Hydroxychloroquine in nephrology: current status and future directions

The origin of hydroxychloroquine (HCQ) dates back to the fifteenth century when the medicinal properties of a Peruvian tree bark that was used by the indigenous people to treat fever were discovered by a Jesuit priest [1]. In 1820, quinine was isolated from this so-called “cinchona bark” and the drug soon gained widespread use as a therapeutic and prophylactic agent for malaria [2]. During World War II, soldiers with rheumatic illnesses using quinacrine (a synthetic derivative of quinine) for malaria prophylaxis reported improvement in their symptoms, and this serendipitous discovery paved the way for the use of antimalarials as disease-modifying agents in systemic lupus erythematosus (SLE) and other rheumatological conditions [3]. Antimalarials are now a part of guideline-recommended therapy for patients with SLE and lupus nephritis, in addition to steroids and other immunosuppressive medications [4, 5]. Hydroxychloroquine is the most commonly used drug of its class, when compared to quinacrine, chloroquine, and other quinine derivatives, due to its better tolerability and safety profile. Its use in SLE has been shown to improve remission rates, reduce disease flares, reduce the incidence of cardiovascular and thrombotic events, preserve bone mass, and improve pregnancy outcomes [6].

Despite decades of clinical use, the mechanisms of action of HCQ are still incompletely elucidated. While traditionally believed that HCQ functions by increasing lysosomal pH and thus, inhibiting self-antigen presentation, it is now evident that the immunomodulatory and anti-inflammatory effects of HCQ are much more complex. With a clearer understanding of its action now beginning to emerge, there has been renewed interest in the possible role of HCQ in other kidney diseases. This review aims to provide a comprehensive overview of the mechanism of action of HCQ and the current evidence for its use in lupus nephritis, IgA nephropathy and other kidney diseases. We also discuss the adverse effects of HCQ and the existing recommendations for monitoring drug toxicity.

Hydroxychloroquine (C₁₈H₂₆ClN₃O) is a 4-aminoquinolone and a racemic mixture of R- and S- enantiomers and is available as 200 mg tablets of hydroxychloroquine sulphate, which is equivalent to 155 mg of HCQ base [7, 8]. The drug has an oral bioavailability of approximately 70%, a large volume of distribution (as high as 47,257 L of blood volume) due to uptake by tissues, and a long half-life of 40–50 days [9, 10]. It undergoes N-dealkylation in the liver by CYP3A4 to the active metabolite desethylhydroxychloroquine and two inactive metabolites, desthylchloroquine and bidesethylchloroquine [11]. Concurrent administration of drugs that induce CYP3A4 enzymes, such as rifampicin and phenytoin, may reduce the blood levels of HCQ, while enzyme inhibitors (ketoconazole, diltiazem, clarithromycin etc.) increase levels [11]. Approximately 40–50% is excreted through the kidneys and so, a dose reduction has been recommended in patients with kidney impairment [4].

Hydroxychloroquine is extensively sequestered in melanin-containing tissues such as skin and retina, which explains the retinopathy and skin hyperpigmentation that has been observed with long-term use, and in other sites such as heart, liver, kidney, brain, and muscles [12]. Although HCQ can cross the placenta and is secreted in breast milk, no toxic effects have been observed and the drug is considered safe in pregnant women and during breastfeeding [13].

Lupus nephritis, which occurs in 20–60% of SLE patients over their lifetime, is one of the most severe forms of SLE [33]. It is associated with high morbidity and mortality and is the most common cause of disease-related mortality in SLE [34, 35]. The first evidence of HCQ efficacy in SLE patients with kidney involvement came from the Lupus in Minorities: Nature versus nurture (LUMINA) cohort, a prospective multi-ethnic cohort of SLE patients, in which it was observed that HCQ administration was associated with a lower risk of lupus nephritis [36, 37]. Fessler et al. reported that, in this cohort of 518 patients, the occurrence of nephritis was higher among those not taking HCQ (P < 0.0001) [37]. It was also found that the severity of renal involvement may be attenuated with HCQ, with a lower frequency of Class IV glomerulonephritis among HCQ users, compared to those who were not on HCQ (9.9 vs 33.3%, P = 0. 0003) [38].

