KP IX General Information

Transcription

1 General Information

2

3 KP IX 1817 General Information 1. Capillary Electrophoresis Determination of Bulk and Tapped Densities Disinfection and Sterilization Methods Guideline for Setting Dissolution Specification of Oral Dosage Forms Guideline to Pharmaceutical Quality Control Using Near Infrared (NIR) Spectroscopy Guideline of Limits for Residual Solvents of Pharmaceuticals Guideline of Validation of Analytical Procedures for Pharmaceuticals Isoelectric focusing Particle Size Determination Powder Particle Density Determination Preservatives-Effectiveness Tests Specific Surface Area Determination Method Terminal Sterilization and Sterilization Indicators

4 1818 General Information 1. Capillary Electrophoresis Capillary electrophoresis is a physical method of analysis based on the migration, inside a capillary, of charged analytes dissolved in an electrolyte solution, under the influence of a direct-current electric field. The migration velocity of an analyte under an electric field of intensity E, is determined by the electrophoretic mobility of the analyte and the electroosmotic mobility of the buffer inside the capillary. The electrophoretic mobility of a solute (µ ep ) depends on the characteristics of the solute (electric charge, molecular size and shape) and those of the buffer in which the migration takes place (type and ionic strength of the electrolyte, ph, viscosity and additives). The electrophoretic velocity ( ep ) of a solute, assuming a spherical shape, is given by the following equation: ep ep q V E 6 r L q: effective charge of the solute, : viscosity of the electrolyte solution, r: Stoke's radius of the solute, V: applied voltage, L: total length of the capillary. When an electric field is applied through the capillary filled with buffer, a flow of solvent is generated inside the capillary, called electro-osmotic flow. The velocity of the electro-osmotic flow depends on the electro-osmotic mobility (µ eo ) which in turn depends on the charge density on the capillary internal wall and the buffer characteristics. The electro-osmotic velocity ( eo ) is given by the following equation: eo eo V E L ε: dielectric constant of the buffer, : zeta potential of the capillary surface. The velocity of the solute ( ) is given by: ep eo The electrophoretic mobility of the analyte and the electro-osmotic mobility may act in the same direction or in opposite directions, depending on the charge of the solute. In normal capillary electrophoresis, anions will migrate in the opposite direction to the electroosmotic flow and their velocities will be smaller than the electro-osmotic velocity. Cations will migrate in the same direction as the electro-osmotic flow and their velocities will be greater than the electro-osmotic velocity. Under conditions in which there is a fast electroosmotic velocity with respect to the electrophoretic velocity of the solutes, both cations and anions can be separated in the same run. The time (t) taken by the solute to migrate the distance (l) from the injection end of the capillary to the detection point (capillary effective length) is given by the following expression: t ep l eo l L ( ) V In general, uncoated fused-silica capillaries above ph 3 have negative charge due to ionized silanol groups in the inner wall. Consequently, the electroosmotic flow is from anode to cathode. The electroosmotic flow must remain constant from run to run if good reproducibility is to be obtained in the migration velocity of the solutes. For some applications, it may be necessary to reduce or suppress the electro-osmotic flow by modifying the inner wall of the capillary or by changing the concentration, composition and/or ph of the buffer solution. After the introduction of the sample into the capillary, each analyte ion of the sample migrates within the background electrolyte as an independent zone, according to its electrophoretic mobility. Zone dispersion, that is the spreading of each solute band, results from different phenomena. Under ideal conditions the sole contribution to the solute-zone broadening is molecular diffusion of the solute along the capillary (longitudinal diffusion). In this ideal case the efficiency of the zone, expressed as the number of theoretical plates (N), is given by the following equation: ep eo ( ep eo) V l N 2 D L D: molecular diffusion coefficient of the solute in the buffer. In practice, other phenomena such as heat dissipation, sample adsorption onto the capillary wall, mismatched conductivity between sample and buffer, length of the injection plug, detector cell size and unlevelled buffer reservoirs can also significantly contribute to band dispersion. Separation between two bands (expressed as the resolution, R S ) can be obtained by modifying the electrophoretic mobility of the analytes, the electroosmotic mobility induced in the capillary and by increasing the efficiency for the band of each analyte, according to the following equation: R S N ( 4 ( epb ep ) eo epa µ epa and µ epb : electrophoretic mobilities of the two analytes separated, )

5 KP X 1819 ep : mean electrophoretic mobility of the two 1 analytes ( ep ( epb epa)) 2 Apparatus An apparatus for capillary electrophoresis is composed of : (1) a high-voltage, controllable direct-current power supply, (2) two buffer reservoirs, held at the same level, containing the prescribed anodic and cathodic solutions, (3) two electrode assemblies (the cathode and the anode), immersed in the buffer reservoirs and connected to the power supply, (4) a separation capillary (usually made of fused-silica) which, when used with some specific types of detectors, has an optical viewing window aligned with the detector. The ends of the capillary are placed in the buffer reservoirs. The capillary is filled with the solution prescribed in the monograph, (5) a suitable injection system, (6) a detector able to monitor the amount of substances of interest passing through a segment of the separation capillary at a given time. It is usually based on absorption spectrophotometry (UV and visible) or fluorometry, but conductimetric, amperometric or mass spectrometric detection can be useful for specific applications. Indirect detection is an alternative method used to detect non UV-absorbing and nonfluorescent compounds, (7) a thermostatic system able to maintain a constant temperature inside the capillary is recommended to obtain a good separation reproducibility, (8) a recorder and a suitable integrator or a computer. The definition of the injection process and its automation are critical for precise quantitative analysis. Modes of injection include gravity, pressure or vacuum injection and electrokinetic injection. The amount of each sample component introduced electrokinetically depends on its electrophoretic mobility, leading to possible discrimination using this injection mode. Use the capillary, the buffer solutions, the preconditioning method, the sample solution and the migration conditions prescribed in the monograph of the considered substance. The employed electrolytic solution is filtered to remove particles and degassed to avoid bubble formation that could interfere with the detection system or interrupt the electrical contact in the capillary during the separation run. A rigorous rinsing procedure should be developed for each analytical method to achieve reproducible migration times of the solutes. 1. Capillary Zone Electrophoresis In capillary zone electrophoresis, analytes are separated in a capillary containing only buffer without any anti-convective medium. With this technique, separation takes place because the different components of the sample migrate as discrete bands with different velocities. The velocity of each band depends on the electrophoretic mobility of the solute and the electroosmotic flow in the capillary. Coated capillaries can be used to increase the separation capacity of those substances adsorbing on fused-silica surfaces. Using this mode of capillary electrophoresis, the analysis of both small (M r < 2000) and large molecules (2000 < M r <100000) can be accomplished. Due to the high efficiency achieved in free solution capillary electrophoresis, separation of molecules having only minute differences in their charge to mass ratio can be effected. This separation mode also allows the separation of chiral compounds by addition of chiral selectors to the separation buffer. Optimization Optimization of the separation is a complex process where several separation parameters can play a major role. The main factors to be considered in the development of separations are instrumental and electrolytic solution parameters. Instrumental parameters (1) Voltage A Joule heating plot is useful in optimizing the applied voltage and column temperature. Separation time is inversely proportional to applied voltage. However, an increase in the voltage used can cause excessive heat production, giving rise to temperature and, as a result thereof, viscosity gradients in the buffer inside the capillary. This effect causes band broadening and decreases resolution. (2) Polarity Electrode polarity can be normal (anode at the inlet and cathode at the outlet) and the electroosmotic flow will move toward the cathode. If the electrode polarity is reversed, the electro-osmotic flow is away from the outlet and only charged analytes with electro-osmotic mobilities greater than the electroosmotic flow will pass to the outlet. (3) Temperature The main effect of temperature is observed on buffer viscosity and electrical conductivity, and therefore on migration velocity. In some cases, an increase in capillary temperature can cause a conformational change in proteins, modifying their migration time and the efficiency of the separation. (4) Capillary The dimensions of the capillary (length and internal diameter) contribute to analysis time, efficiency of separations and load capacity. Increasing both effective length and total length can decrease the electric fields (working at constant voltage) which increases migration time. For a given buffer and electric field, heat dissipation, and hence sample bandbroadening, depend on the internal diameter of the capillary. The latter also affects the detection limit, depending on the sample volume injected and the detection system employed. Since the adsorption of the sample components on the capillary wall limits efficiency, methods to avoid these interactions should be considered in the development of a separation method. In the specific case of proteins, several strategies have been devised to avoid adsorption on the capillary wall. Some of these strategies (use of extreme ph and ad-

6 1820 General Information sorption of positively charged buffer additives) only require modification of the buffer composition to prevent protein adsorption. In other strategies, the internal wall of the capillary is coated with a polymer, covalently bonded to the silica, that prevents interaction between the proteins and the negatively charged silica surface. For this purpose, ready-to-use capillaries with coatings consisting of neutral-hydrophilic, cationic and anionic polymers are available. Electrolytic solution parameters (1) Buffer type and concentration Suitable buffers for capillary electrophoresis have an appropriate buffer capacity in the ph range of choice and low mobility to minimize current generation. Matching buffer-ion mobility to solute mobility, whenever possible, is important for minimizing band distortion. The type of sample solvent used is also important to achieve oncolumn sample focusing, which increases separation efficiency and improves detection. An increase in buffer concentration (for a given ph) decreases electroosmotic flow and solute velocity. (2) Buffer ph The ph of the buffer can affect separation by modifying the charge of the analyte or additives, and by changing the electro-osmotic flow. In protein and peptide separation, changing the ph of the buffer from above to below the isoelectric point (pi) changes the net charge of the solute from negative to positive. An increase in the buffer ph generally increases the electro-osmotic flow. (3) Organic solvents Organic modifiers (methanol, acetonitrile, etc.) may be added to the aqueous buffer to increase the solubility of the solute or other additives and/or to affect the degree of ionization of the sample components. The addition of these organic modifiers to the buffer generally causes a decrease in the electroosmotic flow. (4) Additives for chiral separations For the separation of optical isomers, a chiral selector is added to the separation buffer. The most commonly used chiral selectors are cyclodextrins, but crown ethers, polysaccharides and proteins may also be used. Since chiral recognition is governed by the different interactions between the chiral selector and each of the enantiomers, the resolution achieved for the chiral compounds depends largely on the type of chiral selector used. In this regard, for the development of a given separation, it may be useful to test cyclodextrins having a different cavity size ( -, -, or -cyclodextrin) or modified cyclodextrins with neutral (methyl, ethyl, hydroxyalkyl, etc.) or ionizable (aminomethyl, carboxymethyl, sulfobutyl ether, etc.) groups. When using modified cyclodextrins, batch-to-batch variations in the degree of substitution of the cyclodextrins must be taken into account since it will influence the selectivity. Other factors controlling the resolution in chiral separations are concentration of chiral selector, composition and ph of the buffer and temperature. The use of organic additives, such as methanol or urea can also modify the resolution achieved. 2. Capillary Gel Electrophoresis In capillary gel electrophoresis, separation takes place inside a capillary filled with a gel that acts as a molecular sieve. Molecules with similar charge to mass ratios are separated according to molecular size since smaller molecules move more freely through the network of the gel and therefore migrate faster than larger molecules. Different biological macromolecules (for example, proteins and DNA fragments), which often have similar charge to mass ratios, can thus be separated according to their molecular mass by capillary gel electrophoresis. Characteristics of Gels Two types of gels are used in capillary electrophoresis: permanently coated gels and dynamically coated gels. Permanently coated gels, such as cross-linked polyacrylamide, are prepared inside the capillary by polymerization of the monomers. They are usually bonded to the fused silica wall and cannot be removed without destroying the capillary. If the gels are used for protein analysis under reducing conditions, the separation buffer usually contains sodium dodecyl sulfate and the samples are denatured by heating a mixture of sodium dodecyl sulfate and 2-mercaptoethanol or dithiothreitol before injection. When non-reducing conditions are used (for example, analysis of an intact antibody), 2-mercaptoethanol and dithiothreitol are not used. Separation in cross-linked gels can be optimized by modifying the separation buffer (as indicated in the capillary zone electrophoresis section) and controlling the gel porosity during the gel preparation. For crosslinked polyacrylamide gels, the porosity can be modified by changing the concentration of acrylamide and/or the proportion of cross -linker. As a rule, a decrease in the porosity of the gel leads to a decrease in the mobility of the solutes. Due to the rigidity of these gels, only electrokinetic injection can be used. Dynamically coated gels are hydrophilic polymers, such as linear polyacrylamide, cellulose derivatives, dextran, etc., which can be dissolved in aqueous separation buffers giving rise to a separation medium that also acts as a molecular sieve. These separation media are easier to prepare than cross-linked polymers. They can be prepared in a vial and filled by pressure in a wall-coated capillary with no electroosmotic flow. Replacing the gel before every injection generally improves the separation reproducibility. The porosity of the gels can be increased by using polymers of higher molecular mass (at a given polymer concentration) or by decreasing the polymer concentration (for a given polymer molecular mass). A reduction in the gel porosity leads to a decrease in the mobility of the solute for the same buffer. Since the dissolution of these polymers in the buffer gives low viscosity solutions, both hydrodynamic and electrokinetic injection techniques can be used. 3. Capillary Isoelectric Focusing

7 KP X 1821 In isoelectric focusing, the molecules migrate under the influence of the electric field, so long as they are charged, in a ph gradient generated by ampholytes having pi values in a wide range (polyaminocarboxylic acids), dissolved in the separation buffer. The three basic steps of isoelectric focusing are loading, focusing and mobilization. (1) Loading step: Two methods may be employed: (i) loading in one step: the sample is mixed with ampholytes and introduced into the capillary either by pressure or vacuum; (ii) sequential loading: a leading buffer, then the ampholytes, then the sample mixed with ampholytes, again ampholytes alone and finally the terminating buffer are introduced into the capillary. The volume of the sample must be small enough not to modify the ph gradient. (2) Focusing step: When the voltage is applied, ampholytes migrate toward the cathode or the anode, according to their net charge, thus creating a ph gradient from anode (lower ph) to cathode (higher ph). During this step the components to be separated migrate until they reach a ph corresponding to their isoelectric point (pi) and the current drops to very low values. (3) Mobilization step: If mobilization is required for detection, use one of the following methods. Three methods are available: (i) in the first method, mobilization is accomplished during the focusing step under the effect of the electro-osmotic flow; the electro-osmotic flow must be small enough to allow the focusing of the components; (ii) in the second method, mobilization is accomplished by applying positive pressure after the focusing step; (iii) in the third method, mobilization is achieved after the focusing step by adding salts to the cathode reservoir or the anode reservoir (depending on the direction chosen for mobilization) in order to alter the ph in the capillary when the voltage is applied. As the ph is changed, the proteins and ampholytes are mobilized in the direction of the reservoir which contains the added salts and pass the detector. The separation achieved, expressed as ΔpI, depends on the ph gradient (dph/dx), the number of ampholytes having different pi values, the molecular diffusion coefficient (D), the intensity of the electric field (E) and the variation of the electrophoretic mobility of the analyte with the ph (-dµ/dph): pi 3 D ( dph / dx) E ( d / dph) Optimization The main parameters to be considered in the development of separations are: (1) Voltage Capillary isoelectric focusing utilizes very high electric fields, 300 V/cm to 1000 V/cm in the focusing step. (2) Capillary The electro-osmotic flow must be reduced or suppressed depending on the mobilization strategy (see above). Coated capillaries tend to reduce the electro-osmotic flow. (3) Solutions The anode buffer reservoir is filled with a solution with a ph lower than the pi of the most acidic ampholyte and the cathode reservoir is filled with a solution with a ph higher than the pi of the most basic ampholyte. Phosphoric acid for the anode and sodium hydroxide for the cathode are frequently used. Addition of a polymer, such as methylcellulose, in the ampholyte solution tends to suppress convective forces (if any) and electro-osmotic flow by increasing the viscosity. Commercial ampholytes are available covering many ph ranges and may be mixed if necessary to obtain an expanded ph range. Broad ph ranges are used to estimate the isoelectric point whereas narrower ranges are employed to improve accuracy. Calibration can be done by correlating migration time with isoelectric point for a series of protein markers. During the focusing step precipitation of proteins at their isoelectric point can be prevented, if necessary, using buffer additives such as glycerol, surfactants, urea or zwitterionic buffers. However, depending on the concentration, urea denatures proteins. 4. Micellar Electrokinetic Chromatography (MEKC) In micellar electrokinetic chromatography, separation takes place in an electrolyte solution which contains a surfactant at a concentration above the critical micellar concentration (CMC). The solute molecules are distributed between the aqueous buffer and the pseudo-stationary phase composed of micelles, according to the partition coefficient of the solute. The technique can therefore be considered as a hybrid of electrophoresis and chromatography. MECK is a technique that can be used for the separation of both neutral and charged solutes, maintaining the efficiency, speed and instrumental suitability of capillary electrophoresis. One of the most widely used surfactants in MEKC is the anionic surfactant sodium dodecyl sulfate, although other surfactants, for example cationic surfactants such as cetyltrimethylammonium salts, are also used. The separation mechanism in MECK is as follows. At neutral and alkaline ph, a strong electro-osmotic flow is generated and moves the separation buffer ions in the direction of the cathode. If sodium dodecyl sulfate is employed as the surfactant, the electrophoretic migration of the anionic micelle is in the opposite direction, towards the anode. As a result, the overall micelle migration velocity is slowed down compared to the bulk flow of the electrolytic solution. In the case of neutral solutes, since the analyte can partition between the micelle and the aqueous buffer, and has no electrophoretic mobility, the analyte migration velocity will depend only on the partition coefficient between the micelle and the aqueous buffer. In

8 1822 General Information the electropherogram, the peaks corresponding to each uncharged solute are always between that of the electroosmotic flow marker and that of the micelle (the time elapsed between these two peaks is called the separation window). For electrically charged solutes, the migration velocity depends on both the partition coefficient of the solute between the micelle and the aqueous buffer, and on the electrophoretic mobility of the solute in the absence of micelle. Since the mechanism in MEKC of neutral or weakly ionized solutes is essentially chromatographic, migration of the solute and resolution can be rationalized in terms of the retention factor of the solute (k ), also referred to as mass distribution ratio (D m ), which is the ratio of the number of moles of solute in the micelle to those in the mobile phase. tr t0 k' t t0 1 t R mc V K V t R : migration time of the solute, t 0 : analysis time of an unretained solute (determined by injecting an electro-osmotic flow marker which does not enter the micelle, for instance methanol), t mc : micelle migration time (measured by injecting a micelle marker, such as Sudan III, which migrates while continuously associated in the micelle), K: partition coefficient of the solute, V S : volume of the micellar phase, V M : volume of the mobile phase. Likewise, the resolution between two closely migrating solutes (R S ) is given by the following equation: R S N 1 k' b mc 4 k' t b 1 0 S M t 1 ( t 1 ( t mc 0 ) ) k' N: number of theoretical plates for one of the solutes, : selectivity, k a, k b : retention factors for both solutes, respectively (k b > k a ) Similar, but not identical, equations give k and R S values for electrically charged solutes. Optimization The main parameters to be considered in the development of separations by MEKC are instrumental and electrolytic solution parameters. Instrumental parameters (1) Voltage Separation time is inversely proportional to applied voltage. However, an increase in voltage can a cause excessive heat production that gives rise to temperature gradients and viscosity gradients of the buffer in the cross-section of the capillary. This effect can be significant with high conductivity buffers such as those containing micelles. Poor heat dissipation causes band broadening and decreases resolution. (2) Temperature Variations in capillary temperature affect the partition coefficient of the solute between the buffer and the micelles, the critical micellar concentration and the viscosity of the buffer. These parameters contribute to the migration time of the solutes. The use of a good cooling system improves the reproducibility of the migration time for the solutes. (3) Capillary As in capillary zone electrophoresis, the dimensions of the capillary (length and internal diameter) contribute to analysis time and efficiency of separations. Increasing both effective length and total length can decrease the electric fields (working at constant voltage), increase migration time and improve the separation efficiency. The internal diameter controls heat dissipation (for a given buffer and electric field) and consequently the sample zone diffusion. Electrolytic solution parameters (1) Surfactant type and concentration The type of surfactant, in the same way as the stationary phase in chromatography, affects the resolution since it modifies separation selectivity. Also, the log k of a neutral compound increases linearly with the concentration of surfactant in the mobile phase. Since resolution in MEKC reaches a maximum when k approaches the value of t m t0, modifying the concentration of surfactant in the mobile phase changes the resolution obtained. (2) Buffer ph Although ph does not modify the partition coefficient of non-ionized solutes, it can modify the electroosmotic flow in uncoated capillaries. A decrease in the buffer ph decreases the electro-osmotic flow and therefore increases the resolution of the neutral solutes in MEKC, resulting in a longer analysis time. (3) Organic solvents To improve MEKC separation of hydrophobic compounds, organic modifiers, such as methanol, propanol, acetonitrile, etc., can be added to the electrolytic solution. The addition of these modifiers usually decreases migration time and the selectivity of the separation. Since the addition of organic modifiers affects the critical micellar concentration, a given surfactant concentration can be used only within a certain percentage of organic modifier before the micellization is inhibited or adversely affected, resulting in the absence of micelles and, therefore, in the absence of partition. The dissociation of micelles in the presence of a high content of organic solvent does not always mean that the separation will no longer be possible; in some cases the hydrophobic interaction between the ionic surfactant monomer and the neutral solutes forms solvophilic complexes that can be separated electrophoretically.

