Pharmacogenomics

Imagine this: you’re a new mother, having just given birth to healthy baby boy. Unfortunately, the delivery has left you with debilitating post-partum pain. Your doctor prescribes you a painkiller, which really helps but also makes you feel drowsy, tired, and sometimes constipated. At the same time, you notice that your son isn’t breastfeeding very well. The doctor reassures you that everything’s fine and recommends you freeze and store your excess breastmilk for convenience, and you go home relieved.

One day, your child doesn’t wake up. His skin is gray and he’s not breathing. Frantically, you call 911 but it’s too late. Authorities arrive. An autopsy is done. Extremely high levels of morphine are found in your son’s blood. Infanticide is considered. Your life is in shambles.

This is a true case study published in the Lancet back in 2006.^1,2 The culprit ended up being codeine, an opioid analgesic. Codeine is found in Tylenol No. 3, a painkiller prescribed to many new mothers to relieve post-partum pain.^3,4 Yet the vast majority of these breastfeeding mothers have healthy babies. What happened in this case? In this episode of Medicurio, we will explore the concept of pharmacogenomics – how our genes change how we respond to medications.

When drugs enter our body, they are metabolized and structurally modified by enzymes which either activate or deactivate these medications. The most well-known family of drug-metabolizing enzymes are the cytochrome P450 enzymes. There are over 50 different isozymes of CYP450 enzymes, 7 of which are involved in metabolizing over 80% of medications as shown here.^5–7 For this story, let’s focus on CYP2D6.

The CYP2D6 gene sequence varies between people. These genetic variants, or alleles (denoted by numbers),^8,9 result in slight structural changes in the CYP2D6 enzyme, which alter how effective the enzyme is at metabolizing drugs. Broadly, they can be grouped as normal function, decreased function, and non-functional.¹⁰

Since we inherit a copy of each gene from each parent, each person will have two CYP2D6 alleles. A person can be classified as rapid (normal), intermediate, or poor metabolizers based on what combination of CYP2D6 alleles they inherited.¹⁰ A person who is a rapid metabolizer will have at least one functional allele. However, an intermediate metabolizer will have a decreased function and a non-functional allele, and a poor metabolizer will have two non-functional alleles.¹⁰ You can see all the possible combinations here – the majority of people are rapid metabolizers. In rare cases, a person can be classified as an ultra-rapid metabolizer. These people have three normal alleles of CYP2D6 rather than two due to a gene duplication, and so they metabolize drugs extremely rapidly and efficiently because they have more enzymes.¹⁰ Changes in metabolism can alter the effectiveness and toxicity of a drug.

Drugs can either be activated or inactivated after they are metabolized. Codeine itself is a weak painkiller but is still used because CYP2D6 metabolizes around 10% of the codeine into morphine, a potent opioid analgesic.^10,11

Now consider this – if all metabolizer types were given the same codeine dose, an ultra-rapid metabolizer would convert way too much codeine into morphine, and theoretically should be receiving a lower dose to prevent overdose.

It turned out the mother in the case was an ultra-rapid metabolizer and the excess morphine entered her breastmilk and into her son, which likely caused an overdose. High levels of morphine were found in her frozen breastmilk, supporting this explanation.^1,2 A year after this was reported, the FDA added a warning label on codeine use during breastfeeding, recommending that the lowest dose be used in nursing mothers.¹² It later strengthened its warning in 2017 after more reports of opioid overdose from breastfeeding.¹³

The takeaway from this story isn’t to say that codeine is a bad drug and should be banned – it’s quite an effective painkiller if used in the right people. Unfortunately, we just don’t know who the right people are unless we do genetic testing, which is still quite expensive¹⁴ and has implications on privacy and ethical issues.^15,16

But surely this tragedy could have been avoided if the doctors knew the mother was an ultra-rapid metabolizer so she would not be prescribed codeine. This story isn’t just applicable for codeine and CYP2D6. Almost all drugs are metabolized by a combination of CYP450 and phase II enzymes such as TPMT and UGT.^17,18 Many of these enzymes have genetic variations that change their function, altering the effectiveness and toxicity of the drug. We don’t know who has what genetic variation, so there is the risk of patients prescribed medications that could harm them. Genetic variations are a major cause of adverse drug reactions, which are a huge burden on the healthcare system (~5% of hospitalizations,^19,20 >100,000 deaths,²¹ and >$100 billion in healthcare costs²² in the US each year, and these numbers are likely underestimates²³).

