Single‐dose pharmacokinetics of telmisartan oral solution and effect of feeding in dogs

Abstract Telmisartan is an angiotensin II receptor blocker that has great potential to improve the treatment of hypertension, proteinuria, and cardiovascular disease in dogs. A feline‐approved telmisartan oral solution (TOS) is available, but this formulation has not been evaluated in dogs. The aims of this study were to establish the pharmacokinetics of telmisartan administered as TOS and determine the effect of feeding on drug absorption in dogs. In a cross‐over design, seven healthy dogs received 1 mg/kg telmisartan orally as TOS with or without food and underwent serial measurement of plasma telmisartan concentrations over 24 h. Bioequivalence of TOS administered with vs. without food was assessed by the 90% confidence interval method for maximum concentration (Cmax), and the observed and extrapolated areas under the curve (AUC0–t and AUC0–∞). The mean ratios of these parameters were 0.97 (CI 0.74–1.27), 0.92 (0.81–1.03), and 0.90 (0.82–1.00), respectively. Feeding methods were not bioequivalent based on Cmax due to interindividual variation. These results suggest that TOS can be given to dogs with or without food but should be administered in the same way consistently. Additional pharmacokinetic and pharmacodynamic studies are warranted to confirm this recommendation and establish the therapeutic targets for telmisartan in dogs.


| INTRODUC TI ON
Telmisartan is a relatively new angiotensin II receptor blocking drug (ARB) used for the treatment of systemic hypertension (Acierno et al., 2018), proteinuria (Bugbee et al., 2014), renal disease (Lourenço et al., 2020), and cardiovascular disease (Kawaguchi et al., 2019) in dogs, cats, and humans (Michel et al., 2016). Veterinary interest in telmisartan is growing given the mechanistic benefits of ARBs over the traditional angiotensin converting enzyme inhibitors (ACEi) when suppression of the renin-angiotensin-aldosterone system (RAAS) is clinically indicated. ARBs might be superior to ACEi because they selectively inhibit the maladaptive effects of the angiotensin type I receptor (AT 1 R) with high affinity, while still allowing for the protective counterregulatory effects of AT 2 R; this is in contrast to the ACEi, which indirectly inhibit both AT 1 R and AT 2 R. This unique mechanism of action reduces aldosterone synthesis to promote vasodilation, natriuresis and improved attenuation of the pressor response to angiotensin II (Ames et al., 2019;Jenkins et al., 2015), allowing ARBs (telmisartan) to be considered as the first line therapy for diseases requiring RAAS modulation in dogs (Brown et al., 2013).
Telmisartan oral tablets (Micardis®, Boehringer Ingelheim International GmbH) are currently the most commonly used telmisartan formulation in canine medicine. In 2018, telmisartan oral solution (TOS, Semintra®, Boehringer Ingelheim Vetmedica GmbH) gained FDA approval for the treatment of feline systemic hypertension (Food and Drug Administration, 2018). TOS is commercially available in the United States formulated as a non-flavored, viscous, 10 mg/ml oral liquid solution. This low concentration allows more accurate dosing in small breed dogs. Additionally, the new liquid formulation may provide a more attractive option to owners of dogs that resist oral tablet administration, which could improve overall compliance. However, the pharmacokinetics of TOS in dogs are currently unknown, precluding rational guidelines for the use of this product in this species.
Pharmacokinetic profiles have been established for telmisartan oral tablets in dogs (Baek et al., 2013). However, the data cannot be extrapolated to TOS in dogs because differences in formulation could affect parameters, particularly those affected by absorption (Riviere, 2009). Similarly, the pharmacokinetics of TOS have been evaluated in cats, but these cannot be extrapolated to dogs owning to species-specific differences in drug disposition (FDA, 2018;Sutton, 2004). The effect of feeding on absorption must also be critically evaluated in dogs with respect to the specific dose and formulation investigated before accurate dosing recommendations can be advised. In humans and cats, the oral bioavailability of telmisartan is reduced when given with food. This difference did not affect primary outcomes in phase 3 clinical trials, so telmisartan can be administered with or without food in these species (Boehringer Ingelheim International GmbH, 2018;FDA, 2018). However, species differences can greatly affect gastrointestinal physiology and drug absorption under different conditions. To our knowledge, no studies have evaluated the effect of feeding on the oral bioavailability of telmisartan in dogs, let alone for the TOS product. Thus, it would be prudent to determine whether a difference exists and to evaluate the magnitude of that potential effect, further enhancing rational recommendations for telmisartan administration in dogs.
The present study had two specific objectives. Our first objective was to establish the pharmacokinetic profile of telmisartan administered as TOS in healthy adult dogs at a clinically relevant dose. We hypothesized telmisartan administered as TOS would be rapidly absorbed and exhibit a moderate terminal half-life, similar to the pharmacokinetic profile formerly established for the oral tablet formulation in dogs. Our second objective was to determine the effect of feeding on the oral absorption of telmisartan administered as TOS in dogs. We hypothesized that TOS administered to dogs with and without food would be bioequivalent based on telmisartan maximum plasma concentration (C max ) and area under the curve (AUC).

