Integrating Clinical Variability into PBPK Models for Virtual Bioequivalence of Single and Multiple Doses of Tofacitinib Modified-Release Dosage Form.

Tofacitinib is a potent, selective inhibitor of the Janus kinase (JAK) family of kinases with a high degree of selectivity within the human genome's set of protein kinases. Currently approved formulations for tofacitinib citrate are immediate-release (IR) tablets, modified-release (MR) tablets, and IR solution. A once daily MR microsphere formulation was developed for use in pediatric patients. Demonstration of bioequivalence (BE) between the 10 mg once daily (q.d.) MR microsphere formulation and 5 mg twice daily (b.i.d.) IR solution is needed to enable the exposure-response analyses-based bridging to support regulatory approval. To assess BE between MR microsphere and IR solution, an innovative approach was utilized with physiologically-based pharmacokinetic (PBPK) virtual BE trials (VBE) in lieu of a clinical BE trial. A PBPK model was developed to characterize the absorption of different formulations of tofacitinib using Simcyp ADAM module. VBE trials were conducted by simulating PK profiles using the verified PBPK model and integrating the clinically observed intrasubject coefficient of variation (ICV) where BE was assessed with a predetermined sample size and prespecified criteria. The VBE trials demonstrated BE between IR solution 5 mg b.i.d. and MR microsphere 10 mg q.d. after a single dose on day 1 and after multiple doses on day 5. This research presents an innovative approach that incorporates clinically observed ICV in PBPK model-based VBE trials, which could reduce unnecessary drug exposure to healthy volunteers and streamline new formulation development strategies.

Bridging Efficacy of Tofacitinib Immediate-Release to Extended-Release Formulations for Treatment of Ulcerative Colitis: Application of a Model-Informed Drug Development Approach.

Tofacitinib is an oral, small molecule Janus kinase inhibitor for the treatment of ulcerative colitis (UC). We report a model-informed drug development approach for bridging efficacy from immediate-release (IR) to extended-release (XR) tofacitinib formulations in patients with UC. IR-XR efficacy bridging was supported by exposure-response analysis of phase 3 induction/maintenance studies of the IR formulation in UC to identify exposure metrics relevant for efficacy. Pharmacokinetic studies in healthy subjects were used to confirm similarity of relevant exposure metrics of tofacitinib IR 5 mg twice daily to XR 11 mg once daily, and tofacitinib IR 10 mg twice daily to XR 22 mg once daily, thereby bridging efficacy between IR and XR formulations. Food effect was evaluated at both XR formulation dose levels. Exposure-response analysis demonstrated that area under the plasma concentration-time curve (average plasma concentration) was a relevant predictor of efficacy. Pharmacokinetic studies demonstrated that area under the plasma concentration-time curve was equivalent between formulations under single-dose and steady-state conditions, and other exposure metrics were also similar. These results also supported bridging of safety data for IR-XR formulations. Food had no impact on tofacitinib XR exposure. These data support efficacy/safety bridging of IR-XR formulations in patients with UC.

Developing a Biorelevant Dissolution Method for an Extrudable Core System (ECS) Osmotic Tablet.

The objective of this work is to develop a biorelevant dissolution method to support the clinical study for In Vitro In Vivo Correlation (IVIVC) of the first commercially approved single-layer extrudable core system (ECS) osmotic tablet - the 11 mg tofacitinib modified-release tablet. The dissolution conditions were selected through analysis of experimental work including several designed experiments (DoE). The Apparatus 2 (paddles) was selected over the Apparatus 1 (baskets) to minimize the dissolution test variability. The paddle speed was kept at 50 rpm to be conservative and because higher paddle speed did not offer statistically significant improvement in dissolution test variability. The buffer of 50 mM potassium phosphate at pH 6.8 was selected over other buffers at lower or acid pH as the in vivo drug release is expected to occur in the small intestinal region, where the pH is approximately neutral. Finally, the statistically designed experiments proved that use of the Japanese basket sinkers was effective in reducing dissolution variability and eliminating the artificial shift in dissolution profile caused by final pink color-coated tablets sticking to the dissolution vessel. Discriminatory power of the method was verified and the method was validated per ICH and FDA guidelines. Since a Level A IVIVC is established from the analysis of the results of both in vivo clinical study and in vitro dissolution testing, the method is proven to be biorelevant. It also serves a suitable quality control dissolution method.

