Can a Genetic Profile Predict Susceptibility to Addiction?

By Dorri Olds 08/23/16

The genetics that predispose you to one form of addiction—say, heroin—are different than the genetics that predispose you to other forms of addiction, like cocaine or methamphetamine.

Image: 
The Proove Test
It combines a genetic profile with other objective risk factors.

John Jones (pseudonym) had been off drugs for 17 years. Prior to getting clean, he had been a heroin user who’d contracted AIDS from sharing needles. John is on Medicare and went to his clinic where he told a doctor he was suffering from back pain. John laughed when he told me the doctor prescribed Percocet. Slurring his words, he said, “Whenever I play the AIDS card, doctors feel sorry for me and prescribe anything I want.” The next time I saw John, he was wearing long sleeves on a swelteringly hot day this July. “Why are you so overdressed?” I asked, as I dabbed a tissue to my forehead. John looked down at the sidewalk, fidgeted, tugged at his sleeve, and then admitted he’d returned to shooting heroin. “I never should’ve taken that Percocet. I should’ve known better.”

Yes, and the doctor should’ve known better, too.

John never made it a secret from any doctor that his AIDS came from sharing needles he’d used to shoot heroin. That should have been a red hot alarm for any doctor who considered prescribing opioid pain medicine. Unfortunately, John’s story is far too common.

I spoke with Brian Meshkin for The Fix. Meshkin is the president, CEO, and founder of Proove Biosciences, a company that conducts research and develops new laboratory tests for doctors. The company is committed to helping doctors treat patients based on personalized genetic and behavioral information. The idea is to know which patients are at high risk for drug addiction.

“Opioid addiction has become a huge issue,” said Meshkin. “By being able to give risk information to prescribers of opioid pain medication, they can better decide whether they can safely prescribe opioids or whether they need to try something else.”

Proove, explained Meshkin, provides a genetic profile of the brain’s reward system in order to show whether a patient is genetically at a higher risk for misuse of prescription narcotic pain meds. 

But how is Proove’s Opioid Risk Test any different from others already out there? “Based on the definition of addiction from the American Society of Addiction Medicine and the National Institute on Drug Abuse, about 50% of substance abuse is due to genetic factors,” Meshkin said. “But, there is no one addiction gene that makes you an addict. By being able to combine genetics with other factors, we’re able to make the mathematical equation—the algorithm—very predictive.”

When I asked, “Hasn’t that test been on the market for several years?” Meshkin said, “As part of the studies, we analyzed some survey instruments given as the current standard of care. Things like the ORT [Opioid Risk Tool] and the SOAPP [Screener and Opioid Assessment for Patients with Pain]. But the problem with the current standard of care, when you don’t include genetic information, is asking a patient to fill out a survey and to tell the truth.”

We joked about the cliché: “How do you know when an addict is lying? His lips are moving.”

“The point is that when addicts answer questions, they are likely to lie,” Meshkin said. “It’s a dilemma, which is why studies find that the answers are different depending on whether patients fill them out versus when a clinician fills them out. That’s why they’re not very accurate. Our test combines genetic factors with other aspects to develop a composite risk score of whether someone is low, moderate, or high risk for misusing opioids.”

Which aspects are most predictive? “A diagnosis of depression is a huge issue. For this, we didn’t have to rely upon whether the person was being honest. We could pull that out of their medical history. Depression and some other co-occurring mental health disorders—anxiety, medical insomnia, bipolar disorder, schizophrenia, PTSD—all of these substantially raise risk.”

Another factor is history—exposure to alcohol, illegal drugs, or prescription opioids in the past and if they ran into issues with any of those.

“Age and gender also play a role,” said Meshkin. “There are some other variables that go into the equation that we are able to collect. By combining those pieces of information and how they’re scored, along with the genetics and a combination of the two, that’s what then helps us define the risk. So, all that information has to be collected when a doctor orders the test.”

At that point I was a tad confused and asked, “How many different tests have been done and how many people are tested to make sure that what you’re getting from your test scores are accurate assessments?”

