The Role of OX40 Pathway Inhibition as a New Therapeutic Strategy for Atopic Dermatitis

Abstract

Atopic dermatitis is a chronic inflammatory skin disease affecting approximately 15–20% of children and 2–10% of adults worldwide. Epidermal barrier dysfunction and immune dysregulation are central to its pathogenesis, creating a self-perpetuating cycle in which barrier disruption exacerbates inflammation, which in turn further impairs skin barrier integrity. The OX40/OX40L axis, involving the co-stimulatory receptor OX40 expressed on T cells and its ligand OX40L on antigen-presenting cells, plays a critical role in sustaining T cell-driven inflammatory responses in AD. Despite recent therapeutic advances, many patients remain inadequately controlled, and key unmet needs persist. Inhibitors of the OX40/OX40L pathway represent a novel therapeutic approach by modulating multiple effector and memory T-cell subsets implicated in disease pathogenesis. Amlitelimab, an anti-OX40L monoclonal antibody, has demonstrated sustained efficacy and a favorable safety profile in phase IIa and IIb trials. Rocatinlimab, targeting OX40, has also shown promising results in a phase IIb study and has progressed into multiple phase III trials, with supportive top-line data. In contrast, telazorlimab has shown more modest efficacy and has not advanced to later-stage development. Next-generation agents, including IMG-007, STAR-0310, APG990, and APG279, have been engineered with extended half-lives and attenuated antibody-dependent cellular cytotoxicity to support longer dosing intervals and improve tolerability. While these findings are encouraging, direct comparative studies among agents and versus established therapies are lacking, and long-term efficacy and safety data are still needed. This narrative review explores the role of the OX40/OX40L axis in atopic dermatitis pathogenesis and critically evaluates emerging therapies targeting this pathway, aiming to inform their future integration into clinical practice.

Key Points

Targeting the OX40/OX40L signaling pathway offers a novel immunomodulatory approach for treating atopic dermatitis, distinct from therapies currently in clinical use.

Biologic agents such as amlitelimab and rocatinlimab have shown encouraging, long-lasting clinical effects and good safety profiles in mid-stage trials, with multiple phase III trials underway.

Next-generation biologics such as IMG-007, STAR-0310, and APG990 have been bioengineered for longer half-lives and reduced cytotoxicity, supporting extended dosing intervals and improved tolerability.

Introduction

Atopic dermatitis (AD) is a chronic, relapsing inflammatory skin disease that affects approximately 15–20% of children and 2–10% of adults worldwide [1, 2]. It typically begins in childhood and may resolve spontaneously; however, up to 40% of patients continue to experience intermittent flares throughout life [3]. A positive family history of atopy is present in 70% of individuals with AD and is the primary risk factor, increasing susceptibility to other atopic conditions such as food allergies, allergic rhinitis, and asthma [4, 5]. Clinically, AD presents with erythematous, exudative, or lichenified plaques, often accompanied by xerosis and intense pruritus. These lesions are triggered by endogenous and environmental factors in genetically predisposed individuals [6].

The chronic and fluctuating nature of AD, along with its association with skin infections, sleep disturbances, and mood disorders such as anxiety and depression, significantly impair quality of life [7]. The pathophysiology of AD involves a combination of epidermal barrier dysfunction and immune dysregulation. Acute lesions are predominantly characterized by type 2 and type 22 inflammation, whereas chronic lesions are associated with type 1 and type 17 inflammatory responses [8].

Management of AD begins with trigger identification and avoidance, coupled with daily moisturization to preserve skin barrier function. Topical corticosteroids and calcineurin inhibitors serve as first-line therapies for acute flare-ups. In refractory cases, systemic options such as corticosteroids, phototherapy, conventional immunosuppressants, and targeted biologics and non-biologic agents modulating immune pathways may be considered [5].

Advances in understanding the pathophysiology of AD have led to the development of targeted therapies that act on specific cytokines, their receptors, or intracellular signaling pathways. Among these, Janus kinase inhibitors, as well as interleukin (IL)-4, IL-13, and IL-31 inhibitors, are now widely used in patients who do not respond adequately to conventional treatments [9, 10]. Although these agents are generally more effective than traditional therapies, many patients still fail to achieve satisfactory clinical outcomes. In addition, their therapeutic effects may be short-lived in some patients, with symptoms recurring within days to weeks after discontinuation. These therapies are also associated with treatment-emergent adverse events (TEAEs), including infections, paradoxical psoriasis, and arthralgia among others [11–14].

Given these limitations, there is a pressing need for treatments with novel mechanisms of action. The OX40/OX40L signaling pathway plays a key role in both acute and chronic inflammation in AD and has emerged as a promising therapeutic target [15]. This narrative review aims to explore the therapeutic potential of targeting the OX40/OX40L pathway by outlining the current evidence with a focus on their safety, clinical efficacy, and potential role in the treatment of AD.

The Role of the OX40/OX40L Pathway in the Pathophysiology of Atopic Dermatitis

The pathophysiology of AD involves a complex interplay between epidermal barrier dysfunction, immune dysregulation, and microbial imbalance [16, 17]. Impairments in intercellular lipids, tight junction integrity, and key structural proteins, particularly filaggrin (due to gene mutations), contribute to skin barrier dysfunction and increased permeability. This leads to elevated transepidermal water loss and facilitates the entry of environmental allergens and pathogens, which contribute to the formation of eczematous lesions and recurrent infections, further disrupting barrier integrity [17].

Epidermal damage and allergen exposure stimulate the release of epithelial-derived signaling molecules such as IL-25, IL-33, and thymic stromal lymphopoietin. These molecules activate antigen-presenting cells (APCs), which in turn promote type 2 immune polarization through interaction with naïve T cells. This cascade leads to T helper (Th-)2 cell activation and secretion of AD-associated cytokines, including IL-4, IL-5, IL-13, and IL-31, which drive the disease process and facilitate immunoglobulin E class switching. In parallel, these cytokines stimulate group 2 innate lymphoid cells, which independently amplify type 2 inflammation by producing IL-5 and IL-13 [8, 17]. Although Th2 responses dominate the early and acute phases of AD, other Th subsets, including Th1, Th17, and Th22, also contribute to the chronic inflammatory state through their respective cytokine signatures [17]. The OX40/OX40L pathway plays a central role in the immunopathogenesis of AD. OX40 (also known as TNFRSF4 or CD134) is a co-stimulatory receptor primarily expressed on activated T cells, while its ligand, OX40L (TNFSF4, CD252), is found on APCs [18]. During immune activation, both T cells and APCs upregulate various immune checkpoint molecules that modulate T-cell responses. These checkpoints are broadly categorized as co-stimulatory or coinhibitory, either enhancing or dampening immune activation (Fig. 1).

