ANSWER: D

Rationale:

Option D correctly identifies two established pharmacogenomic associations in ASD pharmacology. ABCB1 variants — in the gene encoding P-glycoprotein — have been associated in several cohort studies with differences in ASD response, consistent with the transporter hypothesis: patients with ABCB1 variants that increase P-gp expression or activity would have greater efflux of ASD substrates from brain tissue, contributing to pharmacoresistance. The SCN1A variant rs3812718 influences alternative splicing of the sodium channel, specifically the ratio of neonatal to adult splice variants, and has been associated in pharmacogenomic studies with differential response to carbamazepine and phenytoin — drugs whose binding depends on the inactivated state of the channel, which is influenced by splicing-dependent changes in channel kinetics.

• Option A: Option A is incorrect because lamotrigine is not a CYP2C19 substrate; it is primarily eliminated by UGT1A4-mediated glucuronidation, and CYP2C19 status does not predict lamotrigine levels or toxicity.
• Option B: Option B is incorrect because HLA-B*15:02 is associated with carbamazepine-induced Stevens-Johnson syndrome and toxic epidermal necrolysis, a severe cutaneous adverse reaction — not with pharmacoresistance through sodium channel subunit expression; its mechanism is immunological, not pharmacodynamic.
• Option C: Option C is incorrect because levetiracetam is not a CYP3A4 substrate; it undergoes enzymatic hydrolysis of its acetamide group by blood esterases, not hepatic CYP-mediated metabolism, and CYP3A4 poor metabolizer status does not affect levetiracetam pharmacokinetics.
• Option E: Option E is incorrect because oxcarbazepine is not primarily eliminated by UGT1A4 glucuronidation; it is rapidly reduced to its active metabolite licarbazepine (monohydroxy derivative) by cytosolic keto-reductases, with subsequent glucuronidation; UGT1A4 poor metabolizer status does not produce clinically significant oxcarbazepine accumulation through this pathway.

ANSWER: A

Rationale:

Option A correctly describes the GABA-A receptor subunit composition changes documented in epileptogenic tissue as a mechanism of target hypothesis resistance. In tissue from pharmacoresistant focal epilepsy, studies have shown shifts in GABA-A receptor subunit composition — including reduction in gamma-2 and delta subunits and upregulation of alpha-4 and alpha-6 subunits. These changes have direct pharmacological consequences: the gamma-2 subunit is required for high-affinity benzodiazepine binding at the benzodiazepine modulatory site, so its reduction diminishes benzodiazepine potentiation of GABA-A currents; delta subunit-containing receptors mediate tonic inhibition and are sensitive to low concentrations of neurosteroids and general anesthetics, and their loss reduces extrasynaptic inhibitory tone. Barbiturates act by prolonging chloride channel opening at a distinct site on the beta subunit, and subunit composition shifts can alter the barbiturate response profile as well. Critically, the receptors retain baseline GABA-activated conductance — this is target modification, not target elimination.

• Option B: Option B is incorrect because the target hypothesis involves changes in receptor subunit composition and pharmacological sensitivity, not loss of 80% of receptor protein; an 80% reduction in total receptor number would abolish inhibitory synaptic transmission and produce a profoundly different clinical phenotype than pharmacoresistant epilepsy.
• Option C: Option C is incorrect because while GABA-A receptor desensitization and phosphorylation occur during sustained seizure activity and contribute to benzodiazepine tolerance in status epilepticus, this is not the established mechanism of chronic pharmacoresistance described by the target hypothesis; the target hypothesis in DRE describes stable subunit composition changes, not irreversible phosphorylation locking receptors in a desensitized conformation.
• Option D: Option D is incorrect because barbiturate resistance is not mediated by GABA transaminase upregulation; barbiturates act directly on GABA-A receptor chloride channels at a binding site on the receptor itself, independent of synaptic GABA concentration; accelerated GABA catabolism would affect the efficacy of endogenous GABA but not the direct allosteric effect of barbiturates on the receptor channel.
• Option E: Option E is incorrect because GABA-A receptor subunit changes in epileptogenic tissue are not cell-type specific in the manner described; the subunit composition shifts affect the epileptogenic tissue broadly, and the described selective pattern — preserving interneuron GABA-A function while abolishing pyramidal neuron GABA-A function — is not the established mechanism of benzodiazepine resistance in the target hypothesis literature.

