Using specialized software in conjunction with UPLC®  Technology, an optimized QbD method for the impurities in vancomycin can be developed that will be robust for method validation and transfer. 

Analytical methods are developed at various stages of the drug develoment process for samples of varying complexity. Due to the inherent nature of the method development process, redundant efforts take place across an organization, resulting in a very costly and time-consuming activities. If we can streamline the process by which we develop methods, products can be brought to market faster and in a more cost effective manner. 

Many different approaches are typically used to develop chromatographic methods today including trial and error, method/column scouting, and software approaches such as first principles approaches and simplex optimization procedures. All these approaches suffer from the inability to determine complex interactions effects between method variables or measurably consider method robustness during the method development process. 

Vancomycin is a tricyclic glycopeptide antibiotic derived from 

) and is indicated for the treatment of serious or severe infections caused by susceptible strains of methicillin-resistant (beta-lactam-resistant) staphylococci. Vancomycin is a large molecule (MW 1485.71 daltons) and contains many impurities that are difficult if not impossible to separate. Traditional HPLC gradient methods have shown the ability to separate out as many as 13 of these impurities, while the use of sub-2-μm ACQUITY UPLC Column chromatography has demonstrated the separation of as many as 26 impurities. 

This paper describes a novel method development approach using Quality by Design (QbD) with Design of Experiments to develop a UPLC method for separating 39 impurities in vancomycin resulting in an optimally performing analytical method while simultaneously applying robustness limits to ensure success in final method validation and ultimately in method transfer. 

The vancomycin studies described here were carried out using an automated integrated system consisting of Fusion Method Development Software, Empower 2 Chromatography Data Software (CDS), and an ACQUITY UPLC System with PDA, Column Manager, and Solvent Select Valve allowing for the screening of up to four different column chemistries, six different aqueous buffers/pHs, and two different organic mobile phases in one run.  

Fusion Method Development Software (S-Matrix Corporation, Eureka, CA) is a Quality by Design based LC Method Development software package with built-in robustness metrics. Fusion includes a built-in interface with the Empower 2 CDS Software that controls the ACQUITY UPLC System. Using the chromatographic results collected from Empower 2 CDS, Fusion manages complex statistics and models for method optimization. Fusion builds experiments, analyzes data, and presents results as visual and numerical method predictions. 

The first phase of the method development involves the screening of the major effectors of selectivity, primarily the column chemistry, buffer pH, and organic mobile phase. The variables and ranges screened along with the constant conditions are listed in Figure 1's tables.  

Figure 1. Screened variables and ranges. 

The experimental design is run and data processed on the chromatographic system and the results are imported back into Fusion. The software predicts the optimum LC method after modeling all significant effects – linear, interaction, and complex – on each critical method performance characteristic. The unshaded (white) area of the overlay graph shown for the BEH C

 column with methanol as the organic mobile phase (Figure 2) highlights the experimental region where the mean performance goals are obtained.  

Figure 2. Processed data are imported to Fusion, where an overlay graph illustrates in white the region where the mean performance goals are achieved. 

The Automated Optimizer wizard defines the LC method performance goals and ranks them in order of importance. The software searches for the LC method that meets all the performance goals simultaneously. The best result(s) are reported along with predicted results for an experimental run (Figure 3). These conditions are used for the next stage, Method Optimization.  

Figure 3. Fusion’s Automated Optimizer facilitates determination of the LC method that meets all performance goals. 

Phase 2 experiments use the column (ACQUITY UPLC BEH C8, 2.1 x 100 mm, 1.7 μm) and mobile phase (pH 5.0 buffer, methanol B solvent) results from Phase 1 plus additional variables with tighter ranges to determine the optimum LC method. The experimental design is created using pump flow rate, gradient time, final percent organic, and column temperature as final optimization variables in the ranges shown (Figure 4).  

Figure 4. Fusion determines optimal method conditions and exports this information back to Empower 2 to be run and processed. 

