Cytochrome P450 enzymes (CYPs) are a superfamily of enzymes involved in the metabolism of many endogenous and exogenous compounds in dogs, humans, and numerous other species. In humans, a small number of CYP isoforms contribute to the metabolism of 70% to 80% of clinically relevant drugs.1 Although the percentage of clinically relevant drugs metabolized by canine CYPs is unknown, the contribution is expected to be similar. Changes in the expression and activity of drug-metabolizing CYPs can significantly impact the pharmacokinetics and, consequently, the pharmacodynamics of drugs metabolized by these enzymes. Exogenous and endogenous factors, such as genetic variation, physiological conditions, age, environmental factors, and interactions between drugs or intoxicants, contribute to the modulation of CYP expression and activity. In humans, the interplay among these factors results in substantial interindividual variability in drug disposition and response.2 Recent in vitro research findings3 from our group have demonstrated the considerable variability in hepatic drug-metabolizing CYP abundance within a set of 59 dog livers in a bank representing multiple breeds, with coefficients of variation ranging from 27% to 276% across different CYP isoforms.
In human medicine, genotyping of drug-metabolizing CYPs has been used to help predict the drug-metabolizing phenotype of individual patients (ie, ultrarapid, normal, intermediate, or poor metabolizers) relative to the entire population.4 However, given that interindividual variability in drug metabolism is multifactorial, of which genetics is only one facet, there can be genotype-phenotype discordance when the genotype-predicted metabolic activity of a patient does not match the patient's measured phenotype.5 Currently, the most accurate approach to determining the drug-metabolizing phenotype of an individual at a given time point is through administration of a CYP-specific probe drug and subsequent measurement of the parent drug-to-metabolite ratios (MRs).2,6 The CYP phenotyping cocktails, which are a combination of CYP-specific probe drugs administered together to simultaneously, but independently, assess the activity of different CYP isoforms, have been used in human medicine since the 1990s. These phenotyping cocktails are used to investigate drug interactions, determine the effect of disease states on drug metabolism, and assess genotype effects on CYP activity and phenotype patients in a clinical setting. The first human phenotyping cocktails involved frequent blood sampling procedures to determine areas under the concentration-versus-time curves (AUCs) for probe drugs and metabolites or metabolic clearance of the probe drug. More recently developed cocktails have moved toward using limited or single time point blood sampling or noninvasive sampling of saliva, urine, hair, and exhaled breath to improve clinical applicability.7,8 With veterinary pharmacogenomics in its relative infancy as compared to its human counterpart, such tools to help predict genotype-activity relationships and study the phenotypic variability of CYP activity in the broader canine population do not yet exist but are needed to accelerate the field.
Currently, there is 1 report9 of employing a CYP probe cocktail in dogs to assess drug-drug interactions. However, the cocktail drugs used had only been validated for human CYPs and have since been shown to not be probe drugs for canine CYP orthologs. Additionally, the cocktail was not validated as no interaction studies were undertaken to ensure that the probe drugs did not interfere with the metabolism of each other when administered together in dogs.
The goal of this work was to develop a tool (the Program in Individualized Medicine [PrIMe] cocktail) that can be utilized to evaluate drug-drug interactions, aid in the identification of novel pharmacogenomic variants, assess the effect of environmental and physiological factors on CYP-mediated drug metabolism, and aid in individualized drug dose selection in dogs. Our primary objective was to develop a CYP phenotyping cocktail for dogs using drugs shown to be specific CYP substrates at clinical (or lower) doses to minimize probe drug exposure and the risk for adverse events. The main hypothesis tested was that plasma probe drug area under the concentration-versus-time curve from 0 to 6 hours (AUC0–6h) to plasma metabolite AUC0–6h MRs after single drug administration was not significantly different from cocktail administration of all 3 drugs to the same 12 dogs. A secondary objective was to determine whether alternative sampling methods (saliva and urine) or single time point samples (plasma, saliva, or urine) could be used in place of multiple blood sampling to improve the clinical acceptability of the PrIMe cocktail for larger population studies.
Methods
Subjects
This study was approved by the Washington State University IACUC (protocol No. 6115), and all owners gave written informed consent for their dogs to participate in the study before conducting the study.
Without prior published data for dogs, the required sample size estimation was based on phenotyping probe cocktail development studies in humans using crossover designs, which indicated an expected 5% to 20% intrasubject coefficient of variation for the plasma AUC ratios of the cocktail probe drugs.8 Using a standard bioequivalence acceptance range of 0.8 to 1.25, a plasma AUC0–6h MR probe cocktail-to-AUC0–6h MR single drug ratio value of 1.0 (indicating an absence of interaction), power set at 0.8, and α = 0.05, with an expected variability of 20%, the minimum sample size was estimated to be 10 dogs.10 Consequently, 12 healthy-owned dogs were recruited for the study from the staff, faculty, and student population at the College of Veterinary Medicine at Washington State University. There were 4 male and 8 female dogs of various breeds, including 5 mixed-breed dogs. Median age was 5.7 years (range, 2.8 to 10.2 years) and median weight was 25 kg (range, 21 to 36 kg). This information for individual dogs is provided (Supplementary Table S1). The health status of each dog was determined by physical examination, medical history, CBC, and serum biochemistry profile conducted at the Washington State University Veterinary Teaching Hospital. All dogs were spayed or neutered and were not receiving any medications or supplements, except for lokivetmab injections or monthly flea, tick, and parasite preventatives. For dogs receiving these medications, the study period did not commence until at least 2 weeks after their last dose. Dogs were housed in kennels at the Washington State University Veterinary Teaching Hospital on study days and provided ad libitum access to water.
