key: cord-0992981-zyivs66i
authors: Muller, Romain
title: Systemic toxicity of chloroquine and hydroxychloroquine: prevalence, mechanisms, risk factors, prognostic and screening possibilities
date: 2021-04-24
journal: Rheumatol Int
DOI: 10.1007/s00296-021-04868-6
sha: cb012cbdfb83bc3b871e2ffa5f6762b6dd20af59
doc_id: 992981
cord_uid: zyivs66i

Chloroquine (CQ) and its hydroxylated analog, hydroxychloroquine (HCQ), are 4-aminoquinoline initially used as an antimalarial treatment. CQ and HCQ (4-aminoquinoline, 4-AQ) are today used in rheumatology, especially to treat rheumatoid arthritis and systemic lupus erythematosus. Their mechanism of action revolves around a singular triptych: 4-AQ acts as alkalizing agents, ionized amphiphilic molecules, and by binding to numerous targets. 4-AQ have so pleiotropic and original mechanisms of action, providing them an effect at the heart of the regulation of several physiological functions. However, this broad spectrum of action is also at the origin of various and original side effects, notably a remarkable chronic systemic toxicity. We describe here the 4-AQ-induced lesions on the eye, the heart, muscle, the nerves, the inner ear, and the kidney. We also describe their prevalence, their pathophysiological mechanisms, their risk factors, their potential severity, and the means to detect them early. Most of these side effects are reversible if treatment is stopped promptly. This 4-AQ-induced toxicity must be known to prescribing physicians, to closely monitor its appearance and stop treatment in time if necessary.

Chloroquine (CQ) and its hydroxylated analog, hydroxychloroquine (HCQ), are 4-aminoquinoline initially used as an antimalarial treatment. The fields of application of these molecules have increased considerably in recent years. Indeed, CQ and HCQ (anti-malarial drugs, 4-AQ) are today commonly used in rheumatology [1] [2] [3] , especially for the treatment of systemic lupus erythematosus. Their mechanism of action revolves around a singular triptych: 4-AQ acts as alkalizing agents, in particular in lysosomes, but also as ionized amphiphilic molecules, modifying phospholipidic membranes properties, and finally by binding to numerous targets whose list is constantly increasing. 4-AQ are conventional synthetic Disease-Modifying Anti-Rheumatic Drugs (csDMARDs), which modulate innate and adaptive immunity by acting on several key points of immune regulation (Fig. 1) . Indeed, 4-AQ inhibits the recognition of nucleic acids by TLRs (Toll-Like Receptor), the MHC (Major Histocompatibility Complex) class II-mediated antigen presentation, inflammation-induced cell proliferation, and antiphospholipid antibodies activity, which are major aspects of the pathophysiology of systemic lupus erythematosus for example [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] . But their action is not limited to the regulation of immunity: 4-AQ acts at the heart of the regulation of other primordial physiological functions, such as lipid and glucose metabolism, hemostasis, vasoactivity, and tumor control. This broad spectrum of action could lead to a wide range of undesirable effects, but 4-AQ are overall well-tolerated treatments. Many randomized clinical trials have studied 4-AQ, without highlighting any major risk associated with taking these drugs [14] [15] [16] [17] [18] . The most common side effects were headache and nausea, which could affect around 10% of patients. However, due to their limited size and duration, these studies cannot take into account the rare and late effects of these treatments. Yet, 4-AQ has a cumulative toxic effect on many organs.

Here we describe the systemic toxic effects associated with chronic 4-AQ intake apart from any overdose.

Ocular toxicity is the best known and most studied of the toxicities of 4-AQ. Among the ophthalmic disorders, retinal involvement is both the most frequent and the most likely to have serious functional repercussions [19] .

The toxicity of 4-AQ is characterized by a progressive perifoveolar degeneration of the photoreceptors and cells of the pigment epithelium [20] . This degeneration is usually classified into 4 stages: a first "preclinical" phase corresponding to an eye fundus and normal visual acuity, followed by a "clinical" maculopathy corresponding to anomalies in the eye fundus or acuity visual, then by the stage of Bull's-eye maculopathy and finally blindness.

The exact prevalence of these retinal lesions is difficult to determine. Indeed, the progress made in recent years in screening methods has shown the presence of signs of early-onset retinal toxicity and therefore has led to an increase in its prevalence [21] . Thus, the prevalence of retinal toxicity assessed by visual acuity, fundoscopy, an Amsler grid, and/or color, vision assessments is estimated between 0.5 and 2% [22] . This toxicity is correlated with the duration of use of the treatment [23] . Retinopathy occurs only very rarely in the first 5 years of treatment for doses lower than 6.5 mg/kg/day, and in only 1% in the first 10 years [21] . However, it affects 3.1% of patients treated for more than 20 years. The prevalence of retinal toxicity assessed by more modern and sensitive methods such as spectral-domain optical coherence tomography (SD-OCT) and 10-2 visual field analyser (VFA)is much higher, estimated at 7.5% [24, 25] . More precisely, this toxicity is estimated at less than 2% within the first 10 years of use but at almost 20% after 20 years of use [24] . The daily dose of 4-AQ plays a major role since patients treated with more than 5 mg/kg/day have 5-7 times more toxicity [24, 25] .

