Pharmacogenomics of analgesics in anesthesia practice: A current update of literature

Abstract

The field of pharmacogenomics seeks to understand how an individual's unique gene sequence can affect their response to certain drugs. It is particularly relevant in anesthesia when the interindividual response to pain medication is essential. Codeine and tramadol are prodrugs metabolized by CYP2D6, polymorphisms of which can cause dangerous or even fatal levels of their metabolites, or decrease the level of metabolites to decrease their analgesic effect. Many other opioids are metabolized by CYP2D6 or CYP3A5, of which loss-of-function variants can cause dangerous levels of these drugs. The OCT1 transporter facilitates the movement of drugs into hepatocytes for metabolism, and variants of this transporter can increase serum levels of morphine and O-desmethyltramadol. Many NSAIDs are metabolized by CYP2C9, and there is concern that variants of this enzyme may lead to high serum levels of these drugs, causing gastrointestinal bleeding, however the data does not strongly support this. The ABCB1 gene encodes for P-glycoprotein which facilitates efflux of opioids away from their target receptors. The C3435T SNP may increase the concentration of opioids at target receptors, although the data is not conclusive. Catechol-O-Methyltransferase (COMT) is shown to indirectly upregulate opioid receptors. Certain haplotypes of COMT have been demonstrated to have an effect on opioid requirements. The OPRM1 gene codes for the mu-opioid receptor, and there is conflicting data regarding its effect on analgesia and opioid requirements. Overall, there is a fair amount of conflicting data in the above topics, suggesting that there is still a lot of research to be done on these topics, and that pain perception is multifactorial, likely including many common genetic variants.

Introduction

A daily challenge that faces the clinical anesthesiologist is the great variability of the response of a given patient to a specific dose of a certain medication, and in particular, analgesic drugs. While this extraordinary variability in patient response to pharmacotherapy is certainly multifactorial, much research has been done in the use of molecular medicine to generate predictions regarding clinical response to medication based on the patient's personal DNA signature. This field of study has been termed “pharmacogenomics,” and has pursuits in both pharmacokinetics and pharmacodynamics, with the goal of creating “personalized medicine” in which medical treatment is customized according to an individual's genetic signature.[1]

Methods

This review included the relevant current literature regarding pharmacogenomics as pertaining to the use of analgesics in anesthesiology practice. A systematic search of PubMed, Goggle Scholar, and Cochrane Database was done in May 2016 using the key words Pharmacogenomics, Anesthesia, CYP polymorphism, and metabolizers. The search retrieved 165 titles and abstracts, hence, the search was restricted to only human studies published in English. Letters, commentaries, editorials, and case reports were not included to be part of selected materials to be reviewed. Total of 84 articles were reviewed. All literature related to CYP polymorphism and analgesics were included. However, no formal criteria except those mentioned above were applied for inclusion or exclusion of studies.

The differences in the genotypes associated with CYP polymorphism produce several types of distinct clinical pain phenotypes presented in Table 1.

Table 1.

Relationship between genotypes and phenotypes for CYP activity in opioid metabolizing pathways

Pharmacokinetics – prodrug activation

Codeine: Codeine is a prodrug, which is demethylated by the CYP2D6 enzyme (a subtype of cytochrome P450) to morphine, which has 200 times the affinity for the mu-opioid receptor compared to codeine.[2] The CYP2D6 enzyme is highly polymorphic, with over 100 alleles identified thus far. Some of these alleles can lead to inactivation, reduction in function, or duplication of the enzyme.[3] These various genotypes have been described with their accompanying phenotypes. Two nonfunctioning alleles are associated with poor metabolism, one or two functioning alleles are associated with extensive metabolism, and duplicated alleles or an allele with a promoter function are associated with ultrarapid metabolism.[4] The ultra-metabolizers tend to produce more active metabolites from codeine (e.g., morphine, morphine-6-glucuronides) which are more active on opioid receptor than the codeine substrate which acts in this case as prodrug. As a result, opioid side effects (e.g., respiratory depression) observed in ultra-metabolizers might be much more pronounced that extensive (normal) metabolizers.

