Clinical trial of ABCB5+ mesenchymal stem cells for recessive dystrophic epidermolysis bullosa

Abstract

BACKGROUNDRecessive dystrophic epidermolysis bullosa (RDEB) is a rare, devastating, and life-threatening inherited skin fragility disorder that comes about due to a lack of functional type VII collagen, for which no effective therapy exists. ABCB5+ dermal mesenchymal stem cells (ABCB5+ MSCs) possess immunomodulatory, inflammation-dampening, and tissue-healing capacities. In a Col7a1-/- mouse model of RDEB, treatment with ABCB5+ MSCs markedly extended the animals' lifespans.METHODSIn this international, multicentric, single-arm, phase I/IIa clinical trial, 16 patients (aged 4-36 years) enrolled into 4 age cohorts received 3 i.v. infusions of 2 × 106 ABCB5+ MSCs/kg on days 0, 17, and 35. Patients were followed up for 12 weeks regarding efficacy and 12 months regarding safety.RESULTSAt 12 weeks, statistically significant median (IQR) reductions in the Epidermolysis Bullosa Disease Activity and Scarring Index activity (EBDASI activity) score of 13.0% (2.9%-30%; P = 0.049) and the Instrument for Scoring Clinical Outcome of Research for Epidermolysis Bullosa clinician (iscorEB‑c) score of 18.2% (1.9%-39.8%; P = 0.037) were observed. Reductions in itch and pain numerical rating scale scores were greatest on day 35, amounting to 37.5% (0.0%-42.9%; P = 0.033) and 25.0% (-8.4% to 46.4%; P = 0.168), respectively. Three adverse events were considered related to the cell product: 1 mild lymphadenopathy and 2 hypersensitivity reactions. The latter 2 were serious but resolved without sequelae shortly after withdrawal of treatment.CONCLUSIONThis trial demonstrates good tolerability, manageable safety, and potential efficacy of i.v. ABCB5+ MSCs as a readily available disease-modifying therapy for RDEB and provides a rationale for further clinical evaluation.TRIAL REGISTRATIONClinicaltrials.gov NCT03529877; EudraCT 2018-001009-98.FUNDINGThe trial was sponsored by RHEACELL GmbH & Co. KG. Contributions by NYF and MHF to this work were supported by the NIH/National Eye Institute (NEI) grants RO1EY025794 and R24EY028767.

Keywords: Adult stem cells; Clinical Trials; Stem cells.

Figures

Figure 1. Study summary.

(A and B) Trial design and trial flow chart. ASigned the informed consent form. BFailed to attend the screening visit (due to poor general health, n = 1) or day 0 visit (due to travel restrictions associated with the COVID‑19 pandemic, n = 1). CPatient was prematurely withdrawn from treatment due to occurrence of a hypersensitivity reaction during the second cell infusion. FU, follow-up.

Figure 2. Representative photographs of patients at baseline (Day 0) and after 3 infusions of ABCB5+ MSCs (Week 12).

(A) Right lateral upper arm and right knee of a 24‑year-old Female patient (cohort 1). (B) Shoulder/neck area (back and front) and right hand of a 13‑year-old Female patient (cohort 2). (C) Right lateral upper arm and dorsum of the right foot of a 13‑year-old Female patient (cohort 2). (D) Back shoulder area of a 9‑year-old Male patient (cohort 3). All patients had consented to publication of their photographs.

Figure 3. Changes in EBDASI.

(A) Percent changes in the EBDASI overall score and total activity and damage subscores at 12 weeks (with the last observation carried forward [LOCF] in cases of missing data), expressed as percentage of the baseline value, in the full analysis set (FAS) and the per-protocol set (PP). (B) Percent changes in the EBDASI activity score by visit, expressed as percentage of the baseline value, in the FAS (no LOCF). Data are shown as medians with IQR; P values (2-sided Wilcoxon signed rank test) indicate statistical significance of changes from baseline. Kruskal-Wallis tests followed by Dunn’s multiple comparison tests revealed no statistically significant differences between the 3 postbaseline visits (day 17, day 35, and week 12; P > 0.05). For EBDASI overall and damage score data, see Supplemental Table 1.

Figure 4. Changes in iscorEB.

(A) Percent changes in the iscorEB overall score and iscorEB‑c and iscorEB‑p subscores at 12 weeks, expressed as percentage of the baseline value, in the full analysis set (FAS) and in the per-protocol set (PP). The lower number of data points for iscorEB overall and iscorEB‑c as compared with the iscorEB‑p is due to difficulties with blood sampling; for these patients, the lab values (anemia, albumin, inflammation) required for calculation of iscorEB overall and iscorEB‑c could not be obtained. (B) Percent changes in the iscorEB‑c score by visit, expressed as percentage of the baseline value, in the FAS. Data are shown as medians with IQR; P values (2‑sided Wilcoxon signed rank test) indicate statistical significance of changes from baseline. Kruskal-Wallis tests followed by Dunn’s multiple comparison tests revealed no statistically significant differences between the 3 postbaseline visits (day 17, day 35, and week 12; P > 0.05). For iscorEB overall and iscorEB‑p data, see Supplemental Table 2.

Figure 5. Changes in itch, pain, and impact of RDEB on life quality in the full analysis set (FAS).

(A–C) Changes in: itch score, pain score, and QOLEB score, expressed as percentage of the baseline value. The lower number of data points for the pain score as compared with itch and QOLEB scores at the postbaseline visits (day 17, day 35, week 12) is caused by 2 patients presenting with pain score = 0 at baseline; therefore, for these patients, percent changes from baseline could not be calculated at any postbaseline visit. Please note that the patient who presented with an extreme percent increase in QOLEB score on day 35 and at week 12 (with score category changing from mild [day 0] to very mild [day 17] to moderate [day 35 and week 12]) had received only an incomplete second cell dose (day 17) and no third cell dose (day 35). Data are shown as medians with IQR; P values (2‑sided Wilcoxon signed rank test) indicate statistical significance of changes from baseline. Kruskal-Wallis tests followed by Dunn’s multiple comparison tests revealed no statistically significant differences between the 3 postbaseline visits (day 17, day 35, and week 12; P > 0.05). For the data of the per-protocol set, see Supplemental Figure 2.

Figure 6. HMGB1 serum concentrations.

Each color represents an individual patient. Data are shown as medians with IQR. Kruskal-Wallis tests followed by Dunn’s multiple comparison tests revealed no statistically significant differences between visits (P > 0.05).