Early Tocilizumab Initiation and 12-Month Hospitalisation in Giant Cell Arteritis: A Propensity Matched Real World Cohort Study

Presenter: Krima Patel

This retrospective target trial emulation using the TriNetX US Collaborative Network evaluated whether early tocilizumab (TCZ) initiation reduces hospitalizations compared with GC monotherapy in newly diagnosed giant cell arteritis (GCA)  patients. Eligible patients-initiated treatment within 90 days of diagnosis. After 1:1 propensity score matching, 127 TCZ-treated patients were matched with 127 GC-treated patients (N=254). Median follow-up was 365 days in both cohorts. All-cause hospitalization occurred in 19.7% of TCZ-treated patients vs 30.7% of GC-treated patients, corresponding to an absolute risk reduction of 11.0%, a relative risk of 0.64 (p=0.043), and a 36% relative risk reduction. Mean acute care encounters per patient were significantly lower with TCZ (0.67±2.12 vs 2.23±6.21; p=0.008). Composite vascular events occurred in 11.0% and 11.8% of TCZ- & GC-treated patients, respectively, with no significant difference between grps (HR 0.86; p=0.356).These real-world findings suggest that early TCZ initiation is associated with significantly fewer hospitalizations and acute care encounters than GC monotherapy in GCA, extending the benefits observed in the GiACTA trial beyond relapse-free survival. 

Methotrexate for Remission Maintenance Following Tocilizumab and Glucocorticoids Induction in Giant Cell Arteritis: The Randomized Phase II METEORITICS Trial

Presenter: Valentin Sebastian Schaefer

The phase II METEORITICS trial evaluated whether methotrexate (MTX) could maintain glucocorticoid-free remission following induction with TCZ and glucocorticoids. In this study, 52 patients with GCA in glucocorticoid-free remission were randomized to MTX 17.5 mg/week or placebo for 12 months; 47 patients completed treatment. Fourteen patients (26.9%) experienced at least one relapse, including 6 in the MTX group and 8 in the placebo group (p=0.588). Kaplan-Meier analysis showed no significant difference in time to first relapse. However, only one MTX-treated patient experienced multiple relapses compared with five placebo-treated patients. Total relapses were numerically lower with MTX than placebo (7 vs 18; relative risk 0.39), and cumulative glucocorticoid exposure was lower (910.2 mg vs 1717.8 mg; p=0.401). During weeks 26-52, only one relapse occurred with MTX vs 11 relapses in six placebo-treated patients. MTX dose reduction was required in seven patients. No opportunistic infections or treatment-related serious adverse events were reported. MTX did not significantly prolong time to first relapse. Overall relapses, recurrent relapses, and cumulative glucocorticoid exposure were numerically lower with MTX, but these findings were not statistically significant. No treatment-related serious adverse events were reported

Remission GCA Study – A Two Center Prospective 72 Week Follow Up Study After Standard Tocilizumab Withdrawal in Patients with Giant Cell Arteritis in Imaging Proven Remission

Presenter: Verena Schönau

The prospective, two-center Remission- giant cell arteritis (GCA) study evaluated flare rates following standard tocilizumab (TCZ)  withdrawal in patients with sustained clinical and imaging-confirmed remission, including the absence of active disease on FDG-PET/CT and ultrasound. Forty patients were followed for 72 weeks after TCZ discontinuation. The cohort was 75% female with a mean age of 72 years. Mean time from diagnosis was 2.91 years, mean TCZ treatment duration was 1.91 years, and mean remission duration before study entry was 1.44 years. Baseline inflammatory activity was low, with mean ESR of 6.9 mm/h and CRP of 1.9 mg/L. Despite this deep remission, 22 patients (55%) experienced a flare, while 18 (45%) remained in remission. Only two flares occurred within the first 90 days, whereas most relapses occurred between days 90 and 180 (n=14). Just one flare was documented beyond 365 days. These findings demonstrate that more than half of patients relapse after TCZ withdrawal despite clinical and imaging remission. Neither imaging status nor inflammatory markers predicted flare-free survival, and the clustering of relapses between days 90 and 180 identifies this period as a critical window for monitoring after TCZ discontinuation.

