NeuroGenetics Curriculum·advanced·20 min

Pharmacogenetics in Neurology

An applied guide to pharmacogenetics for the practicing neurologist — covering CYP450 enzyme genetics, drug-gene interactions relevant to antiepileptic and neuropsychiatric drug therapy, HLA-associated hypersensitivity reactions, and the clinical implementation of pharmacogenetic testing.

Tags: Neurogenetics · Advanced

Learning Objectives

  1. 1.Describe how CYP450 enzyme genetic variation creates metabolizer phenotypes and their pharmacokinetic consequences
  2. 2.Identify the most clinically important drug-gene interactions in neurology practice
  3. 3.Explain HLA allele associations with serious antiepileptic drug hypersensitivity reactions
  4. 4.Apply pharmacogenetic principles to antiepileptic drug selection and dosing
  5. 5.Interpret a pharmacogenetic test report and integrate results into clinical practice

01Principles of Pharmacogenetics

Pharmacogenetics is the study of how genetic variation influences drug response — affecting absorption, distribution, metabolism, excretion (ADME), and pharmacodynamic drug targets. Genetic variants in drug-metabolizing enzymes alter plasma drug concentrations, creating a spectrum from toxicity (impaired metabolism → drug accumulation) to therapeutic failure (ultra-rapid metabolism → subtherapeutic levels). The major clinical phenotypes are: poor metabolizer (PM), intermediate metabolizer (IM), normal/extensive metabolizer (NM/EM), and ultra-rapid metabolizer (UM).

Key Points

  • Phase I metabolism: CYP450 enzymes (CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP3A4/5) — oxidation, reduction, hydrolysis; most critical for neurological drug metabolism
  • Phase II metabolism: UGT enzymes (glucuronidation), TPMT (thiopurine methylation), NAT2 (acetylation) — less critical for most neurological drugs but important for valproate (UGT1A6/UGT2B7)
  • Star (*) allele nomenclature: reference allele = *1 (normal function); loss-of-function alleles (e.g., CYP2C9*2, *3); gain-of-function alleles (CYP2D6*1xN gene duplication = ultra-rapid)
  • Copy number variation: CYP2D6 gene can be deleted (PM), duplicated (UM), or multiplied (UM with >2 functional copies); CYP2D6*1xN duplication causes UM phenotype
  • CPIC guidelines (cpicpgx.org): Clinical Pharmacogenetics Implementation Consortium — evidence-based prescribing recommendations based on genotype; freely available and regularly updated

02CYP450 Enzymes Most Relevant to Neurology

Four CYP450 enzymes are most important for neurological drugs: CYP2C9 (phenytoin, valproate, losartan), CYP2C19 (clopidogrel, clobazam, diazepam, omeprazole), CYP2D6 (tricyclics, opioids, atomoxetine, antipsychotics), and CYP3A4/5 (carbamazepine, oxcarbazepine, statins). Polymorphisms in these enzymes are common — CYP2D6 PM phenotype affects ~7–10% of Europeans; CYP2C19 PM affects ~2–5% of Europeans but up to 15–20% of Asians.

Key Points

  • CYP2C9 and phenytoin: poor metabolizers (CYP2C9*2/*3 compound heterozygous) have dramatically reduced phenytoin clearance → toxicity at standard doses; CPIC recommends 25–50% dose reduction and monitoring in PMs; phenytoin has narrow therapeutic index
  • CYP2C19 and clopidogrel: clopidogrel is a prodrug requiring CYP2C19 activation; PMs (CYP2C19*2/*3) cannot convert to active thienopyridine → increased stroke/MI risk; CYP2C19 loss-of-function alleles common in Asians (~50% have at least one); alternative antiplatelet therapy (prasugrel, ticagrelor) for PMs
  • CYP2C19 and clobazam: clobazam is metabolized to active N-desmethylclobazam by CYP2C19; PMs have 5-fold higher N-desmethylclobazam levels → increased sedation risk; dose reduction recommended
  • CYP2D6 and tricyclic antidepressants (amitriptyline, nortriptyline): PMs have very high plasma levels → cardiac arrhythmia, anticholinergic toxicity; UMs have subtherapeutic levels; CPIC recommends alternative antidepressants for PMs/UMs
  • CYP3A4/5 and carbamazepine: CYP3A4/5 metabolizes carbamazepine; also induces its own metabolism (autoinduction); drug interactions with CYP3A4 inhibitors/inducers are complex; CYP3A5*3 reduces activity but clinical impact of genotype is less than for CYP2D6/2C19

