Population Pharmacokinetics of Vamorolone (VBP15) in Healthy Men and Boys with Duchenne Muscular Dystrophy

Abstract

Duchenne muscular dystrophy (DMD) is an inherited neuromuscular disorder occurring in boys and caused by mutations in the dystrophin gene. Vamorolone is a first-generation delta-9,11 compound that has favorable efficacy and side effect profiles relative to classical glucocorticoids. The pharmacokinetics (PK) of oral vamorolone were assessed in parallel group studies in healthy men (Phase I, n =86) and boys with DMD (Phase IIa, n=48) during 14 days of once-daily dosing with a range of doses. Vamorolone exhibits moderate variability in PK with the maximum plasma concentration usually occurring at 2–4 hours and a half-life of approximately 2 hours for all doses and days examined. Population PK modeling of all data together indicated that the PK of vamorolone can be well described by a one-compartment model with zero-order absorption. Both men and boys showed a dose-linearity of PK parameters for the doses examined, with no accumulation of the drug during daily dosing. Ingestion with food results in markedly enhanced absorption of the drug as tested in healthy men. There were similar PK of vamorolone in healthy men and DMD boys with apparent clearances averaging 2.0 in men and 1.7 L/hr/kg in boys. Overall, vamorolone exhibits well-behaved linear PK with similar profiles in healthy men and boys with DMD, moderate variability in PK parameters, and absorption and disposition profiles similar to classical glucocorticoids.

Introduction

Duchenne muscular dystrophy (DMD) is a fatal genetic muscle disease producing progressive muscle weakness in early childhood ¹. DMD is caused by mutations in the dystrophin gene, that links the dystrophin-associated protein complex (DAPC) at the muscle membrane and the actin cytoskeleton, and is responsible for the strength, stability, and functionality of myofibers ². The progression of DMD varies between patients, but muscle weakness is not apparent until ages of 2 – 2.5 years gradually leading to loss of skeletal muscle and ultimately affected individuals become wheelchair-bound by the age of 11–16 years ^2–4.

As DMD causes contraction-induced myofiber injury, these patients exhibit increased inflammation as measured by cytokine and chemokine signaling, leukocyte adhesion and diapedesis, invasive cell type-specific markers, and complement system activation ⁵. Muscle degeneration is further exacerbated by the endogenous inflammatory response with a key role of nuclear factor kappa-B (NF-κB) and other related factors such as tumor necrosis factor alpha (TNF-α) and transforming growth factor beta (TGFβ) ^{4, 6}. Although there is currently no effective treatment for DMD, glucocorticoids (GC) (prednisone, deflazacort) are standard of care in DMD due to their anti-inflammatory role delaying symptoms of DMD ⁷. Unfortunately, treatment with GC is associated with side effects such as Cushingoid appearance, immunosuppression, muscle weakness, as well as long-term adverse sequelae like osteopenia and stunted growth ⁸. These side effects are related to the transactivation activity of GC where a dimeric GC receptor/ligand complex binds to GC responsive elements (GREs) in the promoter regions of target genes and regulate transcription. Recently, the glucocorticoid deflazacort (Emflaza) was approved by FDA for use in DMD (0.9 mg/kg/day).

Vamorolone (formerly VBP15) is a novel anti-inflammatory drug developed by ReveraGen BioPharma Inc. (Rockville, MD), that is structurally related to GC. It is optimized to retain transrepressive activities (e.g. NF-κB inhibition associated with anti-inflammatory efficacy), reduce transactivation activities (GRE-mediated transcriptional activities associated with safety concerns), and improve membrane stability properties ^10–13. Pre-clinical data in multiple mouse models showed efficacy similar to prednisolone, and marked reductions of side effects such as growth stunting and bone fragility ^{10, 14–17}. Phase 1 clinical trials (86 healthy adult males) showed that vamorolone is well-tolerated for all tested doses with pharmacokinetic (PK) and metabolic profiles similar to prednisone and lesser adrenal suppression along with fewer traditional GC side-effects (bone fragility, metabolic disturbance, immune suppression) ¹⁸. Furthermore, Phase IIa first-in-patient studies (48 DMD boys from 4 to 7 years old) showed improved safety compared to GC as shown by reduction of insulin resistance, beneficial changes in bone turnover, and a reduction in adrenal suppression ¹⁹.

