Académique Documents
Professionnel Documents
Culture Documents
ICS 67.050
This draft International Standard is a draft standard developed within the European Committee for
Standardization (CEN) and processed under the CEN-lead mode of collaboration as defined in the
Vienna Agreement. The document has been transmitted by CEN to ISO for circulation for ISO member
body voting in parallel with CEN enquiry. Comments received from ISO member bodies, including those
from non-CEN members, will be considered by the appropriate CEN technical body. Should this DIS be
accepted, a final draft, established on the basis of comments received, will be submitted to a parallel two-
month FDIS vote in ISO and formal vote in CEN.
In accordance with the provisions of Council Resolution 15/1993 this document is circulated in
the English language only.
To expedite distribution, this document is circulated as received from the committee secretariat.
ISO Central Secretariat work of editing and text composition will be undertaken at publication
stage.
Pour accélérer la distribution, le présent document est distribué tel qu'il est parvenu du
secrétariat du comité. Le travail de rédaction et de composition de texte sera effectué au
Secrétariat central de l'ISO au stade de publication.
THIS DOCUMENT IS A DRAFT CIRCULATED FOR COMMENT AND APPROVAL. IT IS THEREFORE SUBJECT TO CHANGE AND MAY NOT BE
REFERRED TO AS AN INTERNATIONAL STANDARD UNTIL PUBLISHED AS SUCH.
IN ADDITION TO THEIR EVALUATION AS BEING ACCEPTABLE FOR INDUSTRIAL, TECHNOLOGICAL, COMMERCIAL AND USER PURPOSES, DRAFT
INTERNATIONAL STANDARDS MAY ON OCCASION HAVE TO BE CONSIDERED IN THE LIGHT OF THEIR POTENTIAL TO BECOME STANDARDS TO
WHICH REFERENCE MAY BE MADE IN NATIONAL REGULATIONS.
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but shall
not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In the
unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
Copyright notice
This ISO document is a Draft International Standard and is copyright-protected by ISO. Except as permitted
under the applicable laws of the user's country, neither this ISO draft nor any extract from it may be
reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, photocopying,
recording or otherwise, without prior written permission being secured.
Requests for permission to reproduce should be addressed to either ISO at the address below or ISO's
member body in the country of the requester.
Contents Page
Foreword............................................................................................................................................................. 4
Introduction ........................................................................................................................................................ 5
1 Scope ..................................................................................................................................................... 5
2 Normative reference ............................................................................................................................. 5
3 Terms and definitions........................................................................................................................... 6
4 Principle ................................................................................................................................................. 8
5 Apparatus and equipment.................................................................................................................... 8
6 Sampling of non-packed food products............................................................................................. 8
7 Sampling of packed products – Statistical principles .................................................................... 11
8 Preparation of the analytical sample and the test portion ............................................................. 13
9 Packaging and labelling of laboratory samples .............................................................................. 13
10 Dispatch of laboratory samples ........................................................................................................ 14
11 Sampling report .................................................................................................................................. 14
Annex A (informative) Laboratory sample sizes of different cereals, oilseeds, and derived
products............................................................................................................................................... 16
Annex B (informative) Estimation of the number of particles in 100 mg test portions after
ultracentrifugal milling ....................................................................................................................... 17
Bibliography ..................................................................................................................................................... 18
3
prEN ISO 21568:2005 (E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO 21568 was prepared by Technical Committee ISO/TC 34, Agricultural food products, Subcommittee SC ,
and by Technical Committee CEN/TC 275, Food analysis - Horizontal methods in collaboration.
Recipients of this draft are invited to submit, with their comments, notification of any relevant patent rights of
which they are aware and to provide supporting documentation.
Since the first draft prEN ISO 21568 has been circulated for the parallel enquiry in 2003, new scientific
approaches to sampling have been elaborated. The results of a large study on distribution of GMOs in loose
material (bulk agricultural commodities) will be submitted for publication within the next months. Additionally, a
sampling protocol based on these findings was published as a Recommendation by the European
Commission in November 2004. In this study it was proved clearly, that different lots of comparable size
showed a large variety of GMO distribution which is of global relevance. This leads to the conclusion that a
protocol which is based on a distribution assumption can only yield erroneous results. Therefore, it was
necessary to take these new developments into account and to publish a second draft European International
Standard.
Based on the new scientific evidence the chapter 6 "Sampling of loose food products" of prEN ISO
21568:2003 was changed as follows:
A new statistical approach was introduced. While the old version suggested a sampling protocol which was
based on a distribution assumption and therefore was based mainly on the size of the laboratory sample to be
analysed, the new protocol is independent of the distribution of GMOs in the lot, and is therefore not affected
by stratification or clustering effects within the lot. In addition to that the new protocol allows an estimation of
the sampling error, which is of great benefit for enforcement authorities as well as for trading companies.