Fig. 1.

Overview of the OX40/OX40L co-stimulatory pathway in atopic dermatitis and the corresponding sites of action of emerging targeted therapies. Binding of OX40L on antigen-presenting cells to OX40 on T cells provides a secondary co-stimulatory signal that promotes effector and memory T-cell activation. Anti-OX40L agents (e.g., amlitelimab, APG990) inhibit ligand binding on the antigen-presenting cell, whereas anti-OX40 agents (e.g., rocatinlimab, telazorlimab, STAR-0310, IMG-007) block receptor-mediated signaling on the T cell. MHC major histocompatibility complex, TCR T-cell receptor

OX40 and OX40L belong to the tumor necrosis factor superfamily. As costimulatory molecules, they contribute to T-cell proliferation, survival, and cytokine production, thereby amplifying type 2 inflammatory responses in AD [8, 19].

The OX40/OX40L pathway is initiated when APCs engage with naïve T cells, promoting their differentiation into effector T cells. Following this interaction, OX40L is upregulated on APCs within 24 h, while OX40 expression on activated T cells emerges 1–5 days later [18, 20]. The subsequent binding of OX40 to OX40L enhances effector T-cell activity by stimulating proliferation and prolonging cell survival. These effects are mediated through activation of intracellular signaling pathways, including nuclear factor-κB1 and PI3K-PKB/Akt, which support cell division and increase the production of anti-apoptotic proteins [26].

In addition to sustaining effector T-cell responses, the OX40/OX40L axis facilitates their transition into quiescent memory T cells. Upon re-exposure to the same antigens, memory T cells rapidly re-express OX40, enabling renewed interaction with OX40L-expressing APCs and reactivation of the immune response. This mechanism contributes to the expansion and persistence of Th cell subsets implicated in AD, thereby sustaining and amplifying chronic inflammation [18].

Targeting the OX40/OX40L Pathway to Treat Atopic Dermatitis

Given the upstream role of the OX40/OX40L pathway in T-cell activation, disrupting this interaction attenuates T-cell-mediated immune responses across multiple Th subsets, including Th1, Th2, Th17, and Th22 cells. This mechanistic insight has driven the development of targeted therapies aimed at inhibiting the OX40/OX40L pathway, primarily through monoclonal antibodies directed against either protein.

Currently, three anti-OX40 monoclonal antibodies [rocatinlimab, IMG-007 and STAR0310 (derived from telazorlimab)] and one anti-OX40L antibody (amlitelimab, APG990) are under clinical investigation for the treatment of moderate-to-severe AD (Fig. 1). Preliminary results from these trials have demonstrated encouraging efficacy and safety outcomes [8, 21, 22].

Amlitelimab

Amlitelimab (KY1005; SAR445229) is a non-depleting, human IgG4 monoclonal antibody that targets OX40L expressed on APCs, blocking its interaction with OX40 on T cells [18, 20]. A phase I, randomized, double-blind, placebo-controlled study (NCT03161288) evaluated the safety, tolerability, immunogenicity, and pharmacokinetics of amlitelimab in 64 healthy participants. Subjects were divided into eight cohorts; each included six individuals receiving intravenous (IV) amlitelimab and two receiving placebo. Single doses (0.006–0.05 mg/kg) were administered in the first three cohorts, while the remaining five received multiple doses (0.15–12 mg/kg), followed by two maintenance doses at weeks 4 and 8 at half the initial dose. No serious TEAEs were reported, and all adverse events (AEs) were mild and self-limiting, with headache being the most common [8, 23].

A phase IIa, multicenter, randomized, double-blind, parallel-group trial (NCT03754309) evaluated the efficacy, safety, and tolerability of IV amlitelimab in 88 adults with moderate-to-severe AD. Participants were randomized 1:1:1 to receive low dose (LD; 200 mg loading, then 100 mg once every 4 weeks [Q4W]), high dose (HD; 500 mg loading, then 250 mg Q4W), or placebo at weeks 0, 4, 8, and 12.

Primary endpoints included the percentage change in the Eczema Area and Severity Index (EASI) and the incidence of TEAEs at week 16. Mean EASI reductions from baseline were 80.1% (LD), −70% (HD), and −49.4% (placebo), with statistical significance observed for the LD group (p = 0.009). The EASI-75 response rates were 59% (LD), 52% (HD), and 25% (placebo). Additionally, 44% (LD) and 37% (HD) of participants achieved a validated Investigator’s Global Assessment (vIGA) score of 0/1 compared with 8% with placebo (p < 0.001 for both).

Notably, 68% of participants who achieved a vIGA score of 0/1 at week 16 maintained this response through week 36, despite declining serum drug concentrations below pharmacological activity. Amlitelimab was well tolerated, with all TEAEs reported as mild or moderate. Common TEAEs included headache, hyperhidrosis, pyrexia, upper respiratory tract infection, elevated aspartate aminotransferase, and iron deficiency anemia [8, 18, 24].

A phase IIb, randomized, double-blind, placebo-controlled trial (NCT05131477) evaluated the efficacy and safety of subcutaneous (SC) amlitelimab in 390 adults with moderate-to-severe AD over a 52-week period. In Part 1 (weeks 0–24), participants were randomized (1:1:1:1:1) to receive placebo or amlitelimab Q4W in one of four dosing regimens: (1) 250 mg with a 500-mg loading dose; (2) 250 mg; (3) 125 mg; or (4) 62.5 mg. The final dose was given at week 20. At week 16 (primary endpoint), all amlitelimab groups showed a significant improvement in EASI compared with placebo (−29.4%). Least-squares mean EASI differences versus placebo were: −32.1% (250 mg + loading, p < 0.0001), −27.3% (250 mg, p < 0.0001), −22.2% (125 mg, p = 0.0002), and −30.2% (62.5 mg, p < 0.0001). Secondary endpoints at weeks 16 and 24 showed higher rates of EASI-75 and vIGA 0/1 among all treatment groups compared with placebo. In Part 2 (weeks 24–52), clinical responders — defined as participants who attained a vIGA score of 0 or 1 and/or EASI-75 — were re-randomized (3:1) to continue or discontinue amlitelimab. At week 52, 71.9% (vIGA 0/1) and 69% (EASI-75) of those continuing treatment maintained a clinical response. Among those who discontinued, 57% and 61.6% sustained vIGA 0/1 and EASI-75, respectively, 28 weeks after the last dose.

Treatment was associated with reductions in AD-related serum biomarkers, including total immunoglobulin E, IL-13, IL-17A, IL-22, and IL-31, through week 52 in both continuers and discontinuers. Amlitelimab maintained a favorable safety profile with no life-threatening AEs reported. Common events included nasopharyngitis, upper respiratory tract infection, coronavirus disease 2019, and headache [25]. Currently, eight clinical trials involving amlitelimab are ongoing as part of the OCEANA Clinical Program, as detailed in Table 1.

