Milestone events in recessive dystrophic epidermolysis bullosa: findings of the PEBLES study

Abstract

Background

In recessive dystrophic epidermolysis bullosa (RDEB), complications like oesophageal strictures, hand contractures, cardiomyopathy and cutaneous squamous cell carcinoma (SCC) may develop. These complications necessitate procedures such as oesophageal dilatation (OD), gastrostomy tube placement and hand surgery.

Objectives

To determine the prevalence and age of onset of milestone events by RDEB subtype, specifically dysphagia, first OD, first gastrostomy tube, first hand surgery, cardiomyopathy, first SCC and death.

Methods

The Prospective Epidermolysis Bullosa Longitudinal Evaluation Study (PEBLES) is a register study of individuals with RDEB that records comprehensive EB- and non-EB-related health information. Age of onset and prevalence of milestone events were analysed in the full cohort with all RDEB subtypes and by RDEB subtype.

Results

Dysphagia occurred in 85% of the 62 total participants, including all with the severe (RDEB-S) and inversa (RDEB-Inv) subtypes. OD was also frequent, in 69% overall (92% RDEB-S, 89% RDEB-Inv). All events were most frequent and occurred earliest in RDEB-S (except cardiomyopathy), with a median age of dysphagia, OD and gastrostomy tube placement in the first decade. Frequent and early dysphagia and OD in RDEB-Inv may suggest this subtype before characteristic flexural skin changes manifest. Of the participants with RDEB-S, 35% had a first SCC at a median age of 27.8 years. Seven participants died during the 10-year study; the median age for the six with RDEB-S was 36 years, and causes included sepsis, metastatic SCC and complications of refeeding syndrome.

Conclusions

Our results detail the frequency and prevalence of important milestone events by RDEB subtype, informing prognostication, increasing understanding of the natural history of RDEB to help inform study endpoints, and potentially serving as proxy control data for future clinical trials.

What is already known about this topic?

• Dysphagia, hand contractures, squamous cell carcinoma, cardiomyopathy and premature death are complications of recessive dystrophic epidermolysis bullosa (RDEB).

• These may require oesophageal dilatation, gastrostomy tube placement and hand surgery.

What does this study add?

• The prevalence and frequency of these milestone events were explored in patients with RDEB classified by disease subtype.

• Milestone events are more common and occurred earliest in severe RDEB (RDEB-S).

• In RDEB-S, 23% of participants died over an 8-year period at a median age of 36 years.

Epidermolysis bullosa (EB) is a group of genetic diseases characterized by skin and mucosal fragility. In people with EB, minor trauma causes blistering, which can result in delayed healing, chronic wounds and, in some forms, scarring.^1,2 There are four main types of EB, i.e. simplex, junctional, dystrophic and Kindler EB, which result from pathogenic variants in at least 16 different genes. Most of these genes encode structural proteins at the dermoepidermal junction.^1,2 Recent data indicate a prevalence for all types of EB in the UK of 34.8 per million population.³

Dystrophic EB can be autosomal dominantly or recessively inherited and results from pathogenic variants in COL7A1. This gene encodes type VII collagen, which is the major constituent of anchoring fibrils at the dermoepidermal junction.^4,5 Different subtypes of recessive dystrophic EB (RDEB) are recognized on the basis of their clinical features and severity, but mucocutaneous scarring after injury is ubiquitous. RDEB has an estimated prevalence of 3.3 per million in the UK.³

In the severe form, RDEB-S, skin blistering is generalized and leads to widespread wounds and internal complications, while the intermediate form, RDEB-I, is also generalized but less severe.¹ In the inversa type of RDEB (RDEB-Inv), blistering and wounds develop a predilection for flexural sites from adolescence onwards and mucosal involvement is prominent.⁶ An uncommon form of RDEB is associated with delayed onset of intense itching and secondary skin damage from scratching, and is termed the pruriginosa subtype, RDEB-Pru.⁷ Pretibial RDEB (RDEB-PT) is a rare subtype that often develops later in life with localized blistering and scarring predominantly on the shins.⁸