9 KP X 1823 (4) Additives for chiral separations For the separation of enantiomers using MEKC, a chiral selector is included in the micellar system, either covalently bound to the surfactant or added to the micellar separation electrolyte. Micelles that have a moiety with chiral discrimination properties include salts of N- dodecanoyl-l -amino acids, bile salts, etc. Chiral resolution can also be achieved using chiral discriminators, such as cyclodextrins, added to the electrolytic solutions which contain micellized achiral surfactants. (5) Other additives Several strategies can be carried out to modify selectivity, by adding chemicals to the buffer. The addition of several types of cyclodextrins to the buffer can also be used to reduce the interaction of hydrophobic solutes with the micelle, thus increasing the selectivity for this type of compound. The addition of substances, capable of modifying solute-micelle interactions by adsorption on the latter, is used to improve the selectivity of the separations in MEKC. These additives may be a second surfactant (ionic or non-ionic) which gives rise to mixed micelles or metallic cations which dissolve in the micelle and form co-ordination complexes with the solutes. Quantification Peak areas must be divided by the corresponding migration time to give the corrected area in order to: (1) compensate for the shift in migration time from run to run, thus reducing the variation of the response, (2) compensate for the different responses of sample constituents with different migration times. Where an internal standard is used, verify that no peak of the substance to be examined is masked by that of the internal standard. Calculations From the values obtained, calculate the content of the component or components being examined. When prescribed, the percentage content of one or more components of the sample to be examined is calculated by determining the corrected area(s) of the peak(s) as a percentage of the total of the corrected areas of all peaks, excluding those due to solvents or any added reagents. The use of an automatic integration system (integrator or data acquisition and processing system) is recommended. System Suitability In order to check the behavior of the capillary electrophoresis system, system suitability parameters are used. The choice of these parameters depends on the mode of capillary electrophoresis used. They are: retention factor (k, only for micellar electrokinetic chromatography), apparent number of theoretical plates (N), symmetry factor (A S ) and resolution (R S ). In previous sections, the theoretical expressions for N and R S have been described, but more practical equations that allow these parameters to be calculated from the electropherograms are given below. Apparent Number of Theoretical Plates t 5.54 R N wh t R : migration time or distance along the baseline from the point of injection to the perpendicular dropped from the maximum of the peak corresponding to the component, w h : width of the peak at half -height. Resolution The resolution (R S ) between peaks of similar height of two components may be calculated using the following expression: 1.18 ( tr2 tr1) RS tr2 tr1 wh1 w h2 t R1, t R2 : migration times or distances along the baseline from the point of injection to the perpendiculars dropped from the maxima of two adjacent peaks, w h1, w h2 : peak widths at half -height. When appropriate, the resolution may be calculated by measuring the height of the valley (H v ) between two partly resolved peaks in a standard preparation and the height of the smaller peak (H p ) and calculating the peak-to-valley ratio: H p / v H Symmetry Factor The symmetry factor (A S ) of a peak may be calculated using the following expression: w0. 05 A S 2 d w 0.05 : width of the peak at 1/20 of the peak height, d: distance between the perpendicular dropped from the peak maximum and the leading edge of the peak at 1/20 of the peak height. Tests for area repeatability (standard deviation of areas or of the area/migration-time ratio) and for migration time repeatability (standard deviation of migration time) are introduced as suitability parameters. Migration time repeatability provides a test for the suitability of the capillary washing procedures. An alternative practice to avoid the lack of repeatability of the migration time is to use migration time relative to an internal standard. A test for the verification of the signal-to-noise ratio for a standard preparation (or the determination of the limit of quantification) may also be useful for the determination of related substances. Signal-to-noise Ratio p v 2

10 1824 General Information The detection limit and quantification limit correspond to signal-to-noise ratios of 3 and 10 respectively. The signal-to-noise ratio (S/N) is calculated using the following expression: H S / N 2 h H: height of the peak corresponding to the component concerned, in the electropherogram obtained with the prescribed reference solution, measured from the maximum of the peak to the extrapolated baseline of the signal observed over a distance equal to twenty times the width at half height, h: range of the background in an electropherogram obtained after injection of a blank, observed over a distance equal to twenty times the width at the half height of the peak in the electropherogram obtained with the prescribed reference solution and, if possible, situated equally around the place where this peak would be found. 2. Determination of Bulk and Tapped Densities Determination of Bulk and Tapped Densities is a method to determine the bulk densities of powdered drugs under loose and tapped packing conditions, respectively. Loose packing is defined as the state obtained by pouring a powder sample into a vessel without any consolidation, and tapped packing is defined as the state obtained when the vessel containing the powder sample is to be repeatedly dropped a specified distance at a constant drop rate until the apparent volume of sample in the vessel becomes almost constant. The bulk density is expressed in mass per unit apparent volume of powder (g/ml). Because the bulk density is one of the measures of packing properties, compressibility and flow properties, and is dependent on the history of the powder, it is essential to document the bulk density to specify how the determination was made. Bulk density The bulk density is an apparent density obtained by pouring a powder sample into a vessel without any consolidation. The determination of bulk density is achieved by measuring the apparent volume of a powder sample having a known mass in a graduated cylinder (Method 1) or by measuring the mass of powder in a vessel having a known volume (Method 2). Method 1 Constant mass method Unless otherwise specified, pass a quantity of sample sufficient to complete the test through a sieve No. 16 (1000 m) to break up agglomerates that may have formed during storage. Weigh accurately about 30 g of test sample, and pour it into a dry 100 ml graduated glass cylinder (readable to 1 ml). Carefully level the powder without consolidation, if necessary, and read the unsettled apparent volume to the nearest graduated unit and calculate the bulk density ρ B by the following formula: M B V 0 B : Bulk density by constant mass method (g/ml) M : Mass of powder sample (g) V 0 : Apparent volume of powder sample (ml) Record the average of 3 determinations and regard the average value as the bulk density according to the constant mass method. If a 30 g sample is too large to determine, adjust the mass of sample so as to provide an apparent volume of ml. Method 2 Constant volume method Unless otherwise specified, pass a quantity of sample sufficient to complete the test through a sieve No. 16 (1000 µm) to break up agglomerates that may have formed during storage. Allow an excess of sample powder to pour into the measuring vessel having the volume of V and mass of M 0. Then, carefully scrape excess powder from the top of the vessel using the edge of a slide glass or other tool by smoothly moving across it. Remove any material from the sides of the vessel, and determine the total mass M t and calculate the bulk density ρ B by the following formula: ( M t M 0 ) B V ρ B : Bulk density by constant volume method (g/ml) M T : Total mass of powder and measuring vessel (g) M 0 : Mass of measuring vessel (g) V: Volume of measuring vessel (ml) Record the average of 3 determinations and regard the average value as the bulk density according to the constant volume method. Measuring vessel

11 KP X 1825 ρ T : Tapped density by constant mass method (g/ml) M: Mass of powder sample (g) V f : Final apparent volume of sample after tapping (ml) Record the average of 3 determinations and regard the average value as the tap density according to the constant mass method. Supplementary cylinder The figures are in mm. This figure is one of the representations of 100 ml vessel and supplementary cylinder. Figure 1. Vessels for the determination of bulk density by constant volume method and for the determination tapped density Tapped density Tapped density is an apparent density obtained by mechanically tapping a measuring vessel containing a powder sample. The determination of tapped density is achieved by measuring the apparent volume of a powder sample having a known mass in a vessel after tapping (Method 1) or by measuring the mass of powder in a vessel having a known volume after tapping (Method 2). Method 1 Constant mass method Unless otherwise specified, pass a quantity of sample sufficient to complete the test through a sieve No. 16 (1000 µm) or a sieve No. 22 (710 µm) to break up agglomerates that may have formed during storage. Weigh accurately about 100 g of test sample, and pour it into a 250 ml graduated glass cylinder (readable to 2 ml) without consolidation. If the sample is not sufficient, proceed according to the same procedure as that described above by using a 100 ml graduated glass cylinder (readable to 1 ml). It is essential to select appropriate masses of the cylinder support, holder and cylinder so as to ensure the dynamic stability of the apparatus during tapping. After attaching the glass cylinder containing the powder sample to the tapping apparatus, carry out tapping under the measuring conditions (tapping rate and drop height) specified for each apparatus. Unless otherwise specified, repeat increments of 50 taps or 1 minute until the difference between succeeding measurements is less than 2%, and determine the final apparent volume, V f and calculate the tapped density ρ T by the following formula: T M V f Method 2 Constant volume method Unless otherwise specified, pass a quantity of sample sufficient to complete the test through a sieve No. 16 (1000 µm) to break up agglomerates that may have formed during storage. Attach a supplementary cylinder to the stainless steel measuring vessel having a known mass of M 0 and a volume of V (Fig. 1), and then pour an excess of the sample into the vessel. After setting up the vessel in an adequate tapping apparatus with a fixed drop height, carry out tapping at the rate and cumulative tap number specified for each apparatus. Then remove the supplementary cylinder from the vessel and carefully scrape excess powder from the top of the vessel by smoothly moving across it the edge of a slide glass or other tool. Remove any material from the sides of the vessel, and determine the total mass M t. Calculate the tapped density ρ T by the formula: ( M t M 0 ) T V ρ T : Tapped density by constant volume method (g/ml) M t : Total mass of powder and measuring vessel (g) M 0 : Mass of measuring vessel (g) V: Volume of measuring vessel (ml) Determine 3 measurements and calculate the average and the relative standard deviation using 3 different powder samples. Regard the average value as the tap density according to the constant volume method. If the relative standard deviation is not less than 2%, repeat the test with further tapping. Note: Use balances readable to the nearest 0.1 g. 3. Disinfection and Sterilization Methods Disinfection and Sterilization Methods are applied to kill microorganisms in processing equipment/utensils and areas used for drug manufacturing, as well as to perform microbiological tests specified in the monographs, and so differ from Terminal Sterilization and Filtration Method described in Terminal Sterilization and Sterilization Indicators. The killing effect on microorganisms or the estimated level of sterility assurance is greatly variable, so the conditions for

12 1826 General Information disinfection and sterilization treatment must be chosen appropriately for each application. Generally, the following methods are to be used singly or in combination after appropriate optimization of operation procedures and conditions, in accordance with the kind and the degree of the contaminating microorganisms and the nature of the items to which the methods are applied. The validation of sterilization in accordance with Terminal Sterilization and Sterilization Indicators is required when the methods are applied to the manufacturing process of drug products. 1. Disinfection methods These methods are used to reduce the number of living microorganisms, but do not always remove or kill all microorganisms present. Generally, disinfection is classified into chemical disinfection with the use of chemical drugs (disinfectants) and physical disinfection with the use of moist heat, ultraviolet light, and other agents Chemical disinfection Microorganisms are killed with chemical drugs. The killing effect and mechanisms of a chemical drug differ depending on the type, applied concentration, action temperature, and action time of the chemical drug used, the degree of contamination on the object to be disinfected, and the series and state (e.g., vegetative bacteria or spore bacteria) of microorganisms. Therefore, in applying the method, full consideration is required of the sterility and permissible storage period of the prepared chemical drug, the possibility of resistance of microorganisms at the site of application, and the effect of residual chemical drug on the product. In selecting a suitable chemical drug, the following items should be considered in relation to the intended use. 1) The antimicrobial spectrum 2) Action time for killing microorganisms 3) Action durability 4) Effect of the presence of proteins 5) Influence on the human body 6) Solubility in water 7) Influence on the object to be disinfected 8) Odor 9) Convenience of use 10) Easy disposability 11) Influence on the environment at disposal 12) Frequency of occurrence of resistance 1.2. Physical disinfection Microorganisms are killed without a chemical drug. (i) Steam flow method Microorganisms are killed by direct application of steam. This method is used for a product which may be denatured by the moist heat method. As a rule, the product is kept in flowing steam at 100 C for 30 to 60 minutes. (ii) Boiling method Microorganisms are killed by putting the object in boiling water. This method is used for a product which may be denatured by the moist heat method. As a rule, the product is put in boiling water for 15 minutes or more. (iii) Intermittent method Microorganisms are killed by heating for 30 to 60 minutes repeatedly, three to five times, once a day in water at 80 to 100 C or in steam. This method is used for a product which may be denatured by the moist heat method. There is another method called the low temperature intermittent method with repeated heating at 60 to 80 C. During the intermission periods between heating or warming, a suitable temperature for the growth of microorganisms of 20 C or higher, must be maintained. (iv) Ultraviolet method As a rule, microorganisms are killed by irradiation with ultraviolet rays as a wavelength of around 254 nm. This method is used for products which are resistant to ultraviolet rays, such as smooth-surfaced articles, facilities, and equipment, or water and air. This method does not suffer from the occurrence of resistance, which is observed in chemical disinfection, and shows a killing effect on bacteria, fungi, and viruses. It must be taken into consideration that direct ultraviolet irradiation of the human body can injure the eyes and skin. 2. Sterilization methods 2.1. Heating methods In these methods, the heating time before the temperature or pressure reaches the prescribed value differs according to the properties of the product, the size of the container, and the conditions. The duration of heating in conducting these methods is counted from the time when all the parts of the product have reached the prescribed temperature. (i) Moist heat method Microorganisms are killed in saturated steam at a suitable temperature and pressure. This method is generally used for heat-stable substances, such as glass, porcelain, metal, rubber, plastics, paper, and fiber, as well as heat-stable liquids, such as water, culture media, reagents, test solutions, liquid samples, etc. As a rule, one of the following conditions is used. 115 ~ 118 C for 30 minutes 121 ~ 124 C for 15 minutes 126 ~ 129 C for 10 minutes (ii) Dry-heat method Microorganisms are killed in dry-heated air. This method is generally used for heat-stable substances, such as glass, porcelain, and metal, as well as heatstable products, such as mineral oils, fats and oils, powder samples, etc. This method is generally conducted in the way of direct heating by gas or electricity or circulating heated air. As a rule, one of the following conditions is used.

13 KP X ~ 170 C for 120 minutes 170 ~ 180 C for 60 minutes 180 ~ 190 C for 30 minutes 2.2. Irradiation methods (i) Radiation method Microorganisms are killed by gamma-rays emitted from a radioisotope or electron beam and bremsstrahlung (X-ray) generated from an electron accelerator. This method is generally used for radiation-resistant substances such as glass, porcelain, metal, rubber, plastics, fiber, etc. The dose is decided according to the material properties, and the degree of contamination of the product to be sterilized. Special consideration is necessary of the possibility of qualitative change of the product after the application of the method. (ii) Microwave method Microorganisms are killed by the heat generated by direct microwave irradiation. This method is generally used for microwave-resistant products such as water, culture media, test solutions, etc. As a rule, microwave radiation with a wavelength of around 2450±50 MHz is used Gas methods Microorganisms are killed by a sterilizing gas. Suitable gases for killing microorganisms include ethylene oxide gas, formaldehyde gas, hydrogen peroxide gas, chlorine dioxide gas, etc. Temperature, humidity, the concentration of gas, and the exposure time differ in accordance with the species of gas used. As sterilizing gases are generally toxic to humans, full consideration is required of the environmental control for the use of gases and the concentration of residual gas. In some of the gas methods, it may be difficult to measure or estimate quantitatively the killing of microorganisms Filtration method Microorganisms are removed by filtration with a suitable filtering device. This method is generally used for gas, water, or culture media and test solutions containing a substance that is water-soluble and unstable to heat. As a rule a filter having a pore size of 0.22 m or smaller is used for the sterilization. However, in this method, a filter with a pore size of 0.45 m of smaller is permitted to be used. 4. Guideline for Setting Dissolution Specification of Oral Dosage Forms 1. Objective This guideline is intended to provide specified approaches for setting dissolution specification of oral solid dosage forms such as tablets and capsules. By securing quality uniformity of drug product, this guideline contributes to production of qualified drug products. 2. Significance of Dissolution Testing Drug absorption from a solid dosage form, such as the tablet and the capsule, after oral administration is governed by a number of factors, including the release properties of the drug substance from the drug product, the dissolution or solubilization properties of the drug under a given physiological condition and the permeability across the intestinal epithelial barrier. A dissolution testing is a test that determines not only the release of the drug substance but also the dissolution or solubilization of the drug. Ultimately, a dissolution testing can lead to the estimation of the performance of the drug product in the biological system. Therefore, when a dissolution testing is undertaken, it is advisable to consider the solubility of the drug, the dissolution, the intestinal permeability, the characteristics of the dosage form, the pharmacokinetic properties as well as physiological factors. In order to mimic physiological conditions after the oral administration, the conditions for dissolution testing vary, e.g., a various dissolution media with varying phs to represent the environments in the gastrointestinal tract, a number of settings in the rotational speed to simulate the peristaltic movement of the gastrointestinal tract and addition of lipids, enzymes and/or surfactants to mimic the physiological conditions in the tract. Sometimes the results from a dissolution test may be used to predict the kinetics of the absorption of the drug (viz, in vitro-in vivo correlation). However, it is noted that the condition of dissolution testing does not always attempt to represent completely physiological conditions in the gastrointestinal tract. That is, useful information on the quality control and the quality assurance of a given drug product can also be obtained from the dissolution test that does not completely represent physiological conditions in human gastrointestinal tract. A. Lot to lot uniformity in the quality B. Development of composition of drug product and of formulation C. Understanding release mechanism of drug from product D. Determination of pharmaceutical equivalence during storage E. Determination of pharmaceutical equivalence of post approval changes in drug substance, in composition, and in method, site and scale of manufacture 3. Glossary a. Conventional release dosage forms (also known as immediate release dosage forms) A conventional release dosage form or immediate release dosage form is a dosage form in which the dissolution profile is primarily governed by the physicochemical property of the drug substance and cannot be intentionally altered simply by changing the amount of the drug substance in the product or the method of manufacture.

14 1828 General Information b. Delayed release dosage forms (also known as enteric coated dosage forms) A delayed release dosage form or enteric coated dosage form is a dosage form in which the dissolution of the drug is delayed for a certain period after the administration so that the drug is not released in the acidic condition of the stomach by means of composing the product with specialized compositions or specialized methods. Pharmacokinetic criteria of the dosage form are such that the dissolution profile of the dosage form is comparable to its conventional release dosage form counterpart except that the release occurs after a certain time. c. Prolonged release dosage forms (also known as extended release dosage forms) A prolonged release dosage form or extended release dosage form is a dosage form in which the release of the drug substance is prolonged, compared with that from the conventional release dosage form of the same route of administration, and the frequency of administration is reduced compared with that in the conventional forms. These dosage forms are typically manufactured by means of specialized compositions and/or specialized methods. 4. General consideration in setting the dissolution specifications a. Classification of dosage forms In this guideline, dosage forms are classified into conventional release dosage forms, delayed release dosage forms and prolonged release dosage forms as specified in section 3. Glossary. b. Apparatus for dissolution testing i. Apparatus Currently, seven types of dissolution apparatuses are specified in the official compendium and the types are illustrated in Fig. 1. Depending on the properties of the dosage form, a suitable apparatus may be selected for the dissolution testing. Two of the most frequently used methods are the Method 1, the rotary basket method, and the Method 2, the paddle method, both specified in the Dissolution Test. The rotary basket method and the paddle method are both simple and robust testing procedure and are highly standardized globally. Therefore, unless a special dissolution testing is required, the rotary basket method and the paddle method are recommended. When a dissolution test is to be conducted by the rotary basket method or the paddle method, the apparatus should conform to the specification of the Dissolution Test as defined in the General Tests of the Korean Pharmacopoeia. When a dissolution test is to be conducted for the development of a new drug product, the selection between the rotary basket method and the paddle method depends on the consideration of the in vitro dissolution pattern and in vivo pharmacokinetic characteristics. In general, the rotary basket methods is recommended for capsules, a dosage form that is not ease to sink, or for slowly dissolving dosage forms. Although dissolutions from tablets are typically tested with the paddle method, the rotary basket method, rather than the paddle method, may be more suitable in case where the disintegrated debris are settled down at the bottom of the flask thereby causing a delay in the dissolution. Also dissolution apparatuses for dissolution testing is recommended to accomplish the testing of Suitability test for dissolution apparatus (see Appendix 1) for an every certain period of time. Apparatus 1 Apparatus 2 (basket) (paddle) Apparatus 3 Apparatus 4 (flow-through cell) (reciprocating cylinder) Apparatus 5 Apparatus 6 (paddle over disk) (cylinder)

15 KP X 1829 Acryl rod Spring holder Teflon cylinder Angled disk Reciprocating disk Apparatus 7 (reciprocating holder) Figure 1. Types of Dissolution Apparatuses ii. Rotational speed In general, the rotational speed is set at 50 rpm for the paddle method and at 100 rpm for the rotary basket method. A rotational speed faster than 100 rpm is necessary for prolonged dosage forms tested by the paddle method. However, it is not always necessary to set the above mentioned speeds in a test. For example, in case where 100% of the drug is dissolved within 10 minutes using the rotary basket method at the rotational speed of 100 rpm for a dissolution test, it is advisable to set a lower rotation speed and/or to change the composition of the dissolution medium so that the dissolution pattern is discernable. Therefore, the rotational speed of a test should be set at a speed that is able to identify potential lack of pharmaceutical bioequivalence of a drug product. iii. Dissolution medium 1) General considerations - Although the dissolution medium, having the ph of 1.2, 4.0, 4.5, 6.8 or 7.4, is generally used, the medium may be set at a different ph and have a different composition, if necessary because of the property of the dosage form (e.g., the stability and/or the solubility of the drug) or of a specific experimental purpose. Under these circumstances, the properties, including the stability and solubility, of the drug substance must be considered. - Select a dissolution medium that is able to properly identify the difference in the dissolution rate between drug products or between lots according to the factors that are composition of drug product, manufacturing process, equipment, drug substance, etc. In general, a dissolution medium that results in a very rapid dissolution is not suitable for the identification and, thus, it is advisable to select a dissolution medium that shows a relatively slower rate of dissolution. - In case where a given drug substance adsorbs to the surface of dissolution apparatus, thereby creating a problem in analyzing the results, use a dissolution medium with a suitable agent that prevents the adsorption. - Suitable surfactants may be added to dissolution medium to enhance solubility of a very poorly soluble drug. However, addition of organic solvent to medium is not recommended for dissolution test. - In case where the identical result cannot be obtained for the test with the deaerated dissolution medium and for that with the nondeaerated dissolution medium, then the deaeration procedure is necessary (see Appendix II). 2) Characteristics of dissolution medium a) Water Since water does not have any buffer capacity, the disadvantages of water as a dissolution medium include the potential changes in the ph and in the surface tension of the medium by the dissolution of drug and/or the additives. Despite the disadvantages, however, water is the most convenient medium and ecologically-acceptable. Therefore, water is the most widely used dissolution medium, in case where the dissolution pattern from the formulation is not affected by using water as the medium. b) ph 1.2 Dissolution medium Use the disintegration 1 st fluid. This medium is prepared using hydrochloric acid and sodium chloride, and the concentration of hydrochloric acid in the medium is set at about 0.1 mol/l. In case where a particular drug becomes unstable in the medium, it is necessary that the testing condition is modified to maintain the stability. In fact, the dissolutions from a number of formulations are tested with the media having the concentration of hydrochloric acid ranging from mol/l to 0.1 mol/l. c) ph 4.0/4.5 Dissolution medium o ph 4.0 Dissolution medium: Use 0.05 mol/l sodium acetate buffer solution [a mixture of 0.05 mol/l acetic acid and 0.05 mol/l sodium acetate (82:18)]. o ph 4.5 Dissolution medium: Use a buffer solution prepared by dissolving 2.99 g of sodium acetate trihydrate and 1.66 g of glacial acetic acid in water to make 1000 ml.

