By David Stukus, MD
Electric. That was the mood when Dr. Gideon Lack presented the LEAP findings to the audience at the American Academy of Allergy, Asthma, and Immunology. It was an honor to be there, and some day I may be telling allergists who haven’t even been born that I was there, like a classic 7th game of a World Series or the recent Super Bowl. By now, almost everyone concerned with the peanut allergy epidemic knows: Dr. Lack and his colleagues George du Toit et al have completed a study showing that they may have found a way to wind down the epidemic a bit. The study has seized the imaginations of health care reporters all over the world. But before we declare victory, let’s take stock of what we know and what we don’t.
The numbers tell a familiar story: Food allergy in children has doubled in the past 10 years, and peanut allergy has nearly quadrupled since 1997. For the roughly 2% of children in the United States currently living with peanut allergy, managing their allergies demands significant lifestyle modification and constant preparation for accidents. Needless to say, quality of life often suffers for parents and children.
Most recent research has focused on treatment, but these results have not been definitive or quite as promising as we had hoped. Causation is another matter. The big picture is too big—the way we eat, the way we drive, the way we grow our food, and the way we treat illness. But what if we can prevent peanut allergies from occurring in the first place? What if we can do something if not for the current generation of food allergic kids, then at least for the next generation? That’s exactly what the researchers involved in the Learning About Peanut Allergy (LEAP) study sought to find out.
As the dimensions of the food allergy epidemic really started to take off, the conventional wisdom for how to prevent these allergies has shifted repeatedly. In 2000, the American Academy of Pediatrics (AAP) recommended that parents refrain from feeding peanuts to infants at high risk for atopic disease until 3 years of age. However, peanut allergy continued to rise, which sent many physicians to reconsider. Careful review of the literature revealed very little, if any, evidence to support the guidelines. In 2008, the AAP published new guidelines in 2008, which in medical terms is overnight, retracting their previous stance, and declaring there was insufficient evidence to avoid peanut and other foods until a certain age, confusing pediatricians and parents alike.
In the midst of these changes, George du Toit, lead author of the LEAP study, identified a dramatic difference in peanut allergy between Israeli children and a similar population of kids in London. What was the difference? Israeli children almost universally consumed a snack called Bamba, prepared with peanuts, during infancy. The British researchers shifted their strategy from avoidance to early introduction. This was all speculation until the LEAP study results were released on February 23, 2015.
Before I dissect the LEAP study, there are a few extremely important things to understand about these findings:
- This was not a study looking at treatment of children with peanut allergy. It does not offer insight into a cure and is not applicable to anyone who has had prior allergic reactions to peanut.
- This should NOT be done at home! Every single child enrolled in this study underwent skin prick testing AND physician supervised oral challenge to peanut. Don’t experiment on your infant!
- The children enrolled in this study were all less than 12 months old, living in the United Kingdom, predominantly white, had severe eczema and/or egg allergy. This must be considered before extrapolating results to other populations, i.e. older children or African American infants in the U.S. with multiple food allergies.
- This study looked at very specific infants with severe eczema and/or egg allergy. There are other risk factors to consider (namely presence of other food allergies such as milk, wheat, soy and/or history of wheezing) that may have variable effects on this protocol.
- The cut off point chosen as upper limit for inclusion by skin prick test size was a 4 mm wheal. There are many infants deemed at risk but will have larger skin test findings. The results of this study cannot be extrapolated to this population.
Now for the study findings. In my opinion, this is one of the best-designed research studies I have read, both in regards to study design, but also by incorporating solid and proper outcome measures. 640 infants aged 4-11 months old (median age 7.8 months) were divided into two groups: those with negative skin prick tests to peanut and those with mildly positive skin tests to peanut, as defined by a wheal size of 1-4 mm. Ten percent of infants were excluded from the study due to wheal size > 4 mm.
These two cohorts were then randomized to either completely avoid peanut or eat Bamba or peanut butter (the equivalent of 8 peanuts per serving) three times a week until 5 years of age. Every infant randomized to eat peanut then underwent an oral challenge before starting the study to ensure they would not have an allergic reaction. One child with negative skin prick and 6 (13%) children with mildly positive skin prick tests had reactions during challenge and were reassigned to avoidance categories. This is exactly why consultation with a board-certified allergist is paramount before trying this at home – children at risk will need to have skin test and/or oral challenge to ensure safety before incorporating peanut into their diet.
The primary outcome measure was proportion of children who reacted to an oral challenge (gold standard to diagnose food allergy) to peanut at the age of 5. Secondary outcomes included immune markers. Ninety-eight percent of participants remained in the study until completion, which is astounding for a study of this magnitude. There was no significant difference in the rate of serious reactions between the two groups, and only one needed epinephrine. Most reactions were mild to moderate.
