药品中色谱快速分离杂质及裂解物检测方案

AmericanPhartnaccuticalRlevicw Isolation and Identification of Process Related Impurities andDegradation Products from Pharmaceutical Drug Candidates， Part I By Karen M. Alsante， Todd D. Hatajik， Linda L. Lohr， and Thomas R. Sharp.Pfizer Global Research & Development Division， Analytical Research and DevelopmentDepartment—Groton， Connecticut The objective of this two-part review article is to provide guidance for isolating and identifyingprocess related impurities and degradation products from pharmaceutical drug candidates.The identification of degradation products can provide an understanding of impurity formationand define degradation mechanisms. If the identification process is performed at an earlystage of drug development， there is adequate time for improvements in the drug substanceprocess and drug product formulation to prevent these impurities and degradants long beforethe filing stage. Impurity and degradant structure elucidation is a collaborative effort involvingthe analytical chemist， process chemist and/or formulator as well as the degradation， massspectrometry and NMR experts. The process described in this two-part article uses adesigned approach for the impurity and/or degradant identification， which focuses onefficiency so that the success of data collection is maximized. There are a number of activitiesother than collecting experimental data， even though the experiments are central to theprocess. PartI of this article describes a process for isolating unknown impurities anddegradants， while Part II will illustrate the role of mass spectrometry and NMR in theidentification process. The Process The process of identification of impurities and/or degradants begins early in drugdevelopment. Early brainstorming sessions should involve the analytical chemist， the processchemist， the formulator and the degradation chemist， as well as the mass spectrometry andNMR experts. It is imperative to involve all that are familiar with the project of interest. Thegroup meets to assess the timelines for completion and to gather all pertinent information.This initial planning and discussion effort can save significant time in the experimental stage.A few questions that need to be answered at this early stage are： Is this an impurity ordegradant problem? At what level is the impurity/degradant present? Is it a process relatedimpurity (PRI)， and if so， at what step of the process is it formed? Is it a degradant， and if so，under what degradation condition is it formed? Are enriched samples with the unknownimpurity/degradant available? By gathering all relevant information， the most efficient methodof isolation and identification can be selected. The first step of the process is to determine at what level the unknown is present. Accordingto the ICH Guidelines on Impurities in New Drug Substances1： The studies conducted to characterize the structure of actual impurities present in the newdrug substance at a level greater than 0.1% (depending on the daily dose， calculated usingthe response factor of the drug substance) should be described. Note that all specifiedimpurities at a level greater than the identification threshold in batches manufactured by theproposed commercial process should be identified. Degradation products observed in stabilitystudies at recommended storage conditions should be similarly identified. When theidentification of an impurity is not feasible， a summary of the laboratory studies demonstratingthe unsuccessful effort should be included in the application. According to the ICH Guidelines on Impurities in New Drug Products2： Degradation products observed in stability studies conducted at recommended storageconditions should be identified when present at a level greater than the identificationthresholds (1% for a maximum daily dose of<1 mg to 0.1% for a maximum daily dose of>2g). Identification of impurities below the 0.1% level is generally not considered to benecessary unless the potential impurities are expected to be unusually potent or toxic1.Therefore， it is imperative to determine the level of the unknown impurity and/or degradantearly in the process. If the unknown is below the 0.1% threshold， then a discussion will needto take place among the project team members in order to determine if isolation and identification are necessary. However， if the unknown is at or above the 0.1% limit， then effortshould be put forth for identification. Once a decision has been made to identify an unknown， the next logical step is to evaluate allknown process related impurities， precursors， intermediates， and degradation products. Byobserving the relative retention times (HPLC) of all known process related impurities，precursors and intermediates (if available)， one can quickly determine whether or not theimpurity of interest is truly unknown. If the relative retention time of the unknown impuritymatches that of a standard， then the unknown can be identified using HPLC with ultra-violet(UV) photodiode array as well as mass spectrometry (MS) detection. The identity is confirmedby correlating the retention time， UV spectra and mass