Table 1.

Summary of completed clinical trials investigating emerging therapies for atopic dermatitis

Study	Design	Endpoints	Results	Safety
Amlitelimab
Phase I (NCT03161288) (23)	Single-center, double-blind, randomized, placebo-controlled trial (N = 64) Participants were randomized 6:2 to amlitelimab or placebo across 8 cohorts	Primary: Safety and tolerability of amlitelimab	Amlitelimab was well tolerated, with no serious AEs reported	Mild headache was the most common. All events were self resolving and of mild-to-moderate severity
Phase IIa (NCT03754309) (24)	Multicenter, double-blind, randomized (1:1:1), placebo-controlled, parallel-group trial (N = 83) LD: amlitelimab 200-mg loading dose + 100 mg Q4WHD: amlitelimab 500-mg loading dose + 250 mg Q4W Placebo	Primary: Percentage change in EASI from baseline to week 16 Incidence of TEAEs Secondary: EASI-50, EASI-75, and EASI-90 response rates. Percentage of patients with a vIGA score of 0/1 at week 16 Change in SCORAD Index, affected BSA, DLQI, and NRS for pruritus	LS mean percentage change in EASI from baseline to week 16: LD: −80.1%; HD: −70.0%; Placebo: −49.4% EASI-75 at week 16: LD: 59%; HD: 52%; Placebo: 25% Percentage of patients with a vIGA score of 0/1 at week 16: LD: 44%; HD: 37% Placebo: 8%	Most frequent TEAEs: headache, upper respiratory tract infection, hyperhidrosis, pyrexia, increased aspartate aminotransferase, and iron deficiency anemia
Phase IIb (NCT05131477) (25)	Multicenter, two part, double-blind, randomized, placebo-controlled trial (N = 390) Part 1 (24 weeks): randomization (1:1:1:1:1). Amlitelimab SC 250 mg Q4W, 125 mg Q4W or 62.5 mg Q4W without a loading dose or 250 mg Q4W with a 500-mg loading dose, and placebo Part 2 (28 weeks): rerandomization 3:1 (clinical responders) to continue their dosing regimen before week 24 or withdrawn from amlitelimab	Primary: percentage change in EASI from baseline to week 16 Secondary: Part 1: percentage change in EASI from baseline to week 24: EASI-75 response at weeks 16 and 24 Percentage of patients with a vIGA score of 0/1 at weeks 16 and 24 Part 2: vIGA score of 0/1 and EASI-75 responses at week 52 in clinical responders of part 1	Difference from placebo in LS mean percentage change in EASI from baseline to week 16: 250 mg with 500 mg of loading dose: −32.1% (p < 0.0001) 250 mg: −27.3% (p < 0.0001) 125 mg: −22.2% (p = 0.0002) 62.5 mg: − 30.2% (p < 0.0001). Higher percentage of patients achieving a vIGA score of 0/1 at week 16 and week 24 in amlitelimab groups vs placebo Higher percentage of patients achieving EASI-75 at week 16 and week 24 in amlitelimab groups vs placebo Similar percentage of clinical responders achieving a vIGA score of 0/1 and EASI-75 at week 52 in pooled amlitelimab groups vs withdrawn group (71.9% and 69%) vs (57% vs 61.6%)	Most frequent TEAEs: nasopharyngitis, COVID-19, and headache
Rocatinlimab
Phase I (NCT03096223) (27)	Single-center, open-label, repeated-dose study (N = 22) 10 mg/kg IV Q2W, during 6 weeks (total of 3 infusions) + 16 weeks of follow-up	Primary: Safety and tolerability of rocatinlimab (incidence of TEAEs up to week 22) Secondary: Clinical efficacy and pharmacodynamics of rocatinlimab	Good safety and tolerability profile EASI change from baseline in percentage (mean ± SD): - 74.12 ± 20.53% at week 22 Percentage of patients achieving vIGA 0/1: 35% at week 22	Mild or moderate TEAEs, mostly due to infusion reactions: pyrexia, chills, and malaise
Phase IIb (NCT03703102) (28)	Multicenter, randomized, double-blind, parallel-group, placebo-controlled study (N = 274)	Primary: Percentage change in EASI from baseline to week 16 Secondary:	LS mean change in EASI from baseline to week 16: 150 mg Q4W: −48.3% 600 mg Q4W: −49.7% 300 mg Q2W: −61.1% 600 mg Q2W: −57.4% Placebo: −15.0% (all p < 0.001)	Most frequent TEAEs: pyrexia, chills, headache, aphthous ulcer, and nausea Serious AEs occurred in 2– 6% of rocatinlimab
	Randomized (1:1:1:1:1) to: SC rocatinlimab 150 mg Q4W SC rocatinlimab 600 mg Q4W SC rocatinlimab 300 mg Q2W SC rocatinlimab 600 mg Q2W Placebo	Achievement of EASI-50, EASI-75, and EASI-90 at week 16 Achievement of a rIGA/vIGA score of 0 or 1 with a ≥2-point reduction from baseline at week 16 Percentage change from baseline in SCORAD: Change from baseline in DLQI Percentage change from baseline in pruritus NRS score Percentage change from baseline in sleep disturbance NRS score	Rocatinlimab outperformed placebo in all secondary outcomes, with the 300-mg Q2W dose showing the greatest results at week 16 Among rocatinlimab-treated patients who achieved EASI-75 at week 36, 73–96% remained relapse free at week 56 after treatment discontinuation Decreased serum levels of TARC, IgE, and IL-22 in the rocatinlimab groups throughout the study	patients and 1.8% of the placebo group
Phase III HORIZON (NCT05651711) (29)	Randomized, double-blind, placebo-controlled, Parallel assignment study (N = 726) Rocatinlimab arm: dose of rocatinlimab SC Q4W + loading dose at week 2 Placebo arm: dose of placebo SC Q4W + loading dose at week 2	Primary: Achievement of a vIGA/rIGA score of 0 or 1 with a ≥2-point reduction from baseline at week 24; achievement of EASI-75 at week 24	EASI-75 at week 24: rocatinlimab group: 32.8% vs placebo: 13.7% (19.1% difference, p < 0.001) vIGA 0/1 in 19.3% of the rocatinlimab group vs 6.6% in the placebo group (12.8% difference, p < 0.001) rIGA 0/1 in 16.4% of the rocatinlimab group vs.4.9% in the placebo group (11.5% difference, p < 0.001)	TEAEs were more common in the rocatinlimab group Most frequent: pyrexia, nasopharyngitis, chills, headache, aphthous ulcer, and nausea
Phase III IGNITE (NCT05398445) (29)	Randomized, double-blind, placebo-controlled study (N = 769) Arm 1: HD of rocatinlimab SC Q4W + loading dose at week 2 Arm 2: LD of rocatinlimab SC Q4W + loading dose at week 2 Placebo SC Q4W + loading dose at week 2	Primary: Achievement of vIGA score of 0/1 with ≥2-point reduction from baseline at week 24 Achievement of EASI-75 at week 24	rIGA score of 0/1 in 22.7% of the HD group and 16.3% of the LD group (14.4% and 8% difference from placebo, respectively, p ≤ 0.01) EASI-75 at week 24: HD group: 42.3% LD group: 36.3% (29.5% and 23.4% difference from placebo, respectively, p < 0.001)	Most frequent TEAEs: pyrexia, chills, and headache
Phase III SHUTTLE (NCT05724199) (30–31)	Randomized, double-blind, placebo-controlled study (N = 746).	Primary: Achievement of rIGA/vIGA score of 0 or 1 with a ≥2-point reduction from baseline at week 24	vIGA score of 0/1 in 26.1% of the HD group and 25.8% of the LD group rIGA score of 0/1 in 23.3% of the HD group and 22.7% of the LD group	$\mathrm{\varnothing }$
	Rocatinlimab HD Q4W + TCS/TCI + loading dose at week 2 Rocatinlimab LD Q4W + TCS/TCI + loading dose at week 2 Placebo Q4W + TCS/TCI + loading dose at week 2	Achievement of EASI-75 at week 24	EASI-75 at week 24: HD group: 52.3% LD group: 54.1% (28.7% and 30.4% difference from placebo, respectively, p < 0.001)
Phase III VOYAGER (NCT05899816) (32)	Randomized, double-blind, placebo-controlled study (N = 221) Rocatinlimab Q4W for 24 weeks + loading dose at week 2 Placebo Q4W for 24 weeks + loading dose at week 2	Primary: Number of participants with a positive anti-tetanus response from week 20 to week 24 Number of participants with a positive anti-meningococcal response from week 20 to week 24	Rocatinlimab does not interfere with responses to tetanus and meningococcal vaccinations	$\mathrm{\varnothing }$
Telazorlimab
Phase IIa (NCT02683928) (34)	Double-blind, randomized, placebo-controlled, exploratory study (N = 62) 10 mg/kg IV infusions of GBR 830 or placebo on day 1 and day 29	Primary: Incidence of TEAEs and changes from baseline in epidermal hyperplasia and active AD mRNA expression signature Secondary: Percentage improvement from baseline in IGA and EASI scores EASI-50 and EASI-75 responses Achievement of IGA score of 0 or 1 Pruritus NRS and DLQI changes from baseline	Significant reductions in epidermal thickness from baseline on day 29 (p < 0.01) and day 71 (p < 0.001) in the GBR 830 group with no significant changes in the placebo group Significant reduction in mRNA expression of biomarkers associated with Th1, Th2, Th17, and Th22 subtypes On day 71, a higher percentage of patients in the GBR 830 group achieved EASI-50 (76.9%) compared with placebo (37.5%)	Most frequent TEAEs: headache, exacerbation of AD, nasopharyngitis, and upper respiratory tract infection
Phase IIb (NCT03568162) (36)	Double-blind, randomized, placebo-controlled, parallel-group, two-part study Part 1: 313 subjects randomized (1:1:1:1):	Primary: Percentage change from baseline in EASI, at week 16 Secondary: Proportion of subjects achieving EASI-75 and EASI-50 IGA score of 0 or 1 with a ≥2-point	LS mean percentage change in EASI from baseline to week 16: 300 mg Q2W (part 1): −54.4% vs placebo −34.2% (p = 0.008) 600 mg Q2W (part 2): −59% vs. Placebo −41.8% (p = 0.008) Achievement of EASI-75, at week 16, in the SC telazorlimab 300-mg Q2W group vs placebo: 23.7% vs. 11.3%	The frequency of TEAEs was similar between telazorlimab and placebo arms Most frequent TEAEs: exacerbation of AD, nasopharyngitis, upper respiratory tract
	Loading dose of 150 mg and 75 mg Q4W of telazorlimab Loading dose of 600 mg and 300 mg Q4W of telazorlimab Loading dose of 600 mg and 300 mg Q2W of telazorlimab Placebo Part 2: 149 subjects randomized 1:1: Loading dose of 1200 mg and 600 mg Q2W of telazorlimab Placebo	reduction from baseline Pruritus NRS score improvement ≥4	Achievement of EASI-75, at Week 16, in 600 mg Q2W of telazorlimab SC group vs. placebo: 25.3% vs. 18.9%.	infection, and headache
Phase Ia (NCT06304740) (21,38)	Double-blind, randomized, placebo-controlled study (N = 16) Three sequential dose cohorts, each receiving escalating doses of SC IMG-007 or placebo	Primary: Safety and tolerability of IMG-007 (incidence of TEAEs) Secondary: PK of IMG-007	Good safety and tolerability profile A single SC dose of IMG-007 showed a mean terminal half-life of 34.7 days (longer than other anti-OX40/OX40L monoclonal antibodies)	Most frequent TEAEs: injection-site reactions TEAEs were more frequent in the placebo arm (75%) than in the IMG-007 arm (25%)
Phase IIa (NCT05984784) (21,38)	Open-label study (N = 13) Patients received 300 mg of IV IMG007 at baseline, week 2, and week 4, followed by a 24-week follow-up period	Primary: Safety and tolerability of IMG-007 (incidence of TEAEs) Secondary: Percentage change from baseline in EASI over time PK of IMG-007 placebo-controlled, parallel-group, three-arm study. 420 participants (estimated) 0/1 with ≥2-point reduction in IGA, at week 24	Mean reduction in EASI of 77% from baseline to week 16 and EASI-75 response of 54% at week 16 Inhibition of serum markers of Th1, Th2, and Th17 cells was sustained during the 24-week follow-up period	No serious adverse events or treatment-related AEs