16 1830 General Information o In case where these buffer solutions interact with a drug to create undesirable problem, a buffer solution prepared from citrate or phosphate may be alternatively used. d) ph 6.0 Dissolution medium Use disodium phosphate and citric acid buffer solution (ph 6.0) of the Korean Pharmacopoeia. e) ph 6.8 Dissolution medium Use the disintegration 2 nd fluid. f) Simulated gastric fluid Use simulated gastric fluid TS of USP. g) Simulated intestinal fluid Use simulated intestinal fluid TS of USP. h) Neutral and basic dissolution medium Although the dissolution medium, having the ph of 1.2, 4.0, 4.5, 6.8 or 7.4, is generally used, the medium may be set at a different ph and have a different composition, if necessary because of the property of the dosage form (e.g., the stability and/or the solubility of the drug) or of a specific experimental purpose, and the dissolution is tested. Examples of the special conditions include the cases when a particular drug is the most stable in basic dissolution media, when basic condition allows a discriminating analysis or when a given drug has a sufficient solubility in basic media. When a dissolution is tested for a product in basic condition, coupled with the HPLC assay using C 18 column, it may be necessary to neutralize or sufficiently dilute the test solution before the introduction to the column because the packing material tends to be unstable at high phs. 3) Volume of dissolution medium Although the volume of dissolution medium is typically 900 ml in the test, it may be adjusted to L, depending on the testing condition. Dissolution test flask ranging from 150 ml to over 1 L in size is currently available. General considerations for setting the volume of dissolution medium are as follows; a) Solubility of drug Volume of dissolution medium is sufficient enough to maintain a sink condition during the testing. In general, a sink condition is a condition where at least three times (generally 5-10 times) the volume for the complete dissolution of a given amount of drug is available. Therefore, under such condition, the concentration of the drug should always be less than 30% of the intrinsic solubility when the drug product is completely dissolved in the volume of medium used in the test. In the dissolution test, it is necessary to set the dissolution specification such that the process of the dissolution of drug to medium is not the rate limiting step in the dissolution rate by maintaining a sink condition in the test. b) Use of surfactant It is advisable to use a sufficient volume of dissolution medium in the test when surfactant is added to the medium to enhance the solubility of a poorly water-soluble drug. c) Ease of assay For a dissolution test with a drug product having a trace amount of drug, it is recommended to use the possible minimum in volume when the concentration of the dissolved drug may be lower than the limit of quantification thereby creating an assay problem. d) Temperature of test medium In general, maintain the temperature of the medium at 37 ± 5 C for the dissolution test of oral solid dosage forms. iv. Dissolution testing of drug product containing poorly soluble drug 1) Determination of impact of medium ph on dissolution rate Conduct a series of dissolution tests with various media such as water, ph 1.2 dissolution medium, ph 4.0/4.5 dissolution medium and ph 6.8 dissolution medium. In these dissolution studies, use faster the rotational speed (for example, the rotational speed of 120 rpm for the rotary basket method and the rotational speed of 75 rpm for the paddle method). Sample a portion of the medium at suitable times after the initiation of the test, calculate the rate of dissolution for the drug and graphically present the results. Based on these results, the dissolution medium is selected for subsequent tests. In case where the dissolution rate of the drug is below the range between 70 and 85% of the drug, select a medium that shows the most rapid dissolution rate and consider addition of surfactants to the medium. 2) Selection of suitable surfactant Interaction between drug and surfactant is governed by the physicochemical properties of the drug and the surfactant. Therefore, for a given drug, a various surfactants, e.g., nonionic, anionic and cationic surfactants, have to be screened for the selection. In this case, the concentration is recommended to set initially at 2% and the suitability is tested for surfactants such as; o Anionic surfactant: Sodium lauryl o sulfate (SLS) Cationic surfactant: Cetyltrimethyl ammonium bromide (CTAB) o Non-ionic surfactant: Polysorbate 80, 40 and 20, lauryl dimethylamine N- oxide (LDAO) 3) Determination of surfactant concentration

17 KP X 1831 Use the lowest possible concentration of the surfactant that shows a sufficient dissolution rate within the specified time of the completion of the test (i.e., in 2 hours or 6 hours). To accomplish this, study the dissolution with a gradual decrease in the surfactant concentration such as from the initial 2%, then 1.5, 1.0, 0.75, 0.5 and finally to 0.25%. v. Dissolution testing for delayed release dosage forms To confirm the function of enteric coating or the diffusion control of the drug by coating, conduct a series of dissolution tests first in an acidic condition (i.e., acid stage) and then in buffer (i.e., buffer stage). As an example for a dissolution test for a delayed release dosage form in the official compendium, dissolution may be tested as directed in the procedure for delayed release dosage forms or first in 0.1 mol/l hydrochloric acid solution, an acidic condition, and, then, in ph 6.8 phosphate buffer in the buffer condition. For the dissolution testing via the two-stage approach, there are next ways; 1) Dissolution is first tested for 2 hours in 750 ml of 0.1 mol/l hydrochloric acid solution at 37 ± 5 C. After 2 hours, a portion of the dissolution medium is sampled for the quantification of the drug in the medium. To the above mentioned 0.1 mol/l hydrochloric acid dissolution medium, add 250 ml of 0.2 mol/l sodium phosphate solution, previously warmed to 37 ± 5 C. If necessary, the ph of the medium may be adjusted with 2 mol/l hydrochloric acid or with 1 mol/l sodium hydroxide, and the dissolution further tested (the adjustment of the ph should be carried out within not more than 5 minutes of the addition of 0.2 mol/l sodium phosphate solution). 2) Dissolution is first tested for 2 hours in 1 L of 0.1 mol/l hydrochloric acid solution at 37 ± 5 C. After 2 hours, a portion of the dissolution medium is sampled and the remaining medium is discarded. To the dissolution flask, transfer 1 L of ph 6.8 phosphate buffer and perform the Dissolution Test (prepared by adding 750 ml of 0.1 mol/l hydrochloric acid solution to 250 ml of 0.2 mol/l sodium phosphate solution, and, if necessary, adjusting the ph to 6.8 by the addition of 2 mol/l hydrochloric acid or of 1 mol/l sodium hydroxide), previously warmed to 37 ± 5 C. Another way is to transfer the paddle or the rotary basket, containing the sample, from the apparatus of the first dissolution study with 0.1 mol/l hydrochloric acid solution to a new dissolution flask containing ph 6.8 phosphate buffer and to continue to study the dissolution in the new set up. vi. Two stage dissolution testing for gelatin capsules For the case of the dissolution test with hard- or soft-gelatin capsules, a two-stage dissolution testing is recommended. The first stage dissolution is tested using typical dissolution medium as described in the previous section. When the dissolution rate is too slow to be tested properly, proceed to the second stage dissolution test with an addition of enzyme to the test medium. Pepsin or pancreatin is typically used as the digestive enzyme for the test, and the suitable enzyme is selected depending on the ph of the medium. When water or a dissolution medium having the ph of less than 6.8 is used in the test, pepsin is added with the activity of pepsin being not more than units per 1 L of the dissolution medium. In this case, the amount of pepsin addition is not more than 3.2 g per 1 L of the medium. For a dissolution medium having the ph not less than 6.8, pancreatin is used as the enzyme with the activity of pancreatin being not more than 1750 units per 1 L of the dissolution medium. In this case, the amount of pancreatin addition is not more than 50 mg per 1 L of the medium. In case where water is used as the dissolution medium at the first stage, 0.1 mol/l hydrochloric acid solution containing pepsin or ph 6.8 phosphate buffer containing pancreatin may be used as the second stage medium. However, if a high concentration of surfactant, such as sodium laury sulfate, is used in the first stage medium, a care should be taken in the second stage test because the presence of surfactant may cause denaturation of the enzymes in the medium. 5. Dissolution Specification a. Conventional release dosage forms There are two approaches in setting the dissolution specification for conventional release dosage forms. i. Singe-point specifications: This approach is typically applicable to drug products containing highly soluble drug substances (Biopharmaceutics Classification System(BCS) Class I and III) in setting the dissolution specification for the purpose of routine quality control. (e.g., Perform the test according to the next, the dissolution rate of Acetaminophen in 30 minutes should be not less than 80%. ) The single-point specification may be set for the dissolution test of a drug product in which not less than 80% of the drug is dissolved within minutes, even in the presence of the lag time (typically found in film coated formulations and capsules), of the initiation of the test. It is not necessary to attempt to construct in vitro-in vivo correlation for such product in the development stage. ii. Two-point specifications: This approach is typically applicable to drug products containing

18 1832 General Information poorly soluble drug substances (BCS Class II and IV) in setting the dissolution specification for the purpose of routine quality control. In general, the two-point specification is required for drug products having the dissolution time is not less than 45 minutes and is more useful than the single-point specification. b. Delayed release dosage forms The dissolution specification is set at not more than 10% for the acid stage and at the level specified in conventional release dosage forms for the buffer stage. c. Prolonged release dosage forms Set the dissolution specification using at least three time points. In this case, the sampling point should be such that the dissolution from the samples represents the initial, middle and final stage of the dissolution. The average rate of dissolution at the final stage should reach as close as possible to 80%. The initial stage time point (typically 1-2 hours of the initiation of the test) is selected so that the sample is able to identify the presence of initial burst of dissolution from the drug product. In general, 20-30% of the drug should be dissolved in the initial stage. The middle point is selected to identify the dissolution of 50% of the drug and the final point is selected to identify the dissolution of not less than 80% of the drug thus confirming the dissolution of the majority of the drug by the final sampling point. In case where the average rate of dissolution does not reach 80% by the completion of the test, however, the time of completion of the test is set at a point when the rate of dissolution does not change any more. For conventional release dosage forms, delayed release dosage forms and prolonged release dosage forms, the drug product should always conform to the dissolution specification during the indicated time of usage (i.e., within the expiration date). 6. Approaches in setting the dissolution specification In general, the dissolution specification is set according to the following approach. However, in case where more appropriate method is available or other suitable dissolution testing condition (e.g., the dissolution apparatus, the rotational speed and/or the dissolution medium etc.) that better represents the characteristics of the dosage form, the other condition may be used in the test. a. Selection of test specimen i. For a drug, select the lot(or equivalent) that has been used for the clinical test or the bioavailability test or the bioequivalence test. ii. Select the lot that the difference between the content of test specimen and its labeled amount is not more than 5% and difference between content of the test specimen and the content of reference is not more than 5%. b. The first sage (a preliminary test) i. Number of test specimen Proceed with 6 test specimens selected as directed in the section 6-a and perform the dissolution test. ii. Interval of sampling of dissolution solution Sample the dissolution solution up to 2 hours of the test at appropriate intervals (e.g., typically at 5, 10, 15, 30, 45, 60, 90 and 120 minutes) for each dissolution medium. However, the test may be terminated at a point when the final rate of dissolution reaches not less than 85%. iii. Conditions for dissolution testing 1) Dissolution apparatus a) Use the paddle method in the Dissolution Test of the Korean Pharmacopoeia as a preferred method. b) Use other test method such as the rotary basket method when, in the paddle method, the disintegrated debris is settled down at the bottom to form aggregates thereby interfering the test. c) A sinker may be used when the test specimen floats so that appropriately reproducible results cannot be obtained. 2) Dissolution medium a) ph 1.2: Use the medium 1 of the dissolution test in the Korean Pharmacopoeia b) ph 4.0/4.5: 0.05 mol/l sodium acetate buffer solution* 0.05 mol/l sodium acetate buffer solution*: a mixture of 0.05 mol/l acetic acid and 0.05 mol/l sodium acetate (82:28). But, the ph and the composition of the test medium may be modified depending on the pka of a given drug substance c) ph 6.8: Use the medium 2 of the dissolution test in the Korean Pharmacopoeia d) Water e) Addition of surfactant to dissolution medium: Use the medium in which the final rate of dissolution is less than 85% in all medium medium indicated in the section i)-iv) of 4-b. 3) Rotational speed For the paddle method, initially use 50 rpm and increase the speed to 75 rpm if the rate of dissolution is slow. The rotational speed may be occasionally set at 100 rpm, but the rotational speed not less than 150 rpm is not used. iv. Selection of dissolution test conditions 1) Conventional release dosage forms Use the testing condition that results in 70-85% of the rate of the dissolution. Since the testing condition showing a rapid rate of dissolution is not generally discriminating, select the dissolution medium and the rotational speed that results in a relatively slow rate of dissolution (not more than 1 hour). 2) Delayed release dosage forms Proceed with the same testing condition in the

19 KP X 1833 section 6-b-iv)-1) and consult to the conditions in the section 5-b. In this case, use the media of ph 1.2 and ph 6.8 as the test media. 3) Prolonged release dosage forms Select at least three time points for sampling such that a point of about 20-30% dissolution of the drug is set as the initial point (1-2 hours), a point of about 50% dissolution of the drug is set as the middle point and a point of about 80% dissolution of the drug is set as the final point. c. The second stage (the dissolution test) i. Proceed with the test conditions selected based on the preliminary test and perform the test ii. Number of test specimen Proceed with the test specimens from three lots having 12 test specimens per each lot and perform the test. iii. Dissolution specification 1) Conventional release and delayed release dosage forms a) Time span of test: For conventional release dosage forms, use not more than 1 hour as the time span of the test. For delayed release dosage forms, use not more than 2 hours in the acid stage and not more than 1 hour in the buffer stage. b) Setting specification of dissolution Set the specification of the dissolution for the test product at the value of about 10% less than the average rate of dissolution at the time point that reaches an almost plateau in the graphical representation of the results for the dissolution test in the lot that shows the medium rate of dissolution among the three lots. Single-point specification This specification is applicable to drug product containing a highly soluble drug substance (e.g., BCS Class I or III). Set the specification as the lower boundary of the dissolution rate at a time point that shows the rate in the range of 70-85%. Two-point specification This specification is applicable to drug product containing gently dissoluble drug substance (e.g., BCS Class II). 15 Minutes is generally used as the first point and 30, 45 or 60 minutes as the second point for the specification. For the drug product in which a rapid dissolution may affect the efficacy or the adverse reaction, or which has a narrow therapeutic index (see the Attachment 2 of the Minimum Requirements for Bioequivalence Test), the specification is set using the two-point testing (if necessary, set both upper and lower boundary values). 2) Prolonged release dosage forms Select at least three time points for sampling such that a point of about 20-30% dissolution of the drug is set as the initial point (1-2 hours), a point of about 50% dissolution of the drug is set as the middle point, and a point of about 80% dissolution of the drug or a point when the rate of dissolution does not change any more is set as a the final point. The rates of dissolution for the initial point and the middle point are set to the average rate of dissolution rate in ± 15% and the rate for the final point being set to the average rate of dissolution in 10%. 3) Established test the dissolution according to the preliminary specification using the reference drug and verify the validity of the specification, including the analytical procedure. 7. Preparation of validation data for dissolution test a. Dissolution medium Briefly summarize the reason for the selection of the dissolution medium. b. Dissolution test procedure i. Describe the procedure for the dissolution test and analysis according to the following items. o Dissolution apparatus o o Preparation of the standard solution Preparation of the test solution: Sampling time of the dissolution solution, dilution, filtration (if there is a filtration procedure involved, present the data for the recovery evaluation) o Analytical procedure for the quantification in the dissolution solution (e.g., HPLC etc.) o Formula for the calculation o Dissolution specification ii. Submit chromatograms or spectra of the blank test solution, the standard solution and the test solution. c. Validation Describe the validation data for the dissolution test and the analytical procedure. i. Data related to the stability of the test solution during the analysis ii. Analytical procedure o Perform the test according to the Guideline of Validation of Analytical Procedures for Pharmaceuticals. - Range : Demonstrate linearity, accuracy and precision within the specification of the dissolution ± 20%. (For example, for a delayed release dosage form having the dissolution specification of not more than 20% at 1 hour, not less than 90% at 9 hours, submit the data demonstrating the accuracy, precision and linearity in the range of 0-110% of the labeled amount). Appendix I Suitability test for dissolution apparatus

20 1834 General Information Suitability test has to be conducted to determine whether a frequently used dissolution apparatus can be properly used and the test is carried out by an adequate test procedure. It is recommended that the suitability test be conducted typically twice a year. In addition, the test has to be performed when the dissolution apparatus is moved or a significant change has been made to the apparatus. Furthermore, the suitability test is required when the test method is modified from the rotary basket method to the paddle method and vice versa. The suitability of a dissolution apparatus for the rotary basket method or for the paddle method is tested using two standard calibrators (USP dissolution calibrator: The test is conducted using one tablet of the non-dissolvable salicylic acid standard tablet and one tablet of the dissolvable prednisone standard tablet. For the case of the salicylic acid tablet, 50 mmol/l phosphate buffer, ph 7.4 is used as the dissolution medium and, for the case of the prednisone tablet, water is used as the medium. 1. Suitability test for dissolution apparatus using the 300 mg salicylic acid standard tablet (lot number O) a. Use 900 ml of 50 mmol/l phosphate buffer as the dissolution medium. In this case, the ph of the buffer is adjusted to 7.4 ± 0.05 in the room temperature. b. To deaerate the dissolution medium, heat the solution to 41 C while stirring, and filter the solution under vacuum using the filter paper of the pore size of 0.45 µm. During the filtration procedure, continue stirring the filtrate. After the filtration procedure, cap the flask and continue the stirring for 5 additional minutes under vacuum. c. Transfer the dissolution medium to the flask of the dissolution apparatus. During the transfer, a special care should be taken not to trap air bubbles in the medium. Allow the dissolution medium to reach an equilibrium temperature of 37 C (it is not recommended that rotate the paddle to facilitate the medium to achieve the equilibrium). d. When the temperature of the test medium reaches at 37 C, initiate the dissolution test using the rotational speed of 100 rpm. e. After 30 minutes of the initiation of the test, sample a portion of the test medium, filter and determine the absorbance at 296 nm and calculate the rate of dissolution (%) of salicylic acid. If analytically necessary, the filtrate may be diluted with the dissolution medium. f. If the rate of dissolution is within the range specified in the table below, the dissolution apparatus is regarded to be suitable. Note) Dissolve a portion of Salicylic acid RS in ethanol to render the concentration of not more than 1%, dilute the solution with the dissolution medium to obtain a series of the standard solutions of known concentration and use the standard solutions for the analysis. In case where the test solution is filtered, test for the potential adsorption of salicylic acid to the filter medium. Table 1. Dissolution specification of the salicylic acid standard tablet (lot number O) Type of dissolution apparatus minutes of the dissolu- Rate of dissolution at 30 tion test (%) Apparatus 1 (the rotary basket method) Apparatus 2 (the paddle method) 2. Suitability test for dissolution apparatus using the 10 mg prednisone standard tablet (lot number N) a. Use 500 ml of water as the dissolution medium. b. To deaerate the dissolution medium, heat the solution to 41 C while stirring, and filter the solution under vacuum using the filter paper of the pore size of 0.45 µm. During the filtration procedure, continue stirring the filtrate. After the filtration procedure, cap the flask and continue the stirring for 5 additional minutes under vacuum. c. Transfer the dissolution medium to the flask of the dissolution apparatus. During the transfer, a special care should be taken not to trap air bubbles in the medium. Allow the dissolution medium to reach an equilibrium temperature of 37 C (it is not recommended that rotate the paddle to facilitate the medium to achieve the equilibrium). d. When the temperature of the test medium reaches at 37 C, initiate the dissolution test at the rotational speed of 50 rpm. e. After 30 minutes of the initiation of the test, sample a portion of the test medium, filter and determine the absorbance at 242 nm and calculate the rate of dissolution (%) of salicylic acid. If analytically necessary, the filtrate may be diluted with the dissolution medium. f. If dissolution rate(%) is within the next table, it meets the requirements of the Suitability test for dissolution apparatus. Note) Dissolve a portion of Prednisone RS in ethanol to render the concentration of not more than 5%, dilute the solution with water to obtain a series of the standard solutions of known concentration and use the standard solutions for the analysis. In case where the test solution is filtered, test for the potential adsorption of prednisone to the filter medium. Table 2. Dissolution specification of the prednisone standard tablet (lot number N) Type of dissolution ap- Rate of dissolution at 30

21 KP X 1835 paratus Apparatus 1 (the rotary basket method) Apparatus 2 (the paddle method) minutes of the dissolution test (%) The values of rate of dissolution indicated tables 1 and 2 do not represent the average rate of dissolution but represents the rate of dissolution for individual 6 or 12 dissolution testing apparatuses. In addition, the specific value for the rate of dissolution changes with the lot number of the standard tablets as indicated in table 3 below. In addition, if the dissolution testing is solely conducted by, for example the paddle method, then the suitability test may be conducted for the paddle method only. Table 3. Specification of suitability test for dissolution apparatus using the prednisone standard tablet (lot number J) and the salicylic acid standard tablet (lot number K) Calibrator Rate of dissolution at 30 minutes of the Prednisone Salicylic acid Apparatus 1 (the rotary basket method) dissolution test (%) Apparatus 2 (the paddle method) 50 rpm 100 rpm 50 rpm 100 rpm Appendix II Deaeration of test medium Deaeration of the test medium is not always necessary for all drug products for the dissolution testing. In case where the identical result is obtained for the test with the deaerated dissolution medium and for that with the non-deaerated dissolution medium, then the deaeration procedure is not necessary. However, under certain cases, the air dissolved in the dissolution medium may participate in a chemical reaction with drug substance, and air bubbles trapped in the medium may affect the dissolution of the drug. For example, for the case of captoril, the air dissolved in the test medium may facilitate the oxidation of captopril thereby creating a stability problem. In addition, a significant amount of air bubbles trapped in the rotary basket generally decreases significantly the rate of dissolution of drugs. Therefore, in these cases, the air dissolved in the test medium has to be eliminated before the initiation of the test. Table 4 lists typical drug products in which the deaeration is required for the dissolution test. Table 4. List of drug products in which deaeration is required for dissolution testing Dosage form Drug substance Tablet Captopril Tablet Tablet Capsule Capsule Tablet Clonazepam Ergotamine tartrate Etoposide Isotretinoin Meprobamate 1. Deaeration method I - While stirring, heat the test medium to the temperature of 41 C and filter the medium under a reduced pressure with the filter paper having the pore size of 0.45 µm. During the filtration procedure, continue the stirring of the filtrate with a magnetic stirrer. - After the filtration procedure, continue the stirring for 5 additional minutes under vacuum (sonication may be used instead of the stirring). - Transfer the test medium to the dissolution apparatus to render the temperature of 37 C. For the case of poorly soluble drugs, the test medium may contain surfactant such as polysorbate 80 or sodium lauryl sulfate. In this case, foam may be formed during the deaeration procedure, especially in the vacuum filtration step. Therefore, it is recommended that the medium, in the absence of surfactant, is first deaerated and, later, added with surfactant for the dissolution testing of poorly soluble drugs or sustained release dosage forms. 2. Deaeration method II The other typical method for the deaeration of the test medium is helium sparging. The main component of air dissolved in the test medium is oxygen and nitrogen. Helium gas, which is more inert and less water soluble than these gases, may be used to purge oxygen and nitrogen from the test medium. In general, it is sufficient that 50 ml/min of the purging rate is sufficient for the initial minutes and then, later, the rate may be reduced to 5-10 ml/min. The deaeration by helium generally takes minutes for a completion. In fact, when helium sparing deaeration is used, it is expected that the relative standard deviation of the rate of dissolution is lowered with the temperature equilibration being achieved sooner. Other deaeration method, involving a sonication of test medium containing surfactant in a reduced pressure, may also be used. Appendix III Sinkers used in dissolution testing When the dissolution is tested using the paddle method, certain dosage forms, for example using capsules, do not settle down at the bottom of the flask but float to the surface of the dissolution solution. In this case, capsules may be placed in a sinker made of wire and the dissolution is tested. In most cases, sinkers are

