Anti-Saccharomyces cerevisiae IgG and IgA antibodies are associated with systemic inflammation and advanced disease in hidradenitis suppurativa

Abstract

Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease that affects 1% to 4% of the Western population and is associated with major impact on quality of life. Its pathogenesis, combining hair follicle dysregulation and both skin and systemic inflammation, involves a genetic background interacting with environmental factors (eg, bacterial skin dysbiosis, overweight, smoking).1Frew J.W. Vekic D.A. Woods J. Cains G.D. A systematic review and critical evaluation of reported pathogenic sequence variants in hidradenitis suppurativa.Br J Dermatol. 2017; 177: 987-998Crossref PubMed Scopus (49) Google Scholar TNF-α, which is enhanced in HS, is upregulated in other immune-mediated inflammatory diseases (IMIDs) such as severe psoriasis vulgaris (PV), inflammatory bowel disease (IBD), and spondyloarthritis.2Vossen A.R.J.V. van der Zee H.H. Prens E.P. Hidradenitis suppurativa: a systematic review integrating inflammatory pathways into a cohesive pathogenic model.Front Immunol. 2018; 9: 2965Crossref PubMed Scopus (68) Google Scholar Evidence for an increased prevalence of IBD and spondyloarthritis in patients with HS raised the issue of the commonalities across these IMIDs.3Egeberg A. Jemec G.B.E. Kimball A.B. Bachelez H. Gislason G.H. Thyssen J.P. et al.Prevalence and risk of inflammatory bowel disease in patients with hidradenitis suppurativa.J Invest Dermatol. 2017; 137: 1060-1064Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar Recently, serologic detection of anti–Saccharomyces cerevisiae antibodies (ASCAs) was identified as a biomarker for Crohn disease4Quinton J.F. Sendid B. Reumaux D. Duthilleul P. Cortot A. Grandbastien B. et al.Anti-Saccharomyces cerevisiae mannan antibodies combined with antineutrophil cytoplasmic autoantibodies in inflammatory bowel disease: prevalence and diagnostic role.Gut. 1998; 42: 788-791Crossref PubMed Scopus (518) Google Scholar and in spondylarthropathy.5Maillet J. Ottaviani S. Tubach F. Roy C. Nicaise-Rolland P. Palazzo E. et al.Anti-Saccharomyces cerevisiae antibodies (ASCA) in spondyloarthritis: prevalence and associated phenotype.Joint Bone Spine. 2016; 83: 665-668Crossref PubMed Scopus (19) Google Scholar Therefore, we prospectively investigated the seroprevalence of ASCAs in patients with HS and with severe PV, respectively. A multicentric, national transversal study was conducted between 2012 and 2017 in 3 major French Dermatology tertiary centers for skin IMIDs (Saint-Louis Hospital Paris and university hospitals in Lille and Besançon). The patients were consecutive adult (aged ≥18 years) outpatient candidates for systemic therapy who presented with HS or PV (see Fig E1 in this article’s Online Repository at www.jacionline.org). The exclusion criteria were a history of IBD or any systemic immunomodulatory or immunosuppressive treatment. All of the patients with HS were candidates for systemic therapy (antibiotics or TNF inhibitors), whereas the patients with PV were candidates for conventional or biologic therapy approved in France. Sera from healthy blood donors were collected (Etablissement Français du Sang, Hôpital Saint-Louis) during the same time period, after patients had filled out a questionnaire so that individuals with a personal history of psoriasis, HS, IBD, or systemic immunomodulatory or immunosuppressive treatment could be excluded (see the Methods section in this article’s Online Repository at www.jacionline.org). The characteristics of the 469 individuals are summarized in Table E1 (available in this article’s Online Repository at at www.jacionline.org). The patients with PV had a longer median duration of disease compared with the patients with HS (19 vs 9 years); their median Psoriasis Area Severity Index score at the time of inclusion was 10.2 (range 7.7-14.5). On the other hand, 38 of the patients with HS (25.7%) were Hurley stage I, (ie, with mild disease), whereas the 110 others were either stage II (n = 55 [37.2%]) or stage III (n = 55 [37.2%]). Therefore, 74.4% of