(T1030-08-46) Factors to Consider for Increasing HPLC Column Lifetime

Purpose: High pressure liquid chromatography (HPLC), ultrahigh pressure liquid chromatography (UHPLC), and HPLC coupled to mass spectrometry (LC-MS) are among the most common separation, purification and analytical characterization methods in drug substance manufacturing of both synthetic small molecules and biopharmaceutical large molecules. It is well known that the proper sample preparation technique is critical to achieving high-quality data; however, certain sample preparations are commonly omitted in the preparation of pharmaceutical samples for HPLC or LC-MS for characterization. The hypothesis of this study was that omitting filtration within these workflows would cause instrument wear and reduce column lifetime, ultimately leading to high replacement costs, instrument downtime and lower quality data. Further, a study was conducted to investigate the influence of membrane type, pore size and manufacturer on the particulate retention of different syringe filters, as a factor that also correlates to column lifetime.

Methods: The retention of 0.5um and 0.24um diameter polystyrene beads was determined on multiple lots of syringe filters, 0.2um and 0.45um pore sizes, made of either hydrophilic polytetrafluoroethylene (PTFE) or regenerated cellulose (RC), all from four different manufacturers. After 3mL of 0.05% solution of beads were filtered, retentate was collected, characterized spectrophotometrically, and compared to a standard concentration curve. Column lifetime was tested on a UHPLC (Waters Acquity H-Class) with an HPLC column (Ascentis Express C18, 2.7um particle size) by monitoring column backpressure changes, a measure of HPLC column lifetime, while injecting unfiltered and filtered solutions of beads (also a 0.05% solution) from 0.45um syringe filters from the four manufacturers (column conditions in Figure 1). Backpressure (psi) was monitored over 500 column injections or up to 8000 psi, whichever was achieved first.

Results: Retention studies demonstrated the variability in performance that can occur when using syringe filters of different materials, of different pore sizes, and from different manufacturers (Table 1). For example, MFR-2 demonstrated low retention of 0.24um beads, 49.8±31.8%, while showing 100±0.10% retention of 0.5um beads. Further, RC from MFR-4 demonstrated consistently lower retention than hydrophilic PTFE from MFR-1, MFR-2, and MFR-3. Variation from lot to lot was observed in those samples that had less than 95% retention. Injecting samples from syringe filters that demonstrate low retention or injecting unfiltered samples can damage downstream chromatographic analyses, ultimately leading to replacement costs and potentially low-quality data. This is demonstrated experimentally by the changes in column backpressure, which indicates the degree of particulate clogging occurring within the column (Figure 2). All the PTFE filters had high retention (Table 1), which led to no particulate matter entering the column, and no change in column backpressure; however, the RC filter, which demonstrated lower retention, lasted for only 71 injections before the column pressure was too high to continue. This was slightly longer than when injecting unfiltered samples, which lasted for 37 injections before the column needed to be changed, showing the detrimental effect that particulate matter can have on column lifetime and performance. It is important to note that these injections were only 10uL injection volume; 71 and 37 injections corresponded to only 710 and 370uL total injection into the instrument before a new column was needed.

Conclusion: Typical columns should last greater than 1000 injections; however, this study demonstrated that if particulates are unintentionally introduced into the system even at low concentrations and volumes, this lifetime could be severely shortened. Particulate matter can be introduced by either omitting filtration or by using syringe filters that demonstrate low particle retention. This retention can be affected by choice of membrane material, pore size, manufacturer and even lot number. Thus, to protect column lifetime, instrument integrity, and to produce consistent, high-quality data, filtration should be considered a vital part of sample preparation for HPLC, UHPLC and LC-MS workflows for characterization of pharmaceuticals.

Acknowledgements: The authors would like to acknowledge Maricar Dube and Cory Muraco. There are no conflicts of interest.


Figure 1: UHPLC conditions.


Table 1: Retention of 0.22um and 0.45um syringe filters of different materials and sizes.


Figure 2: Changes in backpressure vs. injection number for MFR-1 PTFE (purple circle), MFR-2 PTFE (blue square), MFR-3 PTFE (green triangle), MFR-4 RC (yellow diamond) and unfiltered (pink dashes).