AD atopic dermatitis, AEs adverse events, BSA body surface area, COVID-19 coronavirus disease 2019, DLQI Dermatology Life Quality Index, EASI Eczema Area and Severity Index, HD high dose, IgE immunoglobulin E, IL interleukin, IV intravenous, LD low dose, LS least-squares, mRNA messenger RNA, NRS Numerical Rating, Scale, PK pharmacokinetics, Q2W once every 2 weeks, Q4W once every 4 weeks, rIGA revised Investigator Global Assessment, SC subcutaneous, SCORAD SCORing of Atopic Dermatitis, TARC thymus and activation-regulated chemokine, TCI topical calcineurin inhibitors, TCS topical corticosteroids, TEAEs treatment-emergent adverse events, Th T helper cell, vIGA validated Investigator’s Global Assessment

Rocatinlimab

Rocatinlimab (KHK4083/AMG 451) is a non-fucosylated human IgG1 monoclonal antibody that targets OX40 on activated T cells, inhibiting their clonal expansion and suppressing inflammation driven by Th1, Th2, Th17, and Th22 cells [18, 26]. A phase I, open-label, single-center study (NCT03096223) assessed the safety and tolerability of IV rocatinlimab in 22 adults with moderate-to-severe AD. Participants received 10 mg/kg every 2 weeks (Q2W) for 6 weeks (three total infusions), followed by a 16-week observation period.

