GMO Testing: A Procurement and Compliance Buying Guide for Food-Safety Labs

GMO testing (genetically modified organism testing) is the laboratory analysis used to detect and, where required, quantify genetically engineered DNA or proteins in food, feed, seed, and raw agricultural commodities. For a food-safety lab, ingredient supplier, or contract manufacturer, the decision is rarely “should we test” — it’s which method, which accredited lab or kit supplier, and against which regulatory or voluntary threshold. This guide is written for the person making that procurement and compliance decision, not for a general audience asking whether GMOs are safe.

What GMO testing actually measures

Nearly all GMO testing in commercial and regulatory use relies on detecting either the introduced DNA sequence (genomic methods) or the protein it expresses (protein-based methods). The two are not interchangeable, and choosing the wrong one for your matrix or your compliance target is the single most common procurement mistake.

• PCR (polymerase chain reaction), qualitative — detects the presence or absence of a specific genetic element (commonly a screening target like the CaMV 35S promoter or NOS terminator, or an event-specific sequence). Answers “is it there,” not “how much.”
• PCR (real-time/quantitative, qPCR) — required when a threshold-based labeling regime applies, because it reports GMO content as a percentage of the target species’ DNA, not just detect/non-detect. This is the method regulatory and Non-GMO Project Verified testing programs actually rely on for threshold compliance.
• ELISA (enzyme-linked immunosorbent assay) — detects the expressed protein rather than DNA. Faster and cheaper than PCR, but proteins degrade with processing (heat, fermentation, refining), so ELISA is unreliable on highly processed ingredients (oils, syrups, lecithin) where DNA-based methods remain necessary.
• Lateral-flow strip tests — a rapid, field-deployable ELISA variant, typically qualitative, used for incoming-goods screening at a receiving dock or grain elevator before a sample is sent for lab confirmation. Not a substitute for accredited quantitative testing on a certificate of analysis.

Which method a supplier or lab uses should be driven by your matrix (raw grain vs. a multi-ingredient processed food) and by what you actually need to prove — detect/non-detect screening, or a defensible quantitative percentage against a specific legal or voluntary threshold.

The regulatory and voluntary thresholds you are actually testing against

“GMO testing” is not a single pass/fail standard — the threshold that matters depends entirely on which market and which claim you’re supporting. Build your test program around the specific requirement, not a generic assumption.

• EU labeling regime — Regulation (EC) No 1829/2003 and 1830/2003 set a 0.9% adventitious-presence threshold per ingredient, above which GMO labeling is mandatory for approved GM material; unapproved GM events are subject to a zero/near-zero tolerance approach. Compliance testing in the EU is quantitative PCR against certified reference materials.
• US National Bioengineered Food Disclosure Standard (NBFDS) — administered by USDA’s Agricultural Marketing Service, this sets a 5% inadvertent-or-technically-unavoidable presence threshold per ingredient before disclosure is required, and covers “bioengineered food” specifically as defined in the rule (not every GM-adjacent ingredient is in scope — refined, highly processed ingredients where modified genetic material is no longer detectable are generally excluded).
• Non-GMO Project Verified — a voluntary, private certification (not a government standard) with its own published Product Verification Program Standard, using a 0.9% action threshold per ingredient and requiring testing at ISO 17025-accredited labs using validated quantitative PCR methods for high-risk inputs.
• Codex Alimentarius — the joint FAO/WHO body publishes guidance (CAC/GL 74-2010) for the labeling of foods derived from modern biotechnology that many national frameworks reference or adapt, rather than a single binding global threshold.

A procurement decision should start by naming the specific standard you’re testing against, because it determines the required method sensitivity, the reference materials the lab must use, and how “compliant” gets defined on the certificate of analysis you receive.

Accreditation and method standards to require from a lab or kit supplier

Before evaluating price or turnaround, confirm the lab or kit is actually qualified to produce a defensible result. The baseline documentation to request:

• ISO/IEC 17025 accreditation — the general competence standard for testing and calibration laboratories. This is the credential that Non-GMO Project Verified and most retailer/brand-owner supply-chain programs require by name; a lab without it should not be your source of record for a compliance claim, even if its results look correct.
• ISO 21569 / ISO 21570 / ISO 24276 — the ISO series specifically covering GMO detection and quantification in foods: qualitative PCR-based methods (21569), quantitative PCR-based methods (21570), and general requirements and definitions for the whole detection/quantification/analysis process (24276). Ask which of these your lab’s method is validated against, not just “we use PCR.”
• Certified reference materials (CRMs) — quantitative results are only meaningful if calibrated against a certified reference material for the specific GM event being measured (e.g., materials distributed by the European Reference Laboratory for GM Food and Feed or an equivalent certified source). A supplier who can’t name their reference material source is not producing a legally defensible quantitative result.
• Method validation and limit of detection/quantification (LOD/LOQ) — request the published LOD/LOQ for the specific event or screening target, not a generic sensitivity claim. This determines whether the method can actually resolve a 0.9% or 5% threshold with confidence.

In-house testing vs. contract lab: how to decide

Rapid lateral-flow or ELISA screening at receiving is commonly brought in-house because it’s cheap, fast, and doesn’t require a legally defensible quantitative result — it’s a gate to catch obvious problems before a shipment is accepted, not the record you’d rely on for a labeling claim. Confirmatory quantitative PCR testing, especially anything supporting a Non-GMO Project Verified claim, an EU export declaration, or an NBFDS disclosure decision, is almost always better placed with an ISO 17025-accredited contract lab, for three practical reasons:

• The instrumentation (real-time PCR thermocyclers, extraction platforms) and the ongoing proficiency-testing/method-validation burden needed to keep accreditation current is a real, recurring cost most food businesses can’t justify for occasional testing volume.
• A third-party accredited certificate of analysis carries evidentiary weight with auditors, certification bodies, and customers that an internal result does not, regardless of how carefully the internal test was run.
• Event-specific and multiplex panels (testing for many known GM events simultaneously) require reference materials and validated protocols that are impractical to maintain in-house at low volume.

