Legislation
Resolution

Resolution - RE nº 475, of March 19, 2002
(Official Journal of 20/03/2002)

The Director of the Collegiate Board of Directors of the National Sanitary Surveillance Agency, in the use of the attributions vested in him under Administrative Rule 724, issued by the Director-Chairman, on October 10, 2000,

WHEREAS

paragraph 3 of article 111, of the Bylaws approved by Administrative Rule 593, of August 25, 2000, re-published in the Official Journal of the Union of December 22, 2000;
that the matter was submitted to the examination of the Collegiate Board of Directors, which approved the matter in a meeting held on March 13, 2002, decides:

Article 1 - To determine the publication of the "Guide for Validation of Analytical Methods", attached.

Article 2 - This Resolution enters into force on the date of its publication.

GONZALO VECINA NETO

GUIDE FOR THE VALIDATION OF ANALYTICAL METHODS - 1/2002

1. GENERAL CONSIDERATIONS

a) the information contained in this guide is applicable to bioanalytical methods such as gas chromatography (GC), high efficiency liquid chromatography, and these combined with mass spectometry, used for the quantitative drug determination and their metabolites in biological matrixes, such as blood, serum, plasma or urine. It is also applicable to other analytical techniques, like microbiological and immunological methods, or for other biological matrixes, although in these cases, a high degree of variability can be observed;

b) The validation must guarantee, through experimental studies, that the method meets the requirements of the analytical applications, ensuring the reliability of the results. For this, it must present suitable precision, accuracy, linearity, sensitivity and specificity. Thus, it is important to point out that all equipment and materials must be properly calibrated and the researchers must be qualified and properly trained;
c) reference chemical substances and/or biological standards made official by the Brazilian Pharmacopoeia or by any other code authorized by the current legislation must be used. Studies using secondary standards will be admitted provided their certification is proved, in the absence of reference chemical substances and/or biological pharmacopeial standards;

d) for the bioavailability and bioequivalence studies the internal standard must be used whenever the chromatographic methods are used. The impossibility of its use must be justified;

e) for the purpose of this guide, analytical run is the complete set of standards, samples and quality controls.

1.1. Precision

1.1.1. The method repeatability is verified through a minimum of 9 (nine) determinations contemplating the variation limit of the procedure, in other words, 3 (three) concentrations, low, medium and high, with 3 (three) responses each, or by 6 (six) determinations considering that the medium concentration corresponds to 100% of the expected;

1.1.2. the precision must be determined on one same run (intra-run precision) and in different runs (inter-run precision);

1.1.3. it may be expressed as relative standard deviation (RSD) or coefficient of variation (CV%), not accepting values greater than 15%, according to the formula:

where, SD stands for standard deviation and DMC stands for the determined mean concentration.

1.2. Accuracy

1.2.1. The accuracy of the method must be determined after the establishment of the linearity, its variation limit and specificity, being verified through a minimum of 9 (nine) determinations contemplating the variation limit of the procedure, in other words, 3 (three) concentrations, low, medium and high, with 3 (three) responses each. The assays must be undertaken on the same day (intra-day accuracy) and in different days (inter-days accuracy);

1.2.2. the accuracy must be determined in the same analytical run (intra-run accuracy) and in different runs (inter-run accuracy);

1.2.3. the accuracy is determined by the relation between the average concentration determined experimentally and the correspondent theoretical concentration:

1.3. Calibration/linearity curve

1.3.1. It is recommended that its determination should be done through samples analysis from the proper matrix, a minimum of 5 (five) different concentrations. Alternatives procedures must be justified;

1.3.2. when there is linearity, the results must be analyzed by proper statistical methods, as for example, the linear regression calculation by the least squares method. The obtained curves (experimental and the result of the mathematical treatment), the coefficient of linear correlation and the straight line intercept must be shown.

