Cathy Wolf read the report carefully. She had every right to be skeptical—in the 13 years since her amyotrophic lateral sclerosis diagnosis, she'd read dozens like it: Celebrex; minocycline; vitamin D—you name it, it could slow the progress of ALS. That is until it was tested in a clinical trial.

Many patients around the world also read the report when it was published online February 4, 2008, in Proceedings of the National Academy of Sciences—a top scientific journal. Some even got wind of the shocking conclusion earlier, when an Italian version of the abstract set off Google News alerts: Carbonato de litio rallentar la progressione della sclerosi laterale amiotrofica—Lithium carbonate delays the progression of ALS.

Lithium carbonate (often abbreviated to lithium) is a white salt used to treat mania—the upside of bipolar disorder. It was discovered serendipitously in 1948 by Australian psychiatrist John Cade during a series of crude but well-controlled experiments centered on the injection of urine into guinea pig abdomens. Urine from manic patients killed the guinea pigs faster than normal urine did, causing Cade to suspect that higher concentrations of uric acid might contribute to mania. So he added lithium salt to buffer the manic acid. To his delight, it made the injection more tolerable. After taking a dose himself to make sure lithium was safe for humans, Cade ran a small-scale trial in some of his manic patients. The calming effect was so robust that he concluded (wrongly) that lithium deficiency caused mania.

In Cade's time and the 50 years that followed patients would get their information about drugs from their doctors. But nowadays they can read the same reports their doctors do, contemporaneously. They find them online and use social networks to discuss them with other patients. This phenomenon has been called the "patient 2.0 movement," and it is changing the way doctors and scientists do their jobs.

Using forums like PatientsLikeMe (an online patient community), Facebook and Twitter, patients with ALS, cancer and a range of other conditions are turning biomedical research—a process once mystified by jargon, privacy, peer review and expensive journal subscription fees—into anopen-source and collaborative effort. Although the movement is empowering patients, it worries some doctors. "The Internet is really good for sharing information, but not so good for verifying it," says Richard Bedlack, director of Duke University Medical Center's ALS clinic. The movement, he says, promotes uncorroborated results and opens the window for fraud, giving medical con artists the perfect platform to tout or sell bogus treatments detailed in counterfeit reports, uncritiqued by experts in the field.

The lithium report, however, was not a fraud—there was good reason to suspect lithium might slow ALS, a neurodegenerative disease that causes paralysis and death, usually within five years. In addition to its stabilizing effects on the excitatory neurotransmitter glutamate (the actual way lithium exerts it mood-stabilizing effects), the drug induces autophagy—a process by which a cell can "eat" its own components—including those, which in the case of ALS, might be faulty. The report was a collection of evidence from mice and patients that lithium protected dying motor neurons, preserved their function and delayed their death.

But the study did have limitations and flaws. The sample size was small (only 16 patients took the drug), not blinded and not placebo-controlled (the subjects and the researchers knew who was getting the drug). And patients in the treatment group appeared to have a more slowly progressing form of ALS before the study started, which begged the question: Might these patients have fared better even without lithium?

Because ALS has no cure, the U.S. Food and Drug Administration's Investigational New Drug application (better known as "compassionate use") lets patients request FDA-approved drugs off-label, meaning for purposes other than those for which they were approved. Faced with a death sentence (most patients die in three to five years), ALS patients often feel they have nothing to lose in experimenting with alternative and off-label treatments. "We're dealing with a very vulnerable patient population. Of course, they would be significantly swayed by anything that looks promising," Bedlack says.

But certain off-label drugs have accelerated the disease or confounded it with painful side effects like headaches and muscle cramps. In 2004 a promising report on the effects of minocycline (an antibiotic routinely prescribed for acne) prompted ALS patients to take it off-label. Subsequent clinical trials revealed that patients taking minocycline actually fared worse than those taking a placebo.

"I hate that term, 'compassionate use'. It suggests that people who don't prescribe a particular off-label drug aren't compassionate, which is not the case at all," Bedlack says. "This is one of the worst diseases to affect mankind, and the only way we're going to find a treatment is by doing clinical trials."

