Could uveitis in patients with melanoma being treated with immune checkpoint inhibitors be a good thing?

Uveitis is not a condition that anyone wants to celebrate having. It involves inflammation of the uvea, which consists of the iris, ciliary body, and choroid. The symptoms range from mild eye redness, pain, photophobia, floaters, blurred vision to, in extreme cases, blindness. The causes range from a number of autoimmune diseases to infections to adverse reactions to medications.

But there seems to be a circumstance when uveitis — at least a mild, treatable case of it — might be welcome.

According to research findings presented at the annual meeting of the American Society of Retina Specialists (ASRS) in Montreal last week and earlier this year in the journal Ophthalmology Retina, patients with melanoma who are being treated with immune checkpoint inhibitors who develop uveitis have a better mortality rate than patients who don’t develop uveitis.

Why? Although it will take more research to know for certain, researchers say that one possible explanation for the association between uveitis and improved survival is that uveitis is a clinical marker of immune system activation that has been enabled by the immune checkpoint inhibitors.

The lead author of the brief report in Ophthalmology Retina, Leila Chew, M.D., of the Stein Eye Institute at the University of California, Los Angeles, and her colleagues noted that in their study the higher rates of uveitis were also associated with increased use of the combination of Yervoy (ipilimumab) and Opdivo (nivolumab).

Beyond the intriguing theories explaining why uveitis might be a good sign for melanoma patients, Chew and her colleagues say the findings “may challenge the practice of pausing or discontinuing ICIs [immune checkpoint inhibitors] upon uveitis onset, even at higher grades.” They note that most cases of uveitis can be managed with topical eye drops and periocular and intraocular injections.

As their name suggests, the immune checkpoint inhibitors work by latching on to the proteins on immune system’s T cells that normally function to regulate the immune system, functioning as a kind of “off switch.” By turning the off switch off, the immune checkpoint inhibitors channel the killing power of the T cells so it is direct attumor cells. The immune checkpoint inhibitors include the PD-1 inhibitors, such as Keytruda (pembrolizumab) and Libtayo (cemiplimab); the PD-L1 inhibitors, such as Tecentriq (atezolizumab) and Imfinzi (durvalumab); and the CTLA-4 inhibitors, such as Yervoy and Imjudo (tremelimumab).

The findings presented at the ASRS meeting by lead author Zoe Tsoukas, a student at the University of Miami Miller School of Medicine, and her colleague were based on data in the TriNetX U.S. network database. Tsoukas and her colleagues identified a total of 153,352 adult cancer patients who started treatment with an immune checkpoint inhibitor therapy between 2014 and 2025. They then filtered through the data to identify patients who developed uveitis within a year of starting the treatment. It turned out to be a relatively small number — just 286 patients, which is 0.2%. Even so, when Toukas and her colleagues looked at just the patients with melanoma who had been treated with an immune checkpoint inhibitor, they found that the five-year mortality was 16.7% among those with uveitis compared with 38.6%, and the comparison met the standard for statistical significance.

When they looked at non-melanoma patients, the group with uveitis had a slightly lower five-year mortality rate (45.3% vs. 51.8%), but the difference was not statistically significant. When used the TriNetX data to examine different combinations of immune checkpoint inhibitor treatments. The mortality benefit persisted, they wrote in the abstract presented at ASRS, and it indicated that the “observed mortality benefit in melanoma” was not due to differences in immune checkpoint inhibitor regimens.

UCLA study

The study conducted by Chew and her colleagues was limited to patients treated at the University of California, Los Angeles, Health. They identified 5,966 patients who had been treated with immune checkpoint inhibitors from January 2015 to March 2024. Among those patients, 29 patients (50 eyes) developed uveitis with a median time to onset of 397 days since the first immune checkpoint inhibitor treatment. Patients with melanoma were more likely to develop uveitis, but Chew and her colleagues said once they stratified by treatment, the association of melanoma didn’t hold up and could be explained by treatment with the Yervoy-Opdivo combination. Their analysis did, however, show an association between uveitis and longer survivial. More specifically, patients who developed uveitis had a median survival of 1,008.5 days compared with 444 days for those who did not. The finding held after an analysis to address guarantee-time bias.

“This data supports the hypothesis that ICI-associated uveitis may be a marker of heightened systemic immune activation that coincides with improved oncologic outcomes,” wrote Chew and her colleagues, while also acknowledging a number of limitations of the study, not the least of which is the small number of uveitis patients. Still, they say the findings warrant a prospective study as well as consideration of continuing, not stopping, immune checkpoint inhibitor therapy if uveitis occurs.