1.4. Intervals of the calibration curves

1.4.1. The interval of the calibration curve derives from the linearity study of the method and depends on the purpose of its application. The samples analyzed in the interval of the calibration curve must present linearity, accuracy and compatible precision;

1.4.2. minimum specifications for the calibration curve:

1.4.2.1. drugs and drug products analysis: 80 - 120% of theoretical concentration;

1.4.2.2. contents uniformity: 70 - 130% of theoretical concentration;

1.4.2.3. dissolution test: ± 20% above the specified interval;

1.4.2.4. impurities determination: from the level of the expected impurity to 120% of the maximum limit specified. When they present toxicological significance or unexpected pharmacological effects, the limits of quantification and detection must be adequate to the quantities of impurities to be controlled.

1.5. Specificity/selectivity

1.5.1. In the specificity studies of methods to determine the drug contents, the standard solution of the same must be analyzed, in the presence of known quantities of possible interfering agents (impurities/components/degradation products), demonstrating that the results are not affected by the presence of such components. For this, the results are compared to those obtained from the analysis of similar solutions without the drug . For the tests of impurity determination, the individual separation of the relevant interfering agents must also be demonstrated;

1.5.2. in the absence of standard of the degradation product, sub-product or impurity, the specificity of the method can be determined comparing the results of analysis of samples containing such components with the results of the analysis of the same samples using another well characterized and validated method. When applicable, in these cases, the samples must be submitted to stressful conditions: light, heat, humidity, hydrolysis and oxidation.

1.6. Quantification limit (QL)

1.6.1. Established by means of analysis of solutions containing decreasing concentrations of the drug to the lowest determinable level with acceptable precision and accuracy. It can be represented by the equation:

where: SD stands for the standard deviation of the intercept with the axis of Y of several calibration curves built containing concentrations of the drug near the presumed quantification limit. The standard deviation can also be obtained from the analysis of a suitable number of samples of the white; ic is the slope of the calibration curve;

1.6.2. the ratio of 5:1 between the signal and the noise of the base line can also be used, specifying the method used for the determination of the QL;

1.7. Detection limit (DL)

Established by means of analysis of solutions of known and decreasing concentrations of the drug , to the lowest detectable level. The DL should be 2 to 3 times superior to the noise of the base line. It can be represented by the equation:

where: SD stands for standard deviation of the intercept with the axis of Y of several calibration curves containing concentrations of the drug near the presumed quantification limit. The standard deviation can also be obtained from the analysis of a suitable number of samples of the white; ic is the slope of the calibration curve.

1.8. Robustness

The robustness assessment must be considered during the phase of method development. If susceptibility to variations is verified in the analytical conditions, they must be properly controlled or precautions must be included in the procedure.

Variation examples:

a) Stability of the analytical solutions;
b) time of extraction;
Typical variations in liquid chromatography:
a) influence of the pH variation of the mobile phase;
b) influence of the variation of the mobile phase composition;
c) different columns (different batches and/or manufacturers);
d) temperature;
e) flow rate.

2. RELEVANT SPECIFIC CONSIDERATIONS FOR THE STABILITY STUDY

1.1. The analytical method employed must be an indicator of stability, demonstrating specificity and sensitivity for the degradation products likely to be formed, but not necessarily the same used in the test of content determination;

1.2. the analytical method for the undertaking of the stability study has to be validated observing the parameters of accuracy, precision, linearity, detection limit, quantification limit, specificity, variation limit and robustness. This validation must be carried out in the presence of the sub-products and/or degradation products. In the absence of standards, the samples must be submitted to stressful conditions: light, heat, humidity, hydrolysis and oxidation.

3. STABILITY STUDY OF DRUGS IN BIOLOGICAL LIQUIDS

a) The drug stability in biological liquids depends on its chemical properties, the biological matrix and the packing material used. The stability determined for a type of matrix and a specific packing material can not be extrapolated to others;

b) the stability determinations must use a set of samples, prepared from a recent solution reserve of the drug analyzed, added to a biological matrix free of interference.

