Edward Maa did not plan to become a marijuana researcher. But a few years ago, when the neurologist and epilepsy specialist surveyed his patients about their use of alternative medicines, he discovered that more than a third had turned to marijuana to try to control their seizures. “I had no idea,” says Maa, who is chief of the Comprehensive Epilepsy Program at Denver Health.

Now he is trying to impose some scientific rigor on what has become a very big and unscientific ad hoc experiment in his state, where medical marijuana use is legal. According to the Epilepsy Foundation of Colorado, the widely reported case of Charlotte Figi, a child whose nearly constant seizures were dramatically curtailed with cannabidiol, a marijuana ingredient, has helped trigger an influx of families from around the U.S. seeking similar treatment for their children with seizure disorders. Maa wants to move beyond anecdote and into data. He is monitoring 150 epilepsy patients who all take a product derived from the same strain of marijuana that Figi used, provided by the same source. Over the course of a year, he intends to compare dosage to seizure activity and side effects, as well as patient characteristics, to see if any patterns suggest the drug is effective—or not—in particular situations. “My position is, let's see what's going on,” Maa says. “Let's see if this is helpful and try to understand what we are seeing.”

Understanding the biology and chemistry behind marijuana's claimed medical benefits is becoming extremely important now that 23 states and the District of Columbia allow the use of marijuana to treat some medical conditions, including pain, nausea and glaucoma. Other states are expected to follow suit. Four states and the District of Columbia have legalized it for recreational use as well. Although the federal government still lists marijuana as a Schedule I drug, a class “with no currently accepted medical use,” a body of recent research—most of it done in test tubes and animals, but some done in people—suggests that cannabinoids, which are the active ingredients in marijuana, may have medicinal uses even beyond the approved ones. They might protect the brain from the effects of trauma, ease the spasms of multiple sclerosis and reduce epileptic seizures. Further preliminary work indicates that the chemicals may slow the growth of tumors and reduce brain damage in Alzheimer's disease.

This is your brain on pot

The chemical that induces marijuana's trippy effects, delta-9 tetrahydrocannabinol (THC), was isolated in 1964. Several other components have been described since, including cannabidiol, the compound used by the epilepsy patients, which does not make people high. In the late 1980s and early 1990s scientists began to identify and map two groups of molecules, known as receptors, in the central nervous system and immune system that help cannabinoids bind to cells. That interaction appears to play a critical role in marijuana's various effects. (The brain contains small amounts of its own, naturally occurring cannabinoids, which also bind to these receptors.)

CB1, the more common of the two main receptors, is widely distributed in the brain, with high concentrations in the cortex and the hippocampus (a region important to forming new memories). CB1 receptors also occur in parts of the brain involved in pain perception. There are low levels of CB1 in the brain stem, where cardiac and respiratory functions are regulated; their relative scarcity in this region may explain why, unlike opioids, even heavy doses of cannabinoids do not pose acute threats to the heart or your ability to breathe.

CB2, the other main cannabinoid receptor, is found mostly in the immune system. Its presence there interests scientists because the immune system triggers inflammation, and studies show marijuana can have an anti-inflammatory effect.

In the brain, when the psychoactive component THC links up with CB1, it slows down or blocks the release of a variety of neurotransmitters—signaling molecules released by neurons—including glutamate and dopamine. The result is the high that marijuana is best known for, often along with temporary impairment of short-term memory. Two other well-known effects of the THC-CB1 linkage are the stimulation of appetite, a boon for AIDS patients and others who need to maintain body weight, and the suppression of nausea, a benefit for some cancer patients receiving chemotherapy. THC has also been shown to disrupt the transmission of pain signals.

