Endolysin treatment against Staphylococcus aureus in adults with atopic dermatitis: A randomized controlled trial

Staphylococcus aureus density is increased in many patients with atopic dermatitis (AD) and is thought to contribute to disease pathogenesis, interacting with an altered skin barrier and immunologic changes.1Totté J.E. van der Feltz W.T. Hennekam M. van Belkum A. van Zuuren E.J. Pasmans S.G. Prevalence and odds of Staphylococcus aureus carriage in atopic dermatitis: a systematic review and meta-analysis.Br J Dermatol. 2016; 175: 687-695Crossref PubMed Scopus (153) Google Scholar S aureus might induce or aggravate inflammation through different mechanisms, for example through excretion of virulence factors, even if the S aureus overgrowth is primarily caused by other factors.2Totté J. de Wit J. Pardo L. Schuren F. van Doorn M. Pasmans S. Targeted anti-staphylococcal therapy with endolysins in atopic dermatitis and the effect on steroid use, disease severity and the microbiome: study protocol for a randomized controlled trial (MAAS trial).Trials. 2017; 18: 404Crossref PubMed Scopus (24) Google Scholar Current guidelines only recommend antimicrobial therapy directed against S aureus in patients with clinically infected AD based on a Cochrane review in which no clinical benefit of short-term antimicrobial treatment in patients with noninfected AD was found.3Bath-Hextall F.J. Birnie A.J. Ravenscroft J.C. Williams H.C. Interventions to reduce Staphylococcus aureus in the management of atopic eczema: an updated Cochrane review.Br J Dermatol. 2010; 163: 12-26Crossref PubMed Google Scholar Arguably, long-term antistaphylococcal treatment, such as antibiotics, might reduce symptoms in patients with AD.4Wong S.M. Ng T.G. Baba R. Efficacy and safety of sodium hypochlorite (bleach) baths in patients with moderate to severe atopic dermatitis in Malaysia.J Dermatol. 2013; 40: 874-880Crossref PubMed Scopus (56) Google Scholar However, this is undesired because antibiotics can affect the commensal microbiota and could induce bacterial resistance.5Niebuhr M. Mai U. Kapp A. Werfel T. Antibiotic treatment of cutaneous infections with Staphylococcus aureus in patients with atopic dermatitis: current antimicrobial resistances and susceptibilities.Exp Dermatol. 2008; 17: 953-957Crossref PubMed Scopus (48) Google Scholar In contrast, long-term treatment of AD with an endolysin that targets only S aureus is feasible. It might improve AD symptoms and reduce the use of corticosteroids consecutively.2Totté J. de Wit J. Pardo L. Schuren F. van Doorn M. Pasmans S. Targeted anti-staphylococcal therapy with endolysins in atopic dermatitis and the effect on steroid use, disease severity and the microbiome: study protocol for a randomized controlled trial (MAAS trial).Trials. 2017; 18: 404Crossref PubMed Scopus (24) Google Scholar, 6Totté J.E.E. van Doorn M.B. Pasmans S. Successful treatment of chronic Staphylococcus aureus–related dermatoses with the topical endolysin Staphefekt SA.100: a report of 3 cases.Case Rep Dermatol. 2017; 9: 19-25Crossref PubMed Scopus (40) Google Scholar Therefore we aimed to determine the topical corticosteroid (TCS)–sparing effect and safety of 12 weeks of endolysin treatment against S aureus in patients with AD. We performed a double-blind, vehicle-controlled superiority trial (MAAS trial, ClinicalTrials.gov NCT02840955) in 100 adult patients with TCS-treated, not clinically infected, moderate-to-severe AD, which was defined by an Eczema Area and Severity Index score of 7.1 to 50.0. After a 2-week run-in period to standardize the TCS treatment with triamcinolone acetonide 0.1% cream, patients were randomly assigned 1:1 to a 12-week intervention with either a topical endolysin against S aureus or a vehicle twice daily, followed by an 8-week follow-up period. The vehicle and recombinant chimeric endolysin, Staphefekt SA.100, were provided by Micreos Human Health (Bilthoven, The Netherlands) and topically applied in a cetomacrogol cream. Details on patient inclusion, randomization, and study procedures at 6 assessments are described in a previously published study protocol (see Fig E1 in this article's Online Repository at www.jacionline.org).2Totté J. de Wit J. Pardo L. Schuren F. van Doorn M. Pasmans S. Targeted anti-staphylococcal therapy with endolysins in atopic dermatitis and the effect on steroid use, disease severity and the microbiome: study protocol for a randomized controlled trial (MAAS trial).Trials. 