On the skin, modestly. A phase 3b trial randomized 274 adults with moderate to severe psoriasis and overweight or obesity. Complete skin clearance at week 36 was 40.6% on ixekizumab plus tirzepatide against 29.0% on the biologic alone [1]. That is a risk difference of 11.6%, on a confidence interval running from 0.3% to 22.9%. The trial’s headline endpoint was much larger, and half of it was settled before anyone was treated. No trial has tested a GLP-1 drug on psoriasis without a biologic alongside it.
Between 60% and 78% of people with psoriasis carry overweight or obesity, and weight worsens both the disease and the response to treatment. That relationship is what makes the trial below worth running, and it is also what makes its headline number hard to read. The comorbidity load in psoriasis is counted separately. A phase 3b trial at 72 US sites tested treating both at once, randomizing 274 adults to the biologic ixekizumab with or without tirzepatide [1].
The headline result is large. At week 36, 27.1% of the combination arm reached complete skin clearance and lost at least 10% of their weight. On the biologic alone it was 5.8%, a risk difference of 21.2% (95% CI 12.8%–29.7%).
The number that answers the actual question is the skin-only one. Complete clearance was 40.6% with tirzepatide against 29.0% without, a risk difference of 11.6% whose confidence interval runs from 0.3% to 22.9% (P = .04). The lower bound sits three tenths of a percent above zero. That is a positive result by convention and the most fragile figure in the trial. It is also the one a dermatologist weighing a second drug actually needs.
Two design features deserve weight here. The trial was open-label, so everyone knew who was receiving what, and PASI scoring involves investigator judgment. That combination tends to favor the more intensively treated arm on a visually assessed endpoint. Gastrointestinal events were more frequent on the combination. The trial reports that as consistent with known safety profiles, and it is still a tolerability cost attached to the extra 11.6%. The same class of stomach complaints is described from the mechanism side in nausea and appetite on one circuit.
What the record-based cohort adds, and what it cannot
A US database cohort matched 3,048 psoriasis patients on a GLP-1 drug against 3,048 taking other antidiabetic or antiobesity drugs and followed both for two years [2]. All-cause mortality came out at a hazard ratio of 0.219 (95% CI 0.123 to 0.391) and major adverse cardiac events at 0.561 (95% CI 0.442 to 0.714).
A hazard ratio of 0.219 describes a group dying at roughly a fifth the rate of its comparator. No randomized trial of this class has produced anything close to it. Neither arm measured whether anyone’s psoriasis improved. The full reading, including why the cardiac figure is easier to believe than the mortality one, is in the psoriasis comorbidity cohort.
The adjacent skin disease, and the grade of its evidence
Hidradenitis suppurativa is a different condition with the same obesity link, and it shows what this literature looks like without a randomized comparison. An open-label study of 20 adults on semaglutide for six months reported weight down 21.4 kg, quality of life up 8.8 points and pain down 4.1, all at p<0.001 [3]. There was no control group and no blinding, and three of those four outcomes are self-reported.
A separate matched-record cohort in the same disease reported all-cause mortality at a hazard ratio of 0.24 (95% CI 0.145 to 0.395). Exposure required at least a year of therapy [4]. Anybody who died inside that year could not be counted as exposed. Both studies are set out in the twenty-person pilot and the hidradenitis mortality cohort.
What the trial does establish
The two drugs can be given together for a year without a new safety signal emerging, in a population where doing so was previously untested at this scale. Whether the skin benefit is worth a second injection is a judgment the confidence interval leaves genuinely open. The broader pattern of these drugs producing modest effects across many separate conditions runs through the heart failure evidence and the carpal tunnel comparison.
No telehealth seller prescribes for psoriasis, and nothing here is a reason to start or stop a drug. The record cohort is useful for reading a marketing claim. The risk reductions were larger in the psoriasis group than in groups without it. That is the kind of subgroup finding a landing page quotes without its population attached. Before a first order, the more useful questions are about who is prescribing and what happens at a higher dose, and those are listed in questions they leave open.