A few days ago, Jon Entine posted a story on Slate.com reporting that Caitlin Shetterly’s feature in Elle magazine on genetically engineered (GE) corn (AKA genetically modified, or GM, corn) “just doesn’t withstand the critical scrutiny of science.”

But Entine’s post itself fails to withstand critical scrutiny. He botched his story about GE foods and human allergens on several fronts.

First, any examination of a possible relationship between GE crops and human allergens should include a discussion of StarLink™ corn and Entine’s piece doesn’t.

StarLink was a GE corn that was not approved for human consumption and yet ended up in taco shells and other corn products on U.S. grocery shelves which in turn led to a massive recall of corn products, investigations into reports of humans with allergic reactions, and federal monitoring of U.S. corn crops for levels of the offending protein over nearly eight years.

StarLink was genetically engineered to produce an insecticidal protein from a strain of Bacillus thuringiensis (Bt) bacteria. Other varieties of “Bt corn,” which save farmers from spraying certain insecticides because these varieties make their own, had been commercialized before StarLink and plenty of additional Bt corn varieties have been commercialized since. In fact, the vast majority of corn grown in the U.S. today is Bt corn, including fresh corn available in supermarkets. (And FYI, in almost all cases, the Bt insecticide is produced in every cell in every kernel of every ear of Bt corn.)

Before they can be commercialized in the U.S., Bt corn varieties (or any other GE crops that make their own pesticides) must go through an approval process with the U.S. Environmental Protection Agency (EPA), a process that includes testing each Bt protein for whether it might cause human allergies. Prior to StarLink corn, the Bt protein in every Bt corn variety passed the various tests the EPA ran for allergenicity with flying colors; the Bt protein in StarLink failed them all.

Behaving like a human allergen in the tests didn’t necessarily mean that the StarLink Bt protein actually was a human allergen. Still, the EPA–wisely in my opinion–decided to be cautious and did not allow StarLink to be commercialized for human food.  The agency went on to make what I consider a less wise decision, on the other hand, when it approved StarLink for animal consumption. One reason that decision was unwise is that corn is wind-pollinated. The subsequent contamination of corn meant for human consumption was essentially inevitable.

In the aftermath of StarLink corn contaminating human foods, experts working with the U.S. regulatory agencies concluded that the Bt protein in that GE corn variety had “a medium probability of being a potential allergen” and that the level estimated to be in human food presented a “low probability to sensitize some individuals” to the protein (and thereby possibly cause them to become allergic to it). The U.S. regulatory agencies were concerned enough about the allergenicity of StarLink corn that levels of the StarLink Bt protein in corn ear-marked for human food were monitored until about five years ago. The EPA’s final white paper on exposure to and any potential risk from the StarLink Bt protein was filed in March of 2008.

As I hope I’ve made obvious by now, any scientific scrutiny of the topic of GE corn and human allergies should include a telling of the StarLink corn story. (And it isn’t difficult to find; StarLink corn came up in my first 25 google hits for “genetically modified corn allergen.”) StarLink corn is at least a reminder that the results of genetically engineering a plant can be unexpected and undesired and it therefore remains prudent to use this powerful tool of agricultural biotechnology cautiously, especially as it is now being used to produce GE products that are more varied and complex than Bt corn. At most?…StarLink corn could actually represent an example of how a unique human allergen might have been created through the process of genetic engineering.

Entine’s piece also is inaccurate in its description of what is (or is not) required by FDA of producers of new GE foods, fails to mention that genetic engineering is also, in fact, a mutagenic process, and completely ignores the issue of labeling GE foods, without which tracking down any potential allergens or toxins associated with them would be much more difficult. (To read previous posts I’ve written on these topics, please follow the links I’ve added to the previous sentence.) In short, Entine’s post certainly didn’t “set the record straight” (a request he made of Elle) and wasn’t anything like the “even-handed analysis of the costs and benefits associated with adopting” genetic engineering that he mentioned Amy Harmon’s New York Times article on the plight of the orange industry in Florida had been praised for.

Too bad. What the debate over GE foods could really use is both more even-handedness and accurate record-setting.