2017; 18: 404Crossref PubMed Scopus (24) Google Scholar The primary outcome, the TCS-sparing effect of endolysin treatment against S aureus, was evaluated by the patients, who registered daily use of a TCS (yes/no) over 12 weeks. Secondary outcomes included differences in TCS use measured in grams, clinical efficacy, quality of life using the Skindex-29, S aureus load on the skin, and safety and tolerability of endolysin treatment. Clinical efficacy was measured by using the Eczema Area and Severity Index, Investigators Global Assessment, Patient-Oriented Eczema Measure, and pruritus Numeric Rating Scale and by registration of the number of flares. The S aureus load on the skin was assessed by using a semiquantitative culture and quantitative PCR (qPCR; see Appendix E1 in this article's Online Repository at www.jacionline.org). Generalized linear mixed-effects models for repeated measurements were used to analyze the primary and dichotomous secondary outcomes, and linear mixed-effects models were used for continuous secondary outcomes. Data were analyzed as intention-to-treat and per-protocol. Furthermore, a subgroup analysis was performed in patients with a positive S aureus skin culture at 2 time points before start of the intervention. Eighty-eight (88.0%) patients completed the intervention, and 87 (87.0%) completed follow-up (13% dropout rate, see Fig E2 in this article's Online Repository at www.jacionline.org). Patients' characteristics were comparable between the endolysin and vehicle groups (Table I). Over the 12-week intervention period (corresponding to 8400 days for 100 patients), patients in the endolysin group used a TCS for 1889 (45.0%) days compared with 1566 (37.3%) days in the vehicle group. There was no statistically significant difference in the probability of TCS use per day between the groups in the intention-to-treat and per-protocol analyses and in the subgroup of S aureus–positive patients (P = .97, P = .40, and P = .08, respectively; Table II). Sensitivity analyses showed no differences in the odds ratio of TCS use per assessment day. Except for the number of doctor-reported AD flares during the intervention period (per-protocol: n = 2 in the endolysin group vs n = 10 in the vehicle group; P = .03), no statistically significant differences were found in the secondary outcomes after both intention-to-treat and per-protocol analyses (see Tables E1 and E2 in this article's Online Repository at www.jacionline.org). At baseline, 62 (64.6%) patients had positive results for S aureus based on skin culture and 24 (24.7%) by qPCR. Both methods showed no significant difference in S aureus reduction (see Tables E3 and E4 in this article's Online Repository at www.jacionline.org). During the study, 1 serious adverse event occurred in the endolysin group 8 weeks after the last application of endolysin cream (pleural effusion with hospitalization), which was considered unlikely to be related to the study intervention (see Tables E5 and E6 in this article's Online Repository at www.jacionline.org).Table IBaseline characteristicsTotal (n = 100)Staphefekt (n = 50)Vehicle (n = 50)Age (y), median (IQR)33.5 (25.5-47.5)36.5 (25.0-51.0)32.5 (24.0-44.0)Male sex, no. (%)55 (55.0)24 (48.0)31 (62.0)Race, no. (%) American Indian or Alaska Native5 (5.0)2 (4.0)3 (6.0) Asian10 (10.0)2 (4.0)8 (16.0) Black or African American8 (8.0)5 (10.0)3 (6.0) White77 (77.0)41 (82.0)36 (72.0)Atopic disease, no. (%) Food allergy43 (43.0)18 (36.0)25 (50.0) Rhinoconjunctivitis63 (63.0)28 (56.0)35 (70.0) Asthma47 (47.0)25 (50.0)22 (44.0)EASI score, median (IQR) Screening (V1)12.9 (9.2-19.0)13.7 (8.9-19.1)12.5 (9.2-19.0) Baseline (V2)8.0 (5.0-13.5)*8.3 (5.0-14.7)†8.0 (4.9-12.9)‡IGA, median (IQR), baseline (V2)2.0 (2.0-3.0)*2.0 (2.0-3.0)†2.0 (2.0-3.0)‡POEM, mean (SD), baseline (V2)12.9 (6.2)*14.5 (8.3-17.0)†13.0 (8.0-15.0)‡Pruritus NRS, median (IQR), baseline (V2)3.0 (2.0-4.0)*3.0 (2.0-4.0)†3.0 (2.0-4.0)‡Skindex-29, mean (SD), baseline (V2)35.1 (17.1)*37.3 (15.3)†32.9 (18.7)‡Use of TCSs at screening, no. (%) Class 13 (3.0)1 (2.0)2 (4.0) Class 213 (13.0)6 (12.0)7 (14.0) Class 2-311 (11.0)3 (6.0)8 (16.0) Class 344 (44.0)22 (44.0)22 (44.0) Class 3-43 (3.0)2 (4.0)1 (2.0) Class 418 (18.0)11 (22.0)7 (14.0) Unknown8 (8.0)6 (12.0)4 (8.0)Staphylococcus aureus skin culture,§Positive is defined as having a positive culture at visit 1 and visit 2a, intermediate is defined as having 1 positive culture and 1 negative culture at visit 1 and visit 2a, and negative is defined as having 2 negative cultures at visit 1 and visit 2a. no. (%) Positive56 (56.0)32 (64.0)24 (48.0) Intermediate20 (20.0)7 (14.0)13 (26.0) Negative20 (20.0)9 (18.0)11 (22.0) Missing4 (4.0)2 (4.0)2 (4.0)EASI, Eczema Area and Severity Index; IGA, Investigators Global Assessment; IQR, interquartile range; NRS, Numeric Rating Scale; POEM, Patient-Oriented Eczema Measure.Missing values: *n = 3 (3.0%), †n = 2 (4.0%), and ‡n = 1 (2.0%).§ Positive is defined as having a positive culture at visit 1 and visit 2a, intermediate is defined as having 1 positive culture and 1 negative culture at visit 1 and visit 2a, and negative is defined as having 2 negative cultures at visit 1 and visit 2a. Open table in a new tab Table IIGeneralized linear mixed-effects model results for the difference in TCS use per day during the 12-week intervention periodAnalysisTime period or time point in days from baselinePatients included in analysisTCS use “yes,” no. (%)OR (95% CI)P value∗Overall effect of endolysin treatment during the intervention period was calculated with a likelihood ratio test, and the effect per time point was calculated with a Wald test with a t-distribution.Staphefekt, no.Vehicle, no.Staphefekt, no. (%)Vehicle, no. (%)Intention-to-treatIntervention14428440373037373426 (65.00)26 (70.27)18 (60.00)20 (54.05)22 (59.46)21 (61.76)0.99 (0.95-1.03)0.99 (0.78-1.24)1.04 (0.87-1.23).97.49.91.68Per-protocolIntervention14428422171321181314 (63.64)12 (70.59)7 (53.85)10 (47.62)11 (61.11)5 (38.46)1.09 (0.98-1.21)1.36 (0.84-2.20)1.69 (1.14-2.51).40.10.21.01Staphylococcus aureus positive†Defined as having a positive culture at both visit 1 and visit 2a (endolysin, n = 32; vehicle, n = 24).Intervention14428426221816171517 (65.38)16 (72.73)11 (61.11)7 (43.75)8 (47.06)11 (73.33)0.87 (0.78-0.96)0.85 (0.49-1.48)1.29 (0.84-1.96).08.01.56.24Sensitivity analyses, performed by adding/subtracting 0.25 times the SD to/from the odds, showed no differences in the odds ratio of TCS use per assessment day. Note: Given the low number of patients using escape medication, ie, TCS with higher potency (n = 5), we did not correct for its use.OR, Odds ratio.∗ Overall effect of endolysin treatment during the intervention period was calculated with a likelihood ratio test, and the effect per time point was calculated with a Wald test with a t-distribution.† Defined as having a positive culture at both visit 1 and visit 2a (endolysin, n = 32; vehicle, n = 24). Open table in a new tab EASI, Eczema Area and Severity Index; IGA, Investigators Global Assessment; IQR, interquartile range; NRS, Numeric Rating Scale; POEM, Patient-Oriented Eczema Measure. Missing values: *n = 3 (3.0%), †n = 2 (4.0%), and ‡n = 1 (2.0%). Sensitivity analyses, performed by adding/subtracting 0.25 times the SD to/from the odds, showed no differences in the odds ratio of TCS use per assessment day. Note: Given the low number of patients using escape medication, ie, TCS with higher potency (n = 5), we did not correct for its use. OR, Odds ratio. Our results are in accordance with data from a Cochrane review showing no significant effect of short-term anti–S aureus therapy in patients with noninfected AD.3Bath-Hextall F.J. Birnie A.J. Ravenscroft J.C. Williams H.C. Interventions to reduce Staphylococcus aureus in the management of atopic eczema: an updated Cochrane review.Br J Dermatol. 