No Evident Systemic Terminal Complement Pathway Activation in Hidradenitis Suppurativa

Abstract

In hidradenitis suppurativa (HS) lesional skin, activation of both innate and adaptive immunity has been described, characterized by upregulation of a multitude of cytokines and complement components (van der Zee et al., 2011van der Zee H.H. de Ruiter L. van den Broecke D.G. Dik W.A. Laman J.D. Prens E.P. Elevated levels of tumour necrosis factor (TNF)-α, interleukin (IL)-1β and IL-10 in hidradenitis suppurativa skin: a rationale for targeting TNF-α and IL-1β.Br J Dermatol. 2011; 164: 1292-1298Crossref PubMed Scopus (332) Google Scholar). Higher levels of C5a and soluble C5b-9 (sC5b-9) in the plasma of patients with HS have been reported, indicating systemic late-phase complement pathway activation in HS (Kanni et al., 2018Kanni T. Zenker O. Habel M. Riedemann N. Giamarellos-Bourboulis E.J. Complement activation in hidradenitis suppurativa: a new pathway of pathogenesis?.Br J Dermatol. 2018; 179: 413-419PubMed Google Scholar). This formed the basis for two (ongoing) clinical trials—NCT03001622 and NCT03487276—targeting C5a and C5aR1 in patients with HS. In addition, deposition of complement components C1q, C3b, and C4d and enhanced expression of complement receptors and complement pathway genes have been found in HS lesions (Blok et al., 2016Blok J.L. Li K. Brodmerkel C. Jonkman M.F. Horváth B. Gene expression profiling of skin and blood in hidradenitis suppurativa.Br J Dermatol. 2016; 174: 1392-1394Crossref PubMed Scopus (36) Google Scholar; Gudjonsson et al., 2020Gudjonsson J.E. Tsoi L.C. Ma F. Billi A.C. van Straalen K.R. Vossen A.R.J.V. et al.Contribution of plasma cells and B cells to hidradenitis suppurativa pathogenesis.JCI Insight. 2020; 5e139930Crossref PubMed Scopus (39) Google Scholar; Hoffman et al., 2018Hoffman L.K. Tomalin L.E. Schultz G. Howell M.D. Anandasabapathy N. Alavi A. et al.Integrating the skin and blood transcriptomes and serum proteome in hidradenitis suppurativa reveals complement dysregulation and a plasma cell signature.PLoS One. 2018; 13e0203672Crossref PubMed Scopus (40) Google Scholar). The finding of systemic complement activation in HS is remarkable, because elevation of other markers of systemic complement activation, such as CRP and IL-6, is rare in HS. We therefore evaluated systemic complement activation in the plasma of 76 patients with HS and that of 10 controls (five males, five females) (Table 1) by quantifying the levels of C3, C3d, C5a, and sC5b-9.Table 1Patient Characteristics and Complement LevelsPatient Characteristics and MeasurementsTotal Population with HS (n = 76)Hurley Stage I (n = 23)Hurley Stage II (n = 37)Hurley Stage III (n = 16)Healthy Controls (n = 10)Patient characteristicsSex Female, n (%)45(59.2)15(65.2)21(56.8)9(56.3)5(50.0) Male, n (%)31(40.8)8(34.8)16(43.2)7(43.8)5(50.0)Age, mean (±SD)40.3(11.5)39.1(13.3)39.8(10.6)43.4(11.0)69.6(13.6)Age of onset, median (IQR)20.0(15.0‒29.8)22.0(14.0‒33.0)20.0(16.0‒28.5)17.5(15.0‒40.3)Disease duration, y, median (IQR)15.5(7.3‒25.8)15.0(4.0‒22.0)15.0(9.5‒26.0)20.5(10.3‒27.5)BMI, median (IQR)29.5(24.9‒35.9)31.1(25.4‒37.9)28.2(24.9‒35.0)29.6(24.2‒38.3) Missing, n6510Smoking status Current or ex-smoker, n (%)53(69.7)14(60.9)29(78.4)10(62.5) Never smoked, n (%)23(30.3)9(39.1)8(21.6)6(37.5)IHS4 Mild (≤3)26(52.0)15(75.0)9(45.0)2(20.0) Moderate (4‒10)14(28.0)5(25.0)7(35.0)2(20.0) Severe (≥11)10(20.0)04(20.0)6(60.0) Missing, n263176MeasurementsCRP (mg/l), median (IQR)6.0(3.2‒11.5)4.7(1.4‒6.5)5.6(3.1‒9.8)16.6(3.3‒31.9)C3 (mg/ml), median (IQR)1.4(1.2‒1.6)1.4(1.2‒1.7)1.3(0.6‒1.5)1.4(0.7‒1.7)1.3(1.2‒1.4)C3d (μg/ml), median (IQR)5.2(2.8‒8.1)7.0(5.8‒9.8)3.7(2.4‒6.3)3.8(2.7‒7.6)2.3(1.7‒3.0)C3d/C3 ratio, median (IQR)4.4(3.1‒7.1)5.2(4.0‒7.1)4.4(2.1‒6.5)4.3(2.4‒10.4)1.7(1.2‒2.6)sC5b-9 (ng/ml), median (IQR)58.0(43.0‒96.6)58.0(45.0‒81.0)61.0(40.5‒134.8)44.1(33.8‒90.2)59.7(30.2‒190.3)C5a (ng/ml), median (IQR)1.8(1.3‒2.3)1.9(1.6‒2.3)1.8(1.1‒2.3)1.8(0.8‒2.2)2.0(1.7‒3.2)Abbreviations: BMI, body mass index, HS, hidradenitis suppurativa; IHS4, International Hidradenitis Suppurativa Severity Score System; IQR, interquartile range; sC5b-9, soluble C5b-9. Open table in a new tab Abbreviations: BMI, body mass index, HS, hidradenitis suppurativa; IHS4, International Hidradenitis Suppurativa Severity Score System; IQR, interquartile range; sC5b-9, soluble C5b-9. Data and blood samples were collected after obtaining written informed consent from patients in the HiScreen Registry and Biobank at the Department of Dermatology of the Erasmus University Medical Center (Rotterdam, The Netherlands). The Institutional Review Board (Erasmus University Medical Center) approved the HiScreen Registry and Biobank (MEC-2016-246). C3d was measured by sandwich ELISA as previously described (Peakman et al., 1987Peakman M. Lobo-Yeo A. Senaldi G. Nilsson M. Tee D.E. Vergani D. Quantification of C3d in biological fluids by an enzyme-linked immunosorbent assay.J Immunol Methods. 1987; 104: 51-56Crossref PubMed Scopus (25) Google Scholar). Quantitative antigenic assay for C3 was performed by radial immunodiffusion technique with monospecific antisera (Fijen et al., 1996Fijen C.A. Kuijper E.J. Te Bulte M. van de Heuvel M.M. Holdrinet A.C. Sim R.B. et al.Heterozygous and homozygous factor H deficiency states in a Dutch family.Clin Exp Immunol. 1996; 105: 511-516Crossref PubMed Scopus (37) Google Scholar). Concentrations of sC5b-9 were determined by ELISA as described earlier (Damman et al., 2011Damman J. Seelen M.A. Moers C. Daha M.R. Rahmel A. Leuvenink H.G. et al.Systemic complement activation in deceased donors is associated with acute rejection after renal transplantation in the recipient.Transplantation. 2011; 92: 163-169Crossref PubMed Scopus (65) Google Scholar). C5a in plasma was quantified using a commercially available ELISA kit (HK349, Hycult, Uden, The Netherlands). See Supplementary Materials and Methods for details on data collection, reference values, and statistical analysis. The median plasma C3 concentrations for patients with HS and for controls were within the normal range (Table 1 and Figure 1). The median concentrations of C3 were statistically higher among patients with Hurley stage I than among controls (P = 0.031) and those with Hurley stage II (P = 0.024). A cluster of 15 patients was identified with C3 concentrations below the normal range (Figure 1). These patients were more frequently of Hurley stage II, but no other significant associations were found. Compared with controls, median C3d levels were significantly increased for each Hurley stage (P < 0.001; Hurley stage II, P < 0.001; Hurley stage III, P = 0.041; Table 1). C3d concentrations among patients with Hurley stage I were significantly higher than those among patients with Hurley stage II (P < 0.001) and those with Hurley stage III (P = 0.042). The C3d/C3 ratios were significantly higher in each Hurley stage relative to those in the controls. All concentrations of sC5b-9 and C5a were in the low to normal range and were comparable for controls and patients with HS. There were no significant differences in the concentrations of complement components between sexes or smoking status. The median CRP concentration was significantly higher in patients with Hurley stage III than in those with Hurley stage I (P = 0.04). The CRP levels significantly correlated with concentrations of C3, C3d, and C3d/C3 ratio (P = 0.002; P = 0.001, and P = 0.019, respectively; Table 1). Our study reveals that circulating levels of C5a and sC5b-9 in patients with HS are not significantly increased relative to those in controls. Although concentrations of C3 and C3d as well as those of C3d/C3 ratio were increased in patients with HS, these were all within normal range, except for C3d. However, plasma C3d levels are highly dependent on the concentration of the C3 parent molecule. Therefore, to accurately interpret the C3d levels, C3d should always be additionally reported as a C3d/C3 ratio (Mollnes et al., 2007Mollnes T.E. Jokiranta T.S. Truedsson L. Nilsson B. Rodriguez de Cordoba S. Kirschfink M. Complement analysis in the 21st century.Mol Immunol. 