Extracorporeal photopheresis versus systemic immunosuppression for treatment of immune-related adverse events: clinical outcomes from the prospective two-arm PRIA study.
Wein Lisa L, Ertl Carolin C, Ruf Theresa T, Morak Monika M et al.
Immune checkpoint inhibitor-induced immune-related adverse events (irAEs) can be steroid-refractory (sr) or steroid-dependent (sd), requiring second-line therapy. Evidence guiding optimal management of sr/sd-irAEs while preserving antitumor efficacy is scarce. This study compared extracorporeal photopheresis (ECP) with systemic immunosuppressants (IS) for treatment of sr/sd-irAEs. This prospective two-arm study included 46 patients (23 ECP, 23 IS) with 12 distinct types of irAEs across 6 tumor entities, all classified as steroid-refractory or steroid-dependent. Toxicities affected the gastrointestinal tract, skin, lung, musculoskeletal system, and serosal membranes, with up to seven prior irAE treatment lines. Patients received six cycles of ECP or investigator's choice IS over 12 weeks. Longitudinal assessments included clinical irAE outcomes, quality of life (QoL), and tumor response. ECP was more frequently used in patients with multi-toxicity (39% vs 9%; p=0.02) and multi-resistance to prior second-line immunosuppression (39% vs 17%; p=0.19). At week 12, ECP showed a lower cumulative corticosteroid exposure (395 mg vs 1,260 mg prednisolone equivalent; p=0.03), significantly improved QoL (p=0.01), and a numerically higher irAE response rate without statistical significance compared with IS (94% vs 81%; p=0.85). In advanced cutaneous melanoma, ECP was associated with superior overall survival (15 vs 10 months; p=0.02), and longer progression-free survival (9 vs 3 months; p=0.01). No significant safety concerns were observed with ECP; one fatality in the IS group was infection-related. ECP demonstrated clinical outcomes comparable to IS in sr/sd-irAEs, with significantly lower cumulative corticosteroid exposure and improved QoL. Furthermore, ECP was associated with a favorable safety profile and showed no evidence of compromised antitumor activity. A multicenter trial is planned for further investigation. NCT05700565.