Study design
This was a single-center, open-label, randomized, 4-way crossover study performed at the Washington State University Veterinary Teaching Hospital. Each dog received delayed-release omeprazole tablets (CYP3A12 probe; 40 mg; Prilosec; Procter & Gamble); dextromethorphan HBr capsules (CYP2D15 probe; 30 mg; Robitussin Long-Acting Cold; Haleon Group); bupropion hydrochloride tablets, USP (CYP2B11 probe; 75 mg); or all 3 drugs in combination (PrIMe cocktail). The treatment order was randomized so that the same treatment sequence was never repeated among the 12 dogs. To avoid any possible carryover effects, a washout time of at least 14 days was employed between each study period. On the morning of each study period, dogs arrived fasted overnight and were given a brief physical exam. If dogs were deemed healthy, they were offered their normal morning food ration, and 2 hours later, the study commenced. The individual drugs or cocktail were wrapped in a small piece (roughly 6.5 cm2) of commercially available sliced processed cheese product, effectively serving as a pill pocket, and orally administered to the dogs. Six hours after the drug or drug combination was administered, the dogs were released to their owners. Owners were contacted 24 hours later to inquire about possible observed adverse events.
Heparinized blood samples (3 mL) were collected by cephalic, saphenous, or jugular venipuncture before and at 1, 2, 4, and 6 hours after drug administration. Blood tubes were centrifuged at 1,800 X g for 8 minutes, and the plasma was transferred to cryovials, which were stored at −80 °C until analysis.
Saliva (approx 1 mL) was collected using a saliva collection system (Salivette; cortisol; Sarstedt AG & CO) before, immediately after, and at 0.5. 1, 1.5, 2, 4, and 6 hours after drug administration. Saliva was collected by inserting one end of the swab into the cheek pouch of the dog for 90 seconds on each side while holding a piece (approx 2.5 cm) of low-fat mozzarella cheese near the nose of the dog allowing the cheese to be licked but not ingested. After centrifugation (1,000 X g for 2 minutes at 4 °C) per the manufacturer's instructions, the supernatant was collected and stored at–80 °C until analysis.
Urine was collected by free catch before and 6 hours after drug administration. Urine was split into 2 samples with 1 sample titrated to pH 10 with 4 M sodium carbonate buffer and the other not treated. Urine samples were stored at −80 °C until analysis.
Bioanalysis
Plasma, saliva, and urine concentrations of the cocktail probe drugs (bupropion, dextromethorphan, and omeprazole) and their metabolites (6-hydroxybupropion, dextrorphan, and omeprazole sulfone, respectively) were determined by LC-MS/MS. Sample preparation was modified from previously published methods.8,11 Briefly, 25 µL of plasma or saliva was mixed with 100 or 75 µL, respectively, of an internal standard solution (50:50 vol/vol acetonitrile:methanol containing 12.5 ng/mL each of pantoprazole and GW340416A, a chemical analog of bupropion generously provided by GlaxoSmithKline), vortexed, incubated on ice for 10 minutes, and centrifuged (15,000 X g for 10 minutes at 4 °C). To measure deconjugated dextrorphan, 12.5 µL of plasma or saliva was combined with 12.5 µL ammonium acetate (100 mM, pH 5) containing 1,000 U/mL β-glucuronidase/sulfatase from Helix pomatia Type HP-2 to achieve 12.5 U of enzyme per reaction and incubated overnight at 37 °C. Deconjugated samples were then prepared as described above. The plasma or plasma preparations were diluted 1:1 with ultrapure water, and 20 µL was injected into the LC-MS/MS system.
For urine analysis, 3 different preparations were performed: 1 for buffered urine (bupropion/6-hydroxyburpropion and omeprazole/omeprazole sulfone), 1 for unbuffered urine (bupropion/6-hydroxybupropion, dextromethorphan/dextrorphan, and omeprazole/omeprazole sulfone), and 1 for unbuffered urine after enzymatic deconjugation (dextromethorphan/dextrorphan). Previous analytical assays to quantify omeprazole and metabolites in human urine buffered urine samples to a pH greater than 7 to prevent acidic degradation of the analytes.12,13 For buffered urine, 25 µL of buffered urine was mixed with 25 µL of ultrapure water. Then, 10 µL of the mixture was transferred to a clean microcentrifuge tube and mixed with 100 µL of urine internal standard solution (50:50 vol/vol acetonitrile:methanol containing 25 ng/mL each of pantoprazole and GW340416A), vortexed, incubated on ice for 10 minutes, and centrifuged (15,000 X g for 10 minutes at 4 °C). For unbuffered urine, the same protocol for buffered urine was employed except that the urine was initially mixed with 25 µL of ammonium acetate (100 mM, pH 5) instead of ultrapure water. To measure deconjugated dextrorphan in the unbuffered urine, 25 µL of unbuffered urine was combined with 25 µL of ammonium acetate (100 mM, pH 5) containing 1,000 U/mL β-glucuronidase/sulfatase from H pomatia type HP-2 to achieve 125 U of enzyme per reaction and incubated overnight at 37 °C. Then, 10 µL of the deconjugated urine mixture was prepared as with the other 2 urine protocols. For all urine protocols, supernatants were diluted 1:1 with ultrapure water and 10 µL was injected into the LC-MS/MS.