| Subject recruitment
Eight healthy adult dogs weighing ≥14 kilograms were recruited in September 2021 from the pet population of the students, faculty, and staff at the University of Illinois College of Veterinary Medicine with informed owner consent. All dogs were deemed systemically healthy during an initial screening visit 2-3 weeks prior to the study. A complete physical examination was performed by a boardcertified small animal veterinary internist in addition to a complete blood count, serum biochemistry, urinalysis, and oscillometric blood pressure measurement (Cardell Insight Veterinary Monitor, Midmark Corporation). Additionally, each dog was voluntarily offered a sample of Hill's Prescription Diet® i/d canned formulation (Hill's Pet Nutrition, Inc.) and an oral bolus of water equal to the volume of TOS administered during the sampling phases of the study. Dogs were excluded if they had clinically relevant abnormal laboratory results, systolic blood pressure >160 mmHg, or were receiving any medications other than monthly heartworm and flea/tick preventatives.
Dogs that were unwilling to voluntarily consume the diet or were resistant to oral liquid administration were also excluded. This study was approved by the University of Illinois Institutional Animal Care and Use Committee (protocol #19228).

| Experimental design
This study was performed in a two-way, balanced, cross-over design.
Dogs were randomly assigned to one of two groups by blindly drawing lots out of an envelope after successful completion of the initial screening visit. Group 1 underwent serial plasma drug monitoring following a single 1 mg/kg oral dose of telmisartan as TOS (0.1 mg/ ml) on an empty stomach; then, followed by a 7-day washout period, underwent serial plasma drug monitoring following a single oral dose of 1 mg/kg telmisartan as TOS with food. Group 2 underwent serial plasma drug monitoring following a single oral dose of 1 mg/kg telmisartan as TOS with food first, followed by administration on an empty stomach after the 7-day washout period. The 7-day washout period was determined based on the mean half-life plus 3 standard deviations (5.43 h + [3 × 1.75 h] = 10.7 h) reported for dogs orally administered telmisartan tablets, which are approved for human use (Baek et al., 2013). Because the dose used in the present study was smaller and sustained-release effects were not expected for an oral solution, we anticipated the half-life of TOS would be similar to or shorter than what has been previously reported for the tablets. Additionally, the Food and Drug Administration recommends a washout period at least 10 times the half-life between treatments for pharmacokinetics studies (10.7 h × 10/24 h/day = 4.5 days).

