Malignant melanoma is often treated with immune checkpoint inhibitors (ICIs) and MAP-kinase pathway inhibitors (MAPKis), but the subsequent development of uveitis is not uncommon. This study classified clinical presentations of drug-induced uveitis to help clinicians identify which drug or drugs are most likely to be the causative agent and determine appropriate modifications to their treatment plan.
Study Design
This observational study used adverse drug reaction data from the French Pharmacovigilance Database to classify features of drug-induced uveitis thought to be associated with the use of ICIs or MAPKis. A multiple correspondence analysis was performed first on the variables of interest (e.g., location of inflammation, associated ophthalmological signs), followed by hierarchical clustering analysis to identify homogeneous subgroups of patients.
Outcomes
Among the 113 cases included, 41% were related to at least 1 ICI and 59% to at least 1 MAPKi (BRAF inhibitors or MEK inhibitors). Melanoma was the only indication for MAPKi use; indications for ICI use included melanoma and, less frequently, bronchopulmonary cancer.
Based on common clinical features, 4 clusters of uveitis cases were identified:
- Cluster 1 (35 cases): Isolated posterior uveitis (100%) with subretinal detachment (86%)
- Cluster 2 (12 cases): Isolated posterior (67%) and intermediate uveitis (75%) with papilledema (42%) and retinal vasculitis (25%)
- Cluster 3 (25 cases): Panuveitis (100%) with iris synechiae (24%), macular edema (20%), papilledema (16%), and subretinal detachment (20%)
- Cluster 4 (41 cases): Isolated anterior uveitis (100%) with iris synechiae (34%)
Different clinical trends were seen based on the suspected responsible drug, with BRAF inhibitors presenting with anterior uveitis and panuveitis, MEK inhibitors with posterior involvement (subretinal detachment) and relatively short time to onset (median: 26 days), and ICIs with anterior uveitis, panuveitis, and posterior uveitis frequently complicated by optic neuropathy and retinal vasculitis.
Subretinal detachments in patients in Cluster 1 tended to resolve spontaneously after drug discontinuation without any steroid treatment. Due to more severe ophthalmic involvement, patients in Clusters 2 and 3 were more likely to be treated with systemic steroids after discontinuation of the suspected drug. Patients in Cluster 4, in most cases, did not require drug discontinuation and were treated only with local steroids.
Limitations
The main limitations of the study are its retrospective nature and its reliance on pharmacovigilance data, which can suffer from underreporting, reporting bias, and inadequate details.
Clinical Significance
The use of ICIs and MAPKis, either alone or in combination, has markedly increased over the recent years. As these drugs can significantly improve the life prognosis of patients with systemic malignancies, their use is expected to continue to increase. This study presents patterns of uveitis associated with each type of drug that may help clinicians determine which medication is most likely responsible for uveitis development and whether discontinuation of that agent, with or without the addition of steroid treatment, is likely necessary for patient improvement.
Financial Disclosures: Dr. Arman Mashayekhi discloses no financial relationships.