Ketamine

Ketamine is a non-selective, non-competitive antagonist of the NMDA receptor. Its major circulating metabolite, norketamine, acts at the same receptor with lower affinity and is about one third as active as ketamine in animal testing [1].

After an intravenous dose, blood concentrations fall with an initial half-life of 10 to 15 minutes, which corresponds to the anaesthetic effect, then decline more slowly as the drug redistributes out of the brain into other tissues, with a redistribution half-life of 2.5 hours. Clearance is by liver metabolism: ketamine is N-dealkylated to norketamine primarily by CYP2B6 and CYP3A4 and to a lesser extent by other enzymes, after which norketamine is hydroxylated to hydroxynorketamines and conjugated [1].

Which enzyme leads the first step has been argued over for twenty years. Using pooled human liver microsomes and expressed enzymes at clinically relevant concentrations, one group concluded that CYP3A4 is the principal enzyme for ketamine N-demethylation, with CYP2B6 and CYP2C9 minor contributors [4]. Later work with genotyped microsomes fitted a two-enzyme model in which the high-affinity enzyme matched CYP2B6 and the low-affinity enzyme matched CYP3A4, and found that a CYP2B6 inhibitor and an antibody against CYP2B6 blocked the reaction at clinically relevant concentrations while a CYP3A4 inhibitor and anti-CYP3A4 antibody did not, which points to CYP2B6 as the main enzyme at the concentrations that matter [2]. A third study identified CYP3A4, CYP2C19, CYP2B6, CYP2A6, CYP2D6 and CYP2C9 as all able to form norketamine, with CYP2B6 and CYP2A6 the only two that then hydroxylated it [5]. The label reflects the unresolved state of this by naming both enzymes [1].

CYP2B6 is highly polymorphic, with dozens of described alleles. CPIC groups them into normal function (CYP2B6*1), decreased function (including *6 and *9), no function and uncertain or unknown function. CYP2B6*6 is the most frequent decreased-function allele, with a minor allele frequency between 15% and 60% depending on ancestry, and its reduced activity is attributed largely to aberrant splicing caused by c.516G>T (rs3745274, p.Gln172His) [6]. In 1000 Genomes data the T allele at rs3745274 is on 37.4% of gene copies in Africans, 38.1% in South Asians, 37.3% in admixed Americans, 23.6% in Europeans and 21.5% in East Asians [7], so a great many people carry at least one copy.

What the laboratory shows

The in vitro effect on ketamine is substantial and specific. In human liver microsomes genotyped for CYP2B6*6, the intrinsic clearance of both ketamine enantiomers by the high-affinity enzyme was at least 2 times higher in CYP2B6*1/*1 microsomes than in CYP2B6*1/*6, and at least 6 times higher than in CYP2B6*6/*6. Comparing the expressed proteins directly, CYP2B6.1 had about 160% of the maximum reaction rate of CYP2B6.6 and about 70% of its Michaelis constant, meaning the variant protein both binds ketamine less tightly and turns it over more slowly [2].

What happened in people

Thirty volunteers, ten each with CYP2B6*1/*1, *1/*6 and *6/*6, received a subsedating oral dose of ketamine, with plasma and urine concentrations of ketamine and its CYP2B6-dependent metabolites measured by mass spectrometry. The primary outcome was the ratio of norketamine to ketamine exposure. There was no significant difference between genotypes in that ratio or in any secondary measure, including ketamine exposure, peak concentration, apparent oral clearance and metabolite formation clearances. Participants' self-assessments did differ in energy and awareness. The authors' conclusion was that while CYP2B6*6 diminishes ketamine metabolism in vitro, it did not affect metabolism, clearance or pharmacokinetics after a single low oral dose in people, and that in vitro observations need clinical validation [3].

So the strongest link in this chain is the enzymology: CYP2B6 is on the pathway twice, the variant protein is measurably worse at the reaction, and the allele is common. The weakest link is the only human test of the idea, which was negative. Its design limits how far that negative goes, since it used a single low oral dose in healthy volunteers with ten people per group, and oral ketamine undergoes extensive first-pass metabolism that may behave differently from an intravenous anaesthetic dose. But a null primary outcome in the one study that asked the question directly is the most relevant piece of evidence available, and it is why the rows below should be read with low confidence.

Gene2Rx wrote this guideline itself and the recommendations are exploratory. They are written for a prescriber, and ketamine given medically is administered by clinicians who titrate to effect and monitor vital signs, which is a stronger safeguard than any genotype. The strength the report assigns each row is shown with it. Do not start, stop or change a medicine on the basis of a genetic result without talking to your prescriber.

The intermediate and poor metabolizer rows describe the in vitro findings faithfully [2], but the predicted higher concentrations and longer sedation were not seen in the one human study [3]. A strength of strong on the poor metabolizer row is therefore higher than the evidence supports. The ultrarapid and rapid metabolizer rows are correctly given a strength of none: no ketamine study has examined increased CYP2B6 function.

Ketamine is one of the four substances in the Starter Report, alongside alcohol, caffeine and nicotine, because it is widely encountered both as a medicine and outside medicine, and because the gene behind it is one Gene2Rx already reports for other drugs. CYP2B6 is the enzyme in the guideline for good reasons: the label names it, it appears twice on the metabolic pathway, and its most common variant allele has a measured effect on the ketamine reaction in liver preparations [1][2].

The case against making much of it is equally clear. The in vivo test was null, the relative roles of CYP2B6 and CYP3A4 are still debated, and medical ketamine is titrated and monitored in a way that absorbs modest differences in clearance [3][4]. A guideline that implied a CYP2B6 result should change an anaesthetic dose would be overreaching.