2007; 44: 3838-3849Crossref PubMed Scopus (97) Google Scholar). This ratio is a more robust marker of complement activation because C3d has a half-life of hours in contrast to C5a and sC5b-9, which have half-lives of minutes. C3d profoundly influences acquired immune responses through the C3d receptor (CD21) expressed on B lymphocytes and follicular dendritic cells (Merle et al., 2015Merle N.S. Noe R. Halbwachs-Mecarelli L. Fremeaux-Bacchi V. Roumenina L.T. Complement system part II: role in immunity.Front Immunol. 2015; 6: 257Crossref PubMed Scopus (520) Google Scholar). Hence, C3d could contribute to B-cell activation in HS and could potentially aid in the production of pathogenic (auto)antibodies (Byrd et al., 2019Byrd A.S. Carmona-Rivera C. O’Neil L.J. Carlucci P.M. Cisar C. Rosenberg A.Z. et al.Neutrophil extracellular traps, B cells, and type I interferons contribute to immune dysregulation in hidradenitis suppurativa.Sci Transl Med. 2019; 11: 5908Crossref Scopus (51) Google Scholar; Gudjonsson et al., 2020Gudjonsson J.E. Tsoi L.C. Ma F. Billi A.C. van Straalen K.R. Vossen A.R.J.V. et al.Contribution of plasma cells and B cells to hidradenitis suppurativa pathogenesis.JCI Insight. 2020; 5e139930Crossref PubMed Scopus (39) Google Scholar). In contrast to the study by Kanni et al., 2018Kanni T. Zenker O. Habel M. Riedemann N. Giamarellos-Bourboulis E.J. Complement activation in hidradenitis suppurativa: a new pathway of pathogenesis?.Br J Dermatol. 2018; 179: 413-419PubMed Google Scholar, we did not identify elevated systemic levels of C5a in patients with HS. C5a is difficult to detect in vivo because it is biologically highly active and binds rapidly to C5a receptors (Oppermann and Götze, 1994Oppermann M. Götze O. Plasma clearance of the human C5a anaphylatoxin by binding to leucocyte C5a receptors.Immunology. 1994; 82: 516-521PubMed Google Scholar; Wagner and Hugli, 1984Wagner J.L. Hugli T.E. Radioimmunoassay for anaphylatoxins: a sensitive method for determining complement activation products in biological fluids.Anal Biochem. 1984; 136: 75-88Crossref PubMed Scopus (124) Google Scholar). However, if an increase in C5a was masked by this rapid binding, increased concentrations of sC5b-9 could serve as a proxy for elevated levels of C5a. Yet, we did not find increased levels of sC5b-9. Remarkably, plasma levels of C5a and sC5b-9 in controls were 4‒10 times higher than those found in our study (Kanni et al., 2018Kanni T. Zenker O. Habel M. Riedemann N. Giamarellos-Bourboulis E.J. Complement activation in hidradenitis suppurativa: a new pathway of pathogenesis?.Br J Dermatol. 2018; 179: 413-419PubMed Google Scholar). One other study reported low to normal levels of C5a and C5b-9 in patients with HS (Li et al., 2020Li C, Shah A, Ebsworth K, Dunlap C, Kilgour J, Sullivan K, et al. Complement activation and C5aR elevation in hidradenitis suppurativa patients lesions. Paper presented at: Symposium on Hidradenitis Suppurativa Advances SHSA Conference. 10–11 October 2020; Montreal, Canada.Google Scholar). These substantial differences are likely caused by the use of different ELISA systems. Therefore, comparing ELISA results of complement levels should be done with caution because of the different properties of the capture and detection antibodies used. Cross-reactivity with nonactivated but neoepitope-expressing C5 native protein, artificially increasing the levels of C5a, should be ruled out (Nilsson et al., 2017Nilsson P.H. Thomas A.M. Bergseth G. Gustavsen A. Volokhina E.B. van den Heuvel L.P. et al.Eculizumab-C5 complexes express a C5a neoepitope in vivo: consequences for interpretation of patient complement analyses.Mol Immunol. 