Further, in those with lupus nephritis, HCQ may improve renal remission and reduce risk of renal flares. A study of 29 patients with Class V lupus nephritis found higher complete renal remission rates at one year in patients on HCQ, compared to those not on HCQ (64 vs. 22%; P = 0.036) [39]. This finding was confirmed by Mejia-Vilet et al. who reported that complete remission rates were more than twofold higher in Class V nephritis patients treated with antimalarials than among those not on antimalarials (HR 2.46, 95% CI 1.08–5.64; P = 0.032) [40]. Data from the Spanish Society of Rheumatology Registry of Patients with Systemic Lupus Erythematosus (RELESSER) registry, that also included patients with proliferative forms of lupus nephritis, suggested that HCQ users had a 60% higher probability of achieving complete response, compared to non-users (OR 1.61, 95% CI1.10–2.36; P = 0.014) [41]. A retrospective cohort study of 60 patients with lupus nephritis and renal impairment (eGFR < 60 mL/min/1.73 m²) found that renal recovery at 6 months was more likely among those on HCQ (OR 3.891, 95% CI 1.19–12.65; P = 0.024) [42]. More recently, a randomized placebo-controlled trial of 60 children with Class III/IV lupus nephritis treated with mycophenolate mofetil and steroids found that a higher number of patients in the HCQ arm achieved remission (complete or partial) at 12 months, compared to the placebo group (97 vs. 83%, P = 0.002) [43]. The Canadian Hydroxychloroquine Study Group, in a landmark randomized control trial of 47 patients with quiescent lupus, found a reduced risk of renal flares in the HCQ-continuation group, as compared to the HCQ-withdrawal group (RR 0.26; 95% CI 0.03–2.54) at a follow-up of three years [44].

Hydroxychloroquine may also delay CKD progression and reduce risk of end-stage kidney disease (ESKD). Data from the LUMINA cohort suggested that HCQ was associated with a reduction in risk of renal damage, defined as eGFR < 50%, proteinuria ≥ 3.5 g/day lasting for at least 6 months, and/or ESKD (HR 0.12, 95% CI 0.02–0.97; P = 0.0464) [38]. Sisó et al. reported that the proportion of patients with high creatinine values of > 4 mg/dL and ESKD was lower among those who received antimalarials prior to the development of nephritis, as compared to those who did not (2 vs 11%; P = 0.044) and (2 vs 11%; P = 0.029), respectively [45]. Pokroy-Shapira et al. too reported a lower risk of CKD (stage 3 and above) with HCQ use (HR 0.40; 95% CI 0.20–0.90, P = 0.02) in a single-center study in Israel [46]. Antimalarial use was also associated with improved renal survival in a retrospective study of patients with membranous lupus nephritis (P = 0.007) [47]. Further, an analysis of lupus nephritis patients from the Aspreva Lupus Management Study (ALMS) found that lack of treatment with antimalarials was associated with a two-fold higher likelihood of treatment failure, defined by a composite of death, ESKD, sustained doubling of serum creatinine, renal flare and requirement of rescue therapy [48]. Kidney biopsy studies have also demonstrated a lower degree of tubulointerstitial scarring among lupus nephritis patients on HCQ [49]. In contrast, Wu et al., in a retrospective analysis of a nation-wide cohort of SLE patients, reported no significant difference in the risk of developing CKD among those using HCQs for > 90 days, compared to those who used it for < 90 days (HR 1.295, 95% CI 0.40–4.25) [50].

There is also evidence to suggest that antimalarials reduce the risk of infections in patients with lupus nephritis. Sisó et al. reported fewer infections (11 vs. 29%, P = 0.006) in a retrospective cohort study of lupus nephritis patients [45]. Feldman et al. in a study of 33,565 Medicaid beneficiaries with SLE, with a sub-cohort of 7113 lupus nephritis patients, reported a reduced risk of serious infections requiring hospitalization in HCQ users, compared to never-users both in the overall cohort (HR 0.73; 95% CI 0.68–0.77) and in the lupus nephritis cohort (HR 0.78; 95% CI 0.71–0.87) [51]. Similarly, findings from the Grupo Latino Americano De Estudio del Lupus (GLADEL) cohort study (HR 0.69; 95% CI 0.48–0.99; P = 0.044) also support the protective effect of HCQ on serious infections [52].

Additionally, HCQ may be associated with lower mortality rates in patients with lupus nephritis, as reported by Sisó et al. (2 vs 13%, P = 0.029) [45]. A reduction in all-cause mortality has also been reported by Mok et al. (HR 0.58, 95% CI 0.34–0.99; P = 0.048) and Zheng et al. (HR 0.20; 95% CI 0.05–0.82, P = 0.026) [53, 54].

