Long-term follow-up and treatment outcomes of conjunctivitis during dupilumab treatment in patients with moderate-to-severe atopic dermatitis

Long-term follow-up and treatment outcomes of conjunctivitis during dupilumab treatment in patients with moderate-to-severe atopic dermatitis Roselie Achten, MD*, Daphne Bakker, MD*, Lieneke Ariens, MD, Amanda Lans, MD, Judith Thijs, MD, PhD, Jorien van der Schaft, MD, PhD, Joke de Boer, MD, PhD, Deepak Balak, MD, PhD, Marlies de Graaf, MD, PhD, Chantal van Luijk, MD*, and Marjolein de Bruin-Weller, MD, PhD*

Ophthalmologist-confirmed conjunctivitis during dupilumab treatment was observed in 33 of 167 (19.8%) of patients with atopic dermatitis. Most patients still suffered from mild-to-moderate conjunctivitis during long-term follow-up despite treatment. Dose adjustment or discontinuation of dupilumab was needed in 10 of 33 (30%) and 3 of 33 (9%) patients, respectively.
Dupilumab is the first biologic treatment for atopic dermatitis (AD), and its effectiveness and safety are proven. 1 Although conjunctivitis is the most frequently reported side effect during dupilumab treatment in both clinical trials and daily practice, data on the clinical course of conjunctivitis during long-term use of dupilumab are lacking. [1][2][3] This prospective daily practice study evaluates ophthalmological characteristics and long-term treatment outcomes of ophthalmologist-confirmed conjunctivitis during dupilumab treatment in patients with moderate-to-severe AD. During a 12-month evaluation period, 167 patients with moderate-to-severe AD were treated with dupilumab 300 mg every 2 weeks, after a loading dose of 600 mg, at the University Medical Center Utrecht, the Netherlands. Patients reporting ophthalmological symptoms that could not be controlled with lubricant drops and/or tacrolimus skin ointment (1 mg/g) for the external eyelids were referred to an ophthalmologist. Further (anti-inflammatory) ophthalmological treatment was prescribed by the ophthalmologist and individually chosen per patient. Conjunctivitis was reported in 66 of 167 (39.5%) patients, of whom 33 were referred to an ophthalmologist. Ophthalmologistconfirmed conjunctivitis was reported in 33 of 167 (19.8%) patients (17 female; mean age 45.7 years, standard deviation [SD] 14.3; mean Eczema Area Severity Index at baseline, 21.7 (SD 9.5); Table E1, available in this article's Online Repository at www.jaci-inpractice.org). History of (allergic) conjunctivitis was present in 24 of 33 (72.7%) patients. None of the 33 patients reported conjunctivitis symptoms at the start of dupilumab. In the 33 referred patients, patient-reported eye symptoms, such as redness, tearing, and itching, developed within a median of 33 days (interquartile range, 28.0-61.0) after starting dupilumab. Ophthalmological characteristics were examined and graded in terms of severity by an experienced ophthalmologist following the Utrecht Ophthalmic Inflammatory and Allergic disease ocular surface score (Table I). Overall conjunctivitis severity was based on grading of different ophthalmological characteristics ( Figure 1, A and B).
The most frequently prescribed ophthalmological treatments during follow-up included corticosteroid eye drops, tacrolimus skin ointment for the external eyelids, and lubricant drops ( During follow-up (mean, 17.5 [SD AE3.4] months), the dosing interval of dupilumab was prolonged to 300 mg every 3 to 5 weeks in 10 of 33 (30%) patients because of conjunctivitis, resulting in improvement of eye symptoms in 6 patients and remission in 1 patient. Discontinuation of dupilumab due to ocular pathology was necessary in 3 of 33 (9.1%) patients, showing improvement or remission in all cases (Figure 1, C). Ineffectiveness of dupilumab led to discontinuation in 2 of 33 (6.1%) patients.
