High dupilumab levels in tear fluid of atopic dermatitis patients with moderate‐to‐severe ocular surface disease

Abstract Background The patho‐mechanism of ocular surface disease (OSD) in dupilumab‐treated atopic dermatitis (AD) patients remains unclear. The aim of this study is to measure dupilumab levels in tear fluid and serum, and relate these findings to the severity of OSD during dupilumab treatment in AD patients. Methods This prospective study included dupilumab‐treated moderate‐to‐severe AD patients who were seen by a dermatologist and an ophthalmologist before the start of dupilumab (baseline), and after 4 and 28 weeks of dupilumab treatment. Dupilumab levels in tear fluid and serum were measured by liquid chromatography coupled with tandem mass spectrometry (LC‐MS/MS). Additionally, a pilot study was conducted to measure dupilumab on conjunctival epithelial cells using flow cytometry and LC‐MS/MS. Results At baseline, 89.6% (n = 43/48) of the patients had OSD, with 50.0% having moderate‐to‐severe OSD. After 28 weeks of dupilumab treatment, the median dupilumab tear fluid levels were 0.55 mg/L (IQR 0.35–1.31) and 0.29 mg/L (IQR 0.16–0.60) in patients with moderate‐to‐severe OSD and patients with no or mild OSD, respectively (p = 0.02). Dupilumab levels could be detected on conjunctival epithelial cells of 5 AD patients treated with dupilumab for 4 weeks. Conclusion Patients with moderate‐to‐severe OSD had higher dupilumab tear fluid levels compared to patients with no or mild OSD, indicating that dupilumab reaches the ocular surface. Dupilumab was also detected in conjunctival cell suspensions and was found to directly bind CD45‐conjunctival epithelial cells. This suggests that AD‐induced changes of the conjunctival epithelium may play a role in the development of OSD as well as increased local drug availability.


| INTRODUCTION
Atopic dermatitis (AD) is a chronic, inflammatory, itchy skin disease with a prevalence up to 10% in adults. 1,2 The first biologic therapy that has been introduced to treat moderate-to-severe AD is dupilumab. Dupilumab is a fully human monoclonal IgG4 antibody that is directed against the interleukin (IL)-4 receptor-alpha (IL-4Rα) subunit, inhibiting the binding of IL-4 and IL-13. 3 It has proven its effectiveness in both clinical trials and daily practice studies. [3][4][5] The most frequently reported adverse event during dupilumab treatment in AD patients is dupilumab-associated ocular surface disease (DAOSD), which has been reported in up to 34% of the patients. [4][5][6] Remarkably, in dupilumab trials in other type-2 inflammatory diseases, such as asthma and chronic rhinosinusitis with nasal polyposis, no increased rates of DAOSD were reported. 6 Recently we found that 90% of the moderate-to-severe AD patients had clinical characteristics of ocular surface disease (OSD) before the start of dupilumab, suggesting that AD patients may have a predisposition to develop DAOSD. 7 It might be possible that this pre-existent OSD is aggravated during dupilumab treatment and is then diagnosed as DAOSD.
Several hypothesis are suggested to be responsible for the development of DAOSD, such as focal scarcity of conjunctival goblet cells (GCs). This might be a result of the IL-13 blocking effect, leading to reduced GC hyperplasia. 6,8 It is also hypothesised that DAOSD incidence may decrease in patients with higher serum dupilumab concentrations, and that local under-treatment of dupilumab in the eyes might play a role in the development of DAOSD. 6,9 At this moment, the exact patho-mechanism of DAOSD remains unclear.
To the best of our knowledge, no data are available on dupilumab levels in tear fluid of dupilumab-treated AD patients. The aim of this study was to measure dupilumab levels in tear fluid and serum, and relate this to the severity of OSD during dupilumab treatment in AD patients.  was reported per patient per visit. Severity of OSD was based on the eye with the highest severity within a patient. During ophthalmological examination, topical anaesthesia eye drops (0.4% oxybuprocaine hydrochloride) were dripped in both eyes. Tear production was measured by a Schirmer's test. 11 Subsequently, Schirmer's strips were air-dried for at least 1 h and stored at −80°C until further analyses.