ANSWER: C

Rationale:

Option C correctly identifies intracranial electrode implantation — stereo-EEG (stereotactic-EEG with depth electrodes) or subdural grid electrodes — as the appropriate next step when non-invasive presurgical studies are discordant. In this patient, the VEEG localizes to the right temporal region while FDG-PET hypometabolism points to the left temporal region, with a non-localizing MRI and bilateral neuropsychological findings — a pattern of genuine discordance that cannot be resolved by repeating non-invasive studies. Intracranial recording allows direct electrocortical monitoring from both temporal regions simultaneously, with stimulation mapping to identify eloquent cortex, providing the spatial resolution and certainty required for safe surgical planning in an MRI-negative patient.

• Option A: Option A is incorrect because proceeding to resective surgery based on scalp VEEG localization alone when other studies are discordant risks resecting the wrong region or causing neurological deficits by missing eloquent cortex boundaries; concordance or invasive recording confirmation is required before surgery in discordant cases.
• Option B: Option B is incorrect because MRI-negative drug-resistant epilepsy does not have surgical success rates below 10%; seizure freedom rates in MRI-negative focal epilepsy following thorough evaluation, including invasive recording, range from 30–50% — substantially lower than MRI-positive TLE but far from negligible, and the discordant workup does not itself establish that no resectable zone exists.
• Option D: Option D is incorrect because FDG-PET hypometabolism represents interictal hypometabolism of the epileptogenic zone, not necessarily the ictal onset zone — ictal SPECT rather than FDG-PET reflects seizure onset, and FDG-PET hypometabolism contralateral to scalp EEG localization is a common finding in bilateral temporal epilepsy; repeating VEEG with additional electrodes would not resolve the fundamental discordance between modalities.
• Option E: Option E is incorrect because while MEG is a valuable additional non-invasive localizing study in MRI-negative epilepsy, it has not been shown in prospective randomized trials to be a definitive tiebreaker superior to intracranial recording; MEG is one of the additional non-invasive studies used before proceeding to invasive recording, not a substitute for it when discordance between multiple modalities is present.

ANSWER: E

Rationale:

Option E correctly states both the efficacy and the patient selection criterion for hemispherotomy: seizure freedom rates of approximately 70–80% in pediatric patients with hemispheric epilepsy syndromes — among the highest of any epilepsy surgical procedure — and the requirement for preexisting contralateral hemiplegia. Hemispherotomy or functional hemispherectomy disconnects the epileptogenic hemisphere from the contralateral hemisphere and subcortical structures, eliminating the seizure source. Because this procedure permanently eliminates motor output from the disconnected hemisphere, it will cause or consolidate contralateral hemiplegia. In patients who already have established contralateral hemiplegia from the underlying pathology — as in this child with Rasmussen encephalitis — the motor cost of the procedure has already been incurred by the disease, making the risk-benefit calculation favorable. In a patient with intact contralateral motor function, newly acquiring a permanent hemiplegia would constitute an unacceptable surgical deficit.

• Option A: Option A is incorrect because hemispherotomy outcomes of 70–80% seizure freedom substantially exceed neocortical resection rates of 30–50%, and the procedure does produce permanent contralateral motor deficit — the claim that it does not produce new motor deficits is incorrect; in patients without preexisting hemiplegia, it would newly cause permanent hemiplegia.
• Option B: Option B is incorrect because the 58–60% seizure freedom figure belongs to adult temporal lobectomy in TLE with mesial temporal sclerosis, not hemispherotomy; hemispherotomy achieves 70–80% seizure freedom, a substantially higher rate, and the comparison with temporal lobectomy is not the basis for patient selection.
• Option C: Option C is incorrect because myelination status of the corpus callosum is not a contraindication to hemispherotomy; the procedure is frequently performed in young children precisely because early intervention — before the epileptic hemisphere causes greater developmental harm to the contralateral hemisphere — may optimize developmental outcomes; age under 6 is not a contraindication.
• Option D: Option D is incorrect because the 50–60% seizure freedom rate cited belongs to neocortical resection outcomes, not hemispherotomy; hemispherotomy achieves 70–80% seizure freedom, and the 40% risk figure cited for new motor deficit is not a recognized statistic from the hemispherotomy literature as presented — the relevant point is that permanent contralateral hemiplegia is an expected consequence, not a probabilistic risk, of the procedure.