A beneficial effect of HCQ on lipid profile has also been reported, with lower levels of total cholesterol and LDL- cholesterol reported in HCQ users, compared to non-users in a Chinese cohort study [55]. Similarly, in the LUMINA cohort, those on HCQ were found to have lower mean LDL levels (110.4 ± 48.0 vs. 138.8 ± 95.8, P = 0.0155), compared to those not on the drug [38].

A prospective, multicenter study of pregnant women with lupus nephritis found that HCQ users had an 85% reduction in the odds of having a small-for-gestational-age baby (OR 0.15, 95% CI 0.03–0.77; P = 0.0023) [56]. Other benefits including a lower frequency of hypertension (32 vs 50%, P = 0.027) and thrombotic events (5 vs. 17%, P = 0.04) were reported by Sisó et al., although there was no reduction in the risk of malignancy (OR 0.23, 95% CI 0.01–4.30), stroke (OR 1.44, 95% CI 0.46–4.55) or ischemic heart disease (OR 1.93, 95% CI 0.41–9.09) [45].

Defective glycosylation of IgA1 leading to increased circulating levels of galactose-deficient IgA1 (Gd-IgA1) is the key factor in the pathogenesis of IgA nephropathy [57]. This leads to the formation of IgG and IgA autoantibodies against Gd-IgA1, and the immune complexes thus formed deposit in the glomerular mesangium leading to renal injury. It is proposed that HCQ, by its targeting of TLR signaling and reduction of levels of cytokines such as IL-6, IFN-α and TNF-α, may be effective in attenuating renal damage in IgA nephropathy (Fig. 2) [58].

Data from observational studies and a single randomized control trial (RCT) indicate that HCQ may have a role as an adjuvant antiproteinuric therapy for IgA nephropathy in patients who have persistent proteinuria despite optimal supportive therapy. A pilot study of 28 patients with IgA nephropathy and proteinuria of 0.5–2.0 g/day found that those on HCQ (in addition to losartan) were more likely to achieve a 50% reduction of proteinuria at the end of 24 weeks of treatment compared to those on losartan alone (42.9 vs. 14.3%, P = 0.004) [59]. Similarly, Yang et al., in a propensity-score matched retrospective cohort study, reported that more than two-thirds of patients on HCQ had a reduction in proteinuria by 30% or more within 6 months, compared to 45.6% of those on RAS inhibitors alone [60]. Further, the authors noted that this effect on proteinuria reduction persisted at 24 months of follow-up, with no significant decline in eGFR or serious adverse events during this period.

Retrospective data showed similar reductions in proteinuria between patients treated with HCQ and those on corticosteroids, suggesting that HCQ may serve as an alternate treatment for IgA nephropathy in patients wishing to avoid side-effects related to steroid use [61, 62]. Bagchi et al. reported that, of 38 patients with IgA nephropathy and persistent proteinuria of > 1 g/day despite at least 6 months of conservative therapy with RAS inhibitors, 21 achieved remission with the addition of HCQ at the end of a follow-up period of 6 months [63]. Interestingly, the authors also observed a relapse of proteinuria in 20% of the patients on HCQ discontinuation, suggesting that long-term use of HCQ may be needed.

More recently, Liu et al. randomized sixty subjects with IgA nephropathy, all of whom had an eGFR of > 30 ml/min/1.73 m², proteinuria of 0.75–3.5 g/day and were on maximal tolerated doses of RAS inhibitors, to receive either HCQ or placebo, and found a significant reduction in proteinuria in the HCQ group at 6 months, compared to the placebo group (percentage reduction of − 48.4 vs. 10%, P < 0.001) [64]. Half of those in the HCQ group had a 50% proteinuria reduction, compared to 14.8% in the placebo group at 6 months (P = 0.006). At the end of 6 months, the trial was stopped and therefore, data on long-term efficacy of HCQ are lacking [64]. This is especially unfortunate since, given the long half-life of HCQ, it takes roughly 6 months for 96% of steady state levels to be achieved and so it would not be unjustified to expect further reductions in proteinuria on longer follow-up [7]. On the other hand, the longer the duration of HCQ administration, the greater the risk of retinal toxicity and other adverse events. It is also important to keep in mind that proteinuria is only a surrogate outcome and hence, it is still unclear if HCQ use in IgA nephropathy can lead to reduction in ESKD and other hard outcomes.