One early case report showed life-threatening opioid intoxication from codeine prescribed for cough in a patient with bilateral pneumonia who was later found to have three functioning CYP2D6 alleles, consistent with ultrarapid metabolism.[5] Subsequent case reports described apnea in a child following oral codeine administration, later found to have CYP2D6 allele with a promoter function,[6] and death in a child following codeine administration following adenotonsillectomy, who was found to have supertherapeutic levels of codeine and functional duplication of CYP2D6.[7] Of note, as with many children undergoing adenotonsillectomy, the child had obstructive sleep apnea, which already increases the risk of hypoxemia. A follow-up case report described several more fatal or life-threatening incidents in children taking codeine following adenotosillectomy.[8] One case report also revealed apnea leading to death in a nursing newborn of a mother taking codeine, who was found to have a genotype associated with ultrarapid metabolism,[9] and a case-control study demonstrated severe respiratory depression in breastfeeding newborns with mothers having an ultrarapid metabolism genotype.[10] Due to this mounting evidence, the Food and Drug Administration issued a Black Box Warning for codeine in May, 2013, listing it as contraindicated in pediatrics following adenoidectomy and/or tonsillectomy, and adding a warning regarding breastfeeding mothers taking codeine. This also led the Clinical Pharmacogenetics Implementation Consortium to publish guidelines regarding the use of genotyping and the prescription of codeine.[11]

In addition, one observational study evaluated the effectiveness of codeine in postpartum analgesia following elective caesarean section, and found that patients with genotypes associated with poor metabolism received no analgesia from codeine, whereas those with genotypes associated with ultrarapid metabolism experienced sedation.[12]

Tramadol: Similar to codeine, tramadol is also a prodrug metabolized by CYP2D6, which converts it to its active metabolite, O-desmethyltramadol, also referred to as (+)-M1, which has mu-opioid activity. Experimental pain studies found that patients with genotypes associated with ultrarapid metabolism experienced reduced discomfort from experimental pain, accompanied by increased serum levels of the tramadol metabolite, whereas patients with genotypes associated with poor metabolism had little clinical effect with minimally detectable tramadol metabolite.[13,14,15] In similar fashion to codeine, there are case reports that describe respiratory depression in patients taking tramadol that are later found to have an ultrarapid metabolizing genotype, both in a patient with renal impairment[16] and in a child after adenotonsillectomy.[17] The results of clinical trials vary as patients with a poor metabolizing genotype receiving tramadol following abdominal surgery required more additional/rescue opioids than patients with an extensive metabolizing genotype,[18] whereas patients with a poor metabolizing genotype receiving tramadol following knee arthroscopy attested to better analgesia than patients with an extensive metabolizing genotype.[19] This suggests that during intense pain the therapeutic effects of tramadol may rely more heavily on its opioid metabolite, but during less-severe pain the nonopioid effects of tramadol itself may be more beneficial.

Pharmacokinetics – elimination

Opioids: Many opioids aside from codeine are also metabolized by cytochrome P450 enzymes for elimination, and thus, have been studied through the lens of pharmacogenomics. Although alfentanil is known to be metabolized by CYP3A5, study on multiple allele variants failed to show differences in systemic clearance.[20] Fentanyl is also known to be metabolized by CYP3A5, and a study in cancer patients transitioning to transdermal fentanyl revealed that those who were homogenous for the decreased function CYP3A5*3 variant had increased fentanyl plasma concentrations and an increased rate of central nervous system side effects.[21] The CYP3A4 enzyme has also been shown to play a role in the elimination of fentanyl, with patients possessing the CYP3A4*1 allele having a significantly lower postoperative fentanyl requirement.[22] Hydrocodone is known to be metabolized by the CYP2D6 enzyme, and a case report documented a case of fatal hydrocodone overdose in a child who was later found to have a genotype associated with poor metabolism.[23] Morphine in also a metabolized by CYP2D6, although its alleles appear to have a counterintuitive effect in a study where patients with a genotype associated with ultrarapid metabolism had decreased morphine requirements after elective surgery.[24]

The organic cation transporter 1 (OCT1) facilitates the uptake of drugs into hepatocytes for metabolism. The serum levels of morphine and O-desmethyltramadol (tramadolactive metabolite) are influenced by variations in OCT1.[25,26] Of note, the activation of tramadol and codeine take place independently of OCT1, and thus, the accumulation of the metabolites can still occur based on variations in the transporter. Clinical studies on serum morphine levels in pediatric patients following adenotonsillectomy showed reduced morphine clearance in patients with loss-of-function OCT1 variants.[27,28]

NSAIDs: Nonsteroidal anti-inflammatory drugs such as naproxen and celecoxib have been studied from a pharmacogenomic standpoint primarily regarding their side-effect profile, specifically gastrointestinal (GI) bleeding. As with the medications already discussed, many NSAIDs are metabolized by cytochrome P450 enzymes, specifically CYP2C8 and CYP2C9. The CYP2C9 polymorphisms were initially studied, and the loss-of-function allele CYP2C9*3, when homogenous, was associated with a two-fold reduction in celecoxib clearance compared to the wild type.[29] A later study failed to demonstrate any difference in plasma levels of naproxen in heterozygous patients of the CYP2C9*3 allele.[30] Studies regarding gastrointestinal bleeding have mixed results, with a few studies showing the CYP2C9*3 allele to be associated with a significantly higher risk of bleeding,[31,32] while other studies failed to show a difference in the risk of GI bleeding,[33,34] although one study showed a significant difference in risk in patients with a specific combined CYP2C8 and CYP2C9 mutation.[35] It should be noted that these mutations are quite rare, thus making it difficult for studies to achieve significant power.