Treatment-emergent adverse events occurred in 77.3% of participants, all mild to moderate, primarily related to infusion reactions (e.g., pyrexia, chills, malaise). An exploratory efficacy analysis showed a mean EASI reduction of 74.1% ± 20.5 from baseline and a vIGA score of 0/1 in 35% of patients at week 22. Rocatinlimab demonstrated a favorable safety profile with sustained clinical responses up to 16 weeks post-treatment [27].

Subsequently, a phase IIb, multicenter, randomized, double-blind, placebo-controlled trial (NCT03703102) evaluated the efficacy and safety of SC rocatinlimab in 274 adults with moderate-to-severe AD. Participants were randomized (1:1:1:1:1) to receive rocatinlimab at 150 mg Q4W, 600 mg Q4W, 300 mg Q2W, 600 mg Q2W, or placebo, administered through week 16. Active treatment continued through week 36, followed by a 20-week treatment-free follow-up to week 56.

At week 16 (primary endpoint), all treatment arms showed significantly greater least-squares mean percentage reductions in EASI compared with placebo (−15%). Reductions were: −48.3% (150 mg Q4W), −49.7% (600 mg Q4W), −61.1% (300 mg Q2W), and −57.4% (600 mg Q2W) [all p < 0.001]). Secondary endpoints also favored all active treatment groups.

Clinical responses were maintained through week 36, particularly in the 300-mg Q2W group. Among participants achieving EASI-75 at week 36, 73–96% maintained their response without relapse through week 56. Serum levels of key biomarkers (thymus and activation regulated chemokine, immunoglobulin E, and IL-22) declined significantly by week 16 and remained suppressed throughout the follow-up.

Rocatinlimab was generally well tolerated. The most common TEAEs included pyrexia, nasopharyngitis, chills, headache, aphthous ulcers, and nausea. Serious AEs occurred in 2–6% of rocatinlimab-treated patients versus 1.8% in the placebo group. No serious hypersensitivity reactions or deaths were reported [28].

ROCKET is a phase III clinical program comprising eight trials designed to evaluate the safety, efficacy, durability, and optimal dosing of rocatinlimab in patients with moderate-to-severe AD. Top-line results from four trials have been released, including the recent publication of the 24-week results from the ROCKET-HORIZON ROCKET-IGNITE trial [29].

ROCKET-HORIZON (NCT05651711), a randomized, double-blind, placebo-controlled trial, enrolled 726 participants who received SC rocatinlimab Q4W for 24 weeks (with a loading dose at week 2) or placebo on the same schedule. Primary endpoints included the proportion of patients achieving EASI-75 and a vIGA score of 0/1 with a ≥2-point reduction from baseline at week 24.

At week 24, EASI-75 was achieved by 32.8% of patients in the rocatinlimab group versus 13.7% in the placebo group (p < 0.001). Similarly, 19.3% of treated patients achieved a vIGA score of 0/1 with a ≥2-point improvement, compared with 6.6% with placebo (p < 0.001). Rocatinlimab also outperformed placebo in the stricter revised Investigator’s Global Assessment (rIGA), with 16.4% achieving an rIGA score of 0/1 versus 4.9% in the placebo group (p < 0.001) [29].

Treatment-emergent adverse events were generally similar between both placebo and rocatinlimab groups. Some of these events, including pyrexia, chills, and aphthous ulcers occurred in at least 4% of treated participants and at rates at least twice those of placebo. Most were mild or moderate, often injection related, and tended to occur more commonly after the first dose [29].

ROCKET-IGNITE (NCT05398445) is a 24-week, phase III, randomized, double-blind, placebo-controlled trial evaluating HD and LD rocatinlimab in 769 adults with moderate-to-severe AD. Participants received rocatinlimab Q4W with a loading dose at week 2 or placebo on the same schedule. Primary endpoints mirrored those of ROCKET-HORIZON [29].

At week 24, EASI-75 was achieved by 42.3% of patients in the HD group and 36.3% in the LD group, corresponding to 29.5% and 23.4% differences versus placebo, respectively (p < 0.001). An rIGA score of 0 or 1 was reached by 22.7% (HD) and 16.3% (LD) of patients, representing improvements of 14.4% (p < 0.001) and 8.0% (p = 0.01) compared with placebo. Moreover, a statistically significant improvement in vIGA was seen in the rocatinlimab group versus placebo (24% in HD, 19% in LD, and 9% in placebo); percentage difference versus placebo of 14.9% (p < 0.001 in HD and p = 0.002 in LD) [29].The safety profile of rocatinlimab was consistent with previous trials. Similar to HORIZON, TEAEs were evenly spread between groups in IGNITE. Most reactions were mild or moderate, frequently injection related, and occurred more often after the first administered dose [29].

ROCKET-SHUTTLE (NCT05724199) is a 24-week, phase III, randomized, double-blind, placebo-controlled trial assessing the efficacy and safety of rocatinlimab in combination with topical corticosteroids or topical calcineurin inhibitors in 746 participants with moderate-to-severe AD. Participants were randomized (1:1:1) to receive HD or LD rocatinlimab Q4W (with a loading dose at week 2) or placebo, alongside topical therapy [30, 31]. The trial used the same primary endpoints as ROCKET-HORIZON and ROCKET-IGNITE.

At week 24, a vIGA score of 0/1 was achieved by 26.1% (HD) and 25.6% (LD), while an rIGA score of 0/1 was achieved by 23.3% (HD) and 22.7% (LD), representing differences of 10.9–13.8% versus placebo (p ≤ 0.002). EASI-75 was achieved by 52.3% (HD) and 54.1% (LD), corresponding to 28.7% and 30.4% improvements over placebo (both p < 0.001).

ROCKET-VOYAGER (NCT05899816) is a 24-week, phase III, randomized, double-blind, placebo-controlled trial evaluating the impact of rocatinlimab on vaccine responses in 221 participants. Participants received rocatinlimab or placebo Q4W, with the primary endpoint being the proportion of participants achieving a positive anti-tetanus and anti-meningococcal vaccine response at week 24 [32]. Top-line results indicated that rocatinlimab did not impair vaccine responses. The remaining ongoing clinical trials in the ROCKET program are listed in Table 2.

