Keratosis pilaris and filaggrin loss‐of‐function mutations in patients with atopic dermatitis – Results of a Finnish cross‐sectional study

Abstract Keratosis pilaris (KP) associates with epidermal barrier defects in atopic dermatitis (AD) but its role in disease severity and concomitant atopic diseases seems to vary between populations. We performed a cross‐sectional observational study with 502 randomly selected AD patients of a Finnish tertiary health care center. At a single clinical examination, disease severity (Rajka Langeland severity score and EASI), clinical signs and patient history were evaluated and total IgE levels and frequent filaggrin (FLG) loss‐of‐function mutations were investigated. There was no link with disease severity (p = 0.649, 95% CI 0.569–0.654), asthma (p = 0.230, 95% CI 0.206–0.281) or atopic sensitization (p = 0.351, 95% CI 0.309–0.392). Keratosis pilaris was significantly associated with palmar hyperlinearity (p < 0.000, 95% CI 0.000–0.006, OR 4.664, 95% CI 2.072–10.496) and the filaggrin loss‐of‐function mutation 2282del4 (p < 0.000, 95% CI 0.000–0.009, OR 4.917, 95%CI 1.961–12.330). The prevalence of KP in the cohort was generally low and KP seems to be infrequent in Finnish AD patients. This may be explained by the fact that the tested FLG loss‐of‐function mutations are rarer in the Finnish population compared for example, with central Europe or Asia. Mutations in other locations of the FLG gene or other genes of the epidermal barrier may play a more important role.

inclusion criterion was AD. During a single clinical examination, in each patient, disease severity (Rajka Langeland severity score [RLS] and EASI), clinical signs (KP, palmar hyperlinearity) and patient history were evaluated. Data on disease onset, hereditary factors regarding first-degree relatives, contact allergy (confirmed by patch testing), prick positivity (positivity to any of following: birch, timothy, mug wort, cat, dog, horse, house dust mite and Cladosporium herbarum), peanut allergy (confirmed by allergy tests and concomitant symptoms) and atopic comorbidities were based on the patient history. At the clinical visit, we investigated the total serum IgE-levels and a panel of FLG null-mutations. Genetic testing was carried out by sequencing four FLG LoF mutations that are frequent in the European population (R501X, 2282del4, R2447X, S3247X), two FLG LoF mutations enriched in the Finnish population (S1020X, V603M), and the 12-repeat allele (rs12730241). In addition, we sequenced 59 functional variants of 10 different genes associated with skin barrier defects (e.g., filaggrin-2). The genetic testing, sequencing methods and complete list of genotyped variants have been described in detail in a former study. 4 All laboratory tests were carried out in the Laboratory of Helsinki University Hospital according to accredited methods. The ethics committee of the University Hospital Helsinki, Finland approved the study protocol.

| RE SULTS
There was a relatively low number of patients with KP in the investigated cohort (28 KP vs. 319 with no KP). KP was not associated with AD severity based on EASI at the clinical visit (p = 0.3232, 95% CI 0.276-0.357) and RLS (p = 0.649, 95% CI 0.569-0.654), representing severity of the past year. In addition, there were no significant links to total serum IgE levels, early AD onset or positive history of hand dermatitis, contact allergy, asthma, prick positivity to frequent aeroallergens or peanut allergy (Table 1, Figure 1). There were no TA B L E 1 Patient characteristics, AD severity parameters and FLG loss-of-function mutations

| DISCUSS ION
The study shows that KP in Finnish AD patients may be associated with the FLG LoF mutation 2282del4 and the clinical sign of palmar hyperlinearity. 5 There have been observations that link KP with asthma, atopic sensitization, hand dermatitis and contact allergies based on mechanistic studies and increased epicutaneous penetration of allergens. [6][7][8] We did not observe similar effects in our cohort although the prevalence of atopic comorbidities (e.g., asthma) was relatively high in the studied Finnish AD patients. In addition, there were no indications that KP represents a marker of AD severity or early onset in Finnish patients which has been discussed by some authors. 9 Limitations of the study were the generally low number of patients with KP and the cross-sectional design which represents only a time point in the patients' history. In addition, the data was mostly based on retrospective patient-derived information and there was only a limited panel of FLG mutations tested. We also did not measure blood eosinophil levels or other possibly laboratory values relevant in atopic dermatitis, because we wanted to concentrate on FLG-mutations and clinical characteristics.
Former studies have underlined the genetic heterogeneity of atopic dermatitis and its association with KP. 3

ACK N OWLED G M ENTS
In memory of Professor Sakari Reitamo, our teacher, colleague, and friend.
We would like to thank medical statistician Olavi Koivisto, University of Helsinki, Finland, for his very valuable support.

CO N FLI C T O F I NTE R E S T
The authors state no conflicts of interest.