spectra of the unknown impurity withthat of the standard. Identifying an unknown by using a standard， as described in the above paragraph， is a quickand easy process. However， what happens when the relative retention time of an unknowndoes not match that of a standard? The next step is to obtain molecular mass andfragmentation data via HPLC-MS. It is essential to determine the molecular mass of theunknown. Not only does the molecular mass help in the identification of the unknown， but italso enables one to track the correct peak by HPLC if isolation becomes necessary. In orderto run LC-MS， a mass spectrometry compatible HPLC method must be available. The mobilephase should contain volatile buffers that are HPLC-MS-compatible. (Note： A discussion ofmass spectrometry compatible mobile phases will be discussed in partIl). If such a method isnot available， then one must be developed， which adds time to the identification timeframe. If the mass spectrometry data evaluation yields sufficient structural information， thiseliminates the need to isolate the impurity in question. If standards of the proposed structuresare available， they can be correlated with the unknown as previously described. If standardsare not available， which is usually the case， the proposed structures can be discussed withthe project team. The project team can then decide if the information is suitable for theirneeds， or if isolation is required. An alternative to isolation is small-scale synthesis. If possible structures have been proposedfrom the mass spectrometry data， one can study the process chemistry and determine atwhich step of the process the impurity and/or degradant is most likely to be formed. Byknowing the process chemistry， the feasibility of the proposed structures can be evaluated.Proposed structures can then be synthesized if a reasonable synthesis is available. It iseasier to synthesize and identify the unknown if the chemistry works quickly (i.e. onestep/straight-forward chemistry). If small-scale synthesis is chosen， the synthesis must be themost efficient route. At this stage of the process， it is frequently necessary to isolate and characterize theunknown. One of the most important factors to consider when approaching an isolationexperiment is the sample origin. It is vital to determine whether the unknown is an impurityand/or degradant， and to locate a sample that contains an enriched quantity of the unknown.Isolating low level impurities can prove to be very cumbersome and time consuming.Therefore， the ultimate goal is to find a sample that contains an enriched quantity of theunknown. Two great resources of enriched samples are retained mother liquor samples andpurposeful degradation/stability samples. If the unknown is a drug substance degradant， thenthe degradation reaction can be scaled-up to generate a large quantity of the unknown. If it isa drug product degradant， then effort should be put forth to form the degradant in the drugsubstance so that extraction from the excipients is not required. Whenever enriched samplesare not available， the unknown must be isolated from the bulk drug substance or drugproduct. A number of methods can be used for isolating impurities and/or degradants. Three of themost utilized techniques are thin-layer chromatography1y ((TLC)， flash chromatography (columnchromatography)， and preparative high performance liquid chromatography (HPLC). Theactual technique used depends upon the nature of the impurity and/or degradant， includingthe amount present in the original material from which it must be isolated. A good startingpoint is to assess the separation that is currently being used by the analytical chemist. Doesthe current methodology provide optimum resolution of the impurity/degradant from the main band and other impurities， and if so， is that method by TLC or HPLC? This is a key factor indetermining which technique to utilize. Each of the three techniques will be discussedseparately. Isolation Techniques Thin-Layer Chromatography Thin-Layer Chromatography (TLC) is a good technique to use when normal phase solventsprovide optimum separation. Typical thin-layer separations are performed on glass plates thatare coated with a thin layer of stationary phase. The stationary phases used in TLCencompass all modes of chromatography including adsorption， normal and reverse phase， ionexchange， and size-exclusion chromatography. The equipment required is simple andinexpensive. It is an ideal technique for the isolation of compounds because of its simplicity.However， in order for TLC to be successful， the impurity and/or degradant should be at orabove the 1% level. Anything below this level is very difficult to isolate on a TLC plate due tohigher detection limits. The steps involved in preparative TLC are： (1) application of the sample onto the plates， (2)development of the plates， (3) detection and location of the compound of interest， and (4)extraction of the compound of interest. Detection is usually by ultraviolet light. When theseparated compound of interest is located on the plate， the band is scraped and the impurityis extracted from