HOUSTON — The drug shows promise: In one study, it reduced seizures for some children with a crippling epileptic condition. If approved, it would be the first medication specifically for those youngsters.
It would also be the first prescription drug in the United States made from cannabis plants.
But getting Epidiolex through clinical development has been a bureaucratic nightmare, some researchers say, with federal drug enforcement agents grilling physicians and demanding improvements in hospital security before they’d allow the medication to be tested. The Obama administration is considering relaxing rules on marijuana this summer. For now, though, researchers say the regulations restrict research into the potential benefits — and potential harms — of cannabis.
Dr. Angus Wilfong, who formerly led the comprehensive epilepsy program at Texas Children’s Hospital here, recalled being interviewed by two besuited agents from the Drug Enforcement Administration — “the men in black,” as he put it — before he could start giving Epidiolex to patients. The DEA, Wilfong said, made the hospital upgrade the security cameras in its pharmacy before it could get the drugs.
Dr. Orrin Devinsky, lead investigator on the Epidiolex trials, said he had to go through several federal and state agencies, host armed agents at his clinic, and purchase a new safe that was so heavy that engineers had to make sure the floor could handle it.
“I think everybody recognizes that the current policy, to put it nicely, is byzantine,” said Devinsky, a neurologist at the New York University Langone Medical Center. “It’s just the most crazy and discoordinated system.”
Epidiolex is made from a derivative of marijuana, which is available medically in about half of all states and recreationally in a growing number. But because the federal government classifies marijuana as a Schedule 1 drug — defined as having no therapeutic value and high potential for abuse — experts need the highest level of permission to study cannabis and its constituents, whether in basic research or a clinical trial with patients.
“There are tremendous disincentives,” said John Hudak, a senior fellow in governance studies at the Brookings Institution. (Other Schedule 1 drugs include heroin and certain hallucinogens. Cocaine and opioids have less restrictive classifications because they have been shown to have some medical use.)
That the Epidiolex trials were able to recruit patients at locations around the country demonstrates that cannabis research is possible and happening. And to some scientists, the extra steps are a reasonable precaution when they’re working with a substance that has been shown to hurt developing brains and upon which some people become dependent.
“There are some people who complain about” bureaucratic hurdles, said Bertha Madras, an addiction researcher at Harvard Medical School who served in the White House Office of National Drug Control Policy under President George W. Bush. But, she said, scientists surmount challenges on “so many issues” beyond cannabis research.
Epidiolex is made from a component of marijuana called cannabidiol, or CBD, that does not make people high. Its possible therapeutic benefits for children with epileptic disorders have lured families to places like Colorado, where recreational marijuana is legal. Other states have legalized CBD oil for certain patients to use. But some doctors warn that in those places, the drug is being meted out without regulations on what else might be in the oil, consistent dosages, or an understanding of its long-term effects.
In recent years, the DEA and the Food and Drug Administration have opened the door a bit to CBD research. And according to the DEA, the number of researchers registered to perform studies with marijuana and its components rose from 161 in April 2014 to 344 in March 2016.
The DEA has also said it will consider rescheduling marijuana in the first half of this year.
But for now, scientists seeking to study cannabis have to clear studies and go through background checks with the DEA and show that the product will be tracked and securely stored, with the excess destroyed. In the United States, all the raw cannabis studied comes from one federally regulated operation at the University of Mississippi, although the research restrictions apply whether a marijuana component is derived from the plant itself or is synthesized in a lab.
All those hurdles, many researchers say, take time and money. It recently took Ryan Vandrey and other researchers six months to get the nod from the DEA for a clinical trial looking at cannabis as a possible therapy for veterans with post-traumatic stress disorder — and the agency had no comments on the proposed study itself, Vandrey said.
“I don’t mind the DEA evaluating us as a risk for diversion,” said Vandrey, a behavioral pharmacologist at Johns Hopkins Medicine. But any further evaluation, he said, “is completely redundant to the reviews that have already been done by appropriate and accredited agencies,” including funding bodies, university review boards, and the FDA.
Dr. Eric Marsh, a pediatric neurologist at Children’s Hospital of Philadelphia, registered with the DEA to study Epidiolex because he’d heard anecdotal reports of CBD oil working for patients, and he wanted to test the drug in a randomized trial with monitored dosages and placebos — something that hadn’t been done before.
But, he said, “If someone came to me and said, ‘You have to do this all again,’ and I knew nothing about this substance, I’m not sure I would do it.”
GW Pharmaceuticals (GWPH), Epidiolex’s manufacturer, reported in March that children with Dravet syndrome — an epileptic condition that affects an estimated 1 in 20,000 individuals — had fewer seizures on the drug than children on a placebo in a 120-person, Phase 3 trial. The company plans to release initial results from another Phase 3 trial next month from patients with Lennox-Gastaut syndrome, an even more rare type of childhood epilepsy in which seizures are so severe and unexpected that some sufferers wear helmets with facemasks so they don’t hurt themselves.
GW hopes to ask the FDA to approve Epidiolex next year.
The FDA has shown a willingness to approve drugs related to cannabis that are shown to help patients. In the 1980s, it approved two synthetic drugs — Marinol and Cesamet — that resemble or are identical to THC, a component of marijuana that gives users a high. The drugs, which treat nausea from chemotherapy among other conditions, are listed as Schedule 3 and Schedule 2, respectively.
If the FDA gives the green light to Epidiolex, it would show that the drug has a medical benefit, meaning it cannot remain a Schedule 1 substance. In that case, the DEA could either put the GW drug itself on a new level or move CBD more broadly to a different classification. (The DEA did not comment about how it would schedule Epidiolex if it is approved.)
Alice Mead, head of US professional relations at GW (which is headquartered and makes its drugs in the United Kingdom), said that federal regulations on marijuana created unique challenges for the company, but “it’s not impossible” to conduct trials with cannabis-derived drugs.
“We have probably a hundred research sites licensed by both the DEA and [state agencies],” she said.
Still, the thought of working with the DEA to get an initial trial running was too much for Zynerba Pharmaceuticals (ZYNE), a Pennsylvania-based company developing synthetic cannabinoid treatments for epilepsy and other conditions. For its Phase 1 trial of a CBD gel, it tapped researchers in Australia, said company president Terri Sebree.
“We knew not having to go through the DEA would speed up the work,” she said.