the patients with HS had moderate-to-severe disease at the time of inclusion, which is in line with the recruitment of tertiary care centers. A positive result of serologic detection of ASCA IgG and/or IgA was significantly more frequent in the HS group than in either the PV or HC groups, the rates of which did not differ significantly (24.3% [17.4%-31.2%] vs 4.4% [1.2%-7.6%] and 4.3% [1.2%-7.5%], respectively; P < 10-4 for both comparisons) (Fig 1, A). This association, which was evident across all Hurley stages, was more frequent in advanced stages (see Table E2 in this article’s Online Repository at www.jacionline.org). Among the patients with HS, IgA isotype seroprevalence tended to be higher than IgG isotype seroprevalence (16.9% [11.6%-23.9%] vs 12.9% [8.3%-19.3%]; P = .33), whereas the IgA+/IgG– pattern was observed more frequently than the IgG+/IgA– pattern among patients with HS showing a positivity for 1 single isotype (11.5% [7.2%-17.8%] vs 7.4% [4.1%-13.0%]; P = .23), but the results were not significant (Fig 1, B). In univariate analysis, detection of ASCAs within the HS group was significantly associated with the absence of active smoking, inflammatory rheumatic disease, HS-related autoinflammatory syndrome, and Hurley stage III (Table I). In addition, absence of ASCAs was significantly associated with a C-reactive protein plasmatic level lower than or equal to 5 mg/L (Table I). All 10 patients with HS-related autoinflammatory syndromes, 5 of whom were previously described,6Gottlieb J. Madrange M. Gardair C. Sbidian E. Frazier A. Wolkenstein P. et al.PAPASH , PsAPASH and PASS autoinflammatory syndromes: phenotypic heterogeneity, common biological signature and response to immunosuppressive regimens.Br J Dermatol. 2019; 18: 866-869Crossref Scopus (18) Google Scholar had serologic detection of ASCAs. Of note, a personal history of antibiotic therapy did not influence ASCA prevalence in patients with HS. In multivariate analysis, factors associated with serologic detection of ASCAs in HS were Hurley stage III and C-reactive protein serum level above 10 mg/L (see Table E3 in this article’s Online Repository at www.jacionline.org). During a median time of 17 months (5.2-47.3), 49 patients with HS (34%) were followed up, especially for any sign of IBD or spondyloarthritis, and none of them experience new development of any other inflammatory disease. Of these patients, 12 (24%) had serologic detection of ASCAs. In a subgroups analysis, 52 patients with HS had their γ-secretase (GS) complex–encoding genes sequenced. Among these patients, only 1 with a familial form of HS showed a previously unreported heterozygous mutation of the NCSTN gene and negative detection of ASCAs. Among the 51 other patients free of mutation in any of the 4 GS genes, 19 (36.5%) had a positive result of serologic detection of ASCAs.Table ICharacteristics of patients in the ASCA-seropositive and ASCA-seronegative groups, respectivelyCharacteristicASCA-negativeASCA-positiveP value∗P values obtained by using the Pearson or Fisher chi-square test as appropriate for categoric variables and the Wilconxon test for continuous variables.Sociodemographic characteristics Female, n (%)68 (60.7)24 (66.7).522 Age, median (IQR)30 (25.75-38)33 (26-38.5).442Clinical features BMI, median (IQR)26.1 (23.2-31.5)27.5 (23.7-35.3).296 Active smoking, n (%)73 (65.8)14 (40).007 Associated inflammatory rheumatic disease, n (%)5 (4.5)6 (16.7).025 HS-related autoinflammatory syndromes, n (%)0 (0)10 (27.8).001PASH3PASS2PAPASH4PsAPASH1 Topography, n (%) Axillary81 (72.3)33 (91.7).016 Inguinoperineal102 (91.1)33 (91.7)1.00 Others39 (34.8)15 (41.7).458 Locations number, n (%)127 (24.1)3 (8.3).123252 (46.4)20 (55.6)≥333 (29.5)13 (36.1) Disease duration (y), median (IQR)9 (4.5-13.5)10 (5-17.5).324 Hurley stage, n (%)I33 (29.5)5 (13.9)<.001II47 (42)8 (22.2)III32 (28.6)23 (63.9) C-reactive protein serum level ≤5 mg/L63 (64.95)8 (25)<.001Previous antibiotherapy, n (%)32 (28.83)12 (34.29).53BMI, Body mass index; IQR, interquartile range; PAPASH, Pyogenic arthritis, pyoderma gangrenosum, acne, and hidradenitis suppurativa; PASH, pyoderma gangrenosum, acne, and suppurative hidradenitis; PASS, pyoderma gangrenosum, acne, and spondyloarthritis; PsAPASH, psoriatic arthritis, pyoderma gangrenosum, acne, and hidradenitis suppurativa.