22 1836 General Information made of wire helix and various products, but including sinker and prolong sinker, may be used as specified in the Dissolution Test. Types of sinker Specification Basket type sinker (2.6 in radius 1.7 cm) 6 helix sinker (typically cm) 4 helix sinker (typically cm) Sinker for controlled release formulation (3.8 1 cm; the size of the clamp of 5 mm 7 mm) 3 extended type sinker from Vankel Sinker specified in the Dissolution Test ( cm) 5. Guideline to Pharmaceutical Quality Control Using Near Infrared (NIR) Spectroscopy I. Near Infrared Spectroscopy Near infrared spectroscopy is a method very useful method for identifying organic compounds by measuring the transmittance or reflectance in a certain wavenumber or wavelength range, when the near infrared light passes through the sample. The range of near infrared is between 750 nm and 2500 nm. Even though the intensity of signal is weaker than that in the mid-infrared region, the appearance of overtones and combinations of the fundamental vibrations of C-H, N-H, O-H, and S-H groups allows identification of materials, qualitative analysis and quantitative analysis. Spectrum is recorded as a graph with x-axis of wavenumbers (or wavelength) and y-axis of transmittance or absorbance, as is in mid-infrared region. When a spectrum acquired with a sample, the spectrum can be influenced by particle size, polymorphism, residual solvents, or humidity and these influences can be removed by applying a proper mathematical pretreatment. 1. Instrument Usually, an instrument consists of a light source compartment, a monochromator compartment, a detector compartment and a display unit. - Light source compartment: a light source providing a whole range or a part of the range of 750 nm to 2500 nm (13333 cm -1 to 4000 cm -1 ), or a laser providing monochromic light is used. - Monochromator compartment: a monochromator such as diffraction grating, filter, or interference filter is used. - Detector compartment: it consists of a detector and a signal processing system. - Display unit: it consists of display, recorder, etc. and mathematical pretreatment on a spectrum is possible. 2. Measurement Proceed as one of the following methods: 1) Transmittance Usually, this method is applied to liquids, diluted or undiluted, and to solids in solution. The sample is placed in an NIR transparent cell with a proper pathlength (usually between 0.5 mm and 4 mm) or a fiber optic probe is inserted into the sample. When recording the spectrum of a liquid sample, any spectral disturbances such as temperature dependent perturbations should be taken into consideration, and in all cases, compensation for background interferences should be made. For example, a reference scan of air (for liquids) or solvent (for solutions) may be subtracted from the sample spectrum. 2) Diffused Reflectance Usually, this method is applied to solids. Sample are analyzed using a proper apparatus. When a fiberopric probe is immersed into the sample to acquire spectrum, a special care should be taken to get a reproducible spectrum for each measurement. In all cases, compensation for background interferences should be made. For example, a reference scan of an internal or external reflectance reference should be subtracted from the sample spectrum. When the measurement is made, the particle size, the state of hydration or of sovation should also be taken into consideration. 3) Transflectance Usually, this method is applied to liquids, diluted or undiluted, to solids in solution or in suspension. The measurement is made with the sample in a cell with a proper diffuse reflector, made of either metal or of an inert material (for example, titanium oxide), not exhibiting a spectrum in the near infrared range and introduced behind the sample. The reflector should show absorbance within the quantitative range. II. Standards for Validation This guideline is based on the contents of the ICH general guideline to the analytical methods and is adapted to allow the acquisition of comprehensive and highly reliable information in applying analytical methods using near infrared radiation, by examining mainly specificity, linearity, range, accuracy and precision, which should be considered in using the near infrared spectroscopy. 1. Specificity

23 KP X 1837 Specificity is an ability to analyze selectively the analyte, despite of probable existence of interfering substances. Generally, a tablet contains lots of substances in addition to the active ingredients. For example, in the Ambroxol Tablet, the content of the principal ingredient, Ambroxol, is 12.5% as 30 mg out of 210 mg of total mass, and the remaining part, 87.5% of the total mass may interfere with the analysis of Ambroxol. These interfering substances are impurities, degredation products, water, residual solvents, etc. and a major part of exceipients. The influence of these substances can be reduced by applying chemometric analytical methods such as the principal component analysis. The specificity can be proved by the following methods. First, acquire spectra of the active ingredients and all of other related substances. If the characteristic wavelengths of the active ingredients do not overlap with the wavelengths of peaks from other substances when the original spectra without any pretreatment are examined, the method can be the easiest one to get the specificity in analyzing the active ingredients. But when this kind of trend is not found with the original untreated spectra, mathematical pretreatment methods or chemometric resolution algorithms can be applied to get the specificity by removing the interferences mathematically. Second, by adding (fortifying the concentration) probable interfering substances to a mixture before tableting or powdered tablet, the interfering effect on the analysis is identified. In other words, it should be examined whether the wavelength showing variation by the concentration of the additives overlaps with the characteristic wavelengths of the active ingredients. Third, in many cases, the contents of the principal ingredient are in the range of not less than 95% and not more than 105% of the labeled amounts. When the linear relationship between the actual values and the estimated values by near infrared spectroscopy of the concentrations outside the allowed range of the labeled amounts such as 90% and 110% can be obtained, this can be the good method to confirm the specificity. Fourth, the most reliable method approaching the specificity is the identification and the qualification of substances. The identification of substances is the comparison of the spectra acquired with the test samples, with the spectra acquired with the (powdered) library test samples having variations caused during the production process. The qualification of substances is the comparison of the mean spectrum of the test samples, with the library spectra. By these procedures, the placebos can be identified. The second derivative operation to spectral data has several advantages for the analysis. The differentiation between the peaks of the active ingredients and probable interfering substances can be achieved, and the background perturbation can be removed. In addition to this, peaks can be sharpened, and any variation caused by the physical states of tablets can be removed, resulting in the increased resolution. Usually, the signal-to-noise ratio is lowered by 2 times by the firstorder derivative operation and by 4 times by the second-order derivative operation. Nonetheless, the stability of recent instruments is excellent and this effect is not a big problem. For the construction of calibration model, the use of more than one wavelength for the analysis can remove the effect caused by the substances other than the active ingredients. An example of chemometric methods is multiple linear regression, which uses more than one wavelength. On the contrary, the methods which use information of a whole range or a certain range of wavelengths are principal component regression, partial least squres regression, and so on. The analysis of the principal components with contents of less than 5% of the tablet is expected to be problematic. Any method for increasing the specificity by minimizing or removing the interference comprising 95% of the tablet may be studied further. 2. Linearity The linearity of analytical process means the ability to get the result directly propotional to the concentration (amount) of analyte in the sample. The linear regression analysis is performed in the range of concentration predictable by the calibration curve of near infrared spectroscopy. Usually, while the calibration curve is obtained from the ratios of peak area in the range of 0% to 150% in the methods of liquid chromatography, the result of near infrared spectroscopy gives the linear relationship between a set of reference values in the narrower range of 90% to 110% found by the conventional methods and a set of estimated values found by near infrared spectroscopy. This linear relationship is evaluated by suitable statistical methods such as calculating the regression line by the least squares method. Data such as correlation coefficient of the regression line, y-intercept, slope and sum of squres of residuals are essential items of the documentation for analyzing the regression data. 3. Range Range is the interval between the maximum and the minimum of concentrations used for measuring linearity, precision and accuracy during the analytical process. The range of analysis is set to 2 times the allowed range, for there are lots of difficulties in analyzing tablets nondestructive and in collecting test samples for calibration. For example, the allowed range of contents of the principal ingredients is given to be 5%, considering the degradation of the principal ingredients during the effective period of use and the error of analytical methods. In this case, the minimum range in analyzing the principal ingredients is 10% of the labeled amount. The more accurate calibration be made if the tablets are collected so that the range is widen to be between 80% and 120% of the labeled amount. In reality, the range suitable for the purpose of analysis is preferred, considering the ranges of 80% to 120% for the Assay of the drug substances or the preparations,

24 1838 General Information 70% to 130% for the Uniformity of Dosage Units and 20% of the range of specification for the Dissolution Test. 4. Accuracy The accuracy of analytical process is the measure expressing the similarity between the values found by the conventional analytical method (with proven accuracy) and the estimated values found by near infrared spectroscopy. The accuracy should be proved in the whole range specified by the analytical method. The accuracy is a parameter evaluated only after establishing precision, linearity and specificity. The accuracy can be evaluated by at least 9 measurements in the specified range. For example, the results measured by manipulating 3 levels of concentration and 3 repeatitions of each concentration level. Three minimum levels of concentrations are selected as the mean value and both of the maximum and the minimum. The accuracy is expressed as the recovery, when the samples spiked with a known level of the analyte are quantitated, or it is examined by comparison between the reference value and the mean value of measurements in the confidence interval, when the comparison with the true value or the certified or agreed value is made. In the quantitative analysis by near infrared spectroscopy, the accuracy is expressed as the standard error of prediction found with the validation set. The calibration set should consist of the samples with the concentrations of the active ingredients produced in the batch process and it should include the whole range of the sample. The selected samples should show the uniform distribution of probability, not the normal distribution. The validation set should consist of the samples with the same composition of the active ingredients as in the calibration set. The set of samples for the parallel test are selected from the independent production batch and the analysis is validated by the time. Therefore, the accuracy of the analysis can be examined using the set completely independent from the calibration set and the validation set of samples. This test is performed once a month with independent set of samples after development of the calibration model. 5. Precision The precision of analytical process means the similarity among the results obtained by a series of repeated analysis with the same sample under the specified analytical conditions. The items of precision are repeatability, within-laboratory reproducibility and interlaboratory reproducibility. The precision is tested with homogeneous and authentic samples. When the homogeneous and authentic samples are not easily available, the test can be done with artificially synthesized samples or solutions. The precision is expressed as the relative standard deviation or the coefficient of variation calculated from a series of measurements. Repeatability is set to not more than 1.0% as the relative standard deviation (or CV) by measuring and performing the whole procedure 6 times in a short period time, as long as there is no thermal degradation of the samples with concentration equivalent to 100% of the tested concentration. The sample homogeneity and the surface homogeneity should be taken into more consideration especially when the contents of the principal ingredients are low. In this case, the transmittance is better than the reflectance, for the area of transmittance is wider than that of reflectance. Meanwhile, the transmittance may introduce more noise when the tablets are thick and the light intensity reaching the detector is low. The representative factors of variation which need examination in within-laboratory reproducibility are the date of experiment, the experimenter, and the instrument used and so on. The inter-laboratory reproducibility is evaluated when the analytical method is needed to be standardized for the collaborative experiments between laboratories. If the inter-laboratory reproducibility is proven, the within-laboratory reproducibility is not needed to be validated, but usually, the within-laboratory reproducibility is mainly evaluated, for the inter-laboratory reproducibility is hard to be achieved. III. Qualification of NIR Spectrophotometers The purpose of qualification of NIR spectrophotometers is to verify the suitability of the instrument for the intended use by comparing with the specification of the instrument. The qualification procedure includes Design Qualification (DQ), Installation Qualification (IQ), Operational Qualification (OQ) and Performance Qualification (PQ). 1. Design Qualification (DQ) The design qualification provides the evidence whether the instrument is properly designed to be operational for the intended use. The various influencing factors of the instrument are tested to verify whether they correspond to the specifications of the instrument. At least, it should be verified whether the factors correspond the manufacturer s specifications of the instrument. 2. Installation Qualification (IQ) The installation qualification verifies whether the instrument is installed according to the design and the expressed specifications. Documentation of the serial number of the hardware, the version of the software and so on is included in the verification. And the environment and the facilities of instrument installation are are verified and the status of assembly, electric power of the instrument and so on are also verified. 3. Operational Qualification (OQ) The operational qualification verifies the instrument once again by the methods used to select the spectrophotometer, such as wavelength accuracy and precision, linearity of the monochromator and noise.

25 KP X Performance Qualification (PQ) The performance qualification verifies whether the instrument continues to be operational. Parts of the OQ items are needed to be applied to verify the instrument before the use or periodically (at least once every 6 months). The PQ should be performed when the maintenance of the instrument is done or the lamp is replaced. IV. Quantitative Analysis In the pharmaceutical industries, the quality of pharmaceuticals are assessed by the repeatition of experiments for several times with the drug substances and the final pharmaceutical preparations. Nonetheless, the conventional analytical methods for the quality control cannot be applied to all of the produced pharmaceuticals, for those methods are usually destructive. Therefore, the weakness that only the representative samples are managed can be a critical obstacle in the matter of safety. Several principles of the analytical process are needed for the optimal maintenance process. The accuracy and the precision are necessary and the sample preparation process is not required or is minimal. Moreover, the simultaneous analysis of many analytes should be possible and the rapid adjustment during the production process should be possible. The near infrared spectroscopy is one of the methods satisfying these principles. The near infrared spectroscopy allows non-destructive, real-time total inspection of tablets during the production process. The nondestructive quantitative analysis of the tablets is to be mentioned among the quantitative analyses of various preparations and the general procedure is as follows: Feasibility test Reference data Selection of the samples Acquisition of sample spectra Construction of calibration model Validation of calibration model Validation of model performance 1. Feasibility test The feasibility test is a procedure to check whether a drug material or a pharmaceutical product can be analyzed quantitatively by NIR spectroscopy. The first step is to observe the spectral response at the used level of concentration of the analyte. After acquiring and examining spectra of all the ingredients of a tablet, if there are no significant differences among the ingredients, the second derivatives may be used to identify characteristic peaks of each ingredient from the derivatized spectra. In addition to this, when the spectra are acquired at the different level of concentrations of samples, any spectral changes caused by the concentrationn are checked with the second derivatives of those spectra. The feasibility test is also needed for the selection of algorithm for the calibration. When the principal peaks of the active ingredients do not overlap with those of the excipients, the multiple linear regression can be used. When there is a significant interference, the pricipal component regression or the partial least squares regression is more effective. Items that are needed to be checked are whether to use transmittance or reflectance, the use of fiber-optics, whether there are reproducible results by repeated measurements of the sample, whether there is robustness in the measurement of the sample, and so on. 2. Reference data The quantitative method by NIR spectroscopy is based on the comparison of the reference data with acquired spectral data. Therefore, if the reference data are not correct, the estimated values by NIR spectroscopy cannot be correct either. The reference data may be obtained by the preparation of real reference standards or the application of reference analytical methods, i.e. conventional analytical methods used for the evaluation, such as gravimetry, chromatography, spectroscopy, and so on, of which accuracy is proven. The linearity, accuracy and precision of the reference analytical methods should be verified. The environmental error can be minimized by concurrent use of the NIR spectroscopy and the reference analytical method. Three repeated measurements with the same sample by the reference analytical method are made to get the reference data 3. Selection of the samples i) Calibration Test Set The calibration test set is the standard for the quantification of the NIR spectral responses against the reference data. The sample should be selected carefully so that the calibration test set includes the whole range of possible concentrations of the analytes, and there is robustness over the variation in the concentration of the excipients, and the maximum variation of the analysis is included. As other analytical method, interpolation within the range of the calibration test set is possible with the NIR spectroscopy, but extrapolation is not possible. ii) Validation Test Set The construction of validation test set is the first step of the validation. The validation test set is used to optimize the model and the concentration used should be within the range of the calibration test set. If any variation outside the calibration test set is included, the correct validation values can be obtained. Partitioning of the calibration test set and the validation test set is done as follows. With concurrent experiments, the calibration test set and the validation test set have the same instrumental error and the environmental error.

26 1840 General Information The test set is selected manually or by software method so that the composition of the analytes is uniformly distributed. The test set can be selected on the basis of the spectral variation by software method. Whichever method is used, the care should be taken to have uniform distribution by the concentration range. When the distribution is not uniform (for example, concentrations at the center are used), there should be a proper explanation. The sample size measurable by NIR spectroscopy is significantly smaller than those by the conventional method. This is not because of the sample measurement compartment but because of the sample area illuminated by the near infrared light. The infrared measurement can detect the inhomogeneity even with the mass of microgram level and a proper measurement method is needed to accommodate this variation. This property can be an advantage useful for testing the Uniformity of Dosage Units, but in many applications, averaging the signal from a larger area is needed. Therefore, to get the averaged information, the sample is moved transversely or rotated during the data acquisition. Among the various preparations, in cases of the tablets, there are two cases: the case of using the products as they are and the case of varying ratios of the active ingredients to the excipients. (1) The case of using the products as they are First, this can be applied only when the transmittance follows the Beer-Lambert s law and it also applied to other tablets of the same composition with different masses. Second, this can be applied when the samples are collected from the normal production and each tablet has the concentration range needed for the usage. (2) The case of varying ratios of the active ingredients to the excipients First, the calibration test set is obtained by mixing the products and the concentration expansion samples. Usually, the products lie between 95% and 105% of the labeled amount, so to expand this range, tablets with expanded concentration range are synthesized in the laboratory. This concentration expansion samples have different ratios of the active ingredients to the excipients and the ranges of between 90% and 110% of the labeled amount or between 85% and 115% of the labeled amount are used. Second, the concentration is adjusted by adding the active ingredients and the excipients to the powdered products. If the active ingredients are low compared to the total mass of the tablet, the better calibration results can be acquired by adding the active ingredients to the powdered tablets. Third, the concentration of the active ingredients is varied using the sample synthesized in the laboratory. This is similar to the use of the powdered tablet, but the concentration range is wider (80% to 120%), for the amount of the active ingredients can be varied from the first stage of the sample preparation. Fourth, the active ingredients and the excipients are added to the powdered tablets and then the mixture is tableted. But in this case, the total mass of the tablet is different from that of the products and it may fail in terms of specificity. Fifth, the model comprising all of the variation factor of the production process by using the global calibration set, which includes all of the existing variation factors to make the calibration models. This method consists of 4 steps: laboratory composite (prepared to have active ingredients with the concentration level of 15% of the labeled amount), preparation of granules, tableting of core tablet, and coating the tablets. 4. Acquisition of sample spectra The transmittance is applied to liquids, diluted or undiluted and solids in the solution. The cells with transmittance pathlength of 0.5 to 4 mm or dip-probes are used and the compensation for background interferences should be made. On the contrary, the reflectance and the diffused reflectance are applied mainly to solids and the test is done to the sample placed in a proper apparatus. When the fiber-optics are used, reproducible spectra can be acquired by fixing the sample properly. As in the transmittance, compensation for background interferences should be made. In cases of solid particles, the physical differences such as particle size, shape, compressibility, etc. can be influential, so the particles are manipulated to be as small and uniform in size as possible. Moreover, to make the effect of the absorbed water and the residual solvent constant, the samples are dried for a certain length of time to have the same condition of the absorbed water and the residual solvent of the samples. When the samples are analyzed by the NIR spectroscopy, the NIR spectra are influenced a lot by the absorbed water and the temperature, so it is desirable to construct the environment of constant humidity and constant temperature so that the experiment can be done in the same environment. Moreover, the difference in the polymorphism of the sample or in the degree of crystalization also can be influential to the quantitative analysis, so the case should be taken. After selecting a proper sample measurement method among various sample measurement apparatus by considering the property, the shape, and the size of the sample, the precision of the spectra acquired by repeated measurement is evaluated. When the measurement is repeated 6 times, it meets the requirement if the maximum value of the relative standard deviations of whole range of wavelengths is not more than 1.0%. 5. Construction of calibration model For the development of a calibration model, a proper mathematical preptreatment is needed to be done, if necessary. The pretreatment procedure is an important