| Drug administration and sample collection
All dogs were fasted (≥8 h) prior to arrival for catheter placement.
Dogs were sedated with dexmedetomidine (Dexdomitor, Zoetis Inc.) at 4 μg/kg and butorphanol (Torbugesic-SA, Zoetis Inc.) at 0.2 mg/ kg intravenously for placement of a sterile jugular venous catheter (MILA International, Inc.) using the modified Seldinger technique (Portillo et al., 2006). Atipamezole (Antisedan, Zoetis Inc.), a reversal agent, was administered intramuscularly at 0.04 mg/kg and the dogs were allowed to recover. Following recovery, the dogs were fed a meal of a highly digestible diet (Hill's Prescription Diet® i/d) and then fasted for a minimum of 12 h overnight in hospital prior to drug administration and sampling the following morning with free access to water.
For the fasted phase, dogs were administered 1 mg/kg telmisartan as TOS. Four hours later, dogs received half of their daily energy requirement in the canned form of Hill's Prescription Diet® i/d. For the fed phase, dogs were fed as described above, followed by administration of telmisartan 5-10 min after meal completion. This protocol was adapted from the previously pub-

lished TOS pharmacokinetic studies used in cats (Food and Drug
Administration, 2018).
In both phases, blood samples (4 ml) were collected via the sampling catheter using the three-syringe technique at the following time points: 0 min (before administration), 15, 30, and 45 min, 1, 1.5, 2, 4, 8, 12, 18, and 24 h. Following collection, all blood samples were placed into heparinized tubes and immediately refrigerated; plasma was separated by centrifugation at 1800 × g at 4°C within 4 h of collection and immediately stored at −80°C until analysis. Oscillometric blood pressure measurements were performed at the following time points: 0, 4, 8, 12, and 24 h. A systolic blood pressure <100 mmHg was considered hypotension.
After vortex, the mixture was subject to centrifugation for 5 min at 8000 rpm. The supernatant was for instrument injection. Samples were analyzed with the 5500 QTrap LC-MS/MS system (Sciex).
Software Analyst 1.7.1 was used for data acquisition and analysis. The 1200 HPLC system (Agilent Technologies) includes a degasser, an autosampler, and a binary pump. The LC separation was performed on an Agilent SB-Aq column (4.6 × 50 mm, 5 μm) with mobile phase A (0.1% formic acid in water) and mobile phase B (0.1% formic acid in acetonitrile). The flow rate was 0.3 ml/min. The linear gradient was as follows: 0-2 min, 100% A; 10-14 min, 0% A; 14.1-19 min, 100% A. The autosampler was set at 10°C. The injection volume was 10 μl. Mass spectra were acquired under electrospray ionization (ESI) with the ion spray voltage of +5500 V. The source temperature was 400°C. The curtain gas, ion source gas 1, and ion source gas 2 were 32, 60, and 60 psi, respectively. Multiple reaction monitoring (MRM) was used for quantitation with the following transitions: telmisartan m/z 515.1 m/z 276.1, d3-telmisartan m/z 518.1 m/z 279.1. The linear range of the assay was 0.25-500 ng/ml. Recovery of telmisartan in blank plasma (n = 5) was evaluated at low (2.5 ng/ml), medium (10 ng/ml), and high (100 ng/ml) control concentrations yielding 95.9 ± 4.7%, 103.9 ± 8.3%, and 105.0 ± 7.5% recovery, respectively.
Precision was also evaluated at these concentrations (n = 5 per concentration) on three separate days yielding intra-and inter-assay coefficients of variations of 3.0-9.7% and 1.1-9.5%, respectively.

