Improved Prediction of in Vivo Supersaturation and Precipitation of Poorly Soluble Weakly Basic Drugs Using a Biorelevant Bicarbonate Buffer in a Gastrointestinal Transfer Model.

The characterization of intestinal dissolution of poorly soluble drugs represents a key task during the development of both new drug candidates and drug products. The bicarbonate buffer is considered as the most biorelevant buffer for simulating intestinal conditions. However, because of its complex nature, being the volatility of CO2, it has only been rarely used in the past. The aim of this study was to investigate the effect of a biorelevant bicarbonate buffer on intestinal supersaturation and precipitation of poorly soluble drugs using a gastrointestinal (GI) transfer model. Therefore, the results of ketoconazole, pazopanib, and lapatinib transfer model experiments using FaSSIFbicarbonate were compared with the results obtained using standard FaSSIFphosphate. Additionally, the effect of hydroxypropyl methylcellulose acetate succinate (HPMCAS) as a precipitation inhibitor was investigated in both buffer systems and compared to rat pharmacokinetic (PK) studies with and without coadministration of HPMCAS as a precipitation inhibitor. While HPMCAS was found to be an effective precipitation inhibitor for all drugs in FaSSIFphosphate, the effect in FaSSIFbicarbonate was much less pronounced. The PK studies revealed that HPMCAS did not increase the exposure of any of the model compounds significantly, indicating that the transfer model employing bicarbonate-buffered FaSSIF has a better predictive power compared to the model using phosphate-buffered FaSSIF. Hence, the application of a bicarbonate buffer in a transfer model set-up represents a promising approach to increase the predictive power of this in vitrotool and to contribute to the development of drug substances and drug products in a more biorelevant way.

Application of an automated small-scale in vitro transfer model to predict in vivo precipitation inhibition.

The majority of NCEs are weakly basic drugs. Consequently, their solubility is highly pH-dependent, with higher solubility in the acidic stomach and poor solubility in the neutral intestinal environment. The gastric emptying of dissolved drug can lead to the intestinal precipitation of the drug. One option of reducing this process is to formulate the drug together with a precipitation inhibitor (PI). The aim of this study was to investigate the effects of different PIs on the intestinal concentrations of ketoconazole and five orally administered kinase inhibitors (i.e. pazopanib, gefitinib, lapatinib, vemurafenib, and a Merck KGaA research compound, MSC-A) with the aid of a predictive small-scale in vitro transfer model. This screening revealed that HPMCAS and Soluplus® were the most effective PIs. Whereas all other drugs precipitated within several minutes, gefitinib expressed highly variable amorphous precipitation which was confirmed by PXRD. During the transfer model experiments, this intermediate supersaturated state was stabilized using HPMCAS and Soluplus®. The PI screening protocol described herein allows to study the effect of PIs for solubility and potential bioavailability improvement of poorly soluble drugs to support formulation development already in early stages.

Automated small-scale in vitro transfer model as screening tool for the prediction of in vivo-dissolution and precipitation of poorly solubles.

Precipitation testing, especially for weakly basic APIs, represents a key parameter in drug substance characterization during early development stages, where the amount of API available is limited. Therefore, it was the aim of this study to develop an automated small-scale in vitro transfer model to characterize the supersaturation and precipitation behavior of two poorly water-soluble drugs. Following automation and scale-down of the standard transfer model, the developed small-scale model was used to assess the impact of gastrointestinal variability, i.e. gastric pH, gastric emptying, and gastrointestinal fluid volumes, on supersaturation and precipitation of two weakly basic model compounds, ketoconazole and a new chemical entity from the research laboratories of Merck KGaA, MSC-A. The experiments revealed that variations in gastrointestinal parameters affected the in vitro behavior of ketoconazole, but not of MSC-A. Elevated gastric pH, as it can result from co-medication with acid-reducing drugs, resulted in lower degrees of supersaturation for both substances. This result is in agreement with the observation that the oral bioavailability of ketoconazole is lowered when proton pump inhibitors are co-administered. The small-scale transfer model presented herein represents a valuable in vitro tool to assess the risk of drug precipitation, additionally covering a broad range of gastrointestinal parameters.

Premature drug release of polymeric micelles and its effects on tumor targeting.

Based on the enhanced permeability and retention (EPR) effect, nanoparticles are believed to accumulate in tumors. In this conjunction, the stability of drug encapsulation is assumed to be sufficient. For clarification purposes, PEGylated poly-(D,L-lactic acid) (PEG-PDLLA) micelles which incorporated the hydrophobic model drug dechloro-4-iodo-fenofibrate (IFF) were investigated. H2N-PEG-PDLLA was synthesized, coupled to 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) and labeled with 111-indium. From this polymeric species, mixed micelles with H3CO-PEG-PDLLA were prepared which encapsulated the 125-iodine or 131-iodine labeled drug IFF. Bioimaging and biodistribution experiments in healthy and AR42J-tumor bearing mice were carried out to quantify the uptake of the drug and its carrier in single organs. As a result, upon injection of this system, a rapid dissociation of the polymeric carrier and the incorporated drug (<10 min post inj.) was revealed. Regardless of the premature release, the drug showed an enhanced tumor accumulation compared to the polymeric carrier. In conclusion, the self-assembling system allowed for successful solubilization of the hydrophobic drug by physical incorporation into micelles whereas the tumor targeting properties of the drug delivery system could not be sufficiently shown.

