8 METHOD DEVELOPMENT Rosario LoBrutto 8.1 INTRODUCTION The primary focus of this chapter in on general approaches and considerations toward development of high-performance liquid chromatography (HPLC) methods for separation of pharmaceutical compounds, which may be applied within the various functions in the drug development continuum. It is very important to understand the aim of analysis and the requirements for a par- ticular method to be developed. The aim of analysis of each HPLC method may vary for each developmental area in the drug development process and specific examples are given in Section 8.2. General method development considerations that apply to all reversed- phase methods are discussed in Section 8.3. These include properties of the analyte, detector, mobile phase, stationary phase, and gradient considerations. Building upon this knowledge, strategies for method development for target analytes in which the structure is known and not known are given as general guidelines. This material is reinforced with several method development case studies emphasizing the approaches used and the shortcomings that were encountered during the method development continuum. Also, a method development flow chart for gradient separations is provided in Section 8.5.6 (Figure 8-37) which can be used as an excellent starting point for the devel- opment of HPLC methods. Also, throughout this chapter we focus on analytical challenges a pharma- ceutical scientist encounters during method development; these include speed 347 HPLC for Pharmaceutical Scientists, Edited by Yuri Kazakevich and Rosario LoBrutto Copyright © 2007 by John Wiley & Sons, Inc. of separation, sample preparation, extraction issues, solution stability, detection sensitivity (effect of pH), and separations for structurally similar species. The final part of the chapter provides a refresher on “pK a from an analy- tical chemist’s perspective,” the drivers for choosing normal phase versus reversed phase as a separation mode for a particular analysis, and instru- ment/system consideration, and it concludes with a very interesting section on column testing within the framework of bonded phase stability (effect of pH and type of buffer) and probing column selectivity. 8.2 TYPES OF METHODS There are many HPLC methods that are developed during the process of drug development. The chromatographer needs to understand the aim of analysis in order to make judicious choices prior to the commencement of method development and the implications it may have on the final method that is developed.The following should be considered: method development time, the maximum run time for analysis, the number of samples expected per week, the complexity of the mixture, the structure of the main analyte (physico- chemical properties), possible degradation pathways (i.e., hydrolysis, oxida- tive, photolysis, dethydration, thermolytic, racemization), and whether the analyte or analytes are ionizable. In Section 8.2, the aim of analysis is emphasized especially for the API (active pharmaceutical ingredient) and the drug product. The workflows and the rationale at major decision points during synthetic processing steps where HPLC can be applied in process development are elaborated upon. For example, a fast method is needed to monitor reaction conversion of two components. However, a more complex method would be needed for stabil- ity-indicating purposes where multiple degradation products, synthetic by-products, and excipient peaks need to be resolved from the active phar- maceutical ingredient. 8.2.1 Key Raw Materials The purity of the starting materials needs to be determined in order to ensure the purity of the final compound and to avoid any undesired secondary reac- tions that may carry forward in the downstream chemistry. Starting materials can be classified as raw materials or key raw materials. The latter are raw mate- rials in which some part of the raw material structure is incorporated into the final structure of the drug substance. For the raw materials, usually an identi- fication test and concentration of the reagent suffices although the purity of these materials is sometimes also deemed as a necessity. However, for key raw materials the purity of this reagent must be known. The analysis of the key raw materials depends on the nature of the substance (i.e., volatility) and a 348 METHOD DEVELOPMENT variety of methods could be used such as GC and HPLC.To ensure the quality of the key raw materials, the level and type of impurities present in these mate- rials need to be determined and appropriate specifications need to be set. These can be determined by running a small-scale synthesis (i.e., front run) with the key raw material; and if an intermediate with an acceptable purity is obtained in the next step of the synthesis, the level and type of impurities present in the key raw materials can be