Effective HPLC method development(Part II)

Effective HPLC method development(Part II)

5. Initial Method Conditions

The objective at this stage is to quickly develop HPLC conditions for subsequent method development experiments.  A common mistake is that scientists spend too much time at this stage trying to get a perfect separation.

5.1 Preliminary HPLC Conditions

In order to develop preliminary HPLC conditions in a timely fashion, scientists should use artificial mixtures of active pharmaceutical ingredients and related substances at relatively high concentrations (e.g., 1-2% of related substance relative to API) to develop the preliminary HPLC conditions. The concentration ratio between API and the related substances should be maintained to ensure the chromatography represents that of a real sample.  Alternatively, a highly stressed sample (e.g., 5% degradation) can also be used at this stage.  With the known composition and high levels of degradation products in the sample, one can evaluate the chromatography to determine whether there are adequate separations for all analyte. The high concentrations of related substances are used to ensure all peaks will be detected.

Computer assisted method development can be very helpful in developing the preliminary HPLC conditions quickly.  Since the objective at this stage is to quickly develop HPLC conditions for subsequent method development experiments, scientists should focus on the separation of the significant related substances (section 3.1.1) instead of trying to achieve good resolution for all related substances.  These significant related substances should be baseline resolved from each other with Rs > 2.0.  After the preliminary method development, the HPLC conditions can be further fine-tuned at a later stage (see section 8, method optimization/ robustness) to achieve the required specificity for the other related substances.

5.2 Aged HPLC Column

An aged HPLC column should be used to develop the initial HPLC conditions.  Usually it is more difficult to achieve the required resolution with an aged column (e.g., column with about 200 injections).  This will reflect the worst case scenario likely to be encountered in actual method uses, and help the long-term method robustness. In general, develop all methods with HPLC columns from the same vendor.  The preferred brand of HPLC column should be selected primarily based on the long term stability and lot to lot reproducibility.

6. Sample Preparation

6.1 Selection of Sample Solvent

This stage focuses on the selection of the sample solvent (for extraction) and the proper sample preparation procedures.  Investigate the effect of sample solvents of different % organic, pH, extraction volume and extraction procedure on accuracy, precision, sensitivity (LOQ) and the changes in the chromatography (e.g., peak shape, resolution). Whenever possible use the mobile phase in the sample preparation.  This will ensure that there will not be any compatibility issues between the sample solution and the HPLC conditions.

6.1.1 Accuracy: To investigate the accuracy in sample preparation (i.e., extraction efficiency), prepare a spiked solution by adding known amounts of related substances into a sample matrix.  Compare responses of the spike solutions and the neat standard solutions to assess the recovery from the sample preparation.  In this stage, since only one particular step is being investigated (i.e., sample preparation), close to theoretical recovery should be observed at this point (e.g., 90-110%).

6.1.2 Precision: Use the stressed sample to represent the worst case scenario and perform replicate sample preparations from the same sample composite.  Investigate the consistency of the related substance profile (i.e., any missing peaks?) and the repeatability results from these preparations.

6.2 Another objective is to determine the sample concentration that gives an acceptable LOQ (Signal to Noise ratio, S/N) in low level spike concentrations.  The sample concentration should be low enough to maintain linearity and precision, but high enough to achieve the desired LOQ. For example, if the ICH reporting limit for this drug product is 0.1%, the LOQ of the method should be less than 0.05% (i.e., desired LOQ, in %).  By using spike sample solutions of very diluted concentrations for the significant related substances, estimate the concentrations that give a S/N of about 10 for the significant related substances.  This estimated concentration is the approximate LOQ concentration (i.e., estimated LOQ concentration, in mg/mL).

The following equation can be used to estimate the target sample concentration for the method:

Target sample concentration =

estimated LOQ concentration (mg/mL) x 1/desired LOQ (%) x 100%

Continue ... Part III