Cendakimab for Eczema: Why a Nominal P Value of .03 Was Not a Confirmed Dose Result
The analysis plan can change what a p value supports
A p value below .05 does not always authorize a study to claim a confirmed treatment effect. When several doses are tested, investigators may follow an ordered testing plan to control false-positive findings. Cendakimab's eczema trial gives a concrete example: a lower-dose nominal result looked favorable, but the testing sequence had already stopped.
The adult dose-ranging trial
The 2024 primary publication reports a randomized, double-blind, placebo-controlled phase 2 trial in adults with moderate-to-severe atopic dermatitis inadequately controlled by topical medication. It randomized 221 participants; 220 received study treatment. The primary outcome was mean percentage EASI change at week 16. The planned order tested weekly 720 mg first, then 720 mg every two weeks, then 360 mg every two weeks.
| Finding | Meaning |
|---|---|
| 720 mg weekly | EASI change −84.4% versus −62.7% placebo; p = .003. The primary test succeeded. |
| 720 mg every two weeks | −76.0% versus −62.7%; p = .06. This interrupted the testing sequence. |
| 360 mg every two weeks | Adjusted treatment difference −16.3; nominal p = .03. Confirmatory significance was not claimed. |
| Population and duration | An adult phase 2 study assessed over 16 weeks, not a lifelong or pediatric safety demonstration. |
Why the lower-dose finding was still interesting
The testing hierarchy does not make the observed lower-dose difference disappear. It limits the strength of the formal claim that can be attached to it. Nominal means the p value is being reported without the full confirmation permitted by the ordered multiple-testing plan. Treating it as exploratory preserves a possible research signal while respecting the method chosen before results were known.
This is different from saying that the whole medicine failed or that every dose succeeded. Both would oversimplify a mixed dose-ranging result. Our guide to eczema evidence explains why a result needs its comparator, endpoint and analysis plan. A mean score change also should not be converted into the percentage of people cured.
A favorable phase 2 result is not a treatment ranking
The study was designed against placebo, not against every other IL-13 therapy. Its percentages cannot be lined up beside a different trial's percentages to declare a winner without accounting for populations, endpoints and background care. Small differences between dose groups also do not establish the best personal regimen. Study doses here are research details, not prescribing advice.
Bring a clear goal to a specialist review
Ask which currently authorized choices fit your diagnosis and which have the strongest evidence for the outcome you need. Keep itch and visible disease separate, and record sleep and daily function. Review current AAD guidance alongside the exact product under discussion. A publication does not, by itself, establish a product's local approval, availability or instructions. Longer-term risks and benefits require evidence beyond a 16-week phase 2 trial.
Daily moisture care remains a separate decision
Dr. Arsy #003 4 oz moisturizer is a cosmetic option for hydration, moisture-barrier support and dry-skin comfort. It is not an eczema medicine, and this trial did not test #003. The sports editorial image does not imply prescription-like efficacy. Ask a clinician about product suitability for a child or actively inflamed skin.
A related evidence question
A different exploratory threshold creates another interpretation question in Tozorakimab for Eczema: Why FRONTIER-2's Numerical Responder Signal Did Not Establish Its Primary Benefit. These are separate studies, not a direct comparison of medicines.