Development and validation of a Level A in-vitro in-vivo correlation for tofacitinib modified-release tablets using extrudable core system osmotic delivery technology.

PURPOSE: To determine if a validated Level A in-vitro in-vivo correlation (IVIVC) could be achieved with the extrudable core system (ECS) osmotic tablet platform. Tofacitinib is an oral JAK inhibitor for the treatment of rheumatoid arthritis. METHODS: Fast-, medium-, and slow-release modified-release formulations of 11 mg tofacitinib ECS tablets, and one formulation of 22 mg tofacitinib ECS tablet, were manufactured. In vitro dissolution of the tofacitinib ECS tablets was performed using USP Apparatus 2 (paddles) and in vivo pharmacokinetic (PK) data were obtained from a Phase 1 study in healthy volunteers. A 5 mg immediate-release formulation tablet was included to support deconvolution of the tofacitinib ECS PK tablet data to obtain the in vivo absorption profiles. A linear, piecewise correlation and a simple linear correlation were used to build and validate two IVIVC models. RESULTS: The prediction errors (PEs) for the linear, piecewise correlation met the Food and Drug Administration's criteria for establishing a Level A IVIVC, with a maximum absolute individual internal PE of 4.6%, a maximum absolute average internal PE of 3.9%, and a maximum absolute external PE of 8.4% obtained. CONCLUSIONS: This study demonstrates that the tofacitinib ECS osmotic tablet platform can achieve a Level A IVIVC, similar to other osmotic delivery systems.

Quantifying and reducing powder shear sensitivity when manufacturing capsules with lubricants.

The purpose of this work was to develop a methodology that quantifies the extent of shear induced during an encapsulation process and show how formulation composition and manufacturing process designs can be changed to reduce the negative impact on drug product quality attributes. The powder feed system used in a dosing disc type pharmaceutical capsule filling machine induced additional shear of the powder prior to slug formation. The shear occurred both in the hopper portion, via the rotation of the feed auger and impeller, and in the powder bowl via the tamping pin agitation and/or shear against the stationary surfaces such as the powder level scraper. The extent of shear was quantified to assess the impact of further dispersing the hydrophobic lubricant, magnesium stearate, in both active and placebo formulations. Stratified samples over the course of the encapsulation run showed suppression in the drug dissolution profiles and decrease in the interparticulate tensile strength of the encapsulated product. The amount of shear (duration and rate) induced during the encapsulation unit operation can be much greater than that from typical bin blending operations and therefore requires consideration during product design and scale-up to ensure product robustness.

Extended-Release Once-Daily Formulation of Tofacitinib: Evaluation of Pharmacokinetics Compared With Immediate-Release Tofacitinib and Impact of Food.

Tofacitinib is an oral Janus kinase inhibitor for the treatment of rheumatoid arthritis. An extended-release (XR) formulation has been designed to provide a once-daily (QD) dosing option to patients to achieve comparable pharmacokinetic (PK) parameters to the twice-daily immediate-release (IR) formulation. We conducted 2 randomized, open-label, phase 1 studies in healthy volunteers. Study A characterized single-dose and steady-state PK of tofacitinib XR 11 mg QD and intended to demonstrate equivalence of exposure under single-dose and steady-state conditions to tofacitinib IR 5 mg twice daily. Study B assessed the effect of a high-fat meal on the bioavailability of tofacitinib from the XR formulation. Safety and tolerability were monitored in both studies. In study A (N = 24), the XR and IR formulations achieved time to maximum plasma concentration at 4 hours and 0.5 hours postdose, respectively; terminal half-life was 5.9 hours and 3.2 hours, respectively. Area under plasma concentration-time curve (AUC) and maximum plasma concentration (Cmax ) after single- and multiple-dose administration were equivalent between the XR and IR formulations. In study B (N = 24), no difference in AUC was observed for fed vs fasted conditions. Cmax increased by 27% under the fed state. On repeat administration, negligible accumulation (<20%) of systemic exposures was observed for both formulations. Steady state was achieved within 48 hours of dosing with the XR formulation. Tofacitinib administration as an XR or IR formulation was generally well tolerated in these studies.

Commercial scale validation of a process scale-up model for lubricant blending of pharmaceutical powders.