ANSWER: B

Rationale:

Option B correctly states the neocortical resection outcome: seizure freedom rates for non-temporal lobe epilepsy — particularly MRI-negative neocortical resections — are approximately 30–50%, substantially lower than the approximately 58–60% achieved by temporal lobectomy in mesial temporal sclerosis. The primary reason is that mesial temporal sclerosis provides a discrete, visible MRI lesion that serves as a reliable anchor for surgical planning — the epileptogenic zone and its boundaries are well-defined. In neocortical epilepsy without a structural lesion, the epileptogenic zone must be defined entirely by electrophysiological and functional mapping, which is less precise, and the resection margins may not completely encompass the zone responsible for seizure generation. Despite lower success rates than TLE surgery, neocortical resection remains a meaningful intervention with approximately one-third to one-half of patients achieving seizure freedom — an outcome that no ASD trial can approach in pharmacoresistant patients.

• Option A: Option A is incorrect because the seizure freedom rates are not equivalent; stereo-EEG localization improves surgical outcomes in MRI-negative cases compared with no localization, but it does not equalize outcomes with MRI-positive temporal lobectomy, where the structural lesion provides additional certainty about resection boundaries.
• Option C: Option C is incorrect because a seizure freedom rate below 10% is a significant underestimate of neocortical resection outcomes; the approximately 30–50% figure from the literature substantially exceeds 10%, and the claim that neuromodulation should precede neocortical resection when a localizable zone has been identified is not consistent with evidence-based epilepsy surgery guidelines.
• Option D: Option D is incorrect because neocortical resection outcomes are not higher than temporal lobectomy outcomes; frontal lobe seizures may have characteristic semiology but the epileptogenic zone boundaries are more difficult to define precisely in the absence of a structural lesion, yielding lower seizure freedom rates.
• Option E: Option E is incorrect because proximity to eloquent cortex affects the completeness of resection and surgical morbidity but is not the sole determinant of seizure freedom; the presence or absence of an MRI lesion independently predicts outcome, with MRI-positive cases consistently showing better seizure freedom rates than MRI-negative cases in surgical outcome registries.

ANSWER: D

Rationale:

Option D correctly describes the physiological basis and appropriate interpretation of interictal FDG-PET hypometabolism in presurgical epilepsy evaluation. In the interictal state, epileptogenic cortex and the surrounding functionally impaired tissue show reduced glucose metabolism — reflected as reduced FDG uptake — due to decreased neuronal activity between seizures. This interictal hypometabolism marks the epileptogenic zone and its functionally connected but dysfunctional surrounding cortex. An important practical point is that the hypometabolic region on FDG-PET is typically larger than the actual epileptogenic zone and the intended resection target; it should be interpreted as a localizing marker that overlaps with the seizure onset zone, not as a precise map of the resection boundaries. When the FDG-PET hypometabolism is concordant with VEEG localization — as in this case, both pointing to the right temporal region — it increases confidence in the localizing hypothesis and may support proceeding to invasive recording targeting that region.

• Option A: Option A is incorrect because the finding described — interictal hypometabolism — is explicitly an interictal finding; FDG-PET is obtained in the interictal state by design (the patient is not actively seizing during scanning), and interictal hypometabolism is the expected and informative signal; the concern about ictal confounding does not apply to this interpretation context.
• Option B: Option B is incorrect because FDG-PET and ictal SPECT measure different physiological phenomena and are not interchangeable; FDG-PET measures interictal glucose metabolism (hypometabolism in epileptogenic zones) while ictal SPECT measures cerebral blood flow during or immediately after a seizure (hypermetabolism at the ictal onset zone); the two provide complementary, not equivalent, information.
• Option C: Option C is incorrect because the mechanism of FDG-PET hypometabolism in epileptogenic cortex is reduced neuronal metabolic activity, not astrocytic glucose sequestration; while astrocytes do play a role in brain glucose metabolism, the hypometabolism seen in epileptogenic zones reflects genuine reduction in neuronal activity and energy demand between seizures, not transporter-mediated sequestration.
• Option E: Option E is incorrect because FDG-PET does not have sufficient spatial resolution for millimeter-scale epileptogenic zone delineation, and its hypometabolic region systematically overestimates the resection target; it provides localizing information but not the precision required for surgical margin definition, which is why it is used to guide — not replace — invasive electrocortical recording in MRI-negative cases.

ANSWER: A

Rationale:

Option A correctly distinguishes the two platforms. AAV-based gene therapy uses a viral capsid — typically a serotype with CNS tropism such as AAV9 or AAVrh10 — to deliver a functional gene construct, an inhibitory RNA sequence, or gene-editing machinery directly into neurons. Because AAV does not integrate into the host genome in postmitotic cells but persists as episomal DNA, it produces stable, potentially long-lasting or permanent transgene expression without requiring periodic redosing. This makes AAV suited for delivering new genetic material — such as a functional SCN1A transgene in Dravet syndrome or an inhibitory RNA targeting a gain-of-function allele. ASOs, by contrast, are short synthetic single-stranded oligonucleotides that bind complementary mRNA sequences through Watson-Crick base pairing and alter mRNA stability, processing, or translation; they do not enter the nucleus or alter genomic DNA, and they are cleared over weeks to months, requiring periodic intrathecal or intravenous redosing. ASOs are well suited to reducing expression from gain-of-function alleles (as in SCN8A epilepsy) or modifying splicing to upregulate expression from a functional allele (as proposed in Dravet syndrome SCN1A upregulation strategies).

• Option B: Option B is incorrect because AAV-based therapy and ASOs do not act through the same mechanism; AAV delivers DNA constructs that are transcribed into RNA and then protein within the transduced cell, fundamentally different from an ASO that acts directly on existing mRNA; they are not equivalent approaches packaged differently.
• Option C: Option C is incorrect because ASOs do not act as templates for homologous recombination; homologous recombination-based genome editing requires double-stranded DNA repair templates and CRISPR or similar machinery, not single-stranded oligonucleotides; ASOs act at the RNA level and do not permanently alter genomic sequence.
• Option D: Option D is incorrect because both platforms can in principle be applied to both gain-of-function and loss-of-function epilepsies; AAV can deliver inhibitory RNA constructs to reduce gain-of-function expression, and ASOs can be designed to upregulate expression (e.g., by blocking a silencing element) in loss-of-function conditions; the restriction described is an oversimplification that does not reflect the current state of the field.
• Option E: Option E is incorrect because AAV-based gene therapy does not require dividing cells; AAV episomal persistence in postmitotic neurons is in fact a key advantage of this vector system for CNS applications, and AAV-based therapies are being developed for adult as well as pediatric genetic epilepsies; age-based restriction of AAV to pediatric patients is pharmacologically incorrect.

ANSWER: C

Rationale:

Option C correctly identifies quinidine as the genotype-directed pharmacological intervention in KCNT1 gain-of-function epilepsy. KCNT1 encodes KNa1.1, a sodium-activated potassium channel; gain-of-function mutations increase potassium channel activity, causing excessive membrane hyperpolarization between action potentials followed by paradoxical hyperexcitability through rebound depolarization mechanisms, driving epileptogenesis in cortical and brainstem circuits. Quinidine, a class Ia antiarrhythmic used in cardiac rhythm disorders, was identified as a KNa1.1 channel blocker and repurposed for KCNT1 epilepsy based on the precision medicine principle that identifying the causal molecular defect enables hypothesis-driven pharmacological targeting. Case series and small cohort studies have reported seizure reduction in some KCNT1 gain-of-function patients, though responses are heterogeneous. This represents a paradigmatic example of genotype-directed therapy in epilepsy.