27 KP X 1841 step in chemometric analysis of the NIR data. The pretreatment is defined as a procedure in which the spectral shape is enhanced by mathematical treatment of the NIR data or unwanted variation is removed prior to the development of a calibration model. A proper pretreatment method can be selected by testing the spectral data prior to the data modeling. For example, various methods of pretreatment are used in parallel and evaluated and the best one is selected among them. There are various methods of pretreatment. For example, they are normalization, smoothing, baseline correction, derivatives, mean centering, variance scaling, autoscaling and so on. After selecting a proper method of pretreatment, regression analysis on the NIR data is done by applying various quantitative algorithms. The calibration is a procedure by which a mathematical model is constructed between the instrumental responses and the properties of the sample (usually concentration). As mentioned previously, the major algorithms for calibration are multiple linear regression (MLR), principal component regression (PCR), partial least squares regression (PLSR) and so on. The prediction is a procedure of predicting the properties of the samples from the instrumental signals using the developed model. In a broad sense, there are two different approaches to develop the calibration model and they are the univariate analysis and the multivariate analysis. The univariate analysis is the most commonly used method in the conventional analysis, which relates the single signal of the analytical instrument is related to the concentration of the single ingredient. This method is not commonly used in the NIR spectroscopy. In the NIR spectroscopy, many signals from the analytical instrument are related to the various properties of the samples and the calibration model is constructed by the multivariate analytical methods. * For constructing a quantitative model, refer to the Appendix I. The Appendix I explains mainly MLR, PCR and PLSR, which are used generally in the NIR spectroscopy. 6. Validation of calibration model The quantitative model can be validated internally or externally. Independent validation test set is used to get the information on the predictability of the developed model. The accuracy and the precision of the NIR spectroscopy is compared with those of the reference analytical method. Standard error of calibration (SEC) and standard error of prediction (SEP) are used to evaluate the quality of the quantitative model. A conventional factor of model evaluation, the correlation coefficient of regression is also used in the NIR spectroscopy, but it is not as important as in the conventional univariate analysis. When the content of a tablet is analyzed, the ratio (%) of the amount of analyte to the mass of the tablet is used as the value of concentration. When the calculated SEP value derived as the relative error obtained by converting the amount (mg) of the analyte to 100 is not exceeding the error range of the conventional method of analysis, the use of the model is recommended. 7. Validation of model performance Not just to evaluate simply the model, the performance of the model is needed to be validated periodically to use that model continuously in the industrial fields. The methods for validating model performance can be classified into two methods, which are the method of using the check samples and the method of comparing with the reference analytical method. First, to use the check samples, the check samples should be stable over time and the short-term and long-term accuracy of the model should be evaluated with these samples. Second, to use the method of comparing with the reference analytical method, the data by the NIR spectroscopy and the reference data are acquired periodically for n months (n=1,2,3, ) over n batches and a paired t-test with the confidence level of 95% is performed. When there is significant difference, the quantitative model should be reconstructed. * Refer to the Appendix II for the validation of the model. * Refer to the Appendix III for the usual quantitative analysis and the maintenance of the quantitative model. V. Qualitative Analysis The near infrared spectroscopy can be used for the identification of substances and the qualitative analysis. Identification : It is used when the chemical identification of the substance is needed. Qualitative analysis : After chemical identification of the substance, the suitability of the sample to the model of substances is measured. The model is developed from the samples representing various information on the variation and it includes water, particle size, solvent and other chemical and physical informations. Both of the identification and the qualitative analysis enable the discrimination of substances in the library. Typical qualitative application of the NIR spectroscopy consists of the following procedures. Feasibility test Selection of the samples Acquisition of sample spectra Construction of library Validation of library Routine use Maintenance of library 1. Feasibility test The feasibility test is performed as a first step, prior to the development of model. For example, optimal method of sample measurement, amount of the sample, the minimum number of scans for effective analysis are

28 1842 General Information examined. A prior knowledge on the organization of samples in the library and the molecular structure is useful in the analysis with the NIR spectroscopy. And when the spectra of the representative samples are acquired, it is checked whether the second derivative spectra of the samples are different from each other. 2. Selection of the samples The selection of the samples is an important procedure for the successful qualitative analysis. The sample set for developing the library and the independent set for the purpose of validation are needed. All of the samples used for building and validating the library are needed to be authenticated. The level of authentication of the sample is varied by the usage and the database consists of the samples showing various source of variation. The samples from the different batches are collected, reflecting the changes in ingredients, suppliers, processes, and storage conditions over a certain period of time. If the chemical and physical stability is proven over the storage period, the samples can be collected regardless of the storage status. The number of batches to be collected is dependent on the complexity of analysis and the substances to be analyzed should include the typical variation of the system. The number of batches for the qualitative analysis is greater than that for the identification. There are many kinds of the sample measurement apparatus, such as cups, vials, fiber-optics and custom-made apparatus. The selection of the apparatus is dependent on the user s needs and the validation of the apparatus is defined in the stage of the design qualification (DQ). The sample measurement apparatus is the source of latent variations during the measurement. Therefore, it should be validated continuously in terms of reproducibility, if possible. 3. Acquisition of Sample Spectra The transmittance is applied to liquids, diluted or undiluted and solids in the solution. The cells with transmittance pathlength of 0.5 to 4 mm or dip-probes are used and the compensation for background interferences should be made. On the contrary, the reflectance and the diffused reflectance are applied mainly to solids and the test is done to the sample placed in a proper apparatus. When the fiber-optics are used, reproducible spectra can be acquired by fixing the sample properly. As in the transmittance, compensation for background interferences should be made. In cases of solid particles, the physical differences such as particle size, shape, compressibility, etc. can be influential, so the particles are manipulated to be as small and uniform in size as possible. Moreover, to make the effect of the absorbed water and the residual solvent constant, the samples are dried for a certain length of time to have the same condition of the absorbed water and the residual solvent of the samples. When the samples are analyzed by the NIR spectroscopy, the NIR spectra are influenced a lot by the absorbed water and the temperature, so it is desirable to construct the environment of constant humidity and constant temperature so that the experiment can be done in the same environment. Moreover, the difference in the polymorphism of the sample or in the degree of crystalization also can be influential to the quantitative analysis, so the case should be taken. After selecting a proper sample measurement method among various sample measurement apparatus by considering the property, the shape, and the size of the sample, the precision of the spectra acquired by repeated measurement is evaluated. When the measurement is repeated 6 times, it meets the requirement if the maximum value of the relative standard deviations of whole range of wavelengths is not more than 1.0%. 4. Construction of Library The following procedures are used to develop the library. Definition of the purpose of developing the library Selection of the sample set for developing the library Data display Selection of the sample set for validating the library Data pretreatment/conversion Construction of the library Setting the threshold 1) Definition of the purpose of developing the library Prior to the development of the library, it is important to set the effective range of the library according to the intended use. This applies to the identification and the qualitative analysis of the substances and it includes checking the chemical similarity of the groups to be analyzed and the number of groups. 2) Selection of the sample set for developing the library For the development of the library, the variation caused by the following factors should be considered. The factors, especially important for the library for the qualitative analysis are as follows: Moisture Particle size Residual solvents Degradation products Compositional change of formulated product Other chemical/physical properties Time Alternative sources of material Retained samples Temperature Operator Presentation, e.g. insertion of tube Between-instrument variation Others

29 KP X 1843 The range of consideration over these factors is dependent on the intended use and the needed ability of classification. 3) Data display The data display checks visually the sample showing strange spectral features and identifies outliers. If possible, outliers should be identified and if there is reasonable analytical explanation, they can be removed and in that case, the clear reason should be documented. 4) Selection of the sample set for validating the library There can be a situation, in which the representative samples are needed to be selected from the large group. In the simple situation, they can be selected manually, but in more complex situation, they should be selected after the determination of substance groups using a proper selection tool (for example, principal component analysis, cluster analysis, etc.). The number of samples for each of the substance groups is dependent on the qualitative algorithm and the state of the system (how accurately the boundaries of groups needed to be set). 5) Data pretreatment/conversion Data are needed to be pretreated mathematically to simplify the spectra more. For example, the algorithms of derivatives and scatter correction can remove the background variation caused by the physical difference. The original untreated spectra are used when the information due to the physical status is important. Mathematical transformation of spectra data can be an artifact and can cause loss of important information, so this should be taken into consideration. The algorithms of data pretreatment and conversion should be understood correctly and the theoretical basis always should be provided whenever any conversion is performed. 6) Construction of the library The structure of the library is dependent on the limitation of the software and the user s needs. In the simplest case, all of the substances are combined into one library. On the contrary, it can be divided into sublibraries to secure the needed level of the specificity. The mathematical transformation procedure may be same for all of the substance groups in the principal library. The transformation may be same within the sub-library but different between the sub-libraries. For example, this situation is met when the grade of the lactose is to be determined after the sample is identified as lactose from the library of excipients. The spectral range can be all of the available wavelengths or can be limited to cetain wavelengths. The range of wavelengths can be limited when a certain measurement apparatus is used or unnecessary spectral information is needed to be removed (outside the dynamic range or regions of high noise level). An analytical method of partitioning the wavelength range is useful in removing unwanted effects or in bringing a small but important difference into relief. Like any other analytical techniques, the NIR spectroscopy cannot differentiate all of the substance groups, especially very similar groups. In this case, two groups are combined to one group or any other control method is used to perform the identification and qualitative analysis. There are many algorithms such as correlation, soft independent modeling of class analogy (SIMCA), Mahalanobis distance, SMV and so on. The user selects a proper algorithm with the consideration of the effective range of the library. When the discrimination of the substances is each, it is recommended to use the simplest algorithm. For example, when the purpose is only the identification of the substance, the physical factors are not needed to be considered, so the second derivatives and the wavelength correlation algorithm can be selected. 7) Setting the threshold The internal validation is done with the value set by the software itself or recommended by the manufacturer. The threshold of the library can be changed when the library is validated internally or the sample is evaluated externally. 5. Validation of Library The purpose of validation of an analytical procedure is to check whether the analysis is suitable for the intended use. Upon this purpose, the factors influencing the needed validation should be determined. 1) Internal validation The performance of the library should be evaluated in the course of constructing any kind of spectral database. This is based on the samples constructing the library (whether these samples can be differenctiated). The internal validation is done by the software and the detailed procedure may be different by the used software, but the basic procedure is as follows: The spectra used to construct the library are validated by a proper method (correlation or distance method) It is checked whether the distribution of the substances in the library overlaps. The cross-validation is used for the construction of the library. 2) External validation When the internal validation is successful, the performance of the database is validated using certified samples not used for the database development. Reproducibility: This is not commonly applied method for the identification of substances. In the reproducibility test for the qualitative analysis, the substances included in the library should be reliably dif-

30 1844 General Information ferentiated from the substances outside the library, using the threshold value. Robustness: This category is dependent on the application and the sample selection technique and the effect of delicate variation in the normal operating condition of the analysis is tested. The use of the design of experiment can maximize the effect of the information on the analyte. The following items are considered: The effect of environmental conditions (temperature, humidity) of the analysis The effect of temperature on the sample The location of sample The depth of probe and the degree of compression/filling of the substance Other effect of the sample measurement apparatus The effect of replacement of parts (lamp) The effect of pretreatment and parameters of the algorithm used to construct the library (gap/segment of the derivative operation, distance threshols, etc.) 6. Routine use When the system is evaluated, it is adjusted to allow the needed functions only. For example, the manager of the NIR system or the library development expert should evaluate the software thoroughly, but general users are required to evaluate only the performance of the general identification of substances In the aspect of developing the library for the spectral identification, the considerable part of the characteristic variation within the substances are to be included. But sometimes, this variation may not be included in the sample set for the development of the library. For example, when a test substance is analyzed to be outside the boundary of one of the models in the library, the test substance is expressed as to be inadequate NIR spectrum for the NIR spectral model. In this case, a proper alternative test is performed and the test substance is certified correctly before making decision that the substance is within the model or inserting the spectrum into the library. * Refer to the Appendix IV for the construction of the library for the qualitative analysis. 7. Maintenance of library 1) Removal of existing substances In normal situations, the removal of substances from the library is not recommended, usually, but when the selection of the samples is proven to be wrong, the removal from the library is made and the library should be evaluated once again. To add new substances into the library, the test set of substances is selected to satisfy the details of sample selection parameters and the library is evaluated again for the validation of the specificity, continuously. 3) Fixing the groups in the library Sometimes, the sample set is fixed in the cases as follows: Change in physical properties of the substance Change in supplier Inclusion of the wider range In each case, new samples are certified by the method other than the NIR spectroscopy, before inserting them in the model and the library is evaluated again for the validation of the specificity, continuously. * Refer to the Appendix 5 when there is any problem in applying the library for the qualitative analysis. VI. Terms Used Calibration: a procedure of construction of a quantitative model Prediction: a procedure of estimating the concentration of unknown sample, using the constructed quantitative model Calibration set: a set of samples used for constructing the quantitative model Validation test set or validation set: a set of samples used for validating the quantitative model Multiplicative Scatter Correction: one of the mathematical methods used for the correction of background variation caused by the physical properties such as particle size of the sample Multiple Linear Regression (MLR): quantitative model constructed with absorbance values at two or more wavelengths. Principal Component Regression (PCR): one of the multivariate regression analysis using principal component analysis, or factor analysis. Partial Least Squares Regression (PLSR): one of the multivariate regression analysis using principal component analysis to which the concentration information of the samples are included. Soft Independent Modeling by Class Analogy (SIMCA): a pattern recognition algorithm used for the identification and the qualitative analysis of the samples. NIST SRM: National Institute of Standards and Technology Standard Reference Material 2) Addition of new substances

31 Appendix I. Selection of an algorithm for the construction of quantitative model. KP X 1845

32 1846 General Information Appendix II. Validation and Evaluation of a Quantitative Model

33 KP X 1847 Appendix III. Routine procedure of quantitative analysis and Maintenance of the quantitative model.

34 1848 General Information Appendix IV. Construction of library needed for qualitative analysis

35 Appendix V. In case of failure of the library for qualitative analysis KP X 1849

36 1850 General Information 6. Guideline of Limits for Residual Solvents of Pharmaceuticals 1. INTRODUCTION The objective of this guideline is to recommend acceptable amounts for residual solvents in pharmaceuticals for the safety of the patient. The guideline recommends use of less toxic solvents and describes levels considered to be toxicologically acceptable for some residual solvents. Residual solvents in pharmaceuticals are defined here as organic volatile chemicals that are used or produced in the manufacture of drug substances or excipients, or in the preparation of drug products. The solvents are not completely removed by practical manufacturing techniques. Appropriate selection of the solvent for the synthesis of drug substance may enhance the yield, or determine characteristics such as crystal form, purity, and solubility. Therefore, the solvent may sometimes be a critical parameter in the synthetic process. This guideline does not address solvents deliberately used as excipients nor does it address solvates. However, the content of solvents in such products should be evaluated and justified. Since there is no therapeutic benefit from residual solvents, all residual solvents should be removed to the extent possible to meet product specifications, good manufacturing practices, or other quality-based requirements. Drug products should contain no higher levels of residual solvents than can be supported by safety data. Some solvents that are known to cause unacceptable toxicities (Class 1, Table 1) should be avoided in the production of drug substances, excipients, or drug products unless their use can be strongly justified in a risk-benefit assessment. Some solvents associated with less severe toxicity (Class 2, Table 2) should be limited in order to protect patients from potential adverse effects. Ideally, less toxic solvents (Class 3, Table 3) should be used where practical. The complete list of solvents included in this guideline is given in Appendix 1. The lists are not exhaustive and other solvents can be used and later added to the lists. Recommended limits of Class 1 and 2 solvents or classification of solvents may change as new safety data becomes available. Supporting safety data in a marketing application for a new drug product containing a new solvent may be based on concepts in this guideline or the concept of qualification of impurities as expressed in the guideline for drug substance (Q3A, Impurities in New Drug Substances) or drug product (Q3B, Impurities in New Drug Products), or all three guidelines. 2. SCOPE OF THE GUIDELINE Residual solvents in drug substances, excipients, and in drug products are within the scope of this guideline. Therefore, testing should be performed for residual solvents when production or purification processes are known to result in the presence of such solvents. It is only necessary to test for solvents that are used or produced in the manufacture or purification of drug substances, excipients, or drug product. Although manufacturers may choose to test the drug product, a cumulative method may be used to calculate the residual solvent levels in the drug product from the levels in the ingredients used to produce the drug product. If the calculation results in a level equal to or below that recommended in this guideline, no testing of the drug product for residual solvents need be considered. If, however, the calculated level is above the recommended level, the drug product should be tested to ascertain whether the formulation process has reduced the relevant solvent level to within the acceptable amount. Drug product should also be tested if a solvent is used during its manufacture. This guideline does not apply to potential new drug substances, excipients, or drug products used during the clinical research stages of development, nor does it apply to existing marketed drug products. The guideline applies to all dosage forms and route of administration. Higher levels of residual solvents may be acceptable in certain cases such as short term (30 days or less) or topical application. Justification for these levels should be made on a case by case basis. See Appendix 2 for additional background information related to residual solvents. 3. GENERAL PRINCIPLES 3.1 Classification of Residual Solvents by Risk Assessment The term tolerable daily intake (TDI) is used by the International Program on Chemical Safety (IPCS) to describe exposure limits of toxic chemicals and acceptable daily intake (ADI) is used by the World Health Organization (WHO) and other international health authorities and institutions. The new term permitted daily exposure (PDE) is defined in the present guideline as a pharmaceutically acceptable intake of residual solvents to avoid confusion of differing values for ADI's of the same substance. Residual solvents assessed in this guideline are listed in Appendix 1 by common names and structures. They were evaluated for their possible risk to human health and placed into one of three classes as follows: Class 1 solvents: Solvents to be avoided Known human carcinogens or strongly suspected human carcinogens, and environmental hazards. Class 2 solvents: Solvents to be limited Non-genotoxic animal carcinogens or possible causative agents of other irreversible toxicity such as neurotoxicity or teratogenecity. Solvents suspected of other significant but reversible toxicities.

37 KP X 1851 Class 3 solvents: Solvents with low toxic potential Solvents with low toxic potential to man; no health-based exposure limit is needed. Class 3 solvents have PDEs of 50 mg or more per day. 3.2 Methods for Establishing Exposure Limits The method used to establish permitted daily exposures for residual solvents is presented in Appendix 3. Summaries of the toxicity data that were used to establish limits are published in Pharmeuropa, Vol. 9, No. 1, Supplement, April Options for Setting Limits of Class 2 Solvents Two options are available when setting limits for Class 2 solvents. Option 1: The concentration limits in ppm stated in Table 2 can be used. They were calculated using equation (1) below by assuming a product mass of 10 g administered daily. Concentration (ppm) = 1000 PDE (1) dose Here, PDE is given in terms of mg/day and dose is given in g/day. These limits are considered acceptable for all substances, excipients, or products. Therefore this option may be applied if the daily dose is not known or fixed. If all excipients and drug substances in a formulation meet the limits given in Option 1, then these components may be used in any proportion. No further calculation is necessary provided the daily dose does not exceed 10 g. Products that are administered in doses greater than 10 g per day should be considered under Option 2. Option 2: It is not considered necessary for each component of the drug product to comply with the limits given in Option 1. The PDE in terms of mg/day as stated in Table 2 can be used with the known maximum daily dose and equation (1) above to determine the concentration of residual solvent allowed in drug product. Such limits are considered acceptable provided that it has been demonstrated that the residual solvent has been reduced to the practical minimum. The limits should be realistic in relation to analytical precision, manufacturing capability, reasonable variation in the manufacturing process, and the limits should reflect contemporary manufacturing standards. Option 2 may be applied by adding the amounts of a residual solvent present in each of the components of the drug product. The sum of the amounts of solvent per day should be less than that given by the PDE. Consider an example of the use of Option 1 and Option 2 applied to acetonitrile in a drug product. The permitted daily exposure to acetonitrile is 4.1 mg/day; thus, the option 1 limit is 410 ppm. The maximum administered daily mass of a drug product is 5.0 g, and the drug product contains two excipients. The composition of the drug product and the calculated maximum content of residual acetonitrile are given in the following table. Component Excipient 1 meets the Option 1 limit, but the drug substance, excipient 2, and drug product do not meet the Option 1 limit. Nevertheless, the product meets the Option 2 limit of 4.1 mg per day and thus conforms to the recommendations in this guideline. Consider another example using acetonitrile as residual solvent. The maximum administered daily mass of a drug product is 5.0 g, and the drug product contains two excipients. The composition of the drug product and the calculated maximum content of residual acetonitrile are given in the following table. Component Amount in formulation Amount in formulation Acetonitrile content Acetonitrile content Daily exposure Drug substance 0.3 g 800 ppm 0.24 mg Excipient g 400 ppm 0.36 mg Excipient g 800 ppm 3.04 mg Drug Product 5.0 g 728 ppm 3.64 mg Daily exposure Drug substance 0.3 g 800 ppm 0.24 mg Excipient g 2,000 ppm 1.80 mg Excipient g 800 ppm 3.04 mg Drug Product 5.0 g 1,016 ppm 5.08 mg In this example, the product meets neither the Option 1 nor the Option 2 limit according to this summation. The manufacturer could test the drug product to determine if the formulation process reduced the level of acetonitrile. If the level of acetonitrile was not reduced during formulation to the allowed limit, then the manufacturer of the drug product should take other steps to reduce the amount of acetonitrile in the drug product. If all of these steps fail to reduce the level of residual solvent, in exceptional cases the manufacturer could provide a summary of efforts made to reduce the solvent level to meet the guideline value, and provide a risk-benefit analysis to support allowing the product to be utilized with residual solvent at a higher level. 3.4 Analytical Procedures Residual solvents are typically determined using chromatographic techniques such as gas chromatography. Harmonized procedures for determining levels

38 1852 General Information of residual solvents as described in the pharmacopoeias should be used, if feasible. Otherwise, manufacturers would be free to select the most appropriate validated analytical procedure for a particular application. If only Class 3 solvents are present, a non-specific method such as loss on drying may be used. Validation of methods for residual solvents should conform to the ICH guidelines Text on Validation of Analytical Procedures and Extension of the ICH Text on Validation of Analytical Procedures. 3.5 Reporting levels of residual solvents Manufacturers of pharmaceutical products need certain information about the content of residual solvents in excipients or drug substances in order to meet the criteria of this guideline. The following statements are given as acceptable examples of the information that could be provided from a supplier of excipients or drug substances to a pharmaceutical manufacturer. The supplier might choose one of the following as appropriate: - Only Class 3 solvents are likely to be present. Loss on drying is less than 0.5%. - Only Class 2 solvents X, Y,... are likely to be present. All are below the Option 1 limit. (Here the supplier would name the Class 2 solvents represented by X, Y,...) - Only Class 2 solvents X, Y,... and Class 3 solvents are likely to be present. Residual Class 2 solvents are below the Option 1 limit and residual Class 3 solvents are below 0.5%. If Class 1 solvents are likely to be present, they should be identified and quantified. "Likely to be present" refers to the solvent used in the final manufacturing step and to solvents that are used in earlier manufacturing steps and not removed consistently by a validated process. If solvents of Class 2 or Class 3 are present at greater than their Option 1 limits or 0.5%, respectively, they should be identified and quantified. 4. LIMITS OF RESIDUAL SOLVENTS 4.1 Solvents to Be Avoided Solvents in Class 1 should not be employed in the manufacture of drug substances, excipients, and drug products because of their unacceptable toxicity or their deleterious environmental effect. However, if their use is unavoidable in order to produce a drug product with a significant therapeutic advance, then their levels should be restricted as shown in Table 1, unless otherwise justified. 1,1,1-Trichloroethane is included in Table 1 because it is an environmental hazard. The stated limit of 1500 ppm is based on a review of the safety data. Table 1. Class 1 solvents (solvents that should not be used in manufacture of pharmaceutical products). Solvent Concentration limit (ppm) Note Benzene 2 Carcinogen Toxic and Carbon tetrachloridmental environ- 4 hazard 1,2-Dichloroethane 5 Toxic 1,1-Dichloroethene 8 Toxic Environmental 1,1, Trichloroethane hazard 4.2 Solvents to Be Limited in Pharmaceutical Products Solvents in Table 2 should be limited in pharmaceutical products because of their inherent toxicity. PDEs are given to the nearest 0.1 mg/day, and concentrations are given to the nearest 10 ppm. The stated values do not reflect the necessary analytical precision of determination. Precision should be determined as part of the validation of the method. Table 2. Class 2 solvents in pharmaceutical products. Solvent PDE Concentration (mg/day) limit (ppm) Acetonitrile Chlorobenzene Chloroform Cumene Cyclohexane ,2-Dichloroethene Dichloromethane ,2- Dimethoxyethane N,N- Dimethylacetamide N,N- Dimethylformamide ,4-Dioxane Ethoxyethanol Ethyleneglycol Formamide Hexane Methanol Methoxyethanol Methylbutyl ketone Methylcyclohexane N-Methylpyrrolidone Nitromethane Pyridine Sulfolane Tetrahydrofuran Tetralin Toluene ,1,2- Trichloroethene Xylene note 1) note 1) usually 60% m-xylene, 14% p-xylene, 9% o-