What the result is genuinely useful for is context. CYP2B6 also clears efavirenz, where its effect is established well enough for a CPIC guideline [6], and reduced function there has real consequences. For ketamine specifically, the practical value is knowing that if you have had an unusually long or deep recovery from ketamine, a reduced-function CYP2B6 result is a plausible contributing factor to mention, not an explanation, and that the better documented risks of ketamine have nothing to do with genotype.

If you are having a procedure, tell the anaesthetist about previous reactions to anaesthesia and about every medicine and supplement you take. That history matters more than a CYP2B6 result, which the label does not ask prescribers to consider.

The label's interaction warnings are worth knowing. Theophylline and aminophylline should not be given with ketamine, because the combination may lower the seizure threshold. Sympathomimetics and vasopressin call for close monitoring of vital signs. Benzodiazepines, opioid analgesics and other central nervous system depressants prolong recovery [1]. Drugs that inhibit or induce CYP2B6 or CYP3A4 could in principle change ketamine concentrations, and esketamine itself modestly induces both enzymes in human hepatocytes [8]; no clinical interaction study with ketamine was found for this page, so that remains a theoretical concern rather than a documented one.

Two risks of ketamine are documented and genotype-independent. Emergence reactions, meaning confusion, vivid imagery, hallucinations or delirium during recovery, occurred in about 12% of patients in the label's experience and usually last a few hours; a benzodiazepine is given to prevent them, and reducing stimulation during recovery helps [1]. Transient rises in blood pressure, heart rate and cardiac index are common, and ketamine is contraindicated when a significant rise in blood pressure would be hazardous [1].

Repeated use carries harms that no genetic result modifies. The label describes hepatobiliary dysfunction, usually cholestatic, along with bile duct dilatation, stricture and obstruction with recurrent use, and reports of cystitis, reduced bladder capacity, ureteral stenosis and obstruction leading to kidney damage that can need urgent intervention [1]. Ketamine is a Schedule III controlled substance, and people with a history of substance use disorder are at greater risk of misuse [1].

Gene2Rx reads the variant calls in an uploaded file and assesses a defined set of CYP2B6 star alleles, so a rare reduced-function variant outside that set is not seen, and a Normal Metabolizer result means only that none of the assessed variants was found among the positions the data covers. Consumer arrays (23andMe, AncestryDNA, MyHeritage) read fixed positions, are not clinical-grade tests, and their raw data is labeled not for medical use; array processing infers some variants by imputation, but a variant at a position the data does not cover cannot be seen.

Beyond the test, the biology limits what a result can say. CYP3A4 shares the first metabolic step, and CYP3A4 activity varies several-fold between people for reasons that are mostly not genetic, so a CYP2B6 result describes part of one of two parallel routes. How someone responds to ketamine also depends on the dose and the rate it is given, the route, other anaesthetics and sedatives, age, liver and kidney function, and the clinical situation. The one human genotype study found no effect on clearance [3], so this result should not be used to predict how long sedation will last, and it says nothing about whether ketamine will help a mood disorder.

Should I tell my anaesthetist about a CYP2B6 poor metabolizer result?

You can mention it, but it is unlikely to change anything. The label does not ask prescribers to consider CYP2B6 genotype, and the one controlled study found no difference in ketamine clearance between genotypes [1][3]. Your history of previous anaesthetics and your current medicines are what the anaesthetist needs.

Does this apply to esketamine for depression?

The same enzymes are involved: esketamine is metabolized to noresketamine mainly by CYP2B6 and CYP3A4, with smaller contributions from CYP2C9 and CYP2C19 [8]. No study of esketamine by CYP2B6 genotype was found for this page, so the guideline above should not be read across to it.

Why is ketamine in the Starter Report?

The Starter Report covers four commonly encountered substances, and ketamine is one of them. Every drug Gene2Rx covers, including ketamine, is also in the Full Report.

1. Ketalar (ketamine hydrochloride) injection: prescribing information. DailyMed. Mechanism, pharmacokinetics, interactions, emergence reactions and the hepatobiliary and urinary harms of recurrent use.
2. Li Y et al. The CYP2B6*6 allele significantly alters the N-demethylation of ketamine enantiomers in vitro. Drug Metab Dispos (2013). Intrinsic clearance by genotype and the comparison of CYP2B6.1 with CYP2B6.6.
3. Rao LK et al. Role of cytochrome P4502B6 polymorphisms in ketamine metabolism and clearance. Anesthesiology (2016). The 30-volunteer study with a null primary outcome.
4. Hijazi Y, Boulieu R. Contribution of CYP3A4, CYP2B6 and CYP2C9 isoforms to N-demethylation of ketamine in human liver microsomes. Drug Metab Dispos (2002). The case for CYP3A4 as the principal enzyme.
5. Portmann S et al. Enantioselective capillary electrophoresis for identification of human cytochrome P450 enzymes which metabolize ketamine and norketamine in vitro. J Chromatogr A (2010). Which enzymes form norketamine and which hydroxylate it.
6. Desta Z et al. CPIC guideline for CYP2B6 and efavirenz-containing antiretroviral therapy. Clin Pharmacol Ther (2019). CYP2B6 allele function groups and the frequency and mechanism of CYP2B6*6.
7. dbSNP: rs3745274 (CYP2B6 c.516G>T). 1000 Genomes allele frequencies.
8. Spravato (esketamine) nasal spray: prescribing information. DailyMed. Esketamine metabolism and its modest induction of CYP2B6 and CYP3A4.