Noninvasive assessment of cytokine and antimicrobial peptide levels in hidradenitis suppurativa using transdermal analysis patches

dear editor, Most of the currently available cytokine and antimicrobial peptide (AMP) expression data from hidradenitis suppurativa (HS) lesions have been obtained through painful, invasive sampling techniques such as biopsies. ¹ In contrast, a Transdermal Analysis Patch (TAP, FibroTx, Tallinn, Estonia) captures and measures biomarkers at the skin surface and as such represents a novel, noninvasive and patient‐friendly diagnostic tool for skin biomarker detection. The aim of this study was to measure the levels of several cytokines and AMPs in both lesional and uninvolved skin of patients with HS and site‐matched skin of controls using TAPs.

Twenty patients with HS and 10 healthy control participants were included. TAPs contained capturing antibodies for interleukin (IL)‐1α, IL‐1RA, IL‐8, IL‐12, IL‐17A, IL‐23p19, C‐C motif chemokine ligand 27, kallikrein‐related peptidase‐5, human β‐defensin (hBD)‐1, hBD‐2, hBD‐3, hBD‐4 and a negative and IgG control. The patches were applied on inflamed lesions in the axilla and groin, on associated uninvolved skin, and on identical sites in controls. ² TAPs remained in place for 20 min, and were stored at –20 °C until further analysis by spot enzyme‐linked immunosorbent assay (ELISA). ³ Using GraphPad Prism (v. 8), differences were analysed with Wilcoxon signed‐rank tests or Mann–Whitney U‐tests. P‐values ≤ 0·05 were considered significant. Approval was granted by the institutional review board of the Erasmus University Medical Center (MEC‐2017‐1126).

Eighty per cent (16 of 20) of the included patients with HS were female, 12 had Hurley stage II or III and 13 patients were current smokers, the mean body mass index (BMI) was 30·7 ± 7·12. Eighty per cent (8 of 10) of controls were female and none were current smokers, with a mean BMI of 23·1 ± 3·3.

Significantly lower concentrations of IL‐1α were found in HS lesional axilla skin compared with uninvolved skin (P = 0·016); see Table 1 for fold changes (absolute levels are available on request). Lesional axillary and uninvolved upper arm skin in patients with HS showed significantly higher concentrations of IL‐17A than in controls (respectively, P = 0·010 and P = 0·009). In patients with HS and controls lower concentrations of IL‐23p19 were found in axilla and groin skin compared with uninvolved skin. Levels of hBD‐2 were significantly increased in lesional axilla and groin compared with uninvolved sites, respectively, P < 0·001 and P < 0·001. Significantly different concentrations of hBD‐1, hBD‐2 and hBD‐3 were seen between uninvolved skin of patients with HS and identical sites in controls.

Table 1.

Fold changes (FC) of inflammatory cytokine and antimicrobial peptides measured with transdermal analysis patches, showing significant results after Benjamini–Hochberg correction for multiple testing

	HS lesional axilla vs. HS uninvolved upper arm			HS lesional axilla vs. HC axilla			HS uninvolved upper arm vs. HC upper arm			HS lesional groin vs. HS uninvolved thigh			HS lesional groin vs. HC groin			HS uninvolved thigh vs. HC thigh
	FC		P‐value	FC		P‐value	FC		P‐value	FC		P‐value	FC		P‐value	FC		P‐value
IL‐1α	0·53	↓	0·017b	3·32		ns	0·99		ns	0·53		ns	1·02		ns	1·52		ns
IL‐1RA	0·48		ns	3·32	↑	0·027	0·70		ns	1·95		ns	15·20	↑	0·007	1·12		ns
IL‐8	20·91	↑a	0·003b	20·91	↑a	0·007	ND		–	11·01	↑a	0·003c	11·01	↑a	0·009	ND		–
IL‐12	ND		–	ND		–	ND		–	ND		–	ND		–	ND		–
IL‐17A	1·19		ns	16·59		0·011	9·22	↑	0·010	0·99		ns	2·25		ns	2·00		ns
IL‐23p19	0·69		ns	1·16		ns	1·18		ns	0·70	↓	0·027	0·92		ns	1·06		ns
CCL‐27	0·57		ns	1·00		ns	0·71		ns	1·00		ns	1·00		ns	1·00		ns
KLK‐5	1·47		ns	3·04		ns	1·85		ns	1·42		ns	2·80		ns	0·95		ns
hBD‐1	0·53	↓	0·006	0·50	↓	0·015	0·72		ns	1·34		ns	1·03		ns	0·53	↓	0·024
hBD‐2	5·43	↑	0·001	9·14	↑	0·022	1·68	↑	0·010	4·24	↑	< 0·001d	8·30	↑	< 0·001	1·96	↑	0·011
hBD‐3	1·10		ns	1·24		ns	1·00		ns	1·09		ns	1·26	↑	ns	1·06		ns
hBD‐4	ND		–	ND		–	ND		–	ND		–	ND		–	ND		–