The solution? Imagine if we can use pharmacogenomic information to prescribe each individual patient the most effective and least harmful medication based on their genetic makeup – in other words, precision medicine. Not only could it prevent a baby dying from toxic breastmilk (codeine), it could also prevent bone marrow damage in a patient with an autoimmune disease (azathioprine), reduce the risk of bleeding for people with bad hearts (warfarin),^18,24,25 and increase the chance a person is protected from future strokes (clopidogrel),^18,24 just to name a few (the FDA has a list of drugs with possible pharmacogenomic implications here).²⁶

But this is easier said than done. The hardest barrier to using pharmacogenomic information in the clinic is relating genes to outcomes.^15,25 While the codeine example seems straightforward, most drugs have such a complex metabolic pathway involving multiple enzymes that it is difficult to predict the patient’s response. Further complicating the situation are genetic changes in drug transporters and receptors,^18,20,27 as well as non-genetic reasons such as age, sex, concurrent diseases, drug-drug interactions, diet, and other environmental factors. All of these also influence a drug’s effect on a person.

There’s still a lot more to learn about pharmacogenomics before it can really help in the clinic, but new links between genes and drugs are being discovered every day. Perhaps in the future, when more research is done and genetic testing becomes more routine, doctors can also use genetic information when prescribing medicine to maximize benefits and minimize harms for every patient.

Thanks for watching, and see you next time on Medicurio.

Interesting tidbits:

The prevalence of this CYP2D6 phenotype varies widely and has been estimated at 0.5 to 1% in Chinese and Japanese, 0.5 to 1% in Hispanics, 1 to 10% in Caucasians, 3% in African Americans, and 16 to 28% in North Africans, Ethiopians, and Arabs. Data are not available for other ethnic groups.¹⁰

What’s up with grapefruit juice? A lot of medications should not be taken with grapefruit juice. That’s because grapefruit juice can inhibit the CYP3A4 enzyme, resulting in decreased metabolism of any drugs that are metabolized by CYP3A4, sometimes leading to toxic overdoses. This happens because grapefruit juice contains furanocoumarins are also broken down by CYP3A4, so it acts like a competitive inhibitor for the enzyme. Grapefruit juice also seems to inhibit drug transporters which could also make certain drugs poorly absorbed and useless.²⁸ Basically, avoid grapefruit juice when taking medications! It doesn’t taste that great anyways.

A different type of adverse drug reaction not mentioned in this video is hypersensitivity (allergy). This is best documented in the anti-HIV drug abacavir. Some patients have a unique HLA allele (see Celiac Disease video to learn more about HLA) which is able to bind to abacavir and abacavir-protein complexes, tricking the immune system into activating. This results in essentially a drug allergy – specifically, a delayed hypersensitivity reaction mediated by T cells (which is different from the “classic” drug allergies resulting in anaphylaxis).^18,24,29

In the US, the FDA-approved “AmpliChip CYP450” test can determine your CYP2D6 allele (as well as many other CYP450 enzymes). It is around $400 in the US and is not covered by insurance.¹⁴

Here are two interesting articles about the ethics of pharmacogenomics.^15,16 One topic that is mentioned is that pharmacogenomics poses an interesting conundrum for pharmaceutical companies. On one hand, being able to stratify patients based on genetics may lead to development of more drugs that are shown to work in a subset of a population. On the other hand, the drugs that only benefit a small population of genetically viable patients may not be as financially beneficial as developing “blockbuster” drugs that can be prescribed to as many people as possible.

