Using GPC/SEC for Excipient Characterization

The properties of polymeric excipients can directly affect the clinical efficacy, safety, and quality of a finished pharmaceutical product and are routinely identified as critical quality attributes (CQAs). This article looks at how gel permeation/size‑exclusion chromatography (GPC/SEC) can be applied to measure characteristics such as molecular weight (MW), MW distribution and structure, and degree of branching. Case study data for the measurement of poly-lactic acid (PLA) and poly‑lactic glycolic acid (PLGA) highlights the detailed information that can be accessed.

Photo Credit: samxmeg/Getty Images

Polymeric excipients are an important addition to the sophisticated tableting blends of today. Ingredients such as poly‑lactic acid (PLA), poly-lactic glycolic acid (PLGA), and hydroxyl methyl cellulose and its derivatives enable formulators to achieve closely controlled drug release profiles using a growing range of manufacturing techniques that includes spray drying, hot‑melt extrusion, and lipid‑based drug delivery. The properties of these polymers can directly affect the clinical efficacy, safety, and quality of the finished pharmaceutical product and are therefore often identified as critical quality attributes (CQAs). CQAs for polymer excipients typically include molecular weight (MW), MW distribution, and structural characteristics such as degree of branching.

Gel permeation/size-exclusion chromatography (GPC/SEC) is a powerful technique for the characterization of polymers and other macromolecules. In this article we examine its application in the analysis of polymeric excipients. Case study data for the measurement of PLA/PLGA highlights the detailed information that can be accessed.

The Vital Role of Excipients
The workflows associated with the development of oral solid dosage forms, whether innovator or generic, are increasingly well established and are rooted in a Quality by Design (QbD) approach.¹ These workflows emphasize the need for detailed characterization of the excipient, as well as the active ingredient, both alone and within the blend. The resulting information supports the development of a detailed understanding of how the drug product will behave and of a specification for each component that will ensure successful drug delivery and the necessary quality control.

Traditionally excipients have been used simply as “bulking agents” - in this case the impact of excipient properties on the safety, efficacy, and quality characteristics of the drug product may be relatively limited. However, in modern formulations polymeric excipients often play a far more active role controlling the drug delivery profile and other aspects of drug product performance. Polymers are now routinely used to:²

• Formulate coatings that control the rate of dissolution of the tablet in the stomach.

• Develop spray-dried solid dispersions for the delivery of sparingly soluble drugs.

• Improve blend flow properties.

• Tailor the taste and texture of tablet for improved patient acceptability.

• Improve the stability of tablets containing moisture-sensitive active ingredients.

Precisely differentiating between grades of excipient, choosing an optimal candidate, and ensuring the consistency of supply is critical to the manufacturing process; however, it can be complicated by two factors. Firstly, a number of pharmaceutical excipients are derived from natural polymers, such as naturally occurring celluloses, which can limit the manufacturer’s ability to control polymer properties. There is a choice of different grades and sources but only a limited ability to precisely tailor features such as branching and MW distribution. The second factor stems from the way in which the pharmaceutical industry is structured. An active ingredient often undergoes substantial development in-house, including the rigorous investigation of all CQAs, resulting in a complete understanding of how the drug substance behaves and, ideally, a detailed definition of the structure-function relationships that define its performance.

In contrast an excipient tends to be bought in, sometimes from multiple suppliers, making it difficult to secure supplies that enable a rigorous investigation of the effect of excipient properties. For example, assessing the impact of branching relies on sourcing excipients with different degrees of branching. Furthermore, once an excipient has been identified it can be quite challenging to scope and control the degree of variability associated with the supply. Therefore, although a polymer excipient may be a more straightforward chemical entity than the active ingredient, and indeed have a less direct impact on drug product performance, there are unique challenges associated with its characterization.

An Introduction to GPC/SEC
GPC/SEC is a two-step analytical technique where samples are first separated into fractions on the basis of hydrodynamic size (by passing through a packed chromatography column), which are subsequently characterized using one or more detectors. Measuring the amount of sample in each sized fraction enables determination of the size distribution, and more importantly MW and MW distribution data.

Traditional GPC/SEC systems use a single‑concentration detector, typically a refractive index (RI) detector. With this set‑up, column calibration with appropriate standards provides the correlation needed to estimate a relative (to the standard) MW distribution. These MW data are only accurate if the relationship between molecular size and weight is the same for the sample as it is for the standard. This is a crucial limitation especially when gathering precise data for the detailed comparison of relatively similar excipients, and when the calibration standards are sub-optimal for the polymers of interest.

The use of multiple detectors can directly address this limitation - for example, a light-scattering detector in combination with a concentration detector enables the direct measurement of absolute MW with minimal, non-specific calibration. Further complementary additions include a viscometer to enable the measurement of structural features such as branching or conformation.

GPC/SEC systems with sensitive multi‑detector characterization capabilities are consequently useful for QbD applications. The following case study demonstrates the application of multi‑detection GPC/SEC for the analysis of PLGA and PLA, polymers that are used routinely in pharmaceutical formulation.

Case Study: Analyzing PLA/PLGA Samples
Produced by polymerizing lactic acid or lactic and glycolic acid respectively, PLA and PLGA are used to formulate controlled drug release systems and to manufacture medical components such as absorbable surgical thread and implants.3 Derived from renewable and natural resources, they are commonly classified as “green polymers” because of their biodegradability and biocompatibility. The properties of PLGA can be controlled by varying the ratio of lactic to glycolic acid used in its production to manufacture polymers with different MW, MW distribution, and structure. Multi-detection GPC/SEC can be applied to measure and sensitively compare the properties of the resulting materials.