| Processing of the Schirmer's strips
Schirmer's strips were eluted to obtain tear fluid for further processing. The elution buffer included PBS, Tween20 0.50%, BSA 1%, and 1 EDTA-free Protease Inhibitor Cocktail tablet (Sigma). Every Schirmer's strip was cut into small pieces and placed in one well of a Falcon plate. Next, 100 μL elution buffer was added per well, the plate was sealed and incubated over night at 4°C on a shaker (230 RPM). The next day, a quick spin down was done for 1 min

| Statistical analysis
Patient characteristics were described using absolute numbers (N) and percentages for categorical variables and median with interquartile ranges (IQR) for non-normally distributed continuous variables. Differences in dupilumab levels between no or mild OSD and moderate-to-severe OSD were calculated using the Mann-Whitney U test. Correlations between dupilumab levels in tear fluid and serum, and UTOPIA scores were described using Spearman's test. A p-value of less than 0.05 was considered statistically significant.

| Dupilumab levels in tear fluid and serum at week 4 and week 28
To investigate whether dupilumab is able to reach the ocular surface,  Table S1). Additionally, a significant correlation between dupilumab tear fluid levels at week 28 and UTOPIA score at week 28 was found (Spearman's correlation 0.505, p < 0.001).
In contrast to dupilumab tear fluid levels, dupilumab serum levels  Table S1).
After 28 weeks of treatment with dupilumab, no significant differences were found in dupilumab serum levels in patients with moderate-to-severe OSD at week 28 compared to patients with no ACHTEN ET AL. or mild OSD at week 28 ( Figure 1D, Table S1). Both at week 4 and week 28 of dupilumab treatment, no correlations were found be- Severity of OSD before the start of dupilumab, a n (%) 0.16 mg/L (IQR 0.09-0.55), p = 0.047) ( Figure 1E, Table S1).
Taken together this demonstrates that although dupilumab serum levels are lower (at week 4) or similar (at week 28) in patients with moderate-to-severe OSD compared to patients with no or mild OSD, local dupilumab levels in tear fluid do increase with increasing OSD severity.

| Dupilumab on conjunctival cells
The presence of dupilumab in tear fluid does not prove its binding and potential direct biological effect on conjunctival cells. Therefore we investigated dupilumab binding on conjunctival cells in a pilot study by analysing CIC suspensions with LC-MS/MS and flow cytometry (Gating strategy in Figure S1, patient characteristics are shown in Table S2). Dupilumab levels could indeed be detected in the conjunctival cell suspensions of 5 AD patients treated with dupilumab for 4 weeks compared to 3 AD controls (Figure 2A).

| DISCUSSION
This prospective study shows that OSD is very frequent in moderateto-severe AD, and that dupilumab-treated AD patients with moderate-to-severe OSD had higher dupilumab tear fluid levels compared to patients with no or mild OSD during dupilumab treatment. Additionally, dupilumab was detected in conjunctival cell suspensions of five AD patients after 4 weeks of treatment with dupilumab.
At week 4 of dupilumab treatment, lower serum dupilumab levels were found in patients with moderate-to-severe OSD. However, the steady-state is achieved after 16 weeks of dupilumab treatment, which may explain the variation in the week 4 levels. 13 After 28 weeks of dupilumab treatment, dupilumab serum levels of patients with no or mild OSD at week 28 were similar to dupilumab serum levels of patients with moderate-to-severe OSD at week 28, and no correlation was found between serum dupilumab levels and UTOPIA scores at week 28. However, tear fluid dupilumab levels were higher in patients with moderate-to-severe OSD at week 28 of dupilumab treatment. These findings are in contradiction with the hypothesis that development of DAOSD is related to local undertreatment by dupilumab in the eyes, based on an inverse relationship between serum dupilumab levels and conjunctivitis. 9 Furthermore, the hypothesis of Akinlade et al., 6 who suggested that conjunctivitis incidence might decrease with higher dupilumab concentrations is also not in line with our findings. Based on our results, higher serum levels of dupilumab do not prevent development of OSD. This is also supported by our previous findings that prolongation of the dosing interval of dupilumab or discontinuation of dupilumab resulted in improvement of DAOSD. 10 In addition, we recently investigated dupilumab serum levels in moderate-to-severe AD patients after 16 weeks of treatment, and did not find an association  probably not influence our results.
In conclusion, dupilumab-treated AD patients with moderate-tosevere OSD had higher dupilumab tear fluid levels compared to patients with no or mild OSD. This might be explained by a disrupted blood-tear barrier. Finally, dupilumab binding on conjunctival epithelial cells was found, indicating that dupilumab reaches the ocular surface. Further research is needed to investigate the implications of these increased levels of dupilumab.

AUTHOR CONTRIBUTIONS
All authors have made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data.
All authors have been involved in drafting the manuscript or revising it critically and have given final approval of the version to be published.

ACKNOWLEDGMENTS
This investigator initiated study was sponsored by Sanofi Genzyme.
The sponsor was not involved in the analyses, interpretation of the data, and preparation of the manuscript.