Placebo response in two long-term randomized psoriasis studies that were negative for rosiglitazone

Previous research has suggested that the thiazolidinedione rosiglitazone may possess anti-psoriatic activity.

To compare the efficacy and safety of rosiglitazone with that of placebo in the treatment of chronic plaque psoriasis.

Two large-scale, randomized, double-blind, multicenter studies (study A, n = 1563; study B, n = 1032) were conducted over 52 weeks (plus optional 44 weeks safety extension) in an outpatient setting. The subjects (aged 18-75 years) had moderate-to-severe chronic plaque psoriasis affecting >or=10% body surface area (BSA) with plaques of any elevation above normal-appearing skin (or >or=6% BSA involvement with marked elevation) and had not used phototherapy during the previous month or thiazolidinediones within the previous 3 months. Rosiglitazone was administered as 2, 4, or 8 mg tablets once daily. The main outcome measure was the proportion of subjects achieving >or=75% improvement from baseline in the Psoriasis Area and Severity Index (PASI 75) at week 26.

Rosiglitazone was well tolerated but no more effective than placebo for moderate-to-severe chronic plaque psoriasis. However, there was a large placebo response unrelated to concomitant rescue medication. Interestingly, subjects had been advised to expect a long period before onset of action. At week 26 and across both studies for subjects receiving placebo, the PASI 75 was 9% (48/506) and the PASI 50 (proportion of subjects who achieved at least 50% improvement from baseline) was 27% (135/506). In addition, few subjects withdrew from placebo or rosiglitazone treatment because of 'lack of efficacy' and the majority persisted in the year-long study.

While these large-scale, robust studies demonstrated that rosiglitazone is not active in psoriasis, they also showed that for a large proportion of subjects receiving placebo, the expectation of a successful treatment, the favorable adverse effect profile of the drug, and the supportive environment of a clinical study conferred beneficial effects. These results may have implications for the design of future placebo-controlled studies in patients with psoriasis.