Four years ago, headlines in newspapers across the United States questioned whether the small number of approved clinical trials on adult stem cell therapies should even go forward, echoing scientists' concerns that animal studies posed too many contradictions to justify human tests.

Research has moved on a long way since then. A recent review in the Journal of the American Medical Association found that clinical trials of stem-cell treatments raised few apparent safety issues1. Few deaths followed the introduction of stem cells into the heart, and the therapy was even marginally successful in treating disease. In contrast, trials of stem cells in autoimmune diseases had more mixed outcomes. Depending on the type of accompanying chemotherapy regime, death rates in these trials ranged from a low of 1% to as high as 23%.

“The cardiac trials are safer across the board because they don't involve chemotherapy,” says Richard Burt, the Northwestern University researcher who spearheaded the report. His team reviewed 323 studies for feasibility and toxicity — 69 of these included data on overall patient outcomes. Although only 33 of the trials dealt with heart and vascular diseases, these represented a total of nearly 2,000 cardiac patients.

“The trials were consistently safe,” says Burt, “and while there is still a lot to be learned, there is also a great potential here to help people with heart disease.”

Osiris employees are preparing stem cells for clinical trials in heart disease Credit: Osiris

A handful of new ground-breaking cardiac trials now hope to demonstrate cell therapy's merits. Osiris Therapeutics of Columbia, Maryland, received approval from the US Food and Drug Administration (FDA) in January for a phase II trial that will administer bone-marrow-derived mesenchymal stem cells to about 200 patients directly after they suffer a heart attack in an attempt to prevent fibrotic scars. Stem cells are taken from young healthy volunteers and are delivered to patients through an intravenous drip at any time from one to eight days after a heart attack.

“The safety profile held up during phase I trials, which is very encouraging,” says Randal Mills, chief executive officer at Osiris. “Now we are evaluating efficacy more closely.”

As Osiris begins its trial this summer, another study, conducted by the cell-therapy company Mytogen, based in Charlestown, Massachusetts, will also be enrolling patients. Skeletal muscle stem cells will be delivered to scarred regions of patients' hearts by a catheter guided by 3D imaging. Each patient's own stem cells will be used, and 165 people will be treated. This trial follows positive phase I findings after tests of the therapy in 23 patients.

Trials of stem-cell therapy are also gaining momentum in the United Kingdom. In February, doctors at the University of Bristol started injecting stem cells collected from heart patients' own bone marrow into scars left by coronary bypass operations. Magnetic resonance imaging will show whether the stem cells have reduced scarring six months down the road. Small studies have shown that the approach is safe, but with 60 patients this study will be the largest placebo-controlled trial of its kind.

Little improvement

Yet as the FDA gives the green light to the next generation of trials, debates over the efficacy of the stem-cell treatments and the mechanism of healing continue to rage within the scientific community.

The first wave of clinical trials showed only modest improvements or no effect on the patients' condition. In the largest study featured in the JAMA report1 — a placebo-controlled trial called REPAIR-AMI — researchers collected bone marrow from 200 heart-attack patients and infused it into the coronary artery. Four months after infusion, the team measured the ejection fraction — or the amount of blood pumped by the heart — but saw only a 3–5.5% improvement.

“If it were a drug, this would be considered a negative result,” says Kenneth Chien, a cardiologist at the Harvard Stem Cell Institute in Boston, Massachusetts, and known for his cautious approach to human trials of stem-cell therapies. He still voices many of the same criticisms as five years ago, when he warned against the phase I trials. “If you don't know how it works, you don't know for sure that it works,” he says.

According to the JAMA review, stem cells can provide a local cell-help-cell effect: they secrete factors that promote healing, prevent cell death or encourage new blood vessels through a process called angiogenesis. “Stimulating new blood vessel formation is a critical goal in many of these trials,” says Burt, “but it's possible that angiogenesis is due to secondary effects.”

And even though the end results might appear positive, researchers eye these effects with caution. Animal studies show that bone marrow stem cells, which are the ones most commonly used, can turn into fat and cartilage as well as into muscle. If the stem cells introduced in cardiac therapies differentiate into something other than cardiac cells, the heart muscle will not become any stronger. “If new tissue isn't made, we can't know for sure that the heart is getting repaired,” says Chien, “even if results show a marginal benefit following treatment.”

Animal studies have shown that stem cells can fuse with heart muscle cells and take on their appearance without fully adopting the ability to beat in synchrony with the rest of the heart. Some studies have found that the stem cells can no longer be found 24 hours after surgery, suggesting that the cells don't survive well after transplantation.

While the treatments remain controversial, the debate has shifted to questions about efficacy.

“Most of the debate over mechanism focuses on data from animal studies,” says Burt. “Even if the underlying improvement in animal studies is still there, the mechanism by which this improvement takes place is controversial.” In animal experiments, hearts are removed weeks after introducing the cells to see whether the introduced stem cells are differentiating correctly. In humans, there is no way to retrieve the stem cells and prove they have helped repair the heart.

