The new technologies for analyzing the genomes of cells will certainly provide exciting new data to help unravel some of the major mysteries of biology. At the same time, some of the data simply allow us to rediscover principles that were defined long ago. For example, a recent publication reported that an analysis of gene expression data demonstrated that “neural and mesenchymal stem cells acquire characteristic large chromosomal abnormalities at a similar, or somewhat lower, frequency to that seen in pluripotent stem cells, sometimes within a few passages in culture” [1]. But are these results surprising? Do they blur over a critical distinction among different cells in culture? Do they mean that, as the authors conclude, “..validating the genomic stability of stem cells of all types in culture is crucial …for their safe implementation in cell therapy.” Therefore, will it be necessary to perform extensive genomic analysis on all culture-expanded cells before they can be used safely in patients? Or have we been here before when observing cells in culture?

Concern about the chromosomal stability of cells in culture arose over 40 years ago when Wi-38 human fibroblasts were first introduced by Hayflick et al. [2] as substrates for viruses to develop vaccines. Manufacturers of vaccines were required to ensure that the cells were free, not only of any adventitious agents but also of what were then referred to as “any neoplastic properties.” As result, beginning in the 1960s, manufacturers were required to test the cells by karyotyping of chromosomes in addition to assaying for tumors after injection into immunodeficient mice. A number of chromosomal abnormalities in cultured Wi-38 cells were quickly detected by the crude karyotyping assays available at the time. However, there was considerable confusion about the significance of the changes. An International Committee that met in Geneva in 1976 could not agree on karyotypic criteria that should be applied for use of the cells [3] and the confusion continued for many years. For example, analysis of 18 cell banks revealed a chromosome 7;12 translocation with a frequency of 0.2%–5.6% in the diploid fibroblasts used to produce vaccines, both Wi-38 cells and similar cells referred to as MRC-5 [4]. However, the translocation was rarely seen at low population doublings (PDLs <30), increased with expansion to reach a plateau at 40-50 PDLs, and disappeared when cultures became senescent. Most importantly, there were no differences in growth characteristics or in tumorigenicity in mice between the cultures that did or did not contain t(7;12).

The interpretation of data on chromosomal stability has been further complicated by the recent widespread use of invasive prenatal diagnosis. Assays with array-assisted comparative genomic hybridization detected copy number variation of unknown significance in 0.5%–1% of samples [5]. In addition, copy number variations of genes were found to be common and as high as 12% of the human genome. Where does this leave us in evaluating the potential dangers of therapies with cells that are expanded in culture?

One guideline is provided by more than three decades of experience using culture-expanded human fibroblasts for vaccine production and by cell karyotyping and testing tumorigenicity in mice. One opinion was the “futility of continuing to do these tests is obvious” [6]. Another opinion was that after initial tests to establish the identity of the cells, chromosomal analysis should not be required before using the cells as substrates for the manufacture of live virus vaccines [4].

What are the appropriate tests? Given that most human cancer cells do not produce tumors in immune deficient mice, the assay for tumorigenicity using nude mice has unacceptably low sensitivity. For example, despite intensive efforts, it has not been possible to develop a mouse model with orthotopic implantation of human osteosarcoma cells that mimics the human disease [7]. At the end of the day, the most reliable test is one of the simplest [8, 9]: do the cells senesce in culture? If they do, they are unlikely to produce tumors or malignancies in patients. Unlikely in the sense that there is a low probability, but not a certainty, that the expanded cultures do not contain a few cells that have acquired an oncogenic mutation. In contrast, cells that are immortal in culture carry a serious risk. Unfortunately, at present, there is no strategy for eliminating this risk. Differentiating the cells in such cultures, or cloning and then differentiating them, does not eliminate the risk as there is no current technology, or any method on the horizon, to detect a few tumorigenic cells in the large-scale cultures that are required for most therapeutic applications. In considering any therapy, we always return to the same question: what is the probability of benefiting versus harming the patient? We can answer this question with greater confidence as our knowledge of biology improves, but we will always face unknowns as we move from the laboratory to the bedside. Embryonic stem cells and induced pluripotent cells do not pass the test of senescence in culture. Therefore, the risk/benefit ratio is likely to be high. In contrast, the adult stem cells referred to as mesenchymal stem cells or mesenchymal stromal cells (MSCs) currently in use in a large number of clinical trials reproducibly senesce in culture. It is not surprising, therefore, that no serious adverse events have yet been reported in the many hundreds, probably thousands, of patients who have received MSCs, although it is important to emphasize the need for formal long-term follow-up of the recipients [9, 10]. The apparent safety of administering MSCs is reassuring since the therapeutic efficacy of the cells has not yet been established definitively for many conditions in which they are being tested.


D.J.P. has consultant/advisory role in the advisory board of Temple Therapeutic LLC, obtained research funding/contracted research (including funds paid to your institution) from Temple Therapeutic LLC, and has ownership interest (stocks, stock options, or other ownership interest excluding diversified mutual funds) of less than 5% equity in Temple Therapeutic LLC. A.K. has a consultant/advisory role in TRT and in Temple Therapeutic and has stock options in TRT.


  1. Top of page
  • 1
    Ben-David U, Mayshar Y, Benvenisty N. Large-scale analysis reveals acquisition of lineage-specific chromosomal aberrations in human adult stem cells. Cell Stem Cell 2011; 9: 97102.
  • 2
    Hayflick L, Perkins F, Stevenson RE. Human diploid cell strains. Science 1964; 143: 976.
  • 3
    Report of Ad Hoc Committee on karyological controls of human cell substrates. J Biol Stand 1979; 7: 397404.
  • 4
    Rosolowsky M, McKee R, Nichols W et al. Chromosomal characterization of MRC-5 cell banks utilizing G-banding technique. Dev Biol Stand 1998; 93: 109117.
  • 5
    Strassberg M, Fruhman G, Van den Veyver IB. Copy-number changes in prenatal diagnosis. Expert Rev Mol Diagn 2011; 11: 579592.
  • 6
    Petricciani JC, Horaud FN. Karyology and tumorigenicity testing requirements: Past, present and future. Dev Biol Stand 1998; 93: 513.
  • 7
    Dass CR, Ek ET, Choong PF. Human xenograft osteosarcoma models with spontaneous metastasis in mice: Clinical relevance and applicability for drug testing. J Cancer Res Clin Oncol 2007; 133: 193198.
  • 8
    Romanov SR, Kozakiewicz BK, Holst CR et al. Normal human mammary epithelial cells spontaneously escape senescence and acquire genomic changes. Nature 2001; 409: 633637.
  • 9
    Prockop DJ, Brenner M, Fibbe WE et al. Simmons PJ, Sensebe L, Keating A. Defining the risks of mesenchymal stromal cell therapy. Cytotherapy 2010; 12: 576578.
  • 10
    Casiraghi F, Remuzzi G, Abbate M et al. Multipotent mesenchymal stromal cell therapy and risk of malignancies. Stem Cell Rev 2012 [Epub ahead of print].

Darwin J. Prockop*, Armand Keating†, * Department of Medicine, Texas A&M Health Science College of Medicine, Institute for Regenerative Medicine at Scott & White, Temple, Texas, USA, † Cell Therapy Program, Princess Margaret Hospital, Toronto, Ontario, Canada