Human multipotent mesenchymal stromal cells (MSCs) are currently tested in a growing number of clinical trials to determine their safety and efficiency as an immune modulating and organ regenerative therapy. Repeated observations of genomic instability in the commonly used fetal bovine serum (FBS)-driven MSC cultures and consecutive tumor formation by transformed MSCs in experimental animals have raised serious safety concerns. We and others have recently established alternative clinical scale MSC expansion protocols with pooled human platelet lysate (pHPL) as a substitute for fetal bovine serum. This study was performed to determine the genomic stability of MSCs expanded under humanized conditions ex vivo. Small volume (14–17mL) bone marrow aspirates of four donors (3 male: 30, 36 and 47 years; 1 female: 13 years) were seeded without manipulation directly in heparinized minimum essential medium just supplemented with pHPL and L-glutamine. Clinical scale propagation was done in a newly developed humanized cell expansion system. MSC quality, identity, purity and function were assessed according to a defined panel of release criteria. Array-comparative genomic hybridization (array-CGH) was carried out using a whole genome oligonucleotide microarray platform with female reference DNA. Samples were labeled and scanned images were analyzed using CGH Analytics software. Results confirmed that pHPL is highly efficient in stimulating MSC expansion resulting in the recovery of 780 ± 150 million MSCs (mean ± SEM) after one culture phase. Starting from 15 ± 0.6 mL bone marrow we were able to produce four application doses of MSCs (defined as 2 mio. MSCs/kg x 100kg) in a unique standardized single culture phase within 13.5 ± 1.0 days with a minimum of manipulation and without antibiotics in three of four expansions. MSC viability was ≥ 95%. Flow cytometry revealed virtually pure MSC products with >95% CD73/90/105 reactivity, <2% hematopoietic contamination with CD14/19/34/45-or HLA-DR-reactive cells and intact adipo-, osteo-and chondrigenic differentiation potential. Microbiologic safety measures included negative bacterial/ fungal/mycoplasma testing despite antibiotic/antimycotic-free culture and endotoxin levels <0.025 EU/mL. Array-CGH analysis revealed balanced profiles for all propagated MSC products. Five copy number variations (CNVs; >60kb) that were detected were not documented in the database of genomic variants. Several small (7kb–1.8Mb; n=33) autosomal CNVs were also observed previously in normal individuals and were not associated with phenotype changes. These data extend earlier results showing that MSCs expanded under humanized conditions did not form tumors in experimental animals in vivo. Our data show that despite high proliferation rate MSCs propagated in a human platelet-derived growth factor-driven system are genomically stable in array-CGH and do not form tumors in vivo. This indicates superior safety of the rapidly available humanized MSC transplants compared to currently used FBS-expanded MSCs.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution