Topical Doxycycline Monohydrate Hydrogel (NANODOX®) 1% targeting proteases/PAR2 pathway is a novel therapeutic for Atopic Dermatitis.

Abstract

Atopic Dermatitis (AD) is characterized by skin barrier disruption and an aberrant immune response. Doxycycline are tetracycline antibiotics broadly used systemically to treat inflammatory-dermatologic conditions. Several studies have shown doxycycline have anti-inflammatory and pro-healing properties, mainly by blocking tissue proteolytic activity. It is our hypothesis that daily application of a novel Doxycycline topical formulation in AD subjects will reduce severity of the disease, by blocking cutaneous proteases activity and restoring skin barrier function and inflammation.

To test this hypotehsis we performed a proof of concept, open-label clinical study. Subjects enrolled in the study (n=15) applied NanoDOX® Hydrogel 1% daily for four weeks on a chosen eczematous area. Investigational drug was well tolerated, no local or systemic adverse events due to investigational drug were reported. Notably, a significant clinical improvement was observed based on a modified Eczema Area & Severity Index (EASI) score of the treated area from start of treatment to 14 and 28 days post treatment (p<0.001). A significant improvement of pruritus was also observed (p=0.02). This proof of concept clinical trial is first to explore the impact of a non-systemic doxycycline treatment on AD patients. Our results provide evidence to investigate novel AD treatment strategies targeting cutaneous proteases activity.

Clinical trial Registration: NCT02910011

Background

Atopic Dermatitis (AD) is a chronic and recurring inflammatory skin disease, affecting about 17% of children and 6% adults in the USA¹. AD is characterized by eczematous rash, xerosis, pruritus and recurrent Staphylococcus aureus infection. The intense pruritus and cutaneous infections contribute to the morbidity of AD and are major drivers of the reduced quality-of-life associated with this disease^2–4. The pathogenesis of AD is complex and multifactorial, explained by two complementary hypotheses: 1) inside out, the immune system defect drives the skin damage, and 2) outside in, in which the skin barrier defect drives the aberrant immune response^1,5. In either case, there is an active cross talk between skin barrier and immune system; thus feeding a vicious cycle that supports the chronicity of the disease and the need to target both the skin barrier and the immune system to ensure optimal treatment of the disease.

Premises

Dermatologists have widely used systemic tetracyclines since the 1950s to treat cutaneous inflammatory disorders due to the anti-inflammatory and anti-collagenase properties (reviewed in ⁶). Although the non-antibiotic mechanisms of tetracycline/doxycycline remain to be fully elucidated, several targets potentially relevant to AD have been identified (Figure 1A). A growing body of evidence suggests that tetracyclines downregulate Protease Activator Receptor (PAR)-2 expression and function directly or indirectly, by inhibiting the proteolytic activation of trypsin like serine proteases ^7–10. Activation of PAR2 has been associated with inflammation, altered epidermal barrier function and pruritus¹⁰. Several studies have reported increased protease activity in AD skin^11,12. An emblematic example of the role of proteases and PAR2 in AD is seen in Netherton Syndrome (NS). NS is a rare autosomal recessive monogenetic disease due to loss-of-function mutation in SPINK5 (serine proteases inhibitor of kazal 5) that encodes for protein named LEKTI-1 (lympho-epithelial kazal type 5). The lack of LEKTI results in unbalanced protease activities in the epidermis, mainly KLK5 and 7. Activation of KLK5 and 7 induce direct and indirect (via PAR2 signaling) skin barrier impairment and release of pro-inflammatory cytokines (e.g. TSLP, IL-8)¹³. Research with engineered murine models has shown that ablation of KLK5 and/or KLK7 rescued the skin barrier abnormality and prevents skin inflammation in LEKTI-deficient mice^14,15. In addition, Doxycycline has been reported to non-selectively inhibit members of the metalloproteinase (MMP) superfamily, and so preventing tryptic kallikreins (KLK) activation by MMPs.