∗ P values obtained by using the Pearson or Fisher chi-square test as appropriate for categoric variables and the Wilconxon test for continuous variables. Open table in a new tab BMI, Body mass index; IQR, interquartile range; PAPASH, Pyogenic arthritis, pyoderma gangrenosum, acne, and hidradenitis suppurativa; PASH, pyoderma gangrenosum, acne, and suppurative hidradenitis; PASS, pyoderma gangrenosum, acne, and spondyloarthritis; PsAPASH, psoriatic arthritis, pyoderma gangrenosum, acne, and hidradenitis suppurativa. This study identifies serologic detection of ASCAs as a biomarker for HS, mainly in its severe and inflammatory forms. The association between ASCA seropositivity and high HS tissue and/or systemic inflammatory burden, provides support for a link between systemic inflammation in the absence of IBD and the development of ASCAs. Interestingly, it has been showed in Crohn disease that positivity for ASCAs is associated with more severe disease and greater need for surgery.7Kim B.C. Park S. Han J. Kim J.H. Kim T.I. Kim W.H. Clinical significance of anti-Saccharomyces cerevisiae antibody (ASCA) in Korean patients with Crohn’s disease and its relationship to the disease clinical course.Dig Liver Dis. 2007; 39: 610-616Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar In the similar way in our study, ASCA detection was associated with Hurley stage III in multivariate analyses. On another hand, the value of ASCAs in Hurley stages I and II is more uncertain, and the hypothesis that it might have prognostic significance requires further investigation. Moreover, the presence of ASCAs in all patients with HS-related autoinflammatory syndromes, regardless of joint involvement, supports this latter model. All but 1 of the tested patients were free of GS gene mutation, supporting the conclusion that known genetic abnormalities in HS are not key drivers of ASCA production. The high prevalence of IgA isotype among the HS population argues for a gastrointestinal tropism of deregulated inflammatory responses. Likewise, although none of our patients reported gastrointestinal symptoms, nor did any of them develop any sign of IBD during follow-up, previous observations reporting histopathologic inflammatory changes of the digestive mucosa7Kim B.C. Park S. Han J. Kim J.H. Kim T.I. Kim W.H. Clinical significance of anti-Saccharomyces cerevisiae antibody (ASCA) in Korean patients with Crohn’s disease and its relationship to the disease clinical course.Dig Liver Dis. 2007; 39: 610-616Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar are in line with this latter hypothesis, raising ASCAs as a candidate biomarker for inflammatory burden in the absence of IBD and for putative digestive dysbiosis, a hypothetic pattern that still needs to be established by investigations of the gut microbiome. As the association of detection of ASCAs with gut dysbiosis and gastrointestinal Candida colonization is established in Crohn disease,8Standaert-Vitse A. Sendid B. Joossens M. François N. Vandewalle-El Khoury P. Branche J. et al.Candida albicans colonization and ASCA in familial Crohn’s disease.Am J Gastroenterol. 2009; 104: 1745-1753Crossref PubMed Scopus (133) Google Scholar such a signature might reflect similar perturbations in HS. Likewise, animal models of transtissue inflammation after modulation of the intestinal microbiome provide support for this hypothesis.9Scher J.U. Littman D.R. Abramson S.B. Microbiome in inflammatory arthritis and human rheumatic diseases.Arthritis Rheumatol. 