| Pharmacokinetic and statistical analysis
Continuous data are presented as mean ± standard deviation.
Non-compartmental pharmacokinetic analysis was performed using Phoenix WinNonlin (Certara L.P.) for both the fasted and fed phases. Statistical comparisons were performed using commercially available statistical software (Prism 9, GraphPad Software Inc.).
Differences in maximum concentration (C max ), observed area under the curve (AUC 0-t ), and extrapolated area under the curve (AUC 0-∞ ) between the fasted and fed phases were assessed using the 90% confidence interval bioequivalence method. Mixed effects regression models were developed for log-transformed data of each parameter using formulation, sequence, and period as fixed effects and individual dog as a random effect. The mean squared error was obtained from the analysis of variance and used in the calculation of the 90% confidence intervals. Results are expressed as the ratio of geometric means of the fed vs. fasted phases, relative to the fasted phase, with 90% upper and lower confidence bounds. The back-transformed difference between means is also reported. The formulations were considered bioequivalent for a parameter if the 90% confidence interval fell within 0.80-1.25 (FDA, 2006). Half-life (t 1/2 ), time to maximum concentration (T max ), and mean residence time (MRT) were also compared between the fasted and fed phases and were evaluated using a Wilcoxon signed rank test as these parameters were non-normally distributed based on the Kolmogorov-Smirnov test. p-values <.05 were considered statistically significant.
The effects of phase (fasted vs. fed) and phase order on the occurrence of gastrointestinal adverse effects (vomiting, diarrhea, or both) were evaluated using a mixed effects model with phase and phase order as fixed effects and individual dog as a random effect.

| Study population
Eight dogs were initially enrolled in the study. One dog in Group 1 inadvertently removed the sampling catheter shortly after time 0 sample collection in the first phase and was excluded from the study.
Therefore, a total of seven adult dogs, including one intact female, four spayed females, and two neutered males were included, with a mean age of 4.4 ± 2.3 years (range 1-8 years). The mean dose of telmisartan administered throughout the study was 1.01 ± 0.004 mg/ kg (range 0.99-1.02 mg/kg) ( Table S1).

| Effect of feeding on telmisartan oral solution pharmacokinetics
Pharmacokinetic parameters for TOS with and without food are presented in Table 1. Plasma drug concentrations and pharmacokinetic parameters for individual dogs are presented in Tables S2-S5. Telmisartan concentration-time curves were similar between fasted and fed phases (Figure 1). The ratio of the geometric means of C max of the fed vs. fasted phase was 0.97 (7.48 ng/ml difference) with a 90% confidence interval of 0.74-1.27. The ratio of geometric means of the AUC 0-t of the fed vs. fasted phase was 0.92 (106 ng*h/ml difference) with a 90% confidence interval of 0.81-1.03. The ratio of the geometric mean of AUC 0-∞ of the fed vs. fasted phase was 0.90 (127 ng*h/ml difference) with a 90% confidence interval of 0.82-1.00. There was no significant difference in t 1/2 (p = .375), T max (p > .999), or MRT (p = .938) between phases.  (Table S6). One of the dogs who experienced both diarrhea and vomiting when fasted was found to have ingested at least two socks prior to study admission which were evident in the vomitus. One dog also developed mild, self-resolving generalized tremors during the fasted phase prior to receiving the TOS product. The owner reported these tremors to reflect a historical behavior in response to anxiety or need to eliminate and are not suspected to be a true adverse effect.

| DISCUSS ION
As telmisartan use increases in veterinary clinical practice, it is essential that rational guidelines for dosing and administration be established. Pharmacokinetic parameters have previously been published for oral telmisartan tablets (Micardis®) in dogs, albeit at higher doses (7.6-9 mg/kg) than the currently recommended starting dose (1 mg/kg q 24 h) for treating canine proteinuria (Baek et al., 2013;Brown et al., 2013). Dose directly impacts the pharmacokinetic profile of a drug; thus, it is prudent to evaluate a drug at the dosages used in clinical practice. The recently approved TOS product represents a new formulation of telmisartan that could allow for more accurate dosing in small dogs. Furthermore, an oral solution may be an attractive option for owners who struggle to administer tablets or capsules. However, formulation can impact absorption, so the previously established pharmacokinetics of the oral tablet in dogs, may not be accurate for TOS. Although TOS has already been evaluated and approved for use in cats, the pharmacokinetic parameters established in this species cannot be directly extrapolated to dogs due to the interspecies variability in drug handling (Baek et al., 2013;Food and Drug Administration, 2018). Given the critical differences in dose, product formulation, and species where telmisartan pharmacokinetics were previously investigated, there is a need to document the pharmacokinetic profile of TOS in dogs to allow better dosing recommendations in this species. Additionally, it is well known that food may interact with drug absorption through a variety of mechanisms (Welling, 1996). In humans, food reduces the bioavailability of telmisartan tablets by 6-20%, depending on dose (Boehringer Ingelheim International GmbH, 2018) and in cats, the C max and AUC 0-∞ were higher in fasted animals administered TOS  (Israili, 2000).