Stock solutions of analytes were prepared in methanol. All stock and internal standard solutions were stored at −20 °C. Working calibrant spiking solutions containing the analytes were prepared by dilution of the stock solutions in methanol and used to enrich blank pooled dog plasma, saliva, and urine to create calibration curves and quality control (QC) samples. Calibration curves and QCs were prepared as described for the samples above. Calibration ranges are provided (Supplementary Table S2). Chromatographic separation, data collection, integration, and quantification were performed as previously described for omeprazole sulfonation with the same instrumentation and conditions for all drugs and metabolites in this study.3 The total run time was 5 minutes. The transitions (m/z) for each analyte were as follows: omeprazole, 346.2 → 198.2; omeprazole sulfone, 362.0 → 150.1; pantoprazole, 384.2 → 200.1; bupropion, 240.2 → 131.1; 6-hydroxybupropion, 256.3 → 238.0; GW340416A, 238.0 → 182.0; dextromethorphan, 272.3 → 215.1; and dextrorphan 258.2 → 133.1. Pantoprazole was used as the internal standard for omeprazole and omeprazole sulfone while GW340416A was used as the internal standard for the remaining analytes. Interassay accuracy (determined as the percent bias) for plasma, saliva, and urine quality control samples ranged from –14.4 to 12.5, and interassay precision (determined as the percent coefficient of variation) was less than 12.4 for all analytes in all matrices. Interassay QC biases and precisions are reported (Supplementary Tables S3–S6). The lower limits of quantification (LLOQs) in plasma were 0.1 ng/mL for dextromethorphan and 0.5 ng/mL for remaining analytes. The LLOQs for saliva were 0.05 ng/mL for dextromethorphan; 0.1 ng/mL for omeprazole, omeprazole sulfone, and bupropion; and 0.25 ng/mL for hydroxy bupropion and dextrorphan. The LLOQs for urine were 1.0 ng/mL for omeprazole, 2.5 ng/mL for bupropion, 6.25 ng/mL for omeprazole sulfone and dextromethorphan, and 12.5 ng/mL for dextrorphan.
Data analysis
The AUC0–6h for drugs and metabolites in plasma and saliva were calculated using the log-linear trapezoidal method with commercially available software (Phoenix 64 WinNonlin v. 8.3; Certara).
Bioequivalence tests were performed using the plasma AUC0–6h MRs (ratio of parent drug AUC divided by drug metabolite AUC) of probe drugs given individually and together as a cocktail with a linear mixed-effects model using the abovementioned pharmacokinetic software. Geometric mean plasma AUC0–6h MRs and 90% CI bioequivalence acceptance limits were set at 0.8 to 1.25.14
The plasma AUC0–6h MRs of probe drugs given individually and together as a cocktail was also assessed for statistical difference using a paired t test on log-transformed data. Single time point MRs (ratios of parent drug concentration divided by drug metabolite concentration determined at the stated time) were determined for plasma, saliva, and urine at each time point collected. Results are presented as geometric mean ratios and 90% CI. Correlations between AUC0–6h MRs and single time point MRs were established using Spearman's rank correlation coefficient. For all statistical tests, a P value of .05 was considered statistically significant. All statistical analyses were performed with commercially available software (Sigma Plot 11.0; Systat Software Inc).
Results
Twelve healthy client-owned dogs of varying breeds, sex, body weights, and age successfully completed the 4 study sessions. None of the subjects were observed to experience adverse effects by the investigators after a single drug or cocktail drug administration at the study site or by the owners up to 24 hours after each session.
Concentrations of each probe drug and metabolite measured in plasma, saliva and urine samples after single drug and cocktail drug administration are provided (Supplementary Table S7).
Plasma AUC0–6h MRs of the probes given alone and as a cocktail were compared using paired t tests on log-transformed data to evaluate potential pharmacokinetic interactions between probe drugs (Table 1 and Figure 1). Plasma AUC0–6h MRs for dextromethorphan, omeprazole, and bupropion given alone or as a cocktail was also not statistically significantly different (P < .05). Bioequivalence testing was also performed for each probe drug by comparing the observed 90% CIs for the geometric mean of the ratios of cocktail-dosed plasma AUC0–6h MRs to the individually dosed AUC0–6h MRs to the limits established by the FDA for drug bioequivalence (0.8 to 1.25). As shown, 90% CIs for both omeprazole and bupropion plasma AUC0–6h MR ratios fell within the established bioequivalence limits, while the lower 90% CI for dextromethorphan (0.76) was slightly below the established lower limit of 0.8.
