Systematic review on antipruritic therapies for patients with Epidermolysis bullosa

Abstract

Introduction

Itch is one of the most burdensome symptoms in epidermolysis bullosa (EB), indicating a hitherto unmet therapeutic need. This review leverages existing data on efficacy of itch treatment in EB to support sound decision making.

Methods

A systematic literature search was performed on 29 March 2022. Studies written later than 1991 and reporting outcomes in patients with EB treated for itch were considered.

Results

Of the 3,099 articles screened, 21 studies met eligibility criteria, comprising 353 patients (65.9%) diagnosed for recessive dystrophic EB. Only two studies (9.5%) evaluated itch as primary endpoint, of which solely one revealed a significant relief of self-reported itch upon topical skin care. In those studies assessing itch as secondary endpoint (19/21, 90.5%), only 36.8% studies (n = 7/19) revealed a statistically significant itch reduction of up to 42%. Methodological limitations (heterogeneity of outcomes, inconsistent data assessment) in addition to limited superiority over control were implicated to account for low treatment efficacy observed in most studies.

Conclusion

Current data quality impairs comparative efficacy analyses of itch treatments in EB. Large scale randomized clinical trials and more personalized approaches applying validated measurement instruments for core outcomes are needed to substantiate evidence-based treatment approaches for EB-associated itch.

Keywords:

• Epidermolysis bullosa
• study outcomes
• itch treatment
• quality of life
• disease severity

Background

Epidermolysis bullosa (EB) comprises a heterogeneous subset of inherited skin fragility disorders with clinical hallmarks including blisters, wounds, and scars, which develop upon minor mechanical trauma (1). Molecular aberrations in at least 16 different genes have hitherto been identified to underlie the classical EB types. These variants cause an impairment of the structural and functional integrity of epithelialised tissues, in particular the epidermis and dermo-epidermal basement membrane zone (BMZ) of the skin, and interfere with key cellular pathways including cell adhesion, differentiation, barrier function and tissue repair (2). Genetic heterogeneity as well as epigenetic, biochemical and environmental factors determine the wide phenotypical variety and severity of EB (3–7). Extracutaneous involvement renders especially the more severe subtypes a systemic disease with multi-organ involvement and high morbidity as well as mortality (8,9).

Itch is one of the most burdensome patient-reported symptoms in EB, causing significant impairment of quality of life (QoL) (10). It affects up to 100% of EB patients (EB simplex, EBS 74%; junctional EB, JEB 100%; dystrophic EB, DEB 93%, i.e., dominant dystrophic EB, DDEB 87.5%, recessive dystrophic EB, RDEB 100%) and usually correlates with disease severity (10–12). In RDEB, mean visual analogue scale (VAS) scores for itch range between seven and eight, thus are comparable to those reported for chronic urticaria, atopic dermatitis and prurigo nodularis (13). Aside from effects of itch on mental health (e.g., sleep disturbances, anxiety, depression), scratching further comprises skin integrity and consequently perpetuates itch through amplification of barrier disruption and release of pro-inflammatory and pruritogenic cytokines and mediators (14–17).

The latter bind to nerve endings in the skin, which themselves promote a neurogenic inflammation (18,19). This distal axon inflammation mediates epidermal fiber damage and loss, and thereby promotes neuropathic itch and pain, stimulating a vicious itch-scratch cycle with negative impact on barrier function and disease severity (19–23).

Against this background, anti-pruritic therapies, through modulation of inflammatory cascades that are induced, dysregulated and chronified in response to the mutation-based disruption of the epidermal barrier, hold potential to improve not only the QoL but also the integrity of skin and mucosa (15,16,24,25).

Indeed, a plethora of pharmacological, behavioral and environmental therapies to treat EB associated itch has been reported in the literature (26,27). Most of these approaches, however, have shown only a moderate or no beneficial effect, underscoring a hitherto unmet therapeutic need (23).

Methodologically, current evidence on treatment modalities for EB largely refers to case reports or case series, which typically reflects inherent challenges in conducting controlled clinical trials in rare disease populations featuring small and heterogeneous sample sizes (23,28). In addition there is a considerable heterogeneity of reported outcomes that poses significant challenges for comparative assessment of available study data (29).

This review systematically evaluates clinical studies of anti-pruritic therapies in EB by summarizing and classifying reported outcomes, interventional impact on QoL and disease severity, as well as safety with the intention to improve the level of scientific and clinical reasoning.

Methods

This research was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (30). The review protocol for reference was registered in PROSPERO (CRD42022348349). Ethical approval was not required for this study.

Eligibility criteria

Comprehensive eligibility criteria were applied following the Population, Intervention, Comparator, Outcome (PICO) framework (Appendix 1) (30). Studies investigating any kind of pharmacological treatment in EB patients that also reported on its impact on symptomatic itch relief, which had been published between 1991, when the first consensus paper on EB terminology and classification had been released, and the search end date of 29 March 2022, were included (31). To account for EB-inherent limitations to high level evidence, retro- and prospective cohort studies, as well as clinical trials were included in analyses. Survey studies were also eligible for inclusion, as they provided standardized data collections of itch as a patient/self-reported outcome in daily care settings. Case reports and case series were not included to focus on reports with standardized procedures and systematic assessments.

Search strategy

We systematically searched the databases MEDLINE (PubMed), EMBASE (embase.com), CINAHL (EBSCO), PsycINFO (EBSCO), Cochrane Central Register of Controlled Trials (CENTRAL) and Web of Science at 29th March 2022. Comprehensive search strategies combining index terms and free text words were developed by two authors (E.W.H.K., S.W.). Searches were limited to articles published between 1991 and 29 March 2022. Animal-only studies, reviews, case reports, case series and meeting abstracts were excluded. The search strategies were reviewed by a second information specialist, using a standardized checklist for Peer Review of Electronic Search Strategies (PRESS) (32). The full search strategies are included in Appendix 2. In addition to the databases searched, cited references of included articles were screened to ensure identification of all potentially eligible studies. The identified citations were uploaded to EndNote 20 (Clarivate Analytics, Philadelphia, USA) and duplicates were removed according to a standardized de-duplication method (30).

Study selection and data extraction

We screened titles, abstracts, and full texts using Rayyan software for consistency with eligibility criteria and performed data extraction along with risk of bias assessment independently in duplicate (T.W. and C.PU.) (33). Titles or abstracts identified as potentially relevant by both screeners were subjected to full-text review. Inter-rater reliability was assessed at each screening stage and disagreements in title or abstract screening, as well as full-text screening, were resolved by discussion between the two screeners and, when required, with a third reviewer (M.L.). Data from all included studies were extracted by two independent reviewers (T.W., C.PU.) by means of a standardized data collection form, which had been developed by the entire review team following a pilot phase. Information on study characteristics, treatment specifications, and outcomes of interest (itch, disease severity, QoL, adverse events), with consideration of the respective measurement instruments, was retrieved from supplementary materials and clinical trial registries. As inconsistent and incomplete data reporting rendered an integrative/pooled statistical assessment or meta-analysis infeasible, statistical data were only indicated if originally reported. Any disagreements between the two reviewers during the extraction process were resolved by discussion with a third reviewer (M.L.).