As with IgA nephropathy, HCQ may also exert an antiproteinuric effect in patients with primary membranous nephropathy. A recent prospective cohort study of 126 patients with phospholipase A2 receptor (PLA2R)-associated membranous nephropathy found that a higher proportion of patients receiving a combination of HCQ and renin angiotensin system inhibitors attained a > 30% proteinuria reduction at 6 months, compared to those on renin angiotensin system inhibitors alone (57.5 vs. 28.9%, P = 0.002). The proportion of patients requiring initiation of immunosuppressive therapy was also lower in the HCQ plus renin angiotensin system inhibitors group (25 vs. 45.8%, P = 0.027). On long-term follow-up, higher rates of clinical remission were noted at two years in the HCQ group (62.5 vs 38.6%, P = 0.013). Importantly, it was found that a greater reduction in anti-PLA2R antibody levels was seen with HCQ therapy, implying that its immunomodulatory effect might be responsible for the observed benefit [65]. Ning et al. recently reported a membranous nephropathy patient with concomitant diabetic nephropathy who was successfully treated with a combination of HCQ and rituximab, although in this case the individual contribution of HCQ to the observed therapeutic effect cannot be ascertained [66]. There is a need for well-conducted RCTs to confirm these findings.

It has been postulated that HCQ may be effective in the treatment of systemic vasculitis based on mechanistic considerations, although evidence remains limited. Casian et al. reported symptomatic improvement in fatigue, joint pain, and cutaneous manifestations in sixteen of twenty-six patients with ANCA-associated vasculitis (AAV) treated with a daily dose of 200 mg of HCQ in a retrospective study. The authors also reported that relapses were less frequent and a third of patients were able to reduce steroid doses [67]. Hydroxychloroquine in ANCA Vasculitis Evaluation (HAVEN) is an ongoing multicenter RCT that will attempt to study the effect of the addition of HCQ to maintenance therapy, compared to placebo on disease activity as assessed by the Birmingham Vasculitis Activity Score (BVAS) [68]. The effect of HCQ on renal outcomes in AAV still needs to be explored.

Hydroxychloroquine is also being studied as a potential therapy for those with X-linked Alport’s syndrome (XLAS). It has been postulated that the benefit of HCQ in XLAS may be mediated by its effect on TLRs, although mechanistic data are lacking [69]. Sun et al., in a retrospective case series, reported a significant reduction in microscopic hematuria and/or proteinuria in eight children with XLAS [69]. A phase 2 single-center study is ongoing in China which will randomize 50 participants with XLAS aged between 3 and 18 years to receive either HCQ at a dose of 6.5 mg/kg/day, in addition to angiotensin-converting enzyme inhibitors or standard care (angiotensin-converting enzyme inhibitors alone) for 6 months [70]. Changes in urinary erythrocyte count, proteinuria, and eGFR will be assessed at week 48 in this study.

Data from preclinical studies also indicate a possible anti-inflammatory effect in anti-GBM nephritis. Torigoe et al. found that in rats with anti-GBM nephritis, administration of HCQ improved renal function, and reduced both proteinuria and microscopic hematuria [71]. Hydroxychloroquine use also resulted in improvement in renal histology, with a significant reduction in fibrinoid necrosis and extracapillary proliferation seen in HCQ-treated rats compared to controls. It was also found that the phosphorylation of Jun N-terminal kinase (JNK) and p38 was significantly lower in rats receiving HCQ. Consequently, the authors postulated that the beneficial effects of HCQ in anti-GBM disease may be mediated through its effect on JNL/p38 mitogen-activated protein kinase (MAPK) signaling [71].

It has been suggested that HCQ may exert a nephroprotective effect in ischemia/reperfusion injury based on animal model studies. Tang et al. found that HCQ reduced renal interstitial infiltration by macrophages and neutrophils, decreased levels of pro-inflammatory cytokines, and mitigated the rise in serum creatinine in mice subjected to ischemia/reperfusion injury, compared to controls [72]. Similarly, Zheng et al., in another mouse model study, found that HCQ inhibited macrophage activation and MAPK signaling via its TLR-9 in a dose-dependent manner and attenuated renal fibrosis [73].