An experimental study was conducted to verify that lubrication mixing in commercial-scale bin blenders can be described by a previously-reported lubrication blending process scale-up model. Specifically, the mixing of two placebo formulations (2:1 MCC:lactose, and 2:1 MCC:DCP) with 1% magnesium stearate in 100, 400, and 2000 L bin blenders at 30% and 70% blend fill levels for several extents of lubricant mixing was examined. The lubricated powder blends were assessed for bulk/tapped density and powder flow, as measured by Hausner's ratio. The blends were then compressed into tablets and evaluated for tensile strength, friability, and disintegration. It was observed that the lubrication rate constant, γ, for tablet tensile strength and for bulk specific volume were similar. Furthermore, powder flow, as measured by Hausner's ratio, improved with increased extent of lubrication. Tablet disintegration and tablet friability were both minimally affected as a result of extended lubrication for the placebos blends evaluated in this study. The results of this study confirm that the lubrication mixing model can be applied to scale-up the lubrication blending process from batches made in 30 mL bottle blenders to batches made in 2000 L bin blenders, which is a range of nearly five orders of magnitude.

A quality-by-design study for an immediate-release tablet platform: examining the relative impact of active pharmaceutical ingredient properties, processing methods, and excipient variability on drug product quality attributes.

The impact of filler-lubricant particle size ratio variation (3.4-41.6) on the attributes of an immediate-release tablet was compared with the impacts of the manufacturing method used (direct compression or dry granulation) and drug loading (1%, 5%, and 25%), particle size (D[4,3]: 8-114 µm), and drug type (theophylline or ibuprofen). All batches were successfully manufactured, except for direct compression of 25% drug loading of 8 µm (D[4,3]) drug, which exhibited very poor flow properties. All manufactured tablets possessed adequate quality attributes: tablet weight uniformity <4% RSD, tablet potency: 94%-105%, content uniformity <6% RSD, acceptance value ≤ 15, solid fraction: 0.82-0.86, tensile strength >1 MPa, friability ≤ 0.2% weight loss, and disintegration time < 4 min. The filler-lubricant particle size ratio exhibited the greatest impact on blend and granulation particle size and granulation flow, whereas drug property variation dominated blend flow, ribbon solid fraction, and tablet quality attributes. Although statistically significant effects were observed, the results of this study suggest that the manufacturability and performance of this immediate-release tablet formulation is robust to a broad range of variation in drug properties, both within-grade and extra-grade excipient particle size variations, and the choice of manufacturing method.

Utilizing quantitative certificate of analysis data to assess the amount of excipient lot-to-lot variability sampled during drug product development.

Understanding variability in excipient physico-chemical properties is becoming an important aspect of Quality-by-Design drug product development. However, present experimental methods have only been able to study a few physico-chemical properties for a few excipient lots due to time, cost, and sample gathering considerations. An alternative analysis method is proposed here that shows how quantitative physico-chemical property data reported in vendor certificates of analysis can evaluate excipient lot-to-lot variability in a comprehensive and low cost manner. Microcrystalline cellulose, spray-dried lactose, and magnesium stearate were selected as commonly-used excipients for this demonstration. The proposed analysis method offers drug product developers several advantages over present experimental methods, including the ability to: (1) examine excipient products for manufacturing site and/or year-to-year variations, (2) quantify a domain of prior experience for each excipient by determining the percentage of excipient lots contained within a multi-dimensional ellipsoid described by the excipient lots used during drug product development, and (3) rationally select excipient lots from the vendors inventory to maximize the domain of prior experience throughout the drug development process. For cases where certificate of analysis data may contain insufficient information, drug product developers and excipient vendors should work together to identify more appropriate datasets for analysis.

Incorporating Turbula mixers into a blending scale-up model for evaluating the effect of magnesium stearate on tablet tensile strength and bulk specific volume.