Pharmacokinetics – Transmembrane transport

The ABCB1 gene encodes for P-glycoprotein (PGP) which facilitates the access of xenobiotics to the brain and causes efflux of opioids away from their target receptors. A specific single nucleotide polymorphism (SNP) of the gene which encodes PGP has been found to modulate its activity. This SNP, labelled C3435T, has been shown to improve the efficacy of opioids in experimental pain.[36] This same SNP has also been shown to decrease postoperative pain in pediatrics[37] and to decrease opioid requirements for postoperative pain[38] and cancer pain.[21,39,40] Thus, the C3435T SNP likely decreases PGP activity, which reduces efflux of opioids away from target receptors, thus increasing concentration at opioid receptors. This is consistent with a study that found C3435T to be associated with opioid-induced respiratory depression,[41] and case reports that linked it to respiratory depression in pediatric patients following adenotonsillecomy.[42,43] However, there are some studies that failed to show a relationship between ABCB1 variations and postoperative pediatric pain scores,[44] analgesic effect of oxycodone on postoperative (PO) pain,[45] or morphine use in patients following caesarean section.[46]

Pharmacodynamics – COMT

COMT degrades catecholamines. A variant of COMT (val158met), where a valine is substituted by methionine at codon 158 is produced by SNP G772A has reduced activity which increases dopamine levels. This increased level of circulating dopamine suppresses endogenous opioid production, which in turn upregulates opioid receptors. This increase in opioid binding sites in response to this variant has been seen in postmortem brain cells,[47] and val158met has been shown to decrease regional/endogenous mu-opioid system response and cause higher pain ratings to experimental pain.[48]

An early study demonstrated val158met heterozygosity as being linked to decreased opioid requirements for cancer pain.[49] These findings were later reproduced for postoperative pain.[50] However, many more studies failed to show significance of the val158met variant as produced by the SNP G772A, and instead showed significance in certain COMT haplotypes, which are a set of polymorphisms that have a tendency toward being inherited together. These haplotypes were found to modulate COMT activity,[51] experimental pain sensitivity,[52,53] and opioid requirements for cancer pain[54,55] and PO pain.[56,57]

Pharmacodynamics – Mu opioid receptor

The mu-opioid receptor is encoded by the OPRM1 gene, which is highly polymorphic, and has been extensively studied. An early study involving patients receiving methadone maintenance found A118G (adenine replaced by guanine at codon 118) to be the most common SNP in the OPRM1 gene. The resultant variant receptor had three times the binding affinity for β-endorphin and endogenous opioid.[58] Studies with experimental pain are inconclusive regarding patients who are homozygous for A118G, with findings ranging from increased[59] to decreased[60] pain threshold, to not finding significance.[61]

In the clinical setting, multiple studies demonstrate A118G homozygosity to be associated with increased opioid requirements for both PO[62,63,64,65,66,67] and cancer pain.[68,69] However many studies have shown decreased PO opioid requirements in patients with A118G homozygosity,[70,71,72] or simply failed to show significance of the A118G variant on PO opioid usage.[73,74,75,76]

It is important to note that the available clinical results are often controversial and contradictory, particularly when large populations are examined in association with clinical significance of detected polymorphisms in opioid pathways.

Perhaps the most striking study to date is one which evaluated opioid dosage in 2294 patients being treated for cancer pain.[77] In this study, none of the 112 SNPs from 25 candidate genes (including OPRM1, ABCB1, and COMT) showed significant associations with opioid dose.[77]

Conclusion

Pharmacogenomics is a rapidly expanding field, but there is not, as of yet, a sufficient body of evidence to support the use of widespread genetic screening to predict individual responses to pain medications and the risk of adverse side effects (with exception of CYP2D6 genome). Pain perception and response to medications is determined by many common genetic variants, that have yet to be discovered. The future of research will include the continued analysis of these many variants and how they relate to one another for pharmacogenomics to become a part of standard clinical practice. The future success of pharmacogenomic testing will depend upon more extensive sequencing strategies, and the characterization of rare mutations with definite biological impact on treatment response, adverse effects, and pain pathology.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.