All other changes to the prEN ISO 21568:2003 were introduced following the comments received during the
first enquiry.
4
prEN ISO 21568:2005 (E)
Introduction
Correct sampling is an operation that requires the most careful attention. Emphasis should be laid on the
necessity of obtaining a representative sample of the goods under investigation. Accurate analytical work and
interpretation of results are wasted if the sample does not accurately represent the lot from which it is taken.
If ad-hoc sampling of food products is undertaken without applying a sampling strategy and without
considering the lot specific properties, the analytical result is only valid for the sample which has been
analysed. It is not possible to extend the result to the rest of the lot.
By applying sampling frameworks to assess the level of compliance of a given lot of products a certain
number of samples has to be taken, and the result of the analysis can be extended to the whole lot. The use
of sampling plans is the only effective way to make correct statements about the nature, in this case the GMO-
content, of the product tested.
The procedures given in this standard are recognised as good practice and it is strongly recommended that
they be followed whenever practicable. It is recognised that it is difficult to lay down fixed rules to be followed
in every case, and particular circumstances may render some modification of the method desirable.
This standard has been established for food products, but could also be applied to other products, e. g. feed
and environmental samples.
NOTE In certain areas there are widely recognised trade associations which specify rules for the sampling plans to
be used in contracts under their auspices. In no case will this standard override the rules laid down in such contracts.
1 Scope
This Standard gives guidance for setting up valid sampling strategies for food products that are to be analysed
for the presence of genetically modified organisms and derived products.
2 Normative reference
This draft standard incorporates, by dated or undated reference, provisions from other publications. The
normative references are cited at the appropriate places in the text, and the publications re-listed hereafter.
For dated references, subsequent amendments to or revisions of any of these publications apply to this draft
standard only when incorporated in it by amendment or revision. For undated references the latest edition of
the publication referred to applies.
ISO 6644, Flowing cereals and milled products – Automatic sampling by mechanical means
EN ISO 21572:2004, Foodstuffs - Methods of analysis for the detection of genetically modified organisms and
derived products - Protein based methods
ISO/FDIS 21569, Foodstuffs - Methods of analysis for the detection of genetically modified organisms and
derived products - Qualitative nucleic acid based methods
EN ISO 21571:2005, Foodstuffs - Methods of analysis for the detection of genetically modified organisms and
derived products - DNA extraction methods.
ISO/DIS 24276, Foodstuffs - Nucleic acid based methods of analysis for genetically modified organisms and
derived products - General guidelines and requirements.
5
prEN ISO 21568:2005 (E)
3.1
Consignment
A quantity of some commodity delivered at one time and covered by one set of documents. The consignment
may consist of one or more lots or part of lots.
[ISO 7002:1986]
3.2
Lot
stated portion of the consignment to be tested for presence of GMO.
3.3
Increment
a quantity of material taken at one time from a larger body of material.
NOTE Increments may be tested individually aiming at estimation of the variation of any characteristic throughout a
lot (or between lots).
[ISO 7002:1986]
3.4
Item
an actual or conventional object (a defined quantity) on which a set of observations may be made.
[ISO 7002:1986]
3.5
Sample
one or more items (or a portion of material) selected in some manner from a lot. It is intended to provide
information representative of the lot, and, possibly, to serve as a basis for decision on the lot.
3.6
File increment sample
an increment that is retained for a specific period of time for further analysis
3.7
Bulk sample
a composite of the increments taken from a lot.
3.8
Laboratory sample
a sample as prepared for sending to the laboratory and intended for inspection or testing
[ISO 7002:1986]
3.9
Test portion
a sample, as prepared for testing or analysis, the whole quantity being used for analysis or testing at one time
[ISO 3534-1]
3.10
Lot size
the number of items or quantity of material constituting the lot
[ISO 7002:1986]
6
prEN ISO 21568:2005 (E)
3.11
Sample size
the number of items or quantity of material constituting the sample
[ISO 7002:1986]
3.12
Sample division
the process of selecting one or more representative subsamples from a sample by such means as riffling or
mechanical dividing
3.13
Estimation error
in the estimation of a parameter, the estimation error is the difference between the calculated value of the
estimator and the true value of this parameter
3.14
Sampling error
part of the total estimation error due to one or several of the following parameters:
3.15
Sampling plan
the predetermined procedure for the selection, withdrawal and preparation of samples from a lot to yield the
required information so that a decision can be made regarding the acceptance of the lot.