2016; 68: 35-45Crossref PubMed Scopus (132) Google Scholar Therefore, the present data support the need for further investigations of the gut mycobiome in HS, as a potential promoting factor for tissue inflammation. Moreover, given the major role of IL-17 in intestinal barrier immunity and homeostasis, it would also be interesting to investigate the incident risk of ongoing development of IL-17 inhibitors for use in treating severe HS. In conclusion, this study identifying ASCAs as a biomarker for inflammatory HS paves the way for further explorations of gut mycobiome in different disease subsets. A multicentric national transversal study was conducted between 2012 and 2017, in 3 Dermatology departments known as major French tertiary care centers for skin IMIDs, including HS, which are located respectively in Saint-Louis Hospital in Paris and the university hospitals respectively from Lille and from Besançon. The patients were consecutive adult (aged ≥18 years) outpatient candidates for systemic therapy requiring a blood analysis and presenting with a diagnosis of either HS (n = 148) or PV (n = 159), ascertained by an expert physician in these diseases (H.B., J.G., G.H., F.A., and E.D.). Given the known high prevalence of ASCAs in IBD, patients with a personal history of Crohn disease or ulcerative colitis were excluded to avoid a major confounding factor. Other exclusion criteria included a personal history of any systemic immunomodulatory or immunosuppressive treatment (eg, methotrexate, ciclosporine, azathioprine) and all biologics, including TNF-α. All patients with HS were candidates for systemic therapy, including antibiotics and/or TNF inhibitors, whereas patients with PV were candidates for either conventional or biologic antipsoriatic therapy approved in France. Sera from healthy blood donors were collected (Etablissement Français du Sang, Hôpital Saint-Louis, Paris, France) during the same period, after patients had filled out a questionnaire to exclude individuals with psoriasis; a history of HS, Crohn disease, or ulcerative colitis; or a personal history of any systemic immunomodulatory or immunosuppressive treatment (eg, methotrexate, ciclosporine, azathioprine, or all biologics, including TNF inhibitors). The information collected from these self-administered questionnaires included smoking status, sex, age, and lastly, weight and height, to calculate body mass index. All patients and healthy donors were enrolled after providing written informed consent. The study protocol was approved by the Comité de Protection des Personnes Ile-de-France II. The research was conducted in compliance with the Declaration of Helsinki. We collected demographic and dermatologic features, such as age, sex, weight and height, smoking status, severity assessment of the diseases (Hurley stages, localization of the disease for HS, and Psoriasis Area Severity Index for PV); previous antibiotherapy; and the eventual established association with other IMIDs (especially inflammatory rheumatic diseases, which have been shown to be more prevalent in patients with HS and patients with PV [spondyloarthritis in the former and psoriatic arthritis in the latter]). Serologic detection of ASCAs and their IgG and/or IgA isotypes was performed by using the Bluewell IgA-IgG Screen ELISA ASCC02-96 Kit (D-Tek, Mons, Belgium). According to the manufacturer, positivity for ASCAs was defined by a serum concentration of at least 25 UA/mL for IgG or for IgA. DNA was extracted from the peripheral blood leukocytes of 52 patients with HS using the GE Healthcare Nucleon BACC3 Genomic DNA Extraction Kit (GE Healthcare, Piscataway, NJ) following the manufacturer's instructions. All exons and at least 100 bp of flanking intronic sequences of PSENEN (NM_001281532), PSEN1 (NC_000014), NCSTN (NM_015331), and APH1 (NM_016022) were amplified by PCR using specific primers. Both DNA strands were sequenced using the Big Dye Terminator Cycle Sequencing Ready Reaction Kit, version 3.1 (Applied Biosystems, Bedford, Mass). Qualitative variables were expressed as frequencies (percent) and continuous variables were expressed as median (interquartile range); they were compared by using the chi-square or Fischer exact test and Wilcoxon rank sum test as appropriate. Pairwise tests were performed between patients with HS, patients with psoriasis, and healthy controls, 2 by 2. Multivariate analysis was performed by using logistic regression. To identify the determinants of ASCA serologic positivity, a multivariate logistic regression was performed by using