PO:35:228 | Polymyalgia rheumatica and giant cell arteritis induced by immune checkpoint inhibitors for cancer treatment: a systematic review and meta-analysis highlighting differences from the idiopathic forms

Elvis Hysa¹, Andrea Casabella², Emanuele Gotelli¹, Rosanna Campitiello¹, Carlo Genova³, Enrica Teresa Tanda⁴, Carmen Pizzorni¹, Alberto Sulli¹, Vanessa Smith⁵, Marco Amedeo Cimmino⁶, Sabrina Paolino¹, Maurizio Cutolo¹. | ¹Laboratory of Experimental Rheumatology and Academic Division of Clinical Rheumatology, Department of Internal Medicine, Genova, Italy; ²IRCCS Ospedale Policlinico San Martino, Genova, Italy; ³Academic Oncology Unit; IRCCS Ospedale Policlinico San Martino, Genova, Italy; ⁴UOC Medical Oncology Clinic 2, Department of Internal Medicine, University of Genova, Genova, Italy; ⁵Department of Internal Medicine, Ghent University, Department of Rheumatology, Ghent University Hospital, Ghent, Belgium; ⁶Rheumatology Outpatient Clinic, Villa Ravenna Chiavari Italy.

Background. Immune checkpoint inhibitors (ICIs), widely used in cancer immunotherapy, can trigger immune-related adverse events, including polymyalgia rheumatica (PMR) and giant cell arteritis (GCA), likely through mechanisms involving altered immune tolerance. We conducted a systematic review (SLR) and meta-analysis to characterize these conditions and explore potential differences from the idiopathic forms.

Materials and Methods. The SLR analyzed Medline and EMBASE databases up to July 2024, comparing ICI-induced PMR and GCA to their primary forms. Where direct comparisons with idiopathic forms were unavailable, we referred to data from prior meta-analyses and large observational cohorts to qualitatively explore potential differences.

Results. From 1,237 abstracts, 46 were included, yielding 358 patients (314 with ICI-PMR and 44 with ICI-GCA). ICI-PMR had a pooled prevalence of 0.3% [95% CI: 0.1%–1.2%] among ICI recipients. Patients were predominantly males (64% [95% CI: 54%–73%]), with a mean age of 71 years [95% CI: 68–74]. PD1/PDL1 blockers were used in 85% [95% CI: 80%–89%] of cases. Inflammatory pain in the girdles was universal (100%), however pelvic girdle involvement was explicitly reported in only 3 studies. Peripheral arthritis in ICI-PMR was present in 26% [95% CI: 9%–54%], and normal inflammatory markers were detected in 26% [95% CI: 15%–40%]. Glucocorticoids (GCs) completely improved symptoms in 83% of patients [95% CI: 66%–92%], with 13% [95% CI: 12%–34%] requiring DMARDs and 18% [95% CI: 9%–33%] experiencing relapses (Figure 1). Two comparative case-control studies showed significant differences between ICI-induced and primary PMR. ICI-PMR patients had milder symptoms, lower C-reactive protein levels, and required lower median prednisone dosages. ICI-GCA prevalence was 0.06% among ICI recipients. Male patients comprised 51% [95% CI: 36%–66%], with a mean age of 71 years [95% CI: 68–74]. About 50% [95% CI: 23%–77%] received anti-CTLA4 blockers (alone or with PD1 blockers), while the rest received PD1/PDL1 blockers. Clinical features included cephalic symptoms (85% [95% CI: 70%–93%]), permanent visual loss (23% [95% CI: 12%–39%]), and large-vessel involvement (62% [95% CI: 40%–80%]). No comparative studies between ICI-GCA vs primary GCA were retrieved. High-dose GCs permitted remission in 95% [95% CI: 73%–99%], though 19% [95% CI: 7%–41%] experienced relapses and 10% [2% - 31%] required DMARDs (Figure 1).

Conclusions. ICI-induced PMR and GCA may differ from idiopathic forms, potentially presenting with milder symptoms and more favorable treatment responses. Large observational data and meta-analyses on primary forms suggest higher relapse rates and prolonged GC use [1,2], yet further robust comparative studies are needed to validate whether ICI-induced syndromes truly represent a distinct and less severe clinical entity.