Fusion Software creates the experimental design and exports it to Empower 2, automatically creating all the necessary instrument methods, method sets, and sample sets. The experimental design is run and data processed on the chromatographic system and the results are imported back into Fusion.  

In addition to the data analysis for method optimization, Fusion applies a combination of Monte Carlo Simulation and Process Capability statistics to evaluate method robustness without running additional experiments.  

Visualizing the results with Fusion's 3D Response Surface Plots demonstrates the combined effects of variables on key chromatographic responses such as resolution, peak tailing, and retention time. Colors represent the magnitude of interaction and the curvature indicates the type of interaction (Figure 5).  

Another option for visualizing the interactions between variables is through Multiple Response Effects Plots. These plots clearly identify simple linear additive effects (1), complex interactions of variables (2), and other types of response effects (Figure 6).  

Articles

Analytical methods are developed at various stages of the drug develoment process for samples of varying complexity.

A QbD with Design-of-Experiments Approach to the Development of a Chromatographic Method for the Separation of Impurities in Vancomycin

"Equivalent" vs. "Equal" column selectivity 

Increase productivity while decreasing costs

Discover software tools to facilitate method translation

ACQUITY UPLC®  and ACQUITY UPLC H-Class  

Method Transfer, Compendial Methods, USP, Method Translation, UPLC, HPLC

Pharmaceutical research and development (R&D) organizations were early adopters who recognized the many benefits of UltraPerformance LC®  (UPLC®) Technology including resolution, sensitivity, throughput, and productivity as compared to HPLC. Today, the number of projects involving new drug entities are increasingly performed utilizing UPLC.  

Adopting UPLC for R&D activities is less complex than for laboratories involved with routine analysis, where its use requires consideration about the need to re-file methods for existing products. Routine analysis areas such as Quality Control (QC) laboratories own a vast supply and variety of HPLC instrumentation. Asset procurement regarding new technologies within these groups often requires convincing financial as well as scientific justification.  

Although information illustrating UPLC's return on investment (ROI) for solvent consumption and analysis per unit time can be convincing for R&D, the QC environment requires key practical-use considerations. Managers and end users within QC laboratories require new instrumentation to provide dual purposes: first, the ability to perform both legacy methods and, second, the ability to use sub-2-μm particle columns and methodology in a routine analytical environment without complications. UPLC's adoption must also strategically provide seamless integration within current laboratory practices and decrease learning curves of the end users.  

In this application, various U.S. Pharmacopeia (USP) compendial methods are used as examples to highlight a new method translation strategy to facilitate the transfer of methods to and from any LC-based instrument with ease. 

United States Pharmacopeia reference standards

      • USP Galantamine Hydrobromide RS and USP Galantamine Hydrobromide Related Compounds Mixture RS  

USP Dietary Supplement: Powdered Soy Isoflavone Extract Method

      • USP Apigenin RS, USP Diadzein RS, 

        USP Diadzin RS, USP Genistein RS, 

        USP Genistin RS, USP Glycitein RS, 

        USP Glycitin RS, and USP Defatted Powdered Soy RS 

      • USP Loratadine RS, USP Loratadine Related Compound A RS, and USP Loratadine Compound B RS, Claritin 

References to LC conditions are addressed as per USP Monographs, whereas specific utilization of LC instrumentation for each application is discussed in the figure captions. 

Successful method translation requires understanding three key chromato-graphic attributes before implementation. The analyst must consider the differences between LC instrumentation, column selectivity, and the resolving capability of the original methodology versus the target methodology. By understanding these three essential aspects of method translation, the benefits of increasing productivity and decreasing costs while maximizing asset utilization of present and future instrumentation can be realized. 

Future-proofing your laboratory: Translating HPLC methodology between LC instrumentation

The QC laboratory frequently utilizes a variety of LC instruments for API and drug product analysis. Therefore, instrumentation flexibility is essential. Direct transfer of methods to newer technology may result in retention time and selectivity differences that may be related to decreases in instrument dwell volume.  

