The conventional anti-malaria drug HCQ/ CQ was regarded as a promising agent once again for its dual function in inflammation modulation and virus inhibition since the beginning of the pandemic. Although lacking of good clinical evidence, HCQ/CQ was recommended by several countries’ health administrations for COVID-19 treatment [13, 15, 16]. However, in the present study, we found out that HCQ/CQ was related to neither less mortality cases nor shorter VSPs.
In the several past decades, several researchers have confirmed the anti-virus effects of HCQ/ CQ in vitro and in vivo [8, 10, 11]. HCQ/ CQ could prevent the coronavirus from entering the host cells by interfering with endosomal acidification essential for membrane fusion. However, coronavirus could invade the host cells via the alternative non-endosomal pathway which is not blocked by HCQ/ CQ [21]. CQ could also interfere with virus post translation modification by PH modulation[22]. At the meantime, HCQ/ CQ shows potential treatment value for COVID-19 by acting on host cells directly. HCQ/ CQ could inhibit glycosylation of the cell membrane protein angiotensin converting enzyme-2, to which the SARS-Cov-2 is attaching [23]. HCQ/ CQ could downregulate the toll like receptor (TLR) on activated immune cells and block TLR signal transduction, and prohibit several inflammatory factors secretion, such as IL-6[9, 24].
By far, a few clinical studies have analyzed the effects of HCQ/ CQ in COVID-19 treatment. Gautret and colleagues reported that most of patients with COVID-19 were virologicaly cured 6 days after HCQ initiation, especially those being treated in combination with azithromycin[25]. However, Gautret et al’s study had a relatively small sample size and two selection bias. First, the patients in the treatment and control group were not from the same medical center. Second, the virus loads in the HCQ treatment were lower compared to those in the control group without HCQ treatment at inclusion, indicating that the patients in the HCQ treatment group were at a later disease phase and were more likely to get disease selflimitation [26]. In a randomized clinical trial (RCT), Chen and colleagues reported that after being treated by HCQ with a dosage of 400mg per day for 5 days, the clinical and radiological manifestation improve rates were higher in the HCQ treatment group (80.6% vs. 54.8%)[27]. In another randomized study with mild to moderate COVID-19 patients, Tang et al noticed that the rates of negative conversion of SARS-Cov-2 were similar in patients with and without HCQ treatment (85.4% vs. 81.3%)[28].In a retrospective study, Mallat and colleagues reported that HCQ treatment was an independent factor for longer VSPs. The median time from confirmed positive to negative nasopharyngeal swab results were 17 days in the HCQ treatment group and 10 days in the control group (P=0.023). It is worth noting that HCQ was administrated at an early stage of the disease course in Mallat’s study[29].
In our study, the number of mortality cases were not statistically different between patients with and without HCQ/ CQ treatment. The result might be ascribed to several factors. Firstly, HCQ/ CQ was administrated in a later phase of the disease. In some patients, we used HCQ/ CQ due to persistent SARS-Cov-2 RNA positivity for salvage treatment purposes. It is widely accepted that anti-virus should be taken as early as possible in influenza and corona virus infection [5, 30]. Secondly, the half-life of HCQ/ CQ is as long as 40-60 days due to the large distribution volume in the blood. And it usually takes several weeks before HCQ/ CQ reaching maximal activity[31]. In COVID-19, HCQ/ CQ treatment only lasted for an average of 10 days. Therefore, HCQ/ CQ might be withdrew before it worked. Thirdly, for ethic factors concern, several kinds of drugs, such as GCs, ribavirin, TCM et al., were administrated empirically and anecdotally at the same time. These concomitantly taken drugs might have covered up the potential treatment values of HCQ/ CQ on COVID-19. Fourthly, due to the small sample, the statistical power was not able to different the mortality rates in different treatment groups. Fifthly, it might be true that HCQ/ CQ didn’t have any effect on lowering mortality rates in COVID-19.
The average VSPs were similar to those reported in the previous study[32]. After the multivariate linear regression analysis, we identified that disease durations at admission and HCQ/CQ treatment were independent parameters related to patients’ VSPs, indicating patients might get better prognosis if being well treated earlier. Though HCQ/ CQ treatment was an independent parameter for longer VSPs, we should bear in mind that in some cases HCQ/ CQ was administrated for salvage purpose due to persistent SARS-Cov-2 RNA positivity. Furthermore, VSPs were not statistically different between patients with longer VSPs (VSPs>22 days) in these two treatment groups. It was interesting that more patients took GCs with HCQ/ CQ treatment. However, after being adjusted by other confounders, neither GCs treatment nor GCs dosage was an independent parameter for VSPs prediction. Actually, the effect of GCs in COVID-19 remains controversial and disputable. In SARS and Middle East Respiratory Syndrome (MERS), GCs administration is related to delayed virus RNA clearance [33, 34]. However in the SARS or MERS studies were either taking rather high dose of GCs[33] or were critically ill[34]. On the other hand, patients with SARS of influenza might benefit from low-to-moderate GCs[35, 36]. In the present study, our patients took a low-to-moderate dose of GCs during a relative short period of time. As a result, we didn’t find correlations between GCs treatment and prolonged VSPs. A team consist of front-line physicians from the Chinese Thoracic Society suggested that after careful benefits and harms evaluation short term use of low-to-moderate dose of GCs could be prudently administrated in patients with COVID-19[37].
One of the major concerns for HCQ/ CQ treatment in COVID-19 is the side effect. HCQ/ CQ related retinopathy always occurs after months even years of HCQ/ CQ administration[31]. Meanwhile, HCQ/ CQ related arrythmia might be lethal. And the risk is rising together with other arrhythmogenic drugs, such as azithromycin[31]. Borba et al. reported that high dose of CQ (600mg twice daily) was related to prolonged QTc interval and should not be recommended in critically ill patients[38]. Lane and colleagues reported that HCQ monotherapy was safe in COVID-19. However, HCQ in addition to azithromycin might result in heart failure and cardiovascular mortality [39]. Tang et al. found that HCQ was safe in patients with COVID-19, the most common adverse effects were diarrhea and vomiting[28]. Similarly, HCQ was safe and tolerable in our patients. On the contrast, among the three patients with CQ treatment, one patient complained about dizziness and blurred vision and another patient had recurrent first-degree AVB and obvious QTc elongation.
The major limitation of the study was the relatively small sample size. The sample size of the patients without HCQ/ CQ was expected to be 28. However, after age, gender and disease severity matching, only 21 patients without HCQ/ CQ treatment met the matching criteria and were finally selected. Secondly, some patients were treated with HCQ/ CQ for persistent SARS-Cov-2 RNA positivity. These patients, per se, are refractory. Therefore, selection bias exists in our patients. Thirdly, due to the retrospective nature of the study, although we found out that HCQ/ CQ treatment was related to longer VSPs, we couldn’t tell whether HCQ/ CQ prolonged SARS-Cov-2 RNA clearance or not.