Guideline 3
V. Guideline For Approval Of A Method Of Analysis For Residues
A. INTRODUCTION
Before approving a carcinogenic or noncarcinogenic compound for use in food-producing animals, FDA generally asks that the sponsor provide an acceptable analytical method (either chemical or biological) capable of reliably measuring the marker residue to ensure that the total residue of toxicological concern is not exceeded. The determination of the total residue of toxicological concern can be accomplished by: (a) choosing a marker residue, (b) establishing a quantitative relationship between the marker residue and total residue of toxicological concern, and (c) calculating the maximum permitted concentration of marker residue, R(M), in the target tissue (the tissue used to monitor for total drug residues in all tissues) to ensure that the total residue of toxicological concert does not exceed the permitted concentration.
The process of obtaining approval of a method consists of three steps:
Step 1. Method development by the sponsor and demonstration that the method satisfies the acceptability criteria (part B of this guideline).
Step 2. FDA desk review of the sponsor's data (part C I of this guideline) to determine suitability of the method for nterlaboratory study.
Step 3. Interlaboratory Study (part C 2 of this guideline) to determine whether the method performs as claimed and thus can be used as a practicable and reliable regulatory tool.
Because methods which appear marginally acceptable after desk review often do not pass the interlaboratory validation trials, sponsors are urged to develop methods that are rugged, and exceed rather than meet the minimal standards of acceptability.
B. EVALUATION CRITERIA
Any method will be characterized by a set of attributes that determine its applicability: specificity (what is being measured), precision (the variability of the measurement), and systematic error (or bias, measured as recovery).
1. Specificity
Specificity is the ability of a method to respond only to the substance being measured. The proposed method must provide for identification of the compound being measured. Certain instrumental techniques such as infrared spectroscopy or mass spectrometry may be sufficiently specific by themselves. If the method is not sufficiently specific, then a confirmatory or identification procedure will be needed. In those cases, the regulatory method will have two components, the "determinative" procedure to quantify a given compound and the "confirmatory" procedure to verify the identity of the compound. Other techniques can achieve comparable specificity if used in combination. For example, specificity may be verified by thin-layer chromatography, element-specific gas liquid chromatography, formation of characteristic derivatives and rechromatography, and characteristic relative retention times with several different chromatographic systems of different polarity. Such procedures must be applicable at the designated marker residue concentration R(M).
The sponsor should take into account the possible presence of other compounds approved for use in the same target species and should demonstrate that they will not interfere with the determination of the sponsored compound by the proposed method.
2. Precision
Precision is an important quantitative performance characteristic of a method. It is a measure of the variability of repetitive measurements. Contributions of variability from numerous sources affect precision, but the major components are those from different laboratories (reproducibility) and those from within a laboratory (repeatability). Precision is usually expressed as a standard deviation, but an even more useful term is the relative standard deviation (or coefficient of variation) because it is relatively constant over a considerable concentration range.
The variability finally achieved in the sponsor's laboratory after considerable experience usually is less than that achievable by less experienced laboratories who may later use the method. The final version of the method should be optimized by such procedures as ruggedness testing (W. J. Youden and E. H. Steiner, Statistical Manual of the AOAC, Association of Official Analytical Chemists, Box 540 Benjamin Franklin Station, Washington, DC 1975, p. 33). If a method cannot achieve a suitable degree of repeatability in the sponsor's laboratory, it cannot be expected to do any better in other laboratories. The method should be performed by an analyst not involved in the development of the method to verify the adequacy of the method's description and the identification of critical parameters.
The within laboratory coefficient of variation should not exceed 10% where the designated concentration of marker residue R(M) is greater than or equal to 0.1 ppm. Where the designated concentration of marker residue R(M) is less than 0.1 ppm, the within laboratory coefficient of variation should not exceed 20%.
3. Systematic Error
Systematic error, or bias, is the difference of the measured value from the true, assigned, or accepted value. It is generally expressed as the percent recovery of added analyte, realizing, in the case of residue analysis, that analyte added to a sample may not behave in the same manner as the same analyte biologically incurred. At relatively high concentrations, recoveries are expected to approach 100%. At lower concentrations, and particularly with methods involving a number of steps which may include extractions, solvent transfers, and adsorption chromatography, recoveries may be lower.
Assuming acceptable precision, an average recovery of 80 to 110% should be obtained when the designated concentration of marker residue R(M) is 0.1 ppm or greater. An average recovery of 60 to 110% will be accepted when the designated concentration of marker residue R(M) is less than 0.1 ppm. Correction factors are not ordinarily acceptable unless they are an inherent part of the procedure, as in definitive radioisotope dilution procedures. Since higher recoveries (greater than 110%) may indicate a lack of specificity, FDA will ask for an explanation for such values. FDA will accept lower recovery values than those presented above for methods employing internal standards.