considered acceptable. Note that the impurities may carry forward in the downstream chemistry and/or may react to form new synthetic by-products. If these synthetic by-products are carried forward to the final API, these impurities must be qualified in the appropri- ate toxicological studies if they are above a certain level (also, any potential genetoxic impurities must be identified and well-controlled with sensitive ana- lytical methods). Different lots from the same manufacturer and/or lots of key raw materials from different manufacturers are usually tested. In the following example, the importance of determining the quality of the key raw material is highlighted. Trimethoprim (TMP) (2-4-diamino-5-(3,4,5- trimethoxybenzyl)pyrimidine) is an antibacterial, folic acid antagonist that is usually used with sulfamethoxazole as a treatment for urinary tract infections. Shown in Figure 8-1 is one synthetic scheme of TMP [1]. The starting mater- ial that is used is 3,4,5-trimethoxy benzaldehyde (a key raw material). Depend- ing on the quality of this key raw material, the impurities from this starting material may further react in the subsequent synthetic steps (downstream chemistry) at the benzaldehyde functionality to produce undesired synthetic by-products. TMP manufactured by five different manufacturers from three countries of origin were analyzed by the same HPLC chromatographic method; two impurities, Impurity 1 and Impurity 2, were found in multiple lots of TMP, and the levels depended upon the country of origin (Figure 8-2). [2]. TYPES OF METHODS 349 Figure 8-1. Synthesis of TMP. (Reprinted from reference 1, with permission.) Impurity 1 and Impurity 2, were identified as 2,4-diamino-5-(-4-ethoxy- 3,5-dimethoxybenzyl) pyrimidine and 2,4-diamino-5-(3-bromo-4,5- dimethoxybenzyl) pyrimidine, respectively (structures shown in Figure 8-3), using LC/MS, MS/MS, and NMR. Both these synthetic by-products were presumed to come from the starting key raw material 3,4,5- trimethoxybenzaldehyde. This emphasizes that appropriate control of the starting material is needed. If the purity of the starting material is not controlled adequately, then recrys- tallization steps may be needed in later steps to remove synthetic by-products in the API that are above a qualified level which may result in decreased yield of the synthesis. Chromatographic analysis of the key raw material should be employed to determine the limits of the impurities in 3,4,5-trimethoxyben- 350 METHOD DEVELOPMENT Figure 8-2. HPLC profiles of TMP drug substance from various manufacturers using the same chromatographic method. (Reprinted from reference 2, with permission.) zaldehyde (individual and total) that would give acceptable levels of the result- ing downstream impurities in the API. This would include determining the purity of the key raw material with a defined method and relating the purity of the key raw material from different lots/manufacturers to the quality of the final drug substance. In general, the detection and identification of impurities present in API and formulations play an integral role in the drug development process and methods need to be developed to adequately resolve these species and quan- titate them. The International Conference on Harmonization (ICH) [3] has published a guidelines on impurities in new drug substances and new drug products (see Table 8-1). The acceptable limit of the impurities in drug sub- stances is dependent upon the maximum daily dose and the qualification threshold; however, lower thresholds are sometimes deemed necessary if the TYPES OF METHODS 351 Figure 8-3. Trimethoprim (API) and two potential synthetic by-products. (Reprinted from reference 2, with permission.) TABLE 8-1. Reporting Thresholds of Impurities [3] Maximum Reporting Identification Qualification Daily Dose a Threshold b,c Threshold c Threshold c ≤2 g/day 0.05% 0.10% or 1.0mg/day intake 0.15% or 1.0mg/day intake (whichever is lower) (whichever is lower) >2 g/day 0.03% 0.05% 0.05% a The amount of drug substance administered per day. b Higher reporting thresholds should be scientifically justified. c Lower thresholds can be appropriate if the impurity is unusually toxic. impurity is known to be unusually potent or to produce toxic or unexpected pharmacological effects through genetoxicity studies, general toxicity studies, and/or in-silico assessment. 8.2.2 Drug Substance (Active Pharmaceutical Ingredient) During process development of a new compound, many samples are gener- ated which include reaction mixtures from process monitoring, batch and waste layers from extractions (aqueous and/or organic), batch concentrates, mother liquors and supernatants, and isolated solid intermediates; in the event that PrepHPLC is used, column fractions are analyzed to determine the yield and purity of each fraction so they may be pooled. Impurities may be formed during the manufacture of the API, and these impurities may be related to the starting materials, process-related impurities, synthetic intermediates, or degradation products, and suitable methods need to be developed to control these processes. 