Meshkin said, “Over the past three years, physicians have ordered more than 60,000 opioid risk tests. With regards to clinical study work, you need to show two different types of data with the diagnostic test: clinical validity data and clinical utility data.

“The validity data is about accuracy. If we say, for example, a patient is at high risk, a validity assessment is about how accurate that is. The utility assessment refers to following up when a doctor had the information and used it—the question is, did the patient get better? Were they able to avoid instances of abuse?”

Proove looked at three different studies that they themselves performed. The first one was with 300 patients. Eighty were opioid dependent and opioid abusers, compared with three control groups.

“This was a retrospective study,” said Meshkin. “We looked at which genetic factors and which of what we call phenotypic factors, were predictive of the outcomes.”

Proove then built an algorithm to be able to divide the subjects into classes of low, moderate, and high risk. The complexity of his answers were giving me a headache so I asked him to break it down and explain things to me as if I were a third grader.

He laughed. “Okay, we did a study, like studying history. For example, say we looked at a bunch of people who were in chronic pain, over a two-year period. Some had been prescribed opioids by their doctor and became dependent. Then we had what you’d call control groups, or comparison groups. These people also received opioids but did not become dependent. Then it’s a contrast: black and white.”

That’s when I got the “A-ha!” The trick for Proove was to then figure out why one group became dependent while the other group did not.

“We looked at the genetics between the two," Meshkin continued. “Then, we also looked at those survey instruments between the two to find out which genes were different and which answers to questions were different. Then we built a mathematical equation that said, ‘Okay, these genetic factors combined with these specific questions showed the biggest contrast.'"

Are there different risks for addiction to, say, crystal meth versus heroin? “Yes,” said Meshkin. “We launched a test two weeks ago and it stratifies the different types of addiction based on genetics. There are certain people who are more susceptible to becoming addicted to alcohol versus becoming addicted to heroin versus becoming addicted to cocaine or methamphetamine. The genetics that predisposes you to one form of addiction is different than the genetics that predisposes to other forms of addiction.”

He continued, “When someone has a genetic predisposition, once they complete a rehab and a detox program, they will still have their genetics. They’re going to be at a higher risk for relapse, which is why some type of ongoing maintenance program is going to be so important for them. It can’t just be a detox over three months or six months or 30 days, and then they’re set. It means you have an underlying genetic predisposition and there needs to be some ongoing effort there. The test also provides guidance based on the genetic predisposition as if they’re going to use some form of medically-assisted treatment—which type of medication they should use and how to dose it.”

The brain is so complex. Will we ever fully figure out how to treat addiction? “This is one of the keys to what we’re doing,” Meshkin said. “We’re adding computers and software analysis.”

Here’s where I asked for an analogy. “In the old days, when you had a problem with your car, you’d take it to a mechanic who’d lift up the hood, look underneath and try to figure out what was wrong. Then you’d want to double-check by going to another mechanic for an additional opinion. Today, when you take your car into the dealership, it is hooked up to a computer that can diagnose exactly what is wrong with the car. Then the auto mechanic knows exactly what to do to fix it. I think medicine is heading there. The clinician becomes a body mechanic and computers will analyze information and be able to figure out exactly what’s wrong within a human body’s system.”

“We should be able to figure out what’s wrong,” Meshkin continued, “because 99.9% of human bodies are almost identical as a system, genetically. Because there’s a huge measure of similarity between human beings, there’s a standardization—like a car—and by analyzing the data, we’ll figure out what’s wrong.”

It sounds like sci-fi, but according to Meshkin, when we get to the point where computers can process data within a human body as part of the human system, narrowing down the problem and what to fix should be just a click away.

Watch this video for a testimonial to the value of the Proove test when it comes to NSAIDs:

Please read our comment policy. - The Fix
Disqus comments
Dorri-Olds-headshot.jpg

Dorri Olds is an award-winning writer whose work has appeared in many publications including The New York Times, Marie Claire, Woman’s Day and several book anthologies. Find Dorri on Twitter, Facebook, and LinkedIn.

Disqus comments