Ministério da Saúde
Acesse o Portal do Governo Brasileiro



Resolution - RE nº 896, of May 29, 2003
(D.O.U of 02/06/2003)

The Deputy of the Collegiate Board of Directors of the Brazilian Sanitary Surveillance Agency, in the use of the attribution vested in him by Administrative Order n. 238, of March 31, 2003,


provided in Article 111, clause II, item “a” of paragraph 3 of the Bylaws approved by Administrative Order 593, of August 25, 2000, re-published in the Federal Official Journal 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 6, 2003, decides:

Article 1 - To determine the publication of the "Guide for relative bioavailability/bioequivalence tests”, attached.

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




The relative bioavailability/bioequivalence studies must observe three stages: clinical, analytical and statistical, and must be planned and presented according to the GUIDE FOR PROTOCOL DESIGN OF RELATIVE BIOAVAILABILITY / BIOEQUIVALENCE STUDY and the GUIDE FOR TECHNICAL REPORT OF RELATIVE BIOAVAILABILITY/BIOEQUIVALENCE STUDY, respectively.

1. Clinical stage
a) the test and the reference drug products to be submitted to the bioequivalence study must, at first, be analyzed according to their monograph registered in the Brazilian Pharmacopoeia and, for lack of it, in other codes authorized by the legislation in force, according to the GUIDE FOR CARRYING OUT PHARMACEUTICAL EQUIVALENCE STUDY AND DESIGN OF REPORT. The difference in the drug proportions between the test and the reference drug products must not be greater than 5% (five per cent).

b) the bioequivalence study is generally undertaken through the quantification of the drug or the active metabolite in the circulation (blood, plasma or serum), or through its quantification in the urine, whenever justified. Alternatively, the study may be undertaken through the comparison of pharmacodynamic measurements.

c) in general, the unaltered drug must always be quantified. The metabolites must be quantified in the cases of analytical limitations for quantification of the unaltered drug, or when the unaltered drugs are active, significantly contributing to the effectiveness and safety of the product, having been formed, significantly, by pre-systemic metabolism. In the cases where the quantification of the drug and metabolite(s) is necessary, both must comply with the criteria established for bioequivalence determination.

d) the conventional study is of the open, random, crossed type. The subjects receive the test and the reference drug products on separate occasions (periods), under a scheme of single or multiple doses. A parallel design may be used whenever necessary.

e) the drug products must be administrated with a standard (usually 200mL) liquid (generally water) volume to the fasting volunteers.

f) the number of periods and the study sequences will be determined by the number of drug products analyzed, in order to ensure the statistical validity, according to the GUIDE FOR PLANNING AND CARRYING OUT THE STATISTICAL STAGE OF RELATIVE BIOAVAILABILITY/BIOEQUIVALENCE STUDIES. The interval between the periods must be, at least, of seven half-lives of the drug or the metabolite elimination.

g) the sample collection schedule must observe a length of time equal or greater than 3-5 times the half-life of elimination of the drug or the metabolite.

h) the number of healthy subjects must always ensure sufficient statistical power in order to guarantee the reliability of the results of the bioequivalence study. The number of volunteers is calculated through the variation coefficient and power of the test (see GUIDE FOR PLANNING AND CARRYING OUT STATISTICAL STAGE OF RELATIVE BIOAVAILABILITY/BIOEQUIVALENCE STUDIES), and a number less than 12 is not allowed. The protocol must establish a sufficient number of subjects in the study, foreseeing possible dropouts.
i) according to the drug product, the studies may be conducted in volunteers aged between 18 and 50 and capable of providing their free and informed consent, male, female or both, in which case the number of men and women must be the same and distributed equally between the sequences.
j) the weight of the subjects must be within a limit of ± 15% of the weight considered normal for men and women, considering height and physical structure.

k) smokers and subjects with a history of alcohol and drug abuse must be avoided. In case smokers are included, these subjects must be identified.

