dependent on the tumor type, AF monotherapy is able to significantly reduce tumor growth in some studies (Table 1). However, targeting different hallmarks of cancer via a combination strategy is a more effective therapeutic approach compared to monotherapies. Therefore, we strongly believe that the true power of AF lies within rationally designed drug combination strategies, as seen by the numerous combination studies in Table 1, to further improve the anti-cancer potential of AF, overcome tumor heterogeneity and limit the ability of cancer cells to adapt and develop treatment resistance. For example, in the same SB28 GBM mouse models, daily oral delivery of AF in combination with another ROS-inducing compound significantly reduced tumor growth and prolonged survival in vivo [45]. In a colorectal xenograft model, daily administration of 6 mg/kg AF via oral gavage in combination with 5-fluorouracil (5-FU) inhibited tumor growth and reduced the number of metastatic lung nodules compared to AF monotherapy [29]. These results confirm the potential of oral administration for AF treatment in vivo in combination regimens.
In vitro experiments already demonstrated that TrxR inhibition is one of the main mechanisms of action of AF [60,61]. Our results confirmed that oral delivery of high doses of AF (10 and 15 mg/kg) led to the inhibition of TrxR activity within SB28 tumors in vivo. This is in-line with other studies showing TrxR inhibition after AF delivery via i.p. injections at varying doses, ranging from 5 mg/kg to 10 mg/kg, using a colorimetric TrxR assay [10,26,33]. Additionally, the combination of AF and the GSH biosynthesis inhibitor BSO also inhibited TrxR activity in vivo, but at lower AF concentrations of around 1.5 mg/kg [31,46]. These data suggest that AF is able to inhibit its primary target TrxR in vivo via both i.p. injections and oral gavage. Despite TrxR inhibition in SB28 tumors, oral administration of AF did not result in the inhibition of SB28 tumor growth. A possible explanation could be that TrxR inhibition was not strong enough and other mechanisms counteracted the perturbed redox status after TrxR inhibition. Our previous in vitro results demonstrated that a low dose of AF-induced TrxR inhibition, subsequently boosted the cell's antioxidant defense capacities by upregulating pro-survival molecules, such as NRF2 and glutathione, to prevent cancer cell death [62].
The recommended long term dosing regimen of AF in adult RA patients is 6 mg daily, in a single dose or divided doses [1]. The toxicity profile and therapeutic effect of AF was monitored in clinical trials with more than 5000 RA patients taking the drug, plus RA patients that were monitored for over 7 years. Overall, AF did not show any evidence of severe cumulative toxicity in RA patients [59,63]. However, much higher doses of AF, alone or in combination with other drugs, were needed in vivo to inhibit TrxR activity in the tumors of mice, as shown by our results and others (Table 1). Therefore, it might prove challenging to obtain sufficiently high concentrations of AF in the patient's tumor without increasing unwanted side-effects [4,64,65]. New technological innovations, such as AF-loaded nanoparticles, could offer a possible solution since they are able to enhance drug localization in the target site and minimize systemic cytotoxicity [66,67]. Alternatively, well-designed combination strategies might limit the need for high AF doses. As listed in Table 1, AF in combination with vitamin C, anti-PD-L1, MK2206, adriamycin, cisplatin, buthionine sulfoximine, cold atmospheric plasma etc. showed statistically significant reduction in tumor growth and/or prolonged survival, without obvious side effects, in several in vivo cancer models [25,26,33,35,37,40,41,45].

Conclusions
Daily i.p. injections of 10 mg/kg AF induced weight loss and gastro-intestinal problems in the 129-mouse model and the use of osmotic pumps resulted in local skin lesions. Therefore, oral gavage was the most appropriate administration route for high doses of AF in the syngeneic GBM C57BL/6J and NSCLC 129-mouse models, without inducing weight loss or showing signs of toxicity. However, it is important to be aware that the administration route of AF should be carefully considered based on the mouse model and type of tumor, since our data show that inappropriate selection can lead to false interpretation of the anti-tumoral effect. A solvent consisting of 50% DMSO, 40% PEG300 and 10% ethanol provided optimal solubility of AF for p.o. administration to mice. In addition, AF was a potent TrxR inhibitor in SB28 GBM tumors at high doses, but failed to inhibit tumor growth. Altogether, our results and in vivo results already described in the literature ( Table 1) highlight the therapeutic value of AF in several in vivo cancer models; however, it requires more standardization for its route of administration, and solvent per mouse model and tumor type.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/pharmaceutics14122761/s1, Figure S1: Effect of daily i.p. injections with 10 mg/kg AF on the intestinal tract of 129-mice.; Figure S2: Effect of continuous delivery of 10 mg/kg AF using s.c. osmotic minipump system on the skin of C57BL/6J mice.