Figure 1. USP Method for galantamine and related substances performed on an AllianceÂ® HPLC 2695 with measured dwell volume of 1.1 mL. An XBridgeâ¢ C18 (L1) column with dimensions 4.6 x 100 mm, 3.5 Î¼m was used. 

To illustrate the flexibility provided by the Waters ACQUITY UPLC H-Class System, the USP method for galantamine hydrobromide and related substances was performed on an HPLC instrument (Figure 1). USP system suitability requirements for the related substances assay specify USP tailing of galantamine NMT 2.0 and a resolution of galantamine and 6-alphagalantamine NLT 4.5. When utilizing the same HPLC column on each instrument, the ACQUITY UPLC Columns Calculator (Figure 2) can be used to calculate the differences in the instrument dwell volume. The resulting data yielded no compromise in chromatographic integrity during the translation of the method for use on a UPLC instrument of less dwell volume (Figure 3).  

Figure 2. Example using the ACQUITY UPLC Columns Calculator for HPLC methodology translation, compensating for differences in dwell volume between two different LC system configurations. In red, the dwell volumes of the original and target instrumentation are entered. Once calculated, the gradient table is adjusted to compensate for the instrument differences. The same HPLC column was used for the new HPLC gradient performed on an ACQUITY UPLC H-Class System. 

Differences between instrumentation dwell volumes can be easily accounted for with calculated adjustments to the gradient table with the ACQUITY UPLC Columns Calculator. However, the challenging method translations of original methodology to a target methodology reside with differences in column stationary phase selectivity.

Pharmaceutical research and development (R&D) organizations were early adopters who recognized the many benefits of UltraPerformance LC? (UPLC?) Technology including resolution, sensitivity, throughput, and productivity as compared to HPLC.

Implementation of Methods Translation between Liquid Chromatography Instrumentation

Wide selectivity range for method development

Superior peak shape for basic compounds due to improved loading capacity in low-ionic strength mobile phases commonly used for LC/MS (i.e., formic acid and acetic acid)

ACQUITY UPLC®  CSH™ and XSelect™ HPLC columns  

ACQUITY UPLC and ACQUITY UPLC H-Class systems 

Method development, LC/MS, ACQUITY UPLC CSH, XSelect, selectivity, pH switching, superficially porous particles

Despite recent advances in column technology and stationary phase chemistry, challenges still remain for routine method development and LC/MS applications. This application note discusses how Charged Surface Hybrid (CSH) Technology provides solutions to the major problems still encountered with many commercially-available stationary phases, while providing a wide selectivity range for method development and general purpose use.  

The reproducible addition of a low-level positive charge allows ACQUITY UPLC CSH and XSelect HPLC columns to provide different selectivity and improved performance under challenging chromatographic conditions. As a result, these columns give superior peak shape for basic compounds due to improved loading capacity under low-ionic strength conditions, as well as rapid re-equilibration after changing mobile phase pH. The practical use of columns built on CSH Technology will be demonstrated using commercial drug substances under routine use conditions. The advantages of using these columns for impurity profiling in stability indicating methods and/or forced degradation studies will be demonstrated using typical LC/MS conditions.  

Acid/base/neutral (ABN) mix: flavone (30 μg/mL), octanophenone (100 μg/mL), imipramine (100 μg/mL), amitriptyline (100 μg/mL), 1-pyrenesulfonic acid (50 μg/mL), diclofenac (100 μg/mL), and fenoprofen (100 μg/mL) in 75:25 H

pH switching mix: 300 μg/mL metoprolol and 100 μg/mL of amitriptyline in 95/5 H

Base/neutral (BN) mix: Mixtures containing 2 to 200 μg/mL each of quetiapine and propiophenone were prepared in 50:50 CH

Impurity mix: 500 μg/mL of imipramine spiked with 0.5 μg/mL of amitriptyline (0.1% impurity). Prepared in H

All LC conditions are specified in the figure captions.  