Risk of bias and level of evidence assessment

Due to the EB-inherent limitations to study design and trial methodology, risk of bias assessment and level of evidence was determined to improve qualitative assessment of individual study results across the study landscape. Risk of bias was categorized as low, high or unclear by two reviewers (T.W., C.PU.) for each study by applying the Cochrane Collaboration’s tool for selection, performance, detection, attrition, and reporting bias (34). Level of evidence was assessed for each study by two reviewers (T.W., C.PU.) according to the Oxford Center for Evidence-Based Medicine (OCEBM) 2011 Levels of Evidence regarding methodological quality (35). Disagreements among reviewers regarding risk of bias or evaluation of evidence were resolved through discussion. If consensus could not be reached, a third reviewer (M.L.) was consulted to make a final decision.

Data synthesis

Data were synthesized descriptively and listed in tables and charts to determine and illustrate treatment effects on itch, QoL, and disease severity as well as safety (i.e., adverse events). Study features (e.g., study design, EB subtypes, number of subjects), details on treatment modalities (e.g., therapeutic compound, mode of administration, dosage) and outcome measurement instruments were collated to increase interpretability of individual study results in consideration of the heterogeneous study landscape.

Results

Selection of literature

Our search strategy yielded a total of 4,984 records, among which 1,725 duplicates were removed. Based on eligibility criteria, 95.1% (3,099/3,259) records were excluded during title and abstract screening, mainly because they lacked reporting on itch (outcomes) or were ineligible based on study design or intervention. Of the remaining 160 records, six full text reports were not retrievable, resulting in 154 full text manuscripts which were finally screened against the eligibility criteria (Figure 1). Inter-rater reliability was 99.6% (12 conflicts out of 3,259 records) and 70.6% (47 conflicts out of 160 studies) during title and abstract screening, and full text screening respectively. All conflicts were resolved by discussion.

Figure 1. PRISMA flowchart with the number of records identified and considered for inclusion.

Study characteristics and risk of bias assessment

A total of 21 studies met the eligibility criteria. Except for two survey studies, all of the identified studies were clinical trials (90.5%, n = 19). All eligible studies reported on interventional effects on itch, ten additionally on QoL, and seven on itch and disease severity, respectively. A large proportion of eligible studies included patients with RDEB (n = 9/21; 42.9%; with in total 102/536 participants, 19%), as well as mixed subtype cohorts (n = 6/21; 28.6%), of whom 61.5% (n = 251/536) were RDEB patients (Table 1).

Table 1. Study characteristics.

	n = 21
	n	%
EB type
EBS	3	14.3%
DEB	11	52.4%
RDEB only	9	42.9%
JEB	1	4.8%
KEB	0
Mixed subtypes (n = 408; RDEB 251, DDEB 22, DEB undefined 6, EBS 69, JEB 56, unknown 4)	6	28.6%
(Sub-) Type distribution
EBS	90	16.8%
JEB	61	11.4%
DDEB	26	4.9%
RDEB	353	65.9%
DEB undefined	6	1.1%
Focus on itch
Primary endpoint	3	14.3%
Secondary endpoint	16	76.2%
Exploratory endpoint	1	4.8%
Not specified	1	4.8%
Level of evidence
OCEBM level 1	0
OCEBM level 2	9	42.9%
OCEBM level 3	10	47.6%
OCEBM level 4	2	9.5%
OCEBM level 5	0
Age group
Adult	2	9.5%
Pediatric	5	23.8%
Any age	13	61.9%
Not specified	1	4.8&
Geographical location
Africa	0
Asia	3	14.3%
Australia	1	4.8%
Europe	7	33.3%
Middle East	0
North America	6	28.6%
South America	0
Multiple continents	4	19.0%
Not specified	0
Center
Monocentre	12	57.1%
Multicentre	9	42.9%
Phase
Phase I	0
Phase I/II	5	23.8%
Phase II	4	19.0%
Phase II/III	1	4.8%
Phase III	3	14.3%
Phase IV	0
Pilot trial	1	4.8%
Not specified	7	33.3%
Randomization
Randomized	9	42.9%
Non-randomized	9	42.9%
Not specified	3	14.3%
Assignment
Single group assignment	9	42.9%
Parallel group assignment	3	14.3%
Cross-over assignment	3	14.3%
Intra-patient controlled	4	19.0%
Not specified	2	9.5%
Blinding
Open label	12	57.1%
Single blind	0
Double blind	7	33.3%
Not specified	2	9.5%
Comparator
Baseline	14	66.7%
Placebo	7	33.3%

Interventional impact on itch was evaluated as a primary endpoint in 2/21 (9.5%) studies and as a secondary endpoint in 19/21 (90.5%) studies. According to the OCEBM evidence level categorization on treatment benefits, a large proportion (19/21; 90.5%) of the studies showed good or moderate evidence (OCEBM level 2–3) (35). However, the majority of studies were monocentric, open-label, and with only one treatment group (Table 1). Therefore, most of the studies (12/21; 57.1%) were classified to harbor a high risk of bias. This is in concordance with a high risk of bias denoted for 46% of the in total six risk items assessed for each individual study. The highest risk was found for items referring to detection, performance, and selection bias (Figure 2). No study was excluded from this review based on a classification to harbor high or unclear risk of bias.

Figure 2. Summary of risk of bias assessment for included studies using the Cochrane Collaboration tool.

The heterogeneity of reported outcomes on symptomatic itch relief impaired a comparative data assessment across studies in terms of a meta-analysis. For instance, a considerable number of studies lacked information on mean values (38.1%, n = 8/21) and statistical significance (38.1%, n = 8/21). Moreover, treatment effects were often reported only verbatim without further quantification (14.3%; n = 3/21). The use of seven different treatment measurement instruments further impeded efforts of pooled analyses. In two studies, no measurement instrument for itch was reported.

Outcome measurement instruments

In total 18 different measurement tools were used across all studies to assess outcomes on itch (n = 7), QoL (n = 5) and disease severity (n = 6) (Figure 3). In four studies, disease severity was evaluated using a second measurement instrument to corroborate and substantiate an observed therapeutic impact of the investigated drug clinically (Table 2) (36–39).

Figure 3. Number of outcome measurement instruments applied to assess itch, QoL or disease severity.

BEBSS: Birmingham Epidermolysis Bullosa Severity Score; DLQI: Dermatology Life Quality Index; EBDASI: Epidermolysis Bullosa Disease Activity and Scarring Index; GSS: global severity score; iscorEB: instrument for scoring clinical outcomes of research for epidermolysis bullosa; NRS: numerical rating scale; PedsQL: Pediatric Quality of Life Inventory; QOLEB: quality of life evaluation in epidermolysis bullosa; TBSA: total body surface area; VAS: visual analogue scale.