l) in the case of studies requiring subjects with different characteristics from those previously mentioned, their inclusion must be scientifically justified.

m) cytotoxic drug products must be tested on voluntary patients, suffering from the pathology for which the drug product is indicated, with their free and informed consent or that of their legal representative, in case of incapability of the patient.

n) in the case of drugs presenting long elimination half life (over 24 hours), an alternative collection schedule may be used, of up to 72 hours, allowing the determination of the area under the fragmented curve (ASC0-72), or a parallel study.

o) multiple dose studies are not generally recommended, since single dose studies are more sensitive to differences in formulations. However, multiple dose studies can be used in cases where, admittedly, they educe the inter-individual variability in the absorption process of the drug.

p) studies with feeding must be carried out for modified release forms (in addition to the fasting study) and for immediate release drug products with known interaction with foods.

q) studies that involve pharmacodynamic effect measurement are indicated in the cases where it is not possible to quantify the drug in circulation precisely and accurately, since its concentration is very low (for example: ophthalmic suspensions, local action creams, local action inhalants, etc).

r) the researcher must fill out a registration form of adverse events and list the adopted procedures for the control or treatment of the adverse effects.

s) the research project, the experimental protocol and the free and informed consent term must be submitted to a Committee of Ethics in Research accredited by the National Committee of Ethics in Research (CONEP) of the National Health Council/Ministry of Health (CNS/MS). The title of the project must include the drug name, the dose per unit, the pharmaceutical form and name of the manufacturers of the test and the reference drug products. This title must also be included in the experimental protocol, in the free and informed consent term, as well as in the report prepared by the Committee of Ethics in Research.

t) the subjects taking part in clinical studies requiring confinement, must stay in an appropriate place that meets the Good Clinical Practices (GCP), under the responsibility of a doctor.

u) in cases where it is necessary to transport the biological samples (plasma, serum or urine) the procedure for good laboratory practice must be followed in order to preserve the characteristics of the material to be analyzed. Use appropriate packing (certified) for preservation and transport. The temperature of the biological sample must be recorded with calibrated device to ensure maintenance of stability during the period of transport.

v) any protocol deviations must be reported.

2. Analytical stage

a) all the stages of the study must be carried out in compliance with international norms of Good Laboratory Practices (GLP);

b) the bioanalytical, chromatographic or other method, used for quantification of the drug in biological liquid, must be described in detail in the form of protocol or standard operational procedure (SOP) and must be validated for its application, according to the GUIDE FOR VALIDATION OF ANALYTICAL AND BIOANALYTICAL METHODS. The use of chromatographic methods is recommended;

c) the relation enters the concentration of the analyte and the response from the bioanalytical method must be reproducible and adequately defined, using a sufficient number of standards for the construction of the calibration curve;

d) analyte stability studies (drug or metabolite) must be carried out in the biological liquids, according to the GUIDE FOR VALIDATION OF ANALYTICAL AND BIOANALYTICAL METHODS;

e) the analytical protocol must contain criteria for the samples reanalysis. No more than 20% of the samples can be reanalyzed;

f) any loss of samples must be justified;

g) the analysis of the samples may be carried out in the following conditions: without replication, in duplicate or triplicate. For the analysis of samples in duplicate or triplicate, the acceptance criteria of the results must be described in the SOP;

h) all determinations with values lower than the Quantification Limit (QL), must be considered equal to zero for the statistical calculations;

i) the analytical protocol must contain the criteria for reintegration of the sample data;

j) any protocol deviations must be reported and justified.