Drug loading of polymeric micelles.

PURPOSE: To gain mechanistic insights into drug loading and lyophilization of polymeric micelles. METHODS: PEGylated poly-4-(vinylpyridine) micelles were loaded with dexamethasone. Three different methods were applied and compared: O/W emulsion, direct dialysis, cosolvent evaporation. Micellar dispersions with the highest drug load were lyophilized with varying lyoprotectors: sucrose, trehalose, maltose, a polyvinylpyrrolidine derivative, and ß-cyclodextrin derivatives. For comparison, other PEGylated block copolymer micelles (PEGylated polylactic acid, polylactic acid-co-glycolic acid, polycaprolactone) were freeze-dried. RESULTS: Drug loading via direct dialysis from acetone was a less effective loading method which led to dexamethasone loads <2% w/w. O/W emulsion technique from dichlormethane increased drug load up to ~13% w/w; optimized cosolvent evaporation increased load up to ~19% w/w. An important step for cosolvent evaporation was solubility screen of the drug prior to preparation. Loading was maintained upon lyophilization with ß-cyclodextrins which proved to be versatile stabilizers for other block copolymer micelles. CONCLUSION: Careful solvent selection prior to cosolvent evaporation was a beneficial approach to load hydrophobic drugs into polymeric micelles. Moreover, ß-cyclodextrins could be used as versatile lyoprotectors for these micelles.

Development of a new method to assess nanocrystal dissolution based on light scattering.

PURPOSE: Nanocrystals exhibit enhanced dissolution rates and can effectively increase the bioavailability of poorly water soluble drug substances. However, methods for in vitro characterization of dissolution are unavailable. The objective of this study was to develop an in situ noninvasive analytical method to measure dissolution of crystalline nanosuspensions based on light scattering. METHODS: Fenofibrate nanosuspensions were prepared by wet media milling. Their solubilities and dissolution profiles in simulated gastric fluid supplemented with 0.1% Tween(®) 80 were measured in a small scale setup with an instrument for dynamic light scattering and the intensity of scattered light as readout parameter. RESULTS: A good correlation was achieved between the dissolution profile of a nanosuspension measured in the light scattering setup and a conventional dissolution experiment. Nanosuspensions of 120-270 nm size could be distinguished by the light scattering method. The suspensions dissolved within 1.9-12.3 min. Over a concentration range of 40-87% of the solubility dissolution profiles of a nanosuspension with 140 nm were monitored and the determined total dissolution times were in good agreement with the Noyes-Whitney dissolution model. CONCLUSIONS: A noninvasive, sensitive and reproducible method is presented to assess nanocrystal dissolution. In situ measurements based on light scattering allow a straightforward experimental setup with high temporal resolution.

Analysis of immediate stress mechanisms upon injection of polymeric micelles and related colloidal drug carriers: implications on drug targeting.

Polymeric micelles are ideal carriers for solubilization and targeting applications using hydrophobic drugs. Stability of colloidal aggregates upon injection into the bloodstream is mandatory to maintain the drugs' targeting potential and to influence pharmacokinetics. In this review we analyzed and discussed the most relevant stress mechanisms that polymeric micelles and related colloidal carriers encounter upon injection, including (1) dilution, (2) interactions with blood components, and (3) immunological responses of the body. In detail we analyzed the opsonin-dysopsonin hypothesis that points at a connection between a particles' protein-corona and its tissue accumulation by the enhanced permeability and retention (EPR) effect. In the established theory, size is seen as a necessary condition to reach nanoparticle accumulation in disease modified tissue. There is, however, mounting evidence of other sufficient conditions (e.g., particle charge, receptor recognition of proteins adsorbed onto particle surfaces) triggering nanoparticle extravasation by active mechanisms. In conclusion, the analyzed stress mechanisms are directly responsible for in vivo success or failure of the site-specific delivery with colloidal carrier systems.

Controlled delivery of nanosuspensions from osmotic pumps: zero order and non-zero order kinetics.

Nanosuspensions have gained great interest in the last decade as a formulation tool for poorly soluble drugs. By decreasing particle sizes nanosuspensions enhance dissolution rate and bioavailability of the active pharmaceutical ingredient. Micro-osmotic pumps are widely used in experimental pharmacology and offer a tool of interest for the sustained release of nanosuspensions via the intraperitoneal or subcutaneous application site. The purpose of the present study was to investigate in-vitro the influence of (1) nanosuspension viscosity, (2) pump orifice position and (3) formulation osmolality on the delivery behavior of formulations in implantable osmotic systems. Therefore fenofibrate nanosuspension, methylene blue and fluorescein sodium solutions were chosen as model formulations. They were released in water or isotonic saline solution and drug/dye concentrations were determined by HPLC/UV. Release of nanosuspension particles in low viscous formulations resulted in a burst whereas increasing the viscosity led to the expected zero order delivery. Pumps with upward-positioned orifices released the nanosuspension in a zero order manner. Within the release of dyes, constant delivery could be ensured up to an osmolality of 486 mO sm/kg; above this value premature release of formulation was observed. The results indicate the requirement of in-vitro experiments prior to in-vivo animal testing for determining the release profiles of osmotic pumps.

