Prediction of Carboxylesterase 1-mediated In Vivo Drug Interaction between Methylphenidate and Cannabinoids using Static and Physiologically Based Pharmacokinetic Models

Yuli Qian; John S. Markowitz

doi:10.1124/dmd.121.000823

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Fig. 1.
Chemical structures of MPH, RA, THC, and CBD and hypothesis of CES1-mediated interactions between MPH and THC/CBD. Dashed lines with flat arrows represent inhibition.
Download figure
Open in new tab
Download powerpoint
Fig. 2.
Inhibition curves for THC (A) and CBD (B) in time-dependent inhibition study. HLS9 (250 µg/ml) was preincubated with or without cannabinoids for 30 minutes before addition of MPH. The formation rate of metabolite (ritalinic acid) from MPH was expressed as a relative value over the control sample without addition of cannabinoids. Data points represent the mean (±S.D.) of triplicate samples from a single experiment. Lines represent model prediction.
Download figure
Open in new tab
Download powerpoint
Fig. 3.
Representative analyses of inhibition kinetics and Lineweaver-Burk plots for THC (A1, A2) and CBD (B1, B2). Incubation was conducted using 250 µg/ml HLS9 with combinations of varying concentrations of MPH and cannabinoids. Data were from one of three independent experiments with duplicate samples. Points represent mean values (±S.D.) of duplicate samples.
Download figure
Open in new tab
Download powerpoint
Fig. 4.
PBPK model simulated and observed concentration-time profiles of MPH after oral administration. The simulated PK profiles were compared with the observed data reported in literature for assessment of model performance. Demographics of the simulated virtual individuals (n = 100) were matched to ones reported in the respective study. Solid lines and shades represent mean and 90% (5^th–95^th) prediction interval of simulated profiles, respectively. Points and error bars represent mean and S.D. of the observed data, respectively. Median values (both simulated and observed) were summarized for the Stage_2017 study.
Download figure
Open in new tab
Download powerpoint
Fig. 5.
PBPK model simulated and observed concentration-time profiles of CBD after oral administration. The simulated PK profiles were compared with the observed data reported in literature for assessment of model performance. Demographics of the simulated virtual individuals (n = 100) were matched to ones reported in the respective study. Solid lines and shades represent geometric mean and 90% (5^th–95^th) prediction interval of simulated profiles, respectively. Points and error bars represent geometric mean and S.D. of the observed data, respectively.
Download figure
Open in new tab
Download powerpoint
Fig. 6.
Simulated outcomes of administration of MPH either simultaneously (A) or with a time difference (B) to a single dose of CBD. The IR formulation of MPH (Ritalin) and the prescriptive oral solution formulation of CBD (Epidiolex) were used in simulation. Under the scenarios of simultaneous administration (A), different strengths of CBD dose (2.5–10 mg/kg) were simulated. Under the scenarios of MPH and CBD administered at different time (B), the dose of CBD was fixed at 10 mg/kg. The resulting influence of CBD on the exposure to d-MPH were summarized as ratios of AUC and C_max as compared with MPH administered alone. The ratios were reported in the forms of geometric means and 90% prediction interval of the simulated 100 virtual subjects.
Download figure
Open in new tab
Download powerpoint
Fig. 7.
Simulated changes in MPH exposure when co-administered with multiple doses of CBD for 7 days. The IR formulation of MPH (Ritalin) and the prescriptive oral solution formulation of CBD (Epidiolex) were used in simulation. In the simulation, MPH was administered twice daily with a time interval of 4 hours between the first and second doses, while CBD (0, 5, and 10 mg/kg) was given every 12 hours. The AUC₂₄ and C_max of d-MPH were summarized on Day 1, 4, and 7. Open circles represent individual parameter values of the simulated 100 virtual subjects. Horizontal bars in the boxplots represent the 5^th, 50^th, and 95^th percentiles of the simulated population.

View popup
TABLE 1
Inhibition of MPH hydrolysis by THC and CBD in time-dependent studies.
IC_50,u (µM)
Cannabinoids Preincubation:
0 min (A) Preincubation:
30 min (B) A/B
THC 0.0441 ± 0.0074 0.0618 ± 0.0097 0.71
CBD 0.157 ± 0.027 0.198 ± 0.029 0.79
IC_50,u, unbound inhibitor concentration that achieves 50% activity. IC_50,u values were presented as estimate ± SE (uncertainty) from nonlinear regression analysis using data from a single experiment with three replicate samples.
View popup
TABLE 2
Determination of kinetic parameters for inhibition of MPH hydrolysis by THC and CBD.
Cannabinoids K_m (µM) K_i,u (µM) α V_max (pmol/min/mg protein)
THC 85.7 ± 12.9 0.031 ± 0.003 16.4 ± 14.7 1056 ± 77
CBD 90.6 ± 10.1 0.091 ± 0.004 >9999 940 ± 150
All values were presented as mean ± S.D. of three independent experiments with duplicate samples.

View popup

TABLE 3

Model evaluation for MPH.

				AUC (ng·h/ml)			C_max (ng/ml)
		Dose	N	Mean observed	Mean predicted	PE (%)	Mean observed	Mean predicted	PE (%)
d/dl-MPH
Calibration Datasets
(Midha et al., 2001)		40 mg Ritalin	24	59.1	90.8	53.7	11.7	16.9	45.1
(Midha et al., 2001)		40 mg Ritalin (fed)	24	68.1	103	51.2	14.3	15.6	9.1
(Patrick et al., 2007)		0.3 mg/kg Ritalin	20	82.9	54.5	−34.3	15.3	8.5	−44.4
(Spencer et al., 2006)		40 mg Ritalin	12	81.9^b	97.1	18.5	14.1	15.7	11.3
(Teo et al., 2004)		20 mg Focalin	15	121	96.1	−20.5	23.7	18.1	−23.7
(Teo et al., 2004)		20 mg Focalin (fed)	15	132	99.6	−24.5	22.1	15.5	−30.0
(Patrick et al., 2013)		0.3 mg/kg Ritalin	24	52.1	53.4	2.5	10.1	8.2	−18.8
(Patrick et al., 2013)		0.15 mg/kg Focalin	24	53.7	53.4	−0.6	10.7	8.2	−23.4
(Parasrampuria et al., 2007)		50 mg Ritalin	49	106	125	17.9	21.1	24.0	13.7
(Parasrampuria et al., 2007)		90 mg Ritalin	49	195	226	15.9	38.8	43.2	11.3
Verification Datasets
(Hysek et al., 2014)		60 mg Ritalin	15	175^a	172	−1.9	30.4	32.4	6.6
(Abbas et al., 2016)		40 mg^c Ritalin	29	132	202	52.6	15.6	22.9	46.8
(Adjei et al., 2014)		25 mg^d Ritalin	24	282	188	−33.3	29.1	18.0	−38.1
(DeVane et al., 2000)		10 mg Ritalin	6	23.7^a	19.7	−16.9	4	4.5	12.5
(Koehm et al., 2010)		20 mg Ritalin	9	35	43	22.9	7	7.1	1.4
(Meyer et al., 2000)		20 mg Ritalin	20	35.8	48.9	36.5	7.6	8.8	15.9
(Patrick et al., 1989)		10 mg^e Ritalin	18	42.0	45.3	7.8	6.4	4.8	−25.3
(Spencer et al., 2012)		40 mg^f Ritalin	26	179^a	185	3.3	25.4	22.6	−10.8
(Stage et al., 2017)		10 mg Ritalin	16	21.4^g	23.2^g	18.7	5^g	3.9^g	−19.2
(Wong et al., 1998)		40 mg Ritalin	21	93.6	90.1	−3.7	17.8	16.4	−7.9
Weighted PE (%)						11.7			-0.8
l-MPH
(Wong et al., 1998)	40 mg Ritalin		21	1.51	1.33	−11.9	0.821	0.54	−34.2
(Patrick et al., 2013)	0.3 mg/kg Ritalin		24	0.77	0.72	−6.5	0.18	0.22	22.2
(Patrick et al., 2007)	0.3 mg/kg Ritalin		20	0.82^a^,^b	0.80	−2.4	0.22^b	0.25	13.6
Weighted PE (%)						-7.0			1.3