Major take home points: An 86% relative risk reduction in peanut allergy between infants with negative skin prick test who regularly consumed peanut compared with those who avoided peanut. Furthermore, there was a 70% reduction in peanut allergy in infants with mildly positive peanut skin prick test (sensitized) who consumed peanut compared with those who avoided.
By the age of five, among the 530 infants who initially had negative skin test results, 13.7% of those who had strictly avoided peanuts developed a peanut allergy, compared to only 1.9% of the children who had been eating peanuts.
Among the 98 infants who had mildly positive skin results, 35.3% of the avoidance group went on to develop a peanut allergy, but only 10.6% of the babies eating peanuts. Impressive, but not 100% effective.
Monitoring adherence to the protocol is crucial for any research conducted largely away from the laboratory. The researchers not only used food diaries, which can be fudged, but also objective measurement of levels of peanut protein in the beds of children in each group. They found significantly more peanut dust from those who were eating peanuts, and estimated 92% adherence for all participants. Furthermore, immune markers were followed over time for all groups. A significant increase in baseline peanut wheal size on skin testing was observed only in the peanut avoidance group. Serum IgE levels increased in both avoidance and consumption groups over time, but there were fewer participants in the consumption group with very high IgE levels at 12, 30 and 60 months. Lastly, levels of peanut-specific IgG and IgG4, the immunoglobulins associated with tolerance, were higher in the consumption group than in the avoidance group. This mirrors the effect seen with standard allergy shots.
So, what does this change? Most importantly, the study demonstrates for the first time that both primary (no sensitization) and secondary (sensitized, not allergic) prevention can occur with early introduction of peanut. There clearly exists a small window of opportunity in which frequent consumption can promote tolerance. It remains to be seen whether this is due to desensitization, i.e. oral immunotherapy, or true prevention, but the authors are already investigating this through their next-phase “LEAP-On” study. For LEAP-On, the participants from LEAP who ate peanut and were not allergic will now stop eating peanut completely and undergo challenge 12 months later. We all can’t wait for those findings!
What does this exciting news mean for the rest of us? I don’t advocate for pediatricians or families to try this without having important information such as skin prick size and oral challenge results, which only allergists can provide. As for working allergists, especially pediatric allergists, they ought to be on the drawing board right now thinking about how to incorporate this new data into their practices. Many families will be discouraged by these thoughts, citing poor access to pediatric allergists. This puts additional responsibility on allergists to find ways of collaborating more closely with pediatricians to enforce higher standards of treatment even as they expand their outreach. Every single study that has enrolled children for food allergy desensitization, treatment, and now prevention has used very specific entry criteria. This method, however low-tech, should be just as conscientious. Even in the research setting, participants still experience allergic reactions. Keep in mind, this study limited participants from those without high risk for development of allergic disease, which we’ve been trying to communicate since 2008. Whenever an experimental therapy is adopted there will always be pressure to step over the line. This study may be called LEAP, but it’s still only one step.
David Stukus, MD, is board certified in Allergy/Immunology and is an Assistant Professor of Pediatrics at Nationwide Children’s Hospital and The Ohio State University in Columbus, Ohio. His clinical and research interests focus on asthma and food allergies, especially improving education and adherence for patients and families. As part of his research, Dr. Stukus has created novel technology and educational tools using mobile health apps to improve the care of patients, for which he was recognized with the Nationwide Children’s Hospital Department of Pediatrics Junior Faculty Award in November 2013. Dr. Stukus has been an active member of the medical advisory team for Kids with Food Allergies since 2009 and was elected to the Board of Directors for the Asthma and Allergy Foundation of America in 2014. Lastly, Dr. Stukus actively engages with food allergy support groups and participates in social media on twitter through @AllergyKidsDoc.
Excellent piece!!!! This is a must read for every non-allergist. The headlines are a bit blinding right now and Dr. Stukus makes some beautiful points with clarity and the warmth we all enjoy.
GIna Mennett Lee says
Dr. Stukus, Thanks you once again for providing clarity on a very important topic. (Thanks to Henry, too, for providing the forum.) I have been following this story since it broke. This is the first piece I’ve seen to give a thorough analysis while also including an expert “take” on it’s impact. I am hopeful about this study’s findings but still very much concerned about the impact “perception” may have on those already managing food allergy. Thanks again!
Elizabeth Bauer says
Nice work Dr. Stukus.
I’m not in the medical field, I’m just a mom so forgive me if I’m missing something. But by moving the 13% that had severe reactions to the avoidance group ,and not counting them( their percentage) in the original group that was eating peanuts seems to be changing the real numbers. Am I missing something? Thanks for your insight.
Thanks for your question. I can certainly understand the confusion. This is what it says in the paper, as opposed to the coverage, including ours: “Participants who had a reaction to the baseline challenge were instructed to avoid peanuts. These participants were included in the intention-to-treat analysis but not in the per-protocol analysis.” Thus, they were not counted when the numbers were crunched, which, as you astutely point out, would have compromised the findings.