Three Years Into Peanut Immunotherapy: Testing Whether the Tolerance Will Hold
A boy who has self-managed peanut oral immunotherapy without incident for a year wants to know whether he can eventually stop — and the only trial that tested stopping found most patients couldn't.
Teo R., a 16-year-old boy, has been the one drawing up his own peanut-OIT dose every morning before school for the past year, a routine he inherited from his mother once his allergist judged him old enough to manage it himself. He started oral immunotherapy for peanut allergy at 13, three years after a reaction to a shared birthday cake sent him to the emergency department, and has been on a stable 300mg daily peanut-protein maintenance dose for the past 18 months without a single reaction. He has no other food allergies, no asthma, and is now looking two years ahead to college housing applications that ask, among other things, whether he needs a peanut-free dorm floor — a question he would like to be able to answer differently than he can today. He plays club soccer, works a part-time job at a hardware store on weekends, and describes the daily dose itself as a non-event at this point, something closer to brushing his teeth than a medical procedure — which is part of why the idea of it simply continuing indefinitely, into a dorm room nobody in this room will be checking, has started to bother him more than the dose itself ever did.
His most recent peanut-specific IgE, once markedly elevated, has fallen to 4 kU/L, and his allergist has raised the possibility of formally testing for sustained unresponsiveness — stopping OIT dosing entirely for a defined interval, then re-challenging to see whether protection holds without the drug on board. POISED, the Chinthrajah trial that tested this question directly, built participants aged 7 to 55 up to a 4,000mg daily maintenance dose for two years and then either stopped peanut outright or dropped them to 300mg daily for a further year. Three months after stopping outright, 35% still passed a 4,000mg challenge; by a full year off, only 13% did — while 37% of those held at 300mg daily were still passing at that same point. The number that should give Teo pause is which arm he actually resembles: his 300mg maintenance dose is POISED's reduced arm, not its maintenance arm, so the trial's own most favorable discontinuation figures were earned from a dose more than ten times his. Sustained unresponsiveness and simple desensitization are not the same finding, a distinction the trial was built specifically to separate: desensitization describes protection that holds only while dosing continues, which Teo has clearly achieved, while sustained unresponsiveness describes protection that survives the drug's absence, which is precisely the harder, less common outcome POISED set out to measure.
Two years before he manages this without anyone checking
Teo has done everything right for three years, and I think it's reasonable to test whether that work has produced something that survives without the drug. POISED, Chinthrajah's trial, is the one that asked this directly rather than assuming an answer: participants stopped dosing entirely and were re-challenged, first at three months and again at a year. If we don't test it, we're guessing indefinitely instead of finding out.
I want to be clear this is a proposal to test, not a decision to stop for good today — a positive challenge tells us something real, and a negative one tells us something just as real, sooner than continuing to wonder.
I take the case for testing seriously, but POISED's own numbers argue for caution about what a positive result would even mean for Teo specifically. The two figures get quoted interchangeably and they shouldn't be: 35 percent of those who stopped outright still passed at three months, but only 13 percent were still passing a year later, and the more favorable 37 percent came from patients who never stopped at all — they stayed on 300 milligrams daily. There is no weekly or intermittent schedule anywhere in that trial, and I want that said plainly before anyone builds one. His IgE trajectory is genuinely reassuring, and POISED's discontinuation arms came off 4,000 milligrams a day, not the 300 he takes — so its reversion rates are, if anything, the optimistic version of his own question, not a matched one. Losing three years of protection to find that out carries real cost if it goes the way POISED's own numbers say it most often does.
The trajectory argument would be stronger if we had a marker POISED itself validated as predicting which patients hold tolerance and which don't — right now we don't, so a falling IgE is suggestive, not determinative, of which group he'd land in.
I think there's a third option nobody's put on the table yet, and it maps onto what Teo is actually going to be doing in two years: not a full stop, and not indefinite full-dose maintenance either, but a stepped-down, less-frequent dosing schedule closer to what a busy college student can realistically sustain without daily parental oversight. That's not what POISED tested, so I can't point to trial numbers the way either of you can — but the actual goal here isn't proving a hypothesis, it's keeping a specific 18-year-old protected once he's the only one checking. A schedule he can actually keep is worth more than a theoretically higher-tolerance schedule he quietly stops following in a dorm room.
Agreed: hold at full maintenance for now, revisit the sustained-unresponsiveness question closer to his college departure rather than testing it today, specifically so any transition — a full stop, or the pediatrician's stepped-down schedule — happens under supervision rather than mid-semester. Not agreed: which of those two paths the group should actually aim for when that later conversation happens.
The allergist still wants to formally test for sustained unresponsiveness eventually; the pharmacologist remains skeptical a positive result would mean lasting protection given POISED's own reversion rate; the pediatrician's stepped-down-dosing proposal was written into the chart as a named option for that future visit, not adopted or dismissed today.