CSI Lateral: Understanding drug efficacy in forensic toxicology

Even after death, pharmacogenomics can tell us how particular genetic variants cause fatal drug reactions.

Illustration by  Sarah Nagorcka

Illustration by Sarah Nagorcka


Gene Dosage is a monthly column by Janan Arslan that finds out what genome science is uncovering about each individual's unique response to drugs and pharmaceuticals. Janan is a graduate student and pharmacogenomics researcher with a keen interest in personalised medicine.

In previous columns, we’ve discovered how pharmacogenomics is applied in areas such as pain management, mental health and cardiology. Most of my writing has revolved around preventive medicine – a technology that prevents side effects or lack of efficacy associated with drug use. However, an area of pharmacogenomics application that is not as widely discussed is forensic toxicology.

This field of forensic science focuses on understanding drug toxicity. Part of post-mortem analysis, it can help determine the cause and manner of a person’s death. So, rather than having a preventative application, pharmacogenomics is used here to determine if the presence of particular genetic variants caused the patient to have a fatal overdose or reaction to their medications.

Even forensic scientists can benefit from pharmacogenomics.   Nukamari/Flickr  (CC BY-NC 2.0)

Even forensic scientists can benefit from pharmacogenomics. Nukamari/Flickr (CC BY-NC 2.0)

In forensic toxicology, the parent drug to metabolite concentration is used to decide whether drug consumption prior to death was high (more than the recommended dose) or whether the patient suffered as a result of chronic use. Other factors, such as drug-drug interactions and now drug-gene interactions, are also used as part of the investigation. 

In previous columns, we discussed the genes CYP2D6, CYP2C19, and CYP3A4. There are several drugs of toxicological interest that are metabolised by these genes. These include painkillers like codeine, oxycodone and tramadol, which are all primarily metabolised by the CYP2D6 gene. Methadone – a drug used for both pain and opioid addiction management – is primarily metabolised by CYP3A4. Antidepressants such as amitriptyline, imipramine and citalopram are also of interest; these are primarily metabolised by CYP2C19.

A 2007 paper by Moore and associates showed that oxycodone and methadone were among the top suspect drugs in deaths reported to the US Food and Drug Administration between 1998 and 2005. Other drugs, such as fentanyl (analgesic), acetaminophen (analgesic – known as paracetamol in Australia) and clozapine (antipsychotic) were also noted.

According to the Coroner’s Court of Victoria (CCV), prescription medications “are frequently involved in acute drug deaths,” with particular emphasis on oxycodone, fentanyl and methadone. Death rates caused by use of a single drug were lower than the same drugs used concurrently with other medications. For example, oxycodone alone accounted for 12.5% of deaths; however, the use of multiple drugs including oxycodone resulted in a death rate of 87.5%. The Australian Bureau of Statistics, in a 2001 review of drug-related deaths, concluded that “having a mixture of drugs is more likely to result in adverse health effects including death, than use of a single type of drug.” 

The painkiller oxycodone, pictured here, is more dangerous when combined with other drugs like methadone.   Be.Futureproof/Flickr  (CC BY 2.0)

The painkiller oxycodone, pictured here, is more dangerous when combined with other drugs like methadone. Be.Futureproof/Flickr (CC BY 2.0)

Pharmacogenomics has now taught us that genetics can be an underlying cause of drug-drug interactions. For example, the CCV reported that 8.7% of death-related cases saw the consumption of methadone along with oxycodone. As I mentioned earlier, oxycodone is metabolised by CYP2D6, whereas methadone is metabolised by CYP3A4. However, methadone also has the ability to inhibit the function of CYP2D6. So even if a patient did not have a genetic variant for CYP2D6 and were completely normal for that gene, taking methadone would alter their ability to correctly metabolise oxycodone. Queue potential side effects and toxicity.

The contribution of polypharmacy (the consumption of multiple medications) to death rates is unsurprising. A survey conducted by the Slone Epidemiology Center at Boston University between February 1998 and April 2007 showed that, on average, 82% of US adults are taking at least one medication (prescription or nonprescription drug, vitamin/mineral, herbal/natural supplement), and 29% are taking five or more. People aged 65 years or older continue to be the biggest consumers of medications, with 17-19% in this age group taking at least ten medications in a given week. Polypharmacy has also increased since 2000 from 23% to 29%. It is easy to see why adverse drug reactions are one of the leading causes of morbidity and mortality.

Sallee and associates reported on one of the earliest cases to show that genetic variants can lead to fatal drug intoxication.  A nine-year-old boy treated with multiple psychotherapeutic drugs had died due to fluoxetine intoxication. A post-mortem toxicological analysis showed extremely high concentrations of fluoxetine and norfluoxetine in several tissues. To achieve any therapeutic benefit, fluoxetine must be converted to its metabolites, R-norfluoxetine and S-norfluoxetine, facilitated primarily by the CYP2D6 gene. A pharmacogenomics analysis revealed the child had a defective CYP2D6 gene, which made him unable to properly metabolise the medication.

Drug cocktails are now the norm.   Joshua Thorne/Flickr  (CC BY-NC-ND 2.0)

Drug cocktails are now the norm. Joshua Thorne/Flickr (CC BY-NC-ND 2.0)

Naturally, there are questions regarding the use of any genetic testing in the judicial system. Specific cases, such as the Italian judge who reduced the prison sentence of a convicted murderer for having genetic variants that may have caused his violent behaviour, have made some cautious to use pharmacogenomics in forensics, and understandably so. Although the clinical application of pharmacogenomics has mostly been established in other parts of medicine, its use in forensics still remains in the grey area. People want to use it; they just want more evidence to prove it’s worthwhile.

There are also issues with forensic scientists being able to interpret raw genetic data. In the course of my research, I have encountered some publications (that shall remain nameless) that I was initially very excited about reading, but that excitement quickly waned when I found the genetic interpretations to be incorrect. Of course, this obstacle can easily be overcome with the use of a pharmacogenomics interpretation/testing service. 

Yes, I’ve worked on some forensic cases. No, I won’t talk about them. But I can say this much: pharmacogenomics (from my ever so humble opinion and experience) does work in forensics. It answers questions that would have otherwise remained a mystery. I do hope the technology will be used in the future, and even become standardised. One day, someday.

Edited by Andrew Katsis and Ellie Michaelides