03HLA Alleles and Serious Drug Hypersensitivity in Neurology

Certain HLA alleles confer high risk of severe immune-mediated drug hypersensitivity reactions — Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), and drug reaction with eosinophilia and systemic symptoms (DRESS). Neurologists prescribe several drugs with well-characterized HLA associations. Pre-prescription HLA testing prevents severe, potentially fatal adverse reactions.

Key Points

  • HLA-B*15:02 and carbamazepine SJS/TEN: HLA-B*15:02 is present in ~8–10% of Han Chinese, Thai, and other Southeast Asian populations (rare in Europeans <0.1%); carbamazepine-SJS risk is markedly elevated (>50-fold) in carriers; FDA mandates HLA-B*15:02 testing before carbamazepine use in high-risk Asian ancestry patients
  • HLA-A*31:01 and carbamazepine: common in Northern Europeans (~5%), Japanese; associated with carbamazepine DRESS and maculopapular exanthem; less severe than SJS/TEN but still clinically significant
  • HLA-B*57:01 and abacavir (antiretroviral): hypersensitivity syndrome in ~5% of HIV+ patients; mandated pre-prescription testing in many countries; 100% negative predictive value if absent
  • HLA-B*58:01 and allopurinol: common in Han Chinese (~6–8%); strong association with allopurinol SJS/TEN in Asian populations; screening recommended in high-risk ethnic groups before starting allopurinol for gout
  • Oxcarbazepine cross-reactivity: patients with HLA-B*15:02 who have SJS with carbamazepine are at risk with oxcarbazepine and structurally related AEDs; avoid in B*15:02 carriers; lacosamide and levetiracetam have no known HLA association; lamotrigine is also associated with HLA-B*15:02-related SJS/TEN risk, though the association is weaker than with carbamazepine

04Antiepileptic Drug Pharmacogenomics

Antiepileptic drug (AED) pharmacogenomics encompasses both pharmacokinetic (drug metabolism) and pharmacodynamic (drug target) genetic variation. SCN1A variants that reduce sodium channel sensitivity may explain resistance to sodium channel-blocking AEDs. UGT enzymes metabolize lamotrigine and valproate. POLG mutations contraindicate valproate use. These interactions have direct clinical management implications.

Key Points

  • SCN1A and sodium channel AED response: Dravet syndrome is caused by SCN1A loss-of-function; sodium channel-blocking AEDs (oxcarbazepine, lamotrigine, carbamazepine, phenytoin) may paradoxically worsen seizures in Dravet by further reducing Nav1.1; oxcarbazepine and lamotrigine are the most important to avoid as they are commonly prescribed before a genetic diagnosis — valproate, clobazam, and stiripentol are preferred; see the [[epilepsy|Genetic Epilepsies]] module for comprehensive coverage of genetic epilepsy syndromes and treatment implications
  • POLG (mitochondrial DNA polymerase gamma) mutations and valproate hepatotoxicity: POLG-related disorders (Alpers syndrome, POLG-spectrum disorder) — valproate causes fulminant hepatotoxicity and neurological deterioration; MUST screen for POLG mutations or suggestive features before starting valproate in children with developmental regression or mitochondrial features; see the [[mitochondrial|Mitochondrial Disease]] module for POLG-spectrum disorder clinical features
  • UGT1A4 and lamotrigine: UGT1A4 metabolizes lamotrigine; female sex hormones (pregnancy, oral contraceptives) induce UGT1A4, dramatically increasing lamotrigine clearance; serum level monitoring essential; lamotrigine dose often needs to double during pregnancy
  • Valproate and NAGS/CPS1 (urea cycle): valproate inhibits urea cycle → hyperammonemia in partial UCD carriers; valproate-induced hyperammonemic encephalopathy; consider UCD evaluation before valproate in patients with unexplained hyperammonemia or protein aversion
  • CYP2C9 and phenytoin toxicity: ~0.2–0.4% of Europeans are CYP2C9 poor metabolizers; phenytoin toxicity (nystagmus, ataxia, lethargy) at standard doses should prompt CYP2C9 genotyping and dose reduction