Figure 1:

A) Summary of doxycycline potential activities relevant to AD pathogenesis. B) EASI target (treatment area) trend for each subject at each visit. Subject 3 withdraw at V3 due to AD flare requiring systemic treatment. Subject 4 withdraw after V1 and is not included. Subject 10 and 14 were clear or almost clear at V2 (end of treatment). C) KLK5 was measured in SC extracted by ELISA and normalized to total protein. D) total proteins were eluted from each tape and concentration was measured for each pool (5 tapes each) to evaluate the amount of total protein stripped over time. Data are shown as MEAN ± SEM, n=9 (5 subjects were eliminated from the analysis due to quality of extract in at least one of the time points).

Hypothesis

It is our hypothesis that daily application of a novel Doxycycline topical formulation (NanoDOX® hydrogel) blocking cutaneous proteases activity in AD subjects will reduce severity of the disease, by restoring skin barrier function and reducing skin driven inflammation.

How to test the hypothesis

We have performed a proof of concept clinical trial to investigate safety and efficacy of a novel small particle gel suspension of doxycycline 1% (NanoDOX® Hydrogel, topical Doxycycline Monohydrate, Alchem Laboratories Corp). While systemic tetracyclines have been used for cutaneous inflammatory disorders, topical doses increase local concentrations of the drugs whilst minimalizing the consequences of prolonged systemic antibiotic use and reducing off-target effects.

Preliminary supporting evidence

The study was conducted in the Department of Dermatology, University of Florida (Gainesville, FL) and approved by UF Institutional Review Board (IRB#201601657; NCT02910011). The topical 1% doxycycline monohydrate hydrogel was available for clinical testing through an FDA-cleared IND. Previously, the topical formulation was assessed for safety and efficacy in a controlled trial of chronic, diabetic, lower-extremity foot ulcers. All four chronic ulcers were treated daily and healed with no adverse events¹⁶.

Present study: Table 1 summarizes study population. Of the 15 subjects enrolled, 2 did not complete the study due to AD flare needing systemic steroid. Subjects that were on any topical prescriptions were asked to discontinue the use of those medications in the selected area for one week before starting the study. Subjects were required to apply 1% NanoDOX® Hydrogel, dispensed in single-dose packages, daily at bedtime for four weeks (or until complete clearance) on the chosen target eczematous and adjoining clinically normal (perilesional) area to cover an area of about 4×4 cm. We included non-lesional/perilesional skin to evaluate the safety (e.g. irritant or toxic effects) of the compound on atopic skin, as those subjects have a lower irritancy threshold as compared to non-atopic subjects¹⁷. During the study, subjects were allowed to stay on anti-histamine and topical treatment (excepted for the treatment areas) on a stabilized dose, and were expected to maintain this dose throughout the study.

Table 1.

Demographics of subjects enrolled in this study

D count	Age (in years)	Gender	Swab (V1- Presence of S. aureus)
1	63	Male	1+ S. aureus
2	28	Female	No
3	20	Female	few S. aureus
5	48	Female	No
6	30	Female	No
7	36	Female	No
8	21	Female	1+ S. aureus
9	28	Male	few S. aureus
10	25	Female	few S. aureus
11	34	Male	No
12	22	Female	few S. aureus
13	65	Female	3+ S. aureus
14	22	Female	No
15	52	Female	No

Safety was assessed at each visit and included the incidence and severity of treatment‐emergent adverse events (TEAEs). The presence or absence of skin irritation, redness, atrophy and telangiectasia were noted at each study visit. Subject’s disease status was assessed by a modified Eczema Area and Severity Index (EASI), focused only to the target area, which is the sum of the intensity scores (0 to 3) for four signs: erythema, edema/papulation, excoriation, lichenification¹⁸. Additionally, subjects’ self-assessments of their degree of disease and itching severity control was collected at each visit (5-D Pruritus Scale¹⁹). A skin swab was collected to obtain a semi-quantification of skin bacteria pool, including S. aureus spp. at baseline (V1; Clinical microbiology laboratory at the University of Florida, Rocky Point Lab). Six of fifteen subjects at baseline resulted positive for S. aureus at V1 (Table 1). For exploratory outcomes, repeated tape strippings (Cuderm™ tape) of treated area were collected at each visit. Stratum corneum (SC) cohesion was assessed by the amount of removed protein per five consecutive tapes up to 15 tapes (Micro BCA Protein Assay Kit; Thermo Fisher). KLK5, IL-8, KLK7 and MMP content in the SC extracts was measured by ELISA (Quantikine kit; R&D Systems). Additional methods information in the Supplemental material.

