Background/Purpose: Polymyalgia rheumatica (PMR) is an inflammatory condition that typically affects older individuals for whom long term glucocorticoid (GC) use may be undesirable. Moreover, the comparative effectiveness of sarilumab (SAR) as the only FDA-approved treatment for PMR vs. conventional immunomodulators like methotrexate (MTX) is unclear. We examined the proportion of patients able to discontinue GCs on SAR and evaluate the effectiveness of SAR vs. MTX to achieve GC discontinuation.

Methods: We used augmented commercial claims data from Komodo HealthMap to identify patients with PMR aged ≥50 with ≥2 PMR diagnoses initiating sarilumab as a second line (2L) or third line (3L) therapy following FDA approval 02/2023, requiring a 12 month baseline with full and continuous coverage during follow-up. We descriptively characterized the proportion of patients able to discontinue GCs ( >60 day gap with no GC use) by ≥6 months.  

We then derived two comparator cohorts of MTX (2L) and leflunomide (3L) users that were propensity score (PS) matched 1:1 on age, sex, race/ethnicity, insurance type, baseline GC use and dose, time since first PMR diagnosis, and comorbidities. Standardized mean differences (SMDs) ≤ 0.10 and p values < 0.05 were used to describe differences between cohorts between SAR vs. MTX (2L) and SAR vs. LEF (3L) cohorts. Following confirmation of appropriate balance between the SAR and MTX/LEF cohorts, we described the proportion able to discontinue GCs within 6 and 9 months.

Results: After 02/2023, 235 patients initiated sarilumab, of which 107 had at least 6 months follow-up. Of these, 47.7% (n=51) were able to discontinue GCs within 6 months, with an increasing proportion able to discontinue with additional follow-up time (e.g. 71.4% by 10 months).  After deriving a comparable cohort of patients initiating SAR vs. MTX for 2L treatment, a total of 210 patients were matched. Characteristics of SAR patients: mean (SD) age 67.0 (7.7) years, 56.2% female, 66.7% white, 54.2% commercially insured, mean (SD) starting prednisone dose was 8.4 (5.6) mg/day, and median of 237.0 days from first PMR diagnosis. Based on SMDs and p-values, the 2L SAR-MTX cohorts were well balanced. After restricting to patients with at least 6 months of follow-up with full coverage (n=68 total), SAR and MTX cohort characteristics were similar.

Among these individuals, the proportion of patients using SAR who were able to discontinue GCs within 6 months was numerically greater (59%) than for MTX users (35%). SAR patients required fewer median [IQR] days on GCs (SAR: 122 [61, 195] days vs. MTX: 227 [131,287]) and a numerically lower median (IQR) cumulative dose of GCs: 679mg (440, 1,005) vs. MTX: 867mg (523, 1,248). The 3L SAR vs. LEF cohorts were not well balanced and not further analyzed

Conclusion: Based on these preliminary results derived from large health plan claims data, and consistent with previously published results in an older Medicare-eligible PMR population examining off-label IL-6Ri use, sarilumab appears more effective than MTX as a GC-sparing agent. Analyses are ongoing to augment the sample size and confirm the robustness of these findings.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/comparative-effectiveness-of-sarilumab-vs-methotrexate-as-a-glucocorticoid-sparing-agent-in-patients-with-polymyalgia-rheumatica/