Drs. Dimarakis and Levicar have raised the interesting and well-known subject regarding the use of fetal calf serum (FCS) for the large-scale expansion of cells to be used in cellular therapy (although this use is of a regulatory nature). Unfortunately, serum-free media have not yet been defined for the isolation and expansion of human mesenchymal stem cells (MSCs), and studies using serum-free media refer only to differentiation studies in vitro [1–3]. The serum-free culture experiments of rat MSCs described by Lennon et al. and mentioned by Drs. Dimarakis and Levicar demonstrate their maintenance in culture only after two passages in FCS-supplemented medium . Given that MSCs do exhibit species-specific differences in isolation, growth requirements, morphology, and time of senescence  even maintenance in serum-free conditions may still not be applicable in the human system.
Currently, interest is growing in the use of autologous serum for MSC isolation and expansion. One of the few relevant studies to date defining the optimal dosing conditions suggests that autologous serum is required at a minimum concentration of 10% in the culture medium . For clinical protocols involving low numbers of MSCs, it is feasible to obtain the appropriate volume of peripheral blood. However, most types of cellular therapy require vast numbers of MSCs, which in turn necessitate large amounts of culture media and subsequently forbiddingly large volumes of peripheral blood. Besides, use of pooled human serum could not help to overcome this, because allogeneic serum results in MSC growth arrest and death .
To our knowledge, there are more than a dozen published clinical trials using MSCs for the treatment of various disorders, all of which use FCS for MSC culture (e.g., [8 –13]). Infusion of large numbers of MSCs (even at 109 cells) at different sites (e.g., intravenously, intracoronary, in the middle cerebral artery, or at the surgically exposed spinal cord) has so far not been reported to cause any significant side effects [8 –13]. In addition, recombinant proteins and monoclonal antibodies produced by mammalian cells in the presence of FCS indicated for human use have been approved by regulatory bodies in the U.S. and Europe and are in clinical practice. Although it has been reported that cellular cardiomyoplasty using FCS-cultivated myoblasts has demonstrated significant malignant ventricular arrhythmias , this was not the case when using MSCs expanded in FCS-containing media [12, 13]. In particular, in the study performed by our group , cardioverter defibrillators implanted in five of the patients undergoing MSC infusion failed to detect any ventricular arrhythmia (unpublished data).
In conclusion, the use of FCS does raise potential hazards (although reliable tests, such as for the detection of prion proteins, are now available), but until a suitable alternative arises, this remains the only choice in order to continue clinical trials for treating the variety of irreversible pathological disorders that do not respond to conventional therapies.