Vamorolone was selected among twenty Δ9,11-derivatives due to its NF-κB inhibition potency, reduced transactivation properties, and good bioavailability. Preclinical characterization showed a low-medium clearance (CL) in mice and rats (1.1–1.2 L/hr/kg) and medium CL in beagle dogs (1.5 L/hr/kg). Tissue distribution is moderate in rodents (volume of distribution is 0.76 in mice and 0.77 L/kg in rats) and somewhat higher in dogs (1.9 L/kg). Bioavailability was 53 to 75% in these species. Vamorolone moderately induced CYP3A4, similar to other GC ¹¹.

This report evaluates the pharmacokinetics of vamorolone based on non-compartmental analysis (NCA) and population PK (Pop PK) modeling that considers plasma concentration data after oral dosing of the drug to fasted and fed men (Phase I study) and boys with DMD (Phase IIa study). Parallel group studies assessed a range of doses given once-daily over 14 days in both groups.

Methods

Prior to any study procedures, informed consent was provided by the adults (Phase I) and by the legal guardian of the boys (Phase II) who participated in the study. The vamorolone dosage product was manufactured by Velesco Pharmaceutical Services (Kalamazoo, MI) and supplied as a cherry-flavored oral suspension (4% by weight).

Phase I Study (Healthy Men)

Single Ascending Dose (SAD)

The doses were selected based on the no-observed- adverse-effect-level (NOAEL) in the mouse, the most sensitive species in preclinical studies. The NOAEL in mouse was 30 mg/kg, the human equivalent dose of which is 2.43 mg/kg. Using a safety factor of 25, a starting dose of 0.0972 mg/kg was estimated. Next, the standard dose escalation factor of 3 was utilized for this first-in-human study until attaining a maximum dose of 20 mg/kg. Therefore, the SAD study included 7 Cohorts. in Cohorts 1–5 and 7, 8 subjects received 0.1, 0.3, 1.0, 3.0, 8.0, and 20.0 mg/kg oral doses of either vamorolone (n=6) or placebo (n=2) under fasted conditions. In Cohort 6, 6 subjects received an oral dose of 8 mg/kg of vamorolone under fed conditions. The 54 male subjects were between 19 and 64 years of age, with a BMI range of 21.2 to 31.6 kg/m². In Cohort 6 the single dose of 8 mg/kg vamorolone was given within 30 min of beginning a standard high fat/high calorie meal. Blood was collected pre-dose and for up to 72 hours post-dose for analysis of vamorolone in plasma. The time points were: pre-dose and 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 16, 24, 36, 48, 60, and 72 hours post-dose.

Multiple Ascending Dose (MAD)

For the MAD study, 4 cohorts containing 8 subjects each were randomized to receive vamorolone (n=6) or placebo (n=2) once a day (QD) for 14 days. The dose levels were 1.0, 3.0, 9.0, and 20.0 mg/kg/day. For PK assessment, blood samples were collected pre-dose and for up to 72 hours post-dose after the first and final doses of vamorolone. The time points were the same as the SAD study. The Phase I data were previously published¹⁸. The IRB approval for the Phase I study (VBP15–001) was obtained from the Midlands Independent Review Board. The clinical trial registration number is NCT02415439.Phase IIa Study (Boys with DMD).

