A Rebuttal Lands Fast on a Baricitinib Alopecia Study, and the Reason Is the Study Design Itself

A letter published in the Journal of the American Academy of Dermatology's October 2026 issue does something the peer-review system rarely gets credit for: it argues in public, on the record, about how much weight a single retrospective study should carry. The letter, by Calogero Pagliarello and Carlo Renè Girardelli, responds directly to a retrospective cohort study by Zhang X and colleagues on baricitinib's effectiveness and tolerability in alopecia areata, with or without prior corticosteroid use, appearing in the same JAAD volume (Vol. 95, Issue 4).

The distinction matters for anyone trying to read this exchange correctly. Zhang et al.'s piece is a retrospective cohort study, meaning it looked backward at patients who had already been treated and tried to draw comparisons between those who received corticosteroids before starting baricitinib and those who did not. That design can generate real signal, but it cannot control who got steroids first and who did not, which is exactly the kind of gap a correspondence piece like Pagliarello and Girardelli's typically exists to press on.

Baricitinib itself is not a new or experimental drug in this context. The Janus kinase inhibitor, sold as Olumiant, became the first systemic treatment the FDA approved specifically for severe alopecia areata, clearing that bar in 2022 on the strength of the BRAVE-AA randomized trials. Randomized trials are the reference point that makes a retrospective cohort study look, by comparison, like a lower rung of evidence: useful for real-world texture, weaker for causal claims about whether prior steroid exposure actually changes how well baricitinib works.

That gap between study types is precisely where a reply article like this one operates. Response letters in dermatology journals do not typically bring new patient data; they interrogate the data already in print, asking whether the comparison groups were similar enough to compare, whether confounding by indication was addressed, and whether the conclusions in an abstract match what the numbers in the tables can actually bear. Without the full text of either paper in hand, the specific methodological objections Pagliarello and Girardelli raise cannot be confirmed here, but the existence of the letter itself is a signal worth registering.

What can be said with confidence is narrower than the headline suggests, and that narrowness is the point. A retrospective cohort study on baricitinib and prior corticosteroid use, published in October 2026, has already drawn a formal published rebuttal in the same journal, within the same volume. The claim under scrutiny is not whether baricitinib works for alopecia areata, which the earlier randomized trials addressed; it is the narrower and messier question of what happens when steroids come first, and whether one retrospective chart review can settle that question at all.