Plasma area under the concentration-versus-time curve from 0 to 6 hours (AUC0–6h) of cytochrome P450 (CYP) probe drug-to-metabolite ratios (AUC0–6h MRs) for probe drugs omeprazole (40 mg), bupropion (75 mg), or dextromethorphan (30 mg) administered PO alone or as a combination of all 3 drugs (Program in Individualized Medicine [PrIMe] cocktail) to 12 client-owned dogs during a randomized crossover study with at least 14 days between periods.
| Ratio of MRs | 90% CI for the ratio of MRs | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| CYP | Probe/metabolite | Probe alone | 90% CI | Cocktail | 90% CI | P value* | Cocktail/probe alone | Lower limit | Upper limit |
| CYP3A12 | Omeprazole/omeprazole sulfone | 2.55 | 2.27–2.87 | 2.48 | 2.15–2.87 | .52 | 0.97 | 0.91 | 1.04 |
| CYP2B11 | Bupropion/6-hydroxybupropion | 0.14 | 0.087–0.24 | 0.14 | 0.085–0.24 | .81 | 0.99 | 0.90 | 1.08 |
| CYP2D15 | Dextromethorphan/dextrorphan | 0.026 | 0.018–0.038 | 0.023 | 0.017–0.031 | .20 | 0.84 | 0.76 | 1.05 |
Data are presented as the geometric mean and 90% CI of the AUC0–6h MRs for each probe drug and dosing method (individual probe drug administered alone or together as the 3-probe drug cocktail). Also shown are the geometric mean and 90% CI of the ratios of the cocktail dosed divided by the individually dosed AUC0–6h MR values.
P values obtained using a paired t test on log-transformed data comparing MRs for probe drugs given separately or together as a cocktail.
Plasma area under the concentration-versus-time curve from 0 to 6 hours (AUC0–6h) of cytochrome P450 (CYP) probe drug-to-metabolite ratios (MRs) for omeprazole/omeprazole sulfone (CYP3A12 probe; A), bupropion/6-hydroxybupropion (CYP2B11 probe; B), and dextromethorphan/dextrorphan (CYP2D15 probe; C) after oral administration of parent probe drug omeprazole (40 mg), bupropion (75 mg), or dextromethorphan (30 mg) alone or as a combination of all 3 drugs (Program in Individualized Medicine [PrIMe] cocktail) to the same 12 client-owned dogs during a randomized crossover study with at least 14 days between periods. Open circles are plasma MRs for each dog connected by a line to show the change in the MR that occurred within individual subjects when probe drugs were given as single doses or together as a cocktail. Also shown are P values obtained using a paired t test on log-transformed data comparing MRs of probe drugs given as single doses or together as a cocktail.
Citation: American Journal of Veterinary Research 85, 7; 10.2460/ajvr.24.02.0049
As shown (Figure 2), the plasma concentration-versus-time profiles for each probe drug and their respective metabolite when administered alone were similar to profiles observed after cocktail dosing. The between-subject variability (assessed as percent coefficient of variation) for plasma AUC0–6h MRs varied greatly between probes. The plasma AUC0–6h MR for bupropion (CYP2B11) was found to have the greatest interindividual variability ranging from 143% to 151%, while omeprazole plasma AUC0–6h MR (CYP3A12) had the least interindividual variation ranging from 25.2% to 31.7%). The plasma AUC0–6h MR for dextromethorphan showed intermediate interindividual variability (69.9% to 96.2%). The within-subject variability (comparing single versus cocktail dosing) for the plasma AUC0–6h MRs was considerably less than the intersubject variability at 9%, 12.0%, and 22% for omeprazole, bupropion, and dextromethorphan, respectively.
Plasma concentration-versus-time profiles of cytochrome P450 (CYP) probe drugs (and associated metabolites) obtained after administration separately (as single doses) and together (as a cocktail dose) to the same 12 owned dogs during a randomized crossover study with at least 14 days between periods. The quantified probe drugs and the corresponding metabolites included omeprazole and omeprazole sulfone for CYP3A12 (A), bupropion and 6-hydroxybupropion for CYP2B11 (B), and dextromethorphan and dextrorphan for CYP2D15 (C). Data points and error bars represent the mean and SD of measured concentrations (n = 12).
Citation: American Journal of Veterinary Research 85, 7; 10.2460/ajvr.24.02.0049
Single time point plasma MRs were correlated with the plasma AUC0–6h MRs when given as a cocktail (Table 2). Only correlations for time points in which all 12 dogs had probe drug and metabolite concentration above the LLOQ were examined. These included 4 and 6 hours postdose MRs for omeprazole, 4 hours postdose for bupropion MRs, and 2 to 6 hours postdose MRs for dextromethorphan. The 4-hour single time point plasma MR strongly (r ≥ 0.853) and significantly (P < .001) correlated with the plasma AUC0–6h MRs for all 3 probes and was the only time point for bupropion where all 12 dogs had quantifiable concentrations of the parent drug and metabolites.
Correlations between single time point probe drug-to-metabolite ratios (MRs) and area under the concentration-versus-time curve from 0 to 6 hours MRs (AUC0–6h probe drug/AUC0–6h metabolite) measured in plasma after cocktail (combined) administration of 3 cytochrome P450 (CYP) probe drugs to 12 owned dogs.
| Spearman correlation coefficient (P value) | |||||
|---|---|---|---|---|---|
| Time point (hours) | |||||
| CYP | Probe drug/metabolite AUC0–6h MR | 1 | 2 | 4 | 6 |
| CYP3A12 | Omeprazole/omeprazole sulfone | 0.860 | 0.224 | ||
| (< .0001) | (.47) | ||||
| CYP2B11 | Bupropion/6-hydroxybupropion | 0.965 | |||
| (< .0001) | |||||
| CYP2D15 | Dextromethorphan/dextrorphan | 0.839 | 0.853 | 0.699 | |
| (< .0001) | (< .0001) | (.010) | |||
Shown are the Spearman correlation coefficients (r) and the associated P values for each correlation. Only correlations for time points in which all 12 dogs had probe drug and metabolite concentrations above the lower limit of quantification are shown.