8.2.2.1 Reaction Conversion. Methods that monitor reaction conversion should ensure the resolution of all solvents and in-process impurities from the reactants and the desired intermediate. The goal here is to monitor product A going to product B—in essence, measuring the disappearance of A or com- pletion of the reaction if reagent A is used in excess. Also, the concentration of product B may be needed as well as the purity of product B to control any undesired by-products. In Figure 8-4 the starting material is reacted in a one pot reaction to form the intermediate. At 30 minutes the reaction is only 30% complete. By further monitoring, it was determined that the reaction must be allowed to continue 352 METHOD DEVELOPMENT Figure 8-4. Reaction conversion of starting material to intermediate . for at least 100 minutes in order to reach completion (Figure 8-5). During this process optimization stage, the concentration of the reagents, the temperature of the reaction, mixing conditions, and other processing conditions are opti- mized and HPLC is used as a tool to monitor the reaction. It must be deter- mined how long the reaction should proceed in order to form the desired intermediate in good yield and for how long the intermediate is stable in solu- tion prior to going to the next step (hold point stability). These reaction monitoring analyses should be fast because the reaction time scales may be in the order of minutes to hours. In-line flow injection analysis or spectroscopic methods are sometimes used to monitor reactions (reaction conversion) that are on the minute time scale and for reactions that involve hazardous materials because by the time the samples are analyzed by an off-line chromatographic method, the reaction has gone to completion and/or undesired by-products may have been formed. However, off-line HPLC may still be needed to determine the purity of the desired intermedi- ate present in the reaction solvent before proceeding to the next step of the reaction, and fast HPLC methods can be employed. If the reaction conversion is along the time scale of an hour or greater, fast HPLC methods are used implementing nonporous stationary-phase materials, monolithic columns, and columns packed with sub-2 µm particles (uPLC, xPLC, fPLC) (more informa- tion on fast HPLC methods is provided in Chapter 17). It is advantageous to have short methods to analyze these reaction conversion samples. The in- process samples may have to be diluted with an appropriate solvent to quench the reaction and to have the desired components within the linear range of the chromatographic method.The diluent must be chosen such that it does not react with the components in the mixture. Also, all blank system suitability and standards samples should be run on the chromatographic system prior to injection of the reaction sample to ensure that the HPLC system is working properly and that no downtime in the reporting of the results to the process TYPES OF METHODS 353 Figure 8-5. Reaction monitoring of converting the starting material to intermediate. chemist and/process engineer. For the more conservative chromatographer, two HPLC systems can be set up in the event that an instrument breaks down to ensure no downtime in the reporting of the results. 8.2.2.2 Concentration Determination of In-Process Samples. The concen- tration of the unisolated desired product in solution at a particular interme- diate step may also need to be determined by HPLC. A data calculation sheet such as Excel with the response factors of the standards and the dilution factor of the sample could be incorporated in the data calculation sheet prior to injec- tion of reaction sample to facilitate the results reporting for the concentration of the intermediate in solution. Hence, only the area of the desired interme- diate in solution needs to be populated in the spreadsheet, and the concen- tration result then can be determined. The determined concentration of the intermediate in solution ensures adequate charging of the raw materials used in the further steps of the synthesis. Also, this intermediate in solution is sometimes further concentrated and the concentration is monitored until the desired concentration is obtained. A solvent switch step is sometimes per- formed, and the HPLC method must be able to selectively separate the reac- tion solvents (if they are UV active) from the desired intermediate and potential impurities that may be formed. These reaction solvents may include toluene, inhibited THF with cresol or BHT (if inhibited with BHT, this is very hydrophobic, so proper elution of this additive may be necessary), ethyl acetate, and so on. Sample