A Recipe for Disaster?

It was apparently the philosopher Lucretius – who lived more than 2,100 years ago – who said, “What is food to one, is to others bitter poison.” Certainly, it would be a significant challenge to establish if Lucretius had a sound understanding of allergens back then, but his quote is as true today as it was millennia ago. In fact, the current big issue with food allergies is not so much with labeled ingredients, but those ingredients that enter into the food inadvertently or through fraudulent activity.

Food manufacturers are typically responsible for a range of products with many different recipes, which means that allergens can inadvertently enter products that, according to their ingredients list, do not contain allergenic components. After all, cleaning of product lines is not always 100 percent efficient. You may have seen the label: “produced in a facility that processes X, Y and Z.” In other cases, a more expensive ingredient might be replaced – without an eye for quality, safety, or correct labeling – by a cheaper one. The Food Allergen Resource and Research Programme (FARRP) shared one incident where pine nuts in a pesto were replaced by peanuts – it’s a recipe for disaster.

Clearly, when final products are not labeled as containing allergens, they may consequently cause serious harm. The only way to detect (and potentially quantify) the presence of undeclared allergens is by using analytical tools, which can be grouped into three main areas:

1. DNA-based methods; for example, polymerase chain reaction (PCR)
2. Immunological methods; for example, enzyme-linked immuno sorbent assay (ELISA) and lateral flow devices (LFD)
3. Mass spectrometric assays.

Although all of these techniques tend to work well with unprocessed or lightly processed materials, such as flour, it is more difficult to obtain accurate results with many processed materials. Why? Because the allergens either degrade or react with the matrix and cannot be recovered by extraction. Unfortunately, these allergens may still trigger reactions.

For many processed products, significantly different results can be obtained even within a group of methods. In PCR, target sequences may be single or multi copy, and the polymerase may be more or less efficient. In ELISA, antibodies – and subsequent specificity – can differ, as can extraction methods. Efficient extraction is essential when it comes to processed foods potentially containing egg and/or milk., but all too often assays fail to detect even high quantities of allergens. It is here that mass spectrometric techniques appear to have a significant advantage. Because they work with small, digested peptide fragments, allergens can be detected even in highly processed samples.

For analytical scientists using any of the methods, it is important to look very closely at the validation data – that is to say, what has been validated and how. Is it highly or lightly processed food? Were materials spiked or are the samples incurred? This information allows the analyst to make the best possible choice from the set of available methods; some methods only work with (or have been validated for) raw materials, while others may not be applicable because there is little or no target analyte (for example, aiming to detect chicken DNA in egg white).

Kit and equipment manufacturers often proudly bamboozle us with a myriad of data that demonstrate the supposed superiority of their assay. Unfortunately, in some cases, quantity of data is mistaken for quality. Any data provided should be objectively scrutinized, taking known issues of any of the applicable methods into account.

By now, there are already several European technical specifications (TSs) available for food allergen detection, some of which are based on proprietary assays. While TSs are useful in the absence of any other standardized method, it is much more desirable to develop method performance criteria (MPCs). MPCs are typically derived from a validation study of one or two methods in the particular field (for example, allergen analysis), and include parameters like sensitivity, selectivity, and robustness. However, once the MPCs have been deducted from the validation study, they are no longer tied to those methods, becoming standalone criteria. This has the significant advantage that any new method that meets those MPCs can be used. Such an approach drives the development of new, potentially better methods without the need to use older (potentially outdated) standards or technical specifications.

In fact, the approach has already been in place for some contaminant and residue methods, where European regulations set MPCs (European Commission Decision 2002/657/EC).

In the field of allergen analysis, there have been very encouraging new method developments in the last few years, especially in mass spectrometry, which raises my hopes for a robust and accurate methodology in the not too distant future. This will benefit food manufacturers and, ultimately, the allergic consumer.

Current existing methods are certainly not a recipe for disaster, but it is important for analysts and food manufacturers to understand the limitations of existing methodologies. Both parties should take suitable measures, using appropriate information on test certificates an appropriate product labeling, to prevent adverse health effects caused by inadvertent consumption.