05Clinical Implementation of Pharmacogenetic Testing

Pharmacogenetic testing is increasingly available as preemptive panels that genotype multiple clinically actionable variants before drug prescribing is needed. Implementation requires understanding how to interpret multi-gene reports, recognizing the limitations of current evidence, and integrating results with clinical context. Several health systems have implemented preemptive pharmacogenomics as part of precision medicine initiatives.

Key Points

  • Preemptive vs. reactive testing: reactive testing (at the time of prescribing) requires fast turnaround (days to weeks) which may delay treatment; preemptive panel testing (at first clinical encounter) stores results in EHR for all future prescribing decisions — more cost-effective over time
  • Test report interpretation: reports gene name, diplotype (e.g., CYP2D6*1/*4), predicted phenotype (PM/IM/NM/UM), and drug-specific recommendations; metabolizer phenotypes are substrate-specific (same gene, different recommendations per drug)
  • Evidence tiers: CPIC grades recommendations as A (action required), B (consider modification), C (inform/optional) — not all variants require prescribing changes; distinguish strong associations from weak signals
  • Limitations: most panels cover common variants in European populations; sensitivity lower for non-European ancestries; structural variants (CYP2D6 CNV, CYP2D6-CYP2D7 hybrids) may be missed by simple SNP arrays
  • EHR integration: pharmacogenomic clinical decision support (CDS) alerts at the point of prescribing are most effective; passive reporting without alerts has minimal impact on practice; CPIC guidelines are designed for implementation in EHR-based CDS systems

Quiz Questions

1. A woman with epilepsy on a stable lamotrigine dose delivers her baby. Two weeks postpartum, she develops diplopia, ataxia, and nausea. Her lamotrigine level is found to be twice the pre-pregnancy target. The most likely explanation is:

  1. A.Postpartum depression is causing psychosomatic symptoms that mimic lamotrigine toxicity but are unrelated to drug levels
  2. B.Breastfeeding increases lamotrigine absorption from the gut, leading to supratherapeutic maternal serum concentrations
  3. C.Postpartum estrogen decline reverses UGT1A4 induction, reducing lamotrigine clearance — the pregnancy dose causes toxicity✓
  4. D.Postpartum autoimmune hepatitis has impaired all hepatic drug metabolism including lamotrigine glucuronidation

During pregnancy, rising estrogen levels induce UGT1A4, dramatically increasing lamotrigine clearance and often requiring dose increases of 50-100% or more. After delivery, estrogen levels drop rapidly, and UGT1A4 induction reverses. If the elevated pregnancy dose is not promptly tapered postpartum, lamotrigine accumulates and causes toxicity (diplopia, ataxia, nausea). This is the mirror image of the pregnancy-related clearance increase and illustrates the importance of close lamotrigine level monitoring both during and after pregnancy.

2. A 60-year-old man with neuropathic pain is started on amitriptyline 25 mg nightly. Within days he develops confusion, urinary retention, and QTc prolongation. Pharmacogenetic testing reveals he is a CYP2D6 poor metabolizer (*4/*4). This adverse reaction is best explained by:

  1. A.CYP2D6 poor metabolizers cannot absorb amitriptyline from the GI tract, so the drug accumulates in the intestinal wall causing local toxicity
  2. B.CYP2D6 poor metabolizers have reduced amitriptyline clearance, causing drug accumulation and anticholinergic/cardiac toxicity✓
  3. C.The *4/*4 genotype causes a paradoxical immune-mediated drug allergy to tricyclic antidepressants that mimics toxicity
  4. D.CYP2D6 poor metabolizer status has no clinically significant effect on tricyclic metabolism; the reaction is purely idiosyncratic

Tricyclic antidepressants (amitriptyline, nortriptyline) are primarily metabolized by CYP2D6. Poor metabolizers (e.g., CYP2D6*4/*4, carrying two loss-of-function alleles) have dramatically reduced drug clearance, leading to accumulation of parent drug and toxic metabolites even at standard doses. This manifests as severe anticholinergic toxicity (confusion, urinary retention, dry mouth) and cardiac toxicity (QTc prolongation, arrhythmia risk). CPIC recommends avoiding tricyclics in CYP2D6 poor metabolizers and selecting alternatives not dependent on CYP2D6.