Giant Cell Arteritis with Normal C-Reactive Protein: Clinical Characteristics, Outcomes, and Response to Tocilizumab (The CANON study)

Presenter:Baptiste Chevet

Elevated C-reactive protein (CRP) is a diagnostic hallmark of giant cell arteritis (GCA), yet some patients present with normal CRP despite active disease. This retrospective nationwide multicentre French study compared 74 patients with low-CRP GCA (mean age 72.9 years; 68% women) & 170 patients with elevated CRP to characterize their clinical phenotype, outcomes, & response to tocilizumab (TCZ). Despite CRP ≤10 mg/L, 84% of low-CRP patients fulfilled the 2022 ACR/EULAR classification criteria. Diagnostic yield remained high, with positive temporal artery biopsy findings in 67% and PET-CT evidence of large-vessel involvement in 60%. Compared with high-CRP patients, the low-CRP cohort had significantly lower rates of fever (0% vs 29.4%, p<0.001), fatigue (40.5% vs 72.9%, p<0.001), weight loss (21.6% vs 35.9%, p=0.028), and polymyalgia rheumatica symptoms (24.3% vs 47.1%, p<0.001). Conversely, ophthalmologic manifestations (35.1% vs 22.4%, p=0.037) and ischemic stroke (9.5% vs 2.9%, p=0.049) were more frequent. Relapse-free survival was significantly higher in the low-CRP group (p=0.02). Baseline headache predicted future relapse (90% vs 62.5%, p=0.023), whereas ophthalmologic involvement was associated with a lower relapse risk (15% vs 41.7%, p=0.034). Notably, all 11 TCZ-treated patients (100%) achieved sustained remission without subsequent relapses. These findings identify low-CRP GCA as a distinct phenotype characterized by fewer systemic symptoms but greater ischemic risk. Advanced imaging retained high diagnostic utility, while TCZ demonstrated efficacy, achieving sustained remission in all treated patients (11) and highlighting the responsiveness of localized vascular inflammation to IL-6 receptor blockade despite absent systemic CRP elevation.

Effectiveness of Guselkumab vs Golimumab in Participants with Active Psoriatic Arthritis and Inadequate Response to One Prior Tumor Necrosis Factor Inhibitor: Interim Results at 6 Months in a Phase 3b, Randomized, Open-Label, Pragmatic Trial

Presenter:Alexis Ogdie

Psoriatic arthritis (PsA) patients with an inadequate response (IR) to a first tumor necrosis factor inhibitor (TNFi) may either switch to a different mechanism of action or cycle to another TNFi. The Phase 3b EVOLUTION trial is evaluating these approaches by comparing golimumab (GOL) with guselkumab (GUS) in TNFi-IR patients. This interim analysis included 57 participants randomized to GOL 50 mg every 4 weeks (n=19), GUS 100 mg every 4 weeks (n=17), or GUS 100 mg at Weeks 0 and 4 then every 8 weeks (n=21). At Month 6, achievement of the combined cDAPSA low disease activity/remission and Investigator Global Assessment (IGA) 0/1 endpoint was observed in 26.3% of GOL-treated patients, 23.5% of GUS Q4W patients, and 19.1% of GUS Q8W patients. Mean (SD) cDAPSA improvements were -8.3 (9.5), -9.7 (14.0), and -6.2 (10.1), respectively. Minimal disease activity (MDA) rates were 21.1%, 23.5%, and 9.5%, respectively. Mean improvements in PsAID-12 scores were similar across all treatment groups. At the descriptive 6-month interim analysis, response rates were numerically similar across treatment groups. However, GUS administered every 4 weeks showed numerically greater improvements in cDAPSA and several clinician- and patient-reported outcomes compared with GOL and GUS every 8 weeks. These analyses suggest that both treatment mechanism and dosing frequency may influence outcomes in TNFi-IR PsA, with longer-term efficacy and durability to be assessed at the 12-month primary endpoint.

Cardiovascular Safety of Upadacitinib, Tofacitinib, and Tumor Necrosis Factor Alpha Inhibitors in Rheumatoid and Psoriatic Arthritis

Presenter: Michelle Kasdorf

This TriNetX comparative-effectiveness study evaluated 39,603 patients initiating TNF inhibitors (TNFi; 76.2%), tofacitinib (13.4%), or upadacitinib (10.3%) between 2020 and 2025.Compared with TNFi, tofacitinib was associated with higher risks of major adverse cardiovascular events (MACE; HR 1.65), hospitalized myocardial infarction (HR 2.28), and heart failure (HR 1.35). However, the MACE association attenuated in the intention-to-treat analysis (HR 1.17). Risks were more pronounced before the September 2021 FDA boxed warning (MACE HR 2.73, myocardial infarction HR 3.79) and among biologic-naïve patients (MACE HR 1.96), with no association observed in post-warning or biologic-experienced subgroups. In contrast, upadacitinib showed no association with MACE (HR 0.84), heart failure (HR 0.98), or any sensitivity analysis. Expected increases in herpes zoster risk and no differences in injury or trauma supported adequate adjustment. Overall, tofacitinib, but not upadacitinib, was associated with increased risks of MACE and myocardial infarction, while heart failure and coronary revascularization signals were weaker and analysis dependent. Although consistent with ORAL Surveillance findings, these on-treatment associations were influenced by prescribing patterns and analytical approach, warranting cautious interpretation.