Recent research suggests that THC might also protect neurons from trauma. Early test-tube studies pointed to this effect, and so has one clinical trial published last October. In it, trauma surgeon David Plurad and his colleagues did a retrospective review of 446 traumatic brain injury (TBI) cases treated at Harbor-UCLA Medical Center from January 2010 through December 2012. Their study, reported in the journal American Surgeon, found that 82 of those patients tested positive for THC and two of them died, for a mortality rate of 2.4 percent. The mortality rate among the 364 patients who did not have THC in their system was 11.5 percent, nearly five times higher. After taking into account other factors, such as age, severity of injury and blood alcohol level, the researchers concluded that the link between THC and a lower death rate in these patients stood up. Although the mechanisms are not fully understood, previous research suggests that both THC and cannabidiol may increase blood flow in the brain, bringing needed oxygen as well as nutrients to endangered neurons. Because they inhibit glutamate, they may also prevent toxic effects that occur after brain trauma, when neurons can get overstimulated by the neurotransmitter.

Marijuana, of course, impairs perception and reaction time, so it may have contributed to the accidents that Plurad studied at the same time that it helped some people survive them. The irony is not lost on the surgeon. “There is never going to be one answer for marijuana,” Plurad says. “It's good for you, it's bad for you. It will never be one or the other. It will always be somewhere in between.” Some research, including a recent study funded by the National Institute on Drug Abuse, has shown that heavy use of marijuana (at least four times a week for the past six months in the paper) can lead to adverse changes in parts of the brain associated with reward and decision making. Plurad warns against such heavy use and use by teenagers. “As a clinical person,” he says, “what's interesting to me is that when you get down to the nitty-gritty of taking care of patients, it's cheap. And if it has valuable applications, then we should pursue it.”

Developing a drug

Drug companies are already in pursuit, working on compounds that show the benefits without the cognitive problems. GW Pharmaceuticals, a British firm, has developed two marijuana-derived drugs, Epidiolex and Sativex. Epidiolex, a purified form of cannabidiol, is intended to treat seizures and is being tested in an international clinical trial led by the University of California, San Francisco, Epilepsy Center. It has already been granted orphan drug status—a path to approval based on smaller clinical studies than normal—by the Food and Drug Administration. Sativex, a mouth spray that contains THC and cannabidiol, is approved in Canada and several other countries, but not the U.S., for the treatment of muscle spasticity in multiple sclerosis. It is also being tested as a pain treatment. As Maa and others point out, pharmaceutical cannabinoid medicines offer consistent potency and make dosage easier to control—critical factors in many cases, especially with pediatric patients.

In addition to these conditions, studies in animals and in cells, published in 2014, suggest that cannabinoids might eventually play a helpful role in treating three other kinds of disease. After inducing human breast cancer tumors in mice, researchers in the U.K. found they could shrink the tumors by administering THC. The chemical may disrupt cancer cell growth as it binds to CB2 receptors, which are much more abundant on cancer cells than on healthy ones. At the University of South Carolina, a team discovered that THC could reduce the inflammation associated with autoimmune diseases by suppressing the activity of certain genes involved in the immune response. And at the University of South Florida, researchers working with cells in a lab showed that extremely low concentrations of THC could reduce production of beta amyloid, the protein that forms the plaque abundant in the brains of Alzheimer's patients. It will, however, be several years before scientists learn how well THC will work, if at all, in patients with the condition.

For Lester Grinspoon, the new findings are gratifying but not surprising. Grinspoon, associate professor of psychiatry emeritus at Harvard Medical School, is the American godfather of medical marijuana. Now 86, Grinspoon used marijuana to treat his young son's chemotherapy-related nausea in the 1960s. He wrote a book about marijuana's benefits in 1971 and, after decades of research and controversy, continues to endorse them (see his Web site: RxMarijuana.com). He is pleased the nation at last appears to be catching up with him. “It's about time,” he says. He notes that before World War II, marijuana was listed as a medicine in the country's encyclopedia of drugs, the United States Pharmacopeia. “When marijuana is finally readmitted to the U.S. Pharmacopeia, a place it lost in 1942, it will be seen as one of the safest, least toxic, most versatile drugs of that whole compendium,” he predicts.