TA B L E 1 Summary of noncompartmental analysis of pharmacokinetic parameters after single-dose administration of TOS (1 mg/kg telmisartan) in seven dogs
The terminal half-life of telmisartan is significantly longer in people (t 1/2 = >24 h) compared with dogs. Telmisartan is metabolized almost exclusively via UGT-glucuronosyltransferases in the liver to a pharmacologically inactive metabolite (1-O-acylglucuronide) (Ebner et al., 2013;Plumb, 2019). The rate of this reaction is about 10-fold lower in humans compared with dogs, so differences in clearance likely explain the differences in half-life between these species (Ebner et al., 2013;Toutain & Bousquet-Melou, 2004). This discrepancy could also be explained by extrahepatic glucuronidation in the gastrointestinal tract, alternate hepatic metabolic pathways, or a smaller volume of distribution in dogs but further studies are needed.
We also hypothesized that TOS would be rapidly absorbed in dogs, which was supported by a short T max in both the fasted (1.32 ± 0.66 h) and fed state (1.46 ± 1.18 h). This is similar to the T max for both the commercial telmisartan tablet (0.80 ± 0.36 h) and telmisartan-amlodipine combined tablet (2 h) in dogs (Brown et al., 2013;Wang et al., 2015). Telmisartan T max is also relatively short in humans administered tablets (0.5-1 h) and cats adminis-  Israili, 2000). Whether or not these factors impact extent of absorption is currently unknown since absolute bioavailability has not been established in dogs for any telmisartan formulation to date.
In humans, telmisartan has high inter-individual variability in drug concentration due to its low solubility in biological fluids, low bioavailability (30-60%), and delayed onset of action with high first-pass metabolism following oral administration (Israili, 2000;Patel et al., 2012). High pharmacokinetic variability amongst individuals of the same species makes it challenging to recommend a dose that will be clinically efficacious for all patients. Furthermore, if high inter-individual variability exists within the canine population, it is possible that selecting a clinically efficacious dose for all patients will be even more difficult given further possible differences in drug handling between the various breeds (Fleischer et al., 2008;Sutton, 2004). Telmisartan pharmacokinetics have not been evaluated in a large enough study of dogs to truly characterize population variability. However, we did document a threefold difference between the lowest and highest AUC and a sixfold difference in C max . Thus, similar inter-individual variation in telmisartan disposition could exist in dogs with similar clinical consequences.
Our second objective was to determine the effect of feeding on the absorption of telmisartan in dogs by comparing the C max , AUC 0-t , and AUC 0-∞ between the fasted and fed state. The presence of food in the gastrointestinal tract can affect absorption based on the physiochemical properties of the specific drug by altering the pH and lipophilicity of the intestinal environment, increasing splanchnic blood flow, and prolonging gastrointestinal transit time (Welling, 1996). Species differences in gastrointestinal physiology also influence drug absorption, so administering a drug with food can significantly lower concentration in one species, while having no impact in another (Sutton, 2004). Thus, it is important to evaluate the effect of feeding on telmisartan absorption in dogs despite the available administration recommendations for humans and cats (Boehringer Ingelheim International GmbH, 2018;FDA, 2018). Using the 90% confidence interval method, the telmisartan AUC 0-t and AUC 0-∞ ratios between the fasted and fed phases were within the generally accepted range of 0.80-1.25 but the C max ratio (0.74-1.27) exceeded these bounds (FDA, 2006). By strict interpretive criteria, F I G U R E 1 Plasma telmisartan concentration time course in dogs (n = 7) after oral administration of 1 mg/kg telmisartan as TOS differentiated by phase. Error bars illustrate standard deviation TOS administered with vs. without food cannot be considered bioequivalent. However, the C max confidence interval was wide, reflecting the interindividual variability detected in our study, and fell outside both the lower and upper bounds. Therefore, we cannot make a general recommendation to administer the TOS product with or without food in dogs since the best way to maximize C max is individual-dependent. Based on AUC parameters alone, TOS administered with vs. without food is bioequivalent. As a marker for total drug exposure, AUC is sometimes considered the more important pharmacokinetic parameter, but this should be applied cautiously.
Ultimately, pharmacodynamic studies are necessary to determine which parameters should be targeted (C max vs. AUC vs. time within therapeutic range) to achieve the desired clinical effect. Based on these results, we suggest that TOS can be given to dogs with or without food but should be administered in the same way consistently to avoid day to day variation. Furthermore, dose should be individualized to target clinical parameters such as blood pressure or urine protein:creatinine ratio. (1 mg/kg q 24 h) to provide the initial framework for translating this data to dogs with clinical illness. Finally, we did not perform follow-up biochemical analyses, with specific regard to their renal and electrolyte parameters, which is a documented adverse effect of RAAS inhibition common in dogs and rare in cats and humans (Brown et al., 2013;Food and Drug Administration, 2018;Sadjadi et al., 2009). However, these values were not expected to be abnormal given the one-time administration to healthy adult dogs (Schierok et al., 2001).
Additional studies evaluating TOS and other telmisartan prod- Multidose pharmacokinetic studies would also provide more accurate information about plasma drug concentrations at steady state. The current recommended dose of telmisartan in dogs (1 mg/kg q 24 h) is based on expert opinion (Brown et al., 2013).
Thus, pharmacodynamic studies evaluating markers of RAAS inhibition are ultimately needed to determine rational dosing recommendations in dogs, with follow-up clinical studies evaluating the overall success of treatment and documented incidence of adverse effects. Our study provides the initial framework to advance our understanding of TOS in dogs as it becomes more popular in clinical practice.