The Phase IIa study was the first PK assessment of oral doses of vamorolone in boys with DMD. Blood samples were collected on days 1 and 14 during daily administration of the drug. Twelve subjects were assigned to each of the 4 cohorts and received an oral dose of vamorolone once daily (QD) with 8 ounces of whole milk (or equivalent high fat food portion). The dose levels were 0.25, 0.75, 2.0, and 6.0 mg/kg/day. The blood sampling times were at 0 (pre-dose), 1, 2, 4, 6, and 8 hours on days 1 and 14. There were 96 samples potentially available for PK analyses (48 boys, day 1 and day 14), but 7 samples were unavailable owing to various reasons. The Phase IIa data were previously published¹⁹. The IRB approvals (VBP15–002) were obtained locally at each participating clinical site. The clinical trial registration number is NCT02760264

Bioanalytical methods

Plasma concentrations of vamorolone were measured using a specific and validated liquid chromatography tandem mass spectrometry (LC-MS) assay. Sample processing was performed by liquid-liquid extraction using a sample volume of 50.0 μL. Separation between potential metabolites and interfering endogenous compounds was achieved by UPLC using a Waters Acquity BEH C18 column (2.1 × 50 mm, 1.7 μm particles) at 40°C and isocratic elution with gradient wash using 0.1% HCOOH in H₂O as mobile phase A and 0.1% HCOOH in Acetonitrile as mobile phase B operating at a flow rate of 0.800 mL/min. A triple quadrupole mass spectrometer (AB SciEx Triple Quad 5500) equipped with a Turbo-ionspray source was used for detection in positive ion mode. Quantification was based on multiple reaction monitoring (MRM) of the transitions of m/z 357.3→147.1 for vamorolone and 361.3→147.1 for the internal standard (¹³CD₃ VBP15). A linear calibration curve, ranging from 2.00 to 2000 ng/mL, with a 1/x2 weighing factor was used. The accuracy of the assay within-run bias ranged from −3.8% to 1.4% and overall bias ranged from −2.0 to −0.3%. Precision ranged from 1.5% to 4.4%. The Lower Limit of Quantitation (LLOQ) of the assay was 2.0 ng/mL. Values between 1 and 2 ng/ml were included in the PK calculations.

Non-Compartmental Analysis (NCA)

Apparent plasma concentration values that occurred at pre-dose times (0 hr) were treated as zero (0). All other concentrations were included in the PK analysis. Nominal/Relative times were identical for all data used in this study. The NCA analysis was performed using Phoenix WinNonlin (Certara). The maximum plasma concentration (C_max) and time to C_max (t_max) were taken directly from the data. For the calculation of the AUC_last and AUC_inf, the linear-up log-down trapezoidal method was used with uniform weighting of the data. The elimination rate constant, k_el, was calculated as the negative slope of the terminal log-linear segment of the plasma concentration-time curve. For the cases where the subjects showed a clear washout phase (by visual inspection), at least 3 terminal points were selected for calculation of k_el. For subjects not showing such clear elimination, k_el was assumed to be the average k_el of the rest of the subjects who received the same dose.

The elimination half-life (t_½) was calculated as ln(2)/kel and clearance (CL) uncorrected for bioavailability (F), was calculated as Dose/AUCinf. Dose-proportionality was assessed by fitting the power equation to the C_max and AUC_inf versus Dose data:

$$C_{max}orAU{C}_{inf}=a*Dos{e}^b$$ (1)

For estimating the a and b parameters, the Matlab R2016b^® nonlinear least-squares curve fitting tool cftool was used. Goodness-of-fit was evaluated by calculating R².

Population Pharmacokinetic Model

Nonlinear mixed-effect population (Pop PK) modeling was performed using the first-order conditional estimation with interaction (FOCE-I) implemented in NONMEM 7.3 ²⁰. NONMEM was utilized with Perl-speaks-NONMEM (PsN, version 4.7.0), Xpose (version 4), Pirana (version 2.9.6) ^{21, 22} and data manipulation and plotting was done using R (version 3.4.3) and Matlab R2016b^®_. The Pop PK parameters were modeled in terms of both random (parameter variance) and fixed effects (typical value of population parameter).

All PK data were jointly and suitably described with a one-compartment model (1CM) with a zero-order absorption rate. The differential equations are:

$$\frac{d{A}_0}{dt}=-\frac{A_0}{T_{k0}},\text{with\hspace{0.17em}}{\text{A}}_0\left(\text{t}=0\right)=Dose$$ (2)

$$V\cdot \frac{dC}{dt}=Input-\frac{CL}{V}\cdot C,\text{with C}\left(\text{t}=0\right)=0$$ (3)

$$Input=\left\{\begin{array}{cc}\frac{A_0}{T_{k0}},& t\le T_{k0}\\ 0,& t>T_{k0}\end{array}\right.$$ (4)

where T_k0 is the duration of input, C is plasma drug concentration, CL is clearance, and V is the volume of distribution. Absorption rate (k_a) was calculated as Dose/T_k0. For an individual i the Pop PK model parameters are:

$$T_{K0},i=\widetilde{T_{k0},\iota }\cdot e^{{\eta }_{T_{k0},i}}$$ (5)

$$CL,i=\widetilde{CL,\iota }\cdot e^{{\eta }_{CL},i}$$ (6)

$$V,i=\widetilde{V,\iota }\cdot e^{{\eta }_{V1},i}$$ (7)

where bars indicate typical values of parameters and η’s are parameter variances in the population. Individual specific parameters were assumed to have a log-normal distribution. For evaluating residual variability, an additive plus constant coefficient of variation model was used. The evaluation of the base model without covariates was based on goodness-of-fit criteria including visual inspection, examination of residuals, and the objective function value (see Supplemental Figures S1–S2).

For assessing covariate effects, graphical evaluation was performed initially by plotting parameter estimates and selected patient covariates (Supplementary Figures S3–S4). These graphical plots were used for initial detection of possible effects, investigation of the functional form of the effects (e.g. linear, power, etc), and identification of the direction of effects. Next, potential relationships were evaluated using nonlinear mixed-effect modeling with the stepwise forward inclusion and backward elimination method performed through the stepwise covariate modeling (scm) automated method included in Pirana²¹. For the forward inclusion, statistical significance was considered a decrease of the objective function value (OFV) of at least 3.84 (χ², p< 0.05, df=1) and for backward elimination an increase of OFV of at least 10.83 (χ², p<0.001, df=1). The effects of body weight, food, and age group (e.g. men, boys) were examined.

Comparison between different PK models was performed by comparing the objective function value, Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC). All models were run with FOCE with Interaction.

Results

Demographic and baseline information for the subjects in the Phase I and Phase IIa studies are listed in Supplementary Table S1. In the SAD and MAD adult groups, there were 86 men with an average age of 35 years and average body weight of 80.5 kg. In the DMD group there were 48 boys with average age of 5 years and average body weight of about 19.7 kg. Vamorolone was administered to 66 healthy men and 48 boys in total. The Pop PK analysis included overall 114 individuals and 1922 vamorolone concentrations.

Figures 1 and 2 show the array of PK data for all doses of vamorolone given to the men and DMD boys along with the fittings using the final one-compartment (1CM) population model with zero-order absorption rate. The 1CM (Eq 2–4) with log-normal distributed PK parameters (Eq 5–7) well described the kinetics of vamorolone in all subjects except for some under-prediction of the peak concentrations for the 20 mg/kg dose in men. Typical diagnostic plots supporting the modeling are provided in the Supplementary Materials (Figures S1–S2, S5–S7) and the final parameters in Table S2. The covariates found to significantly affect vamolorone were: a food effect for all PK parameters, and body weight for variation of absorption rates. Inter-individual variability differences between the base model and the final model with covariates is shown in Supplementary Table S3. Typical values of the population model parameters along with inter-individual variances are listed in Table 1 showing reasonable variability of the parameter estimates. Also tested were a two-compartment Pop PK model and a 1CM with first-order absorption kinetics, but the fittings were not as good. Objective function comparisons for the different models tested are shown in Supplementary Table S4.