The eye is a prime target for 4-AQ toxicity, probably due to the high concentration of this molecule in the ocular tissue [26] . This particularly high concentration could be linked to the ability of 4-AQ to bind to melanin [27] , which is particularly abundant in the retina. The mechanism of vision loss due to the toxicity of 4-AQ has been attributed to changes in the retinal pigment epithelium, which causes accumulation of lipofuscin and loss of photoreceptors [28] . Indeed, the accumulation of lipofuscin is implicated in many degenerative diseases of the retina, in particular in the degeneration of photoreceptors [29] , and 4-AQ inhibit the autophagy of the cells of the pigment epithelium, inhibiting the lysosomal storage mechanisms of lipofuscin, so resulting in its intracellular accumulation [30] . In addition, 4-AQ are alkalizing agents, which inhibit the pH-dependent activity of organic anion transporting polypeptide 1A2 (OATP1A2) [31] . This inhibition prevents the recycling of all-trans-retinol in the cells of the pigment epithelium [32] . This accumulation of all-trans-retinol again leads to an accumulation of lipofuscin. However, several studies have suggested that the initial retinal damage does not occur in the retinal pigment epithelium but in the ganglion cells and that the other retinal layers are only affected until after [33, 34] . Indeed, experimental studies in rats and monkeys have shown degeneration of ganglion cells preceding CQ maculopathy [35] . In addition, patients without any clinical sign of 4-AQ maculopathy, and without any degeneration of the cells of the pigment epithelium have significantly lower retinal nerve fiber layer (RNFL) thickness measurements [36] , indicating early loss of ganglion cells [37] . The mechanism of this toxicity is not precisely known to date. 4-AQ also seems to be able to permeabilize the blood-retinal barrier [38] , and have direct toxicity to photoreceptors [39] , without the mechanism being known.

The main risk factors known to date correspond to the dose of 4-AQ administered. First, the daily dose of 4-AQ administered, related to the patient's weight, is a risk factor [40] . Two different methods of calculation of this weight have been proposed [41] . Ideal body weight (IBW) is the oldest method, based on the fact that 4-AQ accumulates mainly in hydrophilic tissues and little in adipose tissue [42] . A daily dose > 6.5 mg/kg IBW constitutes a major risk factor of retinopathy. Nevertheless, "actual" body weight (ABW) appeared to be more efficient than IBW in a 2014 study [24] , where a daily dose > 5 mg/kg ABW was the main risk factor. This superiority of ABW over IBW was not confirmed in a study carried out in 2016 on 565 patients [43] . Note that these estimates exclude the patient compliance rate. Some data suggest that the determination of plasma HCQ, which is correlated with compliance, is a good predictor of retinal toxicity [44] . The total duration of treatment is also a risk factor, especially after 5 years of treatment [45] . The cumulative dose, dependent on the daily dose and the duration of treatment, is so logically correlated with the occurrence of retinopathy [22] . A cumulative dose greater than 600 g seems to be particularly at risk [46] . Chronic kidney disease and tamoxifen [47] also appear to be risk factors. Older age, female gender, high body mass index, and genetic predisposition are discussed factor risks [21, 44, 48] . The toxicity of HCQ appears to be lower than that of CQ [49] .

4-AQ-induced retinopathy can have significant functional implications. The first consequences on vision are the appearance of a peri-foveal annular scotoma [50] , which gradually spreads, and can lead to significant loss of visual acuity or even blindness [51] . The classic late-stage finding of 4-AQ retinopathy, seen as a ring of retinal depigmentation in the parafoveal region is called Bull's-eye maculopathy [52] . In practice, stopping 4-AQ breaks the progression of lesions in the vast majority of cases, even if some rare cases of irremediable progression are reported [53] . Early stopping of 4-AQ is a major prognostic factor [54] : advanced lesions are irreversible and never regress, whereas cessation of 4-AQ before involvement of the external limiting membrane may be associated with regeneration of the photoreceptor layer and with potential functional visual improvement [55] .

Recognition of retinopathy at an early stage, before any retinal pigment epithelium loss, is essential to prevent loss of visual acuity. The latest recommendations suggest initial screening in the first year of treatment but can be done within 5 years [56] [57] [58] . This first screening must determine the existence of previous retinal lesions. All patients should receive baseline examination including fundus photography. The realization in addition of a systematic SD-OCT is recommended by The Royal College of Ophthalmologists [56] . The rest of the follow-up includes VFA plus SD-OCT. Systematic realization of widefield fundus autofluorescence is recommended by The Royal College of Ophthalmologists [56] , whereas this exam is optional according to the American Academy of Ophthalmology [57] . Established retinal toxicity is defined by two test results with abnormalities typical of 4-AQ retinopathy, and should lead to stop the treatment. Patients with one abnormal test result on retinal imaging but normal visual fields should continue to be monitored annually.