• Option A: Option A is incorrect because KCNT1 gain-of-function does not activate mTORC1; the mTOR pathway is regulated by TSC1/TSC2, not by potassium channel activity, and there is no established mechanistic link between KNa1.1 overactivity and mTORC1 hyperactivation; everolimus has no rational basis as a KCNT1-targeted therapy.
• Option B: Option B is incorrect because fenfluramine's mechanism in Dravet syndrome involves serotonin modulation and sigma-1 receptor activity in the context of SCN1A-related interneuron dysfunction; KCNT1 gain-of-function does not impair serotonergic interneuron signaling through the mechanism described, and fenfluramine is not the rationale-based intervention for potassium channelopathy.
• Option D: Option D is incorrect because KCNT1 gain-of-function does not primarily drive epileptogenesis through NMDA receptor disinhibition via magnesium block removal; while membrane potential changes can influence NMDA receptor activity, the primary pharmacological target for KCNT1-mediated hyperexcitability is the potassium channel itself, and memantine targeting NMDA receptors downstream is not the first-line rational intervention.
• Option E: Option E is incorrect because there is no established protein-protein interaction between SV2A and KNa1.1 that would make levetiracetam a rationale-based precision therapy for KCNT1 gain-of-function; the described interaction is pharmacologically fabricated and does not appear in the KCNT1 or SV2A literature.

ANSWER: E

Rationale:

Option E correctly identifies memantine as the mechanistically rational intervention for GRIN2B gain-of-function epilepsy. Memantine is an uncompetitive, voltage-dependent NMDA receptor antagonist that enters the open ion channel pore and blocks current flow in a use-dependent manner — it preferentially blocks channels that are pathologically overactive rather than channels engaged in normal synaptic transmission. In GRIN2B gain-of-function mutations that reduce magnesium block sensitivity and prolong channel open time, the result is excessive calcium influx and sustained NMDA receptor-mediated excitotoxic signaling. Memantine directly antagonizes this overactivation by blocking the channel at the same site — the channel pore — where magnesium normally provides voltage-dependent block. Its use-dependent kinetics mean it preferentially affects channels with prolonged open times, making it mechanistically well-suited to the specific biophysical defect. Preliminary clinical data support its use in GRIN2A and GRIN2B gain-of-function epilepsies as a genotype-directed intervention.

• Option A: Option A is incorrect because GRIN2B gain-of-function does not cause compensatory 5-HT2C receptor downregulation; there is no established mechanistic link between GluN2B overactivation and serotonin receptor regulation that would make fenfluramine a rationally targeted therapy for this specific channelopathy.
• Option B: Option B is incorrect because there is no established protein-protein interaction between SV2A and GluN2B at the postsynaptic density that would confer GluN2B gain-of-function selectivity to levetiracetam; SV2A is a presynaptic vesicle protein and GluN2B is postsynaptic; the described allele-selective inhibition mechanism is pharmacologically fabricated.
• Option C: Option C is incorrect because GRIN2B gain-of-function does not activate mTORC1 through CaMKII phosphorylation of TSC2; while NMDA receptor activation does engage CaMKII signaling, the specific mechanistic link to TSC2 phosphorylation sufficient to produce mTOR pathway hyperactivation comparable to TSC mutations is not established, and everolimus is not a rationale-based precision therapy for GRIN2B channelopathy.
• Option D: Option D is incorrect because valproate's GABA-transaminase inhibitory activity raises inhibitory tone broadly but does not pharmacodynamically correct the specific gain-of-function biophysical defect of reduced magnesium sensitivity and prolonged open time; increasing GABAergic inhibition is a symptomatic approach that any broad-spectrum ASD could provide, not a mechanism-targeted intervention at the NMDA receptor level.

ANSWER: B

Rationale:

Option B correctly describes CBD's antiseizure mechanisms as distinct from sodium channel blockade and operating through multiple non-sodium-channel pathways. The three primary mechanisms best supported by preclinical and clinical evidence are: antagonism of GPR55, a G protein-coupled receptor activated by lysophosphatidylinositol that when activated promotes neuronal excitability and calcium release — GPR55 antagonism by CBD reduces excitatory signaling; modulation of transient receptor potential (TRP) channels, particularly TRPV1 and TRPA1, which influence neuronal calcium signaling and inhibitory interneuron function; and inhibition of adenosine reuptake via equilibrative nucleoside transporter 1 (ENT1), increasing extracellular adenosine concentrations that act on inhibitory A1 and A2A adenosine receptors to reduce neuronal excitability. These mechanisms collectively explain why CBD can be effective in Dravet syndrome despite the contraindication of sodium channel blockers — CBD does not further suppress Nav1.1 in inhibitory interneurons.

