Essential terminology and considerations for validation of non-targeted methods.

Through their suggestive name, non-targeted methods (NTMs) do not aim at a predefined "needle in the haystack." Instead, they exploit all the constituents of the haystack. This new type of analytical method is increasingly finding applications in food and feed testing. However, the concepts, terms, and considerations related to this burgeoning field of analytical testing need to be propagated for the benefit of those associated with academic research, commercial development, or official control. This paper addresses frequently asked questions regarding terminology in connection with NTMs. The widespread development and adoption of these methods also necessitate the need to develop innovative approaches for NTM validation, i.e., evaluating the performance characteristics of a method to determine if it is fit-for-purpose. This work aims to provide a roadmap for approaching NTM validation. In doing so, the paper deliberates on the different considerations that influence the approach to validation and provides suggestions therefor.

Measurement uncertainty interval in case of a known relationship between precision and mean.

Background: Measurement uncertainty is typically expressed in terms of a symmetric interval y±U, where y denotes the measurement result and U the expanded uncertainty. However, in the case of heteroscedasticity, symmetric uncertainty intervals can be misleading. In this paper, a different approach for the calculation of uncertainty intervals is introduced. Methods: This approach is applicable when a validation study has been conducted with samples with known concentrations. In a first step, test results are obtained at the different known concentration levels. Then, on the basis of precision estimates, a prediction range is calculated. The measurement uncertainty for a given test result can then be obtained by projecting the intersection of the test result with the limits of the prediction range back onto the axis of the known values, now interpreted as representing the measurand. Results: It will be shown how, under certain circumstances, asymmetric uncertainty intervals arise quite naturally and lead to more reliable uncertainty intervals. Conclusions: This article establishes a conceptual framework in which measurement uncertainty can be derived from precision whenever the relationship between the latter and concentration has been characterized. This approach is applicable for different types of distributions. Closed expressions for the limits of the uncertainty interval are provided for the simple case of normally distributed test results and constant relative standard deviation.

Development of Non-Targeted Mass Spectrometry Method for Distinguishing Spelt and Wheat.

Food fraud, even when not in the news, is ubiquitous and demands the development of innovative strategies to combat it. A new non-targeted method (NTM) for distinguishing spelt and wheat is described, which aids in food fraud detection and authenticity testing. A highly resolved fingerprint in the form of spectra is obtained for several cultivars of spelt and wheat using liquid chromatography coupled high-resolution mass spectrometry (LC-HRMS). Convolutional neural network (CNN) models are built using a nested cross validation (NCV) approach by appropriately training them using a calibration set comprising duplicate measurements of eleven cultivars of wheat and spelt, each. The results reveal that the CNNs automatically learn patterns and representations to best discriminate tested samples into spelt or wheat. This is further investigated using an external validation set comprising artificially mixed spectra, samples for processed goods (spelt bread and flour), eleven untypical spelt, and six old wheat cultivars. These cultivars were not part of model building. We introduce a metric called the D score to quantitatively evaluate and compare the classification decisions. Our results demonstrate that NTMs based on NCV and CNNs trained using appropriately chosen spectral data can be reliable enough to be used on a wider range of cultivars and their mixes.

Reliability of precision data obtained from interlaboratory studies.

Precision data, such as laboratory-to-laboratory SD (SL) and repeatability SD, obtained from interlaboratory tests are needed to assess analytical test methods. These precision data describing random error are subject to random variation. In order to avoid distorted assessments of test methods, interlaboratory tests must fulfill minimal requirements for achieving, e.g., a desired reliability in S(L). In 2009, McClure and Lee considered reliability of S(L) as a characteristic of an interlaboratory study. They developed an approach to approximate that reliability to make it possible to adapt the study design of an interlaboratory study to a desired reliability in S(L). The McClure and Lee approach introduces the "margin of relative error" to arrive at the magnitude of the uncertainty in S(L). This article discusses their approach and presents a generalized approach. The limitations of McClure and Lee's approximation are shown to result in underestimation of the actual variability of S(L) due to the disregard of the inherent negative bias of S(L). This bias corresponds to the fact that the expected value of the obtained S(L) lies below the true value sigmaL one would obtain in an interlaboratory study with an infinite number of laboratories and replicates. In order to achieve the reported level of reliability in S(L), the actual number of laboratories required is typically approximately 25% higher than that calculated by McClure and Lee. We present a generalized approach using "margins of relative random error," which takes the impact of the bias of the S(L) into account, resulting in a more realistic estimation of the variability of the precision parameter S(L).