4-AQ accumulates in the cornea [59] , causing vortex keratinopathy [60] . These corneal deposits can be symptomatic and cause haloes and glare. These deposits are generally gradually reversible when the treatment is stopped. Concomitant use of amiodarone could be a risk factor [61] .

Chronic intake of 4-AQ is also associated with a thinning of the choroid, associated with a signal void area on the choriocapillaris in the areas of the retinal pigment epithelium defect [62] . These anomalies are probably secondary to retinal toxicity since only observed in the context of toxic retinopathies, but direct choroid toxicity of 4-AQ is not excluded [63] .

Rare cases of diplopia are reported, induced by the neuromuscular toxicity of CQ [64] .

The cardiac toxicity of 4-AQ is polymorphic [65] . Longterm use of 4-AQ is associated with the development of conduction disorders, structural heart disease, sick sinus syndrome [66] , QT prolongation, elevation of cardiac biomarkers [67] and heart failure [68] .

The exact prevalence of 4-AQ cardiotoxicity is difficult to estimate. Indeed, firstly, this cardiac involvement is polymorphic, secondly, it is influenced by the cumulative dose of 4-AQ, and thirdly the estimates fluctuate between studies.

Thus, QT prolongation is observed in 1-10% of patients under 4-AQ, as highlighted by the numerous trials conducted with these treatments in COVID-19 [69, 70] . This prolongation usually appears between the 3rd and 5th day of treatment [71] . Secondary torsades de pointes are rare [72] , but may occur up to several years after initiation of treatment if it is maintained [73] . Conduction abnormalities have rarely been observed in the context of COVID-19 [69, 70] , but seem more frequent in the context of systemic lupus erythematosus, affecting nearly 16% of patients in a study of 453 patients with a cumulative median dose of 1200 g [74] . The conductive anomalies reported are mainly atrioventricular blocks, followed by bundle branch blocks [75] and appear to be more frequent in the short term than after prolonged treatment [74] . Structural anomalies are estimated at almost 12% of patients in this same study. These are mainly ventricular enlargements, more rarely restrictive heart disease. Ventricular hypokinesias are also reported. Valvular and pulmonary artery wall hypertrophy abnormalities seem very rare.

4-AQ has several effects on the heart. First, 4-AQ alkalizes the cardiomyocyte lysosomes, which cause structural heart disease [76] . Indeed, endomyocardial biopsies reveal an accumulation of polysaccharides in cardiomyocytes, probably due to the inhibition of the activity of alpha-galactosidase A, beta-galactosidase, and arylsulfatase, which is reversible after 4-AQ cessation [77] . The pathophysiology of ventricular thickening induced by 4-AQ therefore closely resembles that of Fabry disease, a genetic deficit in alpha-galactosidase A [78] . Ventricular hypertrophy is common to these two pathologies, and, histologically, only the presence of curvilinear bodies in electron microscopy, specific to the toxicity of 4-AQ can make the difference with fabry disease [79] . 4-AQ also exert a "quinine-like" effect due to its common chemical structure with quinidine drugs [80] . Therefore, they are a class 1a antiarrhythmic drug, which, at a physiological dose, can be responsible for a flattening of electrocardiographic T wave, QT prolongation, and cardiac arrhythmia [81] .

Risk factors for the development of 4-AQ-induced cardiotoxicity include advanced age, female gender, long duration of treatment with a high cumulative dose, and renal failure [76, 82] . The presence of retinopathy and clinical signs of myopathy are also correlated with cardiac involvement [75] . Finally, genetic susceptibility is discussed [83] , and Fabry disease seems to constitute a genetic predisposition to the cardiac toxicity of 4-AQ [84] . Arrhythmia, QT prolongation and conduction disorders are more frequent using CQ versus HCQ [85] , and more frequent at the initiation of treatment than after several years. Other risk factors for the appearance of arrhythmia or conduction disorders include the co-prescription of anti-arrhythmic drugs or treatments that prolong QT [86] .

Prognosis in 4-AQ cardiotoxicity can vary from complete improvement in cardiac function to death or cardiac transplantation [87] . According to two studies published in 2018 using data from 42 [88] and 78 [84] patients, treatment withdrawal resulted in complete recovery of heart function in 20-45% of patients but did not prevent the death of 15-45% of them. Meanwhile, regression of conduction disease appears to be rare.

Performing an electrocardiogram (ECG) before then daily for 5 days after starting 4-AQ is recommended to detect conduction disorder or prolonged QT [89, 90] . However, there is no clinical trial demonstrating the relevance of systematic screening for chronic myocardial toxicity. In the context of chronic treatment, Chatre et al. [75] recommend performing an ECG and an echocardiography every 2 years during treatment for screening cardiac structural and conduction disorders. Tselios et al. [67] suggest the evaluation of cardiac biomarkers (troponin, brain natriuretic peptide: BNP) as a screening test for patients using 4-AQ for longer than 5 years. In case of an abnormality in this screeing, the realization of a heart RMI, endomyocardial biopsy or genetic Fabry disease screening are sometimes necessary to confirm the diagnosis of 4-AQ toxicity.