3. Statistical stage

3.1. General methodology

3.1.1. the pharmacokinetic parameters are obtained from the blood concentration curves of the drug versus time, statistically analyzed for the determination of the bioequivalence:

3.2.1 the following pharmacokinetic parameters must be determined: the area under the blood concentration versus time, calculated using the method of the trapezoids, from time zero to time t (ASC0-t), where t is the time related to the last concentration determined through experimentation; the area under the blood concentration versus time, calculated from time zero to time infinite (ASC0-inf), where ASC0-inf = ASC0-t + Ct/lz, where Ct is the last concentration of the drug determined through experimentation and lz is the elimination constant of the terminal stage. The ASC0-t must be equal or greater than 80% of the ASC0-inf, except in the cases where the fragmented ASC is used; the peak of maximum concentration (Cmax) of the drug and/or metabolite and the time to reach this peak (Tmax) must be directly obtained without interpolation of values; the depuration (D), the apparent volume of distribution (Vd) and the half-life of elimination (t1/2) of the drug and/or the metabolite must also be determined, although there is no need for statistical treatment; for studies employing multiple doses the following parameters must be determined:
ASC0-t calculated in the interval of the dose (t) in steady state;

b) Cmax e Tmax, obtained without interpolation of data; minimum drug concentration (Cmin) determined at the end of each interval of the dose in steady state;

c) average concentration of the drug in steady state(C* = ASC0-t /t);

d) fluctuation rate in steady state. for the bioequivalence assessment the parameters ASC0-t, Cmax and Tmax must be employed; in the case of studies with multiple doses it must be proved that the steady state was obtained after the administration of the test and the reference drug products;

3.1.3. the exclusion of more than 5% of the subjects from the study or the lack of over 10% of the values for blood concentration of the drug resulting from administration of each drug product will not be accepted.


a) chart must be presented, containing individual values, average (arithmetic and geometric), standard deviation and variation coefficient of all the pharmacokinetic parameters related to the administration of the test and reference drug products;

b) it is recommended that the ASC0-t e Cmax parameters be transformed into natural logarithm, since, in general, the distribution of the transformed data is closer to a normal distribution in relation to the data not transformed. Justifications must be presented in the cases where it was chosen to carry out the statistical analysis in original scale data;

c) analysis of variance (ANOVA) of the pharmacokinetics parameters ASC0-t and Cmax must be carried out in order to evaluate the effects of sequence, of subjects, of period and of treatment. In addition, the ANOVA chart must be presented, containing source, degree of freedom, sum of squares, average square, F statistic, p figure and the intra and inter individual variation coefficients;

d) it is necessary to build a 90 % confidence interval (CI) for the ratio between the means of the values obtained with the test and the reference drug products, for the ASC0-t- and Cmax parameters. The CI antilogarithm obtained constitutes the CI of 90% for the ratio of the geometrical averages of the parameters

(ASC0-t teste/ ASC0-t referência e Cmax teste/ Cmax referência)

The construction of this CI must be based on the residual average square of the ANOVA obtained according to item c;

e) Tmax is analyzed as the individual difference: test(-)reference, building a 90% CI, using nonparametric test;

f) two drug products are considered bioequivalents when the 90% CI for the following ratios: ASC0-t test/ASC0-ref and Cmax test/Cmax ref between the means the pharmacokinetic parameters of test and reference are between 80 and 125%. Other limits of the 90% CI for Cmax, previously established in the protocol, may be accepted with scientific justifications. When clinically relevant, Tmax is also to be considered;

e) this CI based method is equivalent to the procedure of two corresponding one-sided tests with no hypothesis of bioequivalence, with a level of significance of 5% (µ=0,05);
f) for drugs presenting low therapeutic range, such as carbamazepine, ASC0-t, Cmax and Tmax parameters must be employed in bioequivalence assessment; for valproic acid, clindamycin and others, a 95% CI must be adopted;

g) validated statistical programs must be used;

h) whenever necessary, appropriate statistical models, depending on the kind of bioequivalence study (for example, of multiple doses), must be employed;

i) in cases of subjects presenting different behavior on the absorption parameters, in relation to the other subjects, their exclusion from the study must be justified. The results of the study must be presented with and without the inclusion of their data.

j) inform software programs used for the statistical analysis of the data.

Página Anterior Ínicio da página