2010; 163: 12-26Crossref PubMed Google Scholar We cannot confirm the positive results of other longer-term studies. However, these studies used broad-spectrum antimicrobials and mainly included patients with signs of bacterial infection.4Wong S.M. Ng T.G. Baba R. Efficacy and safety of sodium hypochlorite (bleach) baths in patients with moderate to severe atopic dermatitis in Malaysia.J Dermatol. 2013; 40: 874-880Crossref PubMed Scopus (56) Google Scholar Patients with clinically infected AD were excluded from our study, and a possible effect of anti–S aureus endolysins in this patient group should be determined in future studies. Several hypotheses could explain our results. First, use of triamcinolone in the run-in phase resulted in a decrease in AD severity (Table I), which might have masked a possible benefit of endolysin treatment. Second, daily use of an emollient and good compliance with the treatment could have resulted in a reduction in triamcinolone use in both the endolysin and vehicle groups.7Ng J.P. Liew H.M. Ang S.B. Use of emollients in atopic dermatitis.J Eur Acad Dermatol Venereol. 2015; 29: 854-857Crossref PubMed Scopus (20) Google Scholar Because AD is a heterogeneous disease, anti–S aureus treatment might not be suitable for all patients with AD, indicating the need for subphenotyping. Because only 56% of our study population had 2 consecutive positive S aureus skin cultures (indicating persistent colonization) before the start of the intervention, the target population that would probably benefit the most from endolysin treatment was small. Our data suggest that endolysin treatment has no effect on S aureus in vivo. However, patients might have been recolonized with S aureus from the nose because 73% of them were nasal carriers (data not shown). Alternatively, cetomacrogol as the basis of the endolysin cream might have created a barrier on the skin that prevented the endolysin to reach and subsequently kill S aureus. However, some reduction in S aureus load would have been expected in both treatment groups because of the use of TCSs and emollients in this study, which both have been shown to reduce the S aureus load on the skin.8Angelova-Fischer I. Neufang G. Jung K. Fischer T.W. Zillikens D. A randomized, investigator-blinded efficacy assessment study of stand-alone emollient use in mild to moderately severe atopic dermatitis flares.J Eur Acad Dermatol Venereol. 2014; 28: 9-15Crossref PubMed Scopus (37) Google Scholar, 9Gonzalez M.E. Schaffer J.V. Orlow S.J. Gao Z. Li H. Alekseyenko A.V. et al.Cutaneous microbiome effects of fluticasone propionate cream and adjunctive bleach baths in childhood atopic dermatitis.J Am Acad Dermatol. 2016; 75: 481-493.e8Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar Nonetheless, it is unclear whether complete eradication of S aureus is required for clinical improvement because a case series showed a clear clinical improvement without S aureus reduction using a qualitative culture in patients with S aureus–related dermatoses.6Totté J.E.E. van Doorn M.B. Pasmans S. Successful treatment of chronic Staphylococcus aureus–related dermatoses with the topical endolysin Staphefekt SA.100: a report of 3 cases.Case Rep Dermatol. 2017; 9: 19-25Crossref PubMed Scopus (40) Google Scholar In addition, the discrepant results between culture and qPCR indicate the complexity of the interpretation of S aureus testing. Despite the limitations and outcome, this study provides estimates of AD symptoms, use of TCSs, and percentages of persistent S aureus carriers that can be used for future clinical studies. In conclusion, long-term targeted endolysin treatment against S aureus in this study was well tolerated but had no TCS-sparing effect in patients with AD. However, an effect cannot be excluded because good compliance with the treatment and concurrent application of TCSs, emollients, or both might have masked a clinical benefit. We thank all patients for participating in the study and acknowledge the assistance in data collection provided by K. Stuvel, MD. Download .pdf (.54 MB) Help with pdf files Appendix E1