Exploratory study on the tumorigenicity and oncogenicity of human bone marrow mesenchymal stem cells

Background Mesenchymal stem cells (MSCs), as a popular product in the fields of regenerative medicine and cell therapy, have demonstrated immense potential in the treatment of various diseases due to their characteristics like multi-lineage differentiation potential, immunomodulatory properties, and low immunogenicity. However, safety concerns regarding their clinical application, particularly the potential risk of oncogenicity, have become key factors impeding their rapid and widespread adoption. Methods In this study, we conducted GLP-compliant tumorigenicity ( n = 20) and oncogenicity studies( n = 30) in severely immunodeficient NOG mice in accordance with the requirements of NMPA ‘Technical Guidelines for Nonclinical Studies of Human Stem Cell Products’: The human bone marrow derived mesenchymal stem cell (hBMMSC), the test article, was derived from a single healthy bone donor. The manufacturer Jiuzhitang Maker completed multi-donor comparability studies prior to GLP testing, confirming this donor batch exhibits median representative quality among its allogeneic hBMMSC platform, though findings cannot be generalized to unrelated donors or alternative tissue-derived MSCs. For the oncogenicity assessment, we first established a background database of spontaneous tumors spectrum and incidence in NOG mice to eliminate interference of spontaneous lesions and then accurately evaluate the oncogenicity of the hBMMSC. The reference human clinical baseline dose adopted herein was 0.5 × 10⁶ cells/kg, with a broad clinical safety window supporting routine systemic infusion doses ranging from 0.5 × 10⁶ to 2 × 10⁶ cells/kg across diverse disease indications. We set low (6-fold clinical equivalent) and high (50-fold clinical equivalent) supraclinical doses and investigated lesion outcomes under different total dosing frequencies. Results Tumorigenicity studies revealed that hBMMSC showed no evidence of tumor formation, consistent with previous findings. Histological and qPCR analyses showed no treatment-attributable neoplastic lesions were identified in NOG mice following intravenous administration of hBMMSC at either a low dose (3 × 10⁵ cells/mouse, 6-fold of the clinical dose) administered 10 times or a high dose (2.5 × 10⁶ cells/mouse, 50-fold of the clinical dose) administered 6 times. However, administration of at least 7 high doses resulted in the emergence of tumor types absent from the NOG mouse background data, as well as an increased incidence of hepatocellular adenomas. Immunohistochemistry and qPCR analysis revealed that the tumors were of murine origin, although human cellular components were detected within the tumors. Conclusions These results indicate that under conditions of exaggerated high-dose and high-frequency administrations, MSCs may promote tumor formation in the recipient. These correlative findings do not indicate oncogenic hazards within the standard human clinical dosing range of 0.5 × 10⁶–2 × 10⁶ cells/kg, but warrant clinical caution for protocols featuring extreme repeated high-cell infusions, and long-term tumor surveillance is recommended for all MSC clinical trial subjects.