Table 1.

Influence of genetic factors on response and adverse reactions to AEDs through various mediators: summary of existing findings.

Response
Mediator	Genetic factor	Effect [references]
Pharmacokinetics and pharmacodynamics	Variation in CYP2C9 gene	Risk of developing concentration-dependent neurotoxicity from phenytoin [12], [13]- established evidence
	Variation in CYP2C19 gene	Association with the serum concentration of N-desmethylclobazam and with its clinical efficacy, indicating a gene-dose effect [17], [18], [19], [20], [21]
	Variation in CYP2C19 gene	Ethnic differences in the tolerability profile of phenobarbital [22]
	UGT1A1 variants	Altered clearance of lamotrigine [25]
	Variation in CYP2C19 gene	Risk of adverse reactions from zonisamide [28]
	Variation in SCN1A, ABCC2, UGT2B7 genes	Association with oxcarbazepine maintenance doses [36]
	Variation in CYP1A1 gene	Association with response to first-line antiepileptic drugs in Indian women [37], [38]
	ABCB1 gene (encoding P-glycoprotein, P-gp, multidrug transporter) variants	Drug-resistant epilepsy [40], [41], [42], [43], [44], [45]
	Variation in genes coding for AED targets	No significant association with drug response [49], [50], [51], [52], [53]
Epilepsy genes	Mutations in the SLC2A1 gene causing GLUT-1 deficiency syndrome	Gold standard treatment is the ketogenic diet, treating the symptoms of neuroglycopenia [[57], [58], [59]- established evidence
	Bi-allelic mutations in the ALDH7A1 gene causing pyridoxine (vitamin B6)-dependent epilepsy	Gold standard treatment is pyridoxine or pyridoxal 5′-phosphate supplementation [60], [61]- established evidence
	SCN1A-related epilepsies	Stiripentol is effective in Dravet Syndrome (especially when combined with valproate and clobazam), established in a randomised controlled trial; however, the use of stiripentol, valproate and clobazam does not always yield complete seizure freedom and may cause adverse side effects [70], [71], [72]
		Ranolazine was identified as a selective blocker of persistent currents in mutant NaV1.1 channels expressed in heterologous expression systems [76], [77], [78]- uncertain evidence, recently it was shown that ranolazine does not inhibit the persistent Na+ current more strongly than phenytoin in central neurons
		Clemizole was identified as an effective inhibitor of spontaneous convulsive behaviour and electrographic seizures in zebrafish Nav1.1 (scn1Lab) mutants [81]
		Fenfluramine has serotoninergic effects and was shown to significantly reduce epileptiform discharges in scn1Lab morphants. Evidence of effectiveness in control of convulsive seizures and good tolerability in a small cohort of patients with Dravet Syndrome [83], [84], [85], [86], [87], [88], [89]- ongoing clinical trials in Dravet syndrome
		Use of sodium channel blockers can aggravate seizures in Dravet syndrome. Note, however, that lamotrigine can improve seizure control in some patients with Dravet Syndrome [92], [94]
	GRIN2A causing early-onset epileptic encephalopathy	Use of memantine, a N-methyl-d-aspartate (NMDA) receptor antagonist, can inhibit the increased activity of the NMDA receptor caused by specific mutations [102]
	KCNT1-related epilepsy	Use of quinidine, partial antagonist of KCNT1 used as antiarrhythmic drug, was shown to reduce seizure frequency and improve psychomotor development in one patient with migrating partial seizures of infancy [103] and was shown to reverse the hyperactivity of mutant KCNT1 in Xenopus oocytes [104]
	KCNQ2-related epilepsies	Use of retigabine (ezogabine), a positive allosteric modulator of KCNQ2-5 (Kv7.2–7.5) ion channels, was shown to partially reverse the loss of function, in vitro, of missense KCNQ2 mutations associated with severe epileptic encephalopathy [107], [108], [113]- only preliminary data in humans
		Sodium channel blockers seem effective in KCNQ2-related epilepsy, possibly because voltage-gated sodium channels and KCNQ potassium channels co-localize and are bound at critical locations of the neuronal membrane [111], [114], [115], [116]
	SCN2A-related epilepsies	Sodium channel blockers have shown significant effectiveness in SCN2A-epileptic encephalopathies; pathophysiological considerations suggest that sodium channel blockers should only be applied in patients with gain-of-function mutations [118], [119]
		Treatment with oral 5-hydroxytryptophan, l-Dopa/Carbidopa, and a dopa agonist resulted in mild improvement of seizure control, in a case with a de novo SCN2A splice site mutation associated with epileptic encephalopathy [120]
	SCN8A-related epilepsies	Sodium channel blockers have shown effectiveness in patients with gain-of-function mutations; in rare cases an apparent loss-of-function effect has been described but it remains unclear if these may show gain-of-function effect in vivo [121], [123], [127], [128]
	Tuberous sclerosis complex (TSC) associated with TSC1 or TSC2 mutations	Rapamycin (sirolimus), an inhibitor of the mechanistic target of rapamycin (mTOR) signaling cascade, was shown to prevent the development of epilepsy and premature death in mouse models of TSC and to significantly improve seizure control in children and adults with TSC-associated epilepsy, with a tolerable safety profile [137], [138], [139], [140], [141]
	DEPDC5-related epilepsies	DEPDC5 encodes a protein that is part of the GATOR1 complex, a negative regulator of the mechanistic target of rapamycin complex 1 (mTORC1), in the mTOR pathway; its mutations are predicted to result in upregulation of the mTOR signaling pathway. Inhibition of mTOR pathway with rapamycin was shown to suppress cytomegalic neurons and epileptic seizures in model systems [152], [153], [154]. No studies on treatment with mTOR inhibitors in patients with gene mutations affecting the GATOR1 complex subunit
	PRICKLE-mediated seizures	USP9X, a substrate-specific de-ubiquitinase, is a partner in the mammalian PRICKLE-interactome. Inhibition of USP9X can arrest PRICKLE-mediated seizures [163]. No studies in humans
Adverse reactions
	HLA-B*15:02	Stevens-Johnson syndrome and toxic epidermal necrolysis induced by carbamazepine and other aromatic AEDs in patients from Han Chinese and other South Asian ethnic groups [165], [166], [167], [168], [169], [170]- established evidence
	HLA-A*31:01	Increased risk of carbamazepine-induced hypersensitivity reactions in patients of European ancestry and in the Japanese population [172], [173]- established evidence
	T1405 polymorphism of the CPS1 gene	Increased risk of valproate-induced hyperammonaemia in Caucasian patients [176]
	Val16Ala polymorphism of the SOD2 gene	Elevated serum level of γ-glutamyltransferase induced by valproate in Japanese patients [179]
	Polymorphic LEPR and ANKK1 genes	Weight gain on valproate in Han Chinese patients [180]
	Variation in CYP2C9 and CYP2A6 genes	Risk of toxicity from valproate [181], [182]