39 KP X 1853 xylene with 17% ethyl benzene Isopropyl ether Trifluoroacetic acid 4.3 Solvents with Low Toxic Potential Solvents in Class 3 (shown in Table 3) may be regarded as less toxic and of lower risk to human health. Class 3 includes no solvent known as a human health hazard at levels normally accepted in pharmaceuticals. However, there are no long-term toxicity or carcinogenicity studies for many of the solvents in Class 3. Available data indicate that they are less toxic in acute or short-term studies and negative in genotoxicity studies. It is considered that amounts of these residual solvents of 50 mg per day or less (corresponding to 5000 ppm or 0.5% under Option 1) would be acceptable without justification. Higher amounts may also be acceptable provided they are realistic in relation to manufacturing capability and good manufacturing practice (GMP). Table 3. Class 3 solvents, which should be limited by GMP or other quality-based requirements, in pharmaceutical products. Acetic acid Heptane Acetone Isobutyl acetate Anisole Isopropyl acetate 1-Butanol Methyl acetate 2-Butanol 3-Methyl-1-butanol Butyl acetate Methylethyl ketone tert-butylmethyl ether Methylisobutyl ketone Dimethyl sulfoxide 2-Methyl-1-propanol Ethanol Pentane Ethyl acetate 1-Pentanol Ethyl ether 1-Propanol Ethyl formate 2-Propanol Formic acid Propyl acetate 4.4 Solvents for which No Adequate Toxicological Data was Found The following solvents (Table 4) may be of interest to manufacturers of excipients, drug substances, or drug products. However, no adequate toxicological data on which to base a PDE was found. Manufacturers should supply justification for residual levels of these solvents in pharmaceutical products. Table 4. Solvents for which no adequate toxicological data was found. 1,1-Diethoxypropane 1,1-Dimethoxymethane 2,2-Dimethoxypropane Isooctane Methylisopropyl ketone Methyltetrahydrofuran Petroleum ether Trichloroacetic acid Furthermore, for the Class 2 or Class 3 solvent is used prior to the last step of a manufacturing process of drug substances, it is not necessary to establish the specification of the residual solvents of the drug substances, in case of complying with the requirements of the residual solvents that are not remain in the final active drug substances. (Examples of the recommended data) the content of 3 consecutive batches of residual solvents that is not more than acceptable concentration limit and the data of detection limit. When Class 2 solvents are used, they should be controlled either in a suitable intermediate or in the management data of final active substances depending on the batch scale. 5. GLOSSARY Genotoxic Carcinogens: Carcinogens which produce cancer by affecting genes or chromosomes. LOEL: Abbreviation for lowest-observed effect level. Lowest-Observed Effect Level: The lowest dose of substance in a study or group of studies that produces biologically significant increases in frequency or severity of any effects in the exposed humans or animals. Modifying Factor: A factor determined by professional judgment of a toxicologist and applied to bioassay data to relate that data safely to humans. Neurotoxicity: The ability of a substance to cause adverse effects on the nervous system. NOEL: Abbreviation for no-observed-effect level. No-Observed-Effect Level: The highest dose of substance at which there are no biologically significant increases in frequency or severity of any effects in the exposed humans or animals. PDE: Abbreviation for permitted daily exposure. Permitted Daily Exposure: The maximum acceptable intake per day of residual solvent in pharmaceutical products. Reversible Toxicity: The occurrence of harmful effects that are caused by a substance and which disappear after exposure to the substance ends. Strongly Suspected Human Carcinogen: A substance for which there is no epidemiological evidence of carcinogenesis but there are positive genotoxicity data and clear evidence of carcinogenesis in rodents. Teratogenicity: The occurrence of structural malformations in a developing fetus when a substance is administered during pregnancy. APPENDIX 1. LIST OF SOLVENTS INCLUDED IN THE GUIDELINE Solvent Other Names Classification Acetic acid Ethanoic acid Class 3 Acetone 2-Propanone; Propan-2-one Class 3

40 1854 Monographs, Part II Acetonitrile Class 2 Anisole Methoxybenzene Class 3 Benzene Benzol Class 1 1-Butanol n-butyl alcohol Class 3 Butan-1-ol 2-Butanol sec-butyl alcohol; Butan-2-ol Class 3 Butyl acetate Acetic acid butyl ester Class 3 tert-butylmethyl ether 2-Methoxy-2-methyl- propane Class 3 Carbon tetrachloride Tetrachloromethane Class 1 Chlorobenzene Class 2 Chloroform Trichloromethane Class 2 Cumene Isopropylbenzene; (1-Methyl)ethylbenzene Class 2 Cyclohexane Hexamethylene Class 2 1,2-Dichloroethane sym-dichloroethane; Ethylene dichloride; Ethylene chloride Class 1 1,1-Dichloroethene 1,1-Dichloroethylene; Vinylidene chloride Class 1 1,2-Dichloroethene 1,2-Dichloroethylene; Acetylene dichloride Class 2 Dichloromethane Methylene chloride Class 2 1,2-Dimethoxyethane Ethyleneglycol dimethyl ether; Monoglyme; Dimethyl Cellosolve Class 2 N,N- DMA Class 2 Dimethylacetamide N,N- DMF Class 2 Dimethylformamide Dimethyl sulfoxide Methylsulfinylmethane; Methyl sulfoxide; DMSO Class 3 1,4-Dioxane p-dioxane; [1,4]Dioxane Class 2 Ethanol Ethyl alcohol Class 3 2-Ethoxyethanol Cellosolve Class 2 Ethyl acetate Acetic acid ethyl ester Class 3 Ethyleneglycol 1,2-Dihydroxyethane; 1,2-Ethanediol Class 2 Ethyl ether Diethyl ether; Ethoxyethane; 1,1 -Oxybisethane Class 3 Ethyl formate Formic acid ethyl ester Class 3 Formamide Methanamide Class 2 Formic acid Class 3 Heptane n-heptane Class 3 Hexane n-hexane Class 2 Isobutyl acetate Acetic acid isobutyl ester Class 3 Isopropyl acetate Acetic acid isopropyl ester Class 3 Methanol Methyl alcohol Class 2 2-Methoxyethanol Methyl Cellosolve Class 2 Methyl acetate Acetic acid methyl ester Class 3 3-Methyl-1-butanol Isoamyl alcohol; Isopentyl alcohol; 3-Methylbutan-1-ol Class 3 Methylbutyl ketone 2-Hexanone; Hexan-2-one Class 2 Methylcyclohexane Cyclohexylmethane Class 2 Methylethyl ketone 2-Butanone; MEK; Butan-2-one Class 3 Methylisobutyl ketone 4-Methylpentan-2-one; 4-Methyl-2-pentanone; MIBK Class 3 2-Methyl-1-propanol Isobutyl alcohol; 2-Methylpropan-1-ol Class 3 N-Methylpyrrolidone 1-Methylpyrrolidin-2-one; 1-Methyl-2-pyrrolidinone Class 2 Nitromethane Class 2 Pentane n-pentane Class 3 1-Pentanol Amyl alcohol; Pentan-1-ol; Pentyl alcohol Class 3 1-Propanol Propan-1-ol; Propyl alcohol Class 3 2-Propanol Propan-2-ol; Isopropyl alcohol Class 3 Propyl acetate Acetic acid propyl ester Class 3 Pyridine Class 2 Sulfolane Tetrahydrothiophene 1,1-dioxide Class 2 Tetrahydrofuran Tetramethylene oxide; Oxacyclopentane Class 2 Tetralin 1,2,3,4-Tetrahydro-naphthalene Class 2 Toluene Methylbenzene Class 2 1,1,1-Trichloroethane Methylchloroform Class 1

41 KP X ,1,2-Trichloroethene Trichloroethene Class 2 Xylene note) Dimethybenzene; Xylol Class 2 note) usually 60% m-xylene, 14% p-xylene, 9% o-xylene with 17% ethyl benzene APPENDIX 2. ADDITIONAL BACKGROUND 2.1. Environmental Regulation of Organic Volatile Solvents Several of the residual solvents frequently used in the production of pharmaceuticals are listed as toxic chemicals in Environmental Health Criteria (EHC) monographs and the Integrated Risk Information System (IRIS). The objectives of such groups as the International Programme on Chemical Safety (IPCS), the United States Environmental Protection Agency (USEPA), and the United States Food and Drug Administration (USFDA) include the determination of acceptable exposure levels. The goal is protection of human health and maintenance of environmental integrity against the possible deleterious effects of chemicals resulting from longterm environmental exposure. The methods involved in the estimation of maximum safe exposure limits are usually based on long-term studies. When long-term study data are unavailable, shorter term study data can be used with modification of the approach such as use of larger safety factors. The approach described therein relates primarily to longterm or life-time exposure of the general population in the ambient environment, i.e. ambient air, food, drinking water and other media Residual Solvents in Pharmaceuticals Exposure limits in this guideline are established by referring to methodologies and toxicity data described in EHC and IRIS monographs. However, some specific assumptions about residual solvents to be used in the synthesis and formulation of pharmaceutical products should be taken into account in establishing exposure limits. They are: 1) Patients (not the general population) use pharmaceuticals to treat their diseases or for prophylaxis to prevent infection or disease. 2) The assumption of life-time patient exposure is not necessary for most pharmaceutical products but may be appropriate as a working hypothesis to reduce risk to human health. 3) Residual solvents are unavoidable components in pharmaceutical production and will often be a part of drug products. 4) Residual solvents should not exceed recommended levels except in exceptional circumstances. 5) Data from toxicological studies that are used to determine acceptable levels for residual solvents should have been generated using appropriate protocols such as those described for example by OECD, EPA, and the FDA Red Book. APPENDIX 3. METHODS FOR ESTABLISHING EXPOSURE LIMITS The Gaylor-Kodell method of risk assessment (Gaylor, D. W. and Kodell, R. L.: Linear Interpolation algorithm for low dose assessment of toxic substance. J Environ. Pathology, 4, 305, 1980) is appropriate for Class 1 carcinogenic solvents. Only in cases where reliable carcinogenicity data are available should extrapolation by the use of mathematical models be applied to setting exposure limits. Exposure limits for Class 1 solvents could be determined with the use of a large safety factor (i.e., 10,000 to 100,000) with respect to the no-observed-effect level (NOEL). Detection and quantitation of these solvents should be by state-of-the-art analytical techniques. Acceptable exposure levels in this guideline for Class 2 solvents were established by calculation of PDE values according to the procedures for setting exposure limits in pharmaceuticals (Pharmacopeial Forum, Nov-Dec 1989), and the method adopted by IPCS for Assessing Human Health Risk of Chemicals (Environmental Health Criteria 170, WHO, 1994). These methods are similar to those used by the USEPA (IRIS) and the USFDA (Red Book) and others. The method is outlined here to give a better understanding of the origin of the PDE values. It is not necessary to perform these calculations in order to use the PDE values tabulated in Section 4 of this document. PDE is derived from the no-observed-effect level (NOEL), or the lowest-observed effect level (LOEL) in the most relevant animal study as follows: NOEL x Weight Adjustment PDE (1) F1x F2 x F3x F4 x F5 The PDE is derived preferably from a NOEL. If no NOEL is obtained, the LOEL may be used. Modifying factors proposed here, for relating the data to humans, are the same kind of "uncertainty factors" used in Environmental Health Criteria (Environmental Health Criteria 170, World Health Organization, Geneva, 1994), and "modifying factors" or "safety factors" in Pharmacopeial Forum. The assumption of 100% systemic exposure is used in all calculations regardless of route of administration. The modifying factors are as follows: F1 = A factor to account for extrapolation between species F1 = 5 for extrapolation from rats to humans F1 = 12 for extrapolation from mice to humans F1 = 2 for extrapolation from dogs to humans F1 = 2.5 for extrapolation from rabbits to humans F1 = 3 for extrapolation from monkeys to hu-

42 1856 Monographs, Part II mans F1 = 10 for extrapolation from other animals to humans F1 takes into account the comparative surface area: body weight ratios for the species concerned and for man. Surface area (S) is calculated as: S = km 0.67 (2) in which M = body mass, and the constant k has been taken to be 10. The body weights used in the equation are those shown below in the Table 1. F2 = A factor of 10 to account for variability between individuals A factor of 10 is generally given for all organic solvents, and 10 is used consistently in this guideline. F3 = A variable factor to account for toxicity studies of short-term exposure F3 = 1 for studies that last at least one half lifetime (1 year for rodents or rabbits; 7 years for cats, dogs and monkeys). F3 = 1 for reproductive studies in which the whole period of organogenesis is covered. F3 = 2 for a 6-month study in rodents, or a 3.5- year study in non-rodents. F3 = 5 for a 3-month study in rodents, or a 2- year study in non-rodents. F3 = 10 for studies of a shorter duration. In all cases, the higher factor has been used for study durations between the time points, e.g. a factor of 2 for a 9-month rodent study. F4 = A factor that may be applied in cases of severe toxicity, e.g. non-genotoxic carcinogenicity, neurotoxicity or teratogenicity. In studies of reproductive toxicity, the following factors are used: F4 = 1 for fetal toxicity associated with maternal toxicity F4 = 5 for fetal toxicity without maternal toxicity F4 = 5 for a teratogenic effect with maternal toxicity F4 = 10 for a teratogenic effect without maternal toxicity F5 = A variable factor that may be applied if the noeffect level was not established When only an LOEL is available, a factor of up to 10 could be used depending on the severity of the toxicity. The weight adjustment assumes an arbitrary adult human body weight for either sex of 50 kg. This relatively low weight provides an additional safety factor against the standard weights of 60 kg or 70 kg that are often used in this type of calculation. It is recognized that some adult patients weigh less than 50 kg; these patients are considered to be accommodated by the built-in safety factors used to determine a PDE. If the solvent was present in a formulation specifically intended for pediatric use, an adjustment for a lower body weight would be appropriate. As an example of the application of the equation (1), consider a toxicity study of acetonitrile in mice that is summarized in Pharmeuropa, Vol. 9, No. 1, Supplement, April 1997, page S24. The NOEL is calculated to be 50.7 mg kg-1 day-1. The PDE for acetonitrile in this study is calculated as follows: mg kg day 50 kg PDE 4.2 mg day In this example, F1 = 12 to account for the extrapolation from mice to humans F2 = 10 to account for differences between individual humans F3 = 5 because the duration of the study was only 13 weeks F4 = 1 because no severe toxicity was encountered F5 = 1 because the no effect level was determined Table 1. Values used in the calculations in this guideline. rat body weight pregnant rat body weight mouse body weight pregnant mouse body weight guinea pig body weight Rhesus monkey body weight rabbit body weight (pregnant or not) beagle dog body weight rat respiratory volume mouse respiratory volume rabbit respiratory volume guinea pig respiratory volume human respiratory volume dog respiratory volume monkey respiratory volume mouse water consumption rat water consumption rat food consumption 425 g 330 g 28 g 30 g 500 g 2.5 kg 4 kg 11.5 kg 290 L/day 43 L/day L/day 430 L/day 28,800 L/day 9,000 L/day 1,150 L/day 5 ml/day 30 ml/day 30 g/day The equation for an ideal gas, PV = nrt, is used to convert concentrations of gases used in inhalation studies from units of ppm to units of mg/l or mg/m 3. Consider as an example in which the rat reproductive toxicity study by inhalation of carbon tetrachloride (molecular weight ) is summarized in

43 KP X 1857 Pharmeuropa, Vol. 9, No. 1, Supplement, April 1997, page S9. n P RT 00 x 10 atm x mg mol L atm 1 mol 1 x 8 15 mg 1 8 mg L 5 L The relationship 1000 L = 1 m 3 is used to convert to mg/ m Guideline of Validation of Analytical Procedures for Pharmaceuticals I. OBJECTIVE The purpose of this guideline is to provide the detailed guidance on how to conduct the validation of the analytical procedures necessary for application (approval) for manufacture import of pharmaceuticals quasi-drugs according to notifications of the Korean Food and Drug Administration, such as Regulation of Pharmaceuticals Approval, Notification and Review, and for quality control. II. INTRODUCTION The objective of validation of an analytical procedure for pharmaceuticals is to demonstrate that the procedure, applied in the quality control of the pharmaceutical, is suitable for the intended purpose. The purpose is to direct how to establish relevant validation parameters for each analytical procedure. The scope of analytical procedures that are subjected to the present guideline is as follows: 1) Identification tests in the specification for pharmaceuticals 2) Purity tests: quantitative tests and limit tests for the impurity in a sample 3) Quantification tests: assay for the active component of drug substance or drug product, for other selected component(s) in drug products, for Uniformity of Dosage Units test or for dissolution test The objective of the analytical procedure should be clear since the procedure will govern the validation characteristics which need to be evaluated. Typical validation parameters include specificity, accuracy, precision, detection limit, quantification limit, linearity, range, and robustness. Appropriate validation parameters are selected and evaluated depending on the objective of a given analytical procedure. Furthermore, revalidation may be necessary when there are change in the synthesis of the drug substance, change in the composition of the finished product, and change in the analytical procedure. The degree of revalidation required depends on the nature of the changes. In addition, validation methods, other than those described in this guideline, may be used, although the appropriateness of the alternative method has to be justified. III. GLOSSARY 1. Analytical Procedure The analytical procedure refers to the way of performing the analysis. It should describe in detail the steps necessary to perform each analytical test. The analytical procedure includes but is not limited to: the analyte, the sample, the reference standard, the standard reagents, the use of analytical instruments, the use of the calibration curve, the use of formulae for the calculation in the identification test, purity test or assay. - The term validation of analytical procedure is defined as the process of confirming that the analytical procedure for a quality control test of pharmaceuticals is suitable for its intended use. - Identification Test refers to ensure the identity of an analyte, and generally compare the physicochemical characteristics (spectrum, information from chromatographic methods, and chemical reactivity, etc) of a sample with those of a reference material. - Purity Test is to ensure that all the analytical procedures performed allow an accurate statement of the content of impurities of an analyte, i.e. related substances test, heavy metals, residual solvents content, etc. Testing for impurities can be either a quantification test or a limit test for the impurity in a sample. - Assay (Content or Potency) is to provide an exact result which allows an accurate statement on the content or potency of the analyte in a sample. The assay represents a quantitative measurement of the major component(s) or other selected component(s) (i.e., stabilizer, additive) in the drug substance or in the drug product. The validation principles are also applicable to assay for dissolution test. 2. Specificity Specificity is the ability to assess unequivocally the analyte in the presence of components, e.g., impurities, degradation products, matrix, etc, which may be expected to be present. Lack of specificity of an individual analytical procedure may be compensated by other supporting analytical procedure(s). 3. Accuracy The accuracy of an analytical procedure expresses the closeness of agreement between the value which is accepted either as a conventional true value or an accepted reference value and the value

44 1858 Monographs, Part II found. 4. Precision The precision of an analytical procedure expresses the closeness of agreement (degree of scatter) between a series of measurements obtained from homogeneous sample under the prescribed conditions. Precision may be considered at three levels: repeatability, intermediate precision and reproducibility. - Repeatability expresses the precision of measurements obtained under the same operating conditions (same analyst, same laboratory, same equipment, same sample, etc) over a short interval of time. Repeatability is also termed intra-assay precision. - Intermediate Precision expresses withinlaboratories variations: different days, different analysts, different equipment, etc. - Reproducibility expresses the precision of observed values (collaborative studies, usually applied to standardization of methodology) obtained from a homogenous sample in different laboratories. 5. Detection Limit The detection limit of an individual analytical procedure is the lowest amount of analyte in a sample which can be detected but not necessarily quantified as an exact value. 6. Quantitation Limit The quantitation limit of an individual analytical procedure is the lowest amount of analyte in a sample which can be quantitatively determined with suitable precision and accuracy. The quantitation limit is a parameter of quantitative assays for low levels of compounds in sample matrices, and is used particularly for the determination of impurities and/or degradation products. 7. Linearity The linearity of an analytical procedure is its ability to obtain test results which are directly proportional to the amount (concentration) of analyte in the sample. 8. Range The range of an analytical procedure is the interval between the upper and lower concentration (amounts) of analyte in the sample (including these concentrations) for which it has been demonstrated that the analytical procedure has a suitable level of precision, accuracy and linearity. 9. Robustness The robustness of an analytical procedure is a measure of its capacity to remain unaffected by small, but deliberate variations in method parameters. It provides an indication of its reliability during normal usage. IV. VALIDATION OF ANALYTICAL PROCE- DURES For validation of analytical procedures, the validation procedure and protocol most suitable for their product must be first selected. Then, all data and validation parameters collected during the validation process are used to evaluate the appropriateness. The relevant data collected during validation and formulae used for calculating validation parameters should be presented and justified as appropriate. The validation parameters applicable to the identification test, the control of impurities and the assay procedures is summarized in Appendix 1. Well-characterized reference materials, which documented purity as well as physicochemical and biological characteristics, should be used throughout the validation study. The degree of purity necessary depends on the intended use. 1. Specificity An investigation of specificity should be conducted during the validation of identification tests, the control of impurities and the assay procedures. The procedures used to demonstrate specificity will depend on the intended objective of the analytical procedure. It is not always possible to demonstrate that an analytical procedure is specific for a particular analyte with complete discrimination. In this case, a combination of two or more analytical procedures is recommended to achieve the necessary level of discrimination. A. Identification Suitable identification tests should be able to discriminate between compounds of closely related structures which are likely to be present. The discrimination of a procedure may be confirmed by obtaining positive results, based on the comparison with a known reference material, from samples containing the analyte, coupled with negative results from samples which do not contain the analyte. In addition, the identification test may be applied to materials structurally similar to or closely related to the analyte to confirm that a positive response is not obtained. The choice of such potentially interfering materials should be based on sound scientific judgement with a consideration of the interferences that could occur. B. Assay and Impurity Test(s) For chromatographic procedures, representative chromatograms should be presented to demonstrate specificity and individual components should be appropriately labeled. Similar considerations should be given to other separation techniques.