Validation of a multi-residue method for the determination of several antibiotic groups in honey by LC-MS/MS.

The presented multi-method was developed for the confirmation of 37 antibiotic substances from the six antibiotic groups: macrolides, lincosamides, quinolones, tetracyclines, pleuromutilines and diamino-pyrimidine derivatives. All substances were analysed simultaneously in a single analytical run with the same procedure, including an extraction with buffer, a clean-up by solid-phase extraction, and the measurement by liquid chromatography tandem mass spectrometry in ESI+ mode. The method was validated on the basis of an in-house validation concept with factorial design by combination of seven factors to check the robustness in a concentration range of 5-50 µg kg(-1). The honeys used were of different types with regard to colour and origin. The values calculated for the validation parameters-decision limit CCα (range, 7.5-12.9 µg kg(-1)), detection capability CCß (range, 9.4-19.9 µg kg(-1)), within-laboratory reproducibility RSD(wR) (<20% except for tulathromycin with 23.5% and tylvalosin with 21.4 %), repeatability RSD(r) (<20% except for tylvalosin with 21.1%), and recovery (range, 92-106%)-were acceptable and in agreement with the criteria of Commission Decision 2002/657/EC. The validation results showed that the method was applicable for the residue analysis of antibiotics in honey to substances with and without recommended concentrations, although some changes had been tested during validation to determine the robustness of the method.

Preparation and characterisation of in-house reference material of tylosin in honey and results of a proficiency test.

The analysis of incurred material from animals treated with pharmacologically active substances is an efficient way to check the accuracy of a method. Tylosin A was chosen for the preparation of that material because it is highly effective in controlling active infections of American Foulbrood (AFB), a global threat to apiculture, but residues in honey are not allowed according to European legislation. For this reason an in-house reference material of honey containing the macrolide tylosin A and its degradation product desmycosin (tylosin B) was prepared. After the treatment of a beehive with the appropriate macrolide tylosin A, the honey samples were collected. The incurred honey material was diluted by mixing with blank honey. Concentrations of 25.81 µg kg(-1) for tylosin A and of 19.28 µg kg(-1) for its degradation product desmycosin (tylosin B) were reached. The homogeneity was checked by analysing 12 bottles in duplicate. The stability was tested at different defined temperatures and storage conditions. The reference material described above was homogeneous and stable. Samples of this in-house reference material were used for the realisation of a proficiency test with international participation. All participants accomplished satisfying results with the exception of one laboratory.

Validated determination of eight antibiotic substance groups in cattle and pig muscle by HPLC/MS/MS.

The described multimethod is suited for the determination of 53 substances of eight antibiotic groups-tetracyclines, quinolones, macrolides, sulfonamides, diphenylsulfones, diamino-pyrimidine derivatives, pleuromutilines, and lincosamides-in cattle and pig muscle. All substances were analyzed simultaneously with the same sample preparation and in one HPLC/MS/MS run. The validation of the multimethod was successfully accomplished with the help of an alternative in-house validation concept requiring only 48 experiments. The substances were validated at concentrations of 0.25, 0.5, 1.0, 1.5, and 2.0 x MRL (maximum residue limit) or 5, 10, 20, 30, and 40 microg/kg for substances without an MRL. The calculated relevant validation parameters were based on and comply with the requirements of Commission Decision 2002/657/EC, i.e., the decision limit, detection capability, repeatability, within-laboratory reproducibility, and recovery. The robustness of the method was demonstrated by varying seven factors of the analytical procedure. Several proficiency tests were carried out successfully to provide evidence for the applicability of the method.

The validation of methods for regulatory purposes in the control of residues.

