Dipeptidyl Peptidase-4 Inhibition in Psoriasis Patients with Diabetes: A Double-Blind Randomized Controlled Trial

treat principle. Missing data were replaced using the last observation carried forward method. Between April 25, 2014, and April 22, 2015, 9 participants were randomized to the sitagliptin arm and 11 to the gliclazide arm. Recruitment was halted after 20 participants had been enrolled. The recruitment period and funding had finished at that time. The baseline PASI was similar in the sitagliptin (median 9.5, IQR, 7.8– 9.6) and gliclazide (median 9.4, IQR, 7.6–12) arms (Table 1). There was no significant difference between treatment arms in the change in PASI from baseline to 16 weeks (Fig. 1). The sitagliptin arm had a median reduction in PASI of 0.9 (IQR, 0.4–2.1) compared to a median reduction of 0.8 (IQR, –0.2 to 2.1) for participants treated with gliclazide (U = 43.5, p = 0.648). From baseline to week 32, participants in the gliclazide arm (16 weeks of gliclazide and 16 weeks of sitagliptin) had a median reduction in PASI of 1.8 (IQR, 0.3–3), and those in the sitagliptin arm had a median reduction of 3 (IQR, 2–5.6, U = 29.5, p = 0.128). At 16 weeks, 50% reduction in PASI from baseline (PASI-50) was achieved by one participant per treatment arm (9.1% of the gliclazide arm, 11.1% of the sitagliptin arm). No participants achieved PASI-75 or PASI-90 after 16 weeks of treatment. At 32 weeks, 4 participants (44.4%) in the sitagliptin arm achieved PASI-50 compared to 2 (18.2%) in the gliclazide arm (p = 0.336). One participant (11.1%) in the sitagliptin arm achieved PASI-75 after 32 weeks, compared to 0 (0%) in the gliclazide arm (p = 0.45). No participant achieved PASI-90 or PASI-100 after 32 weeks of treatment. The only statistically significant difference between treatment arms in the QOL endpoints was a small negative median change in the EuroQol 5-item questionnaire (EQ-5D) score in the gliclazide arm after 16 weeks compared to the sitagliptin arm, whose median EQ-5D value did not change (U = 25.0, p = 0.047). After 32 weeks, there was no difference in median change of EQ-5D scores between the two arms. At 16 weeks, the change in high-sensitivity C-reactive protein (hs-CRP) concentration was lower in the sitagliptin arm than in the gliclazide arm: hs-CRP concentration did not change with sitagliptin therapy and increased with gliclazide therapy (U = 21.0, p = 0.030). At week 32, this difference was no longer significant. No other between-group differences were noted in CVD risk factors. At week 16, the increase in active glucagon-like peptide-1 (GLP-1) and decrease in total gastric inhibitory polypeptide (GIP) after sitagliptin treatment was not significant compared to the gliclazide arm, which is interesting as sulphonylurea therapy has not demonstrated effects on GLP-1 and GIP in patients with T2DM [6]. The median insulin decreased significantly in the sitagliptin arm (median –3.91, IQR, –17.52 to 2.61) compared to the gliclazide arm (median 3.24, IQR, 1.36–7.54) at week 16 (U = 9.0, p = 0.02).