The conjunctivitis outcome during a follow-up of 17.6 (SD AE3.5) months was evaluated for 28 of 33 (84.8%) patients who continued dupilumab, by comparing the first conjunctivitis severity category with the latest follow-up category (Figure 1, D). Outcomes were categorized into worsened (worsening with !1 category), stable (unchanged category), improved (improvement with !1 category), or complete remission (no conjunctivitis). Complete remission was seen in 4 of 28 (14%) patients; of these, 2 were still using anti-inflammatory eye drops or tacrolimus ointment for the external eyelids. Improvement of conjunctivitis occurred in 7 of 28 (25%) patients, of whom 6 were still using anti-inflammatory eye drops. Uncontrolled conjunctivitis, meaning stable or worsened conjunctivitis, was seen in 17 of 28 (61%) patients. Ophthalmic anti-inflammatory therapy was prescribed for all of these 17 patients; however, 2 of 17 patients reported being noncompliant.
Literature regarding conjunctivitis during dupilumab is limited by small sample sizes, short follow-up duration, and lack of thorough and standardized ophthalmological investigation. In contrast, all 33 patients of our study underwent standardized examination by an ophthalmologist followed by long-term follow-up.
Several pathomechanisms have been suggested to be responsible for the development or worsening of conjunctivitis during dupilumab treatment in patients with AD, such as rosacea-like conjunctivitis, focal scarcity of intraepithelial goblet cells, and relative ocular undertreatment due to lower tissue distribution of dupilumab in the eyes. 2,4,5 The last hypothesis seems in contradiction with our finding that interval prolongation or discontinuation of dupilumab resulted in improvement of the conjunctivitis.
The management of conjunctivitis during dupilumab treatment is challenging. Previous case series and case reports have described several therapeutic options, including tacrolimus eye ointment, fluorometholone eye drops, cyclosporine eye drops, and lifitegrast eye drops, leading to improvement in most cases. [6][7][8] The majority of our patients received combination therapy, and most patients remained dependent on ophthalmic medication. Anti-inflammatory eye drops and/or tacrolimus ointment for the external eyelids were prescribed most often.
In contrast to clinical trial data, reporting that most conjunctivitis cases recovered or resolved while continuing dupilumab treatment, our results show more persistent ophthalmological signs and symptoms despite adequate ophthalmic treatment. Remarkably, 8 of 33 (24.2%) patients developed limbitis during follow-up, and 6 cases despite adequate ophthalmic anti-inflammatory treatment. Limbal stem cells are vital for corneal healing and the barrier function of the limbus. Chronic limbitis may lead to irreversible limbal stem cell deficiency, which could lead to irreversible long-term visual loss, making adequate monitoring of conjunctivitis necessary. 9 This study has some limitations. First, because all patients were seen in an AD expertise center, the population consisted of patients with more severe AD. As severity of AD may be related to the development of conjunctivitis during dupilumab Overall severity of the conjunctivitis None/mild/moderate/severe conjunctivitis † *None ¼ 0 points; mild ¼ 1 point; moderate ¼ 2 points; severe ¼ 3 points. †0 ¼ no conjunctivitis; 1-4 ¼ mild conjunctivitis (unless the score consists of only Meibomian gland dysfunction and punctate, then the total score is 0); 5-8 ¼ moderate conjunctivitis; !9 ¼ severe conjunctivitis.
treatment, this may have affected the results. 2 Secondly, not all patients may have been compliant with ophthalmic treatment, which might have resulted in undertreatment of the conjunctivitis. Lastly, ophthalmological examination by an ophthalmologist was not performed before starting dupilumab; therefore, pre-existing ophthalmological pathology cannot be excluded.
In conclusion, this study shows ophthalmologist-confirmed conjunctivitis in 33 of 167 (19.8%) patients with AD treated with dupilumab in a 1-year period. During long-term ophthalmological follow-up, the majority of these patients still suffered from mild-to-moderate conjunctivitis despite treatment. Dose adjustment or discontinuation of dupilumab due to ocular pathology was needed in 10 of 33 and 3 of 33 of the patients, respectively.   Other therapy 3 (9.1) Data are n (%) unless otherwise indicated. Multiple therapies per patient. Anti-inflammatory treatment for the external eyelids included tacrolimus skin ointment; corticosteroid eye drops included fluorometholone, dexamethasone, hydrocortisone, softacor, and prednisolone; antihistamine eye drops included ketotifen; other anti-inflammatory therapy (eye drops/eye ointment) included tacrolimus eye ointment and cyclosporine A eye drops; combined anti-inflammatory and antimicrobial treatment (eye drops/eye ointment) is terracortril and tobradex; other therapies are cross-linking, and bandage lens with chloramphenicol.