PK parameters shown as arithmetic mean values unless otherwise noted.
↵^a AUC_0-last
↵^b Obtained from a non-compartmental analysis of digitized population-level data.
↵^c Two doses administered 6 hours apart.
↵^d Three doses administered 4 hours apart.
↵^e Two doses administered 5 hours apart.
↵^f Two doses administered 4 hours apart.
↵^g Values shown as median.

View popup

TABLE 4

Model evaluation for CBD.

			AUC (ng·h/ml)			C_max (ng/ml)
	Dose (Epidiolex)	N	Mean observed	Mean predicted	PE (%)	Mean observed	Mean predicted	PE (%)
Calibration Datasets
(Taylor et al., 2018)	1500 mg	6	1470^a^,b	1486^a	1.1	260^b	214	−17.7
(Taylor et al., 2018)	6000 mg	6	3490^a^,b	3076^a	−11.9	610^b	453	−25.7
(Tayo et al., 2020)	200 mg	8	435^a^,b	648^a	49.0	118^b	87.0	−26.3
Verification Datasets
(Taylor et al., 2019)	200 mg	8	522^a^,b	548^a	5.0	96.4^b	86.2	−10.6
(Crockett et al., 2020)	750 mg	29	1490^b	1526	2.4	154^b	152	−1.3
(Taylor et al., 2018)	3000 mg	6	2540^a^,b	2608^a	2.7	418^b	386	−7.7
(Taylor et al., 2018)	4500 mg	6	3100^a^,b	2968^a	−4.3	677^b	441	−34.9
Weighted PE (%)					6.2			−12.3

PK parameters are shown as geometric mean values.
↵^a AUC_0-last
↵^b Obtained from a non-compartmental analysis of digitized population-level data due to the discrepancies observed between the reported PK parameters and concentration-time plots.

Additional Files

Figures
Tables

Data Supplement

Supplemental Data -
Supplemental Appendix A. Derivation of THC and CBD binding in the incubation mixture in the presence of protein

Supplemental Appendix B. Back-calculation of Intrinsic Clearance of CBD

Supplemental Table 1. Linearity, accuracy, and intraday and interday precision of RA, THC, and CBD.

Supplemental Table 2. Demographics of subjects in the MPH studies used for model development and verification.

Supplemental Table 3. Demographics of subjects in the CBD studies used for model development and verification.

Supplemental Table 4. Drug-specific input parameters for MPH model.

Supplemental Table 5. Drug-specific input parameters for CBD model.

Supplemental Table 6. Unbound concentrations of THC and CBD in the incubation mixture.

Supplemental Figure 1. Binding model of THC and CBD.

Supplemental Figure 2. Chromatograms of RA (red in panel A), THC (red in panel B) and CBD (red in panel C).

Supplemental Figure 3. A workflow of PBPK model development and verification for MPH and CBD and prediction of drug-drug interactions by simulation.

Supplemental Figure 4. Terminal half-lives of MPH reported in studies found from literature review.

Supplemental Figure 5. Sensitivity analysis to assess the parameters that impact CBD disposition after intravenous administration.

Supplemental Figure 6. PBPK model simulated and observed concentration-time profiles of CBD after intravenous administration.

Supplemental Figure 7. Nonspecific binding of THC and CBD in the incubation mixture.

Supplemental References.