Neuromuscular toxicity of 4-AQ is polymorphic [91] . 4-AQ may be responsible for bilateral proximal symmetric muscle deficits, and polyneuropathies [92] .

The prevalence of muscle toxicity has been studied in several studies. Casado et al. [93] defined this toxicity by the association of increased creatinine kinase level and a compatible histological pattern in a study of 119 patients, reporting a prevalence of toxic myopathy induced by 4-AQ at almost 10%, corresponding to an annual incidence of 1%. Avina-Zubieta et al. focused on the clinical assessment of this toxicity, that is to say, the occurrence of muscle deficit, affecting approximately 1 in 100 patient-years treatment in their study [94] .

The prevalence of central and peripheral neuronal damage is unknown but appears to be rare [95] .

The mechanism of lesions induced by 4-AQ on rhabdomyocytes seems similar to that of lesions induced on cardiomyocytes. The lysosomal alkalinization and the ensuing enzymatic inhibition causes an accumulation of membrane phospholipids and glycogen and subsequently curvilinear bodies and lamellar structures [96] . The histological aspect can highlight acid phosphatase-positive granules and vacuoles, as in the context of the Pompe disease, a glycogenosis due to deficiency of lysosomal acid α-glucosidase [97] . Analysis by light microscopy is however most often normal [93] . Electron microscopy reveals curvilinear bodies and myeloid bodies, membranous whorls, and autophagic structures. These cytoplasmic complex lipid bodies constitute the characteristic features of antimalarial myopathy. Only ceroid lipofuscinosis, a rare lipid storage disease, can show a similar pattern [98] . CQ also inhibits cytokine production, resulting in fiber digestion and distinctive cytosome formation with curvilinear profiles and lipofuscin on histopathological findings [99] . Alteration of antioxidant enzymes and alteration of regulation of oxidative metabolites such as nitric oxide (NO) also plays an important role in increasing lipid peroxidation in nerve and muscle tissue [100] . In addition, in vitro, 4-AQ has the property of inhibiting neuromuscular transmission, without the mechanism of action being known [101] . Nerve damage appears to be linked to perineural and Schwann cell damage [102] . Nerve biopsies indeed reveal demyelination associated with cytoplasmic inclusions within Schwan cells [103] .

Risk factors for 4-AQ-induced neuromyotoxicity are not well known. Caucasians and kidney failure are likely risk factors [96] . Concomitant use of other myotoxic drugs, such as statins, proton pump inhibitors or corticosteroids has been reported as a possible risk factor [104] . The duration of treatment is probably a risk factor since treatment has been taken for more than a year in the majority of the cases reported [105] . The presence of cardiac toxicity is correlated with the existence of peripheral muscular toxicity [84] . HCQ probably has less muscle toxicity than CQ [106] .

The neuromuscular toxicity of 4-AQ is generally moderate [105] . Only a few serious cases with life-threatening effects are described [107, 108] . Stopping treatment most often allows almost complete recovery [105, 109] .

There is no clinical trial demonstrating the relevance of systematic screening for chronic neuromuscular toxicity. A complete clinical examination and an annual evaluation of muscular enzymes level is recommended by Casado et al., which has the advantage of being non-invasive and simple to perform, despite an imperfect sensitivity [93] .

In addition to acute skin manifestations probably of allergic origin and non-specific, 4-AQ are associated with chronic pruritus [110] and skin pigmentation disorders: both hyperpigmented macules [111] , which predominates on the anterior side of the shins, and depigmentation of the skin, nails and hair [112] .

The prevalence of CQ-induced pruritus has been estimated up to 50% of patients in several studies [113] , while recently the prevalence of HCQ-induced pruritus has been estimated to be less than 10% [114] . The prevalence of HCQ-induced hyperpigmentation can affect between 10 and 20% of the patients according to recent studies [115, 116] . 4-AQ induced hypopigmentation seems to be rare.

Concerning pruritus, CQ can bind to and activate the masrelated G protein-coupled receptors MrgprA3/MrgprX, which can be coupled to PLC-b3 or TRPA1 and so initiate pruritus [117] . The activation of this receptor is the main known mechanism of CQ-induced pruritus and even constitutes an animal model of itch [118] . More, opiate receptor, serotonin receptor, N-methyl-D-aspartate receptors, and gastrin-releasing peptide receptor are probably involved in this iatrogen itch, but exact mechanisms are unknown [119] . 4-AQ-induced hyperpigmentation has been studied through histological studies, describing melanin granules and hemosiderin deposits in the dermis corresponding to bruises [120] . The main physiopathological hypothesis to date is that these hyperpigmentations are initially triggered by real bruising. Indeed, bruises "which do not disappear" are reported, as well as post-traumatic hyperpigmentations [115] . 4-AQ-induced immunosuppression, including decreased phagocytic function, inhibition of lysosomal function, and decreased secretion of pro-inflammatory cytokines, may inhibits clearance of skin deposits from hemosiderins and decreases the elimination of melanin [121] . Accumulation within the dermis of these deposits can locally stimulate melanogenesis, explaining the localized skin hyperpigmentation [120] . This hypothesis is consistent with the fact that the main known risk factor for these hyperpigmented macules is the taking of antiaggregant or anticoagulant treatment [115, 116] . Concerning the depigmentation of integuments, mechanisms of 4-AQ-induced poliosis and vitiligo remain unclear; the main hypothesis is that 4-AQ bind to eumelanin and pheomelanin and so disturb melanogenesis [122] .

