Potential contributors to placebo rates in hidradenitis suppurativa (HS) clinical trials include underlying disease variability, the relative lack of efficacy of current therapies, and the requirement for rescue therapy and/or concomitant medications. More specifically, the contributors to placebo response lie in the difficulty in evaluating disease severity in HS, lack of non-physician or patient-assessed objective measures of HS improvement, lack of standardized definitions for lesion types, heterogeneous disease presentation, differing responses by disease phenotype or lesion type, and trial size and length, and site selection.
Three-paper comparison of HS trials
Three analyses of the phase 3 PIONEER I and II adalimumab trials looking at placebo responders, hematologic abnormalities, and C-reactive protein (CRP) and adalimumab response, had a shared theme: Lower-burden HS can fluctuate enough to create placebo response, while higher inflammatory burden shows up in abnormal labs and lower odds of biologic response.
A post-hoc placebo-arm analysis of PIONEER I/II through week 12 (Int J Women’s Dermatol, 2024) tackled the question: Is placebo response in HS just noise, or does it reflect real biologic improvement? Among placebo recipients, 27.6% achieved HiSCR. Placebo responders had lower baseline severity and lower baseline CRP, plus a modest CRP decline, versus placebo nonresponders. That suggests some placebo response reflects natural fluctuation or spontaneous improvement in lower-burden HS rather than measurement error alone.
A post-hoc PIONEER I/II analysis of 594 patients with baseline laboratory data, focused on anemia, leukocytosis, and thrombocytosis (JAMA Dermatology, 2024). Of these patients, 42.4% had at least one hematologic abnormality at baseline. The analysis asked if adalimumab improved HS-associated blood count abnormalities, and it found that adalimumab improved hemoglobin, platelet count, and white blood cell count over 12 weeks; placebo did not. This frames HS as a systemic inflammatory disease; successful treatment can improve lab abnormalities, not just skin lesions.
A post-hoc PIONEER I/II analysis of 588 patients with baseline and week-12 CRP data (JAMA Dermatology, 2025), examined if baseline CRP can help predict who responds to adalimumab. Of these patients, 78.9% had elevated baseline CRP. Elevated CRP tracked with higher body mass index (BMI) and more severe disease. Adalimumab helped across CRP strata, but very high CRP predicted lower odds of response among treated patients. This positions CRP as a practical biomarker for inflammatory burden and for considering escalation, drug-level checks, or alternate biologics.
Key considerations for trial design
In the design of phase 2 and 3 clinical trials in HS, it is worth considering whether the following factors could have a net effect of lower placebo noise, cleaner baseline severity, and a more credible estimate of drug effect.
- Minimal lesion counts for entry criteria, creeping over the years, up to Abscesses + Inflammatory Nodules (AN) count of 6. Also, a combination of AN and draining tunnels (dT) counts as a baseline criterion is a possibility.
- Tracking severity of enrolled patients at set timepoints, recording both Hurley stage and a monitor overall average dT and AN count.
- Making photography mandatory, with standardized body-region photos for every patient, and central review.
- Reviewing and documenting rescue criteria, including protocol language for intercurrent event analysis, and educating sites on rescue plans (antibiotics and intralesional interventions).
- Protecting randomization from imbalance. Stratification has occurred for disease severity, smoking, BMI, prior biologic exposure, and region/site cluster.
- Auditing sites and electronic Case Report Form (eCRF) early and often. (Blinded review of baseline distributions and outlier lesion counts; proposed caps for sites until this data is reviewed.)
Response differences by disease severity and/or phenotype
There is quite a bit of difference in response by severity and/or phenotype. There is this idea of a follicular or nodular patient population where they have very high AN count (>10) with few (<3) to no dT. These patients are more likely to be female with greater disease variability, and potentially more responsive to certain medication types.
For all other inflammatory phenotypes, patients generally have longer-standing, more severe HS and are less treatment-responsive. These patients tend to have lower placebo rates but are much harder to treat. Patient populations where males are predominantly affected may not be as responsive to medications.