45 KP X 1859 Critical separations in chromatography should be investigated at an appropriate level to demonstrate that the components are adequately separated. For critical separations, specificity can be demonstrated by the resolution of the two components which elute closest to each other. In cases where a non-specific assay is used, other supporting analytical procedures should be used to demonstrate overall specificity. For example, where a titration is adopted to assay the drug substance for release, the combination of the assay and a suitable test for impurities can be used. The approach is similar for both assay and impurity tests. (1) Impurities are available For the assay, this should involve demonstration of the discrimination of the analyte in the presence of impurities and/or excipients; practically, this can be done by spiking pure substances (drug substance or drug product) with appropriate levels of impurities and/or excipients and demonstrating that the assay result is unaffected by the presence of these materials (by comparison with the assay result obtained on unspiked samples). For the impurity test, the discrimination may be established by spiking drug substance or drug product with appropriate levels of impurities and demonstrating the separation of these impurities individually and/or from other components in the sample matrix. (2) Impurities are not available If impurity or degradation products standards are unavailable, specificity may be demonstrated by comparing the test results of samples containing impurities or degradation products to a second well-characterized procedure. In this approach, analytical procedure having known validation parameter includes pharmacopoeial method, other validated analytical procedure e.g.: pharmacopoeial method of other validated analytical procedure (independent procedure). As appropriate, this should include samples stored under relevant stress conditions light, heat, humidity, acid/base hydrolysis and oxidation. - For the assay, the two results should be compared. - For the impurity tests, the impurity profiles should be compared. Peak purity tests may be useful to show that the analyte chromatographic peak is not attributable to more than one component (e.g., diode array, mass spectrometry). 2. Linearity A linear relationship should be evaluated across the range (see section 3) of the analytical procedure. It may be demonstrated directly on the drug substance (by dilution of a standard stock solution) and/or separate weighings of synthetic mixtures of the drug product components, using the proposed procedure. The latter aspect can be studied during investigation of the range. Linearity should be evaluated by visual inspection of a plot of signals as a function of analyte concentration or content. If there is a linear relationship, test results should be evaluated by appropriate statistical methods, for example, by calculation of a regression line by the method of least squares. In some cases, to obtain linearity between assays and sample concentrations, the test data may need to be subjected to a mathematical transformation prior to the regression analysis. Data from the regression line itself may be helpful to provide mathematical estimates of the degree of linearity. The correlation coefficient, y-intercept, slope of the regression line and residual sum of squares should be submitted. A plot of the data should be included. In addition, an analysis of the deviation of the actual data points from the regression line may also be helpful for evaluating linearity. Some analytical procedures, such as immunoassays, do not demonstrate linearity after any transformation. In this case, the analytical response should be described by an appropriate function of the concentration (amountly or empiraclly) of the concentration (or content) of an analyte in a sample. For the establishment of linearity, a minimum of 5 concentrations is recommended. Other approaches should be justified. 3. Range The specified range is normally derived from linearity studies and depends on the intended application of the procedure. It is established by confirming that the analytical procedure provides an acceptable degree of linearity, accuracy and precision when applied to samples containing amounts of analyte within or at the extremes of the specified range of the analytical procedure. The following minimum specified ranges should be considered: A. Assay of a drug substance or a drug product Normally from 80 to 120 percent of the test concentration. B. Content uniformity Covering a minimum of 70 to 130 percent of the test concentration, unless a wider more appropriate range, based on the nature of the dosage form such as metered dose inhalers, is justified C. Dissolution testing ±20 % over the specified range indicated in dissolution specification of the Specifications and Test

46 1860 Monographs, Part II Procedures of the drug product. For example, if the specifications for a controlled released product cover a region from 20%, after 1 hour, up to 90%, after 24 hours, the validated range would be 0-110% of the labeled content. D. Assay of related substance From the reporting level of an impurity to 120% of the specification. For impurities known to be unusually potent or to produce toxic or unexpected pharmacological effects, the detection/quantitation limit should be commensurate with the level at which the impurities must be controlled. For validation of impurity test procedures carried out during development, it may be necessary to consider the range around a suggested (probable) limit. E. If assay and purity are performed together as one test and only a 100% standard is used, linearity should cover the range from the reporting level of the impurities to 120% of the assay specification. 4. Accuracy Accuracy should be established across the specified range of the analytical procedure. A. Assay (1) Drug Substance Several methods of determining accuracy are available: (a) When the purity is known Application of an analytical procedure to an analyte of known purity (e.g., reference material); (b) When other analytical procedure with the known accuracy is available Comparison of the results of the proposed analytical procedure with those of a second well characterized procedure, the accuracy of which is stated and/or defined (independent procedure); (c) Accuracy may be inferred once precision, linearity and specificity have been established. (2) Drug Product Several methods for determining accuracy are available: (a) Application of the analytical procedure to synthetic mixtures of the drug product components to which known quantities of the drug substance to be analyzed have been added; (b) In cases where it is impossible to obtain samples of all drug product components, it may be acceptable either 1) to add known quantities of the analyte to the drug product, or 2) to compare the results obtained from a second well-characterized procedure, the accuracy of which is stated and/or defined (independent procedure). (c) Accuracy may be inferred once precision, linearity and specificity have been established. B. Quantification of Impurities Accuracy should be assessed on samples (drug substance/drug product) spiked with known amounts of impurities. In cases where it is impossible to obtain samples of certain impurities, and/or degradation products, it is considered acceptable to compare results obtained by an independent procedure. The response factor of the drug substance can be used. It should be clear how the individual or total impurities are to be determined, e.g., weight/weight or area percent, in all cases with respect to the major analyte. C. Recommended Data Accuracy should be assessed using a minimum of 9 determinations over a minimum of 3 concentration levels covering the specified range (e.g., 3 concentrations/3 replicates each of the total analytical procedure). Accuracy should be reported as percent recovery by the assay of known added amount of analyte in the sample or as the difference between the mean and the accepted true value together with the confidence intervals. 5. Precision Validation of tests for assay and for quantitative determination of impurities includes an investigation of precision. A. Repeatability Repeatability should be assessed using: (1) a minimum of 9 determinations covering the specified range for the procedure (e.g., 3 concentrations/3 replicates each), or (2) a minimum of 6 determinations at 100% of the test concentration. B. Intermediate precision The extent to which intermediate precision should be established depends on the circumstances under which the procedure is intended to be used. The applicant should establish the effects of random events on the precision of the analytical procedure. Typical variations to be studied include days, analysts, equipment, etc. The use of an experimental design (matrix) is encouraged. C. Reproducibility Reproducibility is assessed by means of an interlaboratory trial. Reproducibility should be considered in case of the standardization of an analytical

47 KP X 1861 procedure, for instance, for inclusion of procedures in pharmacopoeias. These data are not part of the marketing authorization dossier. D. Recommended Data The standard deviation, relative standard deviation (coefficient of variation) and confidence interval should be reported for each type of precision investigated. 6. Detection limit Several approaches for determining the detection limit are possible, depending on whether the procedure is a non-instrumental or instrumental. Approaches other than those listed below may be acceptable. A. Based on Visual Evaluation Visual evaluation may be used for noninstrumental methods but may also be used with instrumental methods. The detection limit is determined by the analysis of samples with known concentrations of analyte and by establishing the minimum level at which the analyte can be reliably detected. B. Based on Signal-to-Noise This approach can only be applied to analytical procedures which exhibit baseline noise. Determination of the signal-to-noise ratio is performed by comparing measured signals from samples with known low concentrations of analyte with those of blank samples and establishing the minimum concentration at which the analyte can be reliably detected. A signal-to-noise ratio between 3 or 2:1 is generally considered acceptable for estimating the detection limit. C. Based on the Standard Deviation of the Response and the Slope The detection limit (DL) may be expressed as: DL =. * σ / S where σ = the standard deviation of the response S = the slope of the calibration curve The slope S may be estimated from the calibration curve of the analyte. The estimate of σ may be carried out in a variety of ways for example: (1) Based on the Standard Deviation of the Blank Measurement of the magnitude of analytical background response is performed by analyzing an appropriate number of blank samples and calculating the standard deviation of these responses. (2) Based on the Calibration Curve A specific calibration curve should be studied using samples containing an analyte in the range of DL. The residual standard deviation of a regression line or the standard deviation of y-intercepts of regression lines may be used as the standard deviation. D. Recommended Data The detection limit and the method used for determining the detection limit should be presented. If DL is determined based on visual evaluation or based on signal-to-noise ratio, the presentation of the relevant chromatograms is considered acceptable for justification. In cases where an estimated value for the detection limit is obtained by calculation or extrapolation, this estimate may subsequently be validated by the independent analysis of a suitable number of samples known to be near or prepared at the detection limit. 7. Quantification limit Several approaches for determining the quantification limit are possible, depending on whether the procedure is non-instrumental or instrumental. Approaches other than those listed below may be acceptable. A. Based on Visual Evaluation Visual evaluation may be used for noninstrumental methods but may also be used with instrumental methods. The quantification limit is generally determined by the analysis of samples with known concentrations of analyte and by establishing the minimum level at which the analyte can be quantified with acceptable accuracy and precision. B. Based on Signal-to-Noise Approach This approach can only be applied to analytical procedures that exhibit baseline noise. Determination of the signal-to-noise ratio is performed by comparing measured signals from samples with known low concentrations of analyte with those of blank samples and by establishing the minimum concentration at which the analyte can be reliably quantified. A typical signal-to-noise ratio is 10:1. C. Based on the Standard Deviation of the Response and the Slope The quantification limit (QL) may be expressed as: QL = 10 * σ / S where σ = the standard deviation of the response S = the slope of the calibration curve The slope S may be estimated from the calibra-

48 1862 Monographs, Part II tion curve of the analyte. The estimate of σ may be carried out in a variety of ways for example: (1) Based on Standard Deviation of the Blank Measurement of the magnitude of analytical background response is performed by analyzing an appropriate number of blank samples and calculating the standard deviation of these responses. (2) Based on the Calibration Curve A specific calibration curve should be studied using samples, containing an analyte in the range of QL. The residual standard deviation of a regression line or the standard deviation of y-intercepts of regression lines may be used as the standard deviation. D. Recommended Data The quantification limit and the method used for determining the quantification limit should be presented. The limit should be subsequently validated by the analysis of a suitable number of samples known to be near or prepared at the quantification limit. 8. Robustness The evaluation of robustness should be considered during the development phase and depends on the type of procedure under study. It should show the reliability of an analysis with respect to deliberate variations in method parameters. If measurements are susceptible to variations in analytical conditions, the analytical conditions should be suitably controlled or a precautionary statement should be included in the procedure. One consequence of the evaluation of robustness should be that a series of system suitability parameters (e.g., resolution test) is established to ensure that the validity of the analytical procedure is maintained whenever used. A. Typical variation factors common for all analytical procedures - stability of analytical solutions, - extraction time B. Typical variation factors for liquid chromatography - influence of variations of ph in a mobile phase, - influence of variations in mobile phase composition, - different columns (different lots or suppliers), - temperature, - flow rate C. Typical variation factors for gas chromatography - change in columns (different lots and/or suppliers), - temperature, - flow rate 9. System suitability testing System suitability testing is an integral part of many analytical procedures. The tests are based on the concept that the equipment, electronics, analytical operations and samples to be analyzed constitute an integral system that can be evaluated as such. System suitability test parameters to be established for a particular procedure depend on the type of procedure being validated. See General Tests for additional information. Appendix 1) Validation characteristics applicable to analytical procedures - signifies that this characteristic is not normally evaluated + signifies that this characteristic is normally evaluated (1) in cases where reproducibility (see glossary) has been performed, intermediate precision is not needed (2) lack of specificity of one analytical procedure could be compressed by other supporting analytical procedure(s) (3) may be needed in some cases 8. Isoelectric focusing General Principles Isoelectric focusing (IEF) is a method of electrophoresis that separates proteins according to their isoelectric points. Separation is carried out in a slab of polyacrylamide or agarose gel that contains a mixture of amphoteric electrolytes (ampholytes). When subjected to an electric field, the ampholytes migrate in the gel to create a ph gradient. In some cases, gels containing an immobilized ph gradient, prepared by incorporating weak acids and bases to specific regions of the gel network during the preparation of the gel, are used. When the applied proteins reach the gel fraction that has a ph that is the same as their isoelectric point (pi), their charge is neutralized and migration ceases. Gradients can be made over various ranges of ph, according to the mixture of ampholytes chosen. Theoretical Aspects When a protein is at the position of its isoelectric point, it has no net charge and cannot be moved in a gel matrix by the electric field. It may, however, move from that position by diffusion. The ph gradient forces a protein to remain in its isoelectric point position, thus concentrating it; this concentrating effect is called focusing. The applied voltage is limited by the heat generated, which must be dissipated. The use of thin gels and an efficient cooling plate controlled by a thermostatic circulator prevents the burning of the gel whilst allowing sharp focusing. The separation is es-

49 KP X 1863 timated by determining the minimum pi difference (ΔpI), which is necessary to separate 2 neighboring bands: R pi 3 D ( dph / dx) E ( d / dph ) D: Diffusion coefficient of the protein dph/dx: ph gradient E: Intensity of the electric field, in volts per centimeter -dµ/dph: Variation of the solute mobility with the ph in the region close to the pi Type of analytical procedure ASSAY - dissolution (measuremen t only) - content / potency + characteristics Accuracy Precision Repeatability Interm.Pr ecision Specificity(2) Detection Limit Quantification Limit Linearity Range Identification Testing for impurities quan tific ation + + +(1) + -(3) limit (1) Since D and -dµ/dph for a given protein cannot be altered, the separation can be improved by using a narrower ph range and by increasing R, the intensity of the electric field. Resolution between protein bands on an IEF gel prepared with carrier ampholytes can be quite good. Improvements in resolution may be achieved by using immobilized ph gradients where the buffering species, which are analogous to carrier ampholytes, are copolymerized within the gel matrix. Proteins exhibiting pis differing by as little as 0.02 ph units may be resolved using a gel prepared with carrier ampholytes while immobilized ph gradients can resolve proteins differing by approximately ph units. The time required for completion of focusing in thin-layer polyacrylamide gels is determined by placing a colored protein (e.g. hemoglobin) at different positions on the gel surface and by applying the electric field: the steady state is reached when all applications give an identical band pattern. In some protocols the completion of the focusing is indicated by the time elapsed after the sample application. The IEF gel can be used as an identity test when the migration pattern on the gel is compared to a suitable standard preparation and IEF calibration proteins, the IEF gel can be used as a limit test when the density of a band on IEF is compared subjectively with the density of bands appearing in a standard preparation, or it can be used as a quantitative test when the density is measured using a densitometer or similar instrumentation to determine the relative concentration of protein in the bands subject to validation. Apparatus An apparatus for IEF consists of : A controllable generator for constant potential, current and power. Potentials of 2500 V have been used and are considered optimal under a given set of operating conditions. Supply of up to 30 W of constant power is recommended, A rigid plastic IEF chamber that contains a cooled plate, of suitable material, to support the gel, A plastic cover with platinum electrodes that are connected to the gel by means of paper wicks of suitable width, length and thickness, impregnated with solutions of anodic and cathodic electrolytes. Isoelectric Focusing in Polyacrylamide Gels: Detailed Procedure The following method is a detailed description of an IEF procedure in thick polyacrylamide slab gels, which is used unless otherwise stated in the monograph. Preparation of the Gels Mould The mould (see Figure) is composed of a glass plate (A) on which a polyester film (B) is placed to facilitate handling of the gel, one or more spacers (C), a second glass plate (D) and clamps to hold the structure together. 7.5% Polyacrylamide gel Dissolve 29.1 g of acrylamide and 0.9 g of N,N -methylenebisacrylamide in 100 ml of water. To 2.5 volumes of this solution, add the mixture of ampholytes specified in the monograph and dilute to 10 volumes with water. Mix carefully and degas the solution. Practical Aspects Special attention must be paid to sample characteristics and/or preparation. Having salt in the sample can be problematic and it is best to prepare the sample, if possible, in de-ionized water or 2% ampholytes, using dialysis or gel filtration if necessary.

50 1864 Monographs, Part II D C B A Preparation of the mould Place the polyester film on the lower glass plate, apply the spacer, place the second glass plate and fit the clamps. Place 7.5% polyacrylamide gel prepared before use on a magnetic stirrer, and add 0.25 volumes of a solution of ammonium persulfate (1 in 10) and 0.25 volumes of N,N,N,N -tetramethylethylenediamine. Immediately fill the space between the glass plates of the mould with the solution. Method Dismantle the mould and, making use of the polyester film, transfer the gel onto the cooled support, wetted with 2-3 ml of a suitable liquid, taking care to avoid forming air bubbles. Prepare the test solutions and reference solutions as specified in the monograph. Place strips of paper for sample application, about 10 mm 5 mm in size, on the gel and impregnate each with the prescribed amount of the test and reference solutions. Also apply the prescribed quantity of a solution of proteins with known isoelectric points as ph markers. In some protocols, the gel has precast slots where a solution of the sample is applied instead of using impregnated paper strips. Cut 2 strips of paper to the length of the gel and impregnate them with the electrolyte solutions (acid for the anode and alkaline for the cathode). The compositions of the anode and cathode solutions are given in the monograph. Apply these paper wicks to each side of the gel several millimeters from the edge. Fit the cover so that the electrodes are in contact with the wicks (respecting the anodic and cathodic poles). Proceed with the isoelectric focusing by applying the electrical parameters described in the monograph. Switch off the current when the migration of the mixture of standard proteins has stabilized. Using forceps, remove the sample application strips and the 2 electrode wicks. Immerse the gel in fixing solution for isoelectric focusing in polyacrylamide gel. Incubate with gentle shaking at room temperature for 30 minutes. Drain off the solution and add 00 ml of destaining solution. Incubate with shaking for 1hour. Drain the gel, add coomassie staining TS. Incubate for 30 minutes. Destain the gel by passive diffusion with destaining solution until the bands are well visualized against a clear background. Locate the position and intensity of the bands in the electropherogram as prescribed in the monograph. Variations to the Detailed Procedure (Subject to Validation) Where reference to the general method on isoelectric focusing is made, variations in methodology or procedure may be made subject to validation. These include: (1) the use of commercially available pre-cast gels and of commercial staining and destaining kits, (2) the use of immobilized ph gradients, (3) the use of disk gels, (4) the use of gel cassettes of different dimensions, including ultra-thin (0.2 mm) gels, (5) variations in the sample application procedure, including different sample volumes or the use of sample application masks or wicks other than paper, (6) the use of alternate running conditions, including variations in the electric field depending on gel dimensions and equipment, and the use of fixed migration times rather than subjective interpretation of band stability, (7) the inclusion of a pre-focusing step, (8) the use of automated instrumentation, (9) the use of agarose gels. Validation of Iso-Electric Focusing Procedures Where alternative methods to the detailed procedure are employed they must be validated. The following criteria may be used to validate the separation: (1) formation of a stable ph gradient of desired characteristics, assessed for example using colored ph markers of known isoelectric points, (2) comparison with the electropherogram provided with the chemical reference substance for the preparation to be examined, (3) any other validation criteria as prescribed in the monograph. Specified Variations to the General Method Variations to the general method required for the analysis of specific substances may be specified in detail in monographs. These include: (1) the addition of urea in the gel (3 mol/l concentration is often satisfactory to keep protein in solution but up to 8mol/L can be used): some proteins precipitate at their isoelectric point. In this case, urea is included in the gel formulation to keep the protein in solution. If urea is used, only fresh solutions should be used to prevent carbamylation of the protein, (2) the use of alternative staining methods, (3) the use of gel additives such as non-ionic detergents (e.g. octylglucoside) or zwitterionic detergents [e.g., 3-[(3- cholamido propyl)dimethylammonio]-1- propane sulfonate (CHAPS) or 3-[(3- cholamidopropyl)dimethyl ammonio]-2-hydroxy-1- propanesulfate (CHAPSO)], and the addition of ampholyte to the sample, to prevent proteins from aggregating or precipitating. Points to Consider Samples can be applied to any area on the gel, but to protect the proteins from extreme ph environments samples should not be applied close to either electrode. During method development the analyst can try applying the protein in 3 positions on the gel (i.e. middle and both ends); the pattern of a protein applied