High-volume commodity buyers (grain elevators, oilseed crushers) are the more common exception — sustained testing volume can justify in-house quantitative capability, but it still typically runs alongside periodic accredited third-party verification for compliance and customer-facing certificates.

A procurement checklist for evaluating a GMO testing vendor or kit

1. Does the lab hold current ISO/IEC 17025 accreditation, and is GMO/PCR testing specifically within their accredited scope (accreditation is scope-specific, not blanket)?
2. Which ISO 21569/21570/24276 methods, or equivalent validated in-house methods, does the lab use for your target event(s) and matrix?
3. What is the published LOD/LOQ for the relevant target, and does it comfortably resolve the threshold you’re testing against (0.9% EU/Non-GMO Project, 5% NBFDS)?
4. What certified reference materials does the lab calibrate against, and can they document traceability?
5. Can the lab test for the specific GM events relevant to your supply chain (screening targets alone are not event-specific and can produce false positives from approved, non-target GM material)?
6. What is turnaround time, and does it fit your receiving/release schedule — rapid lateral-flow screening for receiving decisions, longer-turnaround quantitative PCR for compliance documentation?
7. What does the certificate of analysis actually state — a bare detect/non-detect, or a quantitative percentage with stated uncertainty against a named reference material?
8. Does the vendor supply chain-of-custody documentation adequate for your certification body or customer’s traceability requirements?

Some ingredient and commodity supply chains route GMO testing procurement through a broader lab-supply or distribution relationship rather than contracting a testing lab directly — for example, LAC Health is one example of a distributor operating in this space; the accreditation and method-validation questions above apply the same way regardless of whether you’re evaluating a testing lab directly or a supplier who arranges testing as part of a broader procurement relationship.

Cost and turnaround: what drives the range

Pricing for GMO testing varies widely by method, target specificity, and matrix, and CASRAI does not publish or endorse specific vendor pricing. As a procurement matter, the variables that actually move cost and turnaround are: qualitative screening vs. quantitative event-specific testing (quantitative costs more and takes longer), the number of distinct GM events tested for (multiplex panels cost more per sample than a single screening target but less than running each event separately), matrix complexity (a raw commodity is cheaper to test than a multi-ingredient processed food where DNA must first be recovered from a complex background), and whether results need to be issued on an accredited certificate of analysis versus an informal report. Build your RFP around these variables explicitly rather than comparing a flat per-sample price across vendors, since a low quote on an unspecified method is not comparable to an accredited, event-specific quantitative result.

Sampling and chain of custody

A GMO test result is only as reliable as the sample it was run on. Because GM material is often unevenly distributed within a bulk commodity lot, sampling protocol — not just the analytical method — is a major source of result variability. Reputable testing programs specify a sampling plan (sample size, number of increments, homogenization procedure) before analysis, and Codex Alimentarius and ISO 24276 both address sampling representativeness as part of the overall testing process, not as an afterthought. When evaluating a vendor, ask what sampling protocol they require or recommend for your commodity and lot size, and confirm chain-of-custody documentation exists from the point of sample collection through to the reported result — this is what makes a result defensible if it’s later challenged by a customer, auditor, or certification body.

Frequently asked questions

Is GMO testing legally required in the United States?

Disclosure is required under the National Bioengineered Food Disclosure Standard for covered bioengineered foods above the 5% inadvertent-presence threshold, administered by USDA’s Agricultural Marketing Service. Whether a specific product and ingredient combination is in scope depends on the rule’s list of bioengineered foods and its exemptions for highly refined ingredients where modified genetic material is no longer detectable — this is a product-specific determination, not a blanket testing mandate.

What’s the difference between screening and event-specific GMO testing?

Screening methods detect common genetic elements (such as the CaMV 35S promoter) present in many different GM constructs, giving a fast presence/absence signal but not identifying which GM event is present. Event-specific methods target the unique DNA sequence created at the insertion site of a particular approved GM event, giving a precise, legally defensible identification and, combined with qPCR, a quantitative result for that specific event.

Can ELISA test kits be used for regulatory compliance testing?

ELISA is commonly used for rapid receiving-dock screening but is generally not sufficient on its own for quantitative regulatory or certification compliance, both because it detects protein (which degrades with processing) rather than DNA, and because most compliance thresholds (EU 0.9%, Non-GMO Project 0.9%) are defined and validated against quantitative PCR methods.

Do organic-certified products still need GMO testing?

Organic certification standards generally prohibit the intentional use of GMOs, but certifying bodies vary in whether and how often they require analytical GMO testing versus relying on input documentation and audit trails; buyers seeking both organic certification and a Non-GMO Project Verified claim should confirm testing requirements with each certification body separately, since the two programs are administered independently and are not identical in their testing requirements.

How often should a supply chain re-test for GMO content?

This depends on the certification program and your own risk assessment rather than a single fixed rule — Non-GMO Project Verified, for example, publishes its own testing frequency requirements by risk category of ingredient in its Product Verification Program Standard. High-risk crops (soy, corn, canola, cotton, sugar beet, alfalfa) generally warrant more frequent verification than low-risk ingredients.