In all forms of RDEB, skin and mucosal damage from minor trauma can lead to sequelae such as oesophageal and urethral strictures, symblepharon and formation of corneal pannus, pseudosyndactyly of the hands and feet, and large joint contractures.⁹ Furthermore, patients are at an increased risk of developing multiple aggressive primary cutaneous squamous cell carcinomas (SCCs), especially in RDEB-S, where they develop from the second decade onwards and are the leading cause of mortality.^10–13 Cardiomyopathy is a rare but recognized complication with a multifactorial aetiology, which is also mostly observed in RDEB-S.^14–17

Such complications are well recognized, and multidisciplinary standard of care aims to monitor for and manage these.^9,16 However, there are few robust data about the onset and progress of these complications in the different types of RDEB, in which severity and phenotype can vary significantly. Previous studies have tended to report complications of all dystrophic EB (DEB) or all RDEB and thereby do not provide granular detail by subtype.

Our prospective register study, the Prospective Epidermolysis Bullosa Longitudinal Evaluation Study (PEBLES), was established to collect detailed information about the natural history of the different forms of RDEB across people of all ages, with an initial review followed by regular follow-up evaluations of participants. Through these reviews, which include specific RDEB-related health issues, patient- and family-reported outcomes, disease severity scores and health economic information, a detailed overview of the disease by RDEB subtype and by age can be assimilated. The data gathered will help apprize prognostication, inform endpoints for studies, and provide proxy control data for clinical trials in the future.

Here, we present the PEBLES findings for the age and frequency of milestone events by RDEB subtype, specifically the onset of dysphagia, first oesophageal dilatation (OD), first gastrostomy tube placement, first hand surgery for contracture release, first SCC, onset of cardiomyopathy and death.

Patients and methods

Study population

PEBLES is a prospective register study established in 2014 to delineate the natural history of RDEB. Participants were recruited from the London EB centres, Great Ormond Street Hospital (children) and Guy’s and St Thomas’ Hospital (adults). An initial index review was performed, with follow-up reviews 6-monthly in those aged < 10 years and annually in individuals aged ≥ 10 years, updating information since the previous review. RDEB was confirmed from genetic testing and/or skin biopsy and was subtyped according to the clinical features. Each review recorded EB- and non-EB-related health issues, itch, pain, quality of life and disease severity scores, laboratory and imaging results, and data regarding costs of care.

Pseudonymized data with the date of birth retained to report participants’ age at each review were recorded in a Research Electronic Data Capture (REDCap) database. PEBLES was ethically approved by the UK Research Ethics Committee and Health Research Authority (IRAS 142032).

Milestone events

Data pertaining to specific milestone events across different subtypes of RDEB are reported for the period November 2014 to November 2022. Specifically, age was recorded at the onset of dysphagia, first OD, first gastrostomy tube placement, first contracture release hand surgery, first cutaneous SCC, onset of cardiomyopathy and death. Also, the frequencies of these events are reported in participants with different RDEB subtypes. Most milestone events were reported retrospectively at the first review.

Statistical analysis

All statistical analyses were conducted using the R programming language version 4.1.x and newer.¹⁸ In particular, this includes the computation of quartiles and medians, for which base R’s summary command was used.

Results

Prevalence of milestone events

Data from 62 participants were analysed, including 26 individuals with RDEB-S, 22 with RDEB-I, 9 with RDEB-Inv, 4 with RDEB-Pru and 1 with RDEB-PT (Table 1). The prevalences of milestone events across all RDEB collectively and by RDEB subtype are shown in Figure 1 and Table 2.

Figure 1

Prevalence of milestone events by recessive dystrophic epidermolysis bullosa (RDEB) subtype. CM, onset of cardiomyopathy; DYS, onset of dysphagia; GT, first gastrostomy tube placement; HS, first hand surgery for contracture release; OD, first oesophageal dilatation; SCC, first squamous cell carcinoma. RDEB severe (RDEB-S), intermediate (RDEB-I), inversa (RDEB-Inv) and pruriginosa (RDEB-Pru). Pretibial (RDEB-PT) is included in All RDEB.

