Hans Keirstead wakes up every morning at his home near Los Angeles and checks CNN. He’s looking for news about the first-ever human test of embryonic stem cells, launched in October by the biotechnology firm Geron. Mostly, he’s looking for bad news. “If someone dies, or is in pain, then it’s over,” he says, pushing a hand through his tawny hair. Keirstead, dressed in a loose linen shirt and wearing a thumb ring, is a biologist at the University of California, Irvine, who has variously been called the “rock star,” “miracle worker,” and “Pied Piper” of stem-cell science. Today he has a corner office in a new $67 million research center paid for in part by California voters, whom he helped persuade to vote for a $3 billion stem-cell spending plan in 2004 with a video of partially paralyzed rats walking again after stem-cell transplants performed in his laboratory.

That same treatment is now being tested in human beings. No wonder Keirstead is anxious. Although he is not directly involved in the clinical trial, the discovery he patented, promoted to Californians, and later licensed to Geron has now become the leading test of whether embryonic stem cells will finally live up to their medical potential. “I’m dying to know if it works,” he says.

As Technology Review went to press, Geron had so far treated two patients: a 21-year-old nursing-school student named T. J. Atchison, who was paralyzed at the chest in a car crash last September, and a second person who has not been publicly identified. The hope is that cells injected into their spinal cords could help mend damaged nerves and restore at least a degree of mobility and sensation. Even if the treatment fails, many researchers believe the test is a critical step toward a time when bodies are healed and regenerated with living cells, not chemical drugs. “Cell therapy is now here to stay,” says Wise Young, a professor at Rutgers University and an expert on spinal-cord injury. “I tell my students that this will be the future—that they will be the first generation of doctors to use cell therapy.”

Thirteen years of public debate, scientific surprises, lawsuits, and presidential decrees have gone by since embryonic stem cells were first isolated, in 1998. Stem cells drawn from early-stage human embryos have the potential to develop into any type of cell in the body. In a lab dish, they can give rise to nerves, skin, even pulsating heart cells. And Geron, a 180-person biotech outfit in Palo Alto, has promised for a decade that treatments based on the cells could be just around the corner. The company says it spent $45 million on amassing the evidence needed to persuade the U.S. Food and Drug Administration to allow the first-of-a-kind human trial to proceed—an effort that included animal tests it calls exhaustive. “The agency told us our application was the largest they’d ever received,” says Geron’s interim CEO, David Greenwood, sweeping his hand over a double-length conference table that once creaked under the weight of all 22,500 pages.

Geron’s success in getting the FDA to green-light the trial has already triggered a small explosion of other embryonic-stem-cell studies. Advanced Cell Technology, a smaller competitor in Marlborough, Massachusetts, has been cleared to begin two trials that will involve replacing cells in the eyes of people going blind from macular degeneration: lab workers will use stem cells to manufacture a type of retinal pigment cell that the disease kills off. Next in the pipeline is a startup company’s effort to transplant lab-grown replacement nerves into infants with a fatal genetic disease called spinal muscular atrophy. That trial is planned by California Stem Cell, which has raised $10 million from wealthy donors and has signed up Keirstead as its chief scientific advisor. Keirstead, bounding through the still empty offices with a tape measure in hand, says he is considering leaving his lab to join the company full time. He thinks that with the Geron trial now under way, other human studies can advance much more quickly and cheaply.

But that depends on what happens in the Geron trial. And even some of stem cells’ most ardent advocates worry that things may be moving too fast. Arthur Caplan, a bioethicist at the University of Pennsylvania and a defender of stem-cell research (see Q&A, September/October 2006), calls the Geron study poorly designed and says it should never have been allowed to proceed. “This is nuts and hugely risky,” says Caplan. “The animal studies are not adequate to justify the trial.” Those studies provide too little proof of safety, he contends, and Keirstead’s original findings in rats offer thin evidence that people will be helped.

Looming large is the history of gene therapy, another advanced biomedical technology, which badly misfired when a young volunteer named Jesse Gelsinger died in a safety study in 1999. Caplan, who was close to those events, sees worrisome similarities (see “The Glimmering Promise of Gene Therapy,” November/December 2006). “If they get an adverse event, there will be hell to pay,” he says.

No Miracle

Spinal-cord injuries cause paralysis by killing off nerves that transmit sensory impulses and leaving others stripped of their myelin sheath, the layer of fatty insulating material that helps nerve signals travel. Geron manufactures its treatment, known as GRNOPC1, by coaxing embryonic stem cells to form what are known as oligodendrocyte precursor cells. Those cells are bottled and frozen, and Geron scientists believe they may help restore some degree of sensation and limb movement to patients if transplanted soon after a spinal-cord injury. That is because oligodendrocyte cells produce myelin and may serve other purposes as well, such as encouraging new blood vessels to form. In Geron’s initial human trial, designed to test the safety of the treatment, doctors plan to inject two million cells each into the spines of 10 people whose legs have been paralyzed in accidents.