Renal failure is associated with skin adverse effect [117] . Itching is more frequent with HCQ than with CQ. Acute pruritus and skin depigmentation exhibit a racial predilection, rather occurring in black subjects [123] . The duration of treatment is correlated with the appearance of hyperpigmented lesions [115] . Thus, their prevalence is multiplied by 3 after 5 years of treatment. Platelet antiaggregants and oral anticoagulants are also predisposing factors to 4-AQinduced pigmentation [115, 116] .

Adverse skin effects are rarely life-threatening [124] . However, these lesions can have a strong aesthetic impact, in particular when they touch the face or the hair. Hair or nail coloring disorders and pruritus completely regress when treatment is stopped [110, 112] . Nevertheless, the regression of hyper and hypo-pigmented macules is only partial after stopping 4-AQ. Q-switched ruby laser could provide regression of lesions [125] .

The patient's skin should be regularly examined [116] . Black patients and those taking treatment which facilitate bruising should be warned of the respective risks of depigmentation and hyperpigmentation of integuments.

4-AQ-induced taste and odor changes are described [126] . Acute and chronic audiovestibular manifestations are also reported [127] including tinnitus, vertigo, hearing loss and deafness [128, 129] . 4-AQ can also cause dysphagia, due to its muscle toxicity [130] .

Prevalence of 4-AQ-induced taste and odor changes is unknown. Acute and transient audiovestibular disorders are reported in 10% of patients when initiating HCQ [131] . Audiovestibular manifestations also correspond to almost 3% of the adverse effects of 4-AQ reported to French pharmacovigilance centers [132] .

The inner ear is an organ rich in melanocytes, the loss of which is responsible for audiovestibular dysfunction [133] . Thus, it has been suggested that, as in the context of skin depigmentation induced by 4-AQ, the binding of 4-AQ to pheomelanin is harmful to the melanocyte and provokes progressive dysfunction of the inner ear [134] . The decrease in melanin could in particular lead to an accumulation of reactive oxygen species toxic to cochlear hair cells [135] . This chronic toxicity could explain that prolonged treatment with CQ leads to progressive impairment of auditory evoked potentials (AEP) [136] . Sudden deafness occurring after years of treatment with 4-AQ could be of ischemic origin. Indeed, several histological studies have reported the destruction of cochlear sensory hair cells, of neuronal population, and support structures, associated with vasoconstriction of the capillaries suggesting an underlying ischemic mechanism [137] . The mechanisms of taste and odor alteration remain unclear.

4-AQ-induced audiovestibular complication can appear from the first administration of 4-AQ [138] , but are more frequent after 1 year of treatment [132] . A prolonged exposure time and high cumulative doses are so probably risk factors [139] . Acoustic trauma could also represent a risk factor [135] . Prognostic 4-AQ-induced taste and odor changes are reversible after interruption of treatment [126] . Acute and mild audiovisual disorders occurring at the start of treatment are often transient without sequelae [131, 140] . Conversely, sudden deafness occurring during 4-AQ treatment, suspected of being of ischemic origin, is most often irreversible [141] , even if rare cases with a favorable outcome after stopping treatment and prescribing corticosteroids are reported [142] . 4-AQinduced chronic and progressive audiovestibular toxicity, suspected of being secondary to the decrease in melanin, appear to be reversible when treatment is stopped early, that is to say when the signs of toxicity appear on the hearing evoked potentials, before any hearing loss on the audiogram [136] .

No screening strategy has proven effective to date. Sudden deafness of vascular origin seems unpredictable. Chronic audiovisual dysfunction can be detected by auditory evoked potentials, which are much more sensitive than the audiogram.

Taking 4-AQ is associated with the onset of chronic renal failure [78, 143] . 4-AQ-induced polyuria is also described [144] .

Prevalence of kidney toxicity is unknown, but seems to be extremely rare since only a few case reports are described.

In case of kidney failure, biopsies reveal renal phospholipidosis, characterized by Zebra bodies and myelin figures at electron microscopy [145, 146] . These aspects are in favor of an accumulation of polysaccharides induced by the alkalinization of lysosomes caused by 4-AQ. Impairment of autophagy flux is also harmful to glomerular and tubular cells [147] . This nephropathy closely resembles that of Fabry disease, all the more so in the event of 4-AQ-induced cardiac toxicity associated [148] [149] [150] . 4-AQ-induced polyuria is multifactorial. On the one hand, 4-AQ downregulates the water channel aquaporin-2 and the Na + -K + -2Cl − -cotransporter (NKCC), causing urinary sodium loss and an increase in diuresis [144] . On the other hand, it is showed that CQ modulates the synthesis of plasma arginine vasopressin [151, 152] and its effect on the kidney [153] , disrupting the urine concentration mechanisms.

