Turning questions to answers in drug testing

Karen Titus

July 2023—As she surveys the opioid epidemic in North America, Christine Snozek, PhD, D(ABCC), could be tempted to think that a ripped-from-the-headlines reality has landed in clinical laboratories as well as on TV crime dramas. With the number of opioid-related deaths increasing in recent years, particularly since the start of the pandemic, drug testing demands have increased for labs as well, says Dr. Snozek, codirector of clinical chemistry and support services and director of point of care and central processing at Mayo Clinic in Arizona.

If only she could turn to the entertainment industry for a technology-based solution. “I wish we had CSI lab capabilities,” she says, referring to the long-running police procedural. “You could run a sample and find all the drugs known to man on one test. If they could go ahead and release that technology, that would be wonderful,” jokes Dr. Snozek, a member of the CAP Toxicology Committee.

Given that laboratories are unlikely to take a meeting with network executives, Dr. Snozek and others in the field will have to look elsewhere. Fortunately, some nonfictional solutions exist.

Dr. Snozek and two coauthors—Danyel Tacker, PhD, D(ABCC), and Sarah Delaney, PhD, D(ABCC)—explored the laboratory’s role in drug testing in a review published recently in Critical Reviews in Clinical Laboratory Sciences (Delaney SR, et al. Crit Rev Clin Lab Sci. 2022;59[5]:309–331). While keeping up with changing-at-warp-speed testing needs might seem insurmountable to any but the largest reference laboratories, Drs. Snozek, Tacker, and Delaney suggest that any—make that every—lab can be a useful resource for clinical colleagues and patients, not only to help stem the opioid crisis but to manage other drug testing as well.

Dr. Snozek draws on her background at two Mayo Clinic sites when she assesses how labs can help. At her former location in Minnesota, the lab had access to the bonanza of resources that fill any large reference laboratory. At her current position in Phoenix, “I wouldn’t call us a small or resource-constrained operation,” she says, “but obviously it’s very different from Rochester’s capabilities.”

For labs that don’t have access to mass analytic methods, such as liquid chromatography with tandem mass spectrometry and high-resolution MS, or dedicated staff to develop and run drug testing assays, there’s still plenty they can do, says Dr. Snozek. Her current Mayo site, for example, is essentially a tertiary center in the Phoenix area. She doesn’t offer mass spectrometry for drug testing, nor does she need to for her patient population. But she can and does offer other drug testing options.

And that’s the point. “This is my soapbox: Labs can be doing more and should be doing more,” Dr. Snozek says.

That runs counter to what many might see as a reasonable response to the relentless opioid crisis that’s lasted longer than the 16-year CSI run: deciding that it’s impossible to keep up. Dr. Snozek says she’s seen such responses, particularly among laboratory staff who lack specific clinical chemistry or toxicology training. “It’s just so overwhelming, they’re not sure they can do anything about it.”

Dr. Snozek offers a brisk tour of basic steps laboratories can take.

Any laboratory, for starters, can look at its test menu and see if it lines up with, as she puts it, “what’s going on in the world.” Start by talking to emergency department colleagues. If the lab doesn’t offer a screen for fentanyl, for example, would their ED colleagues find it useful?

In some cases, it may be reasonable to subtract rather than add. Any lab can ask its local practice if something is no longer useful. Case in point: propoxyphene, which is usually offered in a kit format, such as a point-of-care cup. “Propoxyphene is pretty useless,” Dr. Snozek says, adding that the test wasn’t widely used when it was first made available and is even less so now. “It’s almost unheard-of to see someone with a propoxyphene positivity. There’s no point in testing for it.”

She returns often to this common refrain of any laboratory—as in any laboratory can take such a step, even those that lack staff with extensive toxicology training. “It just involves looking at your lab information system, seeing who’s ordering drug tests, then contacting them” to ask if they’d like to see something different, Dr. Snozek says.