2017; 89: 111-114Crossref PubMed Scopus (12) Google Scholar). The ELISA kit used in our study recognizes a neoepitope on C5a desArg not present in the native C5 component, preventing such cross-reactivity. The finding of complement deposition in HS lesions with enhanced expression of complement receptors and complement pathway genes without systemic spillover is puzzling (Blok et al., 2016Blok J.L. Li K. Brodmerkel C. Jonkman M.F. Horváth B. Gene expression profiling of skin and blood in hidradenitis suppurativa.Br J Dermatol. 2016; 174: 1392-1394Crossref PubMed Scopus (36) Google Scholar; Gudjonsson et al., 2020Gudjonsson J.E. Tsoi L.C. Ma F. Billi A.C. van Straalen K.R. Vossen A.R.J.V. et al.Contribution of plasma cells and B cells to hidradenitis suppurativa pathogenesis.JCI Insight. 2020; 5e139930Crossref PubMed Scopus (39) Google Scholar; Hoffman et al., 2018Hoffman L.K. Tomalin L.E. Schultz G. Howell M.D. Anandasabapathy N. Alavi A. et al.Integrating the skin and blood transcriptomes and serum proteome in hidradenitis suppurativa reveals complement dysregulation and a plasma cell signature.PLoS One. 2018; 13e0203672Crossref PubMed Scopus (40) Google Scholar). It is still unclear whether local complement activation also comprises terminal pathway activation and, if so, why it remains strictly compartmentalized in inflamed HS skin. Therefore, further in-depth research on in situ complement activation, in combination with functional studies, is warranted. In conclusion, our results do not support a prominent role for systemic complement activation in the pathogenesis of HS. The absence of evident systemic and in situ terminal pathway activation argues against the potential benefits of therapeutic intervention in the C5a‒C5aR1 axis, the therapeutic target of two (ongoing) clinical trials in HS. Datasets related to this article are available on request. Lisette M. Prens: http://orcid.org/0000-0002-3751-684X Christine B. Ardon: http://orcid.org/0000-0002-2303-4604 Kelsey R. van Straalen: http://orcid.org/0000-0003-3305-3814 Hessel H. van der Zee: http://orcid.org/0000-0002-2874-7726 Marc A.J. Seelen: http://orcid.org/0000-0003-2240-4682 Jon D. Laman: http://orcid.org/0000-0001-5085-9807 Errol P. Prens: http://orcid.org/0000-0002-8158-660X Barbara Horváth: http://orcid.org/0000-0001-8559-3674 Jeffrey Damman: http://orcid.org/0000-0001-5997-7551 EPP declares grant from AbbVie paid to Erasmus University Medical Center and advisory board membership. BH declares fees from Janssen-Cilag (advisory boards, educational grants, consultations, investigator initiative studies), AbbVie (advisory boards, educational grants, consultations, investigator initiative studies), Novartis Pharma (advisory boards, consultations, investigator initiative studies), UCB Pharma (advisory boards, consultations), Leo Pharma (consultations), Solenne B.V. (investigator initiative studies), Celgene (consultations, investigator initiative studies), Akari Therapeutics (consultations, investigator initiative studies), Philips (consultation), Roche (consultation), Regeneron (consultation), and Sanofi (consultation), for which fees were paid to the institution. The remaining authors state no conflict of interest. The authors thank Anita Meter (University Medical Center Groningen, The Netherlands) for performing the complement measurements. We thank the patients with HS and the healthy volunteers for their participation. The HiScreen Registry and Biobank are financially supported by AbbVie (The Netherlands). The collection of blood and patient data used in this study was approved by the Institutional Review Board of the Erasmus University Medical Center (Rotterdam, The Netherlands) (MEC-2016-246/MEC-2013-337). Conceptualization: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, BH, JD; Data Curation: LMP, CBA, KRVS, MAJS, EPP, JD; Formal