Table 2.

Ongoing program of clinical trials investigating amlitelimab, rocatinlimab, STAR-0310, IMG-007, APG-990, and APG-279 for the treatment of atopic dermatitis

ClinicalTrials.gov identifier	Design	Primary endpoint	Status	Estimated study completion date
Amlitelimab
Phase III COAST-1 (NCT06130566) (40)	Multinational, multicenter, randomized, double-blind, placebo-controlled, parallel-group, three-arm study (N = 420)	Percentage of patients reaching EASI-75 and a vIGA score of 0/1 with a ≥2-point reduction in IGA, at week 24.	Recruiting	2025-11-14
Phase III SHORE (NCT06224348) (42)	Multinational, multicenter, randomized, double-blind, placebo-controlled, parallel-group, three-arm study (N = 496)	Percentage of patients reaching EASI-75 and a vIGA score of 0/1 with a ≥2-point reduction in IGA, at week 24	Recruiting	2026-01-21
Phase III ESTUARY (NCT06407934) (43)	Multinational, multicenter, randomized, double-blind, placebo-controlled, parallel-group study (N = 961), responders from the parent trials (COAST-1, COAST-2, or SHORE)	Proportion of participants who maintain a clinical response (vIGA score of 0/1 and/or EASI-75) at week 48 compared to treatment withdrawal	Recruiting	2027-01-05
Phase III COAST-2 (NCT06181435) (41)	Multinational, multicenter, randomized, double-blind, placebo-controlled, parallel-group, three-arm study (N = 420)	Percentage of patients reaching EASI-75 and a vIGA score of 0/1 with a ≥2-point reduction in IGA, at week 24	Recruiting	2026-02-03
Phase III AQUA (NCT06241118) (44)	Multinational, multicenter, randomized, double-blind, placebo-controlled, parallel-group, three-arm study (N = 249) who have had an inadequate response to prior biologic or oral JAKi therapy	Percentage of patients reaching EASI-75 and vIGA score of 0/1 with a ≥2-point reduction, at week 36 Percentage change in the weekly average of daily sleep disturbance NRS	Recruiting	2026-06-30
Phase II ATLANTIS (NCT05769777) (45)	Open-label, single-group, one-arm, long-term safety study (N = 901)	Percentage of patients who experienced TEAEs from baseline through week 176	Recruiting	2028-10-09
Phase II/III RIVER-AD (NCT05492578) (46)	Open-label, single-group, long-term safety study (N = 1310) who have previously been enrolled in an amlitelimab Oceana Program clinical trial	Percentage of patients who experienced TEAEs from baseline through week 332	Recruiting	2028-12-29
Phase II HYDRO (NCT06015308) (47)	Multicenter, randomized, double-blind, placebo-controlled, two-arm study (N = 215)	Percentage of patients with positive tetanus and pneumococcal vaccine response at week 16	Recruiting	2026-02-27
Phase III ROCKET-ORBIT (NCT05633355) (48)	Open-label, 52-week, non-randomized, single-group assignment study (N = 187), age 12–17 years	Number of participants with treatment-emergent serious AEs up to week 52	Active, not recruiting	2025-07-28
Phase III ROCKET-ASTRO (NCT05704738) (49)	Double-blind, 52 week, re-randomized, sequential assignment, placebo-controlled study (N =532), age 12–17 years	Percentage of patients achieving EASI-75 at week 24; vIGA score of 0/1 with ≥2-point reduction at week 24	Active, not recruiting	2025-11-27
Phase III ROCKET-OUTPOST (NCT06224192) (50)	Open-label, 52-week, randomized, performance study (N = 151)	Proportion of full-dose self-administered rocatinlimab injections among attempted home use injections up to week 16	Active, not recruiting	2026-03-11
Phase III ROCKET-ASCEND (NCT05882877) (51)	Double-blind, 116-week, randomized, parallel-assignment, placebo-controlled study (N = 2200) of those who have completed a parent study within the ROCKET program	Number of participants with TEAEs	Recruiting	2027-05-14
STAR-0313
Phase Ia (NCT06782477) (56)	Randomized, double-blind, placebo-controlled, single ascending dose in healthy adult participants (4 planned cohorts, potentially one additional cohort of Japanese descent)	Safety, tolerability, PK, and immunogenicity of STAR-0310 in healthy adult participants	Recruiting (status verified)	2026-01-31
Phase IIb (NCT07037901) (57)	Randomized, double-blind, placebo-controlled, subcutaneous dose-finding study in adults with moderate-to-severe atopic dermatitis (4 arms: high dose, medium dose, low dose and placebo)	Mean percent change from baseline in EASI at week 2	Recruiting (first patient dosed in July 2025)	Q4 2026
Phase I, monotherapy (54)	Phase I, randomized, double-blind, placebo-controlled, first-in-human, single-ascending dose in healthy adult volunteers	Safety, tolerability, and PK in healthy volunteers	Ongoing (phase I interim data shared, March 2025)	Interim results expected first half of 2025; full phase I completion not yet disclosed
Phase Ib (NCT07027527), APG-777/APG-990 combination therapy (58)	Open-label, randomized, multicenter, active-comparator study evaluating safety, tolerability, and PK of APG-777 + APG-990 vs dupilumab in adults with moderate-to-severe atopic dermatitis	Safety, tolerability, and pharmacokinetic parameters of the combination vs dupilumab	Recruiting (expected start mid-2025)	Duration ~82 weeks per participant; topline data readout expected in second half of 2026

AEs adverse events, EASI Eczema Area and Severity Index, IGA Investigator’s Global Assessment, JAKi Janus kinase inhibitors, NRS Numerical Rating Scale, PK pharmacokinetics, Q quarter, TEAEs treatment-emergent adverse events, vIG-validated Investigator’s Global Assessment

Telazorlimab

Telazorlimab (GBR 830/ISB 830) is a humanized immunoglobulin G1 monoclonal antibody that targets OX40 on T cells, blocking its interaction with OX40L and thereby disrupting downstream signaling [18, 26]. Three phase I studies evaluated the immunogenicity and pharmacokinetics of telazorlimab. In a single ascending dose trial in healthy participants, telazorlimab was administered intravenously at doses of 0.3, 1, 3, and 10 mg/kg. The drug was well tolerated, with comparable pharmacokinetic profiles between healthy individuals and patients with AD. It demonstrated high bioavailability, a prolonged half-life, and no evidence of target-mediated drug disposition [33].