Daily application of 1% NanoDOX® Hydrogel was safe and well tolerated in AD subjects.

No TEAEs were recorded during the study. Only in one subject a transitory light-yellow discoloration was observed on the treatment area; however, it was completely resolved after termination of the study at the 2 weeks follow up. Subjects did not report itch or pain at the site of application, including non-lesional skin. One subject required the use of rescue medication (topical triamcinolone) for 4 days due to a mild AD flare.

Clinical improvement was observed after treatment with 1% NanoDOX® Hydrogel.

We explored the change of EASI target score from pre (V1) to post 14 +/− 3 days (V2) and 28 +/− 3 days (V3) treatment. Except for 2 subjects, all others (n=11) reached at least 50% EASI target improvement from V1 to V2 or V3 (Figure 1B). We then fit a linear model to check whether the improvement of EASI target score was significant. The expected improvements (coefficients, 95% CI) of EASI target from V1 to V2 and V1 to V3 were 4.43 (−5.82, −3.04; p<0.001) and 4.53 (−5.98, −3.07; p<0.001) respectively.

Exploratory Outcomes

Clinically, itch is one of the symptoms that negatively affects patient quality of life. Proteases and activation of PAR2, as well as Th2 inflammation, have been shown to be involved in itch pathways (reviewed in ²⁰). To test the effect of treatment on itch, we utilized proportional odds model to explore the significance of improvement of Itch from V1 to V2 or V3. We observed a significant improvement of itch, based on the 5-D Pruritus Scale, from V1 to V3 (p=0.02).

Interestingly, we observed a reduction of total protein extracted by tape stripping from V1 to V3 (Figure 1C; not statistically significant), thus suggesting an increased in Stratum Corneum (SC) cohesion after 4 weeks treatment. KLK5 is the most abundant in the skin²¹ and increased in the epidermis of patients with AD¹¹. Here we observed a decreasing trend of KLK5 (although not statistically significant) from V1 to V3 (Figure 1D). IL-8, KLK7 and MMP9, were below the detection threshold in over 50% of samples and not further analyzed.

Relevance and perspectives

This proof of concept study is the first to explore the impact of a non-systemic Doxycycline treatment on subjects with mild-moderate AD. No toxicity or safety concerns were reported with the test treatment in our study population. Results from this study showed a significant clinical improvement and reduced itchiness of the treatment areas, as well as a trend in reduction in KLK5 concentration and improved SC cohesion after treatment. Using in silico modeling we investigated Doxycycline docking with plasma KLK and found significant binding to the catalytic pocket (data not shown). These data support a plausible mechanism of the treatment’s effects being through the protease KLK5, but more rigorous experimentation is necessary to confirm. Also, despite the postulated role of S. aureus in AD severity, our patients had clinical improvements independent of the presence of S. aureus before treatment.

Overall, this study is limited due to its small sample size and lack of a placebo arm, but this is by design a proof of concept study. The data are encouranging in that there were definite improvements and a lack of adverse effects or tolerabilty concerns. These findings pave the way to larger studies to investigate novel AD treatment strategies with topical doxycycline and investigations into a hypothetical PAR2 and KLK5-mediated mechanism.

Abbreviations

AD

Atopic dermatitis

EASI

Eczema Area & Severity Index

KLK

Kallikrein-related peptide

LEKTI-1

Lympho-epithelial kazal type 5

NS

Netherton Syndrome

PAR

Protease Activator Receptor

SC

Stratum corneum

SPINK5

Serine proteases inhibitor of kazal 5

TEAEs

Treatment-emergent adverse events