fetching data ... 
Background: Polymyalgia rheumatica (PMR) and giant cell arteritis (GCA) are closely related inflammatory diseases affecting individuals over 50 years of age, with peak incidence in the seventh decade of life. Glucocorticoids (GC) remain the cornerstone of treatment, yet relapses and GC-dependence are frequent. Anti-interleukin-6 receptor (anti-IL6R) therapies, including tocilizumab and sarilumab, have demonstrated efficacy as GC-sparing agents in both GC-dependent PMR and GCA. However, in patients with rheumatoid arthritis, anti-IL6R therapies have been associated with a higher risk of gastrointestinal perforation (GIP) compared to anti-TNF agents. Evidence remains scarce in PMR and GCA, although older age and prolonged glucocorticoid exposure are common characteristics, both recognized as risk factors for GIP.
Objectives: This study aimed to assess the risk of GIP in patients treated with anti-IL6R therapies compared with GC alone.
Methods: We conducted a retrospective cohort study using the French National Healthcare Data System (SNDS), a comprehensive claims database covering nearly the entire French population. Patients were identified based on diagnostic codes for PMR (ICD-10: M35.3) or GCA (ICD-10: M31.5 or M31.6), with initiation of GC-therapy at age ≥50 years between January 1, 2009, and August 31, 2024. Patients who subsequently initiated anti-IL6R therapy (tocilizumab or sarilumab) were retrieved and matched 1:3 to unexposed patients at the anti-IL6R initiation date (index date) based on disease-type (PMR or GCA), sex, age and time since GC-therapy initiation at the index date. Exclusion criteria included inflammatory bowel disease, digestive cancer, and prior gastrointestinal perforation. Patients were followed from index date until GIP, death, or censoring. Event-free survivals were compared using a Cox proportional hazards model stratified by the matched sets to account for the matching between exposed and unexposed patients.
Results: A total of 109347 patients with PMR and 72572 patients with GCA were identified in the SNDS, of whom 3999 received anti-IL6R therapy (tocilizumab or sarilumab) and were matched to 11917 unexposed controls (53 exposed patients were incompletely matched). Baseline characteristics were well balanced between groups (Table 1) except for initial daily GC dose and use of methotrexate. During follow-up, 113 (33 + 80) cases of GIP occurred, corresponding to an incidence rate of 7.12 per 1,000 person-years in the anti-IL6R group versus 4.77 per 1,000 person-years in the GC-only group. The risk of GIP was higher in patients exposed to anti-IL6R agents, with a crude hazard ratio (HR) of 2.03 (95% CI 1.25 to 3.30) and an adjusted HR of 2.10 (95%CI 1.27 to 3.48) after accounting for daily GC dose, NSAID use, and PPI use as time-dependent covariates.
Conclusions: This nationwide study provides the first large-scale evidence regarding the risk of gastrointestinal perforation associated with anti-IL6R therapy in PMR and GCA. Pending final analyses, these findings may help refine the risk–benefit assessment of targeted therapies in routine clinical practice and inform future guideline recommendations.
Patients’ characteristics included in the analysis
| Anti-IL6R group
| GC-only group
|
|
|---|---|---|
| Patients characteristics: | ||
| GCA±PMR, n (%) | 2775 (69.4) | 8253 (69.3) |
| PMR only, n (%) | 1224 (30.6) | 3664 (30.7) |
| Age (years), mean (SD) | 73.1 (8.3) | 74.1 (8.0) |
| Women, n (%) | 2725 (68.1) | 8141 (68.3) |
| Time (months) since GC-start, median (quartiles) | 26 (11 - 62) | 26 (11 - 62) |
| Follow-up (months), median (quartiles) | 11 (5 - 19) | 11 (5 - 23) |
| Investigations (before index date): | ||
| Temporal arteries biopsy, n (%) | 1623 (40.6) | 4556 (38.2) |
| Extracranial artery US, n (%) | 2214 (55.4) | 6067 (50.9) |
| Whole-body PET, n (%) | 2784 (69.6) | 5036 (42.3) |
| Comorbidities (last year before index date): | ||
| Diabetes, n (%) | 707 (17.7) | 2288 (19.2) |
| Heart failure, n (%) | 68 (1.7) | 202 (1.7) |
| Peripheral Vascular Disease, n (%) | 247 (6.2) | 564 (4.7) |
| Cerebrovascular disease, n (%) | 340 (8.5) | 833 (7.0) |
| Chronic pulmonary disease, n (%) | 593 (14.8) | 1837 (15.4) |
| Treatments (last month before index date): | ||
| Methotrexate, n (%) | 917 (22.9) | 1402 (11.8) |
| Daily GC dose (mg), mean (SD) | 19.4 (15.8) | 10.9 (9.7) |
| Cumulative* dose of GC (g), median (quartiles) | 9.8 (6.4 - 15.1) | 8.9 (5.5 - 14.7) |
| Other bDMARD, n (%) | 54 (1.35) | 83 (0.7) |
| NSAID, n (%) | 93 (2.3) | 324 (2.7) |
| PPI, n (%) | 2048 (51.2) | 5687 (47.7) |
| * since GC-initiation until index date |
REFERENCES: NIL.
Acknowledgments: NIL.
Disclosure of Interests: Baptiste Chevet AbbVie, Nowak Emmanuel: None declared, Thierry Marhadour: None declared, Alice Tison: None declared, Baptiste Quere: None declared, Divi Cornec: None declared, Dewi Guellec: None declared, Solene Querellou: None declared, Alain Saraux: None declared, Valerie Devauchelle-Pensec: None declared.