Amlitelimab

Amlitelimab (SAR445229) is Sanofi's anti-OX40 ligand antibody in Phase 3 for moderate-to-severe atopic dermatitis, with regulatory submissions planned for H2 2026. It acts at a different node in the inflammatory circuit than Dupixent - targeting T cell costimulatory activation rather than downstream IL-4/IL-13 cytokine output. Three Phase 3 trials (COAST 1, COAST 2, SHORE) have all met primary endpoints, but the absolute efficacy numbers fall meaningfully below Dupixent's historical benchmarks - COAST 1 monotherapy EASI-75 placebo-adjusted deltas of 17-20 percentage points versus 32-36 for Dupixent's SOLO trials. Sanofi has guided peak sales potential exceeding €5 billion, but sell-side estimates were cut roughly in half after the COAST 1 readout [7]. The commercial positioning challenge is clear: durability, dosing convenience (Q12W versus Dupixent's Q2W), and efficacy in biologic-refractory patients will have to carry the value story, not raw skin clearance.

Amlitelimab has completed three Phase 3 studies in atopic dermatitis. COAST 1 (monotherapy, n=601, reported September 2025) met all primary and key secondary endpoints, with vIGA-AD 0/1 response rates of 21.1% (Q4W) and 22.5% (Q12W) versus 9.2% placebo, and EASI-75 of 35.9% (Q4W) and 39.1% (Q12W) versus 19.1% placebo (non-responder imputation) [1]. COAST 2 (monotherapy, n=589) and SHORE (combination with topical corticosteroids) both reported positive topline results at AAD in March 2026, with COAST 2 vIGA-AD 0/1 of 25.3% (Q4W) and 25.7% (Q12W) versus 14.8% placebo, and SHORE vIGA-AD 0/1 of 28.7% (Q4W) and 32.3% (Q12W) versus 16.8% placebo [2]. Two additional Phase 3 studies - AQUA (biologic-refractory patients, NCT06241118) and ESTUARY (long-term extension with Q12W maintenance dosing) - are expected to read out in H2 2026 [2]. Sanofi has stated global regulatory submissions are planned for H2 2026. No FDA breakthrough therapy, fast track, or other special designations have been disclosed. Beyond AD, amlitelimab is in Phase 2 for asthma (TIDE, completed - see Risks section for dose-response complications) [3], celiac disease (NCT06557772, active), hidradenitis suppurativa, systemic sclerosis, alopecia, and chronic rhinosinusitis with nasal polyps. Sanofi acquired amlitelimab through its $1.1 billion Kymab acquisition in 2021.

In atopic dermatitis, the immune system overreacts to environmental triggers, sending waves of inflammatory T cells into the skin. Most current biologics - Dupixent (dupilumab) chief among them - block IL-4 and IL-13, two signaling molecules that sit partway down this inflammatory cascade. Amlitelimab targets something further upstream: OX40 ligand (OX40L, encoded by the gene TNFSF4), a protein on the surface of antigen-presenting cells (the immune system's scouts) that delivers a "keep going" signal to T cells through the OX40 receptor. When OX40L binds OX40, it tells T cells to survive longer, multiply, and keep producing inflammatory signals [4]. Block that handshake, and you reduce the activation of both Th1 and Th2 T cells - not just the Type 2 inflammation that IL-4/IL-13 blockers address. Importantly, OX40/OX40L costimulation and IL-4/IL-13 signaling are parallel and interconnected pathways rather than a linear hierarchy - OX40 promotes Th2 survival and amplifies IL-4/IL-13 production while also driving Th1, Th17, and Treg biology. This broader immune modulation across multiple T cell subsets is the theoretical differentiator. Amlitelimab is a fully human, non-depleting antibody - it blocks the OX40L signal without killing the cells that express it. Phase 2 biomarker data showed that TARC, eosinophils, and IL-22 (markers of Type 2 inflammation) dropped during treatment and stayed suppressed even after the drug was withdrawn, when serum drug levels had fallen to negligible [5]. That off-treatment durability is biologically distinct from IL-4/IL-13 blockade, where disease typically rebounds once the drug clears. Whether this translates to a clinically meaningful difference in long-term disease control is the key question the Phase 3 program needs to answer.

The key Phase 3 program comprises three completed studies and two ongoing ones, all in patients aged 12+ with moderate-to-severe AD. COAST 1 (NCT05341336) enrolled 601 patients randomized to subcutaneous amlitelimab Q4W (every 4 weeks), amlitelimab Q12W (every 12 weeks, following a Q4W loading phase), or placebo as monotherapy [6]. The co-primary endpoints were vIGA-AD 0/1 (clear or almost clear skin) and EASI-75 (75% improvement in eczema severity) at Week 24. COAST 2 (NCT06181435, n=589) used an identical design. SHORE tested amlitelimab on top of background topical corticosteroids [2]. Two additional studies are running: AQUA (NCT06241118, n=390) specifically enrolls biologic-refractory or JAK inhibitor-refractory patients, reading out the harder-to-treat population, and the long-term extension ESTUARY. The Q12W arm is the commercial play - four injections per year versus dupilumab's every-two-week dosing. The trials are placebo-controlled, not active-comparator, which means cross-trial comparisons to Dupixent are indirect. Sanofi chose not to run a head-to-head study, which is telling. The AQUA trial is the most commercially interesting: if amlitelimab works in biologic failures, it has a clear market niche regardless of how it stacks up against Dupixent in first-line use.