References    

1. Koren, G., Cairns, J., Chitayat, D., Gaedigk, A. & Leeder, S. J. Pharmacogenetics of morphine poisoning in a breastfed neonate of a codeine-prescribed mother. The Lancet 368, 704 (2006).
2. Madadi, P. et al. Safety of codeine during breastfeeding. Can Fam Physician 53, 33–35 (2007).
3. Al-Adhami, N., Whitfield, K. & North, A. Changing Prescribing Culture – a Focus on Codeine Postpartum. Archives of Disease in Childhood 101, e2–e2 (2016).
4. Smolina, K., Weymann, D., Morgan, S., Ross, C. & Carleton, B. Association Between Regulatory Advisories and Codeine Prescribing to Postpartum Women. JAMA313, 1861–1862 (2015).
5. Lynch, T. & Price, A. L. The Effect of Cytochrome P450 Metabolism on Drug Response, Interactions, and Adverse Effects. AFP 76, 391–396 (2007).
6. Zanger, U. M. & Schwab, M. Cytochrome P450 enzymes in drug metabolism: Regulation of gene expression, enzyme activities, and impact of genetic variation. Pharmacology & Therapeutics 138, 103–141 (2013).
7. Kapur, B. M., Lala, P. K. & Shaw, J. L. V. Pharmacogenetics of chronic pain management. Clinical Biochemistry 47, 1169–1187 (2014).
8. CYP2D6. Pharmacogene Variation Consortium https://www.pharmvar.org/gene/CYP2D6.
9. Del Tredici, A. L. et al. Frequency of CYP2D6 Alleles Including Structural Variants in the United States. Front Pharmacol 9, 305 (2018).
10. Dean, L. Codeine Therapy and CYP2D6 Genotype. in Medical Genetics Summaries (eds. Pratt, V. M. et al.) (National Center for Biotechnology Information (US), 2012).
11. Vallejo, R., Barkin, R. L. & Wang, V. C. Pharmacology of Opioids in the Treatment of Chronic Pain Syndromes. Pain Physician 18.
12. Moaddeb, J. & Haga, S. B. Pharmacogenetic testing: current evidence of clinical utility. Ther Adv Drug Saf 4, 155–169 (2013).
13. FDA Drug Safety Communication: FDA restricts use of prescription codeine pain and cough medicines and tramadol pain medicines in children; recommends against use in breastfeeding women. U.S. Food and Drug Administration http://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-restricts-use-prescription-codeine-pain-and-cough-medicines-and (2018).
14. de Leon, J., Susce, M. T. & Murray-Carmichael, E. The AmpliChip^TM CYP450 Genotyping Test. Mol Diag Ther 10, 135–151 (2006).
15. Breckenridge, A. et al. Pharmacogenetics: ethical problems and solutions. Nat Rev Genet 5, 676–680 (2004).
16. Corrigan, O. Pharmacogenetics, ethical issues: review of the Nuffield Council on Bioethics Report. J Med Ethics 31, 144–148 (2005).
17. Ingelman-Sundberg, M. & Rodriguez-Antona, C. Pharmacogenetics of drug-metabolizing enzymes: implications for a safer and more effective drug therapy. Philos Trans R Soc Lond B Biol Sci 360, 1563–1570 (2005).
18. Daly, A. K. Pharmacogenetics: a general review on progress to date. Br Med Bull 124, 65–79 (2017).
19. Kongkaew, C., Noyce, P. R. & Ashcroft, D. M. Hospital admissions associated with adverse drug reactions: a systematic review of prospective observational studies. Ann Pharmacother42, 1017–1025 (2008).
20. Severino, G. & Zompo, M. D. Adverse drug reactions: role of pharmacogenomics. Pharmacological Research 49, 363–373 (2004).
21. Lazarou, J., Pomeranz, B. H. & Corey, P. N. Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA 279, 1200–1205 (1998).
22. Ernst, F. R. & Grizzle, A. J. Drug-related morbidity and mortality: updating the cost-of-illness model. J Am Pharm Assoc (Wash) 41, 192–199 (2001).
23. Schatz, S. N. & Weber, R. J. Adverse Drug Reactions. https://www.accp.com/docs/bookstore/psap/2015B2.SampleChapter.pdf.
24. Pharmacogenomics: Increasing the safety and effectiveness of drug therapy. American Medical Association https://crediblemeds.org/files/3913/6973/9557/pgx-brochure2011.pdf (2011).
25. Gardiner, S. J. & Begg, E. J. Pharmacogenetics, Drug-Metabolizing Enzymes, and Clinical Practice. Pharmacol Rev 58, 521–590 (2006).
26. Table of Pharmacogenomic Biomarkers in Drug Labeling. U.S. Food and Drug Administration https://www.fda.gov/media/124784/download (2019).
27. Zineh, I. Pharmacogenetics of Response to Statins. Curr Atheroscler Rep 9, 187–194 (2007).
28. Bailey, D. G., Dresser, G. & Arnold, J. M. O. Grapefruit–medication interactions: Forbidden fruit or avoidable consequences? CMAJ 185, 309–316 (2013).
29. Llano, A. & Brander, C. Mechanisms involved in the Abacavir-mediated hypersensitivity syndrome. Cell Res 22, 1637–1639 (2012).