Figure 1.

Plasma vamorolone concentration versus time profiles for the adult men for the indicated doses. Symbols are individual subject data and lines depict fitting of data with the Pop PK model.

Figure 2.

Plasma vamorolone concentration versus time profiles for the DMD boys for the indicated doses. Symbols are individual subject data and lines depict fitting of data with the Pop PK model.

Table 1.

Comparison of population pharmacokinetic parameters (mean, standard deviation) between the fasted/fed men and DMD boys considering all doses. The p-values were calculated based on two sample t-tests.

Parameters	Men Fasted	Men Fed	Boys	Men Fasted vs Boys	Men Fed vs Boys
ka [mg/kg∙hr]	5.22 (6.93)	2.13 (0.77)	0.81 (0.87)	p≤0.05	p≤0.05
V/F [L/kg]	8.02 (2.96)	2.66 (0.49)	6.29 (2.29)	p>0.05	p≤0.05
CL/F [L/hr∙kg]	2.05 (0.49)	0.94 (0.23)	1.90 (0.49)	p≤0.05	p≤0.05

An initial NCA assessment of the linearity and stationarity of vamorolone PK was performed. Figure 3 shows AUC_inf and C_max versus dose for the fasted men and DMD boys for Days 1 and 14 with power equation parameters listed in Table 2. Fitting with Eq. 1 resulted in R² values larger than 0.65 for both days (Table 2) indicating reasonable dose-proportionality. The b values had 95% confidence intervals (CI) that included the value of 1.0 reflecting linearity. The PK data for Days 1 and 14 superimposed indicative of stationarity. The sampling times were less intensive in the DMD boys, but the regression results for men and DMD boys generally show overlapping 95% CI ranges. These results graphically demonstrate that the exposure indices of vamorolone obtained in the healthy men and DMD boys are similar and support the use of a 1CM with linear and stationary PK parameters as used in the population modeling.

Figure 3.

Regression analysis of vamorolone C_max and AUC_inf versus dose for men (blue) and parameters for DMD boys (green). Eq 2 was fitted to Day 1 and 14 PK data. The fitted parameters are listed in Table 2.

Table 2.

Regression analysis of C_max and AUC_inf versus dose in men and DMD boys (Equation 1 and Figure 3). Values in parentheses indicate the 95% confidence intervals.

Variable	Day 1		Day 14
	Boys	Men	Boys	Men
AUCinf	a = 321.3 (57.8, 585)	a = 759.8 (410, 1110)	a = 604.8 (409.8, 799.7)	a = 541.5 (101.2, 981.8)
	b = 1.313 (0.84, 1.785)	b = 0.851 (0.69, 1.02)	b = 0.995 (0.804, 1.185)	b = 0.97 (0.68, 1.26)
	R2=0.74	R2=0.85	R2=0.87	R2=0.85
Cmax	a = 86.2 (3.8, 168.7)	a = 120.4 (46.4, 194.4)	a =124.9 (71.9, 177.8)	a = 87.27 (4.51, 170)
	b = 1.28 (0.73, 1.83)	b = 0.95 (0.73, 1.16)	b = 1.14 (0.9, 1.39)	b = 1.13 (0.80, 1.55)
	R2=0.66	R2=0.81	R2=0.86	R2=0.87

Giving vamorolone with a high fat/high calorie meal markedly affects the PK in healthy men. Figure 4 shows the PK data and Pop PK evaluation for 8 mg/kg doses of vamorolone in fed and fasted men. Food produces a higher C_max, t_max, and AUC of the drug. There were significantly lower absorption rates (p= 0.0007), but the smaller apparent volumes of distribution and clearances are likely due to the increased bioavailability (F). The mean values and standard deviations (SD) of the population parameters along with the statistics are listed in Table 1 and Supplementary Figure S4.

Figure 4.

Comparison of PK profiles and model-fitted parameters for the 8 mg/kg doses of vamorolone in fasting and fed men. Bar and whisker plots indicate the medians and first and third quartiles of the individual PK parameters.