But when desperate patients, having read the February 2008 report, swamped clinics with requests for lithium, many doctors obliged. By April 2008, some patients were reporting miraculous results in forum posts. Cathy Wolf, a trained scientist with a PhD in psychology, blamed the placebo effect. But come May, she too asked her doctor for a prescription.

In the past two decades scientists have learned more about the biology underlying ALS than ever before. Sadly, little of this insight has translated into better treatments. And because the disease progresses in such a short period of time, few patients (less than 10 percent in the U.S.) are willing to spend their precious remaining days taking what might turn out to be a placebo.

Still, a randomized, double-blinded, placebo-controlled clinical trial was the only way to prove lithium could actually slow ALS progression. If lithium didn't have an effect, patients using it off-label were opening themselves up to side effects and missing opportunities to try drugs that might work. So, researchers faced an enormous challenge: What would be the incentive for patients, many of whom were so convinced lithium might help them, to enter a trial when they might only get a placebo?

Four independently run lithium trials were launched in 2008, all of which used nontraditional designs in hopes of getting patients to enroll. One allowed patients in the placebo group to switch into the treatment group if their disease progressed; another used a historical control group; and a third allowed control group patients to replace the placebo with lithium, but at a lower dosage. Some patients dropped out of the trials because they thought they were in the control group.

At the same time, patients in the 2.0 movement were running a trial of their own.

PatientsLikeMe introduced the first "real-time, real-world, open, nonblinded, patient-driven" lithium trial. Patients taking the drug off-label were encouraged to enter their blood levels (which were measured weekly because of lithium's toxicity at high doses) and their ALS functional rating scale score (a measure of disease progression) on an online spreadsheet. Patients who weren't taking lithium served as controls. Over 12 months, 149 patients taking lithium for at least two months and 447 controls tracked their disease progression for anyone to see.

"Patients used to think that experimentation was something that scientists and doctors did. I don't think life is like that anymore. Patients say, 'I don't want to wait, I want to try something,'" says Paul Wicks, director of research and development for PatientsLikeMe.*

Cathy Wolf is one of those patients. On May 5, 2008, a nurse crushed her first 150-milligram lithium pill, dissolved the white powder in liquid food, and introduced it directly into her stomach via a tube. By May 16, she reached her target blood level. But a week later, she stopped taking it—it made her "jittery." Although nearly one third of the patients taking lithium stopped early because of side effects, many continued the virtual trial.

The results from three of the four researcher-led trials were presented at the 62nd American Academy of Neurology annual meeting, held this April in Toronto two years after the initial report. All three trials were negative. Like countless drugs that preceded it, lithium just didn't seem to work. At least it was safer than minocycline, and cheaper than stem cell transplants in Mexico—another alternative treatment that some ALS patients are experimenting with against doctors' advice. The PatientsLikeMe trial results were also negative.

Bedlack worries the arrival of the virtual trial will encourage more patients to experiment with alternative and off-label treatments outside of properly designed clinical trials. "There’s still no substitute for the randomized, double-blind, placebo-controlled trial," he says. But he hopes to find a happy medium between the old ways of doing research and new Web-based ones to encourage participation and reduce some of the burden on patients.

In an effort to educate patients in the 2.0 movement about the risks of experimental and even fraudulent treatments, Bedlack created ALSUntangled—a Web-based system monitored by an international team of scientists that investigates new treatments at the request of patients. It uses social networking tools to collect and review information then publishes its findings as open-access articles in a peer-reviewed journal, Amyotrophic Lateral Sclerosis. "We have to help patients fight smart, not just hard," he says. "Ultimately, we all want the same thing. We have to find a way to work together that recognizes the unique strengths and needs of all parties involved: patients, clinicians and scientists." So far, the group has published four investigations and has one more submitted for review. Their list of pending investigations, suggested by patients via Twitter, sits at 27.

*Note (7/12/10): This sentence was edited after publication to correct Paul Wicks's title.