Table 2. Measurement instruments used to measure itch, QoL and disease severity.

Author/year	Treatment approach	Definition of outcome	Measurement instrument	Items	Focus	Creation
Itch
Local
Fivenson et al. 2003	Graftskin therapy	Pruritus at the graftskin treatment site	Not reported	N/A	N/A	N/A
Venugopal et al. 2013	Allogeneic fibroblasts in 2% albumin suspension	Local itch	VAS (0–10)	Single item	Generic	Complete validated questionnaire
Gorell et al. 2015	Helicoll Collagen I Wound Dressing	Level of wound pruritus	VAS 0–4	Single item	Generic	Complete validated questionnaire
Schwieger-Briel et al. 2017	Oleogel-S10 and non-adhesive wound dressing	Wound itch	VAS (1–10)	Single item	Generic	Complete validated questionnaire
Wally et al. 2018	Diacereine 1%	Local pruritus	VAS (0–10)	Single item	Generic	Complete validated questionnaire
Murrell et al. 2020	Allantoin 0% (SD-101–0.0)	Change from baseline in patient-reported itching	Itch Man Pruritus Assessment Tool	Single item	Generic	Complete validated questionnaire
Paller et al. 2020	Allantoin 6% (SD-101–6.0)	Change from baseline in patient-reported itching	Itch Man Pruritus Assessment Tool	Single item	Generic	Complete validated questionnaire
Chiou et al. 2020	Serlopitant	Change in NRS during dressing changes	11- point NRS	Single item	Generic	Complete validated questionnaire
Guttmann-Gruber et al. 2021	Low-dose calcipotriol ointment	Pruritus score	VAS (0–10)	Single item	Generic	Complete validated questionnaire
Systemic
Neufeld-Kaiser et al. 1997	Cyproheptadine	Levels of itching	Not reported	N/A	N/A	N/A
Danial et al. 2015	Mixed	Pruritus relief	5-point Likert scale (–2 = relieves itch a lot, −1 = relieves itch a little, 0 = no change, 1 = increases itch a little, 2 = increases itch a lot)	Singel item	Generic	Complete validated questionnaire
Petrof et al. 2015	Allogeneic bone marrow–derived mesenchymal stromal cells	Pruritus	Pruritus score child version (0–10)	Multiple item	Generic	Complete validated questionnaire
Chiaverini et al. 2015	Erythromycin	Itch severity	VAS	Single item	Generic	Complete validated questionnaire
Chiaverini et al. 2016	Epigallocatechin-3-gallate	Itch	VAS	Single item	Generic	Complete validated questionnaire
Chiou et al. 2020	Serlopitant	Weekly change in the NRS itch score	11- point NRS	Single item	Generic	Complete validated questionnaire
Rashidghamat et al. 2020	Allogeneic bone marrow–derived mesenchymal stromal cells	Itch	Leuven itch scale	Single item	Generic	Complete validated questionnaire
Fujita et al. 2021	Multilineage-differentiating stress-enduring (Muse) cells	Change in VAS for itch	VAS (–100 to 100)	Single item	Generic	Complete validated questionnaire
Kiritsi et al. 2021	Allogeneic ABCB5-positive mesenchymal stem cells (allo-APZ2-EB)	Median itch score	NRS (0–10)	Single item	Generic	Complete validated questionnaire
Lee et al. 2021	Allogeneic umbilical cord blood–derived mesenchymal stem cells	Pruritus	VAS (0–10)	Single item	Generic	Complete validated questionnaire
Chen et al. 2021	Omalizumab	Pruritus score	VAS (0–10)	Single item	Generic	Complete validated questionnaire
Schräder et al. 2021	Cannabinoid-based medicines (CBM) – THC only, THC/CBD, CBD ony	Overall average daily itch	NRS (0–10)	Singel item	Generic	Complete validated questionnaire
		Overall prutius	5-point Likert scale	Singel item	Generic	Complete validated questionnaire
Mosallaei et al. 2022	Gentamicin	Improvement of pruritus	5-D pruritus scale	Multiple item	Generic	Complete validated questionnaire
Quality of life
Fivenson et al. 2003	Graftskin therapy	Quality of life	N/A	N/A	N/A	N/A
Venugopal et al. 2013	Allogeneic fibroblasts in 2% albumin suspension	Quality of life	QOLEB	Multiple item	EB-specific	Complete validated questionnaire
Petrof et al. 2015	Allogeneic bone marrow–derived mesenchymal stromal cells	Quality of life assessed by children	PedsQL score (child version)	Multiple item	Generic	Complete validated questionnaire
		Quality of life assessed by parents	PedsQL score (parent version)	Multiple item	Generic	Complete validated questionnaire
Wally et al. 2018	1% Diacereine	Quality of life	Questionnaire (8 questions)	Multiple item	EB-specific	Self-assembled questionnaire
Rashidghamat et al. 2020	Allogeneic bone marrow–derived mesenchymal stromal cells	Quality of life	QOLEB	Multiple item	EB-specific	Complete validated questionnaire
Fujita et al. 2021	Multilineage-differentiating stress-enduring (Muse) cells	Quality of life	N/A	N/A	N/A	N/A
Kiritsi et al. 2021	Allogeneic ABCB5-positive mesenchymal stem cells (allo-APZ2-EB)	Quality of life	QOLEB	Multiple item	EB-specific	Complete validated questionnaire
Lee et al. 2021	Allogeneic umbilical cord blood–derived mesenchymal stem cells	Quality of life	QOLEB	Multiple item	EB-specific	Complete validated questionnaire
Chen et al. 2021	Omalizumab	Quality of life	DLQI	Multiple item	Generic	Complete validated questionnaire
Mosallaei et al. 2022	Gentamicin	Quality of life	Skindex-16	Multiple item	Generic	Complete validated questionnaire
Disease severity
Petrof et al. 2015	Allogeneic bone marrow–derived mesenchymal stromal cells	Efficacy measure	BEBSS	Multiple item	EB-specific	Complete validated questionnaire
		Efficacy measure	GSS score (0–12)	N/A	N/A	N/A
Rashidghamat et al. 2020	Allogeneic bone marrow–derived mesenchymal stromal cells	Disease severity	BEBSS	Multiple item	EB-specific	Complete validated questionnaire
		Disease severity	EBDASI	Multiple item	EB-specific	Complete validated questionnaire
Kiritsi et al. 2021	Allogeneic ABCB5-positive mesenchymal stem cells (allo-APZ2-EB)	Disease severity	EBDASI	Multiple item	EB-specific	Complete validated questionnaire
		Disease severity	IscorEB	Multiple item	EB-specific	Complete validated questionnaire
Lee et al. 2021	Allogeneic umbilical cord blood–derived mesenchymal stem cells	Disease severity	BEBSS	Multiple item	EB-specific	Complete validated questionnaire
		Body surface affected by EB	TBSA (%)	N/A	N/A	N/A
Chen et al. 2021	Omalizumab	Disease severity	BEBSS	Multiple item	EB-specific	Complete validated questionnaire
Schräder et al. 2021	Cannabinoid-based medicines (CBM) – THC only, THC/CBD, CBD only	Overall EB symptoms	5-point Likert scale	Singel item	Generic	Complete validated questionnaire
Mosallaei et al. 2022	Gentamicin	Disease severity	EBDASI	Multiple item	EB-specific	Complete validated questionnaire