• Option A: Option A is incorrect because CBD does not have high selectivity for Nav1.1 as its primary mechanism; CBD does have some sodium channel activity but this is not Nav1.1-selective, and its established antiseizure mechanisms are the non-sodium-channel pathways described in Option B; framing CBD as a Nav1.1-selective sodium channel blocker mischaracterizes its pharmacology.
• Option C: Option C is incorrect because CBD's antiseizure effects are not mediated exclusively through CB1 receptors on GABAergic interneurons; CBD has low affinity for CB1 receptors relative to THC, and its antiseizure activity persists in CB1 receptor knockout animal models, establishing that CB1 is not the primary antiseizure target; the GPR55, TRP, and adenosine mechanisms are the predominant contributors.
• Option D: Option D is incorrect because CBD is not a GABA-transaminase inhibitor and does not share the mechanism of valproate; these are pharmacologically distinct molecules with different mechanisms, and CBD's clinical pharmacology is not explained by GABA-transaminase inhibition; valproate's teratogenicity is also a clinically important issue, but comparing CBD to valproate on mechanistic grounds in this way is pharmacologically inaccurate.
• Option E: Option E is incorrect because CBD's primary antiseizure mechanisms are not presynaptic voltage-gated calcium channel blockade; ethosuximide blocks T-type calcium channels in thalamic neurons to suppress absence seizures, a mechanism distinct from CBD's GPR55/TRP/adenosine profile; conflating these two drugs' mechanisms misrepresents both.

ANSWER: D

Rationale:

Option D correctly frames the SUDEP risk and the urgency argument for surgical evaluation. Sudden unexpected death in epilepsy (SUDEP) is the leading cause of epilepsy-related mortality in young adults. In drug-resistant epilepsy, annual SUDEP incidence is approximately 1 in 150 to 1 in 1,000 — a range that reflects population heterogeneity — compared with approximately 1 in 11,000 in patients with well-controlled epilepsy. Over the median 10-year delay from the point of pharmacoresistance to surgical evaluation, this excess mortality accumulates into a substantial population-level harm. Surgical seizure freedom substantially reduces SUDEP risk, providing a compelling mortality-based argument for timely surgical referral that goes beyond quality-of-life considerations. Framing surgical evaluation as urgent — not elective — is medically accurate and ethically necessary when counseling patients who are delaying referral.

• Option A: Option A is incorrect because SUDEP risk in drug-resistant epilepsy substantially exceeds the operative mortality of temporal lobectomy, which is less than 0.5% in experienced centers; the risk-benefit calculation strongly favors surgical evaluation, not equivalence between the two approaches.
• Option B: Option B is incorrect because SUDEP occurs across all epilepsy types including focal epilepsy, and drug-resistant temporal lobe epilepsy carries a particularly elevated SUDEP risk because uncontrolled nocturnal generalized tonic-clonic seizures are a primary SUDEP risk factor, which are common in temporal lobe epilepsy with secondary generalization.
• Option C: Option C is incorrect because SUDEP risk does not decline to background population levels after two years of stable drug-resistant epilepsy; the risk is persistent and proportional to ongoing seizure frequency and severity, not time on a stable regimen; long-standing drug-resistant epilepsy does not confer protection against SUDEP.
• Option E: Option E is incorrect because while nocturnal monitoring may provide some safety benefit, no monitoring or safety device has been shown to fully mitigate SUDEP risk, and implementing monitoring measures does not reduce the urgency of surgical evaluation; SUDEP is not prevented by observation alone, and framing monitoring as a substitute for timely surgical referral is clinically misleading.

ANSWER: A

Rationale:

Option A correctly distinguishes all three variants across the four parameters. The classic ketogenic diet specifies a fat-to-combined-protein-and-carbohydrate ratio of 4:1 by weight, requires precise calorie counting and gram-level food weighing, and is implemented with dietitian-supervised meal planning. The modified Atkins diet uses a substantially less restrictive fat ratio (approximately 1:1 to 2:1), limits carbohydrates to 10–20 grams per day without specifying fat intake targets, and does not require food weighing or calorie counting — a significantly lower compliance burden. The low glycemic index treatment (LGIT) uses approximately 60% of calories from fat and restricts carbohydrates not by strict gram limit but by glycemic index — carbohydrates consumed must have a glycemic index below 50, allowing moderate carbohydrate intake from low-glycemic foods. Critically, observational series show that responder rates across all three variants are comparable, making patient-specific factors — compliance capacity, age, food preferences, lifestyle — the primary basis for variant selection rather than efficacy hierarchy.

