A new study likely makes the best estimate yet of the degree of retinopathy risk that patients who take the antimalarial drug hydroxychloroquine (HCQ) can expect, deriving mainly from the cumulative dose taken during the first 5 years of use, according to a study published today in Annals of Internal Medicine.
HCQ works to decrease activity in a patient’s immune system, which is effective in many cases of systemic lupus erythematosus, one of the most common indications for the drug. However, an adverse outcome of treatment can be HCQ retinopathy, a progressive form of vision loss in patients taking HCQ over an extended period (mostly for longer than 5 years). The disease is often asymptomatic, although some patients do present a paracentral scotoma and a decrease in color vision. Patients may also notice flashing shapes in their vision and find that they have difficulty reading. Eventually, HCQ retinopathy can lead to loss of visual acuity, loss of peripheral vision, and loss of night vision.
Researchers from Kaiser Permanente Northern California and Harvard Medical School analyzed 3325 persons who received HCQ for 5 or more years between 2004 and 2020. Their goal was to both characterize the long-term risk for incident HCQ retinopathy and examine the degree to which average HCQ dose within the first 5 years of treatment serves as a prediction of the risk.
The researchers then estimated the risk for developing retinopathy after 15 years, according to patients’ average dosing levels during the first 5 years of therapy. Overall, 81 participants developed HCQ retinopathy with overall cumulative incidences of 2.5% after 10 years and 8.6% after 15 years; the risk was greater for those given a higher dose during the first 5 years of treatment.
The mechanism of how HCQ toxicity may occur is still not completely known. There is evidence that toxicity happens because HCQ binds to melanin in both the retinal pigment epithelium and uvea in high concentrations. HCQ can interfere with lysosomal function, leading to oxidation and accumulation of lysosomes, which can cause dysfunction of the retinal pigment epithelium.
Progressive retinopathy can continue even after the drug is stopped. “It’s thought to be a very mild but important risk,” said Nilanjana Bose, MD, MBA, a rheumatologist with Memorial Hermann Health System in Houston, Texas. “Patients taking HCQ must be screened for retinal issues, most certainly elderly patients and patients with any kind of comorbidities.”
A 2021 joint position statement from the American College of Rheumatology, American Academy of Dermatology, the Rheumatologic Dermatology Society, and the American Academy of Ophthalmology recommends a baseline eye exam within a few months after starting therapy, then additional screening at 5 years on HCQ and annually thereafter.
“Early detection of retinopathy is important in overall visual prognosis, because toxicity can continue even after discontinuation of the medication,” says Rukhsana G. Mirza, MD, professor of ophthalmology and medical education at Northwestern University Feinberg School of Medicine in Chicago. “Examination alone is not sufficient to evaluate early changes, and specialized testing must be done. These include color photos, visual field tests, optical coherence tomography, fundus autofluorescence and in some cases, multifocal electroretinogram. Also, the AAO [American Academy of Ophthalmology] has specific recommendations related to Asian patients as they may have a different pattern of retinopathy that must also be considered.”
More Accurate Risk Measurements
Medscape Medical News asked study co-author April Jorge, MD, assistant professor of medicine in the division of rheumatology, allergy, and immunology at Massachusetts General Hospital and Harvard Medical School, to discuss the study, how it correlates to past research, and what it adds that’s new and useful to rheumatologists and ophthalmologists:
Medscape: Your research found that a higher dose of HCQ in the first 5 years of treatment led to a greater risk of retinopathy. Is there any indication that a lower dose given more frequently, either within that 5-year period or longer, would pose a similar risk?
Jorge: In our study, we assessed the HCQ dose in the first 5 years of use but followed patients who continued the medication longer than 5 years, through up to 15 years of use. Therefore, we compared the risk of HCQ retinopathy associated with different HCQ dosages but for the same duration of use. We found that for any dose of HCQ, the risk of retinopathy increases the longer the medication is used. However, patients who used a higher dose of HCQ had a higher risk of developing retinopathy over time.
Although current guidelines recommend avoiding any HCQ dose over 5 mg/kg/day to reduce the risk of retinopathy, we found a higher risk of retinopathy associated with dosing over 6 mg/kg/day than between 5-6 mg/kg/day and the lowest risk with dosing under 5 mg/kg/day.
Medscape: How does your study align with and/or expand upon previous research regarding HCQ risk?
Jorge: An important prior study of hydroxychloroquine retinopathy was the 2014 study by Ronald Melles, MD, and Michael Marmor, MD, published in JAMA Ophthalmology. Prior to our present study, that was the largest study to use the modern screening method (optical coherence tomography) to detect HCQ retinopathy. That screening tool is more sensitive than older methods, so it can detect early/mild cases of retinopathy that are typically asymptomatic. Compared to older studies, that 2014 study found a much higher risk of HCQ retinopathy than was previously appreciated.
However, that 2014 study did have some key limitations that could affect the risk estimates, such as using prevalent cases. A key feature of our present study is that we took several important steps to generate more accurate risk estimates. This included using an incident user cohort and detecting incident retinopathy cases through serial review of optical coherence tomography (screening) studies.
To achieve a high degree of methodologic rigor in correctly identifying retinopathy outcomes, we had expert ophthalmologists perform masked adjudication of all screening studies, and we assessed the intra-rater reliability of these study interpretations. Therefore, our study adds to the literature more accurate estimates of retinopathy risk. We found a lower cumulative incidence of retinopathy than was identified in the 2014 study, but the risk is still noteworthy.
Also unique to our study, we graded the severity of HCQ retinopathy outcomes. This was important, as we found that the majority of retinopathy cases detected through routine screening are mild and presumed to be asymptomatic. This will likely be reassuring news for patients that we can screen for this adverse event to detect it early and prevent vision loss.
Another important difference was that we assessed the risk of retinopathy associated with using over 6 mg/kg/day, between 5-6 mg/kg/day, and less than 5 mg/kg/day, whereas the highest dosing group assessed in the 2014 study included all patients using over 5 mg/kg/day. The risk was considerably higher in the >6 mg/kg/day group than in the 5-6 mg/kg/day group.
Medscape: How can rheumatologists and ophthalmologists use this new information specifically to better treat their patients?
Jorge: Our study provides more accurate estimates of the risk of HCQ retinopathy than in prior studies. These risk estimates can be used when rheumatologists (and other clinicians who prescribe HCQ) consider the risks and benefits of this otherwise important and well-tolerated medication. The risk associated with different dose ranges could also inform dosing decisions, since dosing over 6 mg/kg/day may be more of a concern than using doses in the 5-6 mg/kg range. Ophthalmologists can also use these new risk estimates to counsel patients of the importance of HCQ retinopathy screening and can also hopefully provide some reassurance to patients that the risk of severe retinopathy is low as long as they are being monitored.
The study authors were supported by grants from the National Institute of Arthritis and Musculoskeletal and Skin Diseases and the Rheumatology Research Foundation. The authors report no relevant financial relationships.
Ann Intern Med. Published online January 17, 2023. Abstract