Placebo response, which has emerged as an important determinant of interpretability and success in atopic dermatitis (AD) clinical trials, is frequently a predictable consequence of conducting high-quality clinical studies in a disease characterized by inherent variability over time. Modern AD programs increasingly acknowledge placebo response as a structural feature rather than an anomaly. Adaptive designs, Bayesian augmented controls, explicit estimand strategies, and careful rescue medication handling are now essential tools to preserve interpretability in the current development environment.
While the last decade has yielded tremendous advances in systemic, and even topical, therapies for AD, few of these medications have been investigated in head-to-head clinical trials. That’s because if you break down RCT data for current AD therapies, they are not as comparable as you might think. Many participant-related, disease-related, and treatment-related factors are at play: inclusion or exclusion criteria, objective measurements used, how screening is designed, disease activity itself, comparator choices, statistical analysis, how you’re handling data, how things are evaluated over time, and what wash out looks like.
There are many areas in which things might go wrong, feeding into the concept of placebo rates being somewhat unpredictable in current studies. For example, if background topical corticosteroids (TCS) is used on an as-needed basis, participants in the placebo arm may use more TCS than participants in the active treatment arm. Placebo responses may also be influenced by the method by which participants obtain rescue treatment in a trial.
Examining placebo response longitudinally
Looking at placebo response in AD trials, longitudinal course is rarely accounted for or systemically captured. The reality is that some patients have episodic disease, some chronic, and others have flare patterns of difference, and that varies around the world and in time. There is no way to control hidden elements inside the disease state over the course of AD studies, which is likely another factor that feeds into placebo response.
However, when placebo response is examined longitudinally, a consistent pattern emerges. Across pooled placebo arms including those from dupilumab trials and subsequent biologic programs, placebo response typically plateaus by approximately week 12. Beyond this point, additional improvement is modest, particularly for objective endpoints such as Eczema Area and Severity Index (EASI) Extending trials beyond the placebo plateau does not inherently reduce placebo contribution and may instead increase variability.
Impact of baseline severity on clinical outcomes
Trends in the last decade of clinical trials include the impact of baseline disease severity on clinical outcomes. Using reported data from 2014 to 2023, an analysis of systemic treatments in adults with AD (Dermatol Ther, 2023) looked at baseline EASI scores in various AD studies, each 16 weeks: 33 studies with baseline EASI scores in overall population; 23 monotherapy studies with baseline EASI scores in placebo groups vs. % EASI reduction; and 14 monotherapy studies with EASI scores at baseline vs rescue medication usage in placebo arms.
The actual baseline EASI scores differed, even when one of the disease severity inclusion criteria was reported as being the same across the trials. EASI scores tended to be lower in more recent trials and may have had a considerable impact on efficacy outcomes, as suggested by potential associations with placebo response and use of rescue medication.
Model-based meta-analysis approach
A technique called longitudinal nonlinear mixed effects modeling, published in 2025, took into account trial phase concomitant therapy and baseline disease severity in order to understand the ideal protocol design in terms of time as well as points. This model-based meta-analysis approach supports the longitudinal effect of key covariates over time.
This model-based approach found that concomitant therapy is a significant driver of placebo response. Trials permitting TCS demonstrate an almost two-fold increase in placebo EASI-75 vs. monotherapy.
Baseline disease severity was found to be another significant driver of placebo response, with an inverse association of +1 increase in baseline EASI reducing EASI-75 placebo rate at Weeks 12/16 by 0.96x.
Time-course modeling suggested that placebo responses plateaued by Week 12. EASI-75 outcomes at Week 12 captured 94% of the projected response at Week 16.
Baseline EASI score’s impact on active treatment arms
Another point that does not relate directly to the placebo question, yet comes up often and is underexplored, is that baseline EASI score may not have the same magnitude of impact on the active treatment arm vs. placebo treatment arm. In a pooled post-hoc analysis of concomitant therapy trials for dupilumab (Ital J Dermatol Venerol, 2022), EASI <24 and EASI ≥24 had a similar response in the treatment arm. However, in the placebo arm, EASI ≥24 had a lower response versus EASI <24.
Recommended approach to placebo-controlled trials
The International Eczema Council (IEC), including Dr. Robert Bissonnette as a coauthor, released a survey-based position statement on the optimization of placebo use in clinical trials with systemic treatments for atopic dermatitis (J Eur Acad Dermatol Venereol, 2019). The statement noted that it requires balancing several, sometimes opposing principles in placebo-controlled trials with systemic treatment for AD, including ethics, methodology, regulatory requirements, and real-world needs.
The IEC paper looked at how concomitant use of TCS facilitates inclusion of more severely affected AD subjects but raises the placebo rate, while on the other hand, long TCS washout periods and avoidance of TCS during trials select for subjects with milder AD and also risks a higher placebo response.
Other results from the consensus e-survey taken by IEC members included recommendations for:
- Selecting sites and assessors with recognized expertise in AD clinical trials
- Clear definition and identification of baseline disease severity
- Minimizing time and proportion of patients on placebo
- Instigating open-label extension studies for enrolment after a predefined time point
- Including outcomes which set a higher bar for disease clearance
- Including patients with milder disease (“eligibility creep”)
- Including environmental influences (sun, temperature, humidity, and seasonality)
Conclusion
Sponsors would do well to recognize placebo response as a predictable structural phenomenon, in AD trials and other dermatology trials, and design studies accordingly. Efforts are ongoing to standardize the design and reporting of future AD clinical trials, so that they will be best positioned to generate interpretable data and make informed development decisions in an increasingly competitive AD landscape. As well as study design, sponsors need to address methodologic factors, patient population factors, concomitant topical therapy, and baseline disease severity going forward.
About the Author
Raj Chovatiya, MD, PhD, MScI, is an Associate Professor at Rosalind Franklin University of Medicine and Science/Chicago Medical School, and the founder and Director of the Center for Medical Dermatology Immunology Research in Chicago. He is a board-certified dermatologist and clinical investigator with subspecialty expertise in atopic dermatitis and inflammatory skin disease. His work focuses on therapeutic development, clinical trial design, and improving interpretability of outcomes in dermatology research. Dr. Chovatiya is actively involved in industry-sponsored and investigator-initiated trials and frequently advises on trial methodology, endpoint selection, and placebo response in moderate-to-severe atopic dermatitis.