4. Collateral Criteria
The method should: (a) utilize commercially available reagents, supplies, instruments (except that new or unusual reagents or standards may be supplied by the sponsor on request); (b) be capable of being performed by reasonably experienced analysts: (c) be capable of being completed within reasonable time periods consistent with regulatory objectives (usually no more than 48 hours of total elapsed time); (d) not need unique instrumentation, large quantities of solvents, reagents, and supplies which would render the method economically impractical; and (e) be capable of being performed safely.
There are several other indications of satisfactory performance that may be helpful in determining if the method is acceptable. These factors may include requirements for a linear calibration (standard) curve and analytical (recovery) curve; effectiveness of extraction; the effect (or noneffect) of specific potential interferences; adequate sensitivity (slope of the calibration curve) and resolution; adequately low and constant blanks; and stability studies.
In practical regulatory use, samples are examined without benefit of treatment history. For interpretation of the analytical response observed in analyses of actual samples, the response for the marker residue should be clearly resolved from any other responses present and should be readily distinguishable above the background signal. (For example, in a gas chromatographic determination, the marker residue peak(s) should be resolved and clearly recognizable from other peaks. At the designated concentration of marker residue, R(M), the response should be a peak at least 20 mm in height and at least 10 times greater than the variability of the background response.)
C. SPECIFIC DATA NEEDED
1. Sponsor's Petition
In development of regulatory methods, sponsors should collect data from three types of samples: (a) "control" target tissue from untreated animals, (b) "fortified" target tissue containing known concentrations of the marker residue added to the sample of control target tissue, and (c) "dosed' target tissue from animals of the target species that have been treated with the drug. The sponsor should determine the baseline (background) response of the method and its variability. The sponsor should demonstrate that the proposed method can satisfactorily recover and identify known amounts of the marker residue which have been added to the target tissue. Finally, the sponsor should demonstrate that the proposed method can satisfactorily recover the biologically incurred marker residue.
In presenting a petition for approval of a proposed regulatory method, the sponsor should provide:
* A complete description of the method including sampling, preparation of analytical samples, storage conditions, reagents, instrumentation, standards, and identification of critical steps and stopping places.
* Quality control criteria that may be needed to verify
and maintain method performance.
* A typical standard curve prepared from marker residue of known purity.
* A typical analytical curve prepared by fortifying (spiking) control tissue with marker residue and observing the resulting analytical responses.
* Data derived from control, fortified, and dosed tissue showing that the method meetsthe specificity, precision, and systematic error attributes.
* Relevant worksheets, calculations, statistical analyses, spectrograms, chromatograms, etc., from the analyses of control, fortified and dosed target tissue.
* Using the proposed determinative procedure of the method, results of analyses of the following samples (as a minimum):
- 5 control tissues
- 5 control tissues fortified with marker residue at 0.5X R(M)
- 5 control tissues fortified with marker residue at 1X R(M)
- 5 control tissues fortified with marker residue at 2X R(M)
- 5 dosed tissues containing biologically incurred total residue at approximately its permitted concentration.
* Using the proposed confirmatory procedure, results of analyses to verify the identity of the marker residue with the following samples (as a minimum):
- 5 control tissues to ensure the absence of false positives
- 5 control tissues fortified with marker residue at R(M)
- 5 dosed tissues containing biologically incurred total residue at approximately its permitted concentration.
2. Interlaboratory Study
After the regulatory method has passed desk review (section 1, above), FDA will conduct an interlaboratory trial of the proposed method to verify that the regulatory method can be employed as a practicable and reliable regulatory tool. At a minimum, the method will be tested in two FDA laboratories and one USDA laboratory using target tissues supplied by the sponsor.
Each of the three laboratories will analyze the following samples by the determinative procedure:
* 5 control tissues
* 5 control tissues fortified with marker residue at 0.5X R(M)
* 5 control tissues fortified with marker residue at 1X R(M)
* 5 control tissues fortified with marker residue at 2X R(M)
* 5 dosed tissues containing biologically incurred total residue at approximately its permitted concentration.
If a separate procedure is necessary to confirm the identity of the marker residue, each of the three laboratories will analyze the following samples:
* 5 control tissues
* 5 control tissues fortified with marker residue at R(M)
* 5 dosed tissues containing biologically incurred total residue at approximately its permitted concentration.
(Whenever possible the confirmatory procedure should be performed on the same extract used for the determinative portion of the method.)
Alternatively, the sponsor may elect to conduct a collaborative study of the proposed method under the auspices of an organization such as the Association of Official Analytical Chemists (AOAC). The study would include the three aforementioned government laboratories and the sponsor's laboratory. The sponsor would be responsible for obtaining the additional collaborators and for supplying the participating laboratories with all the necessary target tissues (control, fortified, and dosed). A minimum of six laboratories would be needed to constitute a collaborative study. The number of tissue samples analyzed by each participating laboratory can be reduced from five to a minimum of two for each group listed above.
|
||||||
 |
||||||
|
||||||