| CON CLUS ION
The results of our study demonstrate that single dose pharmacokinetic parameters of TOS appear similar to those for telmisartan tablets in dogs. Thus, TOS can probably be used in a similar manner in clinical animals with the benefit of allowing for more accurate dosing in small dogs and providing a liquid option for owners who struggle to administer tablets or capsules. Although not bioequivalent based on C max , TOS can likely be administered with or without food in dogs as long the drug is administered the same way each day to promote consistent absorption. Additional pharmacokinetic and pharmacodynamic studies are warranted to further develop rational recommendations for telmisartan use in canine medicine.

AUTH O R CO NTR I B UTI O N S
JR and ZL conceived this study. All authors contributed to its execution and data analysis. AB wrote the initial manuscript draft and all authors contributed to and approved the final version.

ACK N OWLED G M ENTS
We would like to thank Gabriela de Oliveira and Jenny Applebaum for assistance with sample collection and care of study participants. We also thank the owners of the dogs that participated in this study.

Funding was provided through the Department of Veterinary Clinical
Medicine at the University of Illinois College of Veterinary Medicine in Urbana, Illinois.

CO N FLI C T O F I NTE R E S T
We have no financial conflicts of interest or disclaimers to disclose.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that supports the findings of this study are available in the supplementary materials of this article.

E TH I C A L A PPROVA L
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received. The authors confirm that they have adhered to either US or European standards for the protection of animals used for scientific purposes.