Correlation plots between AUC0–6h MRs for saliva and plasma samples following cocktail administration are shown (Figure 3). Saliva and plasma AUC0–6h MRs for omeprazole and dextromethorphan displayed strong, significant correlations (r ≥ 0.825, P <.0001), while saliva and plasma AUC 0-6h MRs for bupropion were not correlated (r = 0.056, P = .85). Single time point saliva MRs were correlated with the plasma AUC0–6h MRs for omeprazole and dextromethorphan when drugs were administered together in the cocktail for the 2 hours and 4 hours time points (Table 3). For omeprazole, there were only 2 time points (4 and 6 h) where all 12 dogs had probe drug and metabolite concentrations exceeding the LLOQs in saliva for the bioanalytical method. However, only the 4-hour saliva time point MR had a statistically significant correlation with the corresponding plasma AUC0–6h MR (r = 0.587; P = .042). A much stronger statistically significant correlation (r = 0.929; P < .0001) was observed at the 2 hours time point between the saliva single time point MR and plasma AUC0–6h MR but only 7 (of 12) dogs had quantifiable concentrations of the parent drug and metabolites at this time point. There were no saliva time points for which dextromethorphan and dextrorphan were above quantifiable concentrations in all 12 dogs. Only 2 saliva time point MRs (2 and 4 h) were found to be significantly correlated with plasma AUC0–6h MR for dextromethorphan (r = 0.782; P = .0052 and r = 0.770; P= .0069, respectively). At these time points, 10 (of 12) dogs had quantifiable concentrations of both dextromethorphan and dextrorphan.
Correlations between probe drug-to-metabolite ratios (MRs) measured in plasma and saliva (A, C, and E) or plasma and unbuffered 6-hour urine (B, D, and F) after cocktail administration of 3 cytochrome P450 (CYP) probes (omeprazole, bupropion, and dextromethorphan) to 12 owned dogs during a randomized crossover study with at least 14 days between periods. A and B—Omeprazole/omeprazole sulfone MRs for CYP3A12. C and D— Bupropion/6-hydroxybupropion MRs for CYP2B11. E and F—Dextromethorphan/dextrorphan MRs for CYP2D15. Dotted lines represent the 95% confidence interval. Also shown are the Spearman correlation coefficients (r), the associated P values, and the number of dogs included in each analysis that had probe drug and metabolite concentrations above the lower limit of quantitation at 6 hours postdose.
Citation: American Journal of Veterinary Research 85, 7; 10.2460/ajvr.24.02.0049
Correlations between single time point probe drug-to-metabolite ratios (MRs) measured in saliva and area under the concentration-versus-time curve from 0 to 6 hours MRs (AUC0–6h probe drug/AUC0–6h metabolite) measured in plasma after cocktail (combined) administration of 3 cytochrome P450 (CYP) probe drugs to 12 owned dogs.
| Spearman correlation coefficient (P value) | |||||||
|---|---|---|---|---|---|---|---|
| Time point (hours) | |||||||
| CYP | Probe drug/metabolite AUC0–6h MR | 0.5 | 1 | 1.5 | 2 | 4 | 6 |
| CYP3A12 | Omeprazole/omeprazole sulfone | 1.000 | 0.750 | 0.929 | 0.587 | 0.531 | |
| (.017) | (.038) | (< .0001) | (.042) | (.071) | |||
| n = 5 | n = 7 | n = 7 | n = 12 | n = 12 | |||
| CYP2D15 | Dextromethorphan/dextrorphan | 0.500 | −0.183 | 0.550 | 0.782 | 0.770 | 0.503 |
| (1.0) | (.61) | (.11) | (.0052) | (.0069) | (.13) | ||
| n = 3 | n = 9 | n = 9 | n = 10 | n = 10 | n = 10 | ||
Shown are the Spearman correlation coefficients (r) and the associated P values for each correlation as well as the number (n) of dogs that had probe drug and metabolite concentrations above the lower limit of quantitation at each time point.
Drug and metabolite concentrations were compared between alkaline buffered urine and unbuffered urine samples for omeprazole and bupropion. Due to acidic requirements for deconjugation, dextromethorphan and dextrorphan concentrations were only determined in unbuffered urine. Bupropion and 6-hydroxybupropion concentrations were not different between buffered and unbuffered urine samples (P = .52 and P = .99, respectively). However, mean omeprazole concentrations in buffered urine were nearly 2-fold higher in buffered urine (28.0 ng/mL) compared to unbuffered urine (14.3 ng/mL; P = .03). In 1 dog, urine omeprazole was below the LLOQ in both buffered and unbuffered urine. No significant difference was found for omeprazole sulfone concentration between buffered and unbuffered urine (P = .46). However, 4 dogs had omeprazole sulfone concentrations that fell below the detectable limit in buffered urine, while all dogs had quantifiable omeprazole sulfone in unbuffered urine samples. Therefore, MR analyses for omeprazole were performed using results from unbuffered urine samples.