Telazorlimab was the first monoclonal antibody targeting the OX40/OX40L pathway to advance to phase IIa evaluation (NCT02683928). In this randomized, double-blind, placebo-controlled repeated-dose trial, 62 participants with moderate-to-severe AD were randomized 3:1 to receive IV telazorlimab (10 mg/kg) or placebo on days 1 and 29.

The study demonstrated that two doses of telazorlimab, administered 4 weeks apart, were safe and well tolerated. Most TEAEs were mild to moderate, with headache, AD exacerbation, nasopharyngitis, and upper respiratory tract infection being the most frequent. Treatment with telazorlimab resulted in significant reductions in epidermal thickness at day 29 (p < 0.01) and day 71 (p < 0.001), with no notable changes in the placebo group. Telazorlimab also significantly decreased the expression of messenger RNA markers associated with Th1, Th2, Th17, and Th22 subsets. Clinically, 76.9% of patients receiving telazorlimab achieved EASI-50 at day 71, compared with 37.5% in the placebo group [34, 35].

A phase IIb, randomized, double-blind, placebo-controlled, parallel-group trial (NCT03568162) further evaluated the safety and efficacy of telazorlimab in adults with moderate-to-severe AD. The primary endpoint was the percentage change in EASI from baseline at week 16 [36].

In Part 1, 313 participants were randomized (1:1:1:1) to receive placebo or one of three SC telazorlimab regimens: (1) a 150-mg loading dose followed by 75 mg Q4W; (2) a 600-mg loading dose followed by 300 mg Q4W; or (3) a 600-mg loading dose followed by 300 mg Q2W for 16 weeks. This was followed by a 38-week open-label extension (300 mg Q2W) and a 12-week treatment-free period. In Part 2, 149 additional participants were randomized (1:1) to receive either a 1200-mg loading dose followed by 600 mg Q2W or placebo for 16 weeks, with a subsequent 38-week open-label period (600 mg Q2W) and a 12-week follow-up.

At week 16, the least-squares mean EASI reduction was greater in the 300-mg Q2W (Part 1, 54.4% vs −34.2% placebo, p = 0.008) and 600-mg Q2W (Part 2, −59.0% vs −41.8% placebo, p = 0.008) arms. These improvements were sustained through week 66, 12 weeks post-treatment. EASI-75 was achieved by 23.7% (300 mg Q2W) and 25.3% (600 mg Q2W) compared with 11.3% and 18.9% in their respective placebo groups.

Telazorlimab was well tolerated with most TEAEs mild to moderate. Common AEs included AD exacerbation, nasopharyngitis, upper respiratory tract infection, and headache, with similar rates in both treatment and placebo groups [36].

Currently, no new clinical trials are evaluating telazorlimab for the treatment of AD. However, a next-generation monoclonal antibody, STAR-0310, has been developed from telazorlimab with the aim of improving its pharmacological and therapeutic properties.

STAR-0310

STAR-0310 is a monoclonal antibody targeting the OX40 receptor, developed from an affinity-matured version of telazorlimab. It incorporates YTE (M252Y/S254T/T256E Fc-engineering) technology to extend the half-life and reduce the potential for antibody-dependent cellular cytotoxicity (ADCC). A phase Ia, randomized, double-blind, placebo-controlled, single-ascending-dose trial has now completed initial evaluation in healthy participants [56]. STAR-0310 demonstrated a best-in-class half-life of up to 68 days, consistent with potential dosing every 6 months. Single subcutaneous doses resulted in durable cytokine inhibition lasting up to 16–20 weeks, supporting sustained target engagement. The therapy was well tolerated across dose levels, with no serious TEAEs or discontinuations, and only mild AEs in a minority of participants that resolved without intervention; notably, no fever or chills were reported [22, 37]. These results strengthen the rationale for further clinical development of STAR-0310 as a long-acting OX40 antagonist, with a potentially favorable safety/tolerability profile and a low-frequency dosing regimen.

IMG-007

IMG-007 is a novel, humanized immunoglobulin G1 monoclonal antibody targeting OX40, designed to inhibit the OX40/OX40L signaling pathway while modulating ADCC without depleting T cells [21]. A phase I, randomized, double-blind, placebo-controlled trial (NCT06304740) evaluated the safety and pharmacokinetics of a single SC dose of IMG-007 in healthy volunteers. IMG-007 demonstrated a favorable safety profile, with injection-site reactions as the most common TEAEs, occurring more frequently in the placebo group (75%) than in the IMG-007 group (25%). These findings align with the intended suppression of ADCC activity. A single SC dose of IMG-007 showed a mean terminal half-life of 34.7 days [21].

Complementing the phase I trial, a phase IIa open-label study (NCT05984784) evaluated the efficacy and safety of IMG-007. Eligible patients received three IV 300-mg infusions Q2W (day 1, week 2, and week 4) and were followed through week 24.

At week 16, interim results showed a mean EASI reduction of 77%, with 54% of patients achieving EASI-75. Sustained inhibition of serum biomarkers associated with Th1, Th2, and Th17 cells was observed throughout the 24-week follow-up. IMG-007 was well tolerated, with no serious or treatment-related AEs reported, including pyrexia or chills; common with monoclonal antibody therapies [21, 38]. A phase IIb clinical trial is scheduled to begin in the first quarter of 2025 [21].

APG990

APG990 is a fully human, half-life-extended, immunoglobulin G1 monoclonal antibody that targets OX40 ligand (OX40L), aiming to disrupt the OX40/OX40L interaction involved in T-cell co-stimulation. It is engineered to prolong systemic exposure, supporting infrequent dosing. In a randomized, double-blind, placebo-controlled, phase I trial involving healthy volunteers, single subcutaneous doses of APG990 were evaluated for safety, pharmacokinetics, and pharmacodynamics. The treatment was generally well tolerated, with no serious or treatment-related AEs reported. The most frequently observed adverse event was mild headache. APG990 demonstrated a terminal half-life of approximately 60 days, indicating potential for dosing intervals of 3–6 months [54].

APG279

APG279 is an investigational combination therapy that unites two monoclonal antibodies, APG777, which blocks IL-13 and is currently in phase II trials for AD, and APG990, which targets the OX40 ligand pathway. Together, these agents aim to modulate distinct but complementary immune mechanisms central to the development of AD. The dual-antibody formulation is designed with extended half-life properties to reduce the frequency of dosing. Preclinical research indicates the potential for enhanced therapeutic activity when the two antibodies are combined. Building on these findings, a phase Ib clinical trial is expected to begin in late 2026 to directly compare APG279 with dupilumab, with a focus on treatment durability, effectiveness, and patient-centered outcomes [55].

Discussion

Given its heterogeneous pathophysiology and unpredictable clinical course, AD remains a therapeutic challenge and continues to significantly impact patients’ quality of life.

