Addiction, The App: The Genetic Underpinnings of Opioid, Cannabis, and Alcoholic Abuse

Understanding heritability, genome wide associated studies, and the genetics of addiction disorders

President Biden’s State of the Union speech got us thinking.

What are we really doing for opioid addicts?

We here at Traitwell decided to answer President Biden’s call by releasing our own addiction app, which relies on the latest genomic science. As always, it’s totally free to the public. We encourage you to spread the word and to write to us if you have any thoughts or suggestions. 

Addiction Profiler: The App and the Opportunity

It is free and informative. Just upload your DNA, and our system will let you know if you have any genetic risk factors for addiction to alcohol, opioids, and cannabis.

http://traitwell.com/addiction.

Knowing that you carry the genes for addiction might make it less likely you’ll ever try substances that could irreparably harm you in the first place.

And while it seems president after president promises a solution to a highly intractable problem, it may be better never to start.

It’s morally wrong to serve alcohol to someone who is predisposed to alcohol addiction, to offer joint to someone who is likely to develop a cannabis use disorder, or push oxycodone on someone at high risk for opioid dependence . There is, of course, the responsibility to know our own addictive tendencies. In the past we talked about family history. Some families seem uniquely affected.

Indeed, that’s the takeaway from Empire of Pain: The Secret History of the Sackler by Patrick Radden Keefe. The book details how the morally bankrupt Sackler family took advantage of millions of Americans through the wanton over-prescription of painkillers. Hundreds of thousands of Americans died. In addition to the loss of life, there’s also the loss of public trust, as the Food and Drug Administration was corruptly complicit. That’s something we are currently experiencing in the discussions around the coronavirus.

I highly recommend the book though it neglects an interesting aspect of opioid abuse and addiction: There’s not much discussion of genetics.

If anyone can get an opioid addiction, perhaps we would take it more seriously. But that isn’t really true, as a chart of drug deaths makes quite clear.

The current opioid crisis here in the U.S. involves many genetic cousins of the Scottish.

Patrick Radden Keefe left unexplored some interesting questions:

• Why do some people get addicted to prescription pain killers and some do not?

• What can we do to identify people’s risk factors for opioid addiction before they take them?

The book is one of the best discussions of addiction I’ve ever read but we ought not keep it abstract.

I recently attended a funeral of a 28-year-old friend of mine who suffered from chronic pain before she took her own life. We can use genetics to help those who suffer. 

(Psychological v. physiological dependence and addiction is not mutually exclusive and something that might be the next step for Traitwell. Many times addicts are addicts because of a coexisting disorder like depression. The heritability of depression and addiction could make for a more perfect cocktail, and may be something for later exploration in the app.)

Opioids are by no means the only addiction plaguing America.

Talking about the genetic underpinnings of addiction might remove some of the stigma to begin treatment. If some people are more susceptible to substance addiction because of their genetics, perhaps that knowledge might provide the opportunity to step outside their addiction and face it without guilt.

This might prove particularly difficult among communities where a culture of honor predominates but we have to try.

The Heritability of Traits

Almost every human trait is heritable. Sometime a trait is more than 50% heritable, sometimes a bit less. On average 50% is a good guess for any trait. To understand what this means, we have to think very carefully about heritability which captures the extent to which differences in traits are due to differences in genes.

We already know that human behavior is influenced by a combination of forces, including genetic and non-genetic influences. The genetic influences are very similar to the influences on human height, appearance, disease, and the like.

Non-genetic influences are called the “environment”, but this is a misleading term. It does include the things that most people would ordinarily think of, such as neighborhoods, schools, socio-economic status, parenting, and the like. These comprise the “shared environment” because people who grow up in the same family should experience similar levels of it. But because it is a catch-all category, the broader term “environment” also includes what is known as “non-shared environment,” which include  influences that are unique to individuals, but are not genetic.

Indeed, “non-shared environment” is poorly understood right now. We can quantify its effect, but we have not been able to disentangle their constituents yet. They include all random influences specific to the individual. Measurement error is one such random influence, but developmental error is possibly more important. Genes try to accomplish certain ends but they don’t reliably do so. Alas, things don't always go the right way when building proteins and accomplishing behavioral responses. There are random disturbances degrading fidelity, and they accumulate.

To understand genes and environment, we would have to manipulate them experimentally. By altering one variable and keeping another constant we can have good control, disentangling the variables in predictable ways. This is feasible for certain animal experiments and plants, but obviously infeasible and unethical for human subjects.