The Pop PK parameters of vamorolone for the DMD boys are shown in Figure 5 and listed in Table 1. The mean values and variability in these parameters for the DMD boys and fasted men are similar for V/F, and CL/F. The k_a and later tmax values indicate that there aredifferences between fasted men and DMD boys possibly related to giving vamorolone with 8 oz of milk or equivalent high fat food portion in boys. Ingestion of vamorolone with food produced lower values and much less variability in k_a. The food also results in significantly lower and less variable population values of both V/F and CL/F that are primarily due to the larger F value. If it is assumed that the disposition CL of vamorolone is not affected by food, the ratios of mean CL/F (values Fasted 2.05 versus Fed 0.94 L/hr/kg) indicate about twice the bioavailability (F) of vamorolone when ingested with food. It should be also noted that the DMD boys were given vamorolone with 8 oz of whole milk rather than during fasting. The values of V/F and CL/F in Figure 5 were normalized for body weight in kg. Figure 6 shows the spread of these values in relation to the body weights of the individual subjects.

Figure 5.

Population model parameters of vamorolone for fasted and fed men and DMD boys. Circles indicate the individual estimates and bar and whisker plots depict the medians and first and third quartiles.

Figure 6.

Relationships between indicated population pharmacokinetic parameters of vamorolone and body weights.

Discussion

The GC provide crucial anti-inflammatory therapy for children with DMD. Deflazacort is the only FDA-approved therapy for the treatment of patients with DMD, but prednisone is widely prescribed off-label. Although there have been a number of clinical trials for deflazacort (DFZ) and prednisone ^23–29, the PK of these compounds in DMD boys have not been published to enable a comparison of vamorolone pharmacokinetics in relation to the other steroids. However, the FDA Clinical Pharmacology and Biopharmaceutics Review for Emflaza (deflazacort) reports a terminal half-life of the active metabolite 21-desDFZ of 2–3 hours with greater variability in exposures in DMD patients than normal adults. (https://www.accessdata.fda.gov/drugsatfda_docs/nda/2017/208684,208685Orig1s000TOC.cfm)

Vamorolone was developed as a replacement for GC treatment and has shown efficacy equal or superior to prednisone in mouse models of muscular dystrophy ^{10, 30}, lung disease ^{14, 17}, inflammatory bowel disease ¹⁶, and multiple sclerosis ¹⁵. In vitro metabolic stability studies showed that vamorolone did not induce CYP1A2, and only moderately induced CYP3A4 ¹¹ with four metabolites identified in human and monkey hepatocytes and three in dogs ¹¹. The finding of similar PK parameters of vamorolone on Days 1 and 14 (Figure 3, Table 2) during multiple-dosing is consistent with lack of self-enzyme induction.

Phase I studies of vamorolone in healthy men showed a typical apparent clearance of around 2.0 L/hr/kg and an average half-life of 2–3 hr ¹⁸ with dose-linear pharmacokinetics (Figure 3). These values are similar to typical GC such as deflazacort and prednisone active metabolites in adults. Deflazacort is an inactive prodrug and is converted rapidly to its active metabolite 21-desDFZ that has a half-life of 1.3 – 1.9 hr and oral clearance of 1.2 – 1.6 L/hr/kg ³¹. Similarly, prednisone is transformed to its active metabolite prednisolone that has a half-life of 3.5–3.7 hr and oral clearance of around 0.2 L/hr/kg in healthy subjects ³². Deflazacort has dose-proportional PK ³³ while prednisolone PK exhibits complex changes with dose ³².

