Skin-based interventions for AD and FA prevention
An increased understanding of the role of the skin barrier and microbiome in AD and FA pathogenesis has sparked interest in developing skin-based interventions for allergy prevention. In the past decade, randomized controlled trials have explored the use of prophylactic skin interventions from early life to prevent development of AD and FA. A Cochrane individual participant data (IPD) meta-analysis found no benefit of a variety of skin interventions inclusive of emollients, oils, and bathing advice overall for AD and FA prevention.71 Another meta-analysis by Zhong et al. found that prophylactic emollients were effective against AD development in high-risk infants only and in those without an interval between treatment cessation and AD outcome assessment.72
A small study in high-risk infants who received emollients (n=11 in the intervention arm and n=12 controls) from birth to age 6 months found greater skin microbiota diversity, lower skin pH, and shifts inStreptococcus abundance on skin, specifically higher S. salivarius and lower S. mitis relative abundance, compared to controls.73 The relative abundance of S. salivarius also correlated negatively with skin pH, suggesting that the lowering of skin pH due to long-term emollient use may either be one of the mechanisms through which changes in skin microbiota composition impact AD development or that these microbial fluctuations merely occur alongside skin barrier changes due to the intervention. Differentiating between these possibilities could not be achieved from this study alone and will require in-depth mechanistic studies such as in murine models to understand the causal interactions between skin emollients, microbiota and the skin barrier. There have also been no studies investigating the role of the skin microbiome in FA prevention.
Skin microbiome-based approaches for AD treatment
In contrast to broad-spectrum antibiotic therapies, repurposing commensal and symbiotic microbes for therapeutic strategies could allow more targeted antimicrobial effects against pathogenic microbes without unintended collateral impact on the rest of the healthy microbiota. Microbiome studies have now investigated numerous microbial mechanisms by which AD skin could benefit from a microbial driven anti-inflammatory outcome. The earliest studies utilized a bacterial lysate ofVitreoscilla filiformis as AD therapeutics. The proposed mechanism was via host antimicrobial peptides and anti-inflammatory effects. In a randomized, double-blind, vehicle-controlled trial, mild AD subjects (n=75 aged 6-70 years) who received topical applications ofV. filiformis experienced improvements in SCORAD and pruritus scores, as well as downward trend of S. aureus skin burden and drop-outs related to lack of efficacy.74
A subsequent MRSA skin infection murine study demonstrated that CoNS species can play an important role in activating other microbes and can also interact with the epidermis to maintain homeostasis, and interact or compete with S. aureus .75 S. epidermidis is an important CoNS that has been shown to reduce skin inflammation by regulating TLR pathways and can produce sphingomyelinase to aid in the production of ceramides to maintain the skin barrier in mouse and in vitro models. 76, 77 These functions highlight the important role S. epidermidis has on the skin and the potential that could be harnessed for AD. However, the use of S. epidermidis without attenuation would be complex, and potentially controversial, due to its dual roles in protection from pathogens but also ability to stimulate inflammation or cause systemic infections in certain situations.78
Transplantation of CoNS commensal microbe S. hominis from healthy patients onto AD skin has been investigated. CoNS strains from healthy volunteers produced potent AMPs that reduced S. aureus on the skin of 5 adult AD subjects.35 A subsequent follow-up phase 1 clinical trial using the S. hominis A9 (ShA9) strain in 54 adult AD patients demonstrated safety; however, eczema severity was not significantly different when examined among all treated participants.79 A post-hoc analysis showed that a subgroup of patients who were colonized by S. aureus strains which were sensitive to killing by ShA9 showed significant clinical improvement compared with vehicle.79 A phase 2 clinical trial [ClinicalTrials.gov ID: NCT05177328] is ongoing to further investigate this approach.
Another bacteriotherapy approach used Roseomonas mucosa (R. mucosa) , a skin commensal harvested from healthy donors, as topical therapy for AD. An open-label phase I/II trial with ten adult and 5 pediatric subjects who received skin treatment with R .mucosa demonstrated improved disease severity, reduced topical steroid requirements, and decreased S. aureusburden.80 However, the phase II placebo-controlled clinical trial [ClinicalTrials.gov ID: NCT04936113] which combined three strains of R. mucosa for a live therapeutic product (FB-401) in AD patients was terminated because it failed to demonstrate any significant differences in patients for achieving the primary outcome of EASI-50 (the proportion of subjects with a minimum of 50% improvement in AD disease severity as measured by the Eczema Area and Severity Index (EASI) score).
Additionally, there have also been recent potential biotherapeutic approaches using the non-bacterial components of the microbiome such as bacteriophages to target S. aureus but not S. epidermidis .81 Bacteriophage endolysin-based therapies have also been developed as AD therapeutics. Staphefekt SA.100 is a commercialized topically administered recombinant phage endolysin which was able to reduce AD symptom and skin inflammation in early phase studies. However, in a follow-on double-blinded, vehicle-controlled randomized superiority trial in adults with non-clinically infected moderate-to-severe AD [ClinicalTrials.gov ID: NCT02840955], the endolysin intervention failed to demonstrate superiority over placebo in topical steroid usage, clinical efficacy, quality of life, or S. aureus burden.82However, this trial excluded patients with clinically infected AD and only half of the cohort were colonized with S. aureus prior to enrolment.
Due to the individuality of skin microbiomes at the strain level in both healthy individuals and AD patients, bacteriotherapy may be effective only in heavily colonized or clinically infected AD, but may be less effective in other circumstances. It is therefore likely that these microbial therapeutic approaches will require careful patient selection to identify specific subgroups who will benefit from such interventions, and optimization of therapeutic design to achieve sustained clinical responses.79, 82