Phase II designs

According to Mills, however, the first wave of studies was never intended to answer detailed questions about mechanism. “Most of the trials look at safety, which may be why benefits appear marginal in some studies,” he says.

Even Osiris's phase I trial, which boasts more striking results, is far too small to draw firm conclusions. Of the 53 patients involved in the trial, 42% saw an overall improvement in their condition — a sharp contrast to the 11% of placebo patients who also improved. But patients have so far only been tracked for a year. Results have also not been peer-reviewed, although they are being prepared for publication.

The debate over mechanism means that the design of the phase II trials will be carefully scrutinized. Over the next few months Osiris will be choosing which end points to monitor, and Mills admits it won't be easy.

Improvement in ejection fraction does not necessarily predict a better chance of survival or even improved health. And such is the case with most of the end points that can be easily measured during a large clinical trial. “We cast a pretty wide net during our phase I study, and we saw a handful of potential benefits,” says Mills. “Now we need to choose the best ones to measure.”

“This is a problem all young therapies face,” says Joshua Hare, a cardiologist at the University of Miami Miller School of Medicine in Florida who helped move many of the first FDA-approved trials forward. The ultimate test of effectiveness is extension of lifespan, but good survival studies often require a cohort of 5,000 people, and as it can take three to five years to conduct a survival study, such gold-standard trials are very costly.

Hare is spearheading a study sponsored by the National Institutes of Health that uses mesenchymal stem cells from a patient's own bone marrow to treat heart failure. Each of these research projects faces the same dilemma, he says. “When a therapy is being developed, it's common that researchers will have to try out different end points. Small samples and suboptimal end points might explain the perceived marginal benefits seen so far.”

Along with selecting the best signs of healing to measure, trials are also selecting patients that are more dramatically sick, Hare adds. Data from the early trials support the idea that improvements will be more pronounced for the worst cases. In the Osiris phase I trial, for example, patients who had suffered the most severe heart attacks saw a 6.2 point increase in ejection fraction in comparison to 5.2 point for more moderate cases. “We aren't going to make these guys bionic,” says Mills. “All the patients did well, but some didn't have as far to go before they hit normal.”

Searching for a mechanism

With so many trials under way, these theories can be put to the test. Douglas Drachman, a cardiologist at the Massachusetts General Hospital in Boston, believes that this second wave of clinical results may be the only way to advance the mechanism debate.

Given that the types of experiments used to uncover therapeutic mechanisms in animals are not possible in humans, clinical trials may be the best way to test how stem cells function in the body. How the stem cells work is a very different question from whether they heal, admits Drachman, but “at some point you have to just go ahead and see whether the therapy actually helps people.”

Drachman and his colleagues treated their first patient with blood-derived stem cells in February, as part of a national phase II trial run by the Illinois-based pharmaceutical company Baxter. The study aimed to alleviate the effects of chronic myocardial ischemia, and used endothelial progenital cells harvested from the blood of each patient.

Drachman's patients were the last treated in the study, called ACT-34 CMI, which involved more than 150 patients.

If the trial proves successful, the mechanism is implicitly verified, argues Drachman. In ACT-34 CMI, researchers hope that stem cells delivered with an electrode-guided catheter will differentiate and grow into microscopic blood vessels, improving blood flow to damaged areas of the heart. If nuclear imaging and MRI results show blood flow has improved, if echocardiograms show that the heart muscle is contracting better, and if patients perform better on treadmills, it's likely that the stem cells did their job, says Drachman. “Even though you can't implant a microscopic surveillance camera in someone's heart, it would be reasonable to assume that the stem cells did transdifferentiate into blood vessel cells.”

Yet in reality, no results in any trial have yet been this clear. “A 3% or 4% cardiac improvement that wears off after a few months does not answer any questions,” says Chien. And while the trials have not raised safety flags, they do pose the risk of exploiting vulnerable patients.

Although slow enrollment hampers many other types of clinical trial, researchers interviewed for this article say their trials fill up quickly, and that they are inundated by requests. If encouraging data comes in from the new batch of trials, phase III studies will probably start enrolling vastly larger numbers of patients — but researchers say that is years away.

Patients unable to enroll in clinical trials in the United States may make a pilgrimage to Germany, where despite the limited amount of data on efficacy, the government has been known to offer incentives to doctors that perform stem cell surgeries. Currently doctors receive government payments of around 5,000 euros (US$8,000) per stem-cell surgery. Controversial stem cell treatments have been available in Europe and South America and elsewhere for a few years; in Thailand, American patients pay around $50,000.

In spite of the fact that success stories from overseas encouraged phase I trials in the United States, the foreign surgeries have not been tracked with the rigour needed to prove safety and efficacy. This is why Chien is surprisingly supportive of the new wave of trials. Although he still believes the phase I trials were too hasty, he is now advocating for increasedgovernment funding. “It's not clear that it's working, which is why I am proactive about encouraging more research,” he says.

Burt relayed a more optimistic message at a congressional briefing in March. “When you have something that works, you keep going with it,” he said. “The message to Congress is that stem-cell therapies need continued support.”