16/18 (88.8%) of the measurement instruments used were validated for the respective outcome. Notably, information on the person who had completed the self-reported questionnaire (i.e., affected individuals themselves, their caregivers or medical staff) was rarely provided. Likewise, contextually relevant aspects of assessment (i.e., day time; procedural versus interval itch; maximum/peak itch intensity versus mean itch scores; symptomatic quality, e.g., neuropathic) were inconsistently specified. Furthermore, information on whether the used instruments were applied in patients’ native language and age-appropriate was missing. A more comprehensive analysis and assessment of validity (i.e., content-, construct-, criterion-related validity) and reliability of study data was infeasible based on published study information. Retrieving additional data on outcome assessment procedures through e.g., contacting corresponding authors was beyond the scope of this research.

Treatment approaches

In total, 17 different treatment approaches were extracted from 19 clinical trials (Table 3). To facilitate a structured assessment, this study categorized treatments by route of application, i.e., either systemic (oral, intravenous or subcutaneous) or local (topical, intralesional injection or transplantation) as well as by modality (i.e., small molecule, cell or protein therapies).

Table 3. Overview of local and systemic treatment modalities reporting on itch outcomes for EB patients.

Author/year	Treatment approach	Dosage and frequency	EB type	Number of patients	Child/adult	Comperator	Primary purpose	Endpoint assessment	Treatment efficacy itch	Treatment efficacy QoL	Treatment efficacy disease severity	Level of evidence
Cell therapy
Local therapies
Venugopal et al. 2013	Allogeneic fibroblasts in 2% albumin suspension	Intralesional injection into wound base (0.25 mL/cm^2 containing 5 × 10^6 fibroblasts or suspension solution alone)	RDEB	5	Adult	Placebo	Wound healing	Month 12	No change VAS local itch between the treatment groups (p > 0.40)	QOLEB improvement not significant (greater improvement in the psychosocial than the functional dimension)	N/A	2
Fivenson et al. 2003	Graftskin therapy	Single transplantation per site (96 sites in total)	EBS (3), JEB (1), RDEB (5)	9	Child	Baseline	Wound healing	Weeks 4 and 12, and between weeks 20 and 28	Decreased pruritus at the graftskin treatment sites in 5 of 9 patients	Improved quality of life	N/A	3
Systemic therapies
Petrof et al. 2015	Allogeneic bone marrow derived mesenchymal stromal cells	Three i.v. infusions (each dose 1–3 × 10^6 cells/kg) on day 0, 7 and 28	RDEB	10	Child	Baseline	Safety	Days 60, 180	Mean difference pruritus score (0–10) from baseline −1.3 (95% CI −4.2, 1.5), −1.3 (95% CI −4.2, 1.5); n = 3	Mean difference PedsQL (child version, lower is improvement) from baseline −5.2 (95% CI −25.6, 15.2), −2.8 (95% CI −18.6, 13.0); n = 5	Mean difference BEBSS from baseline −5.2 (95% CI −10.7, 0.3), −6.9 (95% CI −12.7, −1.1)	3
										Mean difference PedsQL (parent version, lower is improvement) from baseline −4.4 (95% CI −8.1, −0.7), −2.9 (95% CI −7.5, 1.8); n = 8	Mean difference GSS from baseline −2.4 (95% CI −3.4, −1.4), −1.6 (95% CI −2.96, −0.24)
Rashidghamat et al. 2020	Allogeneic bone marrow–derived mesenchymal stromal cells	Two i.v. infusions 2–4 × 10^6 cells/kg on day 0 and 14	RDEB	10	Adult	Baseline	Safety, wound healing	Days 28 and 60	Leuven Itch Scale – Frequency mean reduction 13.89 (95% CI, 3.76–24.02, significant), 18.75 (95% CI, 9.07–28.43, significant) – Severity mean reduction 15.44 (95% CI, 4.47–26.42, significant), 15.16 (95% CI, −1.74 to 32.05, significant) – Consequences of itch mean reduction 12.64 (95% CI, 0.40–24.88, significant), 17.21 (95% CI, 6.40–28.01, significant)	Mean improvement QOLEB at days 28 and 60 of 1.89 (95% CI, –0.87 to 4.65) and 3.13 (95% CI, –0.26 to 6.51)	Mean reduction BEBSS 0.33 (95% CI, −0.3 to 0.97) and 1.61 (95% CI, −0.05 to 3.27)	3
											Minimal change EBDASI
Fujita et al. 2021	Multilineage-differentiating stress-enduring (Muse) cells	Single infusion 1.5 × 10^7 cells (2.98 ± 0.61 × 10^5 cells/kg)	RDEB (2), DDEB (3)	5	Both	Baseline	Wound healing	Week 4	No change on VAS itch; temporary worsening at weeks 12 and 20 throughout a 52 weeks surveillance time	No change	N/A	3
Kiritsi et al. 2021	Allogeneic ABCB5-positive dermal mesenchymal stem cells (allo-APZ2-EB)	Three i.v. infusions 2 × 10^6 cells/kg on day 0, 17 and 35	RDEB	16	Both	Baseline	Disease severity	Week 12	Median reduction NRS itch from baseline by 14.3% (0%–42.9%, not significant); greatest reduction observed at day 35 from baseline by 14.3% (0%–42.9%, p = 0.033)	QOLEB median change from baseline (higher is worsening) 6.5% (–18.8% to 15.4%; not significant)	EBDASI median reduction 3.4% (0.0%–9.4%, not significant) – EBDASI activity score median reduction 13.0% (2.9%–30%, p = 0.049)	3
											IscorEB median reduction 8.1% (–2.7% to 23.8%, not significant) – IscorEB-c median reduction 18.2% (1.9%–39.8%, p = 0.037)
Lee et al. 2021	Allogeneic umbilical cord blood–derived mesenchymal stem cells	Three i.v. injections (1 × 10^6 to 3 × 10^6 cells/kg) day 0, 14, 28	RDEB	6	Both	Baseline	Safety, disease severity	Day 56	Significant reduction in mean itch VAS by −2 points (95% CI, −3.76 to −0.24)	Significant improvement of QOLEB by 6.2 (CI 95%, –8.69, –3.65)	Significant decrease in mean BEBSS by 5.6 points (95% CI, −7.39 to −3.86)	3
											Significant decrease in mean TBSA by 5.4 points (95% CI, −8.14 to −2.61)
Protein therapy
Local therapies
Gorell et al. 2015	Helicoll Collagen I Wound Dressing	Weekly application to wound	RDEB	10	Both	Baseline	Wound healing	Weekly until week 8 after application	Average wound pruritus VAS (0–4) decrease 0.89, p = 0.004	N/A	N/A	2