Analysis: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, JD; Funding Acquisition: MAJS, EPP, BH, JD; Investigation: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, BH, JD; Methodology: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, BH, JD; Project Administration: LMP, CBA, KRVS, MAJS, EPP, BH, JD; Resources: LMP, CBA, KRVS, MAJS, EPP, BH, JD; Supervision: MAJS, EPP, JD; Validation: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, BH, JD; Visualization: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, BH, JD; Writing - Original Draft Preparation: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, BH, JD; Writing - Review and Editing: LMP, CBA, KRVS, HHVDZ, MAJS, JDL, EPP, BH, JD AbbVie (Hoofddorp, The Netherlands) had no role in the design or conduct of the study, the interpretation of the data, or the decision to submit the manuscript for publication. Data were collected from patients participating in the HiScreen Registry and Biobank at the Department of Dermatology of the Erasmus University Medical Center (Rotterdam, The Netherlands). The HiScreen Registry and Biobank were approved by the Institutional Review Board of the Erasmus University Medical Center (MEC-2016-246). This registry collects data on age, sex, body mass index, smoking status, and disease severity (Hurley stage, abscess and nodule counts, and International Hidradenitis Suppurativa Severity Score System). Routine blood samples were collected at the first visit in 8-ml vacuum EDTA tubes (BD Vacutainer, Plymouth, UK) under aseptic conditions. Directly after collection, blood was centrifuged to separate the plasma, which was subsequently aliquoted and stored at –80 °C until further analysis. Concentrations of the complement components C3, C3d, and C5a and the membrane attack complex soluble C5b-9 were measured in the laboratory of the Department of Nephrology, University Medical Center Groningen (Groningen, The Netherlands). The C3d/C3 ratio was determined by dividing the C3d values in μg/ml by the C3 concentration in mg/ml. Soluble C5b-9 concentrations were determined by in-house sandwich ELISA as described earlier (Damman et al., 2011Damman J. Seelen M.A. Moers C. Daha M.R. Rahmel A. Leuvenink H.G. et al.Systemic complement activation in deceased donors is associated with acute rejection after renal transplantation in the recipient.Transplantation. 2011; 92: 163-169Crossref PubMed Scopus (68) Google Scholar). Reference values were based on a normal distribution of a historical cohort of 33 healthy volunteers for C3d and 16 healthy volunteers for soluble C5b-9. Blood samples of this cohort was collected, processed, and analyzed identically to the blood of controls and patients with hidradenitis suppurativa in this study. All values within three SDs from the mean of these historical control populations were deemed within the normal range. CRP concentrations were measured in a subgroup of patients with hidradenitis suppurativa (Hurley stage I, n = 10; Hurley stage II, n = 10; Hurley stage III, n = 10) in the general laboratory of the Erasmus University Medical Center (Rotterdam, The Netherlands) by turbidimetric analysis, with a cut-off value of 5 mg/l. Patient characteristics are presented as numbers (percentages) for categorical variables and as mean ± SD or median (interquartile range) where appropriate for continuous variables. Concentrations of C3, C3d, C5a, and soluble C5b-9 and CRP are expressed as median (interquartile range). Normality was assessed using the Kolmogorov‒Smirnov test. Differences in plasma levels between Hurley stages, sex, body mass index category, and smoking status were assessed using univariate logistic regression analysis. All statistical tests were two-sided, and P ≤ 0.05 was considered statistically significant. Statistical analysis was performed using IBM Statistical Package for the Social Sciences Statistics for Windows, version 25, 2017 (IBM, Armonk, NY).