Will the treatment be a cure? The odds are against it. In general, most new treatments, never mind highly experimental ones, bomb out early. What’s more, GRNOPC1 faces an uphill fight against medical dogma, which says that it’s impossible to reverse damage to either the brain or the human spinal cord. That means few experts expect a miracle from GRNOPC1. Richard Fessler, a surgeon at Northwestern Memorial Hospital in Chicago who is leading patient recruitment for the Geron trial at seven U.S. medical centers, calls the study a “rational” attempt to reverse spinal-cord damage. But he cautions against expecting too much. “We wouldn’t be doing this if we didn’t have hope, but I don’t want to instill false hope,” Fessler said in a news conference in May, after the second patient received the treatment. “I’m not going to go to one of these patients and say, ‘We’re going to give you a transplant and you’re going to walk.’”

Still, some patients are clamoring to join the Geron study, even though only people with extremely recent injuries—the kind that lab research suggests might be helped—are allowed to participate. A Dutch man offered Geron $1 million to treat his son, and ­Keirstead says he received an even bigger offer from a paralyzed Texas millionaire. “He said he’d pay me whatever millions it takes to set up a clinic in Mexico, and another $2 million for me, just to treat him,” he says. “It made me pause, but not for long.”

One person who made a public plea to enter the trial is Michael Martinez, a 24-year-old jockey who was paralyzed after falling from a horse at San Francisco’s Golden Gate Fields last year. Martinez was refused, in part because his injuries, including three crushed vertebrae, were too extensive. “He is the most challenging candidate for stem cells—if they can have any impact in him, that would be extraordinary,” says David Seftel, the doctor who came to ­Martinez’s aid at the track and has led a campaign to have him treated with stem cells. Seftel complains that the spinal-cord specialists who treated Martinez view stem-cell research with skepticism and were reluctant to get behind the idea. “We experienced a lot of resistance,” he says. “We were told it’s an irresponsible option to present to patients at this time. But the science only advances if people take carefully calculated risks.”

With Seftel’s help, and a letter-writing campaign by other paralyzed people, Martinez is now a candidate (pending Swiss government approval) to join a study in Switzerland sponsored by a California company called StemCells Inc. In that study, doctors are implanting nerve cells obtained from early-stage human fetuses; unlike embryonic stem cells, such fetal cells have already begun to differentiate into other cell types. “We have moved heaven and earth here to make sure he gets in,” says Seftel. If he’s approved, Martinez will travel to Switzerland for screening and then, if he passes, undergo 30 days of additional tests before being cleared to participate in the trial.

Martinez, a Panamanian who speaks little English, says he believes the operation “could help me regain sensation in my legs, and return to capacities I had before.” As with most paraplegics, being unable to walk is the least of his problems. His biggest difficulty is bladder infections, since he must urinate through a catheter. Martinez says he’s aware that there are dangers associated with the stem-cell treatment, but as a jockey, he’s used to long odds. “I know it has certain risks, but I don’t want to think about those,” he says. “I want to stay focused on the positive.”

Too Powerful

The human embryonic stem cell was isolated in 1998 by James Thomson at the University of Wisconsin (Geron, in a farsighted gamble, funded his work). Thomson made two main scientific claims about his discovery. The first, and better known, involved the cells’ capacity to differentiate into any tissue type in the body. Less well understood but equally important was that embryonic stem cells are immortal: they keep dividing, never running down as normal cells do. They are, in short, like no other human cells.

The truth of those claims is evident at Geron’s one-story headquarters in Palo Alto. In its labs, the company grows not only nervous-­system cells but also heart muscle, which is being transplants into 100-pound pigs, and cartilage cells that are being tested in the knees of sheep. Amazingly, all the billions of cells that Geron has grown for its spinal-cord program—including those injected into Atchison’s spine—are direct descendants of the very first supply of stem cells that Thomson created from an embryo, a cell line known as H1. “There is no further destruction of human embryos required to keep this work going, and there hasn’t been since 1998,” Ed Wirth, Geron’s medical director, told a Phoenix audience last year. “[It's] very, very powerful how you can multiply these cells.”

If anything, embryonic stem cells are too powerful. Early on, scientists hoped they would be magic bullets for a variety of diseases. Just inject them—and watch them race to injury sites and fill in for dying cells. In one early study, embryonic stem cells placed in the brains of rats suffering from symptoms of Parkinson’s disease did precisely that. Not only did the cells become new neurons, but they began to squirt out dopamine, the chemical lost in Parkinson’s. The problem was that they often ran amok, multiplying into frightening tumors called teratomas—disorganized mixtures of tissues, such as teeth, hair, and jawbone. Rats that developed such tumors died.

