Stem cell success for diabetes   (February 24, 2008)

Let me start by pointing out that I am easily frustrated by rat and other rodent-research. It's all too common that a story gets planted in the media by pharmaceutical manufacturers or biotech companies, extolling some exciting premise that turns out in retrospect to have been only a quirky finding in animal models.

But a news story this past week caught my attention: The New York Times published a story, Stem Cell Therapy Controls Diabetes in Mice, that seemed worth reading -- although it was clear that the research was sponsored by a biotechnology company, Novocell [the company has subsequently been renamed ViaCyte]. They have a press release about their new research at their website [reproduced at Novocell Reports Successful Use of Stem Cells to Generate Insulin in Mice].

Well, it definitely looks like they indeed have something to brag about. In this study, Novocell scientists took human embryonic stem cells and implanted them into mice, then made the mice diabetic (with a standard chemical called streptozotocin). These implanted stem cells were not pancreatic beta cells; instead the researchers used precursors of beta cells. Amazingly, these precursors "grew up" into cells similar to beta cells, in that they became able to control blood glucose levels by producing and secreting insulin even though the mice's natural beta cells were wiped out. The study results are published on-line at Nature Biotechnology: Pancreatic endoderm derived from human embryonic stem cells generates glucose-responsive insulin-secreting cells in vivo. The mice that had received the human embryonic stem cells continued to produce insulin and control their blood sugar while mice without the implants quickly became diabetic. After about 100 days, the scientists removed the implanted cells from the mice, and blood sugar levels shot up.

This time, mouse research is extremely important, as it shows that indeed embryonic stem cells can be induced to behave as beta cells to control blood sugar levels. It's even more interesting that these were human stem cells, being used in a rodent model, as further research will be possible without risking human health while the inevitable problems are identified and (hopefully) fixed.

One such problem: According to the NY Times, some of the mice in the Novocell experiment developed benign tumors called teratomas: 7 of the 105 mice with the implants developed them. Quite obviously, if these tumors are a frequent side-effect of the therapy, no one will want to take the giant step from animal studies to human trials.

What should we conclude from this research?

First, it's been assumed for years that human embryonic stem cells could have an important role to play in eventually eradicating type 1 diabetes. It's now been demonstrated that it's possible to use embryonic stem cells to mimic the behavior of adult beta cells, making insulin and releasing it as needed to control blood sugar levels.

Second, we'll have to wait for lots more research before this technology can be tried in humans. Researchers will need to replicate these results, first in mice, then in other animal species, to make sure that they are indeed real rather than some unexpected artifact of the study design. And assuming that the results are replicated, there's the huge problem of safety: if five percent of future animals undergoing this therapy continue to develop tumors, that would be a show-stopper.

Third, converting a precursor stem cell into a functioning pancreatic beta cell is a phenomenally complex process, and although the process has now been shown to work, several comments from other researchers stressed that we really don't have an understanding of how the process works.

So, it's obvious that use of stem cells to control blood sugar in people with type 1 diabetes is still many years in the future. But further research with stem cells should be done to verify the findings, to investigate if the tumors are preventable, and to understand how precursor stem cells "grow up."

One point is now crystal-clear: experimentation with human embryonic stem cells will help people with diabetes. The researchers should be applauded, and encouraged to do further studies.