Secukinumab in Hidradenitis Suppurativa
In an exploratory post hoc machine-learning analysis of secukinumab in HS patients, (SUNSHINE/SURISE Phase 3 trials), these clinically recognizable patient clusters emerged:
Cluster 1 (milder HS) comprised 54.1% of the 1,084 patients:
- Predominantly female: 65.4%
- Hurley III: 19.3%
- IHS4 mean 16.3; IN 8.5, abscesses 1.5, tunnels 1.2
- Better baseline PROs: DLQI 11.9, pain NRS 4.3, EQ-5D VAS 67.3
- Less current antibiotic use: 7.3%
Cluster 2 (moderate HS) comprised 17.8% of patients:
- More Asia Pacific, Middle East & Africa (APMA) region: 58.5%; Asian race: 56.0%
- Predominantly male: 58.0%; younger: 30.9 y; lower weight: 82.3 kg
- Hurley III: 37.3%
- IHS4 mean 26.6; IN 10.5, abscesses 2.8, tunnels 2.6
- Shortest time since diagnosis: 8.4 y; lowest family history: 9.8%
Cluster 3 (severe HS) comprised 28.1% of patients:
- Highest prior biologic exposure: 45.9%
- Highest prior HS surgery: 47.5%
- Hurley III: 71.8%
- IHS4 mean 46.4; IN 12.6, abscesses 6.0, tunnels 5.4
- Worst baseline PROs: DLQI 18.9, pain NRS 6.7, EQ-5D VAS 53.8
In all 3 of these phenotypes, both secukinumab every two weeks and secukinumab every 4 weeks outperformed placebo for HiSCR, AN count, flares, and NRS30. The lowest placebo response was in cluster 3 (severe) which showed a trend favoring secukinumab every 2 weeks over secukinumab every four weeks, over a longer period (52 weeks). The clinical implication of this analysis is that more severe, tunnel-rich, treatment-experienced patients may need frequent dosing and longer observation before declaring non-response.
Data ‘wobbles’
In what is known as ‘data wobble,” you cut off a trial at week 12, instead of week 16, you might have met statistical significance because patients did better at week 12, but efficacy fell off at week 16. (Of course, sometimes you cannot choose a long trial period due to regulatory requirements on study length.)
A retrospective view of data wobbles in hidradenitis suppurative clinical trials (Int J Womens Dermatol. 2024) we aimed to see how often study arms in HS programs ‘wobble.’ Reviewing HS clinicaltrials.gov studies of systemic, nonantibiotic drugs using HiSCR, supplemented by press releases and conference presentations, we looked at 21 studies, 35 study-drug arms (27 randomized, 8 open-label), and 14 placebo arms. The review found that HiSCR wobble was more common in randomized arms than in open-label arms: 11/27 (40.7%) vs 0/8 (0%).
We looked at whether HISCR 50 or HISCR 75 might be the better primary end point in HS trials. Two phase 3 studies have used HISCR 75 as a primary outcome (Izokibep, Sonelokimab) to varying degrees of success.
We found that wobble clustered in arms with somewhat less severe baseline disease. Potential contributing factors include a higher proportion of less severe patients at baseline and more female patients; specifically:
- Lower draining fistula counts: 2.3 vs 3.2
- Fewer Hurley stage III patients: 33.2% vs 42.5%
- Lower weighted abscess and nodule counts: 12.1 vs 12.6
- Lower weighted DLQI: 12.5 vs 14.5
- Higher proportion of female patients: 63.9% vs 58.3%
The implication for trial design is that tighter baseline severity confirmation, more standardized lesion counting, and endpoint planning that accounts for lesion-count volatility, may reduce false signal loss at the primary endpoint.
Using technology to improve disease severity assessments
Technology models from other disease states include the requirement for standardized photographs for all patients including at screening visits, and central quality control/adjudication. Benefits of photography in HS trials could include:
- Creating a centralized audit trail
- Possible able to count some types of lesions
- Support Hurley stage confirmation when local staging is uncertain
- May allow better tracking at individual site level, particularly for lesion interventions
- Machine learning algorithms could be applied to phenotype patients.
However, photography is not enough in HS trials, for several reasons:
- 2D and 3D photos cannot assess lesions that are primarily dermal in nature (could lead to unclear adjudication process)
- Skin pigmentation and obesity paired with anatomic areas are significant limitations
- Patients may not be willing to have total body photography (sensitive areas) and without this, no lesion counts or staging could be performed accurately
- Time consuming for sites and patients.
- While ultrasound or MRI would potentially fill the gap for dermal lesions, and basic erythema imaging applications might be helpful, ultrasound has not been shown to be an effective outcome measure, and these other imaging modalities could be even more time consuming and costly, requiring more user knowledge and expertise.
Overall, in HS clinical trials, some improvements can likely be made to ensure a diverse patient population, and other more objective “markers” or “biomarkers” may improve placebo response rates, a critical issue to address in other dermatology trials, too, such as atopic dermatitis.
About the author
Martina Porter, MD, is the vice-chair for research and academics in the department of dermatology at Beth Israel Deaconess Medical Center. She is also the director of the Clinical Laboratory for Epidemiology and Applied Research in Skin (CLEARS), and the co-leader of the Pathogens, Immunology, and Inflammation Translational Research Hub at Beth Israel Deaconess Medical Center. Chair of the American Academy of Dermatology’s Patient Safety and Quality Committee, Dr. Porter specializes in treating immune-mediated dermatologic conditions, including hidradenitis suppurativa, psoriasis, atopic dermatitis, and inflammatory bowel disease–related dermatoses, with biologics and small molecules, and leads both investigator-initiated and phase 2 and 3 industry-sponsored clinical trials in these disease indications.