The brain tumors were a sign that the stem cells were still attempting to carry out their original mission: to form an entire person. Researchers quickly settled on a new strategy. They would use stem cells, but only to manufacture daughter cells restricted to a particular destiny—cells already committed to becoming liver, say, or new muscle. “No one wants to put embryonic stem cells into humans, only the product,” Keirstead explains today. What he worked out at his Irvine laboratory was a recipe for turning embryonic stem cells into relatively pure populations of oligodendrocyte precursors. It’s not easy: his recipe requires 42 days of coaxing, coddling, and adding growth factors at just the right moment. Then, in 2005, Keirstead published a report saying that when he injected the oligodendrocyte cells into the spinal cords of crippled rats, they went from dragging their hind paws to walking again in a matter of days. That result was a bombshell, and Geron, which has poured $1.8 million into Keirstead’s lab, quickly decided that pursuing a stem-cell treatment for people with spinal injuries would become the company’s flagship program.

One of Geron’s challenges has been to create an industrial recipe for growing the large numbers of cells needed for treating patients. The company’s senior director for manufacturing operations, Sean Cullen, says Geron is now, with its technology, where companies such as Amgen and Genentech were with protein and antibody drugs a decade ago, when they began manufacturing them. But if proteins were harder to make than ordinary chemicals, cell therapy is an order of magnitude more difficult still. “Think about it,” Cullen says, pointing through a glass porthole into the clean room, where the cells multiply in jars of pink medium. “The cell is a living thing—you can’t define what it is with the same granularity.” Indeed, the product that Geron makes can’t be characterized like a chemical compound. Rather, it’s a mixture of different types of cells, including oligodendrocytes. The manufacturing process is in many ways still undefined, Cullen says—still an art. When he heard that cells he’d cultured had been injected into someone’s spine, “that brought it home,” he reflects. “Now you know it’s life and death.”

Risks versus rewards

What worries some scientists is that Keirstead’s results in rats have never been independently confirmed and published. That’s not unusual in science, but it may be reason for caution in this case, since many discoveries in the stem-cell field have later unraveled. “I do think it matters if it is replicated,” says Thomas Lane, a neuro­scientist at the University of California, Irvine, who once collaborated with Keirstead to use the cells in mice with symptoms of multiple sclerosis, only to find that the cells didn’t survive and did not appear to produce new myelin. While the two studies can’t be compared head to head, says Lane, “at the end of the day [the cells] didn’t work for us.”

One complication in trying to reproduce the results is that other labs may begin with different populations of embryonic stem cells, and each lab has its own tricks for inducing the cells to differentiate, which makes direct comparisons difficult. Wenbin Deng, a professor at the University of California, Davis, has tried to replicate Keirstead’s recipe, and the results leave him cautious about human tests. “I think it’s still a little bit premature at this point,” Deng says. “Even though this type of cell is ideal for transplant studies, there is still a lot of uncertainty about their safety and efficacy.”

Geron scientists say they have replicated and extended ­Keirstead’s findings, although the data haven’t been published. “We would like to publish, but that is not the focus of the team,” says Anna Krassowska, a stem-cell scientist who now works as Geron’s director of investor relations. “Sometimes there is the perception that our entire trial is based on the seven rats of Hans Keirstead, and that is not true.”

Yet even if the treatment heals rats, it is still unclear exactly what it does. Originally, the theory was that new oligodendrocytes should restore the missing myelin on axons, the projections of nerve cells that transmit electrical signals. But Ann Parr, a spinal surgeon and researcher at the University of Michigan, says the benefits appear so quickly—in a matter of days—that new myelin can’t be the whole story. Maybe the cells emit chemicals that help prevent ongoing damage in some other way. “I think there is pretty good evidence that transplanting the cells can have a beneficial effect, but nobody knows how they work,” says Parr.

For critics such as Caplan, the caveats add up to serious doubts. He says he doesn’t see a reason for human tests given the “unimpressive” results in rodents, whose injuries were not as severe as those of Geron’s human subjects. What’s more, the patients Geron is treating aren’t terminally ill. People who are paralyzed in accidents often adapt after the initial shock and return to relatively normal lives. “At first you think they don’t have much to lose,” Parr says, “but they actually do. They could die.” None of those concerns weighed too heavily on Atchison, Geron’s first patient. He signed the forms to join the trial only 30 minutes after reading them. Since the injection, Atchison has worried more about the prospect of developing a tumor, but he has come to terms with the danger. “Even if I became sick,” he says, “I would still be contributing to the health of someone else, somewhere down the line” (see “The Right Decision“).