Small molecules
Local therapies
Schwieger-Briel et al. 2017	Oleogel-S10 and non-adhesive wound dressing	1 cm or 125 µL or 115 mg Oleogel-S10 per cm2 wound area (corresponds to thickness of approximately 1 mm) and covered with a non-adhesive wound dressing (Mepilex^®) on Day 0. Oleogel-S10 was administered at wound dressing changes about every 24 to 48 hours	RDEB (9), DDEB (1)	10	Both	Placebo	Wound healing	Day 14 or 28	No change on wound itch VAS	N/A	N/A	2
Wally et al. 2018	Diacereine 1%	2.0 g of cream (i.e., 20.0 mg of diacerein) daily to defined skin area	EBS	15	Child	Placebo	Blister formation	Week 4	Localized itch VAS: No change (over time or between groups)	No significant change	N/A	2
Murrell et al. 2020	Allantoin 0% (SD-101-0.0)	Once a day to the entire body	EBS (8), RDEB (40), JEB (17)	87	Both	Baseline	Wound healing	Month 3	Itch Man Pruritus Assessment Tool mean improvement −0.7 (SE 0.2) n = 80	N/A	N/A	2
Paller et al. 2020	Allantoin 6% (SD-101-6.0)	Once a day to the entire body	EBS (18), RDEB (119), JEB (32)	Verum: 82 Placebo: 87	Both	Placebo	Wound healing	Day 7 (month 3 data not shown)	Itch Man Pruritus Assessment Tool mean improvement (p = 0.262) Verum −0.5 (SE 0.15) n = 77 Placebo −0.3 (SE 0.14) n = 79	N/A	N/A	2
Guttmann-Gruber et al. 2021	Low-dose calcipotriol ointment	Daily application 0.05 µg/g calcipotriol in ultraphil ointment	RDEB	6	Both	Placebo	Wound healing	Day 28	Improvement of local itch VAS verum to 1.83 vs placebo 5.52 (p < 0.0001)	N/A	N/A	2
Systemic therapies
Chiaverini et al. 2016	Epigallocatechin-3-gallate	Depend on the patient’s weight (from 400 to 800 mg a day)	RDEB	17	Both	Placebo	Blister formation	Month 4	Evolution of VAS itch (end of the period – beginning of the period) non significant improvement p = 0.38 Verum − 1.17 ± 3.53 (n = 12) Placebo 0 ± 2.16 (n = 14)	N/A	N/A	2
Chiou et al. 2020	Serlopitant	Loading dose of 3 tablets (15 mg), followed by 8 weeks of treatment with 5 mg	RDEB (13), JEB (1)	Verum: 7 Placebo: 7	Both	Placebo	Pruritus	Week 8	0.64-point reduction in itch NRS relative to placebo (p = 0.11)	N/A	N/A	2
									No change in itch NRS during dressing changes (p = 0.85)
Neufeld-Kaiser et al. 1997	Cyproheptadine	Standard dosage daily (Physician’s Desk Reference. 48th ed. Montvale, NJ: Medical Economics Data Production Co; 1994:1513)	EBS	10	N/A	Baseline	Blister formation (number of blisters lanced each day)	Week 6	3 patients reported improvement of itch	N/A	N/A	3
Chiaverini et al. 2015	Erythromycin	<10 kg: 250 mg per day; 10–15 kg: 500 mg per day; 15–25 kg: 750 mg per day; 25–35 kg: 1000 mg per day	EBS	6	Child	Baseline	Blister formation	Month 5	No change on VAS itch	N/A	N/A	3
Chen et al. 2021	Omalizumab	150 mg s.c. every 4 weeks for at least 3 cycles	RDEB	11	Both	Baseline	Clinical improvement	At the last treatment cycle (individual)	Reduction VAS pruritus by 42% (p < 0.0001) at last cycle in ten out of eleven patients (one patient terminated the treatment because of poor response)	Decreased DLQI values (lower is improvement) at last cycle, p < 0.0001	Reduced by 15%, p < 0.0005 at last cycle	3
Mosallaei et al. 2022	Gentamicin	Three patients daily i.v. gentamicinsulfate at a dosage of 7.5 mg/kg daily for 14 days Two patients with the high dosage of 10mg/kg daily for 28 days	JEB	5	Child	Baseline	Change in expression of laminin 332	3 months after last dose	Improvement (3), no change (1), worsening (1) on 5-D pruritus scale	Skindex-16 survey (2): Symptom scores stable, downward trends in emotions and functioning components	EBDASI (2): Data incomplete	3
Survey studies
Schräder et al. 2021	Cannabinoid-based medicines (CBM) – THC only, THC/CBD, CBD only	Predominantly inhaled, topical and oral, and sublingual; mostly at least once a day	RDEB (37), DDEB (8), EBS (17), JEB (8)	71	Both	Baseline	Effects on symptoms	N/A	Median pruritus change-score: − 3, IQR: − 1.25 to − 5 [p < 0.001, 95% CI 2.59–3.10])	N/A	46/48 (95.8%) improvement, 1/48 (2.1%) no change, 1/48 (2.1%) worsening of overall EB symptoms on 5-point likert scale	4
									40/44 (90.9) improvement, 3/44 (6.8%) no change, 1/44 (2.3%) worsening on overall pruritus 5-point Likert scale

In addition, two cross-sectional survey studies evaluated a total of nine different treatment approaches (e.g., bathing products, topical products), with at least two to a maximum of 16 different therapeutic products (e.g., bleach, pimecrolimus, hydrocortisone). Notably, the variability and incomplete reporting of substances, their content and composition in addition to inconsistent data on modes of application impaired an unambiguous allocation of the two survey studies to any category.

Treatment outcomes

A comprehensive overview on treatment outcomes is provided in Table 3.

Cell therapy

Local therapies

A statistically non-significant reduction in local itching within the treatment area was reported in five out of nine children with EB (aged 4 months to 14 years) after transplantation of an allogeneic viable bilayer skin construct derived from human neonatal foreskin keratinocytes and dermal fibroblasts. As a limitation, results were reported only verbatim and measurement instruments were not specified by the authors. Consequently, age-appropriateness of the assessment approach remains unclear. Considering the underage study population, the outcomes thus should be interpreted with caution (41).

Similarly, intralesional injection of allogeneic fibroblasts in five adult RDEB patients failed to show a significant reduction in local itch at 12 months, while demonstrating the highest level of evidence (i.e., randomized, placebo controlled testing) in this category (42).

