By Monya Baker

The failure of late-stage clinical trials that used a stem-cell therapy to treat a rare condition has dealt a setback to stem-cell research. The two studies looked at graft-versus-host disease (GvHD), a potentially fatal complication of bone-marrow transplantation in which immune cells from the donated marrow attack the recipient. The trial sponsor, Maryland-based Osiris Therapeutics, announced on 8 September that its Prochymal drug performed no more effectively overall than a placebo.

"I do think it's a blow to the stem-cell community," says Edward Tenthoff, a research analyst at Piper Jaffray in New York City. The good news, Tenthoff says, is that "it's a failed trial, not a safety concern". Last month, embryonic stem-cell company Geron received a second clinical hold on its proposed trial for spinal-cord injury after animal data revealed microscopic cysts around the site of injury.

After the announcement on Tuesday, shares of Osiris closed 34% lower compared with the weekend's close price. Shares in a variety of other stem-cell companies, including Australia's Mesoblast, and California's Geron and StemCells, and Michigan's Aastrom Biosciences, were unaffected.

Unusual approach

Unlike many other stem-cell-based therapies in development, designed to replace or supplement degenerating cells, Osiris's Prochymal is intended to subdue an aggressive immune response that can occur in bone-marrow transplantation. The drug consists of cultured mesenchymal stem cells, a cell type which, along with blood-forming stem cells, is found in the bone marrow.

Last year, a separate group led by Katarina Le Blanc of the Karolinska University Hospital in Stockholm, published results indicating that cultured mesenchymal stem cells could benefit patients with GvHD -- even if transplanted cells were not tissue-matched to the patients who received them1. Compared to that 55-patient trial, Osiris's trials were much larger and included placebo groups. One, with 192 patients, tested Prochymal as an initial treatment of GvHD in combination with standard steroid therapy. The other tested Prochymal in 260 patients for whom steroid therapy had failed to control the disease. In neither of the double-blind studies did patients receiving the experimental drug live longer than patients receiving placebo.

Promising start?

In a conference call to investors, Osiris chief executive Randy Mills hastened to point out signs that the drug might work in subgroups of patients. While Prochymal did not seem to affect GvHD involving the skin -- by far the most common form -- the drug had a statistically significant effect for steroid-resistant disease affecting the liver or the intestinal tract.

Mills said that the company's scientists were still analyzing the data and hoped to talk with the US Food and Drug Administration soon about whether these results could be used to gain approval for Prochymal in these cases or be used to design another clinical trial. Osiris is also conducting tests of Prochymal in Crohn's disease, diabetes, heart disease and lung disease. But plans for a late-stage trial of Crohn's disease, a severe inflammation in the gut, were scaled back after disappointing preliminary results.

"The results are a promising first randomised trial" on mesenchymal stem cells, says Le Blanc, and, like her previous study, support an effect on treatment of GvHD in the liver and the gastrointestinal tract.

"Even though it seems to the business world to be a setback, I think [researchers] can learn and go on," says Pranela Rameshwar, who studies the immune properties of mesenchymal stem cells at New Jersey Medical School in Newark. Rameshwar emphasizes that mesenchymal stem cells, which interact with many kinds of immune cells, also react to the environment they are placed in, and the inflammation in GvHD patients can be highly variable. "Their approach is like a one-size-fits-all," she says, "maybe the treatment has to be personalized." Both scientists say that Osiris's studies indicate that mesenchymal stem cells can be an effective treatment -- if the optimal conditions are identified.

But James Ferrara, director of blood and marrow transplant at the University of Michigan in Ann Arbor, takes a more skeptical view, "They probably don't work, except maybe for the liver," he says. "I think the majority of people feel that they don't work and we need to go on and find something different."