The efficacy of HCQ for the prevention of cardiovascular disease and atherosclerosis in CKD is also under investigation. [74] A phase 2 RCT is currently evaluating the effect of HCQ on atherosclerosis, inflammation, and vascular stiffness in patients with CKD. [75]

The commonly used dose for SLE and lupus nephritis is 200–400 mg/day of HCQ salt (155–310 mg base), given as a single dose or two divided doses. At a dose of 4–5 mg/kg/day, the reported prevalence of retinopathy is < 2% in the first ten years of use, increasing to 20% after 20 years of use [93]. Based on these data, major rheumatological and ophthalmological guidelines have recommended that the optimal daily dose of HCQ be ≤ 5 mg/kg/day (of actual body weight) [4, 94]. Since existing literature on HCQ in lupus nephritis is based on higher doses, the clinical implications of lower doses is yet unknown. Notably, a recent study reported higher flares with an HCQ dose of 5 mg/kg/day or lower, compared to higher doses (adjusted OR 1.98, 95% CI 1.03–3.79); however, the retrospective nature of the study and the lack of data regarding adherence to the drug preclude a definite conclusion [95]. At present, the 2021 KDIGO guidelines suggest an HCQ starting dose of 6.5 mg/kg of ideal (not actual) body weight initially, followed by a maintenance dose of 4–5 mg/kg of ideal body weight [5].

Another area of debate is the need for dose modification for patients with kidney impairment. Although not recommended by drug manufacturers, based on the fact that a significant proportion of the drug is excreted through the kidneys, the KDIGO guidelines recommend a dose reduction of at least 25% for those with an eGFR of < 30 mL/min/1.73 m², while the EULAR/ERA-EDTA guidelines recommend a 50% dose reduction [5, 96].

Fundus examination or color fundus photography, along with spectral-domain optical coherence tomography (SD-OCT) has been recommended at baseline, followed by annual screening after 5 years of use (Fig. 3.) [97]. Patients on higher doses of HCQ, having renal impairment with an eGFR of < 50 mL/min/1.73 m², pre-existing retinal/macular disease or are taking tamoxifen are at higher risk of retinopathy and would require more frequent ophthalmological screening [93, 97] Annual screening should ideally include automated visual field testing, SD-OCT, and widefield fundus autofluorescence (Fig. 1).

There is a growing body of evidence regarding the clinical utility of monitoring blood levels of HCQ. Petri et al. reported that higher whole blood HCQ levels, measured by liquid chromatography-tandem mass spectroscopy, were associated with a higher risk of retinopathy (P = 0.0124) [98]. Studies have also found that the lower the HCQ levels, the higher the disease activity in SLE, with a systematic review reporting a 58% lower risk of active lupus with HCQ levels of ≥ 750 ng/mL [99]. Blood level monitoring may also aid in identifying defaulters, with Costedoat-Chalumeau et al. suggesting a cut-off of < 200 ng/mL for severe non-adherence [100].

Treatment with HCQ (or other equivalent antimalarials) in lupus nephritis is a Class I recommendation as per the KDIGO guidelines and indeed most nephrologists include antimalarials as background therapy for lupus nephritis. However, the existing data for HCQ use in lupus nephritis are mostly observational in nature and based on low-quality evidence (Table 1). It is unlikely, however, that an RCT would ever be conducted since, given the large effect sizes for improved kidney and patient outcomes observed in observational studies and the evidence of increased flares on HCQ withdrawal, such a study would be considered unethical.

A recent systematic review concluded that HCQ was a safe and effective drug for proteinuria reduction, in addition to existing supportive treatment for IgA and could be an alternative, or an add-on treatment for those with inadequate response to immunosuppression [101]. Nevertheless, it would be premature to recommend routine use of HCQ in IgA nephropathy, until additional data from RCTs, or from large observational studies with long-term follow-up are available. Moreover, studies so far have mainly been conducted in the Chinese population, and therefore, lack of generalizability may be a problem. It is also unclear if the use of HCQ can delay progression to ESKD. Research on the use of HCQ for membranous nephropathy, Alport’s disease and other kidney diseases is still in the preliminary stages, and further studies are required to better understand its role in these conditions.

In conclusion, HCQ may have applications in nephrology practice beyond lupus nephritis; however, given the limitations of the evidence included in this narrative review, a broader routine use of HCQ cannot be recommended at present. There is an urgent need for well-designed studies to evaluate its role in other kidney diseases. A more comprehensive understanding of its mechanism of action is also needed. With the emerging role of blood level monitoring, there is reason to hope that the long-term safety of HCQ use may improve. This will not only enable individualized drug dosing but will also provide a tool to ensure drug compliance. However, additional data are needed to elucidate clinically relevant thresholds and frequency of monitoring.