Systemic therapies

Infusion of allogeneic bone marrow derived mesenchymal stromal cells (BM-MSCs) significantly improved mean severity (–15.44 points) as well as frequency (–13.89 points), and consequences of itch (–12.64 points, e.g., lesions from scratching, sleep disturbances) on the Leuven itch scale (range for each item 0–100) in ten adult RDEB patients aged 26–55 years (37). Treatment was associated with a non-significant trend toward improvement of QoL (quality of life evaluation in epidermolysis bullosa, QOLEB) and disease severity (Birmingham Epidermolysis Bullosa Severity Score, BEBSS; Epidermolysis Bullosa Disease Activity and Scarring Index, EBDASI) at day 60 (37).

Intravenous allogeneic human umbilical cord blood derived mesenchymal stem cells (hUCB-MSCs) demonstrated a significant reduction in mean itch by two points on VAS 0–10 in six RDEB patients aged eight to 60 years. In addition, significant mean improvements in QoL and disease severity by 6.2 points in QOLEB and 5.6 points in BEBSS were observed, respectively (39).

Infusion of allogeneic ATP-binding cassette subfamily B5 positive dermal mesenchymal stem cells (ABCB5⁺ MSCs) resulted in a significant median reduction in itch by 14.3% on day 35 (numerical rating scale, NRS), in a cohort of 16 RDEB patients. In this study, a significant improvement in disease severity was only observed in analyses of distinct subscores (instrument for scoring clinical outcomes of research for epidermolysis bullosa IscorEB clinician subscore −18.2%, EBDASI activity score −13.0%), while no relevant change in the QOLEB was determined (38).

A single infusion of allogenic multilineage-differentiating stress-enduring (Muse) cells in two RDEB and three DDEB patients (aged 17–49 years) revealed no improvement in itch after four weeks. In contrast, a temporary aggravation was noted between weeks 12 and 20 (43).

Protein therapy

Topical applications of protein therapies have been investigated for a potential effect on local itching only. For this purpose, an acellular skin substitute matrix with purified type I bovine collagen was applied weekly to ten RDEB patients with chronic wounds with six-months to six-years of persistence. Seven participants (aged 8–24 years) completed the study, and reported a significant reduction of average wound itch on week eight (–0.89, VAS 0–4) compared to baseline (44).

Small molecules

Local therapies

All five studies identified demonstrated a good level of evidence (OCEBM level 2). Two publications reported on the same registered clinical trial (ESSENCE study), referring to the largest study cohort analyzed in this review, with a total of 169 patients (SD-101 cream, NCT02384460) (45,46). One of these clinical investigations reported only on the vehicle arm (allantoin 0%, SD-101-0.0) (46), while the other publication reported on both, the vehicle and verum (allantoin 6%, SD-101-6.0) arms (45). In total, the trial included 82 patients in the verum cohort, while 87 participants had received placebo. Both, children and adults, with either EBS, DEB, or JEB were enrolled. Trial participants applied the topical treatment to their entire integument on a daily basis. Compared to baseline, antipruritic effects were detected in both cohorts already at the first visit on day seven as well as after three months, with a mean improvement of −0.5 and −0.3 (Itch Man Pruritus Assessment Tool 0–5) for verum and placebo on day seven, respectively. After three months, mean improvement was −0.7 for placebo, while for this timepoint no verum data was reported. Overall, treatment effects on itch did not reach statistical significance when comparing verum to placebo (45,46).

In a study on daily application of low-dose calcipotriol in ultraphil ointment in six patients with RDEB, a significant reduction in wound itching down to 1.83 (VAS, 0–10) was demonstrated on day 28 compared to placebo (5.52, p < 0.0001) (47).

Two additional studies, oleogel-S10 in ten DEB patients and diacereine 1% in 15 EBS children, revealed no reduction in localized itching within the treatment area (48,49).

Systemic therapies

In a single-arm, open-label study, the humanized monoclonal anti-IgE antibody omalizumab was subcutaneously administered in eleven children older than seven years and adults with RDEB and high serum IgE levels. The treatment regime was 150 mg every four weeks. A significant itch reduction of 42% compared to baseline (VAS 0–10, p < 0.0001) was observed in ten patients that completed the trial and received omalizumab for at least three treatment cycles. Itch relief was accompanied by a significant improvement in QoL (Dermatology Life Quality Index DLQI, p < 0.0001) and disease severity (BEBSS reduction by 15%, p < 0.0005) after the last cycle (50). As a drawback, the QoL measurement instrument (DLQI) was not validated for underage subjects (n = 4/10; 40% of the cohort that completed the trial) (51).

Of the six publications on systemically administered small molecules, one clinical trial focused on itch relief in patients with RDEB (n = 13) and JEB (n = 1) upon treatment with the neurokinin 1 receptor (NK1) receptor antagonist serlopitant. It is the only clinical trial in this systematic review that evaluated itch as a primary endpoint. However, the randomized, double-blind, placebo-controlled trial revealed only a non-significant trend toward reduction of generalized itch after eight weeks on the NRS scale, and localized itching at sites of dressing changes did not improve (52).

Orally administered epigallocatechin-3-gallate (green tea extract) showed a non-significant itch reduction in 17 RDEB patients (53). In a study on cyproheptadine, using ‘standard dosing’ (which, however, was not further specified in the publication), three out of ten patients reported ‘relief of itching’ (not further quantified) after six weeks (54).

Daily intravenous infusion of gentamicin for up to 28 days in five children with JEB (two with intermediate JEB and three with severe JEB) did not show consistent improvement of itch, QoL or disease severity (55). Similarly, six children with EBS who were given erythromycin orally for five months did not experience any relief from itching (56).

Survey studies

The survey studies retrospectively analyzed patient-reported data on various topical and systemic treatment modalities. Both studies were classified as having a low level of evidence (OCEBM level 4), which reflects inherent limitations in terms of objectivity (lack of systematic application and evaluation, absence of reproducible and comparable conditions). Danial et al. included 146 individuals with EB to examine the impact of various products, applications, and techniques on itch alleviation. Patients were asked to name the treatments they had used and to rate their effectiveness in relieving itch (57). Commonly used treatments included ‘creams,’ ‘topical corticosteroids,’ ‘oils,’ oral hydroxyzine, topical diphenhydramine and vaporizing rubs. Significantly lower itch intensity correlated with higher frequency of ‘creams’ (p = 0.05) or ‘lotions’ (p = 0.04) use but was independent of the product used (40).

Another survey study was conducted to investigate the effect of cannabinoid-based medicines (CBM) on EB-specific symptoms in 71 patients with EBS, JEB, and DEB. Patients’ reported data revealed a median improvement on NRS by three points (IQR: −1.25 to −5, p < 0.001, 95% CI 2.59–3.10) upon CBM intake (which was predominantly inhaled, followed by topical and oral, and sublingual application). 40/44 (90.9%) and 46/48 (95.8%) patients indicated an improvement of generalized itch and overall EB symptoms on a 5-D Likert scale, respectively (58). Notably, the authors were unable to objectively verify the Δ9-tetrahydrocannabinol (THC) content and composition of CBM products, which, however, impacts the mechanism of action and scope of potential side effects (59).

