As an emergency room doctor, Kyle Kingsley felt frustration at seeing the drug-addicted patients return to his ER over and over, trying to find a doctor who would prescribe them more opioid painkillers.

Those pain pills cause twice as many deaths in Minnesota as street heroin, even though both are made from same plant, the poppy. Now Kingsley is hoping to use a different plant — cannabis — to divert pain patients from opioids to marijuana drugs. But the ambitious plan will require approval from one of the toughest bureaucracies on the planet: The U.S. Food and Drug Administration.

"This is a real alternative," said Kingsley, whose Minneapolis company Vireo Health LLC is one of two in Minnesota with a license to cultivate medical marijuana and convert it to oils and pills. "Anytime you are talking about potential FDA-track medicines, it takes years for this to happen. But in parallel to our current production of medications in Minnesota and New York, we're going to pursue these FDA endeavors."

The effort is separate from last week's announcement that the Minnesota Health Department is adding intractable pain to the list of reasons why patients can be prescribed medical marijuana next year.

Rather, Vireo's long-term goal is to figure out how to make non-smokeable marijuana drugs with clinically reliable potency and solubility that can be tested in controlled studies at academic medical centers. The study data would support an application for FDA approval, which would make Vireo's patented cannabis drug legal in all 50 states, according to Kingsley and some attorneys.

"I think if it is FDA approved, they would have to consider it lawfully available in all jurisdictions," said Minneapolis FDA regulatory attorney Mark DuVal, who is not representing Vireo. "It's a function of the Interstate Commerce clause to the Constitution."

However, marijuana is classified as a Schedule 1 drug, meaning that it is highly addictive and has no accepted medical use. Either Congress, the FDA, or the Obama administration would have to take a step like moving it down to Schedule 2, which includes addictive drugs that do have medical uses.

The data to show clinical safety have been elusive in part because the potency and soluability of the active chemicals in marijuana vary widely. Vireo chemists believe the key to making stable (and patentable) marijuana drugs lay in a chemical compound called Captisol.

Last month, Vireo announced a deal to buy Captisol from its maker, Ligand Pharmaceuticals, which has supplied the compound to major drugmakers like Merck & Co. and Amgen to produce drugs that were eventually approved by the FDA.

"We hope that Captisol will enable these medications to be more effective, and it will allow us to pursue proprietary drug pathways with these medicines," Kingsley said of the cannabis drugs. "We are very excited about this possibility. It could change pain medicine as we know it."

Even longtime critics of medical marijuana can applaud a rigorous approach toward gathering safety and effectiveness data on marijuana-based drugs.

"I'm all for it when companies legitimately try to come to market through the FDA in a legitimate process," said Dr. Eric Voth, a Kansas physician and chairman of the Institute on Global Drug Policy. "I think that is the way meds should be brought to the market, not by popular vote. And that is what we've had across the country, is medicine by popular vote."

Regulated clinical trials like the ones Vireo is contemplating could help quantify some of the reported negative effects of cannabis drugs, like changes in psychiatric disorders and increased driving fatalities. But Voth is skeptical that Vireo's effort will produce any groundbreaking treatment for chronic pain.

The scant data from reliable studies so far don't seem to show a major benefit for the treatment of pain, relegating the drug to a secondary treatment used in combination with another therapy, he said.

"As far as it being the new panacea that eliminates narcotics, I don't think anybody is suggesting that at all," said Voth, whose clinical specialties include addiction and pain medicine.

Such work is costly. A January report from the Tufts Center for the Study of Drug Development says it now costs $2.6 billion for pharmaceutical companies to develop and gain marketing approval for one new drug.

"We are not planning on $2 billion," Kingsley said, when asked how much Vireo plans to spend to get approval. "We have a lot of academic (research) partnerships, and we are excited that we can do this in a very cost-effective way. We are not going into this naively. We know this could potentially be very expensive."

Vireo is a private company established earlier this year, and doesn't reveal its finances. It's the parent company of Minnesota Medical Solutions, a 3-year-old company that had a $13 million private stock offering in April.

Securities filings said 83 investors had already chipped in money. MinnMed started running Minnesota dispensaries this year.

The FDA says much more information is needed on the efficacy of using marijuana drugs to treat conditions like glaucoma, AIDS wasting syndrome, chronic nerve pain and multiple sclerosis.

The need for research is underscored by the fact that unapproved substances like marijuana vary considerably in their potency and purity, FDA officials say.

The FDA "has an important role to play in supporting scientific research into the medical uses of marijuana and its constituents in scientifically valid investigations as part of the agency's drug review and approval process," the agency's website says. "As a part of this role, the FDA supports those in the medical research community who intend to study marijuana."

This summer President Obama announced a change in rules that would streamline medical marijuana research. The National Institute on Drug Abuse oversees the cultivation of marijuana for medical research, while the Drug Enforcement Administration licenses and monitors study sites.

Meanwhile, the FDA also issued warning letters to companies last February warning them not to promote products as cannabis drugs that can cure, treat or mitigate diseases.

Although the FDA's website says it has never approved a product containing natural marijuana, the agency has approved two older drugs, Valeant's Cesamet and AbbVie's Marinol, which are man-made forms of cannabis that can treat severe nausea from cancer chemotherapy.

In October, London-based GW Pharmaceuticals announced that international clinical trials of its investigational drug Sativex, made from natural cannabis to treat pain in cancer patients, failed to show that the drug benefited patients more than a placebo. The company is now working with the FDA "on a potential path forward" after a sub-analysis of the same data found that patients in the U.S., who the company said were less frail than their European counterparts, did benefit.

GW Pharmaceuticals is also testing the cannabinoid drug Epidiolex for treatment of seizures in children with a rare form of epilepsy, and an intravenous cannabis drug for treatment of brain injury from oxygen deprivation in neonatal patients, among other ­projects.

Seattle attorney Robert McVay, who advises marijuana businesses in the states where it's legal, noted that many small- and medium-sized distributors are not clamoring to see marijuana turned into an FDA-approved pill or inhaler.

"Everyone wants to see the research," he said. "But what I'd say the industry doesn't want to see is marijuana taken over by the pharmaceutical industry, especially in states where it legal for all purposes."

McVay noted that several large pharmaceutical and tobacco companies have "irons in the fire" internally on development of marijuana drugs, but they appear to be waiting for more legal clarity before moving forward.

Joe Carlson • 612-673-4779

Twitter: @_JoeCarlson