ing of the origins and embryonic development of stem cells, but we do need to know the answers to some fundamental questions:
What causes stem cells to maintain themselves in an undifferentiated state?
What cues do cells use to tell them when to start or stop dividing?
What genetic and environmental signals affect differentiation?
What physiological properties guide the functional integration of newly generated tissues into existing organs?
The scientific investigations that will answer those questions need to be comprehensive and repeated before researchers can make strong claims about the capabilities of stem cells. Because stem cell research is relatively new, it is important to build a scientific foundation that can support the research community’s ability to evaluate and confirm new findings and demonstrations. The pillars of this foundation were identified at the workshop. They include markers that characterize specific types of stem cells; markers that distinguish stages of a stem cell’s commitment to differentiate into a particular cell lineage; profiles of gene expression in stem cells and their progeny; standard procedures for isolating stem cells from the body; techniques to propagate them reliably; and consensus on the physiological or other criteria that confirm restoration of tissue function following stem cell transplantation.
As knowledge of stem cells grows, investigators will be able to ask meaningful questions about therapeutic approaches, including whether to implant cells in an undifferentiated state or a differentiated state, and which of the various sources of stem cells are best suited to address a specific clinical need. It may become apparent that combined therapies (transplanting multiple stem cell types, or using gene therapy in combination with stem cells transplants) will be needed, depending on such factors as the stage of a particular disease or the age of a patient. For now, all of these questions must wait for the establishment of a more firm scientific foundation.
Because observing the behavior of tissue in vivo is difficult and the