Such steps, though useful, hardly close the case. Dr. Snozek is quick to acknowledge that drug testing can feel overwhelming because it is, in many ways, overwhelming. Keeping up with current trends can feel like trying to outswim a rip tide.

“If you’re talking about novel drugs that come and go, it is difficult to get that information in anything even remotely resembling real time,” Dr. Snozek says. A truly new drug that hasn’t been well characterized may appear as a pocket of overdoses that show up in a metro area and then peter out. The drug will remain unrecognized for weeks or months until someone has the ability to analyze it and confirm it.

Dr. Snozek commiserates with labs that might feel steamrolled by such scenarios. At that point, she says, “It probably is appropriate for most labs to throw up their hands and say, ‘I don’t have the resources to keep up.’ And honestly, if we can’t test for it, we aren’t the best resource,” she says. “As far as bringing up a test for these novel agents that come and go quickly, it’s not feasible for most laboratories to do.”

Coauthor Dr. Tacker also laments the speed with which changes occur in the field. Simply put, “We can’t be as nimble as the changes occurring at the street level in drug use,” she says. “We often find things incidentally and in tragic conditions and circumstances. It’s usually something terrible that clues us in that we need to change something in the lab,” says Dr. Tacker, clinical professor of pathology, medical director of clinical chemistry and mass spectrometry laboratories, and CLIA medical director of blood gas laboratories, J.W. Ruby Memorial Hospital, West Virginia University Health System, Morgantown.

Nevertheless, other clues might appear earlier. Regional trends relate to which drugs are prominent in local circulation, she says; likewise, the CDC’s Morbidity and Mortality Weekly Report can indicate outbreaks or clusters of toxicities, offering useful information about symptoms and which agents are responsible. These reports “can help labs triangulate how their services may align, or may even expose gaps in their services and prompt them to try to fill them,” says Dr. Tacker, who is also CLIA medical director at Fairmont (WVa.)Medical Center.

Dr. Tacker has done this herself, noting how xylazine has made its way into street blends of opioids and other drugs. “It’s incredibly toxic, and there’s no standard laboratory screening reagent for it. There aren’t any point-of-care tests that are going to test this for us.” Basically, she says, providers are figuring out the drug based on its presentation in the clinic, then calling the lab to ask for confirmation.

She first heard about xylazine when she spoke with coauthor Dr. Delaney, clinical biochemist at Unity Health Toronto, as they were working on the review article and discussing what examples they’d use. Dr. Delaney mentioned the drug’s impact in the Toronto area. Says Dr. Tacker: “I hadn’t even heard of it yet, but now it’s made its way to West Virginia.” Observing the movement of these drugs is interesting, “but also frightening,” she says. “How do we serve clinicians in a way that’s productive? How do we answer questions for them?”

Longer trends have been easier to track, from heroin to semisynthetic and fully synthetic opioids. But within those decades-long climatic shifts, the specifics are chaotic—one year’s drought is next year’s deluge.

West Virginia has long been an epicenter of the opioid crisis, Dr. Tacker notes, spurring harm reduction efforts. “The most common questions we get on a day-to-day basis are, ‘Is this person using? And what can we do to help with harm reduction?’” she says.

The story of her lab’s response makes a compelling tale of its own.

The lab placed point-of-care drug cups in behavioral medicine clinics, which allows providers a first pass at screening and enables them to have an immediate talk with the patient if something has changed.

But first the lab had to convince a key player: itself. “We originally pushed back against point-of-care–based drug testing,” Dr. Tacker concedes. The lab had doubts about the technology and feared its use would lead to more problems. She was happy to be proved wrong. “It’s been a boon for the clinic—a huge benefit.” Having the results immediately redirects conversations with patients quickly, who tend to respond well, she says. “Usually the patient will come clean because they know this is their chance to get help. It turns things around for them.”

Based on her experience, Dr. Tacker says she encourages labs to collaborate with behavioral medicine specialists in the clinic setting. “We’ve built a strong relationship with them just on creating the point-of-care program.”