preparation here is also important, and the appro- priate diluent must be determined to ensure solubility of all components and no reactivity with the sample analyte. Sometimes when extractions are performed to remove undesired by- products, the concentration of the desired product in the organic and aqueous layers are also determined.The concentration in the organic layers are deemed the most important (contains the desired product), and the concentrations in the aqueous layers are determined later to ensure mass balance and overall yield of the reaction. The aqueous layers are usually enriched with the unde- sired by-products and are good samples to use during the development of the HPLC method in the early stages of the synthetic development. The pH of these layers is usually checked as well to ensure that the proper amount of acid or base has been added to the reaction mixture either to quench the reac- tion or to drive the desired product into the organic layer. 8.2.2.3 Purity of Intermediates. Determining the purity of the desired product in the organic layers is important to ensure that an adequate number of aqueous washes removed the unwanted by-products. This organic layer may be carried forward to the next step without any further isolation. However, if the intermediate will be isolated, then the purity and weight percent of the isolated intermediate needs to be determined to ensure mass balance and determine overall yield of the reaction. The purity of the intermediates needs to be evaluated in order to determine if synthetic by-products generated in a 354 METHOD DEVELOPMENT prior step will react in the downstream chemistry or, if they do not react in further steps of the synthesis, to determine what their rejection will be in sub- sequent steps of the synthesis. Eventually, these methods may also have to be transferred to a manufacturing facility and will be used as critical process con- trols before proceeding to the next step of the synthesis. Methods should be able to separate the API from the synthetic process impurities including the penultimate intermediate and potential degradation products and any poten- tial solvents that may be observed, depending on the type of detection employed. The final method should be able to resolve all impurities from the active as well as from each other. The optimal goal is to have one method to resolve all isolated intermediates, penultimate products, synthetic by-products, and potential degradation products from the API. Eventually, a stability-indicating method needs to be validated and included in the IND (investigational drug substance), IMPD (investigational medicinal product dossier), and NDA (new drug application). Stability- indicating methods must be able to resolve potential degradation products from the active substance that may increase or form during the storage (certain temperature/certain humidity). There are generally two types of stability studies that are performed: long- term and accelerated.The purpose of the accelerated stability studies is to help set the shelf life at the recommended storage conditions and predict the amount of degradation that could be anticipated under long-term storage con- ditions. However, at the predicted shelf life timepoint (number of months or years) the sample stored at the recommended storage condition must be run to confirm the shelf-life prediction from the accelerated stability studies. Therefore the stability-indicating methods should be rugged and robust and meet all validation requirements that will allow for determining the purity of the active pharmaceutical ingredient throughout the duration of the support- ive stability studies. Method validation is discussed in detail in Chapter 9. 8.2.3 Drug Product Methods that are used during the development of a drug product formulation should be able to assess reacting or catalyzing excipients and any undesired reactions leading to degradation products. Methods should be able to sepa- rate the active pharmaceutical ingredient (API) from the drug product degra- dation products, excipients, excipient degradation products, and any synthetic impurities that are present in the API. These are usually performed during early phase development and are known as excipient compatibility studies. Both binary (API + excipient) and excipient mixtures + API are used to assess which excipient or formulation blend will lead to the most stable drug for- mulation. Moreover, other preformulation studies that are also performed include solubility studies and solution stability studies at various pH values (Figure 8-6). HPLC with UV detection is the most applicable technique to use TYPES OF METHODS 355 for these studies. Fast gradient methods on short columns are usually used in most cases for solubility determination. Some software programs such as ACD (advanced chemistry development software) can be used to estimate the sol- ubility as a function of pH and can be used as a starting point to estimate the appropriate dilution of the different solutions prepared at the different pH values. For example, in Figure 8-7 if the target concentration for the HPLC assay is 1 mg/mL for this basic compound, then the sample at pH 1 (8 mg/mL 356 METHOD DEVELOPMENT Figure 8-6. Solution pH stability. Figure 8-7. ACD prediction versus HPLC experimental results for the solubility of a basic compound. [...]