Endogenous cortisol is regulated by a feedback mechanism between the hypothalamus-pituitary- and adrenal glands (the HPA axis) ³⁴. Administered synthetic GC act on this feedback mechanism inhibiting corticotrophin and ultimately secretion of cortisol ^{35, 36}. The degree of inhibition is highly dependent on the dose, potency, and time of administration ³⁷. Owing to the relatively short half-life, several pre-clinical and clinical studies have shown that daily dosing of GC can enhance their beneficial effects with minimal safety concerns ³⁸. The first-in-human safety data obtained for vamorolone in healthy adult men showed no changes in bone-turnover markers, no evidence of drug-induced hyperglycemia, and no loss of blood lymphocytes at any dose tested ¹⁸. Adrenal suppression evaluated by first-in-morning cortisol was only evident in a subset of men on the highest doses of vamorolone. Along with improved safety biomarker profiles compared to GC, the Phase IIa studies in DMD boys showed good dose-related anti-inflammatory effects of vamorolone based on exploratory efficacy biomarkers ¹⁹.

DMD is frequently accompanied by respiratory, orthopedic, and cardiac complications ³⁹. These patients also have disrupted GI function with gastric emptying half-times and oro-cecal transit times significantly longer than healthy controls ⁴⁰. If these symptoms were present in the boys in this study, there is no indication that the oral absorption of vamorolone was impacted.

Vamorolone PK data were fitted using a one-compartment model with zero-order absorption. Zero-order absorption indicates that a constant amount of drug enters the circulation per unit time (e.g. like an infusion). Typical zero-order absorption produces a sharp peak in plasma and then a quick decline while first-order absorption produces a round peak. The PK data both appear zero-order and were better fitted as such. It is not possible to ascribe mechanistic reasons based solely on plasma data.

Similar to traditional GC and hydrocortisone, vamorolone has low solubility in aqueous solutions ^41–43. Vamorolone has high CaCo2 cell permeability (11.2 · 10⁻⁶ cm/s), but its water solubility is 0.187 mM. These properties would place vamorolone in FDA BCS Class II. The highest dose of 20 mg/kg ingested with 240 ml of water in a 80 kg person would produce a concentration of 1600 mg per 240 ml or 18.7 mM. The dose given in suspension thus far exceeds the solubility of the drug, which may explain the moderate bioavailability of vamorolone and is the likely reason that a high fat meal (40 g fat) enhances the absorption of the drug. Despite this solubility limitation, like many orally dosed GC, oral use of a suspension formulation (4% vamorolone by weight) showed reasonable, but unknown, apparent bioavailability. As there was a significant food effect with an increase in AUCinf by nearly 250% (Figure 4), absorption in the fasting state must be incomplete. Food is known to change the rate and extent of absorption of drugs in both directions ⁴⁴. While a first-pass effect cannot be ruled out, it is unlikely that dosing with food would enhance first-pass availability.

In studies of DMD boys, the clinical trial design team considered pragmatic issues of controlling for food in daily morning dosing in 4 to 7 year-old children, and opted for giving the drug with 8 oz of whole fat milk or fat equivalent. Since whole milk contains 3.25% fat (see Wikipedia), 8 oz of whole fat milk constitutes about 7.4 g of fat for a 20 kg boy. For men, vamorolone was ingested after a high calorie/high fat meal that is composed of 40 g of fat and is more viscous. Thus the PK data for the DMD boys might be expected to have a lesser food effect with absorption closer to fasting than for a high fat meal. A food effect was also observed for deflazacort where giving tablets with a high fat meal reduced C_max by nearly 30% and delayed t_max by one hour relative to fasting ⁴⁵. No effect was observed on the overall systemic absorption as measured by the AUC. Food taken with prednisone tablets delays the tmax by 2 hr but does not affect overall bioavailability ⁴⁶.

In summary, the analyses presented in this report describe the pharmacokinetics of oral vamorolone in healthy men and boys with DMD. The PK of vamorolone was well fitted with a 1-compartment model with zero-order absorption. The PK of vamorolone is rather rapid and resembles that of other commonly used GC such as deflazacort and prednisolone. The PK of vamorolone was linear and stationary in adult men and DMD boys, exhibited reasonably consistent PK, and was generally similar in adult men and boys with DMD.