Dupilumab-Associated Ocular Surface Disease: Clinical Characteristics, Treatment and Follow Up

Abstract

Purpose:

A consecutive case series of patients with dupilumab-associated ocular surface disease (DAOSD) that describes common ocular symptoms and signs, proposes a symptom disease severity grading system, and describes treatment strategies of DAOSD patients.

Methods:

Retrospective chart review of patients with concomitant dupilumab-treated atopic dermatitis (AD) and DAOSD with ophthalmic evaluation between January 2014 and May 2019.

Results:

29 patients (mean age 46 years, M/F: 12/17) with 57 ophthalmic exams were identified. The most common ocular symptoms included irritation/pain (n=28, 97%), redness (n=24, 83%), pruritus (n=18, 62%), discharge (n=18, 62%) and light sensitivity (n=6, 21%). The most frequent signs included conjunctival injection (n=18, 62%), superficial punctate keratitis (n=16, 55%) and papillary reaction (n=8, 28%). Topical corticosteroids (TCS) (n=23, 79%), tacrolimus (n=6, 21%), and artificial tears (n=7, 24%), were the most commonly utilized therapy. Of those with follow-up documentation (n=21), 20 were noted to have partial or complete response with TCS based on symptoms and reduction of signs. Using our proposed symptom-based grading scale, scaled 1 to 5 based on presence of the common symptoms listed above, 66% (n=19) requiring topical immunomodulating therapy were found in the ‘severe’ group (≥3 symptoms) and 17% (n=5) were found in the ‘mild’ group (≤2 symptoms)

Conclusions:

This study provides insight into the common presenting ocular signs and symptoms associated with DAOSD, and highlights the efficacy of TCS and other immunomodulators in improving symptoms associated with DAOSD. Based on our findings we propose a symptom-based grading system that can guide non-ophthalmic physicians regarding ophthalmology consult.

Introduction:

There is a well-established association between severe atopic dermatitis (AD) and ocular surface diseases—including allergic conjunctivitis, blepharitis, and keratitis—with incidence rates increasing with increased AD severity.^1-3

In recent years, several new systemic medications have been trialed for those with severe AD. Dupilumab is the first biologic medication approved in the United States for treatment of moderate-to-severe AD in patients ≥12 years of age. It is a monoclonal antibody that inhibits the type 2 helper T-cell (Th2) mediated inflammatory pathway by antagonizing the shared receptor component for interleukin (IL)-4 and IL-13, thereby inhibiting IL-4 and IL-13 signaling. The pathogenesis of AD is multifactorial, driven by a complex interplay between environmental factors, epidermal barrier defects, immune response dysregulation and genetic polymorphisms.⁴ Research indicates that IL-4 and IL-13 are important key mediators of atopic diseases, including AD.^{5, 6} Dupilumab has demonstrated efficacy and safety for the treatment of several atopic conditions—including AD^7-9 and asthma^{10, 11}—further supporting the involvement of IL-4 and IL-13 in the pathophysiology of Th2 inflammatory conditions.

While few adverse events were reported in clinical trials, increased incidence of conjunctivitis with dupilumab compared to placebo was consistently observed.^{8, 9, 12} A recent small case series also identified conjunctivitis following dupilumab therapy in two clinical practices, with up to 50% of patients developing dupilumab-associated conjunctivitis.¹³ The etiology and risk factors for the development of dupilumab-associated conjunctivitis are unclear but seem to be specific to the AD population—no clear association has been observed in asthma trials of dupilumab.¹¹ Previous studies have suggested that conjunctivitis develops in patients with a higher baseline severity AD as well as a strong atopic history. The conjunctivitis usually develops in patients whose AD is greatly responsive to dupilumab.^{13, 14} Other dupilumab-associated ocular surface diseases (DAOSD) have been described—including keratitis, eye pruritus, and dry eye—but are far less frequent than conjunctivitis.^{15, 16} There are currently no data regarding the natural course of DAOSD in clinical practice. Treatment for DAOSD has not been well established—case reports and small case series have described partial response to topical low-potency corticosteroids (TCS), tacrolimus, and lifitegrast.^{13, 15} Dupilumab’s efficacy in controlling severe AD has propelled the increased clinical use of dupilumab and the resultant increased incidence of DAOSD.

The primary aim of this study was to investigate the most common presentation of patients developing DAOSD, and elucidate the treatment strategies of DAOSD. Additionally, this study proposes a systematic grading system and flow chart for the use of treatment modalities and ophthalmology referral to improve early recognition of disease by non-ophthalmic trained physicians.