Table 1

Participant characteristics at index review by recessive dystrophic epidermolysis bullosa (RDEB) subtype (N = 62)

Category	All RDEB	RDEB-S	RDEB-I	RDEB-Inv	RDEB-Pru	RDEB-PT
Participants	62	26	22	9	4	1
Age group (years)
0–9	10 (16)	8 (31)	2 (9)	0	0	0
10–17	5 (8)	3 (12)	2 (9)	0	0	0
18–39	24 (39)	12 (46)	4 (18)	6 (67)	2 (50)	0 (0)
≥ 40	23 (37)	3 (12)	14 (64)	3 (33)	2 (50)	1 (100)
Age (years)	32 (21–48)	23 (7–31)	46 (29–60)	36 (28–48)	46 (35–57)	69 (69–69)
Sex
Female	34 (55)	13 (50)	14 (64)	6 (67)	1 (25)	0
Male	28 (45)	13 (50)	8 (36)	3 (33)	3 (75)	1 (100)
Ethnicity
White	52 (84)	19 (73)	20 (91)	8 (89)	4 (100)	1 (100)
Asian	7 (11)	5 (19)	1 (5)	1 (11)	0	0
Mixed	2 (3)	2 (8)	0	0	0	0
Other	1 (2)	0	1 (5)	0	0	0
Number of reviews	6 (4–7)	7 (5–9)	6 (3–7)	7 (7–7)	5 (3–6)	5 (5–5)
Period of reviews (years)	7 (4–8)	7 (5–8)	7 (3–8)	7 (7–7)	5 (3–8)	5 (5–5)

The results are presented as the median (interquartile range) or n (%). Subtypes: I, intermediate; Inv, inversa; Pru, pruriginosa; PT, pretibial; S, severe.

Table 2

Prevalence of milestone events by recessive dystrophic epidermolysis bullosa (RDEB) subtype (N = 62)

Subtype	All RDEB	RDEB-S	RDEB-I	RDEB-Inv	RDEB-Pru
Participants	62	26	22	9	4
Milestone eventa
Dysphagia	53 (85)	26 (100)	16 (73)	9 (100)	2 (50)
Oesophageal dilatation	43 (69)	24 (92)	11 (50)	8 (89)	0
Gastrostomy tube	14 (23)	12 (46)	1 (5)	1 (11)	0
Hand surgery for contracture release	21 (34)	15 (58)	6 (27)	0	0
Squamous cell carcinoma	13 (21)	9 (35)	3 (14)	0	1 (25)
Cardiomyopathy	2 (3)	1 (4)	1 (5)	0	0
Death	7 (11)	6 (23)	1 (5)	0	0

The results are presented as n (%). Subtypes: I, intermediate; Inv, inversa; Pru, pruriginosa; S, severe. Pretibial RDEB (one patient) is included in all RDEB; no milestone events were observed. ^aOnset or first occurrence.

Dysphagia was common, with a prevalence of 85% for all RDEB; it occurred in all of the participants with RDEB-S and RDEB-Inv, but was relatively less frequent in those with RDEB-I and RDEB-Pru (73% and 50%, respectively). Reflecting the high frequency of dysphagia, 69% of all participants with RDEB had had at least one OD, particularly those with RDEB-S (92%) and RDEB-Inv (89%). Gastrostomy tube insertion was reported for just under one-quarter of patients across all RDEB but was most prevalent in RDEB-S at 46% of individuals.

First hand surgery for contracture release was also most prevalent in RDEB-S (58%), compared with 34% overall. In total, 35% of participants with RDEB-S had a first SCC, but this was also a complication for 14% and 25% of individuals with RDEB-I and RDEB-Pru, respectively. Cardiomyopathy was uncommon, occurring in just one individual each with RDEB-S and RDEB-I.