Lack of experimental data means that we have to look to observational data to understand how genes and environment combine to affect behavior. In the real world, people differ naturally in their genetic endowments. Some people are twins, others are siblings, aunts, uncles, or grandparents. People also differ in their environmental influences. They may be adopted into families that they have no genetic relationship with. There are even a small number of cases in which identical twins have been given up for adoption and raised separately from a very young age. These people share all their genes but (within limits) none of their environment.

Until recently, heritability estimates were solely based on large family studies exploiting the above facts, combined into very big meta-analyses. These are still the best broad source of information about the relationship between “nature” and “nurture.” When all the data from studies up till 2012 is combined, there are 12 million subjects we can use to derive heritability estimates. This allows us to state with a high degree of confidence that almost every human trait of any importance is heritable to some extent, usually moderate. But first notice that the data has limitations. We are only able to study the influence of natural variation that we happen to encounter in our data sets. People have different environments, but not completely different. There is a limited range of variation. Conceivably that range could be much wider in a different world, as it may have been in the past and might be in the future.

Heritability estimates are intrinsically linked to the specific populations they were derived from, but not too much should be made of this. Twin studies have been conducted all over the world, and they give similar results. Heritability estimates capture our current world well and are our best estimates of the relative influences of nature (genes) and nurture (anything non-genetic including random errors).

If a trait is heritable it must to some extent be under genetic control, but the converse is not necessarily true. There are traits that are clearly under genetic control, like two-legged-ness among humans, that are hardly heritable simply because there is almost no natural variation in genes for legged-ness. Heritability is not the same thing as “under genetic control” although but it does imply it, and it does measure the extent of that control in the world we observe.

Heritability and Genes

Genome-Wide Association Studies (GWAS) now allow us to go beyond heritability estimates to try and find the specific genes involved in traits like substance use disorders. Commonly available DNA chip arrays have been used to genotype large numbers of people. You may have done an 23andMe or Ancestry spit kit, which rely on such chip arrays for results. These chips contain a sub-sample of the genome thought to be important, and do not include those elements which rarely vary. This means that they cannot track down all the heritability found using twin and family studies. We’ll need whole genome sequencing to move that effort forward, and we’ll discuss that at another time.

Nevertheless, by forming very large datasets with tens, hundreds or even millions of subjects, along with measures of the traits being studied, a substantial number of specific genes can be found via GWAS. More properly, Single Nucleotide Polymorphisms (SNPS) can be identified. These are locations within the genome which vary, and strictly speaking need not be within genes, but that is a technical nuance we will not go further into.

Think of it all as genetic information that matters. Traitwell’s Addiction Profiler is an app based on GWAS studies from the Psychiatric Genetics Consortium.

Genetics and Addiction

Addiction Profiler checks your DNA to see if you carry genetic information associated with addiction to alcohol, opioids, and cannabis. Substance addiction is a general term for two kinds of behaviors or traits: use disorders and dependency. These overlap to some extent but have different foci.

“Use disorder” refers to what is commonly called “being hooked.” Regular use of a drug such as nicotine or morphine produces noticeable problems and life outcomes. The user may go to special lengths to get the substance, and may get in some trouble as a result. However the user is not necessarily dependent on the substance, just very fond of it.

“Dependency” refers to use of a substance to the point where it becomes pathological and interferes more seriously with the life of the sufferer. Tolerance develops and ever-increasing doses may be needed. Use is now habitual. This may involve serious criminal behavior, over and above the crime of obtaining the substance if it is illegal. Alcohol dependency may lead to domestic violence, for example.

In practice “use” and “dependency” may become blurred together. Sometimes people will not be classified as “dependent” unless their use of a substance is pathological to some extent and so regular that it may be thought a habit. But this blurring should not present too much difficulty in practice, as there is a lot of distance between the average user and the average dependent.[JA4] 

Indeed, substance addictions are substantially heritable like all other traits, at an average of over 70%. There is little to no influence of shared environment on addiction, but there is substantial non-shared environmental influence. The best-known of these addictions is probably nicotine, which is itself about 50% heritable. It is of course extremely widespread, with serious degradation to health for many people. Similarly, caffeine is used by 80% of the world's population, but unlike nicotine, it rarely leads to problems, though some may have withdrawal symptoms. Mild use of caffeine is about 40% heritable, but heavy use (at least 5 brewed cups a day) is 70% heritable.

We consider four important kinds of substance addiction: Cannabis Use Disorder, Alcohol Use Disorder, Alcohol Dependency and Opioid Dependency. These are all studied by the Psychiatric Genetics Consortium.