... optimal dilution scheme for each sample at a particular pH Moreover, the predicted ACD values and the experimental values from pH 1–8 in this example were in very good agreement This is used only for estimation since sometimes the ACD prediction of the solubility for the ionized form of bases and acids shows greater deviation than for the solubility predicted for the neutral form of the molecules Reaction... upon storage at specified long-term storage conditions in the drug product Therefore, for the drug product formulations, as per the current ICH guidelines on impurities in new drug products [5], the reporting thresholds are 0.1% for drugs with maximum daily dose of ≤1 g/day and 0.05% for drugs with maximum daily dose of >1 g/day For more description on the reporting, identification, and qualification of impurities... reversed-phase and normal-phase HPLC method development in the pharmaceutical industry is carried out with UV detection In this section the practical use of UV detection will be discussed A wavelength for UV detection must be chosen so that an accurate mass balance may be determined Therefore, if area% normalization is to be used, then all the impurities and the active pharmaceutical ingredient must... carboxylic acid as a result of acid/base hydrolysis or degradation due to microenviromental pH of the formulation Therefore, in an eluent that has a high pH, the potential acidic impuritiy may be in its ionized form which may result in the elution of the potential degradation product with or even before the void volume If the target analyte is ionizable, the pKa of the analyte should be determined or... 363 compound is a prerequisite for any salt selection program Salt formation during a salt selection program provides a means of altering the physicochemical and resultant biological characteristics of a drug substance without modifying its chemical structure, and most compounds with a suitable acidic or basic functionality can potentially be transformed into its salt form The free acid/free base and... Guidance for the Industry: Q3B(R): Impurities in New Drug Products (shown in Table 8-2) Eventually, a stability-indicating method needs to be validated and to be included in the IND, IMPD, and NDA These methods should be rugged and robust and meet all validation requirements at each particular stage of drug development This method may be used also for content uniformity, and TABLE 8-2 Thresholds for Degradation... degradation products assuming a separate stability-indicating method has been established The methods for dissolution and content uniformity may offer the chromatographer some advantages in terms of high-throughput analysis by allowing chromatographer to use shorter methods employing shorter columns, and other fast HPLC approaches because the resolution of the impurities from each other may not be deemed necessary... Enantiomers have identical physical properties except for the direction of optical rotation Diastereomers are basically stereoisomers that are not enantiomers of each other A pair of enantiomers exists for all molecules containing a single chiral center and have the opposite configuration at each of the stereo centers The maximum number of stereoisomers for a compound with n stereo centers is 2n Diastereomers,... accomplished using achiral stationary phases Another alternative is the use of chiral columns for the separation of diastereomers in either the reversed-phase or normal-phase mode The use of achiral bonded phases without chiral additives, such as phenyl and alkyl bonded phases for the separation of diastereomeric pharmaceutical compounds, is acceptable Different selectivities can be obtained by employing... known information about the analyte such as its structure, physical and chemical properties, toxicity, purity, hygroscopitiy, solubility, and stability should be determined These data may be available from preformulation reports, early drug discovery sample screening reports, from the literature on similar compounds, or from past experience with similar compounds However, many times this information . development; these include speed 347 HPLC for Pharmaceutical Scientists, Edited by Yuri Kazakevich and Rosario LoBrutto Copyright © 2007 by John Wiley & Sons,. considerations toward development of high-performance liquid chromatography (HPLC) methods for separation of pharmaceutical compounds, which may be applied