Brensocatib in hidradenitis suppurativa: a DPP1 inhibitor that worked in bronchiectasis and lost to placebo here

What was tried

Insmed ran CEDAR (NCT06685835), a Phase 2b, randomized, quadruple-masked, placebo-controlled trial of the oral DPP1 inhibitor brensocatib in adults with moderate to severe hidradenitis suppurativa. The study enrolled 214 patients across 72 sites in 11 countries [1]. Participants were randomized in a 1:1:1 ratio to brensocatib 10 mg, brensocatib 40 mg, or placebo once daily for 16 weeks [2]. The primary endpoint was the percent change from baseline in total abscess and inflammatory nodule (AN) count at Week 16, with HiSCR50 and HiSCR75 among the secondary endpoints [1]. Brensocatib (INS1007, AZD7986, CHEMBL3900409) is a reversible inhibitor of dipeptidyl peptidase 1, the lysosomal enzyme encoded by CTSC that activates the neutrophil serine proteases neutrophil elastase, proteinase 3, and cathepsin G [3]. The trial record lists the reason for stopping as "Lack of Efficacy" [1].

The biological hypothesis

Brensocatib was built for neutrophil-driven disease. By blocking DPP1, it prevents maturation of neutrophil serine proteases inside developing neutrophils, lowering the proteolytic load those cells release at sites of inflammation. That mechanism earned its first wins in bronchiectasis. In the WILLOW Phase 2 trial, brensocatib prolonged time to first exacerbation with adjusted hazard ratios of 0.58 at 10 mg and 0.62 at 25 mg versus placebo [4]. In the ASPEN Phase 3 trial of 1,721 patients, brensocatib lowered the annualized exacerbation rate to 1.02 at 10 mg and 1.04 at 25 mg, against 1.29 on placebo, with rate ratios of 0.79 and 0.81 [5]. The hidradenitis hypothesis extended that logic. HS lesions carry dense neutrophil infiltrates and draining tunnels full of proteases, so suppressing serine protease activation looked like a route to fewer abscesses and nodules. The genetic anchor for this target in HS was thin from the start. Open Targets scores the CTSC association with hidradenitis suppurativa at 0.075, near the floor of its scale [6]. Loss-of-function CTSC mutations cause Papillon-Lefevre syndrome, and the CEDAR protocol excluded those patients, which underlines that the human genetics of CTSC point to periodontal and palmoplantar disease rather than HS [1].

What actually happened

At Week 16, total AN count fell 45.5 percent in the brensocatib 10 mg arm and 40.3 percent in the 40 mg arm, against a 57.1 percent reduction on placebo [2]. Both active arms finished below placebo. The study missed its primary endpoint and every secondary efficacy endpoint at both doses [2]. Safety was unremarkable. Brensocatib was well tolerated with no new safety signals, including at the 40 mg dose, the highest Insmed has tested [2]. Insmed discontinued the HS program and kept the bronchiectasis franchise, where the drug is approved [2]. Pharmacovigilance data are consistent with a clean profile so far. OpenFDA FAERS holds 16 reports naming brensocatib, of which 8 were serious and 1 recorded a death, with top terms covering balance disorder, electrolyte and creatinine changes, and bradycardia [7]. Those counts are small and reflect early post-approval bronchiectasis use rather than the HS trial.

Failure mechanism, best guess

The most parsimonious read is translational mismatch. Brensocatib silences a proximal driver of neutrophilic airway disease, but hidradenitis suppurativa is not primarily a neutrophil serine protease disease. The treatments with controlled evidence in HS hit TNF (adalimumab) and IL-17 (secukinumab and bimekizumab), and current pathophysiology places IL-1 beta, TNF, IL-17, and interferon gamma at the center, with neutrophils acting as one downstream effector among many [8]. Blocking serine protease maturation leaves the upstream cytokine network intact. A second factor compounded the first. The placebo arm improved 57.1 percent in AN count, a large nonspecific response that narrows the detectable window for any added effect [2]. When the placebo bar sits that high and the target sits that far downstream, a real but modest pharmacodynamic effect cannot separate from noise. The weak CTSC association in HS was a quantitative signal that the target lacked human validation in this indication before the trial opened [6].

