Design: phase 2 randomised trial of two dosing schedules, no placebo arm (NCT03811912)
Population: 57 adults with severe alopecia areata, mean age 40.1 years, 66.7% female
Primary outcome: relative change from baseline in Severity of Alopecia Tool score at week 24
Effect: 46.4 (90% CI 34.3–58.4) with 8 mg twice daily vs 18.0 (90% CI 7.6–28.4) with 16 mg once daily; more patients reached SALT ≤20 and ≤10 on the split schedule

Deuruxolitinib is a Janus kinase 1/2 inhibitor for severe alopecia areata, and the practical question in a phase 2 programme is how to give it. This trial randomised 57 adults with severe disease (mean age 40.1 years, 66.7% women) to 8 mg twice daily or 16 mg once daily — the same 16 mg total — and measured the relative change in Severity of Alopecia Tool (SALT) score at week 24.

The split dose was clearly better: mean relative SALT improvement 46.4 (90% confidence interval 34.3 to 58.4) on 8 mg twice daily against 18.0 (90% CI 7.6 to 28.4) on 16 mg once daily, with more patients reaching SALT ≤20 and ≤10 on the twice-daily schedule. Treatment-emergent adverse events were mild or moderate in both groups.

The interesting part is what it implies about the drug class. If total daily exposure were what mattered, the two arms would have performed alike. They did not, which points to sustained target coverage rather than peak concentration driving hair regrowth — and that is a reasonable prior to hold about other Janus kinase inhibitors in this disease, though it has not been tested for them. At 57 patients this is a dose-selection study, not an efficacy trial, and the phase 3 programme is where the effect size gets established.

Read this as dose-schedule evidence, not as evidence that deuruxolitinib works — there was no placebo arm.
Where a Janus kinase inhibitor in alopecia areata has a twice-daily licensed schedule, do not consolidate it to once daily for convenience.
Use SALT scoring and photographs at baseline so response at 24 weeks is measurable rather than impressionistic.
Set the expectation that meaningful regrowth takes months, and review at 24 weeks rather than earlier.
Apply the usual Janus kinase inhibitor screening and monitoring regardless of schedule.

Why it matters

It suggests sustained target coverage rather than peak concentration is what regrows hair, which is a question worth asking of the whole class.

The statistics, in plain English

These are 90% confidence intervals, not the usual 95%, which is conventional in dose-finding work but means the intervals are narrower than they would otherwise be — treat them as less conservative than they look. Even so, the two intervals (34.3–58.4 and 7.6–28.4) do not overlap, so the difference between schedules is not a chance finding. What 57 patients cannot tell you is the size of the benefit against no treatment, because there was no placebo arm and both groups received active drug.