• Option B: Option B is incorrect because classic KD and MAD are not equivalent in fat ratio; the defining distinction between them is precisely the fat ratio (4:1 vs 1:1–2:1) and the associated compliance requirements; and LGIT has not been shown superior to classic KD in randomized controlled trials — the randomized controlled trial evidence base in dietary epilepsy therapy is strongest for the classic KD (Neal et al. 2008).
• Option C: Option C is incorrect because the fat ratio in ketogenic dietary variants is not age-determined; the choice between variants is based on compliance feasibility and patient-specific factors, not age-based metabolic differences; a 4-year-old and a 40-year-old may both be on classic 4:1 KD or MAD depending on their clinical circumstances and adherence capacity.
• Option D: Option D is incorrect because mandatory hospitalization with a fasting period is no longer standard practice for classic KD initiation; contemporary protocols initiate the diet in the outpatient or day-hospital setting without mandatory fasting, since fasting does not improve long-term seizure outcomes; and MAD does not require weekly clinic visits as a standard protocol.
• Option E: Option E is incorrect because MAD and LGIT do not have equivalent carbohydrate restrictions — MAD restricts carbohydrates to 10–20 g/day regardless of glycemic index, while LGIT allows more carbohydrates provided they have a glycemic index below 50; and the classic KD is used across all age groups including ambulatory children and adults, not exclusively in medically fragile infants.

ANSWER: C

Rationale:

Option C correctly explains the pharmacogenomic mechanism of phenytoin toxicity in this patient. Phenytoin is primarily metabolized by CYP2C9 (and to a lesser extent CYP2C19) via aromatic hydroxylation to its major inactive metabolite, p-hydroxyphenytoin (HPPH), which is then conjugated and renally excreted. CYP2C9 accounts for the majority of this metabolic clearance, and the *3 allele encodes an enzyme with markedly reduced catalytic activity. Homozygous *3/*3 poor metabolizers have substantially lower phenytoin clearance than extensive metabolizers, causing the drug to accumulate to toxic concentrations at standard doses. This is further compounded by phenytoin's saturable (zero-order, Michaelis-Menten) kinetics at therapeutic concentrations: as plasma levels approach the Km of the metabolic enzymes, small increases in dose or decreases in enzyme capacity produce disproportionately large increases in plasma concentration — the pharmacokinetic basis for phenytoin's notoriously narrow therapeutic index. In CYP2C9 poor metabolizers, dose reductions of 20–50% may be required to achieve therapeutic levels.

• Option A: Option A is incorrect because CYP2C9 does not mediate renal tubular secretion; CYP2C9 is a hepatic (and intestinal) oxidative enzyme, not a renal transporter; phenytoin glucuronide excretion involves UGT enzymes and renal transporters, not CYP2C9, and enterohepatic recirculation of phenytoin is not a clinically significant pharmacokinetic process.
• Option B: Option B is incorrect because CYP2C9 is not a significant contributor to intestinal first-pass metabolism of phenytoin; phenytoin bioavailability is primarily determined by formulation dissolution and gastrointestinal absorption characteristics, not intestinal CYP2C9 activity; CYP2C9 poor metabolizer status does not increase bioavailability through this intestinal mechanism.
• Option D: Option D is incorrect because phenytoin is not a prodrug requiring CYP2C9 activation; phenytoin itself is the pharmacologically active sodium channel blocker, and CYP2C9 mediates its inactivation to HPPH — the consequence of CYP2C9 poor metabolizer status is reduced inactivation and drug accumulation, not failure to activate a prodrug.
• Option E: Option E is incorrect because CYP2C9 does not affect albumin synthesis; albumin is synthesized in the liver independently of CYP enzyme activity, and CYP2C9 poor metabolizer status has no established effect on plasma albumin concentration or phenytoin protein binding; hypoalbuminemia is a separate clinical concern requiring free phenytoin level monitoring but has nothing to do with CYP2C9 genotype.