[ISO 7002:1986]
3.16
Acceptance quality limit (AQL)
quality level that is the worst tolerable process average when a continuing series of lots is submitted for
acceptance sampling
NOTE This concept only applies when a sampling scheme with rules for switching and for discontinuation, such as in
ISO 2859 or ISO 3951, is used.
NOTE Although individual lots with quality as bad as the acceptance quality limit may be accepted with fairly high
probability, the designation of an acceptance quality limit does not suggest that this is a desirable quality level. Sampling
schemes found in International Standards such as this part of ISO 2859, with their rules for switching and for
discontinuation of sampling inspection, are designed to encourage suppliers to have process averages consistently better
than the AQL. Otherwise, there is a high risk that the inspection severity will be switched to tightened inspection under
which the criteria for lot acceptance become more demanding. Once on tightened inspection, unless action is taken to
improve the process, it is very likely that the rule requiring discontinuation of sampling inspection pending such
improvement will be invoked.
[ISO 2859-1]
3.17
Inspection level
the inspection level relates the sample size to the lot size and hence to the discrimination afforded between
"good" and "poor" quality.
NOTE A sampling scheme involves "switching" between normal, tightened and reduced inspection sampling plans,
(see e.g. Table 1 of ISO 2859-1:1989 and Table 1-A of ISO 3951:1989).
7
prEN ISO 21568:2005 (E)
3.18
Specification limit
a limit (value) set by a contract or legal requirements above which a decision shall be made, e.g. regulatory
compliance
3.19
Non conforming item or increment
a sample from which a test result above the specification limit has been obtained.
4 Principle
In this standard sampling is considered to consist of the following steps:
homogenisation and reduction of the particle size by appropriate means to form the analytical sample;
Samples shall be representative of the lots from which they are taken. Therefore, as the composition of a lot is
seldom uniform, a sufficient number of increments shall be taken and carefully mixed, thus giving a bulk
sample from which the laboratory sample is obtained by successive divisions or otherwise.
If it is necessary to determine the sampling error, file increment samples should be kept for further analysis.
All sampling operations shall be carried out over a sufficiently short period of time so as to avoid any alteration
in the composition of the samples.
Sampling shall be carried out in such a manner as to protect from adventitious contamination the samples, the
sampling instruments and the container in which the samples are placed. Special attention shall be paid to
avoid cross-contamination during the sampling procedure between different lots. Sampling apparatus shall be
cleaned carefully, e. g. by using DNA-destroying agents. Material adhering to the outside of the sampling
instrument shall be removed before the contents are discharged.
Lot distribution properties affect the efficiency of sampling strategies [1]. In case the variable of interest is
randomly distributed within the lot under investigation, the sampling error can be estimated according to the
binomial distribution [2]. However, in reality, food product lots may show non-random distributions [3], and lot
heterogeneity needs to be taken into account when defining sampling procedures statistically. This sampling
8
prEN ISO 21568:2005 (E)
protocol also provides a way to estimate the sampling error associated with the overall GMO content of a lot,
without imposing any distribution assumption. Indeed, estimates of sampling error as function of individual lot
properties, are a pre-requisite to achieve a given risk level under any possible lot properties scenario.
The Standard Deviation (SD) of the increment GMO content estimates may be interpreted as an indicator of
the lot heterogeneity and in highly heterogeneous lots an indication of the sampling error, generally the larger
the standard deviation, the larger the sampling error in highly heterogeneous lots. When an estimation of the
sampling error of the bulk sample is required, the users can execute the second step of the sampling protocol
to determine the sampling error estimate.
When the analytical result obtained from the bulk sample is close to the specification limit and variability of the
measurement is required, analysis of the individual file increment samples can be used to determine the error
estimate related to this result.
NOTE: As an example for a specification limit = 3% and the acceptable range of the estimate is +/- 50%, it is
recommended to proceed with the analysis of individual file increment samples if the estimate (x) falls within
the range 1,5% < x < 4,5%.
6.2 Procedure
6.2.1 General
This protocol is based on a two step procedure. The first step is designed to provide an estimate of the GMO
content of the lot based on a bulk sample. The second step is designed to provide an estimate of the
associated sampling error, if required.
Increments should be sampled following the technical principles given in ISO 6644:2002 as described in 6.3
and ISO 13690:1999 as described in 6.4.
