Was this a Mistake? HCQ Dose Reductions per AAOS Guidelines Save

In the wake of the 2016 American Academy of Ophthalmology Guidelines, rheumatologists have dutifully reduced the dose of hydroxychloroquine (HCQ) take by patients with systemic lupus erythematosus (SLE) to some number south of 5mg/kg per day. Though I believe the ophthalmologists correctly interpreted a 2014 paper in JAMA Ophthalmology that quoted a much-higher rate of HCQ associated retinopathy than had previously been appreciated, they and subsequent guidelines published in the field of rheumatology lacked a critical piece of information: what actually happens when you do this?

The dam finally broke during ACR Convergence 2022, with multiple abstracts and sessions highlighted the potential costs of this. 

The most notable of these was presented by Jacquelyn Nestor in Abstract #1654, which used a creative case-crossover design to identify associations between HCQ dosing and hospitalizations for SLE flares. I recently profiled a similar paper on my own podcast, which had been published in JAMA as a research letter. This new project extended their prior findings to a critical outcome measure in SLE: hospitalizations for SLE flares. In this abstract they identified a 4-5 fold higher risk of hospitalization for and SLE flare for patients who were on low dose (ie under 5mg/kg/day) as opposed to high dose (i.e., over 5mg/kg/day) HCQ. I would be genuinely surprised to find that any of the expensive biologic agents we use for SLE could overcome such a risk. 

Another abstract with similar findings was Abstract # 0982, which was presented by Jordan Jacquez and evaluated a prospective cohort of 286 patients who had been followed for 10 years. They evaluated the association between average HCQ dose and disease damage over time on the SLICC/ACR damage index (SDI) and with respect to major adverse cardiovascular events (MACE) or stroke. Patients who had lower HCQ doses per year were significantly more likely to develop damage on the SDI, suffer from major adverse cardiovascular events, or experience a stroke. There was no difference with respect to eye toxicity, but longer follow up and a larger sample would likely have been necessary to evaluate this. There was also no information about HCQ dosing with respect to the 5mg/kg per day threshold, but the association between HCQ dose and SLE-badness seems clear. 

Finally, one of my concerns about the increased emphasis on HCQ retinopathy is that side effects – very understandably – scare people.

As a fellow in 2017 who staffed and SLE clinic, I frequently counseled patients about the “new understanding” of HCQ retinopathy risk and encouraged them to decrease their doses of HCQ. Multiple patients responded by asking to be taken off completely, which is understandable, but it made me wonder if pushing fear about HCQ retinopathy might increased nonadherence.

An interesting article from Shivani Garg Abstract # 0344 evaluated HCQ thresholds for adherence using blood levels. Patients who were prescribed an even 400mg daily – which corresponds to 5mg/kg/d for an 80kg patient – were significantly more likely to be therapeutic. I wonder if patients instructed to take wonky doses, like 200mg / 400mg every other day, are less likely to be adherent? More research would be necessary, as this is just speculation at this point, but anything that “rocks the boat” of adherence (scary risks, funky dosing, excessive monitoring) seems likely to me to lower it. 

What is not speculation anymore, in my opinion, is that dose reductions may cause harm. I should emphasize “may,” because HCQ retinopathy is real and for many patients who are on a stable dose of HCQ and in low disease activity, dose reductions may be entirely appropriate. Limiting exposure to medications is always good practice. That said, the risk of reducing HCQ in SLE will be one of the issues I remember most from ACR 2022.