Correlation plots between 6-hour (unbuffered) urine MRs and the corresponding plasma AUC0–6h MRs are shown (Figure 3). Bupropion and dextromethorphan 6-hour urine MRs correlated significantly with plasma AUC0–6h MRs with probe drug and metabolite concentrations above the LLOQ in urine samples from all dogs. Conversely, 6-hour urine MRs for omeprazole did not correlate with omeprazole plasma AUC0–6h MR and (as mentioned above) not all dogs had quantifiable concentrations of omeprazole or omeprazole sulfone in urine regardless of urine sample treatment (buffered or unbuffered).
Discussion
In this study, the combination of probe drugs for simultaneous phenotyping of 3 major canine drug metabolizing CYPs in plasma, saliva, and urine was investigated. To our knowledge, this is the first probe cocktail developed and validated for dogs using canine CYP-specific probe drugs. Our primary goal was to develop a phenotyping cocktail using clinical (or lower) doses of probe drugs to minimize the risk of adverse effects and potential drug-drug interactions so that the cocktail could also be broadly and safely used. While only a small number of dogs were enrolled in this study, the selected drug combination and doses did not lead to any observed adverse events in this study population. The probe drugs used in this study are commercially available and accessible globally, which should enhance adoption by other research groups.
An important characteristic of a robust phenotyping cocktail is that there should be minimal pharmacokinetic interactions between individual drugs that could obfuscate the true phenotype of a subject. Historically, such pharmacokinetic interactions have been evaluated statistically using paired t tests (comparing MRs between single and cocktail dosing) and/or by bioequivalence analysis (determining whether the 90% CI of the ratio of MRs falls within the limits accepted by the FDA, usually 0.80 to 1.25).8 Paired t test analysis confirmed that the plasma AUC0–6h MRs of all 3 drugs were not statistically different when dosed alone compared with dosing together. This finding is not surprising given that the drugs selected for this cocktail were known to be selectively metabolized by different CYP isoforms, which would minimize the risk for a pharmacokinetic drug-drug interaction resulting from P450 enzyme inhibition.3,15,16 Bioequivalence was also met for cocktail versus individual dosing of omeprazole and bupropion. However, the lower 90% confidence interval for dextromethorphan was 0.76, falling just below the 0.8 cutoff. Inhibition of dextromethorphan metabolism by another drug in the cocktail is unlikely since this should have resulted in an increase rather than a decrease in the dextromethorphan plasma AUC0–6h MR for cocktail versus single-dose administration. This study was designed to include a sufficient sample size to establish bioequivalence in a crossover study with the assumption that intraindividual variability in the ratio of plasma AUC0–6h MRs should range from 5% to 20%. While this was true for omeprazole and bupropion, intraindividual variability for dextromethorphan was 22%. Future studies using dextromethorphan as a phenotyping probe in dogs should be designed to account for this unexpectedly high variability.
To enhance the utility of the PrIMe cocktail for much larger population studies, a single sampling time point would be ideal as has been the case with human phenotyping cocktails. A single blood sampling time point at 4 hours postdose for all 3 CYP isoforms with good correlations between the MR at 4 hours and plasma AUC0–6h MR was identified. Time points before or after 4 hours could not be used because not all dogs had quantifiable concentrations for all drugs and metabolites. It is unknown whether the single time point would be sufficient to assess CYP induction and inhibition or whether the 4-hour correlation would remain stable in conditions of altered CYP function, and thus, this should be investigated in future studies.
Noninvasive sampling would also increase the owner acceptability of cocktail probe studies. Saliva AUC0–6h MRs for omeprazole and dextromethorphan were strongly and significantly correlated with corresponding plasma AUC0–6h MRs indicating that frequent saliva sampling rather than plasma sampling for omeprazole and dextromethorphan phenotyping may be used. Furthermore, single saliva time point MRs at 4 hours for omeprazole and dextromethorphan were found to be significantly correlated with the respective plasma AUC0–6h MRs suggesting that a single 4-hour saliva time point could also be used for these probes.
Bupropion and dextromethorphan 6-hour urine MRs significantly correlated with plasma AUC0–6h MRs indicating that a single 6-hour urine sample could be used for phenotyping instead of multiple blood sampling. As with a single time point sampling of plasma, it is unknown whether the single time point sampling saliva or urine identified here would be sufficient to assess CYP induction and inhibition or whether the single time point correlations would remain stable in conditions of altered CYP function, and thus, this should be investigated in future studies.
Dextromethorphan, an antitussive, is a specific substrate for CYP2D15 in vitro but to our knowledge has not yet been used for phenotyping dogs as a single probe drug or as a cocktail.3,15,16 Dextromethorphan is occasionally used in dogs with atopic dermatitis to reduce repetitive behavior and is rarely used as an antitussive due to a short half-life and poor oral bioavailability.17 Therapeutic doses (2 mg/kg, twice daily) of dextromethorphan are well tolerated in dogs and adverse effects are uncommon, but lethargy has been reported.18 In the PrIMe cocktail, 30 mg of dextromethorphan was administered to all dogs (weight range, 21 to 36 kg) resulting in a weight normalized dose range of 0.8 to 1.4 mg/kg, which is about half of the clinically used dose. For smaller dogs, a single gel capsule of dextromethorphan (15 mg) could likely be used.
One dog was found to have high dextromethorphan MRs suggesting slow metabolism. Interestingly, this dog's littermate appeared to be an intermediate metabolizer compared to the other dogs. The littermates were English Springer Spaniels. It is possible that the outlier was homozygous for a CYP2D15 mutation associated with reduced activity while the littermate was heterozygous.19 While dogs were not genotyped for any known CYP mutations in this study, it does show the sensitivity of the probe to detect differences in CYP activity, which may be helpful in identifying new clinically relevant genotypes.
Bupropion, an antidepressant, is an established probe drug for human CYP2B6 that we have recently validated as an in vitro probe for the canine ortholog CYP2B11.15 Although bupropion is not routinely used clinically in dogs, no observable adverse effects were reported at chronic doses as high as 150 mg/kg/day in preclinical toxicology studies.20 In this cocktail, bupropion was administered at a dose of 75 mg to all dogs (weight normalized dose range 2 to 3.6 mg/kg), which is well below the highest tested (no observable adverse effect) dose for any sized dog. Bupropion MRs showed the largest interindividual variation in MRs among the 3 probe drugs (143% to 151% variability). We have previously reported breed-associated differences in CYP2B11 abundance in liver microsomes but not for other CYPs.3 Given the diversity of breeds included in this study, a large interindividual variation in CYP2B11 activity is somewhat expected. Additionally, this further demonstrates the sensitivity of the PrIMe cocktail to detect differences in CYP-mediated metabolism.
Sulfonation of omeprazole, a proton pump inhibitor, is an established probe activity for CYP3A4 metabolism in humans.21 However, omeprazole is more often used in humans as a probe for CYP2C19 activity, which is responsible for the hydroxylation of omeprazole to 5-hydroxyomeprazole. We have previously shown omeprazole sulfonation to be an in vitro probe for canine CYP3A12 activity.15 Omeprazole is commonly used in dogs to treat gastroduodenal ulcers, Helicobacter spp infection, gastritis, reflux esophagitis and to prevent or treat gastric erosions caused by ulcerogenic drugs such as NSAIDs.22 Although omeprazole is typically used in dogs at a dose of 0.5 to 1 mg/kg daily, doses up to 2.6 mg/kg once daily or up to 2 mg/kg twice daily may be used for certain conditions, such as severe gastrinoma or esophagitis.22 The most common adverse effects are diarrhea and related gastrointestinal distress.22 In the PrIMe cocktail, all dogs received 40 mg of omeprazole resulting in a weight-normalized dose range of 1.1 to 1.9 mg/kg, which is less than the highest dose that is used clinically. As with dextromethorphan, a smaller dose (20 mg) could be administered to dogs under 21 kg to maintain doses at or below those used clinically.
Unfortunately, we were not able to confirm the utility of urine MRs for all P450 probes. In urine, omeprazole MRs were not correlated with that observed in plasma. Assay of omeprazole using buffered urine samples has been reported for P450 cocktail phenotyping in humans, although assay of plasma samples is more common.23 Omeprazole is acid labile, and while we examined both unbuffered and buffered urine, intended to help stabilize omeprazole, it is possible that the omeprazole degraded in the bladder before collection resulting in poor correlation between plasma AUC0–6h and 6-hour urine MRs.
In conclusion, we have shown that the combination of 3 phenotyping probe drugs contained in the PrIMe cocktail can be administered simultaneously without causing significant pharmacokinetic interactions between drugs. The probe drugs given individually or together were also well tolerated in this study population, which varied in breed, age, sex, and weight. Single time point sampling of blood at 4 hours after dosing could be used as an alternative to multiple blood sampling for all probe drugs. Furthermore, noninvasive sampling of urine (for dextromethorphan and bupropion MRs) and saliva (for omeprazole and dextromethorphan) could also be viable alternative collection techniques. Potential future applications of this technology would include investigations of dog breed-associated differences in P450 drug metabolism, as well as identification of other intrinsic (genes, sex, age, and disease) and extrinsic (diet, dietary supplements, and coadministered drugs) factors that result in variable drug pharmacokinetics in dogs.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
None reported.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
This work was supported by the American Kennel Club Canine Health Foundation (grants 2242 and 2529) and the William R. Jones Endowment at Washington State University College of Veterinary Medicine.
References
- 1.↑
Zanger UM, Schwab M. Cytochrome P450 enzymes in drug metabolism: regulation of gene expression, enzyme activities, and impact of genetic variation. Pharmacol Ther. 2013;138(1):103–141.