The number of sampling points, where the increments for creating the bulk sample and the file increment
samples are taken is defined according to the lot size. In case of lots from 50 to 500 tonnes, the size of the
bulk sample should be 0.01% of the total lot size. In case of lots smaller than 50 tonnes, the size of the bulk
sample should be 5kg. In case of lots larger than 500 tonnes, the size of the bulk sample should be 50 kg.
(see Table 1).
At each sampling interval (systematic sampling) or sampling point (static sampling) an increment of 1kg
should be collected and split into two portions of 0.5 kg: one to be used as an increment for the production of
the bulk sample, the other to be stored as a file increment sample.
Note according to ISO 6644 the maximum bulk sample size is 100 kg and the increment size 1 kg, with a maximum
lot size of 500 t. As a two step procedure is recommended to estimate the sampling error the increment of 1 kg is splitted
in two leading to an increment size of 0,5 kg. The resulting bulk sample of 50 kg equals 0,01% of the maximum lot size as
defined in ISO 6644. In order to maintain the sampling frequency at the same level, the bulk sample should be 0,01%
independent of the lot size. For lots <50 t and lots >500 t the bulk sample size is fixed due to statistical and economical
reasons, respectively.
9
prEN ISO 21568:2005 (E)
The bulk sample shall be formed by combining the increments and mixing them thoroughly.
The file increment samples are sealed and stored for further analysis as described in 6.1, if necessary.
Large sampling units e. g. potatoes, squash, coconuts, sugar beets, papaya, etc. should be sampled
according to 7.
A laboratory sample is prepared from the bulk sample by coning and quartering or by using sample dividers as
described in ISO 13690:1999. The size of the laboratory sample should be selected as specified in Annex A.
The analyses shall be carried out according to ISO/DIS 21571:2002, ISO/DIS 21569:2002,
ISO/DIS 21570:2003 and/or ISO 21572:2004.
If the analytical result of the laboratory sample obtained from the bulk sample is close to the specification limit
an estimation of the sampling error is recommended. This should be carried out as follows:
20 file increment samples should be randomly selected from those collected and analysed according to
ISO/DIS 21571:2002, ISO/DIS 21569:2002, ISO/DIS 21570:2003 and/or ISO 21572:2004. If there are fewer or
equal to 20 file increment samples, as in case of small lots, all samples should be analysed.
Analytical results from these 20 file increment samples are used to estimate the content of GMO derived
material of the lot and its associated sampling error expressed as standard deviation (SD). If this error is
acceptable with respect to e.g. a contract between buyer and seller or legal requirements, no additional
analysis of the remaining file increment samples is necessary. If instead the level of associated error is not
acceptable, further analysis of additional file increment samples might be necessary. The analytical process
stops when the level of error associated to the GMO content estimate reaches a level acceptable according to
e.g. a contract between buyer and seller or legal requirements, or all the collected file increment samples have
been analysed.
The sampling error (estimator of the standard deviation) can be calculated using equation (1):
− 2
æ ö
i=n ç xi − x ÷
s= å è ø (1)
i =1 n − 1
where
n is sample size 20
− i =n
xi
with x = ån
i =1
Whenever possible sampling should be performed on moving goods, e. g. during loading or unloading.
Systematic sampling should be used to collect increment samples from the lot, and the sampling period
should be defined according to ISO 6644.
Mechanical sampling equipment shall be capable of taking increments from as much of the cross-section as
possible of the flowing bulk material. It should be adjustable such that the size of increments and the sampling
10
prEN ISO 21568:2005 (E)
frequency can be varied over a desirable wide range. The sampling equipment should be accessible to
facilitate inspection and cleaning. Examples of automatic sampling devices are given in ISO 6644.
The increments shall be taken from as much of the cross section of the flowing material as possible, in such a
way that every part of the lot has an equal opportunity of entering the inlet of the sampling device.
Each wagon, lorry, barge or ship should be sampled according to the principles described in ISO13690:1999.
Increments shall be taken throughout the whole depth of the lot. A suitable instrument shall be used to
achieve this requirement. The number of increments to be taken is described in 6.2.
The lot should be sampled throughout its whole depth and using a grid system, like that used for wagons,
barges or ships (see ISO 13690:1999). A suitable instrument shall be used to achieve this requirement. The
number of increments to be taken is described in 6.2.
If the depth of the lot does not permit the use of this method, sampling should be carried out from moving
goods according to 6.3.
The lot from which the samples are taken has to be defined, e.g. a given batch or all packets from one day of
production. An acceptable quality level (AQL) has to be defined for the acceptance of a lot. As an example, in
table 2 sample sizes matching different AQLs are given according to the principles described in ISO 2859-
1:1999, where more details are provided.