Kidney toxicity is associated with prolonged treatment and a cumulative dose of 4-AQ [146] .

Prognostic 4-AQ-induced kidney toxicity appears to be reversible upon discontinuation of 4-AQ [146] .

There is no validated strategy for monitoring the kidney toxicity of 4-AQ. However, given the very progressive onset of kidney failure, annual creatinine testing seems both effective and minimally invasive.

4-AQ have pleiotropic and original mechanisms of action, acting at the heart of the regulation of immunity, lipid and glucose metabolism, hemostasis, vasoactivity, and tumor control. However, this broad spectrum of action is also at the origin of various and original side effects, notably including ocular, cardiac, neuromuscular or cutaneous toxicity ( Table 1) . Most of these side effects are reversible if treatment is stopped early, but can have serious consequences if stopped too late. The pathophysiology and risk factors of these toxicities are becoming better known, allowing identifying the riskiest situations (Table 2) , and proposing risk reduction strategies (Table 3) . For instance, prescribing HCQ instead of CQ seems to reduce retinal, cutaneous, and neuromuscular toxicity, and regular monitoring of muscle enzymes appears to be effective in detecting neuromuscular toxicity. This 4-AQ-induced toxicity most often occurs after chronic treatment and must be known to prescribing physicians. It is important to closely monitor its appearance and, if necessary, to stop the treatment before the benefit-risk balance becomes unfavorable. 

The author(s) declare that they have no competing interests. 

Clinical efficacy and side effects of antimalarials in systemic lupus erythematosus: a systematic review

Treatment of sarcoidosis

EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2016 update

Lysosomotropic drugs enhance pro-inflammatory responses to IL-1β in macrophages by inhibiting internalization of the IL-1 receptor

L'autophagie garante de l'immunité et de l'inflammation-Tout est bien, tout va bien, tout va pour le mieux qu'il soit possible

Antimalarial drugs inhibit phospholipase A2 activation and induction of interleukin 1beta and tumor necrosis factor alpha in macrophages: implications for their mode of action in rheumatoid arthritis

Hydroxychloroquine protects the annexin A5 anticoagulant shield from disruption by antiphospholipid antibodies: evidence for a novel effect for an old antimalarial drug

The ectodomain of toll-like receptor 9 is cleaved to generate a functional receptor

The lysosomal cysteine proteases in MHC class II antigen presentation

PPT1 promotes tumor growth and is the molecular target of chloroquine derivatives in cancer

Immune stimulation mediated by autoantigen binding sites within small nuclear RNAs involves toll-like receptors 7 and 8

Cutting edge: antimalarial drugs inhibit IFN-β production through blockade of cyclic GMP-AMP synthase-DNA interaction

Acid sensing ion channels (ASICS): potential targets for the discovery of novel therapeutics in disease management

Chloroquine for influenza prevention: a randomised, double-blind, placebo controlled trial

Effects of hydroxychloroquine on symptomatic improvement in primary Sjögren syndrome: the JOQUER randomized clinical trial

Comparative study of efficacy and safety of hydroxychloroquine and chloroquine in polymorphic light eruption: a randomized, double-blind, multicentric study

Effects of hydroxychloroquine in patients with cutaneous lupus erythematosus: a multicenter, double-blind, randomized. Parallel-Group Trial Arthritis

Chloroquine as weekly chemoprophylaxis or intermittent treatment to prevent malaria in pregnancy in Malawi: a randomised controlled trial

Hydroxychloroquine and chloroquine: assessing the risk of retinal toxicity

Ocular toxicity of hydroxychloroquine

Hydroxychloroquine retinopathy-implications of research advances for rheumatology care

Rates and predictors of hydroxychloroquine retinal toxicity in patients with rheumatoid arthritis and systemic lupus erythematosus

The incidence of irreversible retinal toxicity in patients treated with hydroxychloroquine: a reappraisal

The risk of toxic retinopathy in patients on long-term hydroxychloroquine therapy

Somatotype, the risk of hydroxychloroquine retinopathy, and safe daily dosing guidelines

Tissue distribution of chloroquine, hydroxychloroquine, and desethylchloroquine in the rat

The melanin binding of drugs and its implications

Different effects of chloroquine and hydroxychloroquine on lysosomal function in cultured retinal pigment epithelial cells

Senescence in the pathogenesis of age-related macular degeneration

Inhibition or stimulation of autophagy affects early formation of lipofuscin-like autofluorescence in the retinal pigment epithelium cell

pH-dependent transport kinetics of the human organic anion-transporting polypeptide 1A2

Retinopathy in mice induced by disrupted all-trans-retinal clearance

Ultrastructural alterations in rat and cat retina and pigment epithelium induced by chloroquine

Histopathologic observations by electron microscopy

Amyloid precursor protein, a beta and amyloid-associated proteins involved in chloroquine retinopathy in rats-immunopathological studies

Retinal nerve fibre layer thickness measurements in patients using chloroquine

A possible early sign of hydroxychloroquine macular toxicity

Chloroquine and hydroxychloroquine increase retinal pigment epithelial layer permeability

