QT PROLONGATION WITH HYDROXYCHLOROQUINE AND AZITHROMYCIN FOR THE TREATMENT OF COVID-19: THE NEED FOR PHARMACOGENETIC INSIGHTS

We read with great interest the recent study by Maraj et al¹, which showed an increased risk of QT prolongation with hydroxychloroquine and azithromycin for the treatment of COVID-19. The authors emphasized that “delineation of the determinants of significant QTc prolongation and proarrhythmic risk for hydroxychloroquine/azithromycin is very important.” However a major determinant is missing from this study: genetics. Genetics explains ~30% of the variability in drug-induced QT prolongation.² Although Maraj et al¹ stated that baseline QTc > 460ms is considered a risk factor for QTc prolongation, several mutations in cardiac ion channels display low penetrance that results in clinically asymptomatic individuals with normal QT intervals at baseline.³ In the presence of certain drugs, these individuals seem to be more susceptible to QTc prolongation than the rest of the population.

Hydroxychloroquine and azithromycin are known to induce QT prolongation via a human Ether-à-go-go–related (HERG) gene potassium channel blockade. Indeed, multiple genetic variants in KCNH2 – the gene encoding HERG – have been significantly associated with drug-induced QT prolongation and arrhythmias.⁴ Other genetic variants in cardiac ion channels (and their subunits) have been repeatedly shown to be involved in drug-induced QT prolongation (e.g., rs1805128 in KCNE1 and rs79299226 in SCN5A).⁴ Most genetic variants in cardiac ion channels are rare, but some of the high-risk variants are relatively common. For example, rs1805124 in SCN5A has a minor allele frequency of 22% in European, 31% in African, and 10% in East Asian ancestries. Genetic variants of other proteins and solute carrier transporters that are highly expressed in the heart have also been associated with drug-induced QT prolongation in genome-wide association studies (e.g., rs4959235 in SLC22A23, rs62624461 in ACN9, and rs7142881 in NUBPL).⁴

In addition to these cardiac mechanisms, genetic variants can also significantly affect pharmacokinetic mechanisms, i.e., the function of the metabolic enzymes and transporters responsible for the clearance of hydroxychloroquine and azithromycin.⁵ Patients possessing variants in pharmacokinetic genes have significantly higher plasma concentrations of hydroxychloroquine and/or azithromycin than their counterparts, even at similar doses. Individuals with the C allele of rs1135840 in CYP2D6 – encoding a defective metabolic enzyme – had ~30% higher ratio of active metabolite:parent hydroxychloroquine plasma concentrations (allele frequencies 30%−52%).⁵ Individuals with the G allele of rs1045642 in ABCB1 – encoding a drug transporter – had 2-fold higher peak plasma concentrations of azithromycin (allele frequencies 43%−85%).⁵

These examples demonstrate the need for pharmacogenetics to be considered in studies of drug-induced QT prolongation. Since patients’ genetic profiles can be obtained prior to the administration of QT-prolonging drugs, pharmacogenetics has tremendous potential to decrease the risk of drug-induced QT prolongation, not only in patients with COVID-19, but in other conditions as well.