Still on Prednisone After Fourteen Months: What Counts as Failing Benralizumab
A steroid-dependent patient whose biologic has clearly not done its job, with CHEST's own algorithm and a cautionary switch trial pulling the next decision in two different directions.
Aaron P. runs the line at a busy diner kitchen, six shifts a week on his feet in front of a flat-top grill that does his asthma no favors on a good day. He started benralizumab fourteen months ago after years of maintenance prednisone, hopeful enough about coming off steroids entirely that he mentioned it to his whole kitchen crew before the first dose. He has not come off them. He is still on 5 mg of daily prednisone, still had two courses of burst-dose steroids in the past year, and his own read — “I feel about the same as before, just with more needles now” — is not far from what his chart shows.
CHEST’s definition of a good clinical response sets a real bar: roughly a 50% reduction in exacerbations alongside a 50% reduction in maintenance OCS dose, or a 50% exacerbation reduction alone with OCS unchanged. Aaron clears neither. His exacerbation rate has fallen only modestly — two burst courses this year against three the year before starting — and his OCS dose has not moved at all in fourteen months on treatment, still the same 5 mg he walked in on. By the guideline’s own language, this is not a borderline call requiring interpretation — it is the case its four-to-six-month reassessment window exists to catch, run considerably past that window because his prior team kept extending the trial hoping the numbers would still turn. They have not turned, and the cushingoid changes showing up on his face now are their own quiet argument against waiting any longer to find out.
Fourteen months in, deciding what “not working” actually means
He clears CHEST’s own definition of inadequate response on both counts — his exacerbation reduction falls short of the 50% bar, and his OCS dose has not moved at all in fourteen months. For a steroid-dependent patient in exactly this situation, the guideline’s stated preference is dupilumab over tezepelumab, specifically because of its demonstrated ability to reduce maintenance OCS dosing. I would not treat this as a close call.
I agree he has failed benralizumab by any reasonable reading of his numbers. Where I would pause is on where to go next — IL-5 signaling and IL-4/13 signaling both sit inside the same broad type-2 inflammatory pathway, and he has already shown us that pathway is not doing what we need for him. Tezepelumab works through a genuinely different mechanism upstream of both. After one failure within a pathway, I would want real consideration of stepping outside it rather than one lateral move within it.
I am not disputing that dupilumab has real OCS-sparing evidence in general — I am asking whether that evidence was built in patients like him, who have already shown the broader type-2 pathway underperforming.
The mechanistic caution is fair in the abstract, but I do not think it is the strongest evidence actually in front of us for Aaron specifically. CHEST’s algorithm was built for exactly his situation — a documented anti-IL5/5Rα non-responder who is steroid-dependent — and it names dupilumab first, on real OCS-sparing data, not as a default lateral step but as a specific recommendation for this scenario. NIMBLE’s caution is real, but it answers a different question: whether an already well-controlled patient can be switched onto a new dosing interval without losing ground. That is not what we are deciding today. We are deciding what a documented non-responder should try next, and the guideline built for exactly that question already has an answer.
Agreed: discontinue benralizumab, start dupilumab, with a formal reassessment at four to six months against CHEST's own response criteria — not extended informally past that window the way the prior trial was.
Agreed also, and stated explicitly rather than left implicit this time: prednisone taper attempts wait for confirmed dupilumab response, so a failed taper cannot be mistaken for a failed biologic.