When treatment triggers uveitis

Ocular inflammation triggered by medication is an infrequent but clinically significant finding, and a recent systematic review in Clinical Ophthalmology set out to map which drug classes are most often responsible. Pooling data from 317 published articles covering 690 individual patient cases, the review's authors searched PubMed, Scopus and the Cochrane Library for reports of noninfectious uveitis linked to a specific medication.¹

Study design and patient characteristics

Led by Raghuram Jasti of Drexel University College of Medicine, the research team noted that drug-induced uveitis is thought to represent somewhere between three and five cases per 1,000 seen at specialist referral centres. Across the pooled cohort, patients were, on average, in their mid-fifties at the time of diagnosis, and slightly more than half were women. Inflammation affecting both eyes was documented in nearly two out of three patients, and roughly three-quarters of cases involved anterior-segment disease. From first exposure to the offending drug to the appearance of symptoms took an average of about six and a half months, while resolving the inflammation, once identified and treated, took closer to two months on average.¹

Drug classes most implicated

No single category dominated more than antineoplastic agents, which were linked to nearly three in ten cases. Vaccines and antibiotics followed as the next largest contributors, together accounting for close to another third of the cohort. Rounding out the list, in descending order of frequency, were intraocular pressure-lowering eye drops, bisphosphonates, VEGF inhibitors, antiviral medications and disease-modifying antirheumatic drugs.¹

Digging into the antineoplastic category specifically, immune checkpoint inhibitors were behind just over half of those cases, with combination BRAF/MEK inhibitor regimens and other targeted cancer therapies making up most of the rest. Compared with the overall cohort, uveitis triggered by cancer drugs tended to take considerably longer to resolve than cases caused by anti-VEGF injections, antibiotics or bisphosphonates, and was also more likely to involve both eyes and deeper inflammation (panuveitis) rather than anterior-segment disease alone.¹

Two studies published since the Jasti review add further texture to the checkpoint inhibitor findings. A database analysis spanning more than 120 million patient records found that people being treated for metastatic melanoma face a notably steeper risk of checkpoint inhibitor-associated uveitis than those treated for other malignancies.² A separate cohort of nearly 6,000 patients at a single US centre found that uveitis onset during checkpoint inhibitor therapy was associated with improved overall survival.³

Time to onset and resolution

How quickly uveitis appeared, and how long it took to clear, varied considerably depending on the drug involved. Symptoms surfaced fastest with anti-VEGF injections and vaccines, often within two to four weeks. Antibiotics sat in the middle of the range, while disease-modifying antirheumatic drugs and pressure-lowering eye drops showed the longest lag, with average exposure times stretching beyond 600 days before uveitis developed. A similar pattern held for recovery: bisphosphonates, anti-VEGF agents and antibiotics were tied to the fastest resolution, whereas cases linked to antineoplastics, antirheumatic drugs and antivirals often took several months to clear.¹

Route, laterality and immune status

The review also looked at how a drug was administered and a patient's broader health status in relation to uveitis presentation. Medications taken orally were tied to a markedly higher chance of bilateral disease, while drugs delivered by injection into the muscle, the eye itself, or beneath the conjunctiva were associated with a lower likelihood of both eyes being affected. Patients with a history of cancer were more likely than expected to develop bilateral uveitis, whereas those living with HIV/AIDS were less likely to.¹

Limitations and clinical implications

The authors noted several limitations of working almost entirely from published case reports and small case series, cautioning that this approach may not capture the true scope of drug-induced uveitis in the wider population. Differences in how long individual physicians followed up with patients likely affected the accuracy of reported resolution times, and many of the studies reviewed lacked the detail needed to formally score the likelihood that a given drug caused the uveitis using standard causality assessment tools.¹

With biologic and targeted therapies occupying an ever-larger share of modern treatment regimens, the authors concluded that recognising which drug classes carry the highest inflammatory risk, and how quickly that risk tends to manifest, will be important for optimising patient outcomes. They recommended tighter collaboration between ophthalmologists and the specialists prescribing these therapies, particularly in oncology and rheumatology, to catch and manage ocular complications before they threaten vision.¹

References

1. Jasti R, Wang Z, Zhou L, Nguyen BV, Berkenstock MK. Rates of drug-induced uveitis: a review by medication class. Clin Ophthalmol. 2026;20:564171. doi:10.2147/OPTH.S564171

2. Bellanda V, Schulgit MJ, Hassan KA, et al. Immune checkpoint inhibitor-associated uveitis: insights from comparative analyses across malignancies and drug classes. Am J Ophthalmol. doi:10.1016/j.ajo.2026.05.046

3. Chew L, Feng J, Hutchins E, et al. Immune checkpoint inhibitor-associated uveitis: predictors and survival impact. Ophthalmol Retina. 2026;10(2):215-217. doi:10.1016/j.oret.2025.11.001