Materials and Methods:

We performed a retrospective chart review of pediatric and adult patients with concomitant dupilumab-treated AD and DAOSD seen at Oregon Health & Science University (OHSU) from January 2014 to May 2019. Patients seen in an outpatient dermatology clinic with AD (captured by ICD-10: L20) treated with dupilumab who were referred to ophthalmology were independently identified using i2b2 web-based Cohort Discovery System. Chart review was performed to confirm presence of AD, use of dupilumab, history of ophthalmology evaluation at OHSU, diagnosis of DAOSD, and gather additional health information. Only those who had ophthalmic evaluation at OHSU with subsequent documentation of ocular surface disease were included in the analyses.

A 3-category scale of response to ophthalmic treatments was generated based on the clinical evaluations reported by 2 ophthalmologists (JC and WC): no response, partial response, and complete response. These categories were generated based on both subjective reports from patients and clinical slit lamp exam findings of reduced signs.

All data were entered into the Research Electronic Data Capture application (REDCap). Analysis was performed in the STATA statistical analysis program, version 15. The institutional review board at OHSU approved this study (IRB no. 19623). A waiver of consent was granted given the retrospective design and minimal risk.

Results:

Table 1 lists the demographics and medical history of this patient cohort. Through the Cohort Discovery System, we identified 57 patients using the criteria outlined above. Of the 57 charts reviewed, 29 demonstrated symptoms and signs consistent with DAOSD upon ophthalmic examination with a total of 57 ophthalmic exams. The mean age of the cohort was 46 ± 15.2, consisted of 17 females (58.6%), and was predominantly Caucasian (n=22, 79%).

Table 1.

Demographics and Medical History

Sex, n (%)
Male	12 (41.4)
Female	17 (58.6)
Age (y)
Mean (SD)	45.5 (15.2)
Min, max	11.8, 72
Ethnicity, n (%)
Non-Hispanic	25 (89.3)
Hispanic	3 (10.7)
Race, n (%)
White	22 (78.6)
Asian	5 (17.9)
Black	0 (0)
American Indian or Alaska Native	0 (0)
Native Hawaiian or Other Pacific Islander	0 (0)
Two or more races	1 (3.6)
Smoking history, n (mean pack years)	5 (7.7)
Other medical conditions, n (%)
Diabetes	2 (6.9)
Hypertension	7 (24.1)
Dyslipidemia	5 (17.2)
Leukemia/Lymphoma	1 (3.4)
COPD	2 (6.9)
Other	9 (31.0)

^*All data are reported as n (%) unless indicated otherwise

Table 2 lists the atopic medical history of this cohort. Twenty-six (90%) and 19 (66%) had a history of facial and eyelid AD, respectively. Average body surface area affected by AD was 48% before dupilumab (n=21), and 31% at the time of ophthalmology referral (n=15). Nine patients (32%) had a history of allergic conjunctivitis, and 2 (7%) had a history of atopic keratoconjunctivitis. Additionally, 26 (93%) and 22 (79%) patients reported a concurrent atopic history of allergic rhinitis and asthma, respectively. The average number of weeks from dupilumab initiation to ophthalmology referral were 40.4±37.3. Dupilumab was discontinued in 3 (10%) patients, due to the severity of DAOSD.

Table 2.

Atopic Dermatitis History^*

Years with AD (n=20)
Average years (SD)	43 (15.3)
Min, max	11.8, 69.6
Pertinent atopic history (n=28), n (%)
Asthma	22 (78.6)
Allergic rhinitis	26 (92.9)
Allergic conjunctivitis	9 (32.1)
Food allergy	12 (42.9)
Facial AD, n (%)	26 (89.7)
Eyelid AD, n (%)	19 (65.5)
Dupilumab
Discontinued, n (%)	3 (10.3)
Duration from initiation to ophthalmology referral (n=27), average weeks (min, max)	40.4 (3.7, 128.1)
AD severity
BSAå, mean
Before dupilumab (n=21)	48.3
At time of ophthalmology referral (n=15)	30.5

^*All data are reported as n (%) unless indicated otherwise; categories without 29 records indicate missing data

AD, atopic dermatitis;

^åBSA, body surface area (percent of body surface affected by atopic dermatitis)

Table 3 lists the ophthalmic symptoms, signs, diagnosis and management of the cohort. The most common ocular symptoms included irritation or pain (n=28, 96.6%), redness (n=24, 83%), pruritus (n=18, 62%), and tearing (n=18, 62%). The most frequent ocular signs included conjunctival injection (n=18, 62%), superficial punctate keratitis (n=16, 55%) and papillary reaction (n=8, 28%). Other signs included follicular reaction (n=4, 14%), mucous discharge (n=3, 10%), and ropey discharge (n=1, 3%).

Table 3.