Adverse events

Of the 21 trials assessed, only 13 reported on treatment-related adverse events. Overall, there was no statistically significant increase of adverse events reported for verum cohorts. (Table 4). In one study on ABCB5⁺ MSCs, two serious adverse events (classified as hypersensitivity reactions) were reported, which resolved shortly after withdrawal of the treatment (38).

Table 4. Adverse events reported in clinical studies.

Author/year	Treatment approach	Number of patients	Adverse events	Type of adverse event	Treatment related adverse events
Cell therapy
Local therapies
Venugopal et al. 2013	Allogeneic fibroblasts in 2% albumin suspension	5	N/A	N/A	0
Fivenson et al. 2003	Graftskin therapy	9	1	N/A	Postoperative wound infection (1)
Systemic therapies
Petrof et al. 2015	Allogeneic bone marrow derived mesenchymal stromal cells	10	163	Most frequent: DMSO odor (28), skin/mucosal blsiters/wounds (18), respiratory tract infections (10), diarrhea (9), skin infections (7), vomiting (6), pruritus (4)	40 – DMSO odor (28), Skin/mucosal blisters/wounds (2), pruritus (2), nausea (2), fine hair growth (1), milia (1), rash (1), abdominal pain (1), increased appetite (1), bradycardia (1)
Rashidghamat et al. 2020	Allogeneic bone marrow–derived mesenchymal stromal cells	10	9	N/A	0
Fujita et al. 2021	Multilineage-differentiating stress-enduring (Muse) cells	5	N/A	Stomach pain (grade 3), acquired lacrimal stricture (grade 2), fever (grade 1), gastroenteritis (grade 1), upper respiratory tract infection (grade 1), paraesthesia of the upper arms (grade 1)	Paresthesia 24h after infusion resolved within 14 days (1)
Kiritsi et al. 2021	Allogeneic ABCB5-positive mesenchymal stem cells (allo-APZ2-EB)	16	72	Most frequent: Infection or infestation (8), gastrointestinal disorders (6), blood and lymphatic disorders (4), eye disorders (2), hypersesnitivity (2)	Hypersensitivity reactions (2, serious but resolved without sequelae shortly after withdrawal of treatment), mild lymphadenopathy (1)
Lee et al. 2021	Allogeneic umbilical cord blood–derived mesenchymal stem cells	6	13	Wound infection (4), general condition deterioration (1), dizziness (1), acute gastritis (1), nausea (1), fever (1), H. zoster (1), upper resp tract infection (1), lacrimal duct obstruction (1), epistaxis (1)	2
Protein therapy
Local therapies
Gorell et al. 2015	Helicoll Collagen I Wound Dressing	10	N/A	Most frequent: Pain during dressing changes (5), pruritus at a target wound (4), and foul-smelling exudate (2)	N/A
Small molecules
Local therapies
Schwieger-Briel et al. 2017	Oleogel-S10 and non-adhesive wound dressing	10	The substance was tolerated very well without any complaints of stinging or burning	N/A	N/A
Wally et al. 2018	Diacereine 1%	15	Verum 5 (none in the treatment area) Placebo 8 (none in the treatment area)	Most frequently: Pruritus (verum 1, placebo 0), Fever (verum 1, placebo 3)	0
Murrell et al. 2020	Allantoin 0% (SD-101-0.0)	87	N/A	N/A	N/A
Paller et al. 2020	Allantoin 6% (SD-101-6.0)	Verum: 82 Placebo: 87	Verum 71 Placebo 61	AEs Occurring in ≥ 5% of Patients, most frequent: worsening of pruritus (verum 9, placebo 8), nasopharyngitis (verum 11, placebo 3)	Verum 15 (TrAEs occurring in ≥2% of Patients, most frequent: generalized or localized pruritus 6, generalized pruritus 2, wound 2, blister 2) Placebo 19 (TrAEs occurring in ≥2% of Patients, most frequent: generalized or localized pruritus 5, generalized pruritus 0, wound 2, blister 2)
Guttmann-Gruber et al. 2021	Low-dose calcipotriol ointment	6	N/A	N/A	0
Systemic therapies
Chiaverini et al. 2016	Epigallocatechin-3-gallate	17	Verum: 26 Placebo: 16 p = 0.256	Most frequent: Upper respiratory tract infection (verum 5, placebo 3), diarrhea (2), esophageal pain (verum 2, placebo 0), epidermoid carcinoma (verum 2, placebo 1)	Verum: 5 Placebo: 3
Chiou et al. 2020	Serlopitant	Verum: 7 Placebo: 7	N/A	N/A	Most frequent: nausea (verum 2, placebo 1), diarrhea (verum 1, placebo 0), headache (1)
Neufeld-Kaiser et al. 1997	Cyproheptadine	10	N/A	Drowsiness, increased appetite, irritability	N/A
Chiaverini et al. 2015	Erythromycin	6	5	N/A	0
Chen et al. 2021	Omalizumab	12	N/A	N/A	N/A
Mosallaei et al. 2022	Gentamicin	5	0	N/A	N/A
Survey studies
Schräder et al. 2021	Cannabinoid-based medicines (CBM) – THC only, THC/CBD, CBD only	71	45	Dry mouth (20), cough/wheezing (13), dry/red eyes (12), fatigue (10), dizziness/lightheadedness (7), paranoia (3), problems with memory/attention (3), problems with coordination (3)	N/A
Danial et al. 2015	Bathing products	146	N/A	N/A	N/A
	Moisturizers
	Topical products
	Oral medication taken for pruritus
	Oral medication taken for pain
	Oral medication taken for sleep
	Dressings
	Complementary/alternative methods

Discussion

In this systematic review we gathered data on the efficacy and safety of itch treatments from EB studies with the aim to improve the scientific and clinical decision making process for targeting one of the most burdensome and therapeutically challenging symptoms in this population. Twenty-one studies were classified eligible for analyses. However, methodological shortcomings inherent to rare disease clinical research, the heterogeneity and incomplete reporting of outcomes, as well as lacking information on applied measurement instruments hampering validity assessment, significantly impaired efforts of data pooling and comparative assessment.

We identified a significant heterogeneity of outcomes and outcome measurement instruments reported in the literature for the assessment of the impact of study interventions on itch. Variations in trial design (e.g., endpoints, inclusion and exclusion criteria, frequency and time points of measurements) as well as incomplete information on contextually relevant aspects of assessment procedures (e.g., day time, symptom quality, questionnaire respondent, age and language validation) impaired subtype- and age- specific stratification of treatment efficacy as well as feasibility of data pooling and cross trial comparison/comparative analyses. Occasionally, measurement instruments were used that lacked validation for distinct subcohorts of the study population (50), which may have had a confounding impact on effect sizes (60,61).