the stationary phase (i.e. silica gel) with an appropriate solvent. Theextracted material is filtered or centrifuged， and the solvent collected is evaporated to yieldthe isolated material. It is essential to remove silica gel and other interferences that mayinhibit the identification of the compound. The isolated material is then submitted for LC-MSanalysis. One of the main disadvantages of using TLC for preparative isolations is that limited amountsof material can be isolated from the plates. Using preparative TLC plates can circumvent thisproblem. Preparative plates contain thicker films of stationary phase， thus allowing largeramounts of sample to be applied. Even in cases where preparative plates are used， chancesare good that not enough material can be isolated to obtain traditional NMR analysis(including 1H and 13C NMR). Therefore， TLC is most useful when an impurity and/ordegradant is identifiable by LC-MS.In cases where NMR analysis is essential foridentification， flash chromatography and/or preparative HPLC are more suitable techniques. Flash Chromatography (Column Chromatography) When an existing normal phase TLC method provides adequate resolution of theimpurity/degradant to be isolated， then flash chromatography can be a useful technique.Flash chromatography is a simple absorption chromatography technique for the routinepurification of organic compounds. It allows for separations of samples weighing 0.01-10.0 gin 10 to 15 minutes. Flash chromatography is a rapid， inexpensive and easily performedtechnique with a large sample capacity (approximately 5 times the load of reverse phasepacking materials). Normal phase flash chromatography is ideal when the sample is soluble innonpolar or moderately polar solvents such as hexane， chloroform， and dichloromethane.These volatile solvents allow easier concentration of impurities and degradants. Beforechoosing flash chromatography as the separation technique， use chemistry knowledge toassess the potential stability of the isolated product prior to isolation in order to determine ifspecial collection conditions are necessary. For example， collect thermally unstable productsin chilled flasks. The first step in developing a flash chromatographic separation is to determine the optimumsolvent composition by analytical TLC. A solvent system is chosen that provides goodresolution and moves the desired impurity and/or degradant to Rf= 0.354. TLC can provide aguide to suitable solvent systems， but development work may be needed in order to optimizethe separation when it is scaled-up to the larger I.D. columns used in flash chromatography. Asuitable column is selected based upon the resolution of the impurity and/or degradant andthe amount of sample to be purified. The amount of sample that can be purified on a columnis dependent upon the resolution of the impurity and/or degradant， and it is proportional to the cross sectional area. If less resolution is required to separate a key degradant or impuritysample load can be significantly increased. Typically， each run on a flash column canchromatograph gram quantities of material depending on the column size. The column load istypically much higher than that of reverse phase chromatography. The columns are generally packed with silica gel. In order for the separation to be successful，the size of the silica gel should be 40-63 mm4. A concentrated solution of the sample isprepared. The sample solution is applied at the top of the column， and the walls of the columnare washed with a few milliliters of eluent. Solvent is added to the column， and air pressure isapplied at a flow rate of 2 inches/minute to rapidly elute the desired impurity and/ordegradant. Separation is based upon the differential interactions between the solutemolecules and the adsorbent surface of the silica gel. Fractions are continuously collectedand monitored by chromatographic techniques (HPLC with UV detection， GC， or TLC). Thefractions containing the compound of interest are combined and evaporated to dryness. Theisolated material is cleaned (post-isolation cleanup， such as small scale column or analyticalHPLC re-injection， is essential) and submitted for LC-MS and NMR analysis. Flash chromatography is a fast and inexpensive technique for isolations requiring onlymoderate resolution4. Typically， compounds having Rf > 0.15 can be cleanly separated usingthis technique， and separations at Rf @ 0.10 are possible. In cases where high resolution isrequired， flash chromatography can be used as a preliminary purification step. It can be agood method of concentrating and partially purifying complex mixtures， thus making the finalisolation much easier by preparative HPLC. For example， the main band can be isolated fromthe rest of the mixture； therefore， the impurity mixture needing optimum resolution can beinjected with higher load on HPLC. In addition， automated flash chromatographic systemswith UV detectors and fraction collectors are available that further simplify the isolationprocess. Preparative HPLC Preparative chromatography is the process of using liquid chromatography to isolate asufficient amount