AOAC Gluten Quantitative Validation Guidance-Round 1(Nov 2023)

208

4.

Method Developer Validation Study

Quantitative methods are those whose result is the amount (mass or concentration) of the analyte. This guidance has been 209 developed for use with candidate methods that are designed to quantify gluten. If a candidate method’s intended use is not 210 covered by this document or existing standard method performance requirements (SMPRs), the standing AOAC expert review 211 panel (ERP) for gluten, or other qualified agency, may determine the appropriate cross-reactivity/interference panels, and 212 performance requirements. 213 Method developers may prepare study test materials in-house for the single laboratory validation (method developer study), 214 but all test materials and test portions must be blind-coded and randomized. Analyses conducted by the method developer must 215 be performed by an independent analyst without prior knowledge of the test materials undergoing analysis. Ideally, all test 216 materials for the independent laboratory and collaborative studies should be prepared by an external entity independent from 217 the method developer. At least one incurred test material for the independent laboratory and collaborative studies must be 218 prepared by an external entity independent from the method developer. In situations where an independent entity is unavailable 219 to prepare all of the test materials for the independent laboratory and collaborative studies, or their use is impractical for all test 220 materials, method developers may produce and distribute test materials as long as detailed information is provided on 221 procedures used to prevent bias (preparation, coding, etc.), and justification is provided for failing to use an independent entity 222 to prepare all of the test materials. 226 performance of a candidate method in the following areas: (1) calibration fit, (2) selectivity, (3) precision (repeatability and 227 intermediate precision), (4) LOD/LOQ, (5) recovery, and (6) robustness. These studies are generally conducted within a method 228 developer laboratory. 229 Gluten has multiple potential sources – wheat, rye, barley, oats and their hybrids and crossbreeds – and multiple regulatory 230 levels. Developers must determine which of these sources and levels their method is intended to detect, and perform matrix 231 studies for each claimed gluten source. 232 4.2 Calibration Fit Study 233 Analyze calibration standards as they are included in the test kit, or prepared as described in the test method. Analyze at 234 least four replicates of each concentration defined for the calibration curve. Fit the calibration curve as described in the method 235 instructions and/or kit insert. Full descriptions must be provided with respect to performing the calibration function calculations, 236 including any transformations conducted and the regression model used. Full calibration curve plots and calibration functions 237 must be shown. 238 From the calibration curve function, determine the calculated concentrations for each of the standards. Calculate the 239 residuals for each concentration standard. Residuals are the difference between the observed value and the predicted value for 240 each dependent variable in the calibration curve. (Residual = observed value - predicted value.) Residuals should be calculated 241 from the instrument response. For most quantitative gluten methods, instrument response would be optical density (absorbance) 242 values. 243 Plot the residuals versus concentration. Residuals should have random distributions and be centered on zero. If a non- 244 random pattern is observed, the calibration function or measurement range may not be appropriate. Residuals should generally 245 also be <15% of the measured response, and up to 20% at the lowest non-zero calibration standard. 246 4.3 Selectivity Study 247 The selectivity study is intended to provide information on potential sources of cross-reactivity and interference. The 248 information related to cross-reactivity and interference should be reported in the validation report or in the package insert from 249 the method developer 250 Breadth 251 This section of the validation is intended to provide information to end users on the method’s performance with less 252 common varieties of gluten-containing grains, such as einkorn, spelt and emmer. 253 The materials identified in Annex A, Table 1, should be tested at three times the LOQ of the method (as long as that is equal 254 to or below 20 mg/kg, otherwise test at 20 mg/kg) in a rice flour matrix. Test six test portions per test material. 255 The absorbance or optical density (OD) values for all test portions and standards must be reported. The mean gluten 256 concentration for each gluten source must be reported. Mean concentrations below the LOQ should be reported as Below the 257 223 224 4.1 Scope 225 A Single Laboratory Validation (SLV) study (also referred to as a Method Developer Study), is intended to evaluate the