Comparative investigations on in vitro serum stability of polymeric micelle formulations.

PURPOSE: Stability of polymeric micelles upon injection is essential for a drug delivery system but is not fully understood. We optimized an analytical test allowing quantification of micellar stability in biofluids and applied it to a variety of block copolymer micelles with different hydrophobic block architechtures. METHODS: Polymeric micelles were prepared from four different polymers and investigated via encapsulation of two fluorescent dyes. Samples were incubated in human serum; changes in Foerster Resonance Energy Transfer (FRET) were recorded as a function of time. This fluorescence-based approach was supported semi-quantitatively by results from Asymmetrical Flow Field-Flow-Fractionation (AF4). RESULTS: After incubation experiments, micellar stability was determined by calculation of two stability-indicating parameters: residual micellar fractions (RMFs) and in vitro serum half-lives. Both parameters showed that PEG-PVPy micelles rapidly destabilized after 3 h (RMF < 45%), whereas PEG-PLA, PEG-PLGA and PEG-PCL micelles were far more stable (RMFs 65 to 98%). CONCLUSION: This FRET-based assay is a valuable tool in evaluating and screening serum stability of polymeric micelles and revealed low serum stability of PEG-PVPy micelles compared to polyester-based micelles.

Comparison of different protein concentration techniques within preformulation development.

Highly concentrated antibody solutions are of increasing importance in the pharmaceutical industry. During production highly concentrated solutions are usually prepared by tangential flow filtration (TFF). Since this technique is often not applicable in the early phase of formulation development, where the available amounts of protein are commonly very small, small scale techniques like dialysis or ultrafiltration with stirred cells or centrifugal filters have to be employed. In this study the small scale techniques were compared to tangential flow filtration, with regard to the quality and stability of the concentrated products. The achievable concentration of a protein, when starting from a model antibody solution with 10mg/ml, was also assessed. Concentrations above 100mg/ml could be obtained with all techniques, however with different product qualities. The stability of the highly concentrated solutions (100 mg/ml) was analyzed by turbidity measurements, size exclusion chromatography (SEC), SDS-PAGE and isoelectric focusing (IEF) after storage at 25 and 40°C for 8 weeks. Solutions prepared by dialysis exhibited the smallest degree of instability, whereas those manufactured by centrifugal filtration revealed the best comparability to products obtained by tangential flow filtration with regard to the results of isoelectric focusing, turbidity measurements (UV-vis) and size exclusion chromatography. Stability differences were observed within all analytical methods, primarily after storage and not directly after the concentration process.

Relevant shaking stress conditions for antibody preformulation development.

In protein formulation development, shaking stress is often employed to assess the physical stability of antibody formulations against aggregation. Since there are currently no guidelines describing suitable test conditions, very different shaking stress designs are used. These different designs may influence the resulting stability data. The aim of this study was to establish a shaking stress design within the protein range of 2-5mg/ml which can rapidly distinguish between antibody formulations of poor stability and those with potential for further development. Small scale shaking stress experiments were performed with different monoclonal IgG antibodies (as buffered solutions or marketed formulations). Variables were the filling degree of the sample containers, the container type and size and the shaking intensity. The stability of the samples was assessed by visual inspection, UV-VIS spectrophotometric turbidity measurements and size exclusion chromatography. All tested parameters had a strong influence on the stability results. The most discriminating conditions were obtained when shaking of the formulations was performed at 200rpm in a 2ml injection vial filled with 1ml protein solution. This experimental setup led to clearly different stability results for buffered solutions and marketed products. Moreover, this setup required only relatively small amounts of protein solution which is advantageous in prefomulation studies.

[Evaluation and further development of a pyrogenicity assay based on human whole blood]

When cells of the immune system, i.e. primarily blood monocytes and macrophages, come into contact with pyrogens (fever inducing contaminations) they release mediators transmitting the fever reaction within the organism. A new pyrogen test exploits this reaction for the detection of pyrogens: human whole blood taken from healthy volunteers is incubated in the presence of the test sample. In case of pyrogen contamination, the formation of interleukin-1 is induced, which is determined by ELISA. According to the various pharmacopoeia, the rabbit pyrogen test determines the fever reaction following injection of a test sample. In comparison, the new whole blood assay is more sensitive, less expensive and determines the reaction of the targeted species. In contrast to the well established in vitro alternative, i.e. the limulus amebocyte lysate assay (LAL), the blood assay is not restricted to endotoxins of Gram-negative bacteria and is not to the same extent disturbed by endotoxin-binding blood proteins. Here, interim results of the ongoing optimisation and prevalidation are demonstrated. Preliminary data of the evaluation for biological and pharmaceutical drugs are presented.