Turbula bottle blenders are often used in lab-scale experiments during early-stage pharmaceutical product development. Unfortunately, applying knowledge gained with these blenders to larger-sized diffusion mixers is limited by the lack of blending models that include Turbula mixers. To address this need for lubrication blending scale-up, 2:1 blends of microcrystalline cellulose and spray-dried lactose or dibasic calcium phosphate were mixed with 1% magnesium stearate using Turbula bottle blenders, varying bottle volume, V (30-1250mL); bottle headspace fraction, F(headspace) (30-70%); and the number of blending cycles, r (24 to ∼190,000 cycles). The impact of lubrication blending on tensile strength and bulk specific volume quality attributes, QA, was modeled by:where QA(0) is initial QA value, ß is sensitivity of QA to lubrication, γ is formulation-specific lubrication rate constant, and L is characteristic mixing length scale (i.e. 1.5V(1/3) for Turbula blenders, V(1/3) for simple diffusion mixers). The factor of 1.5 captures the bottle dimensions and the more complex mixing dynamics of the Turbula blender. This lubrication blending process model is valid for scale-up from 30-mL to 200-L blenders. Assessing bulk specific volume may provide a simpler, more material-sparing means for determining γ than tensile strength, since these QAs exhibited similar γ values.

Examining the impact of excipient material property variation on drug product quality attributes: a quality-by-design study for a roller compacted, immediate release tablet.

A quality-by-design study examining the impact of variability in excipient material properties on the quality attributes of an immediate release tablet was performed. A literature review and risk analysis identified particle size of microcrystalline cellulose (MCC), spray-dried lactose (SDL), and magnesium stearate (MgSt), and polymorph and specific surface area of MgSt as potential high-risk material properties. The following results were obtained with laboratory-scale processing equipment: (1) a 32-µm increase in d(50) (mean particle diameter) of MCC and SDL led to a ∼ 30-µm increase in blend and granulation d(50) and a statistically significant increase in the blend and granulation flow function coefficients, and (2) a 32-µm increase in d(50) of MCC and SDL, a 4.4 m(2)/g increase in the surface area, and a 19-µm decrease in the particle size of MgSt yielded an 18%-28% increase in ribbon tensile strength and tablet hardness. Confirmatory experiments with kilo-scale equipment showed impact of excipient variability on granulation particle size and tablet hardness was ∼ 50% smaller. Although the impact of these differences on overall manufacturability and performance of the tablets examined here were deemed low, the presence of statistically significant effects supports examining excipient variability as part of the design and control strategy of new drug products.

Improving the hardness of dry granulated tablets containing sodium lauryl sulfate.

The impact of the addition of a wetting agent, the surfactant sodium lauryl sulfate (SLS), on the tablet hardness of a dry granulated, solid oral dosage form was investigated. In three batches, SLS was added concurrently with: (1) a poorly soluble, highly hydrophobic active pharmaceutical ingredient (API) and the other excipients prior to the initial blending step, (2) magnesium stearate prior to roller compaction, or (3) magnesium stearate prior to tableting. A fourth batch, which did not contain SLS, served as a control. The maximum hardness of 100 mg, 1/4″-SRC tablets for the four batches--SLS added initially, prior to roller compaction, prior to tableting, and no SLS--were 61±3, 71±3, 89±5, and 86±3N, respectively, suggesting reduced processing of SLS improves tablet hardness by ∼50%. Dissolution of the tablets in 900 ml of simulated gastric fluid with paddles at 75 rpm showed that: (1) there was no impact on the insertion point of SLS into the process on API dissolution, and (2) that the presence of SLS improved dissolution by 5% compared to the control tablets. Adding SLS just prior to tableting can improve tablet hardness and yield similar dissolution performance relative to SLS addition prior to the initial blending step.

Scale-up model describing the impact of lubrication on tablet tensile strength.

Lubrication of 2:1 and 1:1 blends of microcrystalline cellulose and spray-dried lactose or dibasic calcium phosphate (DCP) with 0.33% or 1% magnesium stearate, as model free-flowing pharmaceutical formulations, was performed in rotary drum blenders. Blender process parameters examined in this study included type (Bin, V, and Turbula), volume (0.75-Quart to 200-L), fraction of headspace in the blender after the blend is loaded (30-70%), speed (6-202 rpm), and time (up to 225 min). Based on analysis of the experimental data, the following model for the impact of the lubrication process on tablet tensile strength at 0.85 solid fraction, TS(SF=0.85), was obtained, TS(SF=0.85)=TS(SF=0.85,0) [ßexp(-γ×V(1/3)×F(headspace)×r)+(1-ß)], where V is blender volume, F(headspace) is the headspace fraction, r is the number of revolutions (i.e. speed × time), TS(SF=0.85,0) is the initial tensile strength of the blend, ß is the sensitivity of the blend to lubrication, and γ is the lubrication rate constant of the formulation. This model can be used to maintain tensile strength during scale-up, by ensuring that (V(1/3)F(headspace)r)(1)=(V(1/3)F(headspace)r)(2). The model also suggests that formulations with DCP are less sensitive to lubrication and more slowly lubricated than formulations with spray-dried lactose (i.e. smaller ß and γ values).

