Pierre Boulas

Development of a NIR-based blend uniformity method for a drug product containing multiple structurally similar actives by using the quality by design principles

The aim of this study is to develop an at-line near infrared (NIR) method for the rapid and simultaneous determination of four structurally similar active pharmaceutical ingredients (APIs) in powder blends intended for the manufacturing of tablets. Two of the four APIs in the formula are present in relatively small amounts, one at 0.95% and the other at 0.57%. Such small amounts in addition to the similarity in structures add significant complexity to the blend uniformity analysis. The NIR method is developed using spectra from six laboratory-created calibration samples augmented by a small set of spectra from a large-scale blending sample. Applying the quality by design (QbD) principles, the calibration design included concentration variations of the four APIs and a main excipient, microcrystalline cellulose. A bench-top FT-NIR instrument was used to acquire the spectra. The obtained NIR spectra were analyzed by applying principal component analysis (PCA) before calibration model development. Score patterns from the PCA were analyzed to reveal relationship between latent variables and concentration variations of the APIs. In calibration model development, both PLS-1 and PLS-2 models were created and evaluated for their effectiveness in predicting API concentrations in the blending samples. The final NIR method shows satisfactory specificity and accuracy.

Engineered particles demonstrate improved flow properties at elevated drug loadings for direct compression manufacturing

Optimizing powder flow and compaction properties are critical for ensuring a robust tablet manufacturing process. The impact of flow and compaction properties of the active pharmaceutical ingredient (API) becomes progressively significant for higher drug load formulations, and for scaling up manufacturing processes. This study demonstrated that flow properties of a powder blend can be improved through API particle engineering, without critically impacting blend tabletability at elevated drug loadings. In studying a jet milled API (D50=24μm) and particle engineered wet milled API (D50=70μm and 90μm), flow functions of all API lots were similarly poor despite the vast difference in average particle size (ffc<4). This finding strays from the common notion that powder flow properties are directly correlated to particle size distribution. Upon adding excipients, however, clear trends in flow functions based on API particle size were observed. Wet milled API blends had a much improved flow function (ffc>10) compared with the jet milled API blends. Investigation of the compaction properties of both wet and jet milled powder blends also revealed that both jet and wet milled material produced robust tablets at the drug loadings used. The ability to practically demonstrate this uncommon observation that similarly poor flowing APIs can lead to a marked difference upon blending is important for pharmaceutical development. It is especially important in early phase development during API selection, and is advantageous particularly when material-sparing techniques are utilized.

At-line determination of pharmaceuticals small molecule's blending end point using chemometric modeling combined with Fourier transform near infrared spectroscopy

This article summarizes the development and validation of a Fourier transform near infrared spectroscopy (FT-NIR) method for the rapid at-line prediction of active pharmaceutical ingredient (API) in a powder blend to optimize small molecule formulations. The method was used to determine the blend uniformity end-point for a pharmaceutical solid dosage formulation containing a range of API concentrations. A set of calibration spectra from samples with concentrations ranging from 1% to 15% of API (w/w) were collected at-line from 4000 to 12,500cm-1. The ability of the FT-NIR method to predict API concentration in the blend samples was validated against a reference high performance liquid chromatography (HPLC) method. The prediction efficiency of four different types of multivariate data modeling methods such as partial least-squares 1 (PLS1), partial least-squares 2 (PLS2), principal component regression (PCR) and artificial neural network (ANN), were compared using relevant multivariate figures of merit. The prediction ability of the regression models were cross validated against results generated with the reference HPLC method. PLS1 and ANN showed excellent and superior prediction abilities when compared to PLS2 and PCR. Based upon these results and because of its decreased complexity compared to ANN, PLS1 was selected as the best chemometric method to predict blend uniformity at-line. The FT-NIR measurement and the associated chemometric analysis were implemented in the production environment for rapid at-line determination of the end-point of the small molecule blending operation.

Impact of Shear History on Powder Flow Characterization Using a Ring Shear Tester

In this study, we have investigated the impact of repeated shear displacement on powder flow properties. We show that when multiple yield loci are obtained using the same bulk solid specimen by stepping through different stress levels (i.e., stress walk [SW]), the shear deformation of the powder in a rotational shear cell, that is, Schulze Ring Shear Tester, is maximized, reducing the powder shear strength. This approach is material and time sparing; however, it imprecisely predicts better powder flowability. The magnitude of the change in the unconfined yield strength, σc, due to this prolonged shear displacement appears to be material-dependent, being less impactful for free-flowing powders. Using the SW and the individual yield loci-generated flow properties, we have demonstrated that in hopper design, the shear displacement effect impacts the computed critical arching diameter more than the critical mass flow angle. This knowledge of powder flow properties highlights limitations associated with the SW. An exponential function was found to describe the relationship between the change in σc at the highest major principal stress and the density weighted flowability, ffρ, with an R2 of 0.98. Such a model could be a valuable tool for correcting shear strength results obtained from SW, saving time, and material.

A systematic evaluation of dual functionality of sodium lauryl sulfate as a tablet lubricant and wetting enhancer

ABSTRACT Appropriate lubrication is important in tablet manufacturing as it lowers punch sticking propensity and protects tooling by reducing friction between die wall and tablet during tablet manufacturing. Most commercial lubricants negatively impact tabletability and dissolution. A delicate balance is usually attained by trial and error to identify the optimal level of lubricant in a tablet formulation. In this work, we have evaluated the effectiveness of sodium lauryl sulfate (SLS), a surfactant, as a tableting lubricant. If adequate lubrication efficiency is achieved, the use of SLS may be suitable to mitigate problems associated with hydrophobic lubricants. Results show that SLS, when applied in the proper amount to typical pharmaceutical powder mixtures, achieved lubrication efficiency comparable to a grade of magnesium stearate (MgSt) without deteriorating tabletability. Moreover, SLS‐containing tablets of celecoxib also exhibited improved in vitro dissolution compared to MgSt‐containing tablets. The enhancement in dissolution properties was attributed to the improved wetting by the dissolution medium due to the presence of SLS.

Low colonic absorption drugs: risks and opportunities in the development of oral extended release products

ABSTRACT Introduction: Currently numerous drugs have been observed with lower colonic absorption than small intestine absorption, which can significantly impact in vivo performance of their oral extended release (ER) products. Areas covered: We reviewed over 300 publications, patents, book chapters, and commercial reports of drug products from regulatory agencies for low colonic absorption (LCA) drugs and critical findings are discussed. The focuses of this article are (1) current findings on the causes of low colonic absorption to support early assessment of LCA candidates, and (2) current knowledge on successful ER strategies and technical platforms used for LCA drugs in commercial drug products to facilitate oral ER product development. Expert opinion: Colonic drug absorption is one of the critical considerations in successful development of oral ER products. The root causes of low colonic absorption in many LCA drugs are still unclear. It is recommended to evaluate colonic drug absorption of drug candidate at early stage of oral ER product development. After evaluation, the selection of a formulation platform to develop an oral ER product needs to be carefully considered for LCA drugs. Based on the current commercial oral ER formulation platforms for LCA drugs, compounds are first divided into five types (I-V) and different ER formulation approaches with higher success rate are recommended for each type.