51 KP X 1865 at opposite ends of the gel may not be identical. A phenomenon known as cathodic drift, where the ph gradient decays over time, may occur if a gel is focused too long. Although not well understood, electroendoosmosis and absorption of carbon dioxide may be factors that lead to cathodic drift. Cathodic drift is observed as focused protein migrating off the cathode end of the gel. Immobilized ph gradients may be used to address this problem. Efficient cooling (approximately 4 C) of the bed that the gel lies on during focusing is important. High field strengths used during isoelectric focusing can lead to overheating and affect the quality of the focused gel. Reagents and Solutions Fixing solution for isoelectric focusing in polyacrylamide gel Dissolve 35 g of 5-sulfosalicylic acid dihydrate and 100 g of trichloroacetic acid in water to make 1000 ml. Coomassie staining TS Dissolve 125 mg of coomassie brilliant blue R-250 in 100 ml of a mixture of water, methanol and acetic acid (100) (5:4:1), and filter. Destaining solution A mixture of water, methanol and acetic acid (100) (5:4:1). 9. Particle Size Determination Particle Size Determination is a method to determine directly or indirectly morphological appearance, shape, size and its distribution of powdered pharmaceutical drugs and excipients to examine their micromeritic properties. Optical microscopy or analytical sieving method is used according to the measuring purpose and the properties of the test specimen. Method 1. Optical Microscopy The optical microscopy is used to observe the morphological appearance and shape of individual particle either directly with the naked eye or by using a microscopic photograph, in order to measure the particle size. The particle size distribution can also be determined by this method. It is also possible with this method to measure the size of the individual particle even then different kinds of particles mingle if they are optically distinguishable. Data processing techniques, such as image analysis, can be useful for determining the particle size distribution. This method for particle characterization can generally be applied to particles not less than 1 μm. The lower limit is imposed by the resolving power of the microscope. The upper limit is less definite and is determined by the increased difficulty associated with the characterization of large particles. Various alternative techniques are available for particle characterization, outside the applicable range of optical microscopy. Optical microscopy is particularly useful for characterizing particles that are not spherical. This method may also serve as a base for the calibration of faster and more routine methods that may be developed. Apparatus - Use a microscope that is stable and protected from vibration. The microscope magnification (product of the objective magnification, ocular magnification, and additional magnifying components) must be sufficient to allow adequate characterization of the smallest particles to be classified in the test specimen. The greatest numerical aperture of the objective should be sought for each magnification range. Polarizing filters may be used in conjunction with suitable analyzers and retardation plates. Color filters of relatively narrow spectral transmission should be used with achromatic objectives and are preferable with apochromats and are required for appropriate color rendition in photomicrography. Condensers corrected for at least spherical aberration should be used in the microscope substage and with the lamp. The numerical aperture of the substage condenser should match that of the objective under the condition of use, in the other words this is affected by the actual aperture of the condenser diaphragm and the presence of immersion oils. Adjustment - The precise alignment of all elements of the optical system and proper focusing are essential. The focusing of the elements should be done in accordance with the recommendations of the microscope manufacturer. Critical axial alignment is recommended. Illumination - A requirement for good illumination is a uniform and adjustable intensity of light over the entire field of view; Kohler illumination is preferred. With colored particles, choose the color of the filters used so as to control the contrast and detail of the image. Visual Characterization - The magnification and numerical aperture should be sufficiently high to allow adequate resolution of the images of the particles to be characterized. Determine the actual magnification using a calibrated stage micrometer to calibrate an ocular micrometer. Errors can be minimized if the magnification is sufficient that the image of the particle is at least 10 ocular divisions. Each objective must be calibrated separately. The ocular scale, the stage micrometer scale and the ocular scale should be aligned to calibrate. In this way, a precise determination of the distance between ocular stage divisions can be made. When the particle size is measured, an ocular micrometer is inserted at the position of the ocular diaphragm, and a calibrated stage micrometer is placed at the center of the microscope stage and fixed in place. The ocular is attached to the lens barrel and adjusted to the focus point of the stage micrometer scale. Then, the distance between the scales of the two micrometers is determined, and the sample size equivalent 1 division of the ocular scale is calculated using the fol-

52 1866 Monographs, Part II lowing formula: The particle size equivalent 1 division on the ocular scale ( m) = Length on the stage micrometer ( m)/number of scale divisions on the ocular micrometer The stage micrometer is removed and the test specimen is placed on the microscope stage. After adjusting the focus, the particle sizes are determined from the number of scale divisions read through the ocular. Several different magnifications may be necessary to characterize materials having a wide particle size distribution. Photographic Characterization - If particle size is to be determined by photographic methods, take care to ensure that the object is sharply focused at the plane of the photographic emulsion. Determine the actual magnification by photographing a calibrated stage micrometer, using photographic film of sufficient speed, resolving power, and contrast. Exposure and processing should be identical for photographs of both the test specimen and the determination of magnification. The apparent size of a photographic image is influenced by the exposure, development, and printing processes as well as by the resolving power of the microscope. Preparation of the Mount - The mounting medium will be selected according to the physical properties of the test specimen. Sufficient, but not excessive, contrast between the specimen and the mounting medium is required to ensure adequate detail of the specimen edge. The particles should rest in one plane and be adequately dispersed to distinguish individual particles of interest. Furthermore, the particles must be representative of the distribution of sizes in the material and must not be altered during preparation of the mount. Selection of the mounting medium must include a consideration of the analyte solubility. Crystallinity characterization - The crystallinity of a material may be characterized to determine compliance with the crystallinity requirement where stated in the individual monograph of a drug substance. Unless otherwise specified in the individual monograph, mount a few particles of the specimen in mineral oil on a clean glass slide. Examine the mixture using a polarizing microscope: the particles show birefringence (interference colors) and extinction positions when the microscope stage is revolved. Limit Test of Particle Size by Microscopy - Weigh a suitable quantity of the power to be examined (for example, 10 to 100 mg), and suspend it in 10 ml of a suitable medium in which the power does not dissolve, adding, if necessary, a wetting agent. A homogeneous suspension of particles can be maintained by suspending the particles in a medium of similar or matching density and by providing adequate agitation. Introduce a portion of the homogeneous suspension into a suitable counting cell, and scan under a microscope an area corresponding to not less than 10 g of the powder to be examined. Count all the particles having a maximum dimension greater than the prescribed size limit. The size limit and the permitted number of particles exceeding the limit are defined for each substance. Particle Size Characterization - The measurement of particle size varies in complexity depending on the shape of the particle and the number of particles characterized must be sufficient to insure an acceptable level of uncertainty in the measured parameters 1). For irregular particles, a variety of definitions of particle size exist. In general, for irregularly shaped particles, characterization of particle size must also include information on the type of diameter measured as well as information on particle shape. Several commonly used measurements of particle size are defined below (see Fig. 1): Fig. 1 Commonly used measurements of particle size Feret's Diameter - The distance between imaginary parallel lines tangent to a randomly oriented particle and perpendicular to the ocular scale. Martin's Diameter - The diameter of the particle at the point that divides a randomly oriented particle into two equal projected areas. Projected area Diameter - The diameter of a circle that has the same projected are as the particle. Length - The longest dimension from edge to edge of a particle oriented parallel to the ocular scale. Width - The longest dimension of the particle measured at right angles to the length. Particle Shape Characterization - For irregularly shaped particles, characterization of particle size must also include information on particle shape. The homogeneity of the powder should be checked using appropriate magnification. The following defines some commonly used descriptors of particle shape (see Fig. 2):

53 KP X 1867 Fig. 2 Commonly used descriptions of particle shape Acicular - Slender, needle-like particle of similar width and thickness. Columnar - Long, thin particle with a width and thickness that are greater than those of an acicular particle. Flake - Thin, flat particle of similar length and width. Plate - Flat particles of similar length and width but with greater thickness than flakes. Lath - Long, thin, and blade-like particle Equant - Particles of similar length, width and thickness; both cubical and spherical particles are included. General Observations - A particle is generally considered to be the smallest discrete unit. A particle may be a liquid or semisolid droplet; a single crystal or polycrystalline; amorphous or an agglomerate. Particles may be associated. This degree of association may be described by the following terms: Lamellar - Stacked plates. Aggregate - Mass of adhered particles. Agglomerate - Fused or cemented particles. Conglomerate - Mixture of two or more types of particles. Spherulite - Radical cluster. Drusy - Particle covered with tiny particles. Particle condition may be described by the following terms Edges - Angular, rounded, smooth, sharp, fractured. Optical - Color (using proper color balancing filters), transparent, translucent, opaque. Defects - Occlusions, inclusions. Surface characteristics may be described as: Cracked - Partial split, break, or fissure. Smooth - Free of irregularities, roughness, or projections. Porous - Having openings or passageways. Rough - Bumpy, uneven, not smooth. Pitted - Small indentations. Method 2. Analytical Sieving Method The analytical sieving method is a method to estimate the particle size distribution of powered pharmaceutical drugs by sieving. The particle size determined by this method is shown as the size of a minimum sieve opening through which the particle passes. Sieving is one of the oldest methods of classifying powders and granules by particle size distribution. When using a woven sieve cloth, the sieving will essentially sort the particles by their intermediate size dimension (i.e., breadth or width). Mechanical sieving is most suitable where the majority of the particles are larger than about 75 m. For smaller particles, the light weight provides insufficient force during sieving to overcome the surface forces of cohesion and adhesion that cause the particles to stick to each other and to the sieve, and thus cause particles that would be expected to pass through the sieve to be retained. For such materials other means of agitation such as air-jet sieving or sonic sifting may be more appropriate. Nevertheless, sieving can sometimes be used for some powders or granules having median particle sizes smaller than 75 m where the method can be validated. In pharmaceutical terms, sieving is usually the method of choice for classification of the coarser grades of single powders or granules. It is a particularly attractive method in that powders and granules are classified only on the basis of particle size, and in most cases the analysis can be carried out in the dry state. Among the limitations of sieving method are the need for an appreciable amount of sample (normally at least 25 g, depending on the density of the powder or granule, and the diameter of test sieves) and difficulty in sieving oily or other cohesive powders or granules that tend to clog the sieve openings. The method is essentially a two-dimensional estimate of size because passage through the sieve aperture is frequently more dependent on maximum width and thickness than on length. This method is intended for estimation of the total particle size distribution of a single material. It is not intended for determination of the proportion of particles passing or retained on one or two sieves. Estimate the particle size distribution as described under Dry Sieving Method, unless otherwise specified in the individual monograph. Where difficulty is experienced in reaching the endpoint (i.e., material does not readily pass through the sieves) or when it is necessary to use the finer end of the sieving range (below 75 m), serious consideration should be given to the use of an alternative particle-sizing method. Sieving should be carried out under conditions that do not cause the test sample to gain or lose moisture. The relative humidity of the environment in which the sieving is carried out should be controlled to prevent moisture uptake or loss by the sample. In the absence of evidence to the contrary, analytical test sieving is normally carried at ambient humidity. Any special conditions that apply to a particular material should be detailed in the individual monograph. Principles of Analytical Sieving - Analytical test sieves are constructed from a woven-wire mesh, which is of simple weave that is assumed to give nearly square apertures and is sealed into the base of an open cylindrical container. The basic analytical method involves stacking the sieves on top of one another in ascending degrees of coarseness, and then placing the test powder on the top sieve.

54 1868 Monographs, Part II The nest of sieves is subjected to a standardized period of agitation, and then the weight of material retained on each sieve is accurately determined. The test gives the weight percentage of powder in each sieve size range. This sieving process for estimating the particle size distribution of a single pharmaceutical powder is generally intended for use where at least 80 % of the particles are larger than 75 m. The size parameter involved in determining particle size distribution by analytical sieving is the length of the side of the minimum square aperture through which the particle will pass. TEST SIEVES Unless otherwise specified in the monograph, use the sieves listed in the Table 1. Sieves are selected to cover the entire range of particle sizes present in the test specimen. A nest of sieves having a 2 progression of the area of the sieve openings is recommended. The nest of sieves is assembled with the coarsest screen at the top and the finest at the bottom. Use micrometers or millimeters in denoting test sieve openings. [Note - Mesh numbers are provided in the table for conversion purposes only.] Test sieves are made from stainless steel or, less preferably, from brass or other suitable non-reactive wire. Calibration and recalibration of test sieves is in accordance with the most current edition of ISO ). Sieves should be carefully examined for gross distortions and fractures, especially at their screen frame joints, before use. Sieves may be calibrated optically to estimate the average opening size, and opening variability, of the sieve mesh. Alternatively, for the valuation of the effective opening of test sieves in the size range of 212 to 850 m, Standard Glass Spheres are available. Unless otherwise specified in the individual monograph, perform the sieve analysis at controlled room temperature and at ambient relative humidity. Cleaning Test Sieves - Ideally, test sieves should be cleaned using only an air jet or a liquid stream. If some apertures remain blocked by test particles, careful gentle brushing maybe used as a last resort. Test Specimen - If the test specimen weight is not given in the monograph for a particular material, use a test specimen having a weight between 25 and 100 g, depending on the bulk density of the material, and test sieves having a 200 mm diameter. For 76 mm sieves the amount of material that can be accommodated is approximately 1/7 th that which can be accommodated on a 200 mm sieve. Determine the most appropriate weight for a given material by test sieving accurately weighed specimens of different weights, such as 25, 50, and 100 g, for the same time period on a mechanical shaker. [Note - If the test results are similar for the 25 g and 50 g specimens, but the 100 g specimen shows a lower percentage through the finest sieve, the 100-g specimen size is too large.] Where only a specimen of 10 to 25 g is available, smaller diameter test sieves conforming to the same mesh specifications (table 1) may be substituted, but the endpoint must be re-determined. The use of test samples having a smaller mass (e. g. down to 5 g) may be needed. For materials with low apparent particle density, or for materials mainly comprising particles with a highly isodiametrical shape, specimen weights below 5 g for a 200 mm screen may be necessary to avoid excessive blocking of the sieve. During validation of a particular sieve analysis method, it is expected that the problem of sieve blocking will have been addressed. If the test material is prone to picking up or losing significant amounts of water with varying humidity, the test must be carried out in an appropriately controlled environment. Similarly, if the test material is known to develop an electrostatic charge, careful observation must be made to ensure that such charging is not influencing the analysis. An antistatic agent, such as colloidal silicon dioxide and/or aluminum oxide, may be added at a 0.5 percent (m/m) level to minimize this effect. If both of the above effects cannot eliminated, an alternative particle-sizing technique must be selected. Agitation Methods - Several different sieve and powder agitation devices are commercially available, all of which may be used to perform sieve analyses. However, the different methods of agitation may give different results for sieve analyses and endpoint determinations because of the different types and magnitude of the forces acting on the individual particles under test. Methods using mechanical agitation or electromagnetic agitation, and that can include either a vertical oscillation or a horizontal circular motion, or tapping or a combination of both tapping and horizontal circular motion are available. Entrainment of the particles in an air stream may also be used. The results must indicate which agitation method was used and the agitation parameters used (if they can be varied). End point Determination - The test sieving analysis is complete when the weight on any of the test sieves does not change by more than 5 % or 0.1 g (10 % in the case of 76 mm sieves) of the previous weight on that sieve. If less than 5 % of the total specimen weight is present on a given sieve, the endpoint for that sieve is increased to a weight change of not more than 20 % of the previous weight on that sieve. If more than 50 % of the total specimen weight is found on any one sieve, unless this is indicated in the monograph, the test should be repeated, but with the addition to the sieve nest of a more coarse sieve intermediate between that carrying the excessive weight and the next coarsest sieve in the original nest, i.e., addition of the ISO series sieve omitted from the nest of sieves. SIEVING METHODS Mechanical agitation Dry Sieving Method - Tare each test sieve to the nearest 0.1 g. Place an accurately weighed quantity of

55 KP X 1869 test specimen on the top (coarsest) sieve, and replace the lid. Agitate the nest of sieves for 5 minutes. Then carefully remove each from the nest without loss of material. If there is some fine powder on the down surface of each sieve, take if off by the brush gently, and combine it with the sieve fraction retained on each next down sieve. Reweigh each sieve, and determine the weight of material on each sieve. Determine the weight of material in the collecting pan in a similar manner. Reassemble the nest of sieves, and agitate for 5 minutes. Remove and weigh each sieve as previously described. Repeat these steps until the endpoint criteria are met (see Endpoint Determination under Test Sieves). Upon completion of the analysis, reconcile the weights of material. Total losses must not exceed 5 % of the weight of the original test specimen. Repeat the analysis with a fresh specimen, but using a single sieving time equal to that of the combined times used above. Confirm that this sieving time conforms to the requirements for endpoint determination. When this endpoint has been validated for a specific material, then a single fixed time of sieving may be used for future analyses, providing the particle size distribution falls within normal variation. If there is evidence that the particles retained on any sieve are aggregates rather than single particles, the use of mechanical dry sieving is unlikely to give good reproducibility, a different particle size analysis method should be used. Air Entrainment Methods Air jet and Sonic Shifter Sieving - Different types of commercial equipment that use a moving air current are available for sieving. A system that uses a single sieve at a time is referred to as air jet sieving. It uses the same general sieving methodology as that described under the Dry Sieving Method, but with a standardized air jet replacing the normal agitation mechanism. It requires sequential analyses on individual sieves starting with the finest sieve to obtain a particle size distribution. Air jet sieving often includes the use of finer test sieves than used in ordinary dry sieving. This technique is more suitable where only oversize or undersize fractions are needed. In the sonic sifting method a nest of sieves is used, and the test specimen is carried in a vertically oscillating column of air that lifts the specimen and then carries it back against the mesh openings at a given number of pulses per minute. It may be necessary to lower the sample amount to 5 g, when sonic shifting is employed. The air jet sieving and sonic sieving methods may be useful for powders or granules when mechanical sieving techniques are incapable of giving a meaningful analysis. These methods are highly dependent upon proper dispersion of the powder in the air current. This requirement may be hard to achieve if the method is used at the lower end of the sieving range (i. e., below 75 m), when the particles tend to be more cohesive, and especially if there is any tendency for the material to develop an electrostatic charge. For the above reasons endpoint determination is particularly critical, and it is very important to confirm that the oversize material comprises single particles and is not composed of aggregates. INTERPRETATION The raw data must include the weight of test specimen, the total sieving time, and the precise sieving methodology and the set values for any variable parameters, in addition to the weights retained on the individual sieves and in the pan. It may be convenient to convert the raw data into a cumulative weight distribution in terms of a cumulative weight undersize, the range of sieves used should include a sieve through which all the material passes. If there is evidence on any of the test sieves that the material remaining on it is composed of aggregates formed during the sieving process, the analysis is invalid. 1) Additional information on particle size measurement, sample size, and data analysis is available, for example, in ISO ) International Organization for Standardization (ISO) Specification ISO ; Test sieves-technical requirements and testing Principal sizes ISO Nominal Aperture Supplementary sizes R 20/3 R 20 R 40/3 US Sieve No. Recommended USP Sieves(mesh) European Sieve No. Korean Sieve No mm mm mm mm 9.50 mm 9.00 mm 8.00 mm 8.00 mm 8.00 mm 7.10 mm 6.70 mm 6.30 mm 5.60 mm 5.60 mm 5.60 mm mm 4.75 mm 4

56 1870 Monographs, Part II 4.50 mm 4.00 mm 4.00 mm 4.00 mm mm 3.35 mm mm 2.80 mm 2.80 mm 2.80 mm mm 2.36 mm mm 2.00 mm 2.00 mm 2.00 mm mm 1.70 mm mm 1.40 mm 1.40 mm 1.40 mm mm 1.18 mm mm 1.00 mm 1.00 mm 1.00 mm m 850 m m 710 m 710 m 710 m m 600 m m 500 m 500 m 500 m m 425 m m 355 m 355 m 355 m m 300 m m 250 m 250 m 250 m m 212 m m 180 m 180 m 180 m m 150 m m 125 m 125 m 125 m m 106 m m 90 m 90 m 90 m m 75 m m 63 m 63 m 63 m m 53 m m 45 m 45 m 45 m m 38 m

57 KP X Powder Particle Density Determination The Powder Particle Density Determination is a method to determine particle density of powdered pharmaceutical drugs or raw materials of drugs, and the gas displacement pycnometer is generally used. The powder density by this method is determined with an assumption that the volume of the gas displaced by the powder in a closed system is equal to the volume of the powder. The bulk density at loose packing or the tapped density at tapping express the apparent densities of the powder, since interparticular void volume of the powder is considered to be a part of the volume of the powder. On the contrary, the pycnometric particle density expresses the powder density nearly equal to the crystal density, since the volume of the powder, that is deducted with void volume of open pores accessible to gas, is counted. Powder particle density is expressed in mass per unit volume (kg/m 3 ), and generally expressed in g/cm 3. Apparatus The schematic diagram of particle density apparatus for gas displacement pycnometric measurement is shown in Figure. The apparatus consists of a test cell in which the sample is placed, a reference cell and a manometer. Generally, helium is used as the measurement gas. The apparatus has to be equipped with a system capable of pressuring the test cell to the defined pressure through the manometer. Calibration of apparatus The volumes of the test cell (V c ) and the reference cell (V r ) must be accurately determined to the nearest cm 3. In order to assure accuracy of the results of volume obtained, calibration of the apparatus is carried out as follows using a calibration ball of known volume for particle density measurement. The final pressure (P f ) are determined for the initial empty test cell followed by the test cell placed with the calibration ball for particle density measurement in accordance with the procedures, and V c and V r are calculated using the equation described in the section of Procedure. Calculation can be made taking into account that the sample volume (V s ) is zero in the first run. V r : Reference cell volume (cm 3 ) V c : Test cell volume (cm 3 ) V s : Sample volume (cm 3 ) M : Manometer (cm 3 ) Figure. Schematic diagram of a gas displacement pycnometer used to determine powder particle density out between 15 and 30 C, and the temperature must not vary by more than 2 C during the course of measurement. Firstly, weigh the mass of the test cell and record it. After weighing out the amount of the sample as described in the individual monograph and placing it in the test cell, seal the cell in the pycnometer. Secondly, introduce the measurement gas (helium) into the test cell, in order to remove volatile contaminants in the powder. If necessary, keep the sample powder under reduced pressure to remove the volatile contaminants in advance and use it as the test sample for measurement. Open the valve which connects the reference cell with the test cell, confirm with manometer that the pressure inside the system is stable, and then read the system reference pressure (P r ). Secondly, close the valve that connects to the two cells, and introduce the measurement gas into the test cell to achieve positive pressure. Confirm with the manometer that the pressure inside is stable, and then read the initial pressure (P i ). Open the valve to connect test cell with the reference cell. After confirming that the indicator of the manometer is stable, read the final pressure (P f ). Calculate the sample volume (V s ) with the following equation. V V r s Vc Pi Pr 1 Pf Pr V r : Reference cell volume (cm 3 ) V c : Test cell volume (cm 3 ) V s : Sample volume (cm 3 ) P i : Initial pressure (kpa) P f : Final pressure (kpa) P r : Reference pressure (kpa) Repeat the measurement sequence for the same powder sample until consecutive measurements of the sample volume agree to within 0.5 %, and calculate the mean of sample volumes (V s ). Finally, unload the test cell, weigh the mass of test cell, and calculate the final sample mass by deducing the empty cell mass from the test cell mass. The powder particle density ρ is calculated by the following equation. m V s ρ: Powder particle density (g/ cm 3 ) m: Final sample mass (g) V s : Sample volume (cm 3 ) Procedure The measurement of the particle density is carried