Safety and tolerability of bone marrow–derived mesenchymal stem cells in lupus animal models and a phase I clinical trial in humans

Objective Several clinical trials aimed at treating various autoimmune diseases, including systemic lupus erythematosus (SLE), by introducing mesenchymal stem cells (MSCs) have been conducted. However, with refractory lupus nephritis (LN), the outcomes of MSC transplantation are not well known, and further validation is required. In particular, data concerning the safety and efficacy of LN treatment using bone marrow-derived MSCs (BM-MSCs) are still lacking. Methods We identified characteristics of BM-MSCs in terms of cell morphology, chromosomal stability, differentiation capacity, and phenotype through cell passages. The in vivo stability of BM-MSCs was evaluated by single-dose and repeated-dose toxicity tests, tumorigenicity tests, and biodistribution tests using lupus mouse models. Based on the encouraging nonclinical results, we conducted a nonrandomized, open-label, single-arm phase I clinical trial to evaluate the tolerability and safety of a single administration of haploidentical allogeneic BM-MSCs (CS20AT04) in seven LN patients (NCT03174587). We used a classical three + three design to find the optimal dosage. The starting dose was 2.0×10 6 cells/kg and escalated to 3.0×10 6 cells/kg if there was no dose-limiting toxicity (DLT). Evaluation of the safety and tolerability was assessed 28 days after the infusion, and the maximum tolerated dose was determined. Results Properly cultured BM-MSCs showed high proliferation and multipotency, but chromosomal changes were not found. There were two deaths by a rapid administration rate in the high-dose group (2.0×10 6 cells/head) in a single administration test. BM-MSCs were distributed in the kidneys until Day 7. In the phase I clinical trial, seven LN patients were enrolled. Participants received BM-MSCs through intravenous infusion. There was no DLT at both initial dose (2.0×10 6 cells/kg) and escalated dose (3.0×10 6 cells/kg). One patient was not administered the full 2.0×10 6 cells/kg dose because of a technical error during infusion. This patient did not show DLT. Three adverse events were reported, namely, one diarrhea, one toothache, and one arthralgia, and all were considered NCI-CTC grade I events. Conclusion We defined the characteristics of BM-MSCs and identified their safety and tolerability in both animal models and a phase I clinical trial. The maximum tolerated dose was determined to be 3.0×10 6 cells/kg in patients with LN.