The AQL gives the number (or percentage) of non conforming units in a lot above which this lot will be
rejected. The criteria for non conformity should be laid down e. g. in the contract between the buyer and the
seller. Non conformity in this context can be a qualitative positive test result or failure to match a certain
specification limit, i. e. the content of GMO derived material above this limit.
11
prEN ISO 21568:2005 (E)
Table 2 — Sample size for sampling packed products and maximum number of non-conforming units
at four examples of AQLs using the standard sampling approach according to ISO 2859-1
The units (bags, packets) shall be tested individually, since the packet is the level at which compliance with
the specification limit is tested.
Based on experience or on information regarding the nature of the lot under investigation ISO 2859-1allows a
change of the standard procedure to a reduced or tightened sampling.
Normal inspection is designed to protect the producer against having a high proportion of lots rejected when
the quality of the product is better than the AQL. However, if two out of any five (or fewer) successive lots are
not accepted, then tightened inspection must be introduced. On the other hand, if production quality is
consistently better than the AQL, sampling costs may be reduced (at the discretion of the responsible
authority) by the introduction of reduced-inspection sampling plans. For details see ISO 2859-1.
12
prEN ISO 21568:2005 (E)
Apparatus and equipment used to prepare the analytical samples shall ensure that the analytical sample is
homogenous. In the case of material composed of discrete entities, care should be taken to ensure that the
particle size is reduced to the appropriate size (for examples see Annex B).
Special requirements regarding the grinding and homogenisation step are described in ISO/DIS 21571:2002.
After grinding and/or homogenisation of the sample the test portion (at least 200 mg, see ISO/DIS
21571:2002) is gained by dividing the analytical sample.
9.1 General
The laboratory samples shall be packed in containers suitable for the purpose, according to the tests to be
undertaken.
9.2 Information
Subject to the requirements of the contractor, the following information should be recorded as applicable:
point of departure;
destination;
quantity of consignment;
type of goods;
name of seller;
name of buyer;
13
prEN ISO 21568:2005 (E)
date of sampling;
Samples should be dispatched and stored under appropriate conditions so that the sample integrity is
maintained.
11 Sampling report
The sampling report should include the following as applicable:
name and signature of the owner of the sample or his legal representative
producer
lot number
14
prEN ISO 21568:2005 (E)
sample destination, e.g. name and address to which samples are sent
additional remarks, e. g:
identification of the transport vehicle (licence plate number, name of the ship, etc.)
15
prEN ISO 21568:2005 (E)
Annex A
(informative)
Recommended sizes of laboratory samples of different cereals and oilseeds are given in table A.1. The
calculation of the sample size is based on the grain/seed weight and a grain/seed number of 10 000 per
sample according to [4].
16
prEN ISO 21568:2005 (E)
Annex B
(informative)
Factors to be considered: The heating associated to the milling process increases as function of the desired
particle size reduction. This can result in the degradation of the analyte (protein or DNA) and/or in a pasty
sample (i.e. with soybeans). For practical reasons mesh sizes between 0,75 mm (maize kernels) and 0,5 mm
(soybeans) are suitable; if smaller particle sizes of maize kernels have to be obtained, a two step milling has
to be performed as indicated above.
1) This information is given for the convenience of users of this standard and does not constitute any endorsement by
CEN. Equivalent products may be used if they can be shown to lead to the same results.
17
prEN ISO 21568:2005 (E)
Bibliography
ISO 7002:1986 Agricultural food products – Layout for a standard method of sampling from a lot
ISO 2859-1:1999 Sampling procedures for inspection by attributes – Part 1: Sampling schemes indexed by
acceptance quality limit (AQL) for lot-by-lot inspection.
ISO 3534-1:1993 Statistics - Vocabulary and symbols - Part 1: Probability and general statistical terms
[1] Paoletti, C., Donatelli, M., Kay, S., and van den Eede, G.: Simulating kernel lot sampling: the effect of
heterogeneity on the detection of GMO contaminations, 2003, Seed Sci. Technol. 31, 629-638
[2] Codex Alimentarius Guidelines on sampling ALINORM 04/27/23 approved at the Codex Commission in
Geneva, 28 June –3 July 2004
[3] Lischer, P.: Sampling procedures to determine the proportion of genetically modified organisms in raw
materials. Part I: correct sampling, good sampling practice. Mitt. Lebensm. Hyg. 92: 290-304.
[4] Hübner, P.; Waiblinger, H.-U.; Pietsch, K. and Brodmann, P.: Validation of PCR methods for the
quantification of genetically modified plants in food, 2001,J. AOAC Int. 84: 1855-1864.
18