The topic of validation is diversified. This review outlines the validation strategies which can be found in national, international and supranational regulations, compares them with one another and aims to elaborate on the main principles. European regulations and legislation, Codex alimentarius guidelines, the official methods program of the AOAC, and naturally the relevant ISO standards, particularly the ISO 5725 series, are taken into consideration. The objective of every validation is to demonstrate fitness for purpose. This varies of course in its characteristics for the diverse uses. However, all approaches have in common the objective of harmonisation of food control by using effective and reliable methods. To this end, criteria are determined and validation models developed and made compulsory. ISO 5725 is the central basis for validations for quantitative methods with its validation specifications through method collaborative studies. On the contrary, there are no valid uniform international method specifications for qualitative methods. Collaborative studies are in opposition to single-lab-validations with different sources of error. Whereas laboratory errors are predominant in collaborative studies, the single-lab-validation or in-house validation concentrates particularly on time and processing errors (intermediate precision). In new statistical models for in-house validations, the matrix mismatch error is also considered. The validation models presented here are of a general nature and can be used in principle for all analytical methods. Correct and appropriate statistical modelling is very important.

In-house validation and factorial effect analysis of a liquid chromatography-tandem mass spectrometry method for the determination of steroids in bovine muscle.

Anabolic steroids are banned from use in food-producing animals in the EU (Council Directive 96/22/EC). To control the zero-tolerance concept a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for the screening and confirmation of most of the relevant natural and synthetic estrogenic and androgenic steroids in bovine muscle was developed and validated. The method permits to confirm and quantify almost all steroids below 1 microgkg(-1). The validation was carried out according to Commission Decision 2002/657/EC, chapter 3.1.3 "alternative validation", by applying a matrix-comprehensive in-house validation concept. Decision limit CCalpha, detection capability CCbeta, recovery, repeatability, within-laboratory reproducibility and the uncertainty of measurement were calculated. Furthermore, a factorial effect analysis was carried out to identify factors that have a significant influence on the method. Factors considered to be relevant for the method in routine analysis (e.g., operator, storage duration of the extracts before measurement, different cartridge lots) were systematically varied on two levels. The factorial analysis showed that different cartridge lots and different storage durations of the extracts before measurement can exert a relevant influence on the method.

Development and in-house validation of an LC-MS/MS method for the determination of stilbenes and resorcylic acid lactones in bovine urine.

Stilbenes and zeranol are nonsteroidal estrogenic growth promoters which are banned in the European Union (EU) for use in food-producing animals by Council Directive 96/22/EC. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was developed for the screening and confirmation of stilbenes (diethylstilbestrol, dienestrol, hexestrol) and resorcylic acid lactones (zeranol and its metabolites taleranol and zearalanone as well as the mycotoxins alpha-zearalenol, beta-zearalenol and zearalenone) in bovine urine. The method permits the confirmation and quantification of stilbenes and resorcylic acid lactones at levels below 1 microg L(-1) and 1.5 microg L(-1), respectively. The validation was carried out according to Commission Decision 2002/657/EC, Chap. 3.1.3 "alternative validation" by a matrix-comprehensive in-house validation concept. Decision limit CCalpha, detection capability CCbeta, recovery, repeatability, within-laboratory reproducibility and the uncertainty of measurement were calculated. Furthermore, a factorial effect analysis was carried out to identify factors that have a significant influence on the method. Factors considered to be relevant for the method in routine analysis (e.g. operator, matrix condition, storage duration of the extracts before measurement, different cartridge lots, hydrolysis conditions) were systematically varied on two levels. The factorial analysis showed that different cartridge lots, storage durations and matrix conditions can exert a relevant influence on the method.

Matrix-comprehensive in-house validation and robustness check of a confirmatory method for the determination of four nitrofuran metabolites in poultry muscle and shrimp by LC-MS/MS.

An already well-described method for the determination of nitrofuran metabolites 3-amino-5-methyl-morpholino-2-oxazolidinone (AMOZ), 3-amino-2-oxazolidinone (AOZ), semicarbazide (SEM) and 1-aminohydantoin (AHD) was adapted to the needs of our laboratory and checked for its robustness regarding sample conditions and the processing step. Using the same data, the method was validated and the measurement uncertainty was estimated. All criteria and requirements of Commission Decision 2002/657/EC were fulfilled. The CC(alpha) determined lies between 0.1 and 0.7 microg/kg, the CC(beta) lies between 0.1 and 0.9 microg/kg, the measurement uncertainty was estimated as being between 7 and 17% taking into account matrix, time and sample preparation influences.