Intravenous allogeneic multilineage‐differentiating stress‐enduring cells in adults with dystrophic epidermolysis bullosa: a phase 1/2 open‐label study

To the Editor,

Epidermolysis bullosa (EB) is a group of genodermatoses characterized by generalized blisters from mutations in the genes encoding the basement membrane zone (BMZ) proteins.¹ The infusion of allogeneic cells such as mesenchymal stem/stromal cells (MSCs), which have intact BMZ genes, is a promising treatment.², ³, ⁴, ⁵, ⁶ We here focus on multilineage‐differentiating stress‐enduring (Muse) cells, which were initially found as a stem cell marker stage‐specific embryonic antigen (SSEA)‐3‐positive subpopulation of bone marrow‐derived MSCs.⁷ Papers have demonstrated that SSEA‐3(+) Muse cells have higher pluripotency than the SSEA‐3(−) population of MSCs, which suggests that Muse cells are preferable to bulk MSCs for regenerative medicine.⁷, ⁸ Very recently, CL2020, a clinical‐grade Muse cell product (Life Science Institute, Inc., Tokyo, Japan), demonstrated safety and efficacy in human patients with acute myocardial infarction.⁸ In addition, we reported that human Muse cells can differentiate into epidermal keratinocytes with the expression of human BMZ proteins in vivo.⁹ We herein conducted a pilot study on CL2020 for the treatment of EB in human adults.

This study was designed as an open‐label, non‐randomized, single‐arm, non‐controlled clinical trial (JapicCTI‐184563). In brief, EB patients with refractory ulcers and recurrent ulcers lasting for more than 4 weeks were selected, measured and followed for 52 weeks after a single infusion of CL2020 containing 1.5 × 10⁷ cells (2.98 ± 0.61 × 10⁵ cells/kg). The reduction rate of the selected skin ulcers per patient (%) was calculated as follows:¹⁰

$$=\frac{{\sum }_{\mathrm{lesion}=1}^n\left(\mathrm{Baseline}\mathrm{ulcer}\mathrm{area}\left({\mathrm{cm}}^2\right)‐\mathrm{Ulcer}\mathrm{area}\mathrm{at}\mathrm{evaluation}\left({\mathrm{cm}}^2\right)\right)}{{\sum }_{\mathrm{lesion}=1}^n\left(\mathrm{Baseline}\mathrm{ulcer}\mathrm{area}\left({\mathrm{cm}}^2\right)\right)}\times 100.$$

Five patients [one male and four female, ages 17–49 (median: 22)] with 13 ulcers from two institutes were finally enrolled and followed (Table 1). All cases showed mild or self‐limiting adverse effects, and the following were reported: stomach pain (Grade 3), acquired lacrimal stricture (G2), fever (G1), gastroenteritis (G1), upper respiratory tract infection (G1) and paraesthesia of the upper arms (G1). One patient showed paraesthesia within 24 h after infusion and resolved in 14 days, which was considered a possible CL2020‐related side‐effect. Two patients showed a >50% ulcer reduction rate per patient at 4 weeks after CL2020 administration (Wk04, Fig. 1a,b). The average reduction rate at Wk04 was 46.32 [95% confidence interval (CI): −13.59 to 106.22]%. A tendency for the ulcers to improve was observed from Wk02 to Wk08, but overall, the ulcer size returned to the baseline at Wk12. The change in the total size of the selected ulcers was −9.98 (95% CI: −17.87 to −2.09) cm² at Wk04, and statistically significant improvement was found (P = 0.017, Fig. 1c). Regarding scores for pain, itch and quality of life, only the pain score seems to have improved, and it did so moderately at the early phase (P = 0.003 at Wk02, Fig. 1d,e). Blood examinations revealed that liver dysfunction, probably due to chronic inflammation, improved after the CL2020 infusion in two patients (Fig. 1f,g).

Table 1.

The details of the participants

Patient	1	2	3	4	5
Age, sex	26, F	22, F	20, F	17, M	49, F
Diagnosis	Intermediate RDEB	Intermediate RDEB	Intermediate DDEB	Localized DDEB (pretibial)	Intermediate DDEB
Bodyweight (kg)	42.3	46.2	52.4	75.9	46.0
Number of selected ulcers	2	4	3	2	2
Average ulcer size (cm2, min–max)	1.50 (0.7–2.3)	16.00 (1.6–49.2)	1.93 (0.9–3.8)	8.10 (2.7–13.5)	2.20 (1.2–3.2)
Dose of CL2020 (×105 cells/kg)	3.55	3.25	2.86	1.98	3.26
Average reduction rate at Wk04 (%)	100	59.5	37.9	43.2	−9.1

DDEB, dominant dystrophic epidermolysis bullosa; RDEB, recessive dystrophic epidermolysis bullosa.

Figure 1.

(a) Representative clinical images of the patient (Patient 1, right lower leg). During the observation period from Wk‐4 to Wk00, the area of skin erosion was largely unchanged. After CL2020 administration, the area of the erosion rapidly improved. (b) The time course for the reduction rate of ulcer area per patient. Summary statistics (mean and standard deviation) and two‐sided 95% confidence intervals for the mean were calculated, and a transition chart was prepared for each subject. Adjusted mean changes from the baseline (95% confidence intervals) were determined using a linear mixed‐effects model for all efficacy outcomes. (c) Change in the combined size of selected ulcers from the baseline (cm²) per patient. (d) Change in visual analogue score (VAS) for pain. At Wk02, a significant reduction is observed. (e) VAS for itch. Itch score increases are noted at Wk12 and Wk20. (f, g) Regarding blood examinations, improvement of liver enzymes is noted in two patients. It may be explained by the tissue regeneration of Muse cells in the liver or the anti‐inflammatory effects of Muse cells. Bars: standard deviation. *P < 0.05.

We obtained a skin sample at Wk04 from 1 recessive DEB patient and performed immunofluorescence analysis and electron microscopic investigation.¹⁰ No increases in type VII collagen fluorescence intensity or in anchoring fibrils were observed (data not shown).

The limitations of the current study are given here. (i) The single administration of a rather small number of cells might have resulted in the limited clinical efficacy. (ii) Ulcers of somewhat small sizes tended to be included, and three of the five enrolled patients had dominant DEB, which could have adversely affected the endpoint results by frequent scratching. (iii) Only one biopsy was conducted in the trial. (iv) No interventions were performed on children. CL2020 administration is a well‐tolerated therapy and Muse cells are a potentially promising regenerative medicine for adults with severe EB.

Author contributions

YF conceptualized the study. YF, TN, ST, KN and AI performed data curation. YF and OW performed formal analysis. HS obtained funding. YF, ST, KN and HN investigated the study. YF, ST, KN, SS, TT, MA, AI and HS designed methodology. YF, MA, AI and HS administered the project. YF, TN, ST, KN and AI provided resources. MA, AI and HS supervised the study. YF, ST, KN, SS, AI and HS performed validation. YF, ST, KN and HN performed visualization. YF wrote the original draft. All authors reviewed and edited the manuscript.

Conflicts of interest

YF and HS hold a patent on the use of Muse cells for treating EB. OW is an employee of Life Science Institute, Inc. (LSII). AI and HS received a research grant from LSII. YF and HS received medical advisor fees from LSII.

Funding source

This work received funding from Life Science Institute, Inc. (to HS).