Alcohol is a drug, even though it is widely available in most societies and usually not thought of as a (serious) drug by most people. Around 92% of women and 97% of men report trying it at least once. A majority of people may take it or leave it, but a substantial sub-section of society becomes strongly addicted to it and engages in problematic use. Alcohol disorders are at least 50% heritable.

Cannabis, also known as marijuana and by a legion of euphemisms in a myriad of alternative forms, is well-known as a “soft” drug. It is thought to be non-addictive physically but in practice it is psychologically addictive for many, leading to disorders. It is known to be linked to psychosis. Cannabis disorders are at least 70% heritable.

Opioids include opium, OxyContin, Fentanyl, and other derived drugs, and are strongly addictive for a sub-section of the population. Opioid disorders are between 40% and 60% heritable.

Addiction Profiler is free and informative, and relies on the latest genomics research on alcohol, cannabis, and opioid addiction. Check out the referenced studies and further reading suggestions to learn more. And be sure to upload your DNA at http://traitwell.com/addiction and let us know what you think.

---

Studies Used

1. Opioid Dependence

Leveraging genome-wide data to investigate differences between opioid use vs. opioid dependence in 41,176 individuals from the Psychiatric Genomics Consortium.

4503 dependency cases, 4173 opioid-exposed controls, and 32,500 opioid-unexposed controls, of European and African ancestry.

https://pubmed.ncbi.nlm.nih.gov/32099098/

Polimanti R, Walters RK, Johnson EC, et al. Journal of Molecular Psychiatry. 2020 Aug;25(8):1673-1687. doi: 10.1038/s41380-020-0677-9.

(See also Wang SC, Chen YC, Lee CH, Cheng CM. Opioid Addiction, Genetic Susceptibility, and Medical Treatments: A Review. International Journal of Molecular Science. 2019;20(17):4294. Published 2019 Sep 2. doi:10.3390/ijms20174294 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6747085/)

2. Alcohol Use

Genome-Wide Association Study Meta-Analysis of the Alcohol Use Disorders Identification Test (AUDIT) in Two Population-Based Cohorts.

https://pubmed.ncbi.nlm.nih.gov/30336701/

Participants from UK Biobank [N=121,604] and 23andMe [N=20,328]). European ancestry. Using the Alcohol Use Disorders Identification Test (AUDIT). AUDIT-P score was significantly positively genetically correlated with schizophrenia (rg=0.22), major depressive disorder (rg=0.26), and attention deficit hyperactivity disorder (rg=0.23), whereas AUDIT-C score was significantly negatively genetically correlated with major depressive disorder (rg=-0.24) and ADHD (rg=-0.10).

Sanchez-Roige S, Palmer AA, Fontanillas P, Elson SL; 23andMe Research Team, the Substance Use Disorder Working Group of the Psychiatric Genomics Consortium, Adams MJ, Howard DM, Edenberg HJ, Davies G, Crist RC, Deary IJ, McIntosh AM, Clarke TK. American Journal of Psychiatry. 2019 Feb 1;176(2):107-118. doi: 10.1176/appi.ajp.2018.18040369.

3. Alcohol Dependence

Transancestral GWAS of alcohol dependence reveals common genetic underpinnings with psychiatric disorders.

https://pubmed.ncbi.nlm.nih.gov/30482948/

14,904 individuals with alcohol dependence and 37,944 controls from 28 case-control and family-based studies were meta-analyzed, stratified by genetic ancestry (European, n = 46,568; African, n = 6,280).

Walters RK, Polimanti R, Johnson EC, et al. Nature Neuroscience. 2018 Dec;21(12):1656-1669. doi: 10.1038/s41593-018-0275-1.

4. Cannabis Use

A large-scale genome-wide association study meta-analysis of cannabis use disorder.

https://pubmed.ncbi.nlm.nih.gov/33096046/

20 916 case samples, 363 116 control samples. Psychiatric Genomics Consortium Substance Use Disorders working group, iPSYCH, and deCODE.

Johnson EC, Demontis D, Thorgeirsson TE, et al. Lancet Psychiatry. 2020 Dec;7(12):1032-1045. doi: 10.1016/S2215-0366(20)30339-4.

Further Reading

Kenneth S. Kendler and Carol A. Prescott.
Genes, Environment, and Psychopathology: Understanding the Causes of Psychiatric and Substance Use Disorders(Guilford, 2006).

Valerie S. Knopik, Jenae M. Neiderhiser, John C. DeFries, Robert Plomin
Behavioral Genetics (Worth, 2017).