The job of balancing the evidence for and against stem-cell therapy fell to the U.S. Food and Drug Administration in 2008, when Geron first submitted its application to test the treatment in people. For the agency, which is charged with ensuring the safety of all medicines, embryonic stem cells were not only a charged political subject but a huge technical challenge. When the FDA called together its top advisors that year, at the Hilton Hotel in Gaithersburg, Maryland, to discuss whether to approve Geron’s treatment, one participant called stem cells “probably the most complex biological therapeutic humanly imaginable.”

Everyone was well aware of how some previous attempts to alter the body’s cellular and genetic makeup had gone wrong. In addition to the infamous gene-therapy death, there was the case of Parkinson’s patients who began to experience uncontrolled movements after receiving transplants of tissue from fetuses. Also worrying was a French study a decade ago in which transplants of genetically altered bone marrow had cured children of severe combined immunodeficiency, or “bubble boy” disease, only to cause leukemia years later. Unlike ordinary drugs, whose action quickly fades, these treatments threatened to get stronger. “For some products,” agency officials noted, “unchecked proliferation is a real possibility.”

The FDA’s overriding worry was that a stray embryonic stem cell could cause a tumor. After the 2008 meeting, the agency told Geron that its trial couldn’t proceed. The problem: some of Geron’s rats had developed tiny cysts where the treatment had been injected. Jane Lebkowski, Geron’s chief scientific officer, says the growths were harmless masses of epithelial cells, like “microscopic water balloons.” Harmless, maybe. But they didn’t belong in the spinal cord, and who knows how much they might grow during a human lifetime. Lebkowski says Geron adjusted its manufacturing recipe to eliminate the unwanted tissue. But it took the company another two years—and a hundred or so more rats—to persuade the FDA that its product was safe enough to test in people. Even so, the FDA demanded unusual safety precautions—stipulating, for example, that the patients be tracked for years to come.

Such delays add up, and some believe the FDA is creating a roadblock. Several companies have gained the agency’s approval to test injections of stem cells taken from immature human fetuses, an older but related technology that has also raised concerns at the FDA. “The problem is that the agency is overworked and understaffed and isn’t so familiar with cell therapy,” says Richard Garr, CEO of NeuralStem, a company that recently began tests of fetal spinal-cord cells in patients with Lou Gehrig’s disease, or ALS. For NeuralStem’s study, like Geron’s, the agency required that patients be treated at least 30 days apart, to allow time to tell whether problems would arise. It will take the company at least a year and a half to complete the study, given that 18 patients are expected to enroll. “And here is a disease that kills you in three to five years, on average, after diagnosis,” says Garr. “So it feels like they are slowing you down. I think the FDA believes the caution is justified. I can tell you that the patient advocacy groups are frustrated.”

Indeed, many patients opt not to wait. Unregulated clinics, cranks, con men, and quacks have popped up from Cancún to Beijing, tempting patients to pay as much as $40,000 for the chance of a stem-cell miracle. But medical tourists who dodge U.S. safety regulations for overseas injections run unknown risks. In 2009, Israeli doctors treating a 13-year-old boy reported the first case of a brain tumor caused by a stem-cell therapy. The boy’s parents had taken him to a fly-by-night Moscow clinic where cells gathered from human fetuses had been injected into his brain.

Patients who join the Geron study, by contrast, will be subject to a battery of MRIs, blood tests, medical exams, and follow-ups lasting 15 years. Perhaps because of the demands put on candidates, the trial has been moving at a crawl. In April, the company surprised investors when it reported that in six months it had managed to enroll only one patient. The good news was that the patient, Atchison, had suffered no unexpected side effects. The bad news, even after the second patient was enrolled in May, is that at this pace it will take Geron an agonizing three years to finish. CEO David Greenwood says that the company has asked the FDA to loosen the strict criteria for subjects. “Cell therapies are new, and the agency appropriately, I think, takes a very conservative posture,” he says. But, he adds, “you can narrow your funnel so much you don’t get any patients.”

Back at his office at UC Irvine, Keirstead says he has received several phone calls from people who were considering joining the Geron trial: “They were looking for a level of confidence, a feeling from me. Is it really going to work, and is it safe?” The calls have put Keirstead in a difficult spot. “My ridiculous sense of optimism may be clouding my judgment,” he says. “But I tell them we’ve done everything we can possibly do scientifically and in animals. And we still don’t know if it works in humans.”

Antonio Regalado is the Latin America contributor to Science magazine. He is based in São Paulo, Brazil.