Beyond that, the central lab has expanded its urine drug panels, which are offered hospitalwide and in clinics. It’s an 11-component drug panel—covering fentanyl, methadone, buprenorphine, oxycodone, and opiates—with creatinine as a check for integrity. They’ve also created the ability to automatically reflex positive results in the panel to confirmatory testing, primarily on liquid chromatography–mass spectrometry. That’s critical, Dr. Tacker says. “They may be taking an over-the-counter medication that is similar enough to cause a false-positive.”

Or, if a person has an unexpected negative result, Dr. Tacker continues, that could spur conversations about noncompliance. “So the laboratory service has integrated itself without even trying into these care scenarios,” she says. “We’re just trying to support decision-making for our clinicians. There’s so many of them in so many different practices.”

How did Dr. Tacker and her colleagues develop their panel? “I wish I could say it was a deliberate effort, done in a year,” Dr. Tacker says. Instead, it evolved over time, mostly as clinicians asked about testing for new substances. When she joined WVU at the end of 2010, she recalls, colleagues were asking about tests for so-called bath salts, which are synthetic cathinones. “I had to look up ‘bath salts.’ I thought they were literally talking about what you’d get at Bed, Bath & Beyond,” she recalls with a laugh. “I had no idea.”

The lab was also asked about testing for methadone, which eventually made its way onto the menu.

But the drug cups were the catalyst, she reiterates. “Behavioral medicine approached us and said, ‘Look, we have to have something else.’” That launched serious discussions about POC testing. “It caused me to start taking a hard look at what was in the cups, so we could choose the right one. And that cross-informed me [about] what we should be doing with our panels in the main lab.” If a test was important to behavioral medicine specialists, she realized, it would also be important to obstetricians, family medicine physicians, and others who might be looking to ensure compliance for, say, barbiturate prescriptions. By about 2014, the elements started to coalesce, and the lab selected a POC cup with cutoffs that were compatible with the central lab.

Several years later, an upgrade in the lab enabled it to perform mass spectrometry toxicology testing, and Dr. Tacker began to research an opioid panel. The goal was to address the so-called heavy hitters, she says, with an emphasis on high sensitivity.

Cue mea culpa No. 2.

The lab met its goal, Dr. Tacker says, launching a super-sensitive panel. It was so sensitive, in fact, that colleagues from behavioral medicine began calling to let the lab know the panel was detecting morphine in patients who’d undergone minor surgeries a week earlier—i.e. clearly not drug use candidates.

These days, Dr. Tacker makes a point of doing a careful annual review of the panel. She pulls every report submitted in the previous year for every orderable configuration of the confirmatory testing. “I look at the cutoff, how many results came in near the cutoff, whether they were clinically relevant, where they came from—was it an inpatient, was it the addiction clinic setting?” In short: How did the test results affect care and decision-making?

She then shares her summaries with behavioral medicine. “We think we’ve settled on a good set of cutoffs,” thanks to these discussions, she says. “We’ve started to inform our process based on what’s going on in the clinics.” For example, the lab now offers tramadol confirmatories for patients receiving pain control for neuropathies and other conditions. And it’s no longer flagging patients who’ve undergone minor surgeries.

How difficult is it to capture this information? “I’m an Excel spreadsheet girl,” Dr. Tacker answers, adding, “I don’t have any major training in data analytics.” Her approach is simple: “I go into Epic, I run a bench report for the orderable tests that we have built, and it extracts a report for me that I pull into Excel. And then I start counting.” She uses a few simple equations to ask how many results were below cutoff, at or around cutoff, and high above cutoff. It does take time, she acknowledges. In her case, it requires about six weeks, “putting in an hour here, an hour there.”

The close relationship she’s forged with clinical colleagues has paid off in another way, she says. When the lab wanted to bring in-house many of the confirmatory tests it was sending out, it had neither the staff nor the equipment to meet the high demand. When they approached administration for funding, behavioral medicine joined the lab in making the case.