Ophthalmic Symptoms, Signs, Diagnosis, and Treatment (n=29)^*

Total ophthalmic exams, n	57
Symptoms (n=29)
Irritation/pain	28 (96.6)
Redness	24 (82.8)
Watering/tearing/discharge	18 (62.1)
Pruritus/itching	18 (62.1)
Light sensitivity	6 (20.7)
Swelling	3 (10.3)
Physical exam signs, n (%)∆
Conjunctival (n=29)
Papillary reaction	8 (27.6)
Follicular reaction	4 (13.8)
Erythema/injection	18 (62.1)
Ropey discharge	1 (3.4)
Mucous discharge	3 (10.3)
Corneal (n=29)
SPK/PEE	16 (55.2)
Ophthalmic DAOSD diagnoses, n (%)
Conjunctivitis	27 (93.1)
Keratitis	2 (6.9)
Blepharitis	2 (6.9)
Topical therapies used n=29, n (%)
Steroidså	23 (79.3)
Cyclosporinef	1 (3.4)
Lifitegrast	3 (10.3)
Tacrolimus¥	6 (20.7)
Artificial tears	7 (24.1)
Antihistamines∂	4 (13.8)
Response to topical steroids n=21, n (%)
No response	1 (4.8)
Partial response	16 (76.2)
Complete response	4 (19.0)
Response to observed at follow-up day n=, mean days (min, max)	54.8 (3, 145)

^*All data are reported as n (%) unless indicated otherwise

SPK/PEE, superficial punctate keratitis/punctate epithelial erosions; DAOSD, dupilumab-associated ocular surface disease

^∆Diffuse conjunctival/lid swelling, corneal infiltrates, corneal melt, limbal edema, and corneal thinning were not observed in this population

^åFluorometholone; 1% prednisolone acetate; 0.2%, 0.5%, or 1.0% loteprednol etabonate; 3/1 mg/ml tobramycin/dexamethasone

^f0.05%

^¥0.1% or 0.03%

^∂0.025% ketotofen fumarate; 0.1%, 0.2% or 0.7% olapatadine hydrochloride

Twenty-seven (93%) were diagnosed with conjunctivitis, 2 (7%) with blepharitis, and 2 (7%) with keratitis. Common therapies included TCS (n=23, 79.3%), topical tacrolimus (n=6, 20.7%), and artificial tears (n=7, 24.1%). Of the 21 patients with follow-up documentation, 1 (4.8%) had no response, 16 (76.2%) partial, and 4 (19.0%) demonstrated complete response to TCS. A response to TCS was noted and recorded at patient’s first follow up appointment, which occurred at an average of 70.8±83.0 days from initial presentation. Figure 1 demonstrates a patient with moderate to severe DAOSD. This patient failed mild steroid therapy and experienced an intraocular pressure increase on more potent TCS, but ultimately did well on combination topical tacrolimus and loteprednol etabonate.

Figure 1.

Photograph of a patient with moderate-severe ophthalmic findings associated with Dupilumab use before (A) & 6 months after (B) topical steroid and tacrolimus therapy. 67-year-old female patient who had poor response to mild topical steroid (0.1% fluoromethololone) and elevated intraocular pressure response to stronger steroid treatment (1% Prednisolone Acetate) and was transitioned to topical tacrolimus (0.1% ointment to eyelids and 1% loteprednol etabonate).

Discussion:

Patients with AD treated with dupilumab are at a known increased risk of developing conjunctivitis when compared to placebo.^{8, 9, 12, 17} While the exact pathogenesis of DAOSD is still poorly understood, baseline AD severity and prior history of conjunctivitis are both independent risk factors for the development of DAOSD.¹⁷ Additionally, certain biomarkers (TARC, IgE and eosinophils) have been found to be associated with the development of conjunctivitis in patients with more severe AD.^{18, 19} When compared to the general population, AD is associated with a higher prevalence of ocular co-morbidities.¹ Based on these observations, it is possible that patients with AD who have preexisting ocular disorders may have a lower threshold for the development of DAOSD or that DAOSD may be an exacerbation of pre-existing milder conditions in some patients. Interestingly, 90% of the patients in our cohort had a history of facial AD and 66% had eyelid AD as determined by the referring dermatologist, which are probable predictors for patients who develop symptomatic conjunctivitis. Large scale prospective trials are required to clarify these relationships.

In the present study, we observed a predominance of conjunctival DAOSD when compared to alternative diagnoses—such as keratitis and blepharitis—which parallels prior studies.¹⁶ This cohort had a higher prevalence of severe DAOSD compared to prior studies. A recent study by Akinlade et al. evaluating the incidence of conjunctivitis in dupilumab trials found severe conjunctivitis in ≤0.5%, with the majority of DAOSD conjunctivitis being reported as mild to moderate.¹⁷ The discrepancy may reflect a selection bias, as this cohort represents patients who were referred and evaluated by a university ophthalmology practice. Other dupilumab treated patients from our dermatology department with perhaps milder eye symptoms or established ophthalmologist/optometrist relationships outside the university were not evaluated. Thus, the determination of the prevalence of conjunctivitis in a dupilumab treated AD population is beyond the scope of this retrospective study, however prevalence has been established in other studies.^6,12,13,14. Our cohort does represent findings and outcomes in the more severe cases of a university dermatology practice that are most likely to present to the ophthalmologist.