How to prevent this next time

Two quantitative tools would have reframed the decision. The first is an association-anchored prior. Setting the prior probability of success from the Open Targets CTSC association of 0.075 rather than from bronchiectasis efficacy would have lowered the expected value of an HS expansion before any patient enrolled [6]. The second is a Bayesian predictive probability of success computed at a preplanned interim, which integrates the posterior over the treatment effect with the probability of a positive final result:

PP(success)=∫θ​P(final success∣θ,interim data)⋅π(θ∣interim data)dθ

A futility interim built around this quantity, with a placebo response prior centered near the observed 57.1 percent, would have flagged the drug-below-placebo trend early and capped exposure [2]. Powering should have rested on the placebo-adjusted delta, not the within-arm reduction, because a 40 to 45 percent active response means little next to a 57 percent placebo response [2]. The single highest leverage change would have been to set the prior probability of success from the CTSC association of 0.075 and the TNF and IL-17 anchored biology of hidradenitis suppurativa, not from the bronchiectasis exacerbation data, before committing a 214-patient Phase 2b [6].

What this means for similar programs

DPP1 inhibition stays validated where neutrophil serine proteases are proximal, which means airway diseases such as bronchiectasis, not mixed-cytokine dermatoses. Programs reading brensocatib across to other neutrophil-rich skin conditions should treat the CEDAR result as a prior, not a footnote. The general rule is that a drug validated in one organ system needs its own target-to-disease anchor before label expansion, and an association score near zero is a reason to demand a biomarker-defined responder hypothesis first.

Open questions

The trial has no posted results record yet, so several questions stay open. Was there a neutrophil-high or tunnel-dominant HS subgroup that moved, even as the overall population did not? Did skin or blood neutrophil serine protease activity fall as expected, confirming target engagement while efficacy failed? What drove the 57.1 percent placebo response, and is AN count too noisy at 16 weeks for a Phase 2b readout? These answers decide whether DPP1 is dead in HS or only dead in unselected HS.

Sources

1. ClinicalTrials.gov. NCT06685835, CEDAR study record (enrollment 214, quadruple masking, AN-count primary endpoint, reason stopped "Lack of Efficacy"). https://clinicaltrials.gov/study/NCT06685835.

2. Insmed. Insmed Provides Clinical Update on Phase 2b CEDAR Study, 7 April 2026 (AN reduction 45.5 percent at 10 mg, 40.3 percent at 40 mg, 57.1 percent on placebo, endpoints not met, HS program discontinued). https://investor.insmed.com/2026-04-07-Insmed-Provides-Clinical-Update-on-Phase-2b-CEDAR-Study.

3. ChEMBL. Brensocatib, CHEMBL3900409, dipeptidyl peptidase I inhibitor. https://www.ebi.ac.uk/chembl/compound_report_card/CHEMBL3900409/.

4. Chalmers JD, et al. Phase 2 Trial of the DPP-1 Inhibitor Brensocatib in Bronchiectasis. N Engl J Med 2020, 383:2127-2137.

5. Chalmers JD, et al. Phase 3 Trial of the DPP-1 Inhibitor Brensocatib in Bronchiectasis. N Engl J Med 2025, 392:1569-1581.

6. Open Targets Platform. CTSC (ENSG00000109861) association with hidradenitis suppurativa, overall score 0.075. https://platform.opentargets.org/target/ENSG00000109861/associations.

7. OpenFDA FAERS. medicinalproduct brensocatib, 16 reports, 8 serious, 1 death. https://api.fda.gov/drug/event.json.

8. Sabat R, et al. Hidradenitis suppurativa. Lancet 2025, 405:420-438.