of material for other experimental or functional purposes. This sectiondescribes the use of preparative HPLC in isolating tens of milligrams of pure unknowncompound(s) for the purpose of structure elucidation by spectroscopic techniques， which isoften referred to as semi-preparative HPLC. This section will focus primarily on preparativeHPLC methods with the following parameters： Column Length： 15-50 cm Column Width： 10-40 mm Particle Size： 5-60 mm Column Load： 10-1000 mg Flow Rate： 5-100 mL/min The scale of preparative HPLC is normally larger than that of conventional HPLC. Therefore，a practical starting point is to develop an analytical separation that optimizes the isolationconditions. Optimization of the analytical method implies seeking conditions which combinemaximum resolution of the peak of interest and minimum elution time， under the restriction ofa limited pressure drop. The optimized conditions determine the column， mobile phase， flowrate and sample loading capacity for the particular column. The conditions may be eithernormal phase or reversephase. The mobile phase should be chosen carefully to avoid saltcomplexation with the compound to be isolated. Volatile acid salts such as trifluoroacetic acid，formic acid and acetic acid are acceptable mobile phase additives， and the ammoniumcounter-ion is preferred for pH adjustment to any of these acids. Once the analytical scale method conditions are optimized， the next step is to choose acolumn and scale-up the analytical HPLC parameters so that preparative chromatographycan be performed and the unknown compound(s) isolated for identification by MS and NMR.For ease of transition， a preparative column consisting of the same packing material andparticle size should be chosen. The column is the most important component of the process.The column determines the amount of material that can be loaded onto the column for thedesired purity and recovery. An important step in the scale-up procedure is determining the maximum load on the analytical column. The maximum analytical load is essential indetermining the loading capacity of the preparative column. When an appropriate column ischosen， the analytical isolation can be scaled up using Equation 1： The scale-up factor is used to predict the loading capacity and flow rate for the preparativecolumn. For example， if a separation optimized on a 4.6 mm x 150 mm column was scaled toa 20 mm x 300 mm column， the scale factor for the sample load would be 38. Thus， the scale-up factor multiplied by the maximum analytical load estimates how much material can beloaded on the preparative column (see Equation 2). To maintain the same resolution when scaling-up a method， the flow rate also needs to bescaled proportionally. The preparative flow rate can be estimated by using the scale-up factor.For this estimation， the scale-up factor is multiplied by the analytical flow rate to estimate thepreparative flow rate (see Equation3). YMC Inc. has simplified the process of scaling-up an analytical separation for preparativeisolation by developing matched R&D column sets6. Each R&D column set contains ananalytical methods development column and a preparative isolation column packed from thesame lot of packing material. This provides assurance that any separation developed on theanalytical column will scale-up directly on the matched preparative column without furthermethod modification. The use of these column sets eliminates potential selectivity differencescaused by different types of silica and different particle and pore size packings by providingmatched columns. When the preparative method has been optimized， injections are made and the compound ofinterest is typically collected using a fraction collector. The fractions are pooled together in acollection vessel. The stability of the isolated product should be assessed prior to isolation inorder to determine if special collection conditions are required. In addition， be sure to useclean glassware since contamination can occur if the glassware has not been thoroughlycleaned. Also， carefully select the tubing used for collection of the unknown. For example，avoid using Tygon tubing because it contains phthalates， which can contaminate the isolatedcompound. The isolated product is concentrated using conventional sample concentrationtechniques (i.e. distillation at normal or reduced pressure， precipitation， freeze-drying， solventextraction， and membrane filtration). Rotary evaporation and flash distillation are the two mostcommonly used techniques to recover isolated products from the mobile phase. After theproduct has been recovered， it should be dried under high vacuum to remove all solvents. Ananalytical clean up of the isolated sample is critical prior to NMR analysis. A clean sampleimproves the purity and quality of NMR data. As was mentioned earlier， volatile acid saltssuch as trifluoroacetic acid， formic acid and acetic acid are often used as mobile phaseadditives， which may cause salt formation if the pH of the mobile phase is adjusted (i.e. usingammonium hydroxide). In addition， mobile phase solvents may also contain low