3.1. Short term stability

3.1.1. Stability in freezing and defrosting cycles

The drug stability must be tested during three freezing and defrosting cycles using a minimum of three samples of low and high concentrations determined in the validation of the analytical method, in the following conditions: the samples must be frozen at -20 ºC, or other temperature indicated for the storage (for example, -70ºC) and maintained for 24 hours, after which they are defrosted at room temperature. When completely defrosted, the samples must be frozen again at -20ºC, for 12 to 24 hours, and then successively, till the three cycles are complete, the drugs in the samples being quantified after the third cycle.

3.1.2. Stability in time and analysis conditions

3.1.2.1. The drug must remain stable during the analysis. To verify this property, a minimum of three samples of low and high concentrations determined in the validation of the analytical method are used. Each one of them will be submitted to natural defrosting, at room temperature, and kept in this condition for the maximum expected time of the analysis duration;

3.1.2.2. If equipment is used that employs automatic systems of sampling/injection, with devices for temperature control (for example, refrigeration), a study of the drug stability must be undertaken, in the sample used in the analysis procedure, including the internal standard, at the temperature in which the test is being carried out.

3.2. Long term stability

3.2.1. The period of storage for the study of long term stability must exceed the interval of time comprised between the first sampling and the analysis of the last, according to the schedule presented in the protocol of the bioequivalence or the bioavailability study;

3.2.2. The temperature used in the assay must reproduce the one recommended for the samples storage, normally equal to -20ºC. The results must be compared with the average of those verified on the first day of the study. To verify this property, a minimum of three samples of the low and high concentrations determined in the validation of the analytical method must be used.

3.3. Stability of the standard solutions

3.3.1. The standard solutions stability of the drug and the internal standard in the biological liquid, at room temperature, must be evaluated after a minimum of six hours of its preparation.

3.3.2.such solutions must be refrigerated or frozen for seven to fourteen days, or other appropriate period of time;

3.3.3. The results of this test must be compared with the ones obtained using recent solutions prepared from the drug and internal standard in the biological liquid.

3.4. Statistic analysis of the results

Whatever the statistic method used to evaluate the stability study results, it must be described clearly in the standard operational procedure (SOP).

4. RELEVANT SPECIFIC CONSIDERATIONS FOR BIOANALYTICAL METHODS

4.1. Pre-study validation

4.1.1. Specificity

a) Analyze samples from the biological matrix (blood, plasma, serum, urine or other) obtained from six subjects, four being normal samples, one lipaemic and one hemolytic, under controlled time conditions, feeding and other important factors for the study. Each sample of the white must be tested using the procedure and chromatographic and spectrophotometric conditions proposed. The results must be compared with the ones obtained with aqueous solution of the analyte, in concentration levels near to QL;

b) Any sample of the white presenting significant interference in the period of time of retention of the drug , metabolite or internal standard, must be rejected. If one or more of the analyzed samples present such interference, new samples of other six subjects must be tested. If one or more of the samples of this group present significant interference in the period of time of the drug retention, the method must be altered with the purpose of eliminating it;

c) The interference agents must be components of the biological matrix, metabolites, decomposition products and drug products used concomitantly to the study. The interference of the nicotine, caffeine, OTC products and metabolites must be considered whenever necessary;

d) If the method purpose is the quantification of more than one drug , each one must be injected separately to determine the period of time of individual retention and insure that impurities from one drug did not interfere in the other's analysis.

4.1.2.Calibration/linearity curve

4.1.2.1. The calibration curve (standards) represents the ratio between the response of the instrument and the known analyte concentration. A calibration curve must be built for each drug using the same biological matrix proposed for the study. The calibration curve must include the analysis of the sample of the white (biological matrix exempt of the drug and internal standard), of the sample zero (biological matrix plus the internal standard) and five to eight samples containing drug standard and internal standard, contemplating the expected variation limit (80% of the lowest concentration and 120% of the highest concentration intended to be analysed), including the QL.