Overall, 11% of study participants had died during the 10-year reporting window: six participants with RDEB-S (two from metastatic SCC, two from sepsis, one from cerebral oedema from refeeding syndrome, one cause unknown) and one with RDEB-I (from metastatic SCC).

Age at onset of milestone events

The median age at the onset of the first milestone event by all RDEB and by RDEB subtype is shown in Table 3. Figure 2 illustrates median ages for participants with RDEB-S. All milestone events except the onset of cardiomyopathy occurred earliest in participants with RDEB-S. In particular, in this subgroup, the median ages (with interquartile range, IQR) at the onset of dysphagia, first OD and first gastrostomy tube insertion were within the first decade of life, at 2.6 (1.3–5.9), 8.3 (4.3–21.8) and 6.4 (4.8–9.8) years, respectively. Onset of dysphagia and first OD occurred earlier in patients with RDEB-Inv than in those with RDEB-I, reflecting the predilection for mucosal involvement in this subtype.

Figure 2

Median age of milestone events in severe recessive dystrophic epidermolysis bullosa. CM, cardiomyopathy; GT, first gastrostomy tube placement; HS, hand surgery for contracture release; OD, oesophageal dilatation; SCC, squamous cell carcinoma.

Table 3

Age (years) at milestone events by recessive dystrophic epidermolysis bullosa (RDEB) subtype (N = 62)

Characteristics	All RDEB	RDEB-S	RDEB-I	RDEB-Inv	RDEB-Pru
Participants	62	26	22	9	4
DYS	5.5 (2.5–15.4)	2.6 (1.3–5.9)	10.9 (4.8–26.8)	9.0 (4.5–16.0)	20.8 (13.9–27.6)
OD	18.3 (6.5–29.3)	8.3 (4.3–21.8)	42.9 (18.5–45.3)	23.4 (13.3–29.6)	–
GT	7.0 (5.5–10.3)	6.4 (4.8–9.8)	9.3 (9.3–9.3)	14.6 (14.6–14.6)	–
HS	18.0 (7.4–20.9)	13.1 (6.5–19.7)	19.6 (18.3–23.3)	–	–
SCC	31.2 (26.6–41.8)	27.8 (24.9–33.6)	41.8 (35.2–56.7)	–	42.1 (42.1–42.1)
CM	24.2 (23.0–25.5)	26.7 (26.7–26.7)	21.7 (21.7–21.7)	–	–
Death	42.4 (28.6–50.9)	36.0 (28.2–46.6)	61.2 (61.2–61.2)	–	–

The results are presented as the median (interquartile range). CM, onset of cardiomyopathy; DYS, onset of dysphagia; GT, first gastrostomy tube placement; HS, first hand surgery for contracture release; OD, first oesophageal dilatation; SCC, first squamous cell carcinoma. Subtypes: I, intermediate; Inv, inversa; Pru, pruriginosa; S, severe. Pretibial RDEB (one patient) is included in all RDEB; no milestone events were observed.

The ages at first gastrostomy tube insertion were 9.3 years and almost 15 years in the single participants with RDEB-I and RDEB-Inv, respectively, who had this procedure. The median age of a first SCC in RDEB-S was 27.8 years (24.9–33.6), compared with a later onset in RDEB-I and RDEB-Pru (both approximately 42 years). The two participants with cardiomyopathy first experienced this in their 20s. The median age of death in the six participants with RDEB-S was 36.0 years (28.2–46.6), compared with 61 years in the single individual with RDEB-I who died during the study.

Discussion

This study delineated the frequency of significant milestone events (onset of dysphagia, first OD, first gastrostomy tube placement, first hand surgery, first cutaneous SCC, onset of cardiomyopathy and death) in a cohort of paediatric and adult patients with RDEB stratified by disease subtype. Our results indicate that many of these events and complications occur frequently and early on in life, particularly in individuals with RDEB-S.

