BE BOLD: Bimekizumab Superiority Over Risankizumab in PsA Joints

Bimekizumab, a dual IL-17A/F inhibitor, demonstrated superiority over risankizumab, an IL-23 inhibitor, on the primary joint endpoint of ACR50 at week 16 in patients with active psoriatic arthritis (PsA), according to late-breaking results from the phase 3b BE BOLD study presented at the European Alliance of Associations for Rheumatology (EULAR), held June 3-6 in London, United Kingdom — the first head-to-head trial to show superiority in a joint-focused outcome between 2 biologic disease-modifying antirheumatic drugs in PsA.^1,2

Iain B. McInnes, CBE, PhD, Muirhead Chair of Medicine and Versus Arthritis Professor of Rheumatology at the University of Glasgow, College of Medical, Veterinary and Life Sciences, Glasgow, United Kingdom, presented the data and discussed the findings in further detail with RheumatologyLive.

BE BOLD (NCT06624228) randomized 553 adults with active PsA per CASPAR criteria (tender joint count ≥3/68, swollen joint count ≥3/66) who were biologic DMARD-naïve or had inadequate response or intolerance to 1 prior TNF inhibitor 1:1 to subcutaneous bimekizumab or risankizumab. Bimekizumab dosing was based on baseline psoriasis severity per label: most patients received 160 mg every 4 weeks (Q4W); those with moderate to severe psoriasis (body surface area ≥10%, Investigator's Global Assessment ≥3, and PASI ≥12) received 320 mg Q4W through week 16 then Q8W. Risankizumab 150 mg was administered at baseline, week 4, and week 16 regardless of psoriasis severity.¹

Baseline characteristics were comparable between arms. Mean age was 51.1 (standard deviation [SD], 13.6) years for bimekizumab (n = 277) and 50.7 (SD, 12.7) for risankizumab (n = 276); mean tender joint count was 17.3 (SD, 12.7) and 16.6 (SD, 12.2), and mean swollen joint count was 10.1 (SD, 7.4) and 9.5 (SD, 6.8), respectively. Study completion rates through week 16 were high: 96.0% for bimekizumab and 93.1% for risankizumab.¹

For the primary endpoint of ACR50 at week 16, bimekizumab was superior to risankizumab (49.1% vs 38.4%; Δ10.7; P = .0078). The first ranked secondary endpoint, minimal disease activity (MDA) at week 16, was numerically higher with bimekizumab but did not reach statistical significance (43.0% vs 39.9%; P = .4408), halting the sequential testing procedure. McInnes noted that the trial was not powered to show a difference in MDA given the relatively low proportion of patients with moderate to severe psoriasis, and that detailed examination of MDA subcomponents in future analyses would be of interest. Subsequent secondary endpoints were therefore descriptive only: ACR50+PASI100 at week 16 in patients with baseline BSA ≥3% was achieved by 33.5% versus 24.4% (nominal P = .0800), and ACR50 at week 4 by 19.9% versus 7.2% (nominal P <.0001). Additional descriptive outcomes at week 16 included PASI100 in patients with baseline BSA ≥3% (53.4% vs 46.6%; nominal P = .2395) and DAPSA low disease activity or remission (65.3% vs 54.7%; nominal P = .0066).¹

Safety data were collected through a February 23, 2026 cut-off, at which point 83.4% of patients had completed double-blind treatment. Treatment-emergent adverse events (TEAEs) occurred in 57.0% of bimekizumab- and 52.0% of risankizumab-treated patients. Serious TEAEs were reported in 5 (1.8%) and 8 (2.9%) patients, respectively; severe TEAEs in 5 (1.8%) each. Candida infections were more frequent with bimekizumab, consistent with the known mechanism-of-action effect of IL-17 inhibition; all were mild or moderate, none were serious or systemic, and none led to study discontinuation. Rates of serious infection, inflammatory bowel disease, malignant tumors, neutropenia, and cardiovascular disorders were low across both treatment arms. 1 death occurred, attributed by the investigator to myocardial infarction in a patient with coronary artery disease, hypertension, and hyperlipidemia, and deemed unrelated to treatment. There were no cases of suicidal ideation or behavior, anaphylaxis, or active tuberculosis.¹

RheumatologyLive: What do the ACR50 differences between bimekizumab and risankizumab tell you at a mechanistic level about dual IL-17A/F inhibition versus IL-23 inhibition in the joint?

Iain McInnes, MD, PhD: It is well known that IL-23 sits upstream as a regulatory cytokine whereas the IL-17 family of cytokines are downstream effectors - in essence they serve to activate the immune and stromal cells in the joint and skin, whereas IL-23 sits upstream to orchestrate matters - as such we would expect different clinical responses (consider downstream effects that are independent of upstream IL-23 mediated regulation) and indeed rate of onset of response as we inhibit these discrete parts of the immune regulatory network - the data therefore make mechanistic sense.

How about the speed of joint response and is early response at Week 4 a clinically meaningful signal?

Early responses are appreciated by patients and clinicians alike, especially with troublesome signs of skin and musculoskeletal disease. We do however always pay close attention to robust responses in the medium term.

How do you interpret the MDA differences not reaching statistical significance (and the subsequent nominal secondary end points)?

The primary question of this study was to establish if the joints were differentially affected by drug treatment hence an ACR50 primary outcome - and so it proved in favour of the IL-17A/F inhibitor. However the number of patients with moderate to severe psoriasis was relatively low and as such we must recall that the trial study was not designed for, nor likely to show, a difference in MDA. Detailed examination of the sub components of MDA in future would be of interest.

What does it mean for the field — and the guidelines — that an IL-17A/F inhibitor has now shown superiority over an IL-23 inhibitor on a joint endpoint, when IL-23 inhibitors had been increasingly positioned as the preferred biologic in PsA because of their skin efficacy and favorable safety profile? 

I think the 2 MOAs are at least equivalent in skin and refer you to H2H studies and RWE database analyses in this regard - and we now have superiority for articular endpoints with an IL-17A/F inhibitor over an IL-23 inhibitor in a formal RCT. The safety issue seems to focus mainly on candida infections which are not a major issue in routine clinical practice and as such it is unclear whether safety will differentiate the MOAs in practice. The presence of IBD may however be a differentiator in favor of IL-23 inhibition. In due course, learned taskforces will determine the impact on guidelines or recommendations - H2H data are always helpful in the SLRs that inform such algorithms. I would not wish to prejudge their conclusions that said.

This transcript has been edited for clarity.

References

1. McInnes IB, Gossec L, Gottlieb AB, et al. Bimekizumab efficacy and safety versus risankizumab in patients with active psoriatic arthritis: 16-week results from a head-to-head, multicentre, randomised, phase 3b study (BE BOLD) [abstract LB0001]. Ann Rheum Dis. 2026. doi:10.1136/annrheumdis-2026-eular.LBA_B.10

2. Gossec L, et al. EULAR recommendations for the management of psoriatic arthritis with pharmacological therapies: 2023 update. Ann Rheum Dis. 2024;83:706–719.