Heterogeneity in skin treated with low-frequency ultrasound.

Recent experimental evidence using colored, fluorescent permeants suggests that skin treated with low-frequency sonophoresis (LFS) is perturbed in a heterogeneous manner. Macroscopic and microscopic visualization studies, topical penetration studies, transdermal permeability studies, and skin electrical resistivity measurements have shown that discrete domains, referred to as localized transport regions (LTRs), which are formed during LFS treatment of the skin, possess greatly reduced barrier properties, and therefore exhibit increased permeant skin penetration, compared to the surrounding regions of LFS-treated skin. The transformation of LTR formation from a heterogeneous to a homogeneous phenomenon has the potential benefit of increasing the maximum level of transdermal permeability or of reducing the area of skin required to deliver a desired dose of drug transdermally. Future studies, aimed at elucidating both the mechanisms of LTR formation and the limits of nondamaging formation of LTRs in the skin, are required to incorporate these proposed improvements to enhance the efficacy and practical utility of low-frequency sonophoresis.

Evaluation of hydrophilic permeant transport parameters in the localized and non-localized transport regions of skin treated simultaneously with low-frequency ultrasound and sodium lauryl sulfate.

The porosity (epsilon), the tortuosity (tau), and the hindrance factor (H) of the aqueous pore channels located in the localized transport regions (LTRs) and the non-LTRs formed in skin treated simultaneously with low-frequency ultrasound (US) and the surfactant sodium lauryl sulfate (SLS), were evaluated for the delivery of four hydrophilic permeants (urea, mannitol, raffinose, and inulin) by analyzing dual-radiolabeled diffusion masking experiments for three different idealized cases of the aqueous pore pathway hypothesis. When epsilon and tau were assumed to be independent of the permeant radius, H was found to be statistically larger in the LTRs than in the non-LTRs. When a distribution of pore radii was assumed to exist in the skin, no statistical differences in epsilon, tau, and H were observed due to the large variation in the pore radii distribution shape parameter (3 A to infinity). When infinitely large aqueous pores were assumed to exist in the skin, epsilon was found to be 3-8-fold greater in the LTRs than in the non-LTRs, while little difference was observed in the LTRs and in the non-LTRs for tau. This last result suggests that the efficacy of US/SLS treatment may be enhanced by increasing the porosity of the non-LTRs.

First-principles, structure-based transdermal transport model to evaluate lipid partition and diffusion coefficients of hydrophobic permeants solely from stratum corneum permeation experiments.

To account for the effect of branched, parallel transport pathways in the intercellular domain of the stratum corneum (SC) on the passive transdermal transport of hydrophobic permeants, we have developed, from first-principles, a new theoretical model-the Two-Tortuosity Model. This new model requires two tortuosity factors to account for: (1) the effective diffusion path length, and (2) the total volume of the branched, parallel transport pathways present in the SC intercellular domain, both of which may be evaluated from known values of the SC structure. After validating the Two-Tortuosity model with simulated SC diffusion experiments in FEMLAB (a finite element software package), the vehicle-bilayer partition coefficient, K(b), and the lipid bilayer diffusion coefficient, D(b), in untreated human SC were evaluated using this new model for two hydrophobic permeants, naphthol (K(b) = 225 +/- 42, D(b) = 1.7 x 10(-7) +/- 0.3 x 10(-7) cm(2)/s) and testosterone (K(b) = 92 +/- 29, D(b) = 1.9 x 10(-8) +/- 0.5 x 10(-8) cm(2)/s). The results presented in this paper demonstrate that this new method to evaluate K(b) and D(b) is comparable to, and simpler than, previous methods, in which SC permeation experiments were combined with octanol-water partition experiments, or with SC solute release experiments, to evaluate K(b) and D(b).