The diversity in key trial features is in concordance with data from a recent scoping review revealing great heterogeneity among 1,280 reported outcomes pertaining to 80 different outcome domains (the ‘what to measure’) for EB. Moreover, these outcome domains had been assessed by a diverse range of in total 200 measurement instruments (the ‘how to measure’) in 207 studies (29). In the context of EB, which features a high intra- and inter-individual disease heterogeneity per se, such heterogeneity constitutes a major shortcoming to clinical research efficacy. Considering the growing trial landscape competing for a limited number of eligible patients, initiatives to foster a more homogenous reporting and assessment of core outcomes are thus highly warranted (62).

A small number of eligible patients is a prototypic feature of EB trials. It requires an optimal balance between maximum bias reduction (through e.g., restrictive inclusion criteria, timing and frequency of measurements, blinding, randomization) and feasibility (acceptable trial burden e.g., invasiveness of investigations, access to verum arm, travel burden, individual risk to benefit assessment) for the EB patient cohort that is typically severely affected, vulnerable, geographically dispersed and often underage (Table 1) (28). As the inter- and intra-individual heterogeneity of patient cohorts and a variable disease course additionally favor random imbalance in covariates and bias, trials were often designed to be open-label (n = 12/21, 57.1%), and allocated to a single group (n = 9/21, 42.9%) while baseline was serving as comparator (n = 14/21, 66.7%). Consequently, the overall formal quality of data evidence of the assessed studies was mediocre. Moreover, randomization was often infeasible or actively omitted (n = 12/21, 57.1%) to ensure distribution of verum to all participants, thereby raising recruitment, compliance, adherence and trial efficacy. Being less affected by these factors, reporting and attrition bias, however, were often low for the reported outcomes (Figure 2).

Despite a wide variety of topical and systemic treatment modalities (n = 83) that have been evaluated in this work, statistically significant anti-itch effects were observed in only eight studies (38.1%). These trials mainly involved (R)DEB patients, the EB cohort who is reported to suffer most from pruritus (Table 3) (12). Notably, these studies were actually powered for primary endpoints other than symptomatic itch relief, assessing e.g., wound healing, disease severity, safety, and blister formation. Conversely, the only one trial reporting on an interventional impact on itch as primary endpoint, failed to reach statistical significance for this outcome (52). These data further underscore that the formal quality of data evidence for efficacy of antipuritic treatment modalities is currently sparse.

With regard to topical therapeutics, significant improvement of itch within the treatment area was only observed upon weekly application of collagen type I wound dressings as well as daily low-dose calcipotriol ointment (44,47). Notably, narrow therapeutic impact or non-superiority of topical applications compared to control has also been linked to the observation that regular (i.e., intensified) per protocol application of topical (skin care) products commonly serving as control or placebo, has the potential to effectively and independently reduce itch. Occasionally, control thereby may even outperform verum (45,46,57).

Remarkably, oral agents did not yield statistically significant itch relief in this analysis, despite mediating a presumably generalized, systemic effect.

Considering the interpretation of reported effect sizes, we want to emphasize that statistical significance per se does neither imply clinical meaningfulness (i.e., a change that the patient can discern) nor a minimally important difference. The latter refers to the point at which a change in symptoms or experienced disease burden becomes large enough that most patients would want to change their therapy to get this benefit (63). This outcome also covers aspects of patient-centeredness, particularly with regard to treatment burden (such as feasibility and invasiveness of route of administration considering e.g., tolerability of oral intake in the background of mucosal EB lesions or tolerability of topical applications to large, vulnerable skin areas). Data supporting such an impact of any therapeutic modality was not extractable.

To corroborate observed effect sizes, value assessment may include the correlation of symptomatic itch relief, QoL and disease severity outcomes, as these scores have shown a strong interdependence (22,23). However, these items were simultaneously reported in only a limited number of studies to underpin patient-reported data. Improvement of parameters for itch (VAS), QoL (DLQI, QOLEB) and disease severity (BEBSS; total body surface area, TBSA) was concurrent and significant in two trials (omalizumab, hUCB-MSCs), and for itch and disease severity in one trial (ABCB5⁺ MSCs), respectively (38,39,50).

Overall, these analyses revealed no statistically significant increase of adverse events reported for verum cohorts (Table 4). However, adverse events were only inconsistently and incompletely reported, which impeded a thorough risk-benefit assessment. Such an analysis, however, would be of particular relevance considering the vulnerable and often underage study population and need for long-term therapy. Those adverse events reported were rarely rated serious, with the exemption of hypersensitivity reactions associated with infusions of ABCB5⁺ MSCs. The latter, however, were observed only in isolated cases and without any reported sequelae.

Individualized approaches employing (potentially) predictive biomarkers (e.g., pruritogenic lead cytokines used as surrogates for the response to certain immunomodulatory therapies) constitute an innovative approach to effectively target itch mediators. This option seems particularly attractive in view of the rapidly expanding immunomodulatory therapy options. Profiling of inflammatory signatures and identification of lead cytokines in patients with EB pruriginosa prompted the administration of omalizumab (50). Furthermore, repurposing of IL-4/IL-13 antibody dupilumab, janus kinase inhibitors (i.e., baricitinib, upadacitinib), as well as the phosphodiesterase-4-inhibitor apremilast, induced a self-reported itch reduction, however, hitherto only in case reports and case series (14,40,58,64). Methodological issues (e.g., detection, performance, and reporting bias; optimal dosing), as well as dose finding and determination of long term outcomes remain to be addressed by large scaled studies.

Conclusions

As translational perspectives and therapeutic potential for EB broadens, a steadily growing number of innovative therapeutic components from bench as well as repurposed drugs have entered clinical testing to correct, modify or palliate the symptoms of this devastating genodermatosis, including burdensome itch (65). Our data underscore the considerable lack of high level evidence based on large-scale studies sufficiently powered to demonstrate a clinically meaningful effect. In addition, the use of a variety of different outcome measurement instruments impeded comparative assessment of heterogeneous and small-sized study populations. Therefore, current evidence is insufficient to unequivocally support recommendations on treatment prioritization for EB associated itch based on efficacy and safety data from clinical trials. Against this background, common efforts to boost international collaboration, coordination and consensus, including consistent use of core outcomes with adequate and validated measurement instruments, are crucial for efficient data generation out of size-limited EB patient cohorts. A collaborative approach will likewise foster the identification and validation of pilot data on predictive biomarker signatures and therapeutic targets, which hold the potential to more effectively meet the challenges of disease heterogeneity by addressing patient-specific pathogenic traits of EB.

Author contributions

T.W., C.PU, E.W.H.K., S.W., C.P., M.C.B. and M.L. developed the concept and designed the systematic review. T.W., C.PU., E.W.H.K., and M.L. contributed to acquisition, analysis, or interpretation of data. T.W., C.PU., and M.L. wrote the initial manuscript. All authors contributed to critical revision of the manuscript for important intellectual content. All authors read and approved the final manuscript. M.L. and M.C.B. provided supervision.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this review are available from the corresponding authors upon reasonable request.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.