Resolution of any chromatographic separation depends on the efficiency, retentitvity, and selectivity of the column and conditions being used. Efficiency can be increased mechanically by using smaller particle columns and instrumentation with lower dispersion. Retentivity and selectivity can be altered chemically through the use of different base particles, ligand coverage, and ligand type for the packing material. Therefore, during method development experiments, exploring a variety of stationary phases with different selectivity provides the best chance to separate compounds in a mixture. However, quantitating selectivity changes across a range of stationary phases presents a challenge due to the high dependency on the conditions and analytes being used.  

In order to quantitate selectivity differences, we employed a previously described method

  that compares the square of the correlation coefficient (R

) to calculate the selectivity difference (S-value) between two sets of conditions (i.e., two columns, high and low pH mobile phase, organic solvent, etc.). The S-value will be 0 for two conditions that give identical selectivity, and 100 for completely orthogonal selectivity.  

This method was used to compare the selectivity of ACQUITY UPLC CSH Columns under the same set of conditions with a diverse set of compounds (Figure 1). The ACQUITY UPLC BEH C

 Column was included in the comparison as a benchmark. The calculated S-values are shown to illustrate the differences in selectivity between the four stationary phases.  

Figure 1. Analysis of the ABN mix on ACQUITY UPLC CSH and ACQUITY UPLC BEH columns under low pH conditions. The ACQUITY UPLC System was equipped with a PDA detector and four-position column manager. Mobile phase A was 0.1% formic acid in water, and mobile phase B was 100% acetonitrile. The gradient was from 5 to 95% B in 5 minutes. Flow rate was 0.5 mL/min. UV detection at 254 nm. Injection volume was 2 Î¼L. Column temperature was 30 Â°C. Column dimensions were 2.1 x 50 mm, 1.7 Î¼m.			

The first observation is that the application of a low-level surface charge on the CSH C

 column results in selectivity differences compared to the BEH C

 column for the ionized analytes. Secondly, the three ACQUITY UPLC CSH Columns have very different selectivity from each other, which makes them ideal choices for incorporation into a column selection protocol. Finally, this low-level charge vastly improves the peak shape for the basic compounds in low-ionic strength mobile phases, which will be discussed in more detail in the following sections.  

Chromatographic selectivity can also be manipulated through the use of different mobile phase pH and elution solvent. This further enhances the capability of the columns to resolve multiple peaks in a sample. Finding stationary phases that are chemically stable above a mobile phase pH of 8 is challenging due to the pH limitation for high purity silica phases. 

 and ACQUITY UPLC CSH Phenyl-Hexyl columns are stable at elevated pH (up to pH 11) and can thus be used with a wide range of separation conditions to yield dramatically different selectivity (Figure 2). Changing the pH of the mobile phase affects only the ionizable compounds, while the neutral compounds do not change retention time. Changing the elution solvent from acetonitrile to methanol results in longer elution times for all compounds as expected, and can change the selectivity of the separation, as is shown in Figure 2 at high pH.  

Figure 2. Analysis of the ABN mix on ACQUITY UPLC CSH C18 using different mobile phase pH and elution solvent. The ACQUITY UPLC System was equipped with a PDA detector and four-position column manager. The low pH mobile phase was 0.1% formic acid in water, and the high pH mobile phase was 0.1% ammonium hydroxide in water. Mobile phase B was either 100% acetonitrile or 100% methanol. The gradient was from 5 to 95% B in 5 minutes. Flow rate was 0.5 mL/min. UV detection at 254 nm. Injection volume was 2 Î¼L. Column temperature was 30 Â°C. Column dimensions were 2.1 x 50 mm, 1.7 Î¼m.			