Evaluation of the porosity, the tortuosity, and the hindrance factor for the transdermal delivery of hydrophilic permeants in the context of the aqueous pore pathway hypothesis using dual-radiolabeled permeability experiments.

The aqueous pore pathway hypothesis has been modified to include both transient and steady-state domains of diffusive transport to evaluate the porosity, the tortuosity, and the hindrance factor of the skin aqueous pore channels from an individual dual-radiolabeled permeability experiment. Using these theoretical and experimental methods, the porosity (epsilon), the tortuosity (tau), and the hindrance factor (H) of the skin aqueous pore channels were evaluated as a function of: (i) the radius of the selected model hydrophilic permeants (urea, mannitol, raffinose, and inulin), and (ii) the extent of skin perturbation present in untreated skin, skin pretreated at a low dose, and a high dose, with a simultaneous application of 20 kHz ultrasound and the surfactant sodium lauryl sulfate (SLS), and the dermis. The results of this investigation revealed that the tortuosity decreased, and only the hindrance factor for inulin was significantly less than 1, over the range of permeant radii examined. Furthermore, only the porosity increased over the range of skin perturbation examined (over 100-fold), suggesting that a surface-related phenomenon is primarily responsible for the observed enhancement in the transdermal permeability of hydrophilic permeants induced by the simultaneous application of ultrasound and SLS.

Dual-channel two-photon microscopy study of transdermal transport in skin treated with low-frequency ultrasound and a chemical enhancer.

Visualization of transdermal permeant pathways is necessary to substantiate model-based conclusions drawn using permeability data. The aim of this investigation was to visualize the transdermal delivery of sulforhodamine B (SRB), a fluorescent hydrophilic permeant, and of rhodamine B hexyl ester (RBHE), a fluorescent hydrophobic permeant, using dual-channel two-photon microscopy (TPM) to better understand the transport pathways and the mechanisms of enhancement in skin treated with low-frequency ultrasound (US) and/or a chemical enhancer (sodium lauryl sulfate--SLS) relative to untreated skin (the control). The results demonstrate that (1) both SRB and RBHE penetrate beyond the stratum corneum and into the viable epidermis only in discrete regions (localized transport regions--LTRs) of US treated and of US/SLS-treated skin, (2) a chemical enhancer is required in the coupling medium during US treatment to obtain two significant levels of increased penetration of SRB and RBHE in US-treated skin relative to untreated skin, and (3) transcellular pathways are present in the LTRs of US treated and of US/SLS-treated skin for SRB and RBHE, and in SLS-treated skin for SRB. In summary, the skin is greatly perturbed in the LTRs of US treated and US/SLS-treated skin with chemical enhancers playing a significant role in US-mediated transdermal drug delivery.

Experimental demonstration of the existence of highly permeable localized transport regions in low-frequency sonophoresis.

Recent advances in low-frequency sonophoresis have focused on the existence of hypothesized localized transport regions (LTRs). However, there has been no actual experimental demonstration that the hypothesized LTRs are, in fact, localized regions of high permeability. Through a series of low-frequency sonophoresis experiments conducted with full-thickness pig skin, in the presence of the surfactant sodium lauryl sulfate (SLS), in which we have separately measured the transport of calcein through the LTRs, which have areas ranging from 10 to 40 mm(2), and the surrounding regions of the skin (the non-LTRs) by means of a novel masking technique, we demonstrate that the calcein permeability through the LTRs is approximately 80-fold higher than the calcein permeability through the non-LTRs, suggesting that the LTRs are structurally perturbed to a greater extent than the non-LTRs from the exposure to the ultrasound/SLS system. In addition, we propose basic models to predict the total skin transdermal permeability from the transdermal permeabilities of the LTRs and the non-LTRs, and then compare the predictions to the experimental data obtained from the masking experiments. We also demonstrate that both the LTRs and the non-LTRs exhibit significant decreases in skin electrical resistivity relative to untreated skin ( approximately 5000-fold and approximately 170-fold, respectively), suggesting the existence of two levels of significant skin structural perturbation due to ultrasound exposure in the presence of SLS. Finally, an analysis of the porosity/tortuosity ratio (epsilon/tau) values suggests that trans-cellular transdermal transport pathways are present within the highly permeable, and highly structurally perturbed, LTRs.