Approximately 75% of patients report dissatisfaction with current treatments, citing frequent flare-ups, the need for ongoing pharmacologic adjustments, and the slow onset of action associated with newer biologics [39]. These limitations underscore the need for novel therapies that target alternative pathogenic pathways.

The OX40/OX40L axis is a key co-stimulatory immune pathway involved in sustaining T-cell-mediated inflammation in AD. Inhibiting this interaction offers a novel therapeutic mechanism distinct from currently approved targeted therapies. Several agents targeting this pathway have shown promising efficacy and safety in early-phase trials.

In two phase II trials, amlitelimab demonstrated significant and sustained efficacy in patients with moderate-to-severe AD, meeting the primary and most secondary endpoints [24, 25]. The ongoing phase III Oceana Program trials are expected to provide further insights into the long-term efficacy and safety of amlitelimab [40–47].

Rocatinlimab also showed robust clinical efficacy in a phase IIb trial, with the 300-mg Q2W regimen yielding the most favorable results. The study met its primary endpoint, along with several key secondary endpoints [28]. Although Q2W dosing provided strong outcomes, Q4W dosing may offer greater practicality and improve treatment adherence. Top-line results from the ROCKET program further support the efficacy and safety of rocatinlimab in the treatment of AD, with recently published results of HORIZON and IGNITE [29, 48–51].

Telazorlimab was the first anti-OX40 monoclonal antibody to show promising results in AD, prompting further investigation. In the phase IIb trial, the higher dose regimens (300 mg and 600 mg Q2W following a loading dose) produced greater reductions in EASI at week 16 compared with placebo. However, overall clinical efficacy was limited, with no significant proportion of patients achieving EASI-75 [36].

Three novel anti-OX40 monoclonal antibodies, STAR-0310, IMG-007, and APG990 (the latter also part of a co-formulated combination therapy, APG279, which includes APG777, an IL-13 inhibitor), have been developed with Fc bioengineering modifications designed to extend the half-life and silence ADCC, thereby aiming to reduce the risk of adverse effects [52, 53]. STAR-0310, derived from telazorlimab, is currently being evaluated in a phase Ia study, and results are still awaited [37]. In a phase IIa trial, IMG-007 achieved a mean EASI reduction of 77% and an EASI-75 response in 54% of patients at week 16 following three 300-mg Q2W doses [38]. APG990 demonstrated favorable safety and tolerability in a phase I trial, with no serious or treatment-related AEs reported [54]. IMG-007 exhibited a half-life of 34.7 days, STAR-0310 a half-life of 26 days, and APG990 approximately 60 days — all notably longer than the typical 10–14 days observed with standard immunoglobulin G1 antibodies [21, 22]. Pharmacokinetic properties suggest that these agents may offer advantages in AD treatment through extended dosing intervals and reduced ADCC-related risks.

The clinical efficacy of OX40/OX40L pathway inhibitors has been shown to persist for 3–9 months after the final dose, despite serum drug concentrations falling below pharmacologically active levels. This durability was accompanied by sustained reductions in serum biomarkers associated with Th1, Th2, Th17, and Th22 subsets, suggesting a potential disease-modifying effect, that warrants further evaluation and definition [24, 25, 28, 36].

All OX40/OX40L pathway inhibitors have demonstrated favorable safety profiles, with no major AEs reported. The most common TEAEs were pyrexia and chills [23–25, 27–36]. Adverse effects associated with monoclonal antibodies are often linked to their immunomodulatory activity, which may increase susceptibility to infections. However, as OX40 is primarily expressed on effector T cells, inhibition of the OX40/OX40L pathway preserves homeostasis of naïve T cells, thereby avoiding generalized immunosuppression and reducing infection risk [26]. Currently approved biologics for AD target the type 2 inflammation pathway, which may lead to compensatory shifts toward Th1 or Th17 responses, contributing to adverse effects such as rosacea, alopecia, psoriasis, and arthralgia. In contrast, OX40/OX40L inhibitors modulate all Th cell subsets through their upstream mechanism of action, offering broader immune regulation and reducing the risk of adverse effects linked to skewed immune pathway activation [18, 26].

When considered alongside existing treatment options, OX40/OX40L inhibitors may represent a different therapeutic position within the AD landscape. Although currently approved advanced therapies have substantially improved moderate-to-severe disease management, they remain constrained by the need for continuous administration, heterogeneity in long-term durability, and variability in patient outcomes. Early evidence with OX40/OX40L inhibitors suggests a range of clinical responses broadly comparable to existing agents, while potentially offering longer treatment-free intervals, extended administration intervals, and a broader degree of immune recalibration. Key uncertainties, including which patients are most likely to benefit, the consistency of remission across populations, and how these therapies will compare directly with established agents, remain important areas for future study.

Nevertheless, the full results of ongoing phase III trials are needed to confirm the long-term efficacy and safety of these agents. Head-to-head studies comparing different OX40/OX40L inhibitors, as well as comparisons with established therapies, will be essential to define optimal dosing strategies and support more effective, personalized approaches to the treatment of AD.

Conclusions

Despite ongoing advances in AD treatments, a substantial proportion of patients remain unresponsive to available therapies. Monoclonal antibodies targeting the OX40/OX40L pathway, particularly amlitelimab and rocatinlimab, have emerged as promising options because of their potential disease-modifying effects. The dermatology community now awaits the results of ongoing phase III trials, as well as studies investigating the next-generation anti-OX40 antibodies IMG-007, STAR-0310, and APG990, to validate their long-term efficacy. Based on current clinical trial data, these novel therapies offer the potential to transform AD management, providing hope for sustained disease control and meaningful improvements in patient’s quality of life.

Funding

Open access funding provided by FCT|FCCN (b-on).

Declarations

Funding

No funding was received for the preparation of this article.

Conflicts of Interest/Competing Interests

David Oliveira dos Santos, Amir Mohamed, Adam Mohamed, and José Miguel Alvarenga have no conflicts of interest that are directly relevant to the content of this article. Tiago Torres has received consultancy and/or speaker’s honoraria from and/or participated in clinical trials sponsored by AbbVie, Amgen, Almirall, Arena Pharmaceuticals, Biocad, Biogen, Boehringer Ingelheim, Bristol Myers Squibb, Celgene, Fresenius-Kabi, Janssen, LEO Pharma, Eli Lilly, MSD, Mylan, Novartis, Pfizer, Samsung-Bioepis, Sanofi-Genzyme, Sandoz, and UCB.

Ethics Approval

Not applicable.

Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

Availability of Data and Material

Not applicable.

Code Availability

Not applicable.

Authors’ Contributions

All named authors (DOS, AM, AM, JMA, and TT) meet the International Committee of Medical Journal Editors criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given the final approval for the version to be published.