all variables significant at the .20 level in univariate analysis. A stepwise forward procedure was used to derive the multivariate model. The odds ratios were computed with their 95% CIs. Statistical significance was defined as a P value less than or equal to .05. The analysis was performed with Stata software (version 14.0) and R software (version 3.5.1). Table E1Patients' baseline characteristics and seroprevalence of ASCA across the 3 study groupsCharacteristicPatients with HS (n = 148)Patients with PV (n = 159)HCs (n = 162)P value∗P values obtained by using the Pearson or Fisher chi-square test, as appropriate for categoric variables and the Wilcoxon test for continuous variables for patients with HS vs for HCsP value∗P values obtained by using the Pearson or Fisher chi-square test, as appropriate for categoric variables and the Wilcoxon test for continuous variables for patients with PV vs for HCsP value∗P values obtained by using the Pearson or Fisher chi-square test, as appropriate for categoric variables and the Wilcoxon test for continuous variables for patients with HS vs for patients with PVFemale, n (%)92 (62.2)65 (40.9)88 (54.3).016.163<10-4Age at inclusion (y), median (IQR)31.5 (26-39)45 (33-56)33 (26-47.8)<10-4.035<10-4Age at disease onset (y), median (IQR)20 (16-27)27 (18-40)Disease duration (y), median (IQR)9 (5-14.75)19 (12-29)BMI, kg/m2, median (IQR)26.5 (23.2-33.0)26.6 (24.0-30.2)23.6 (21.5-25.8)<10-4<10-4.134Active smoking, n (%)87 (59.6)63 (40.4)35 (21.7)<10-4<10-4.005Hurley grades, n (%) I38 (25.7) II55 (37.2) III55 (37.2)PASI, median (IQR)10.2 (7.7-14.5)Inflammatory rheumatologic disease, n (%)11 (7.4)30 (19)BMI, Body mass index; HC, healthy controls; IQR, interquartile range 25-75; PASI, Psoriasis Area Severity Index.∗ P values obtained by using the Pearson or Fisher chi-square test, as appropriate for categoric variables and the Wilcoxon test for continuous variables Open table in a new tab Table E2ASCA seropositivity odds ratio among all patients with HS and patients at each Hurley stage versus controlsPopulationnOR95% CIP valueAll Hurley stages1487.123.24-17.98<.01Hurley stage I383.350.94-11.17.049Hurley stage II553.771.29-11.28.015Hurley stage III5515.926.59-43.09<.001OR, Odds ratio. Open table in a new tab Table E3Factors associated with positive serologic detection of ASCA among patients with HS in multivariate analysisFactorOR95% CIP valueHurley stage III3.541.6-8.1.003CRP level >10 mg/L2.581.1-5.9.02CRP, C-reactive protein; OR, odds ratio. Open table in a new tab BMI, Body mass index; HC, healthy controls; IQR, interquartile range 25-75; PASI, Psoriasis Area Severity Index. OR, Odds ratio. CRP, C-reactive protein; OR, odds ratio. Anti–Saccharomyces cervisiae antibodies in hidradenitis suppurativa: More than a gut feelingJournal of Allergy and Clinical ImmunologyVol. 146Issue 2PreviewAnti–Saccharomyces cervisiae antibodies (ASCAs) are pANCA antibodies against the mannose portion (mannans) of the fungal cell wall.1 They have a high sensitivity and specificity as a diagnostic biomarker for Crohn disease and as Assan et al1 report are associated with severe disease activity in hidradenitis suppurativa (HS). HS is a chronic, inflammatory dermatoses manifesting in painful nodules and abscesses in intertriginous areas often with development of purulent chronically draining epithelialized dermal tunnels. Full-Text PDF Anti–Saccharomyces cerevisiae antibodies could reflect distinct phenotypes in hidradenitis suppurativaJournal of Allergy and Clinical ImmunologyVol. 146Issue 2PreviewIn their recent publication Assan et al1 have shown that both IgG and IgA anti–Saccharomyces cerevisiae antibodies (ASCAs) are associated with hidradenitis suppurativa (HS).1 As the authors state, ASCA positivity has long been associated with Crohn disease and has recently been identified as a biomarker for spondyloarthritis (SpA). However, both Crohn disease and SpA comprise distinct phenotypes. The phenotype in Crohn disease can change from an inflammatory phenotype, to a fibrostenotic or penetrating phenotype with respectively severe stricturing or fistulation. Full-Text PDF