[Cardiovascular regenerative medicine at the crossroads. Clinical trials of cellular therapy must now be based on reliable experimental data from animals with characteristics similar to human's].

It is now over 4 years since early reports of murine models raised high expectations that bone marrow cell transplantation to the postischemic myocardium could produce physiologically significant myocardial regeneration. In quick succession, a flurry of publications documented the capacity of a variety of other types of adult cell to produce similar results. These publications were all controversial from the start because none addressed the mechanisms involved in the differentiation of transplanted cells. In addition, each report raised at least as many questions as it answered. Despite these obvious weaknesses, the first phase-I clinical trials were started immediately without any further animal experimentation. Today the results of more than a dozen trials are already in the public domain but we still do not have a single piece of solid data documenting whether any of the approaches used is capable of regenerating contractile cells in the human myocardium. This is one of the main reasons why the controversy over the effectiveness of this therapeutic approach is becoming increasingly heated. Moreover, skepticism about the efficacy, and even the feasibility, of inducing clinically relevant myocardial regeneration has increased to the point where it threatens the future of this nascent field. The present situation in myocardial generation contrasts sharply with that in neural regeneration. Although there is a solid and extensive body of knowledge on the origin, phenotype, and regulatory mechanisms of neural stem cells, the first clinical trials have only recently been started. To move this field forward it is necessary to distinguish between the procedures needed to establish proof-of-concept and those that have the potential for widespread clinical application. In addition, the technique must be implemented in such a way that it continues to add to existing knowledge. It is our belief that, if the necessary information is to be acquired, we need: a) significantly more extensive experimental data from animals whose anatomical and physiological characteristics are similar to human's, including data on, for example, dose-effect relationships, the best form of administration, and the duration of therapeutic responses; and b) better understanding of the molecular mechanisms that determine whether cardiac stem cells and transplanted cells will either remain as stem cells or differentiate. In summary, if we are to progress systematically in this area, we need better understanding of myocardial biology. Without it, we run the risk of holding back the field for decades, as happened with the first human heart transplants and with trials of gene therapy.