CCL‐27, C‐C motif chemokine ligand 27; hBD, human β‐defensin; HC, healthy control; HS, hidradenitis suppurativa; IL, interleukin; KLK, kallikrein‐related peptidase; ND, not detected in both groups; ns, not significant; bold, significant (P ≤ 0·05).

^aAnalysed using half of the lower limit of detection for one of the groups where data was under limit of detection; ^bmissing, n = 3; ^cmissing, n = 2; ^dmissing, n = 1.

The IL‐1 pathway is known to be activated in HS, yet our study showed decreased levels of IL‐1α in HS lesional skin compared with uninvolved skin. Lower IL‐1α levels in lesional HS skin could reflect intracellular location and consumption of IL‐1α at sites of inflammation. ⁴ , ⁵ However, further investigation is needed to understand the complex regulation of IL‐1 in the skin and in HS.

In contrast to the elevated concentrations of hBD‐2 and hBD‐3 in HS lesional skin, the level of hBD‐1 was decreased compared with controls. Lower levels of hBD‐1 in lesional tissue have been described previously by Hotz et al. ⁶ In contrast to hBD‐2 and hBD‐3, expression of hBD‐1 is not induced by inflammatory stimuli such as tumour necrosis factor‐α, IL‐1β and interferon‐γ. ⁷ The mechanisms behind the decreased levels of hBD‐1 in HS pathogenesis remain elusive. The significantly higher levels of hBD‐2 in uninvolved HS skin compared with identical sites in controls is suggestive for an activated innate immune system in the uninvolved skin in patients with HS.

Relatively high concentrations of IL‐23p19 were found in all samples and no significant differences were seen between patients with HS and controls. The detection of increased levels of IL‐23p19 in lesional HS biopsies, which capture both epidermis and dermis, and the lack of differences in lesional skin compared with controls when measured with TAPs, might indicate that IL‐23 secreted by keratinocytes plays a lesser role in HS pathogenesis than IL‐23 released by dermal dendritic cells. ¹

The detection of biomarkers from the topmost layers of the skin could be seen as a limitation in HS as the deep dermal infiltrate, characteristic of the disease, is probably not fully captured by a TAP. Nevertheless, keratinocytes are active players in HS inflammation and assessing keratinocyte‐related inflammatory mediators in a noninvasive manner could provide new insights into their role in the pathogenesis of HS.

In conclusion, this is the first study assessing the expression of cytokines and AMPs using TAPs in patients with HS. We showed that TAPs can be used effectively to evaluate a wide range of HS‐associated biomarkers in the skin. Additionally, the noninvasive, patient‐friendly nature of this technique could prove to be of specific value for monitoring the inflammatory state of the skin over time both in clinical trials and daily practice. The results found in this study support a role for anti‐IL‐17A in the treatment of HS, but raise questions on the benefit of anti‐IL‐1α. Overall, our results confirm previously published data obtained with invasive techniques and highlight the role of the epidermis and subclinical inflammation in HS pathogenesis.

Author Contribution

Christine B. Ardon: Formal analysis (lead); Investigation (lead); Methodology (equal); Visualization (equal); Writing‐original draft (lead); Writing‐review & editing (supporting). Cong Wang: Formal analysis (supporting); Investigation (supporting). Errol P Prens: Conceptualization (lead); Methodology (equal); Supervision (equal); Writing‐review & editing (supporting). Kelsey R. van Straalen: Formal analysis (supporting); Methodology (equal); Supervision (equal); Visualization (equal); Writing‐review & editing (lead).