Our model puts this drug's chances of eventual approval at 42%. It starts from the historical approval rate for Phase 3 drugs in this area - about 61% - then adjusts based on ten specific facts about the trial and sponsor. The estimate goes up because of the trial's non-randomized design and open-label blinding, and comes back down because enrollment is smaller than typical for this phase and earlier-phase results were weak. The remaining factors are close to average, so they don't move the number much from there.

The primary risk is commercial, not regulatory. Amlitelimab will almost certainly get approved - three positive Phase 3 trials with consistent safety is a strong package. But "approved" and "successful" are different things in a market where Dupixent generated €15.7 billion in 2025 revenue [8]. The efficacy gap is real. Analysts characterized the COAST 1 data as falling well below Dupixent's performance and below investor expectations, with Leerink cutting amlitelimab sales projections roughly in half [7]. Sanofi has guided peak sales potential exceeding €5 billion, but post-COAST 1 sell-side estimates are significantly lower. If amlitelimab can't match Dupixent on skin clearance, it needs to win on something else: dosing convenience (Q12W vs Q2W), durability (off-treatment response maintenance), or positioning in biologic-refractory patients. Each carries risk. The Q12W dosing story is strong but the ESTUARY extension data isn't in yet. The off-treatment durability from Phase 2 is intriguing but unconfirmed in larger trials. And the biologic-refractory study (AQUA) hasn't read out. The asthma indication faces a more complicated picture than initially expected. The TIDE Phase 2 showed a non-monotonic dose-response: the medium dose (125 mg) reduced annualized asthma exacerbation rate by 47.8% versus placebo (p=0.008), but the highest dose (250 mg) showed only a 24.5% reduction (p=0.22) [3]. The 250 mg dose did show FEV1 and ACQ-5 improvements (p<0.05), suggesting possible dissociation between lung function and exacerbation endpoints. This inverted dose-response raises questions about therapeutic window and complicates Phase 3 dose selection - Sanofi will likely advance the medium dose, but a non-monotonic dose-response in a class where multiple biologics already have strong evidence is a risk flag. The competitive landscape shifted materially in March 2026 when Kyowa Kirin discontinued rocatinlimab (the only other OX40-class drug in late-stage development) after identifying emerging malignancy concerns including Kaposi's sarcoma cases in clinical trials [9][10]. Amgen had already returned rocatinlimab rights to Kyowa Kirin in January 2026. On one hand, this eliminates the nearest class competitor; on the other, a malignancy signal in a closely related mechanism (OX40 receptor vs. OX40 ligand) raises broader class-level questions about long-term T cell costimulation blockade. Amlitelimab's safety database across >2,000 patients shows malignancy rates <1% and comparable to placebo, and its non-depleting mechanism is mechanistically distinct from rocatinlimab's approach - but the theoretical concern with reduced immune surveillance will require years of post-marketing data to fully resolve. The Dupixent franchise's patent cliff provides important context for timing. Key dupilumab compound patents are expected to expire around 2031, with biosimilar entry anticipated in the early 2030s (Formycon's FYB208 biosimilar candidate has completed preclinical development) [13]. This gives amlitelimab a window of 4-5 years as a differentiated branded biologic before the market dynamics shift fundamentally with dupilumab biosimilar entry.

Worth watching, with caveats. Amlitelimab will likely get approved in 2027 and carve out a position in the AD market, but it's not going to be a Dupixent-killer - and Sanofi knows this, which is why they're developing it as a complement rather than a replacement. Sanofi has guided peak sales potential exceeding €5 billion, but post-COAST 1 analyst revisions suggest the realistic range is likely €2-3 billion at peak, contingent on the AQUA and ESTUARY readouts [7]. The real signal to watch is the AQUA readout (H2 2026): if amlitelimab delivers meaningful efficacy in patients who've already failed dupilumab or JAK inhibitors, it has a defensible commercial niche. If the biologic-refractory data is soft, amlitelimab gets relegated to a convenience play (Q12W dosing) that will face price and access headwinds against a Dupixent franchise with massive payer infrastructure. The off-treatment durability story from Phase 2 biomarker data is the most scientifically distinctive feature of this drug - if Phase 3 extension data confirms it, that changes the treatment paradigm from chronic suppression to potential disease modification. The rocatinlimab discontinuation in March 2026 is a double-edged sword: it removes the nearest class competitor but raises class-level safety questions that will shadow amlitelimab's regulatory review [10]. Sanofi's internal framing of amlitelimab as part of its immunology portfolio succession is driven by the Dupixent patent cliff - key compound patents expire around 2031, with biosimilar entry expected in the early 2030s [8][13]. Check back at AQUA and ESTUARY readouts in H2 2026, and watch for Sanofi's pricing strategy at the regulatory submission.