Method: Samples of PLA and of PLGA were analyzed using an OMNISEC GPC/SEC system with triple detector array (RI, light‑scattering, and viscometer detectors) (all Malvern Instruments). Tetrahydrofuran (THF) was used as the solvent for the samples and as the mobile phase. The analyses were performed at 30 ^oC using samples prepared and measured at concentrations in the range 1–5 mg/mL.

THF is a common solvent for many GPC/SEC applications that readily dissolves both PLA and PLGA; however, the RI sensitivity (dn/dc) for this sample/solvent combination is low (approximately 0.045 mL/g to 0.051 mL/g). This means that any change in polymer concentration induces a relatively small change in RI, requiring a very sensitive RI detector to measure concentration changes under these conditions. Likewise, dn/dc values also impact the response of light-scattering detectors. A compromise was therefore forced on the analyst by this dn/dc issue. They would either have to analyze the samples at a high concentration, which overloads the SEC columns and distorts the MW distribution obtained, or have to switch to an alternative solvent, normally acetone, which is far less suitable for the proper dissolution of the sample. However, the results gathered here using a multi-detector GPC/SEC approach demonstrate that it is sufficiently sensitive to enable the use of the preferred THF solvent.

Results: Figure 1 shows typical chromatograms for a PLA sample, with traces from each detector in the array, and derived values of MW and intrinsic viscosity (IV) plotted as a function of retention time. Excellent data quality is seen across all the measurements made with each detector, as evidenced by the clean stable baseline and the smooth well‑defined signal peaks. Similar data quality is also observed in the measurement of the PLGA samples (data not shown).

Using the MW and IV data generated by these measurements it is possible to construct a Mark-Houwink (M-H) plot, a logarithmic plot of IV against MW, to investigate the structural differences between samples. IV is inversely proportional to the molecular density in solution so both the gradient and intercept of a M-H plot reveal information about the structural characteristics of a polymer. Three different types of PLGA sample were measured to investigate the impact of composition on structure. These included copolymers with lactide:glycolide compositions of 50:50; 65:35; and 75:25, respectively. For the 50:50 copolymer, two samples were analyzed that had different overall MW, a low-to-mid MW sample, and a mid-to-high MW sample. This gave four samples in total and for each one duplicate injections were made on the GPC/SEC system to check repeatability.

The results indicate that increasing the amount of lactide in the polymer decreases its coil density in solution. Samples with a 75:25 lactide:glycolide ratio exhibit the highest IV at any given MW while those that have a 50:50 composition have a much lower IV, at an equivalent MW. This means that any change in the copolymer composition will change the molecular structure of the polymer. More practically, the data provide insight as to why these polymer samples will behave differently, thereby supporting the manipulation of polymer properties to control drug delivery performance. It is also worth noting that this easy differentiation of copolymer content by the M-H plot is independent of the molecular weight, as shown by the consistent, overlapping plots of the two 50:50 samples with different molecular weights. In other words, the exact excipient MW distribution and structure (composition) can be determined, or compared to a reference, in a single analysis.

In an extension of the study, multiple injections of the same PLGA sample were performed to directly assess reproducibility. Table 1 shows data from 10 repeat injections, each of 100 µL, for a PLGA sample containing 50% lactic acid, measured at a concentration of 2.132 mg/mL. A relative standard deviation of 0.53% for the 10 adjacent samples demonstrates the excellent repeatability obtained. Such repeatability with automation substantially lightens workload associated with the rigorous and extensive experimentation needed to adopt QbD.

Conclusion
The performance of many sophisticated pharmaceutical products relies on the use of polymeric excipients. The MW, MW distribution, and structural characteristics of such excipients define their behaviour and are therefore identified routinely as CQAs for the product. GPC/SEC is an established technique for the measurement of these properties and as a result has an important role to play in pharmaceutical formulation. The technique when applied with multiple detection systems can provide high sensitivity, which enables the precise differentiation of excipient grades and greater flexibility in solvent choice, and highly automated repeatable measurement, for efficient, high productivity analysis. It can also allow formulators to gather the information needed to fully scope excipient performance, and to swiftly specify an optimal excipient for any given drug product.

References

1. “Analytical techniques with a place in the oral solid dosage formulation toolkit’ Whitepaper available for download at: http://www.malvern.com/en/support/resource-center/Whitepapers/WP141223AnalTechOralToolkit.aspx

1. A. Siew, Pharmaceutical Technology Europe 39(1), (2015).

1. M. Chaubal, Drug Delivery Technology2(5), (2002).

Stephen Ball is Product Marketing Manager, Nanoparticle and Molecular Characterization, at Malvern Instruments. He holds a degree in computer aided chemistry from the University of Surrey, UK, which included a year in industry working as a research chemist for the Dow Chemical Company in Horgen, Switzerland. Before joining Malvern Instruments, he worked for Polymer Laboratories as an applications chemist, before taking on a marketing position as a product manager for light scattering instrumentation at Agilent Technologies.

E-mail: Stephen.ball@malvern.com

Website:www.malvern.comThis article was featured in LCGC's digital magazine, The Column. Click here to view the full issue>>