Performance in low-ionic strength mobile phases

Method development protocols that rely on using the same columns at both high and low pH must be flushed and re-equilibrated prior to use between different pH conditions. With MS-compatible mobile phases such as formic acid and acetic acid, changes in both retention and selectivity are observed for ionized analytes

  at low pH after the same column has been exposed to high pH. This is likely due to a slow surface equilibration on high purity packing materials, which essentially have very little surface charge at low pH. Thus, changes in the surface charge (e.g., caused by switching from low to high pH and back again) can cause drastic shifts in the chromatographic performance, especially in low-ionic strength mobile phases. 

Since ACQUITY UPLC CSH and XSelect HPLC columns already have a fixed low-level charge on the surface, re-equilibration of the stationary phase is very rapid, and no changes in retention and selectivity are observed when switching between low and high pH mobile phases on the same column (Figure 3). On other materials, such as the hybrid coated silica column shown in Figure 3, there is not only a shift in retention time for ionized bases, but also a loss in peak shape, which affects the ability to reproducibly integrate these compounds.  

Figure 3. Analysis of the pH switching mix on XSelect CSH C18 and a silica C18 column with a hybrid coating in formic acid conditions before and after exposure of each column to 119 injections (~6 hours) at high pH. The ACQUITY UPLC System was equipped with a PDA detector and four-position column manager. The low pH mobile phase was 0.1% formic acid in water, and the high pH mobile phase was 10 mM ammonium bicarbonate, pH 10. Mobile phase B was 100% acetonitrile. The gradient was from 5 to 95% B in 2.5 minutes. Flow rate was 0.8 mL/min. UV detection at 260 nm. Injection volume was 2 Î¼L. Column temperature was 30 Â°C. Column dimensions were 2.1 x 50 mm.			

This degradation in performance is not due to a deterioration of the stationary phase or loss of bonded phase, since the retention and peak shape of neutral analytes are not affected.

  It is due to the fact that many commercially-available, high-purity stationary phases take an extremely long time to re-equilibrate in low-ionic strength mobile phases like formic acid after exposure to higher pH mobile phases (pH ≥7). 

Another advantage of CSH columns is the ability to load larger amounts of basic compounds without deterioration in peak shape, which has many advantages for analytical and preparative development.  

Many modern stationary phases suffer from poor chromatographic performance for basic compounds at low pH.

  The performance is even worse for bases analyzed using low-ionic strength mobile phases such as formic acid. While the exact cause for this behavior is not completely known, it is thought that variations in the surface charge of high purity chromatographic materials can have a large impact on the chromatography, so much so that efficiency and peak capacity for bases at low pH can decrease by more than 50%, even at fairly low analytical loads.  

Despite recent advances in column technology and stationary phase chemistry, challenges still remain for routine method development and LC/MS applications.

Practical Applications of Charged Surface Hybrid (CSH) Technology

This application note summarizes a more comprehensive piece on the efficiencies and productivities to be gained by use of Empower™ 2 Method Validation Manager Software. 

Method development, method validation, validation protocols, sample lists, compliance, data security

Method validation is a demanding activity. It requires a large investment in personnel, materials, instruments, supervision, and, most of all, time. Some of the more time-consuming aspects of validation involve the creation of validation protocols and sample lists, tracking of the workflow from protocol to final reporting, the performance of calculations, and the intense need to organize and manage raw and processed data. The potential for errors in the many steps of the validation process is large and the time delays when errors occur can be costly. 

Waters Empower 2 Method Validation Manager (MVM) Software, used in partnership with the ACQUITY UPLC System, can dramatically address these time-consuming elements of analytical method validation. MVM is designed to streamline the setup, execution, calculation, and reporting of a method validation. It provides easy data tracking and complete organization of validation data and results monitored by the built-in oversight of automated error checking.  

MVM is business-critical software that can significantly reduce the time and costs required to perform chromatographic method validation. Because MVM allows the entire chromatographic method validation process to be efficiently performed within Empower 2, fewer software applications need be deployed, validated, and maintained. Software training and support is also minimized. When less software is required, the software that is business-essential can be deployed more quickly and efficiently.  

In addition, Method Validation Manager allows you to be fully compliant with governmental regulations by providing data security, a full set of user privileges, audit trails, and automatic data documentation; providing you with the necessary information and complete data traceability required for final reports and to pass audits and data reviews. 

To illustrate the straightforward operation and comprehensive functionality of MVM, the major steps involved in the workflow for a basic assay validation of the drug product acetaminophen are summarized in this work.  

Here we illustrate the utility of MVM software in the validation of a method for the analysis of acetaminophen. The method system suitability criteria for the analysis of acetaminophen are listed in Table 1. The full method is described in the extended version of this Waters Application Note (2007; 720002401en), available on the Waters website.  

Acetaminophen analysis with the ACQUITY UPLC System and an ACQUITY UPLC BEH C

, 1.7 μm, 2.1 x 50 mm Column, is shown in Figure 1.  

The elements of the written validation protocol for this method were easily transferred into the validation protocol method template of MVM (Figure 2).  

Figure 2. The written protocol can be easily transferred to Empower MVM. 

The following validation tests were performed in this study:  

Intermediate precision (different analyst)

(See the extended version of this piece for all data examples.)

Individual tests and their associated acceptance criteria were configured. 

For the accuracy test configuration, three specified concentration levels with three individual preparations were injected one time. Levels were entered in the levels table. MVM then ensures that samples submitted for accuracy analysis match the defined parameters. For the acceptance criteria for the accuracy test, percent recovery was the test result of interest. The acceptance range for the test is indicated as 95 to 105%. 

Complete sample set methods were constructed and then saved as templates within the validation protocol method, as shown in Figure 3.  

Figure 4. System precision parameters. In this protocol, the % RSD of peak area must be no more than 2%. 

Since the validation protocol called for method suitability parameters to be met by each analysis, system precision requirements were also configured, as shown in Figure 4.   

During the process of test configuration and sample set method construction, errors were automatically caught by MVM, illustrated by a red X in the validation protocol window. Using the update status button and responding to error messages from the message center effectively guides all troubleshooting activity. Thus MVM ensures that all sample set methods are consistent with their respective test configurations.  

The validation protocol method was saved within a validation template project. Next, a validation working project was started and a new study was initiated based on the validation protocol method template. 

Figure 5. The green check mark in this validation manager window indicates that the sample set method is consistent with the user-configured test criteria.

The validation manager window lists the test configurations and acceptance criteria for the validation study. Additional functionality includes indicators that show test status and required approval (Figure 5).   

Since complete sample set methods are contained in the validation protocol method, the study can now be executed. Standards and samples were prepared then analyzed on the ACQUITY UPLC System as the previously established sample set methods.

Figure 6. Repeatability, or intra-assay precision, test result in the validation result review window. 

The analytical method for acetaminophen was evaluated for accuracy, repeatability, intermediate precision, linearity, robustness and stability. All parameters met their predefined acceptance criteria. Examples of the accuracy and linearity screens from MVM are shown in Figures 6 through 9.  

Figure 7. Accuracy validation result review window.

Figure 9. The consolidated results of two separate sample sets are presented in the stability validation test result review window.

Figure 8. Linearity result shown in the validation result review window. 

Method validation is a demanding activity. It requires a large investment in personnel, materials, instruments, supervision, and, most of all, time.

Special Issues-10-15-2010 | LCGC Supplements

A QbD with Design-of-Experiments Approach to the Development of a Chromatographic Method for the Separation of Impurities in Vancomycin

Implementation of Methods Translation between Liquid Chromatography Instrumentation

Practical Applications of Charged Surface Hybrid (CSH) Technology

Empower 2 Method Validation Manager Software: A Tool for Rapid Method Validation