11 December 2009

Antidepressants, CYP2D6, and opioid metabolism

Awhile back I posted this bleg for more information about the interaction between antidepressants and codeine. Peter Nelson emailed me asking whether I had found out anything else. I hadn't, so he did some research of his own which he has kindly agreed to allow me to share.

Take it away, Peter:

[Usual disclaimer: Neither he nor I are medical professionals. Don't take this as medical advice, et cetera.]

Since emailing you I’ve been studying the research literature and it’s crystal clear that codeine will not have any analgesic properties for people either genetically lacking CYP2D6 (6-10% of caucasians, other %’s for other ethnic groups) or who are taking a drug that blocks it.
Many antidepressants, including fluoxetine, paroxetine and bupropion are strong inhibitors of it, as are many other drugs including various antiarrhythmics, antifungals, cancer drugs, etc.

The story on the other synthetic opioids doesn’t look too good either. CYP2D6 plays a critical role in the metabolism of hydrocodone, oxycodone, and tramadol but they have more complex metabolic pathways and even now there are details that remain to be elucidated.

Hydrocodone itself has little affinity for the μ opioid (pain) receptors so it has to get metabolized the main clinically-active metabolite is assumed to be hydromorphone because it’s a known painkiller with a high affinity for the μ opioid receptors. And lack of CYP2D6 blocks that process. That part is clear, but there are unanswered questions.

For example in Kaplan et al, (Inhibition of cytochrome P450 2D6 metabolism of hydrocodone to hydromorphone does not importantly affect abuse liability J Pharmacol Exp Ther. 1997 Apr;281(1):103-8) subjects’ subjective perception of the effects of hydrocodone were unrelated to hydromorphone conversion. Heiskanen et al, (Effects of blocking CYP2D6 on the pharmacokinetics and pharmacodynamics of oxycodone. Clin Pharmacol Ther. 1998 Dec;64(6):603-11. ) performed a similar experiment involving oxycodone with similar results.
But critically, neither experiment looked at pain tolerance. Also Otton et al, (CYP2D6 phenotype determines the metabolic conversion of hydrocodone to hydromorphone - SV - Clin Pharmacol Ther - 01-NOV-1993; 54(5):) performed an experiment similar to Kaplan’s but did find that subjects responded in ways consisten with hydromorphone conversion (again, no pain test).

Based on what we think we know about hydrocodone (i.e., that the active metabolite is hydromorphone), Otton’s results make more sense. But both hydrocodone and oxycodone still have work left to do elucidating the effects of some of the other metabolites that are currently thought to be inactive.

And Heiskanen’s results also make sense because oxycodone – the parent compound - actually appears to have a nontrivial affinity for μ receptors itself, and furthermore some of its other metabolites such as noroxycodone, which may be mediated by a different enzyme – CYP2C19 - may also have high binding affinity. (Lalovic et al, Quantitative contribution of CYP2D6 and CYP3A to oxycodone metabolism in human liver and intestinal microsomes. Drug Metab Dispos. 2004 Apr;32(4):447-54. )
In other words oxycodone may work fine as an analgesic for CYP2D6 impaired patients. BUT that doesn’t mean oxycodone gets us off the hook - instead it appears to have a nastier hook: The oxymorhone is far more readily cleared than the parent compound oxycodone. So without CYP2D6 oxycodone accumulates, potentially becoming toxic or fatal.
Two studies underscore that risk: Jannetto, et al, Pharmacogenomics as molecular autopsy for postmortem forensic toxicology; genotyping cytochrome P450 2D6 for oxycodone cases. J Anal Toxicol 2002; 26:438–447 and Drummer et al, A study of deaths involving oxycodone. J Forensic Sci 1994; 39:1069–1075.

The real bottom lines are these:
1. Work remains to be done in elucidating both the pharmacokinetics and clinical effects of various metabolites for all of the synthetic opioids.

2. As far as I could tell, there seem to be no human studies evaluating analgesic properties of synthetic opioids for patients who either lack the gene for CYP2D6 or for whom CYP2D6 is blocked by a drug-drug interaction.

3. Drug-drug interactions of this type will become more common as the population becomes older and as we use a greater variety of drugs. As it is, bupropion. paroxetine and fluoxetine (all potent CYP2D6 blockers) accounted for roughly 50 million prescriptions in the US alone last year. Other CYP2D6 blockers account for millions more.

4. I’ve spoken to several physicians about this and they all expressed worry and concern that they feel unsure how to do pain management for CYP2D6-impaired patients, especially in postoperative or fracture cases where OTC drugs aren’t enough and the “nuclear options” like fentanyl or methadone (both of which work regardless of CYP2D6) would be overkill and dangerous.

Visit Peter's blog at http://blog.pnart.com/. Thanks again!