Several case studies and case series have described various treatment modalities for DAOSD—including TCS, antibiotics, antihistamines, lifitegrast, mast cell stabilizers, and calcineurin inhibitors.^{15, 17, 20, 21} This study offers added benefit of follow-up data illuminating the efficacy of TCS in managing DAOSD. There are known risks associated with TCS use—including increased intra-ocular pressure, infectious keratitis, and long-term risk of cataract development. Figure 1 illustrates a patient who had successful treatment under ophthalmology care but needed monitoring for intraocular pressure rise after starting TCS in combination with topical tacrolimus. Neither therapy alone was adequate to control her symptoms. Additionally, patients with atopic disease have higher risks for glaucoma, cataracts, and herpetic keratitis independent of TCS.¹ As such, ocular disease necessitating TCS therapy should be managed by an ophthalmologist.

With increasing utilization of dupilumab in clinical practice, an anticipated rise in DAOSD is expected. As such, there is a need for established protocols aiding clinicians in early recognition, management, and criteria for referral to ophthalmologists. We propose a simple symptom-based severity grading scale (Table 4) and management diagram (Figure 2) based on our observations. The proposed system is based on the five most prevalent symptoms observed in our patient population—including light sensitivity, irritation/pain, discharge, redness, and pruritus (LID RP). Using this novel grading system, 7 (31.8%) demonstrated ‘mild’ symptom-based disease (score ≤2) at their first ophthalmic examination, and 22 (75.9%) with ‘severe’ disease (score ≥3). The predominance of severe disease observed in the present cohort can be explained by inclusion of only patients whose ocular disease warranted referral for ophthalmic evaluation by academic dermatology providers. Of those requiring topical immunomodulating therapy, including TCS, 65.5% (n=19) of the patients were found in the severe group where-as only 17.2% (n=5) were found in the mild group. These numbers partially reflect that our cohort had much larger number “severe” category patients and majority of the cohort was started on topical immunomodulators. However, a larger percentage of patients in the “mild” category may not require topical steroids. Figure 2 outlines a flow chart for the management of DAOSD by non-ophthalmic providers, such as dermatologists and primary care physicians. For those with mild disease (score ≤2) we advise against initial use of immunomodulator therapies, and instead recommend a treatment trial of topical antihistamine or mast cell stabilizers and/or over the counter artificial tears with follow-up with the prescribing dermatologist or primary care physician. In our “mild” category cohort, five patients (71% of the mild category) were treated with either topical antihistamines, mast cell stabilizers, or artificial tears and did not require treatment with topical immunomodulators. All of these therapies have demonstrated efficacy in management of disease in prior studies. ^{20, 21} Additionally, all of these medications have a lower side effect profile when compared to TCS, are available in over the counter forms in the US and several countries, and can be managed by non-ophthalmic physicians. For more numerous (or severe) symptoms (score of ≥3), an ophthalmology consult should be obtained. Because atopic disease has associated ophthalmic pathology, some of these patients may already be under an ophthalmologist’s care. The treating dermatologist should inform the patient’s ophthalmologist of their ocular reaction to dupilumab in order to expedite a follow-up exam.

Table 4.

Eye Symptom-Based Grading Scale (LID RP)

Criteria (1 point each)	Light Sensitivity
	Irritation or pain
	Discharge
	Redness
	Pruritus
Grading Scale
Mild	0-2 point
Severe	3-5 points

Figure 2.

Flow Diagram Based on Symptom-Based Grading Scale

This study has several limitations including a relatively small cohort that lacked ethnic diversity. Additionally, the present cohort was gathered exclusively from a tertiary care clinic setting, and may represent patients with more severe dermatologic and ophthalmic disease. The simple grading system was retrospectively tested on our cohort based on multiple iterations of symptoms scores and even weighting certain symptoms, but was found to be most predictive in in its current form. The cohort is small and the population may not provide a broad enough range of disease presentation to be generalizable. Larger studies involving more patients with longer follow-up will clarify complications, the disease course of DAOSD, and provide validation or, modification, for the proposed grading scale.

In conclusion, this study provides insight into the common presenting ocular signs and symptoms associated with DAOSD, and highlights the efficacy of TCS in improving symptoms associated with DAOSD. Additionally, this study proposes a novel symptom-based grading system, which can guide dermatologists and other care providers toward an ophthalmology consult.