What we know, from Maureen L. Condic, associate professor of neurobiology and anatomy at the University of Utah School of Medicine in First Things.

The assertion that embryonic stem cells in the laboratory can be induced to form all the cells comprising the mature human body has been repeated so often that it seems incontrovertibly true. What is missing from this assertion remains the simple fact that there is essentially no scientific evidence supporting it. Experiments have shown that embryonic stem cells are able to participate in normal embryonic development, an observation that is also true of cancerous embryonal carcinoma cells. When injected into early mouse embryos, both embryonic stem cells and embryonal carcinoma cells randomly contribute to every tissue of the developing body.

Even more dramatically, when embryonic stem cells are injected into mouse embryos under specific experimental circumstances (a procedure known as tetraploid complementation), they can be induced to form all the cells of the postnatal body. These experiments prove that embryonic stem cells (and embryonal carcinoma cells) remain capable of responding appropriately to the developmental signals that regulate tissue formation in the embryo, and from these results we can conclude that if embryonic stem cells were intended to provide cell replacement therapies for embryos, they would represent a very promising therapeutic approach. The problem, of course, is that embryos are not the intended targets of stem cell therapies, and there is little reason to believe that the capabilities of embryonic stem cells in an embryonic environment are relevant to their therapeutic potential for non-embryonic patients.

Five years ago, most scientists working in the field of embryonic stem cell research confidently predicted that we would soon determine the precise recipe of molecular factors required to replicate in the laboratory the mysterious inner life of the embryo. David Anderson, a stem cell researcher at Caltech, boldly asserted in a New York Times opinion piece that once science had figured out the factors required to replicate embryonic development, specific molecules could simply be “thrown into the bubbling cauldron of our petri dishes,” where they would transform embryonic stem cells into an unlimited source of replacement cells for any tissue we chose to produce.

Skepticism regarding this claim was well warranted. While there have been hundreds of papers published over the past five years that stridently claim “cell type X produced from embryonic stem cells,” under closer inspection these successes have all been less miraculous than they appeared. It is relatively easy to generate stem cell derivatives in the laboratory that have at least some of the properties of normal, mature cell types. But the test of whether an embryonic stem cell–derived brain cell, for example, is indeed a normal adult brain cell is to put it into the brain of an adult animal and determine whether it survives and contributes to normal brain function. In addition, if laboratory-generated cells are to be therapeutically useful for the treatment of human disease and injury, they must be shown to have therapeutic value in adult animals: It is not sufficient that embryonic stem cell–derived cells merely survive in adults; they must also be able to repair the underlying disease or injury. It is precisely this kind of test that embryonic stem cell–derived tissues have proved unable to pass.