Dysphagia and oesophageal strictures are well-recognized complications of DEB,^16,19–22 although breakdown by DEB or RDEB subtype has only rarely been explored in detail. The National EB Registry (NEBR) collected data from 3280 patients with EB in the USA between 1986 and 2002.²³ Unlike most studies exploring clinical features and complications of EB, information was stratified by RDEB subtype, with the prevalences of dysphagia reported as 94%, 61% and 100% in individuals with RDEB-S, RDEB-I and RDEB-Inv, respectively, similarly to our findings (100%, 73% and 100%).²² Data on the age of onset of dysphagia were not presented.

There were differences in the frequencies of patients having had a prior OD. However, at 36%, 28% and 71% for RDEB-S, RDEB-I and RDEB-Inv, respectively,²⁴ these differed from our frequencies of 92%, 50% and 89%. These differences may reflect a shift in the management of OD in EB over time, moving to a safe balloon technique as the standard of care,^25,26 as opposed to the older bougienage technique, which is associated with greater morbidity.^27–29

In another US study, Feinstein et al. reported 644 patients with EB, of whom 283 had RDEB over a 6.5-year period; 72% of all participants were under 20 years of age.³⁰ Of the RDEB group, reported altogether, 56% had had an OD, with a median age of first procedure at 6.6 years.³⁰

In contrast, our study reported an OD in 69% of all participants, with a median age of 18.3 years (IQR 6.5–29.3). The greater prevalence and later timing probably reflect the higher age of participants in PEBLES (for all RDEB, the median age of the first review was 32 years, IQR 21–48). This indicates longer disease follow-up during which an OD might have been possible. Also, older participants might not have been offered OD as routinely when younger compared with the current standard of care. Gastrostomy tubes had been placed in 37% of Feinstein’s series at a median age of 5.5 years,³⁰ compared with 22.6% and a median age of 7.0 years (IQR 5.5–10.3) in our study.

Another study followed 32 children with EB, of whom 27 had generalized RDEB (subtype unspecified). They had gastrostomies placed at a median age of 7.3 years (IQR reported as ± 6.3 years).³¹ Again, this may reflect changes in practice with more frequent gastrostomy tube insertion in recent years as standard EB practice. This is supported by a recent questionnaire-based study of caregivers, which included 15 patients with DEB from a total of 20 patients with EB; the average age at gastrostomy tube insertion was 2.7 years (SD 2.1).³²

The NEBR data demonstrated a high frequency of pseudosyndactyly across the RDEB subtypes, present in 95%, 51% and 41% of patients with RDEB-S, RDEB-I and RDEB-Inv, respectively.³³ There was an earlier onset in RDEB-S compared with the other subtypes.³³ Feinstein et al. reported hand surgery in 22% of their RDEB group at a median age of 8.1 years.³⁰ In our study, a greater proportion of 34% and a later median age of 18.0 years (IQR 7.4–20.9) were found across all patients with RDEB, but hand surgery was more common (58%) and earlier (median 13.1 years, IQR 6.5–19.7) in the RDEB-S group. The NEBR also reported greater hand surgery rates for RDEB-S (61%) than for RDEB-I (24%) and RDEB-Inv (18%).³³ Again, this may reflect a change in practice with less frequent hand surgery over time due to recurrence of contractures and a more conservative approach.

The cumulative risk of developing cutaneous SCC in patients with RDEB from the NEBR was highest in individuals with the severe subtype, with risk of having at least one SCC increasing from 8% at age 20 years to 68% by 35 years and 90% by 55 years.¹⁰ A smaller series of 11 patients with RDEB-S showed a 26% cumulative risk of SCC at age 20 years, rising to 76% at 35 years.¹¹

In a UK study, patients with RDEB-S developed SCC at a median age of 29.5 years, with those with other EB subtypes developing this complication later (47.1 years).¹² Similarly, a recent Dutch study found that 26% of individuals with RDEB-S developed SCC at a median age of 27.7 years.¹³ The median lengths of survival in RDEB-S after a diagnosis of first SCC were 2.4 years¹² and 3.4 years,¹³ respectively, underscoring the significance and impact of this complication in RDEB, particularly in patients with the severe subtype.