Selective thinning of the perifoveal inner retina as an early sign of hydroxychloroquine retinal toxicity

Incidence of hydroxychloroquine retinopathy in 1,207 patients in a large multicenter outpatient practice

Hydroxychloroquine screening alert: change is in the wind

Pharmacokinetics of hydroxychloroquine and chloroquine during treatment of rheumatic diseases

The prevalence of hydroxychloroquine retinopathy and toxic dosing, and the role of the ophthalmologist in reducing both

Hydroxychloroquine blood levels predict hydroxychloroquine retinopathy

Frequency and clinical characteristics of hydroxychloroquine retinopathy in korean patients with rheumatologic diseases

Assessment of the retinal toxicity of hydroxychloroquine

Hydroxychloroquine toxicity and aromatase inhibitors

Panretinal degeneration associated with long-term hydroxychloroquine use and heterozygous USH2A mutation

Comparison of hydroxychloroquine and chloroquine use and the development of retinal toxicity

Scotoma analysis of 10-2 visual field testing with a white target in screening for hydroxychloroquine retinopathy

Intoxication rétinienne sévère aux antipaludéens de synthèse. /data/revues/01815512/00290006/642/ (Internet) (cited 2020 Mar 30

Bull's-eye maculopathy associated with hydroxychloroquine

Cystoid macular oedema and epiretinal membrane formation during progression of chloroquine retinopathy after drug cessation

Effect of disease stage on progression of hydroxychloroquine retinopathy

Progression of hydroxychloroquine toxic effects after drug therapy cessation: new evidence from multimodal imaging

The Royal College of Ophthalmologists (2020) Opthalmologists TRC of clinical guidelines (Internet) (cited 2020 Mar 30

American Academy of Ophthalmology (2016) Recommendations on screening for chloroquine and hydroxychloroquine retinopathy

joint statement on hydroxychloroquine use with respect to retinal toxicity

Chloroquine and hydroxychloroquine toxicity

In vivo confocal microscopy in hydroxychloroquine-induced keratopathy

Amiodarone keratopathy: drug-induced lipid storage disease

Choroidal thinning associated with hydroxychloroquine retinopathy

Toxic effects of hydroxychloroquine on the choroid: evidence from multimodal imaging

Diplopia and loss of accommodation due to chloroquine

Chloroquine cardiomyopathy-a review of the literature

Cumulative dose of hydroxychloroquine is associated with a decrease of resting heart rate in patients with systemic lupus erythematosus: a pilot study

Abnormal cardiac biomarkers in patients with systemic lupus erythematosus and no prior heart disease: a consequence of antimalarials?

Acute left ventricular failure in a patient with hydroxychloroquine-induced cardiomyopathy

Hydroxychloroquine with or without azithromycin in mild-to-moderate Covid-19

Outcomes of 3,737 COVID-19 patients treated with hydroxychloroquine/azithromycin and other regimens in Marseille, France: a retrospective analysis

Predictive factors for cardiac conduction abnormalities with hydroxychloroquine-containing combinations for COVID-19

Effect of hydroxychloroquine in hospitalized patients with COVID-19

Life threatening severe QTc prolongation in patient with systemic lupus erythematosus due to hydroxychloroquine

Electrocardiogram abnormalities related to antimalarials in systemic lupus erythematosus

Cardiac complications attributed to chloroquine and hydroxychloroquine: a systematic review of the literature

Hydroxychloroquine cardiotoxicity presenting as a rapidly evolving biventricular cardiomyopathy: key diagnostic features and literature review

Inhibition of cardiomyocyte lysosomal activity in hydroxychloroquine cardiomyopathy

Chloroquine toxicity misdiagnosed as fabry disease associated with systemic lupus erythematosus and hashimoto thyroiditis

Antimalarial-induced cardiomyopathy resembles fabry disease on cardiac MRI

Cardiotoxicity of antimalarial drugs

Antimalarial drugs: QT prolongation and cardiac arrhythmias

Hydroxychloroquine cardiotoxicity in systemic lupus erythematosus: a report of 2 cases and review of the literature

Hydroxychloroquineinduced cardiomyopathy and heart failure in twins

Heart involvement in a woman treated with hydroxychloroquine for systemic lupus erythematosus revealing fabry disease

Hydroxychloroquine overdose: toxicokinetics and management

Cardiovascular toxicities associated with hydroxychloroquine and azithromycin

Cardiomyopathy related to antimalarial therapy with illustrative case report

Antimalarial-induced cardiomyopathy: a systematic review of the literature

php? specid= 67767 535& typed oc=N

Arrhythmic profile and 24-hour QT interval variability in COVID-19 patients treated with hydroxychloroquine and azithromycin

Hydroxychloroquine neuromyotoxicity. Abstract-Europe PMC (Internet)

Effect of gemcitabine and nab-paclitaxel with or without hydroxychloroquine on patients with advanced pancreatic cancer: a phase 2 randomized clinical trial

