Follow Us:
Opioid 101:
Contact us:
Thank you! We’ve sent you an email to confirm your address.
An Algorithmic Breakthrough In the Fight Against Illicit Opioid Trafficking
By MICHAEL KELLY|April 13, 2018
Quick Takeaway
  • Americans are overdosing not just on fentanyl, but on dozens of chemical variations of it, constantly being invented in illicit crime labs.
  • These fentanyl "analogs" are hard for crime labs to detect and for prosecutors to bring cases against.
  • A new algorithmic breakthrough will help crime labs speed the process of identifying these analogs.

Behind the deadliest and most lucrative sector of opioid trafficking, a global cat-and-mouse game is playing out each day—among chemists.

The good news is that, last month, our side just made a major breakthrough.

But to understand what happened, we have to take a step back and understand the challenge that has now been overcome.

The synthetic chemical known as fentanyl, which is about 30 to 50 times more potent than heroin, was implicated in nearly half of the opioid related overdose deaths reported in 2016, and about 30% of all drug overdose deaths, which works out to 174 a day. A dose the size of a few grains of salt—2.5 mg—can prove lethal, and the substance has been laced into heroin, cocaine, methamphetamine, and counterfeit pills stamped to look like prescription drugs.

But when we say “fentanyl” we’re using a shorthand. The statistics we’re talking about are actually figures drawn from a category that the US Centers for Disease Control and Prevention calls “synthetic” opioids, which encompasses more than just fentanyl proper. It includes all opioids that can be made in a lab, without having to be derived from compounds extracted from the opium poppy. That category includes not just fentanyl but also dozens of fentanyl “analogs”—chemicals that are slight variations on its potentially lethal theme.

Sniffer dogs in the age of fentanyl

We’re talking about drugs like acrylfentanyl, isobutyryl fentanyl, meta-fluorofentanyl, methoxyacetyl fentanyl, ocfentanil, or, even more exotically, MPF 47,700 and U-4TDP. These are illicitly manufactured and trafficked, coming mainly from labs in China. (Technically the “synthetics” category also includes some prescription drugs, like tramadol and lawful fentanyl products, like Duragesic, but the CDC thinks that, for the past several years, such drugs have accounted for a trivial percentage of the synthetic opioid overdoses.)

Last year the United Nations Office on Drugs and Crime published a list of 18 new fentanyl analogues for inclusion in its Table I category of prohibited drugs. But more black market variants emerge every month.

These newly created variants present special problems to both crime labs and prosecutors. For the labs, they’re hard to detect and identify.

And until the lab has done so, it’s hard for the Drug Enforcement Agency or Congress to formally prohibit it. And until that’s happened, prosecutors can’t convict people for trafficking it.

At the Drug Enforcement Administration Special Testing and Research Laboratory in Sterling, Va., about 50 chemists work on analyzing various samples sent from crime labs around the country. Eight of those chemists work exclusively on finding new psychoactive substances, or NPS, including novel versions of fentanyl.

“As new drugs emerge, the labs have to keep up with reference standards,” says NPS supervising chemist Jill Head. “The labs are really much better at it than they were in 2010, when NPS first emerged, but we don’t see [the phenomenon] stopping or decreasing.”

The lab keeps a library of mass spectrometry samples of illegal drugs. As of the end of March, more than 40 fentanyl siblings were listed — four times the number from the previous year — and the list, Head says, just keeps growing.

Detecting new fentanyl analogs is tricky work. The mass spectrometer bombards a sample with electrons, breaking the compounds into fragments and measuring the fragments. It then spits out a line chart characterized by spikes that are, in effect, the compound’s signature.

The charts can then be compared to the reference library for identification, but if there’s no corresponding signature yet in the library, identification becomes more difficult.

“We also have capabilities and expertise that allow us to identify a substance without knowing the reference material,” she said. “Through structural elucidation—measuring, for instance, the number of carbons and hydrogens and how they are connected—we can effect an identification.

“When we know what it is,” she continues, “we publish it online,” so other crime labs around the nation can use it.

That brings us to the breakthrough. Last month a team of chemists at the National Institute of Standards and Technology unveiled a new digital tool they’d come up with: an algorithm that can analyze mass spectrometer results for all fentanyl compounds.

It finds not just known analogs, but those for which there is no known reference, according to Arun Moothy, a member of the team that created it.

“If the possible theoretical limit [in the number of variants] was a million, it could still identify them all,” he asserts.

The algorithm is similar to one previously used by field investigators looking for chemical weapons, according to Stephen Stein, another member of the NIST team.

Alas, using the new algorithm involves more than just pushing a button and getting the answer. Interpreting results takes skill, Stein says. It’s like looking at an X-ray or MRI image. The data is there, but the technician needs experience and insight to know what it means.

Luckily, it seems unlikely that there are a million fentanyl analogues, and, in any case, the upper limit will be determined not by the number of theoretically possible chemical permutations, but by the subset of those that will happen to get drug users high.

As newly identified synthetic drugs pile up, the DEA and lawmakers are racing to keep up. Under the Controlled Substances Act, the DEA has the authority to request an emergency listing on Schedule I—the category of drugs, like heroin, forbidden to be sold for any purpose—but the listing is temporary, lasting two years with a possible one-year extension.

Since 2015, the DEA has issued eight such orders, banning 17 fentanyl analogs. In February, growing frustrated with that approach, the DEA issued a new temporary order that purports to prospectively define and outlaw analogs, even before they’ve been invented. It prohibits molecules exhibiting any one of four chemical modifications to the basic fentanyl template, including things like “replacement of the aniline ring with any aromatic monocycle whether or not further substituted in or on the aromatic monocycle.”

But no one’s sure if that will order will hold up legally.

The Controlled Substance Act itself also has a provision that addresses unspecified chemical variants of fentanyl, but it imposes some additional hurdles of proof and some prosecutors have shied away from using it.

As of the end of March, there were more than a dozen bills in Congress addressing the challenge of fentanyl analogs.

One of the more promising is HR 4922, sponsored by Rep. James Sensenbrenner (R-Wisc.), whose district, which covers Milwaukee and most of its suburbs, has been hard hit by the epidemic. An identical companion bill is being sponsored by Sen. Ron Johnson (R-Wisc.) in the Senate.

These bills list an additional 19 fentanyl analogues, and give the DEA more robust authority to list analogues as they are discovered, Sensenbrenner’s staff said. (The bill’s language appears to, in effect, retroactively authorize what the DEA did in February.)

“It puts the law on a more sound footing, codifies the ability they already have to list new analogues,” a spokesperson for Sensenbrenner says. “Some folks had begun to raise questions about the DEA doing this without specific authority.”

Filed under: Fentanyl/ Legislation/ Research