- 2.↑
Tracy TS, Chaudhry AS, Prasad B, et al. Interindividual variability in cytochrome P450-mediated drug metabolism. Drug Metab Dispos. 2016;44(3):343–351. doi:10.1124/dmd.115.067900
- 3.↑
Martinez SE, Shi J, Zhu HJ, Perez Jimenez T, Zhu Z, Court MH. Absolute quantitation of drug-metabolizing cytochrome P450 enzymes and accessory proteins in dog liver microsomes using label-free standard-free analysis reveals interbreed variability. Drug Metab Disposit. 2019;47(11):1314–1324. doi:10.1124/dmd.119.088070
- 4.↑
Whirl-Carrillo M, Huddart R, Gong L, et al. An evidence-based framework for evaluating pharmacogenomics knowledge for personalized medicine. Clin Pharmacol Ther. 2021;110(3):563–572. doi:10.1002/cpt.2350
- 5.↑
Shah RR, Smith RL. Addressing phenoconversion: the Achilles' heel of personalized medicine. Br J Clin Pharmacol. 2015;79(2):222–240. doi:10.1111/bcp.12441
- 6.↑
Tremmel R, Hofmann U, Haag M, Schaeffeler E, Schwab M. Circulating biomarkers instead of genotyping to establish metabolizer phenotypes. Annu Rev Pharmacol Toxicol. 2024;64(1):65–87. doi:10.1146/annurev-pharmtox-032023-121106
- 7.↑
De Kesel PMM, Lambert WE, Stove CP. Alternative sampling strategies for cytochrome P450 phenotyping. Clin Pharmacokinet. 2016;55(2):169–184. doi:10.1007/s40262-015-0306-y
- 8.↑
Donzelli M, Derungs A, Serratore M, et al. The basel cocktail for simultaneous phenotyping of human cytochrome P450 isoforms in plasma, saliva and dried blood spots. Clin Pharmacokient. 2014;53(3):271–282. doi:10.1007/s40262-013-0115-0
- 9.↑
Zhao Y, Liang A, Zhang Y, Li C, Yi Y, OG N. Impact of tetrahydropalmatine on the pharmacokinetics of probe drugs for CYP1A2, 2D6 and 3A isozymes in beagle dogs. Phytotherap Res. 2016;30(6):906–914.
- 10.↑
Dubins D. Free Analysis Research Tool for Sample Size Iterative Estimation. Version 2.9. Accessed February 21, 2024. http://individual.utoronto.ca/ddubins
- 11.↑
Grangeon A, Gravel S, Gaudette F, Turgeon J, Michaud V. Highly sensitive LC-MS/MS methods for the determination of seven human CYP450 activities using small oral doses of probe-drugs in human. J Chromatogr B Analyt Technol Biomed Life Sci. 2017;1040:144–158.
- 12.↑
Mihaly GW, Prichard PJ, Smallwood RA, Yeomans ND, Louis WJ. Simultaneous high-performance liquid chromatographic analysis of omeprazole and its sulphone and sulphide metabolites in human plasma and urine. J Chromatogr B Biomed Sci Appl. 1983;278:311–319. doi:10.1016/S0378-4347(00)84790-8
- 13.↑
Naesdal J, Andersson T, Bodemar G, et al. Pharmacokinetics of [14C] omeprazole in patients with impaired renal function. Clin Pharmacol Ther. 1986;40(3):344–351. doi:10.1038/clpt.1986.186
- 14.↑
Guidance for Industry #224: Bioequivalence: Blood Level Bioequivalence Study. International Cooperation on Harmonisation of Technical Requirements for Registration of Veterinary Medicinal Products; 2016.
- 15.↑
Martinez SE, Andresen MC, Zhu Z, Papageorgiou I, Court MH. Pharmacogenomics of poor drug metabolism in greyhounds: cytochrome P450 (CYP) 2B11 genetic variation, breed distribution, and functional characterization. Sci Rep. 2020;10(1). doi:10.1038/s41598-019-56660-z
- 16.↑
Shou M, Norcross R, Sandig G, et al. Substrate specificity and kinetic properties of seven heterologously expressed dog cytochromes p450. Drug Metab Dispos. 2003;31(9):1161–1169. doi:10.1124/dmd.31.9.1161
- 17.↑
KuKanich B, Papich MG. Plasma profile and pharmacokinetics of dextromethorphan after intravenous and oral administration in healthy dogs. J Vet Pharmacol Ther. 2004;27(5):337–341. doi:10.1111/j.1365-2885.2004.00608.x
- 18.↑
Dodman NH, Shuster L, Nesbitt G, et al. The use of dextromethorphan to treat repetitive self-directed scratching, biting, or chewing in dogs with allergic dermatitis. J Vet Pharmacol Ther. 2004;27(2):99–104. doi:10.1111/j.1365-2885.2004.00549.x
- 19.↑
Perez Jimenez T, Zhu Z, Court MH. Association of cytochrome P450 2D15 (CYP2D15) nonsynonymous polymorphisms and exon 3 deleted RNA splice variant with CYP2D15 protein content and enzyme function in dog liver microsomes. J Vet Pharmacol Ther. 2023;46(2):77–90. doi:10.1111/jvp.13113
- 20.↑
Tucker WE. Preclinical toxicology of bupropion: an overview. J Clin Psychiatry. 1983;44(5 Pt 2):60–62.
- 21.↑
Böttiger Y. Use of omeprazole sulfone in a single plasma sample as a probe for CYP3A4. Eur J Clin Pharmacol. 2006;62(8):621–625. doi:10.1007/s00228-006-0156-5
- 22.↑
Omeprazole. Plumb's Veterinary Drug Handbook. Published November 2023. Accessed December 13, 2023. https://academic-plumbs-com.er.lib.k-state.edu/drug/pr03EbnvaePROD?source=search&searchQuery=omeprazole
- 23.↑
Petsalo A, Turpeinen M, Pelkonen O, Tolonen A. Analysis of nine drugs and their cytochrome P450-specific probe metabolites from urine by liquid chromatography–tandem mass spectrometry utilizing sub 2μm particle size column. J Chromatogr A. 2008;1215(1–2):107–115. doi:10.1016/j.chroma.2008.10.122
