3.2 Chloroquine and hydroxychloroquine
Chloroquine and its derivative, hydroxychloroquine, are widely used as inexpensive and safe anti-malarial drugs. In particular, the established good tolerability of chloroquine/hydroxychloroquine has made them safe to use even in pregnancy (Villegas et al., 2007). In addition to anti-malarial activity, both drugs have immunomodulating effects and are used for the treatment of autoimmune diseases including systemic and discoid lupus erythematosus, psoriatic arthritis, and rheumatoid arthritis. Chloroquine/hydroxychloroquine concentrate extensively in acidic vesicles including the endosomes, Golgi vesicles, and the lysosomes (Ohkuma & Poole, 1981). This leads to lysosomal membrane permeabilisation or dysfunction of several enzymes including acid hydrolases and palmitoyl-protein thioesterase 1 (Rebecca et al., 2019; Savarino, Boelaert, Cassone, Majori, & Cauda, 2003; Schrezenmeier & Dorner, 2020). Although the precise mechanisms of the anti-viral effects are not fully understood, it has been proposed that chloroquine/hydroxychloroquine can prevent virus infection (pre-infection) by interfering with the glycosylation of cellular receptors and impair viral replication by increasing endosomal pH (post-infection) (Savarino et al., 2003; Savarino et al., 2004; Vincent et al., 2005).
Owing to their efficacy against viruses (mostly demonstrated in vitro ) including influenza, HIV, coronavirus OC43, and SARS-CoV, a large number of clinical trials (>160) have been registered worldwide using chloroquine/hydroxychloroquine alone, or in combination with other drugs (e.g. azithromycin) for the treatment of COVID-19. Despite promising in vitro antiviral results for hydroxychloroquine/chloroquine, there is no convincing evidence of efficacy at present (Gao, Tian, & Yang, 2020; Gautret, Lagier, Parola, Hoang, Meddeb, Mailhe, et al., 2020; Gautret, Lagier, Parola, Hoang, Meddeb, Sevestre, et al., 2020; J. Liu et al., 2020; Magagnoli, 2020; Mathian et al., 2020; Million et al., 2020; Tang, 2020; Yao et al., 2020). A post-exposure prophylaxis randomised controlled trial of 821 participants failed to show any benefit of hydroxychloroquine (n=414) compared with placebo (n=407) (Boulware et al., 2020). At the time of writing, the RECOVERY trial (clinical trial identifier NCT04381936) which is the largest randomised control trial so far conducted for the treatment of COVID, has stopped recruiting to the hydroxychloroquine arm (1542 patients compared with 3132 on standard care) because of no beneficial effect either in terms of mortality or hospital stay (Horby & Landray, 2020). There are still many other trials on-going testing the efficacy of hydroxychloroquine for either prophylaxis or treatment.
Both chloroquine and hydroxychloroquine have been in clinical use for many years for rheumatoid diseases, and thus their safety profile is well established. Dose-dependent retinal toxicity has long been recognized as the major AE with long-term use of chloroquine/hydroxychloroquine (Marmor et al., 2011). Besides retinal toxicity, gastrointestinal, liver and renal toxicity have also been reported (Giner Galvan, Oltra, Rueda, Esteban, & Redon, 2007; Michaelides, Stover, Francis, & Weleber, 2011; Mittal, Zhang, Feng, & Werth, 2018). As both drugs are mainly metabolised in the liver and excreted by renal clearance, their use in patients with liver or renal impairment may worsen the function of these organs. For chloroquine treatment, prescribing information recommends the full dose at all degrees of renal impairment but suggests that monitoring of renal function may be useful (Sanofi-Aventis, 2017 ). For hydroxychloroquine, reductions in dosage are advised for patients with impaired renal function, as well as those taking concomitant medications with known risks of kidney damage (Concordia Pharmaceuticals Inc, 2017).
A serious AE associated with chloroquine/hydroxychloroquine is cardiotoxicity, which can take many forms including cardiomyopathy in rare instances. Prolonged treatment or high dosage of chloroquine/hydroxychloroquine has been shown to increase of the risk of QT interval prolongation, polymorphic ventricular tachycardia, and sudden cardiac death (Chatre, Roubille, Vernhet, Jorgensen, & Pers, 2018). A large epidemiological analysis in patients with rheumatoid arthritis has recently shown that 30-day cardiovascular mortality was increased by more than 2-fold when hydroxychloroquine was combined with azithromycin. The lethal ventricular arrhythmias are primarily due to inhibition of a potassium channel (the inward rectifier Kir2.1 channel) and may occur at low µM concentrations (IC50=8.7 µM) (Rodriguez-Menchaca et al., 2008). While therapeutic doses of chloroquine typically result in plasma concentrations of 2-5 µM, much higher concentrations in the heart are expected based on a 400-fold increase observed in rat PK studies (McChesney, Banks, & Fabian, 1967; Walker, Dawodu, Adeyokunnu, Salako, & Alvan, 1983). The binding of chloroquine to the inward rectifier Kir2.1 channel can be stabilized by negatively charged and aromatic amino acids (Rodriguez-Menchaca et al., 2008). The binding of chloroquine/hydroxychloroquine to proteins is also stereoselective, but whether one of the chloroquine/hydroxychloroquine enantiomers has a stronger interaction with the Kir2.1 channel is not known. Caution is needed when hydroxychloroquine is used in combination with other drugs (including azithromycin), which increase the QT interval because of a pharmacodynamic synergistic interaction.
Given the comorbidities in many patients with COVID-19, especially those with underlying cardiovascular disease, and the fact that COVID-19 itself is associated with cardiac manifestations, this may increase the risk of cardiotoxicity associated with the use of chloroquine/hydroxychloroquine. Indeed, excessive QTc prolongation was observed in 36 % of patients as reported by Bessiere at al. and greater QTc prolongation was also seen in patients taking the combination of hydroxychloroquine and azithromycin than those taking hydroxychloroquine alone, highlighting the importance of pharmacodynamic interactions (Bessiere et al., 2020; Mercuro et al., 2020). Furthermore, a phase IIb trial in Brazil showed that a higher dose of chloroquine (600 mg twice daily) in patients hospitalised with COVID-19 had a higher fatality rate (30 %) compared with 15 % in the lower dose (450 mg twice daily) group (Borba et al., 2020). QTc interval prolongation >500 msec was observed in 19 % of the high dose group compared with 11 % of the low dose group. The US prophylaxis randomised control trial however did not show any increase in cardiovascular AEs (Boulware et al., 2020). We await the publication of the RECOVERY trial to determine whether there was an excess of cardiovascular events. However, it is important to note that despite the size of the RECOVERY trial (n = 1542 patients), it may still be under-powered to identify an excess number of cardiovascular events when compared with standard of care.