4.1.2.2. Factors to be considered in the assessment of the calibration curve:

a) deviation less than or equal to 20% (twenty per cent) in relation to the nominal concentration for the QL;

b) deviation less than or equal to 15% (fifteen per cent) in relation to the nominal concentration for the other concentrations of the calibration curve;

c) at least four of the six concentrations of the calibration curve must comply with the former criteria, including the QL and the largest concentration of the calibration curve;

d) the coefficient of linear correlation must be equal or superior to 0,95.

4.1.3. Quantification limit (QL)

4.1.3.1. No significant interference must be presented by the sample of the white in the period of time of the drug retention. The QL must be at least five times superior to any interference of the sample of the white during the drug retention;

4.1.3.2. the response peak of the drug in the QL must be identifiable and reproducible with 20% (twenty per cent) of precision and 80% (eighty per cent) - 120% (hundred and twenty per cent) of accuracy;

4.1.4. Precision

At least three concentrations (low, medium and high) within the range of the expected limit must be analyzed and at least five replications must be made. The CV should not exceed 15% (fifteen per cent), except for the QL, for which values less than or equal to 20% (twenty per cent) are allowed.

Precision must be determined in a single analytical run (intra-run precision) and in different runs (inter-run precision), according to the description in item 1.1 of this annex;

4.1.5. Accuracy

Is determined by the analysis of samples containing known quantities of drug , in three concentrations (low, medium and high) within the range of the expected limit, performing at least five replications. The deviation should not exceed 15% (fifteen per cent), except for the QL, for which values less than or equal to 20% (twenty per cent) are allowed. Accuracy must be determined in a single analytical run (intra-run precision) and in different runs (inter-run precision), according to the description in item 1.1 of this annex;

4.1.6. Recovery

The recovery measures the procedure efficiency of extraction of an analytical method within a variation limit. Recovery percentages near to 100% are desirable, nevertheless lower values are accepted, for example, 50 to 60%, provided the recovery is precise and accurate. This test must be done comparing the analytical results of samples extracted from three concentrations (low, medium and high) with the results obtained with standard solutions not extracted, presenting 100% of recovery;

4.2. Quality control (QC)

4.2.1. QC of the quantification limit (QC-QL): same concentration as the QL;

4.2.2. QC of low concentration (QCL): less than or equal 3 x QL;

4.2.3. QC of medium concentration (QCM): approximately the mean between QCL and QCH;

4.2.4. QC of high concentration (QCH): 75 to 90% of the largest concentration of the calibration curve;

4.3. Approval criteria

The analytical method is considered valid when it meets the following criteria:

4.3.1. Precision: the VCs calculated from biological matrixes obtained from a minimum of three subjects, for QCL, QCM and QCH must be less than or equal to 15%, and less than or equal to 20% for QC-QL;

4.3.2. Accuracy: must present values comprised within more or less than 15% of the nominal value for QCL, QCM and QCH and more or less than 20% for QC-QL, calculated from biological matrixes obtained from a minimum of three subjects;

4.3.3. Sensitivity: the smallest concentration of the calibration curve can be accepted as the QL of the method when the VC for QC-QL, calculated from biological matrixes of a minimum of three subjects is inferior or equal to 20%;

4.3.4. Specificity: the response for the interfering peaks at the same retention time of the drug must be less than 20% of the QL response. The response for the interfering peaks at the retention time of the drug and the internal standard must be less than 20% and 5%, respectively, from the response in the concentration used;

Obs: with the analytical method validated, its precision and accuracy must be monitored continuously to ensure a satisfactory performance. To reach this target, six samples of quality control (two QCL, two QCM and two QCH) must be analyzed, together with the other samples, at adequate intervals, depending on the total number of samples. The results of the QC samples will serve as base for the acceptation or rejection of the analytical run. A minimum of 67% (four out of six QC samples may present deviation of more or less than 20% of its respective nominal value; 33% (two out of six) QC samples may be outside these limits, but not for the same concentration. All the samples from the same subject must be present in single analytical run.