Introduction

Atopic dermatitis (AD), a chronic immune-mediated disease characterized by intense pruritus, is associated with debilitating effects on patients’ lives. Patients with moderate-to-severe AD report a multidimensional burden including itch, pain, sleep disturbance, anxiety and depression, and impaired health-related quality-of-life (HRQoL) [1].

Dupilumab is a fully human monoclonal antibody that targets the interleukin (IL)-4 receptor-alpha subunit and inhibits signaling of cytokines IL-4 and IL-13, both of which drive, at least in part, atopic diseases including AD [2]. The efficacy and safety of dupilumab was evaluated in two phase 3 clinical trials in adults with moderate-to-severe AD whose disease was inadequately controlled by topical treatment, SOLO 1 and SOLO 2 (ClinicalTrials.gov Number, NCT02277743 and NCT02277769, respectively) [3]. Published results from both trials consistently showed that, compared with placebo, subcutaneous dupilumab 300 mg once weekly (qw) or every 2 weeks (q2w) significantly improved objective signs and symptoms of AD including pruritus, symptoms of anxiety and depression, and HRQoL assessed using a dermatology-specific measure [3].

This analysis further expands on the effects of dupilumab on HRQoL by using the 5-dimension 3-level EuroQoL (EQ-5D) [4], which was included in SOLO 1 and SOLO 2, to evaluate changes in overall health status and specific dimensions that contribute to HRQoL.

Methods

The EQ-5D describes health states through the generation of a utility score based on levels of impairment on 5 dimensions; utility scores represent health states anchored at perfect health (=1) and death (=0), although states lower than 0 are attainable. Since trial designs were identical [3], we report EQ-5D results from both trials in a pooled analysis using censoring after rescue and last-observation-carried-forward. Protocols for both trials were approved by the appropriate institutional review boards/ethics committees at each study site, and all patients provided written informed consent prior to study participation.

Treatments were compared overall and for clinical responders using analysis of covariance (ANCOVA) with baseline as covariate and treatment, region, and baseline Investigator’s Global Assessment (IGA) strata as fixed factors; a study identifier was included as an additional factor for the pooled analysis. Changes from baseline in EQ-5D scores are presented as least squares means, i.e., results from the ANCOVA model that adjusted for the covariate. Clinical responders were defined as patients who achieved at week 16: IGA score of 0 or 1 (clear or almost clear) and a reduction from baseline ≥2 points (primary study endpoint), and improvements in the Eczema Area and Severity Index (EASI) [5] ≥50% and ≥75%; EASI ≥75% response was the key secondary endpoint.

The proportions of patients who reported the different levels of problems on the individual dimension of the EQ-5D were also compared by treatment group using Fisher’s exact test.

Results

Patients (n = 1379) were 57.9% male with a mean (SD) age of 38.3 (14.3) years. Baseline EQ-5D utility scores ranged from 0.611 to 0.629 across treatment groups (Table 1). These scores indicated impaired HRQoL, which was slightly worse than the HRQoL reported for moderate-to-severe psoriasis (0.642) [6], as well as the general population norms for the UK (0.856) and US (0.867) [7]. Patients treated with dupilumab at both dosing regimens reported significant improvements in HRQoL by week 16 as indicated by increases in EQ-5D utility scores (Table 1). These increases resulted in scores that approached population norms [4]; were in the same range as that of biologic agents for psoriasis with the strongest effects on EQ-5D [8]; and were clinically meaningful, as they exceeded the reported minimal clinically important difference of 0.082 [9].

Table 1 Change from baseline in EQ-5D utility scores at week 16, with censoring after rescue treatment and last-observation-carried-forward for imputation of missing data (full analysis set)
Full size table

Changes from baseline were also clinically meaningful when analyzed by response levels (Table 1), showing that IGA and EASI responders have a clinically meaningful change in utility scores, suggesting that improvements in utility scores are likely associated with clinical response. This relationship between clinical improvement and change in utility scores is further supported by the observation that changes in utility scores among dupilumab responders were consistently higher than the overall mean changes in utility scores with dupilumab. It should also be noted that, since clinical responders are defined by the response criteria regardless of treatment allocation, it is not surprising that placebo IGA and EASI responders had meaningful changes in EQ-5D. Improvements in HRQoL in these patients would also be expected to be driven by the clinical improvements in AD. This was evidenced by similar changes in EQ-5D utility scores for the placebo and dupilumab IGA and EASI 75% responders, even though the proportions of responders were significantly higher with dupilumab (Table 2).

Table 2 Patients who reported at least some problems on individual EQ-5D dimensions at baseline and 16 weeks
Full size table

At baseline, patients reported large burdens on the individual EQ-5D dimensions of pain/discomfort, anxiety/depression and usual activities as indicated by substantial proportions of patients with “problems” (Table 2). Improvements in the EQ-5D at week 16 were primarily driven by increased proportions of patients reporting “no problems” in these three dimensions with dupilumab qw (62.2–85.8%) and q2w (62.0–85.0%) relative to placebo (39.2–69.5%; all P < 0.05) (Table 2).

Conclusions

These results show that patients with moderate-to-severe AD have substantially impaired HRQoL, and that dupilumab treatment results in improvements in not just skin-specific but also general HRQoL that are statistically significant relative to placebo and are clinically meaningful. A limitation is that, in order to quantify general health, the EQ-5D simplifies it into 5 dimensions. However, the advantage of assessing general health using the EQ-5D is that it allows comparison of disease burden and the impact of treatment on general health across diseases.