3. A Thai woman with newly diagnosed focal epilepsy needs an antiepileptic drug. She is found to carry HLA-B*15:02. Which of the following AEDs can be safely prescribed without HLA-related SJS/TEN risk?

  1. A.Carbamazepine — the association with HLA-B*15:02 is only relevant for Han Chinese, not Thai patients
  2. B.Oxcarbazepine — it is structurally different enough from carbamazepine to avoid cross-reactivity
  3. C.Levetiracetam — it has no known HLA association with SJS/TEN and is safe regardless of HLA-B*15:02 status✓
  4. D.Phenytoin — aromatic AEDs are safe if the starting dose is low

HLA-B*15:02 is prevalent in Southeast Asian populations including Thai (not just Han Chinese) and confers a markedly elevated risk of SJS/TEN with carbamazepine. Oxcarbazepine has cross-reactivity and should also be avoided in B*15:02 carriers. Lamotrigine also carries some HLA-B*15:02-associated SJS/TEN risk. Levetiracetam has no known HLA association with serious cutaneous adverse reactions and is a safe choice. This scenario reinforces that HLA-B*15:02 screening is relevant across Southeast Asian populations, and alternative AEDs without HLA-mediated hypersensitivity should be selected for carriers.

4. A 4-year-old girl with epilepsy is on clobazam for seizure control. Her seizures are well-controlled but she develops excessive sedation. Pharmacogenetic testing shows she is a CYP2C19 poor metabolizer (*2/*2). The mechanism underlying her sedation is:

  1. A.CYP2C19 PMs cannot activate clobazam, so the parent drug accumulates and causes CNS depression through an off-target sedative mechanism
  2. B.CYP2C19 PMs have ~5-fold higher N-desmethylclobazam levels because CYP2C19 is the primary enzyme clearing this active metabolite✓
  3. C.CYP2C19 PMs have reduced GABA-A receptor sensitivity, requiring higher clobazam doses that then cause paradoxical sedation
  4. D.CYP2C19 genotype does not affect clobazam pharmacokinetics or metabolite levels; the sedation is entirely unrelated to genotype

Clobazam is metabolized to N-desmethylclobazam, a pharmacologically active metabolite with a long half-life. CYP2C19 is the primary enzyme responsible for further metabolism (clearance) of N-desmethylclobazam. In CYP2C19 poor metabolizers, N-desmethylclobazam accumulates to approximately 5-fold higher levels than in normal metabolizers, causing excessive sedation. Dose reduction of clobazam is recommended in CYP2C19 PMs. This is a clinically important interaction in pediatric epilepsy, where clobazam is widely used.

5. A hospital is implementing a preemptive pharmacogenomic testing program. Which statement best describes the advantage of preemptive over reactive pharmacogenetic testing?

  1. A.Preemptive testing is cheaper per test because it only analyzes one gene at a time rather than running a full multi-gene panel
  2. B.Preemptive panel testing stores results in the EHR for all future prescribing, avoiding treatment delays from reactive testing✓
  3. C.Preemptive testing eliminates the need for CPIC guidelines because all drug-gene interactions are automatically flagged by the panel
  4. D.Reactive testing is always preferred because pharmacogenetic evidence changes too rapidly for stored panel results to remain valid

The key advantage of preemptive pharmacogenomic testing is that multi-gene panel results are available in the EHR before any drug is prescribed, enabling immediate pharmacogenomic-informed prescribing decisions without treatment delays. Reactive testing (ordering at the time of prescribing) requires days to weeks for results, which may delay critical drug therapy. Preemptive testing is increasingly cost-effective as panel costs decrease, and EHR-integrated clinical decision support (CDS) alerts at the point of prescribing maximize the clinical utility of stored results. CPIC guidelines are designed specifically for implementation in such EHR-based CDS systems.