Antimalarial myopathy: an underdiagnosed complication? Prospective longitudinal study of 119 patients

Incidence of myopathy in patients treated with antimalarials A report of three cases and a review of the literature

Peripheral neuropathy following chloroquine therapy

Hydroxychloroquine-induced autophagic vacuolar myopathy with mitochondrial abnormalities

Pearls & Oysters: hydroxychloroquine-induced toxic myopathy mimics Pompe disease: critical role of genetic test

Chloroquine-induced cytosomes with curvilinear profiles in muscle

Chloroquine analogues in drug discovery: new directions of uses, mechanisms of actions and toxic manifestations from malaria to multifarious diseases

Hydroxychloroquine-induced oxidative stress on sciatic nerve and muscle tissue of rats: a stereological and biochemical study

The interaction between aminoquinolines and presynaptically acting neurotoxins

Chloroquine neuromyopathies: 4 cases during antimalarial prevention

Peripheral nerve lesions in chloroquine-induced neuromyopathies

Myopathy including polymyositis: a likely class adverse effect of proton pump inhibitors?

Hydroxychloroquine neuromyotoxicity

Chloroquine neuromyotoxicity. Clinical and pathologic perspective

Hydroxychloroquineinduced toxic myopathy causing respiratory failure

Severe hydroxychloroquine myopathy

Antimalarial myopathy in a systemic lupus erythematosus patient with quadriparesis and seizures: a case-based review

A case of hydroxychloroquine induced pruritus

Hydroxychloroquine-induced hyperpigmentation: the staining pattern

Clinical image: blonde by prescription

Chloroquine-induced pruritus in malaria fever: contribution of malaria parasitaemia and the effects of prednisolone, niacin, and their combination, compared with antihistamine

Retrospective analysis of dermatologic adverse events associated with hydroxychloroquine reported to the US Food and Drug Administration

Hydroxychloroquine-induced pigmentation in patients with systemic lupus erythematosus: a case-control study

Hydroxychloroquine-induced hyperpigmentation in systemic diseases: prevalence, clinical features and risk factors: a cross-sectional study of 41 cases

Itching, chloroquine, and malaria: a review of recent molecular and neuroscience advances and their contribution to mechanistic understanding and therapeutics of chronic non-histaminergic pruritus

Characterization of a chloroquine-induced canine model of pruritus and skin inflammation

Peripheral NMDA receptor/NO system blockage inhibits itch responses induced by chloroquine in mice

Drug-and heavy metal-induced hyperpigmentation

Autophagy has a significant role in determining skin color by regulating melanosome degradation in keratinocytes

Chloroquineinduced hair depigmentation

Hydroxychloroquine-related skin discoloration

Fatal toxic epidermal necrolysis associated with hydroxychloroquine

Treatment of nonmelanotic hyperpigmentation with the Q-switched ruby laser

Changes in taste and smell caused by hydroxychloroquine

Hydroxychloroquine ototoxicity in a patient with rheumatoid arthritis

The ototoxicity of chloroquine phosphate

Ototoxicity of chloroquine

A rare cause of dysphagia and weight loss

Effects of hydroxychloroquine on immune activation and disease progression among HIV-infected patients not receiving antiretroviral therapy: a randomized controlled trial

Antimalarial ototoxicity: an underdiagnosed complication? A study of spontaneous reports to the French Pharmacovigilance Network

Ototoxic interaction between noise and pheomelanin: distortion product otoacoustic emissions after acoustical trauma in chloroquine-treated red, black, and albino guinea pigs

Hair cell loss from acoustic trauma in chloroquine-treated red, black and albino guinea pigs

Alterations of auditory evoked potentials during the course of chloroquine treatment

Further observations concerning the toxic effects of streptomycin and quinine on the auditory organ of guinea pigs

Chloroquine ototoxicity: an idiosyncratic phenomenon

Further observations on the use of 4-aminoquinoline compounds in patients with rheumatoid arthritis or related diseases

Quinine-induced hearing loss

Chloroquine ototoxicity: an iatrogenic problem

Chloroquine ototoxicity-a reversible phenomenon?

Chloroquine-induced lipidosis mimicking Fabry disease

Chronic use of chloroquine disrupts the urine concentration mechanism by lowering cAMP levels in the inner medulla

Curvilinear bodies in hydroxychloroquine-induced renal phospholipidosis resembling Fabry disease

Ultrastructural deposits appearing as "zebra bodies" in renal biopsy: Fabry disease?-comparative case reports

The chronic adverse effect of chloroquine on kidney in rats through an autophagy dependent and independent pathways

Hydroxychloroquine-induced renal phospholipidosis resembling Fabry disease in undifferentiated connective tissue disease: a case report

Hydroxychloroquine-induced mimic of renal Fabry disease

Iatrogenic phospholipidosis mimicking Fabry disease

Chloroquine inhibits arginine vasopressin production in isolated rat inner medullary segments induced cAMP collecting duct

Arginine vasopressin mediates the chloroquine induced increase in renal sodium excretion

The effect of chloroquine on renal function and vasopressin secretion: a nitric oxidedependent effect