levelimpurities that become enriched during the concentration process. It is essential to removeany salts and/or impurities from the isolated product. A simple purification method for theisolated product is to re-inject it onto the preparative column using a mobile phase without anyadditives or pH adjustments. By utilizing gradient elution， salts can be removed byincorporating an aqueous rinse at the beginning of the run， and then the organic solvent canbe ramped to elute the desired product. Thus， the isolated peak is purified. Solid-phaseextraction also offers great potential in purifying the isolated product because of itsuniversality7. Additionally， washing the isolated sample with deuterated solvent several timesalso helps to prepare the sample for NMR experimentation. Once the sample has beenpurified， it is submitted for LC-MS and NMR analysis. As was mentioned earlier， isolation of low-level impurities and/or degradants can be quitelabor intensive. Consider a 0.1% level impurity present in a drug substance bulk lot. Based ontraditional NMR experiments， 5 mg of the impurity would be needed to obtain structuralconfirmation. To isolate 5 mg of the impurity from the bulk， a minimum of 5 g of bulk drugsubstance would be needed， assuming 100% recovery. Because actual recoveries aregenerally closer to 50% for low level (0.1% range) isolations， 10 g of bulk drug substancewould generally be requested. In addition to requiring significant bulk material， the timeframe to complete the isolation isconsiderable. If the maximum analytical load for a 4.6 mm x 150 mm column has been determined to be 5 mg of the parent drug substance， assuming the isolation will be performedusing semi-preparative chromatography (20 mm x 300 mm column)， approximately 190 mg ofsample can be loaded onto the preparative column. For a 0.1% level unknown， this translatesto 190 mg of unknown injected onto the preparative column. Therefore， a total of 27 injectionsare required. If the assay time were estimated to be one hour， it would take at least 27 hoursto perform the injections needed to obtain 5 mg (once again assuming 100 % recovery). Thistimeframe does not include the time needed for method scale-up development， concentrationand solubility experiments， as well as mass spectrometryand NMR experimentation. On the other hand， if a sample was available that contained 10% of the unknown， only 1 gramof bulk would be needed and the estimated timeframe of the isolation would be drasticallyreduced. In this example above， 19 mg of unknown can be injected onto the preparativecolumn (assuming the maximum analytical load does not change and resolution is retainedwith the higher level impurity). Therefore， only one injection would be needed to obtain theamount necessary for NMR analysis， reducing the time to 1 hour. Most of the applications discussed in this section deal with semi-preparative HPLC columnsusing steel as well as radial compression columns. Radial compression uses radial pressureapplied to a flexible-wall column to lessen wall effects. Mobile phase has a tendency to flowslightly faster near the wall of the column because of decreased permeability. The solutemolecules that happen to be near the wall are carried along faster than the average of thesolute band， and consequently， band spreading results. Preparative scale radial compressionchromatography columns efficiencies close to those of analytical-scale columns when anadequate radial compression level is used8.Radial compression technology also helps lowerthe cost by substituting reusable column holders in place of expensive steel columns. Conclusion The process described in this article uses a designed approach for the isolation of unknownimpurities and degradants in pharmaceutical drug substances. This approach focuses onefficiency so that the success of data collection is maximized. The isolation of pure material iscrucial when trying to identify the structure of an unknown impurity/degradant. Once theunknown has been isolated， it can be submitted for structure elucidation using massspectrometry and NMR. Part Il of this article will describe the role of mass spectrometry andNMR in the identification process. References 1. International Conference on Harmonization，“Draft Revised Guidance on Impurities in NewDrug Substances，” Federal Register 65 (140)， 45085-45090 (2000). ( 2. International Conference on Harmonization，“Draft Revised Guidance on Impurities in NewDrug Products，" Federal Register 65 (139) 44791-44797 (2000). ) ( 3. Skoog， D . A . ， Principles of Instrumental Analysis， Third Edition ， Saunders CollegePublishing， Philadelphia， 1 985.4. S till ， W . C.； Kahn， M.； Mitra， A . Journal of Organic Chemistry1978， 43，2923. ) ( 5. Knox， J. H.； Pyper， H. M. Journal of Chromatography 1986， 363， 1-3 0 . ) ( 6. Waters Corporation. Waters Preparative L iquid Chromatography Catalog， 19 9 8. ) 7. Porsch， B. Journal of Chromatography 1994，658，179-194. ( 8. Carta， G.； Stringfield， W. B. Journal of Chromatography 1994， 658 ， 407-417. ) ( 9. Gervais， D.P.； Laughinghouse， W.S.； Carta， G . J ournal o f Chromatography 19 9 5， 70 8 ， 41- 53. )