Natural moisturizing factor as a clinical marker in atopic dermatitis

Atopic dermatitis (AD) is a heterogeneous disease with various biological origins and clinical appearances. It is likely that different therapies or treatment intensities are not equally effective for all AD endotypes. The strongest genetic risk factor for AD is a null muta‐ tion in the filaggrin gene (FLG).1 Patients with eczema who carry a FLG null mutation are also prone to more persistent, severe eczema, and earlier onset of AD compared to patients without a FLG null mutation. Stratification of patients based on the FLG null endotype could enable more targeted treatment. Methods to determine FLG null mutations based on genotyping are time consuming and require specialized laboratory infrastructure, further complicated by the existence of over 50 different polymorphisms with widely varying prevalences between ethnic groups.2 In the stratum corneum (SC) filaggrin is enzymatically degraded into its constituting amino acids and their derivatives, together with specific salts and sugars collec‐ tively named natural moisturizing factor (NMF). Decreased NMF provides an accurate surrogate marker for the presence of FLG null polymorphisms.3 This can be measured rapidly and noninvasively by Raman spectroscopy in a clinically compatible test. We have assessed the potential of NMF as a novel clinical marker in AD by examining the association of clinically measured NMF val‐ ues with severity of AD, early onset of AD, and the co‐morbidities of AD: allergic sensitization, food allergy, bronchial hyperreactivity (BHR), asthma, and allergic rhinitis. Of 207 children with AD (0‐18 years of age), NMF values had been measured routinely during a visit to the pediatric atopy cen‐ ter KinderHaven‐Sophia Children's Hospital‐Erasmus MC University Medical Center Rotterdam in The Netherlands. The retrospective study protocol was approved by the medical ethics committee of Erasmus MC (MEC‐2016‐244). AD was diagnosed by a dermatologist according to the UK Working Party's Diagnostic Criteria for Atopic Dermatitis.4 NMF had been measured noninvasively on the palm of the hand by Raman spectroscopy using an in vivo Raman skin ana‐ lyzer (gen2‐SCA, RiverD International BV, Rotterdam). NMF values were classified as normal NMF (>1.14 arbitrary units) or decreased NMF (<0.995 arbitrary units), using a 0.07 confidence interval around the threshold of 1.07 as established by O’Regan et al.3 Patients with a NMF value between 0.995‐1.14 were excluded. The interval was the estimated 95% confidence interval, calculated as the standard error (SE) of the NMF value, averaged over the entire cohort, and multiplied by 1.96. Disease characteristics and comorbidity status were retrieved from the electronic medical patient files by two in‐ dependent researchers (see Appendix S1). Severity (mild to moderate or severe) of AD was measured by proxy of therapy based on the cri‐ teria as described by Wollenberg et al5 (Appendix S1). Associations between NMF status and the clinical parameters were tested by uni‐ variate and multivariate logistic regression models with adjustment for age and gender. Sixty‐seven out of 207 (32.4%) patients had decreased NMF. Figure 1 shows the distribution of disease severity in relation to the groups normal NMF and decreased NMF. Patients with decreased NMF had increased risk of severe AD, OR 2.12 (95% CI 1.02‐4.43), sensitization for food allergens, OR 2.27 (95%CI 1.21‐4.23), sensiti‐ zation for inhalation allergens, OR 2.22 (95%CI 1.13‐4.34), and food allergies, OR 2.79 (95% CI 1.33‐5.86; Table 1 and Table S1). Having decreased NMF did not show an association with early‐onset AD, allergic rhinitis, BHR, asthma and combined asthma, and/or BHR. In this retrospective study, we examined the associations between NMF values and the clinical parameters of the atopic syndrome. NMF