“It helps to have these partnerships,” Dr. Tacker says. “Then it’s not just you asking for cash to build a program.”

Her clinical colleagues have learned to listen to the lab. At one point Dr. Tacker decided it would be useful to add tramadol screening.

“We sold three in the first year,” she says. “We thought, Maybe they just don’t know we offer this test. So we reminded them.” Still no buyers. Lo and behold, providers knew the lab also offered a definitive test by mass spec; in fact, the lab had recommended this test to its colleagues in compliance situations, to avoid the double charge. “Well, they followed that instruction to the T. So we turned off the screening test.”

As Dr. Tacker’s tales tell, creating more responsive drug testing requires time and a few misses as well as hits. It’s doable, especially for those who persevere. Her fear—one her coauthors on the Critical Reviews publication share—is that too few labs are wading in.

Their article drew information, in part, from CAP Surveys data. As Dr. Snozek notes, in late 2021/early 2022, roughly only 50 percent of laboratories that were signed up for the opiate screen also had a specific result for oxycodone, and only about an eighth had a specific result for fentanyl.

This doesn’t necessarily mean labs aren’t running tests for these drugs, she notes; they might be using a different proficiency testing platform for them. Nonetheless, she adds that the majority of laboratories that participate in a broad Survey of this type tend to be those that don’t typically have multiple, dedicated mass spectrometry-type platforms for this testing. “So it seemed likely to be a good snapshot of routine, regular labs.”

Thus the low numbers are cause for concern, given that use of oxycodone and fentanyl is no longer a new phenomenon. “It’s one thing to say routine labs don’t have the bandwidth to update test menus in response to novel psychoactive agents,” says Dr. Snozek. “But when fentanyl has been the main killer for eight-plus years, that’s a significant lag that’s clinically relevant.”

Speaking from her own experience, Dr. Delaney, who is also an assistant professor, Department of Laboratory Medicine and Pathobiology, University of Toronto, offers insight into possible reasons for the slow uptake. There are, she notes, only so many kits and tests available, and many are third-party reagents. Her lab is currently transitioning to the latest model of its analyzers. “But we use a third-party reagent, and there are no parameters that have been worked up for this brand-new instrument, so we’re scrambling.”

Vendors will also have to find the right incentives (the most obvious being money) to develop tests, but vendors, like labs, can also be hamstrung by shifting trends. Dr. Delaney reports Toronto has seen a tremendous, sustained rise in fluorofentanyls this spring. “It came out of nowhere,” she says. But that doesn’t mean it will continue, or that it will make sense for manufacturers to develop a test. On the other hand, the longstanding presence of fentanyl should induce test development for this substance. “There’s a real demand to make that available on confirmatory or mass-spectrometry–based testing, which might be a little easier to optimize and develop.”

The rapid fluctuations in the unregulated drug supply put another burden on laboratories: They have to work even harder than usual to justify the need for new tests, Dr. Delaney says. “You have to build a case for putting the time and energy and money into building a new method for xylazine, for instance. And even if something’s in the drug supply, you have to weigh the pros and cons of implementing testing.”

It’s a conversation she has regularly with clinical colleagues. “Sometimes clinicians will say, ‘We need this,’ but not understand just how dynamic the drug supply is.” The only way to possibly keep up, she says, is to have strong relationships with the providers who use the urine drug screen and to meet with them often. “It’s a challenging dance.”

Dr. Delaney and colleagues have looked at trending positivity rates for various substances over several years, hoping to use the data to decide what to remove from the test menu. It sounds like a good idea, but in practice it may not provide clear-cut answers. Depending on the substance, “you could say, ‘We haven’t seen it in six months—we’re going to remove it from our method.’ But then it will come up and have sustained presentation for another six months.” It’s nearly impossible to predict whether a drug will return, Dr. Delaney says.

With no bright and shiny CSI technology forthcoming, laboratories will need to figure out for themselves how to update their test menus for each methodology.