All in the blend for drug tablets

In addition to the active ingredient that produces the medicinal effect, drug tablets contain a whole load of other ingredients, including excipients that stabilise and enhance the active ingredient and binders that bind everything together. Understandably, these ingredients need to be blended together at just the right concentrations for the resultant tablet to work properly, but this blending process can be affected by a whole range of different factors.

These factors include the physical properties of the ingredients, such as particle size, the blending parameters, such as time and speed, and the environmental conditions, such as temperature and humidity. Fluctuations in any of these factors can alter the blend, with potentially detrimental consequences if this alters the concentration of the active ingredient, as this is one of the main causes of drug recalls.

Hence the need for regularly monitoring drug blends prior to them being turned into tablets. Current methods for doing this involve taking frequent samples of the pharmaceutical blend and then analysing them using an off-line technique such as high-performance liquid chromatography (HPLC). This approach is not ideal, however, both because it’s time consuming and because it can only sample a small subset of the blend, which might not be representative.

So a team of German chemists led by Gabriele Reich at the University of Heidelberg decided to assess the effectiveness of NIR spectroscopy at determining the concentration of the various ingredients making up pharmaceutical blends. As a non-invasive analytical technique, NIR spectroscopy could theoretically continuously analyse the blend as it passes by on a conveyor belt, removing the need to take discrete samples.

To test the ability of NIR spectroscopy, Reich and her team used it to analyse various blends comprising the nonsteroidal anti-inflammatory drug naproxen, as the active ingredient, with microcrystalline cellulose and croscarmelose sodium as the excipients and mannitol as a sweetener. They then used the resultant NIR data to develop models of varying complexity for determining the concentration of the ingredients, reporting their results in the European Journal of Pharmaceutics and Biopharmaceutics.

They found that a basic model was effective at determining the concentration of the active ingredient when environmental conditions remained fairly stable, but that a more complex model was required to determine the concentration of all the ingredients in fluctuating environmental conditions. Nevertheless, with this more complex model, on-line NIR spectroscopy proved to be just as accurate as off-line HPLC at determining the concentration of the ingredients. The only difference was that NIR spectroscopy consistently predicted slightly higher concentrations for the active ingredient than HPLC, perhaps due to HPLC having to analyse discrete samples.

Blog tags: