Viewpoint Ann Rheum Dis: first published as 10.1136/annrheumdis-2020-217480 on 15 April 2020. Downloaded from Festina lente: hydroxychloroquine, COVID-19 and the role of the rheumatologist Elizabeth R Graef,1 Jean W Liew,2 Michael S Putman ,3 Julia F Simard,4,5 Emily Sirotich,6 Francis Berenbaum ,7 Alí Duarte- García,8 Rebecca Grainger,9 Carly Harrison,10 Maximilian F Konig,11 Peter Korsten ,12 Laurie Proulx,13 Dawn P Richards,13 Philip C Robinson,14 Sebastian E Sattui ,15 Manuel Francisco Ugarte- Gil ,16 Kristen J Young,17 Alfred HJ Kim ,18 Jeffrey A Sparks ,19 On behalf of COVID-19 Global Rheumatology Alliance Handling editor Josef S As of the end of March 2020, the COVID-19 left patients with rheumatic disease who require Smolen pandemic has resulted in over 850 000 confirmed HCQ to prevent morbidity and mortality in a 1 For numbered affiliations see cases and an estimated 42 000 deaths worldwide. uniquely vulnerable position. Moreover, we should end of article. All agree that safe and effective therapies for treat- note that poor outcomes in the rheumatic disease ment and prevention are urgently needed. In the community disproportionately affect populations 11 Correspondence to midst of this rapidly progressing crisis, evidence has already facing barriers to healthcare access. For Dr Jeffrey A Sparks, Division of emerged suggesting that antimalarial medications, example, patients living in rural areas or those with Rheumatology, Inflammation, such as hydroxychloroquine (HCQ), may be effica- lower socioeconomic status may not have the addi- and Immunity, Department of Medicine, Brigham and cious for COVID-19 treatment. After amplification tional resources needed to obtain HCQ if not avail- Women’s Hospital, Boston, MA from politicians, news outlets and social media, a able at their local pharmacy. Thus, disadvantaged 02115, USA; rush to acquire supplies of HCQ resulted in world- patients and those who are underinsured/uninsured jsparks@ bwh. harvard. edu wide shortages. Recent government policies may are most susceptible to detrimental outcomes as Dr Alfred HJ Kim; have exacerbated these issues, where wider use in an unintended consequence of COVID-19- related akim@ wustl. edu both COVID-19 treatment and prevention were HCQ shortages. Instead of prioritising access based ERG and JWL contributed authorised or recommended by India, the US Food on demonstrated need or an adequate assessment equally. and Drug Administration and other countries.2–4 In of risk, empiric use in the pandemic has occurred AHK and JAS contributed response to dwindling supplies, several US states without protecting a sufficient supply for continued equally. have issued restrictions on HCQ use including use in people with rheumatic disease. ERG and JWL are joint first limiting dispensation quantities and verifying indi- authors. cations.5–8 Rheumatologists, researchers and patient AHK and JAS are joint senior partners must advocate for the appropriate distribu- MAXIMISE UTILITY authors. tion and use of HCQ, as millions of people with Decades of research support the use of HCQ in rheumatic diseases. As a particularly notable Received 1 April 2020 rheumatic diseases worldwide depend on HCQ to Accepted 8 April 2020 control disease activity and maintain quality of life. example, HCQ use in systemic lupus erythe- http://ard.bmj.com/ 12 Published Online First In doing so, we must also remind ourselves to ‘make matosus (SLE) reduces disease activity, limits 13 14 15 April 2020 haste slowly’ (festina lente). damage accrual and improves survival. In a Emanuel et al9 published a well- timed commen- well- designed placebo-controlled randomised trial, tary suggesting the following principles for fairly discontinuation of HCQ in patients with stable SLE 15 allocating scarce resources during the COVID-19 increased the frequency of disease flares. HCQ is also approved by multiple agencies as a treatment crisis: equal treatment, attempts to maximise bene- on October 4, 2021 by guest. Protected copyright. fits and prioritising the most vulnerable. These option for rheumatoid arthritis (RA), where meta- recommendations echo prior guidance published in analyses suggest unique benefits to improve cardio- 2016 by the WHO on how to address future infec- vascular risk, the major cause of excess mortality in 16 tious disease outbreaks.10 The report cautioned that these patients. ‘special attention should be given to ensuring that The evidence supporting HCQ use in COVID-19 persons who face heightened susceptibility to harm is less compelling. In a small non- randomised pilot or injustice during infectious disease outbreaks are trial including 26 people with COVID-19 initially able to contribute to decisions about infectious treated with HCQ (and azithromycin (AZM) in an disease outbreak planning and response’. This additional subset), nasopharyngeal viral clearance ethical framework offers health systems a structure at day 6 of treatment was reported in 57% of the 14 for approaching the use and distribution of HCQ patients retained in the analysis who received HCQ © Author(s) (or their 17 employer(s)) 2020. No during the COVID-19 pandemic to minimise poten- monotherapy versus 13% of controls. However, commercial re- use. See rights tial impact on patients with rheumatic disease. concerns about the study design and data analysis and permissions. Published have been previously discussed,18 and a potential by BMJ. signal toward increased harm in patients treated To cite: Graef ER, CONSIDER EQUITY with HCQ cannot be dismissed outright. Since Liew JW, Putman MS, Allowing a ‘first come, first serve’ system during an then, another preprint by Gautret et al described et al. Ann Rheum Dis international infectious disease crisis has likely led a selected cohort of 80 patients with COVID-19.19 2020;79:734–736. to disparities in access to HCQ. These deficits have Rather than lending clarity to the ongoing debate, 734 Graef ER, et al. Ann Rheum Dis 2020;79:734–736. doi:10.1136/annrheumdis-2020-217480 Viewpoint Ann Rheum Dis: first published as 10.1136/annrheumdis-2020-217480 on 15 April 2020. Downloaded from this has created more confusion. Almost all (92%) of COVID-19 and RA and exempting patients with other rheumatic conditions subjects examined had mild disease, and four patients were from utilisation management practices. asymptomatic carriers. Such a study population would seem At this time, ballooning infection rates threaten to overwhelm likely to have favourable clinical outcomes and early viral clear- the capacity of healthcare systems worldwide. Safe and effec- ance, regardless of intervention. The absence of a control group tive therapies for COVID-19 are desperately needed. We, like also precludes analysis of any potential harm related to HCQ’s many physicians, hope well- designed studies will demonstrate experimental use in these patients. HCQ’s efficacy against COVID-19 infection. In the meantime, Two subsequent randomised controlled trials have similarly provisions should be underway to ensure adequate supply for provided insufficient clarification. An open- label randomised all indications, particularly our patients with rheumatic diseases. controlled trial of 30 patients recently reported no significant Until such evidence emerges, as rheumatologists we must advo- differences in viral clearance between a group receiving HCQ (13 cate for individuals in whom the safety and efficacy of HCQ out of 15, 87%) and a group receiving standard of care (14 out is established. Lastly, we recognise that fair resource allocation of 15, 93%).20 On the contrary, another single-centre, double- ultimately depends on separation of responsibilities. While it is blind, randomised parallel-group trial suggested improvements our duty to advocate for our patients, it is also our responsibility in clinical outcomes (fever, cough) and pulmonary CT findings to respect the difficult decisions made by the larger medical and in those receiving HCQ versus placebo and usual care.21 This patient communities in these uncertain times. trial also suffered from inconsistencies in its study protocol, inadequate primary outcome measures, a small sample size and Author affiliations 1 inappropriate statistical methodology (eg, since power calcula- Division of Rheumatology and Clinical Immunology, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA tions to justify the reported sample size were not performed, the 2Department of Medicine, Division of Rheumatology, University of Washington, p-values are difficult to interpret). Seattle, WA, United States Fortunately, more rigorously designed randomised controlled 3Department of Medicine, Division of Rheumatology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA trials with adequate power to assess meaningful outcomes for 4 the prophylaxis and treatment of COVID-19 are underway Health Research & Policy, Division of Epidemiology and Department of Medicine, Division of Immunology & Rheumatology, Stanford School of Medicine, Stanford, (eg, NCT04307693, NCT04323631, NCT04315896). These California, USA trials will also provide safety profiles of these therapies when 5Department of Medicine, Clinical Epidemiology Unit, Sweden used for COVID-19. Both HCQ and AZM are known QTc 6McMaster University, Hamilton, Ontario, Canada 7 prolonging agents. Their combined administration in a novel Department of Rheumatology, Sorbonne Université, Paris, France 8Division of Rheumatology, Mayo Clinic, Rochester, Minnesota, USA disease where