The study of Feinstein et al. described SCC in 6.7% of their cohort of all patients with RDEB at a median age of 22.6 years.³⁰ In contrast, our study recorded SCC in 21% of all participants with RDEB at a median age of 31.2 years (IQR 26.6–41.8). This presumably reflects greater representation of adults in our study, but the highest incidence of 35% was seen in RDEB-S at a median age of 27.8 years (IQR 24.9–33.6).

Cardiomyopathy was an uncommon event in our study, occurring in just two individuals. This concurs with NEBR data, which reported low frequencies of cardiomyopathy of 4% and 0.4% in RDEB-S and RDEB-I, respectively,¹⁵ although a higher risk was described in 6 of 61 (10%) patients with DEB in another study.¹⁴

Seven of our cohort (11%) died during the study: six (23%) with RDEB-S at a median age of 36.0 years (IQR 28.2–46.6) and one with RDEB-I at 61.2 years. In the NEBR cohort, 38% of patients with RDEB-S died in the 16 years the project was running, with a median age of 22.1 years.³⁴ This contrasts with RDEB-I (10% at median 30.3 years) and RDEB-Inv (12% at median 52.3 years), which is consistent with the milder phenotypes compared with RDEB-S.

Our single participant with RDEB-PT did not experience any milestone events during the course of the study, reflecting the milder phenotype in this subtype, which is largely confined to skin fragility and scarring on the shins. This, and the diversity of results for all our milestones studied, underscores the necessity to break down RDEB into subtypes in the analysis of clinical features, complications and natural history. Our data enable more accurate prognostication to better inform patients, caregivers and professionals about the anticipated progress of disease over time, and the need for interventions and screening for potentially serious and life-threatening complications. Our data may also give an early indication of the RDEB-Inv phenotype, which generally presents later in adolescence or early adulthood. Although skin involvement is often in keeping with an RDEB-I picture early in life, early-onset dysphagia and the need for OD might signpost that oesophageal mucosal involvement is more severe and will be an early feature of a developing RDEB-Inv phenotype.

Inclusion of, and separation into, different RDEB subtypes is a strength of our study, as is inclusion of both adult and child participants. However, a limitation is the relatively small numbers of participants with the less common RDEB subtypes, making them relatively under-represented. In addition, some of the milestone events are relatively infrequent (e.g. cardiomyopathy, SCC and death), and therefore including greater numbers of participants over a longer period of time is necessary to capture these rare events more fully.

In conclusion, this study provides a comprehensive review of key milestone events in patients with different subtypes of RDEB across all ages. It highlights the prevalence and timing of important complications, which have a direct bearing on an individual’s clinical course throughout life. The cumulative data mean that it is possible to prognosticate about anticipated complications and events relevant to individuals living with EB and to their healthcare teams. In addition to delineating the natural history of RDEB, they will help inform endpoints for future clinical trials or surveillance of new therapies following regulatory approval, and could serve as proxy control data for forthcoming studies.

Funding sources

This study was funded by DEBRA UK (2013–2022) and DEBRA Austria (2022–2024). They were not involved in the study design, data collection, data analysis or manuscript preparation.

Data availability

The data underlying this article cannot be shared publicly because consent was not sought from the study participants. It is planned that pseudonymized data will be made available via a repository to researchers and industry once relevant domains of the study have been published.

Ethics statement

PEBLES was ethically approved by the UK Research Ethics Committee and Health Research Authority (IRAS 142032).

Patient consent

Not applicable.

Author notes

Conflicts of interest: A.E.M. and J.E.M. have undertaken paid consultancy for Amryt Pharma and Krystal Biotech. J.A.M. is the Editor-in-Chief of the British Journal of Dermatology. The other authors declare no conflicts of interest.