Comparative Risk of Incident Malignancy Among Adults with Rheumatoid Arthritis or Psoriatic Arthritis Initiating Upadacitinib, Tofacitinib, or a Tumor Necrosis Factor Inhibitor

Presenter: Olivia Lenzo

Tofacitinib and upadacitinib are JAK inhibitors approved for rheumatoid arthritis and psoriatic arthritis. Following reports of increased malignancy risk with tofacitinib, this TriNetX comparative safety study evaluated incident invasive malignancy among patients initiating tofacitinib, upadacitinib, or TNF inhibitors (TNFi) between 2020 and 2025. The primary outcome was first invasive malignancy excluding non-melanoma skin cancer, assessed for up to 730 days. In the primary modified intention-to-treat analysis, invasive malignancy risk was higher with tofacitinib vs TNFi (HR 1.22). In contrast, no increased risk was observed for upadacitinib vs TNFi (HR 1.05,) or upadacitinib vs tofacitinib (HR 0.89). In the prespecified as-treated analysis, all associations were attenuated and no longer statistically significant: upadacitinib vs TNFi (HR 1.13), tofacitinib vs TNFi (HR 1.20,), and upadacitinib vs tofacitinib (HR 1.12). Positive-control analyses showed the expected increase in herpes zoster risk with both JAK inhibitors, while negative-control outcomes were near null. The ORAL Surveillance-eligible benchmark cohort had insufficient events for HR estimation. Overall, tofacitinib showed a small increase in invasive malignancy risk vs TNFi in the primary analysis, although this finding was not sustained in sensitivity analyses. No statistically significant increase in malignancy risk was identified with upadacitinib; however, potential channeling bias related to post-boxed-warning prescribing patterns should be considered.

JAK Inhibitors in Giant Cell Arteritis: A Systematic Review and Meta-Analysis Supporting Upadacitinib Efficacy and Glucocorticoid Reduction

Presenter:Mireia Sanmartin Martinez

This PRISMA-based systematic review and meta-analysis evaluated the efficacy and safety of JAK inhibitors in GCA, including upadacitinib, baricitinib, and tofacitinib across four studies involving 274 patients.The pooled overall response rate (ORR) was 75.0% at 6 months (37/50; I²=10.5%) and 62.3% at 12 months (31/50; I²=0%). Pooled complete response rates were 72.4% at 6 months (148/244; I²=66.3%) and 51.3% at 12 months (125/244; I²=0%). In the upadacitinib subgroup, complete response rates were 73.0% and 50.2% at 6 and 12 months. One study reported a mean prednisone-equivalent reduction of 7.98 mg/day (n=14). Herpes zoster incidence was 5.0%, while treatment discontinuations due to adverse events occurred in 13.8%. Overall, JAK inhibitors demonstrated meaningful efficacy, manageable safety, and glucocorticoid-sparing potential in GCA. Evidence was strongest for upadacitinib, although complete-response estimates were heavily influenced by the phase 3 SELECT-GCA trial, and ORR findings from smaller cohorts should be interpreted cautiously.

Real World Effectiveness and Safety of Upadacitinib in Giant Cell Arteritis

Presenter: Fabricio Benavides

This retrospective study evaluated the effectiveness and safety of upadacitinib (UPA) in 63 patients with relapsing giant cell arteritis (GCA), many of whom had previously received methotrexate (49.2%), tocilizumab (TCZ) (63.5%), other biologics, or JAK inhibitors. Clinical remission was achieved in 56.7%, 59.1%, 64.7%, and 81.2 of evaluable patients at 1, 3, 6, and 12 months, respectively. EULAR remission rates were 58.1%, 46.1%, 66.7%, and 64.3% (9/14). Absence of systemic inflammation was observed in 63.6% at 6 months and 50.0% at 12 months. PET/CT improvement occurred in 85.7% and 50.0% at 6 and 12 months, respectively. After a median follow-up of 10 months, 11 patients (17.5%) discontinued UPA, including 3 (4.8%) for lack of response and 7 (11.1%) for severe adverse events: sepsis (n=1), urinary tract infection (n=1), pulmonary thromboembolism (n=1), herpes zoster (n=2), acute myeloid leukemia (n=1), and glioblastoma multiforme (n=1). One patient discontinued due to sustained remission. These findings suggest potential effectiveness of UPA and relative safety in relapsing GCA, including patients refractory to multiple prior therapies. The results provide preliminary real-world data in a more treatment-experienced population but require confirmation in larger prospective cohorts.

Lower Corticosteroid Injection Rates with Secukinumab vs Interleukin-23 Inhibitors: A Real-World Study of Corticosteroid Use Among Patients with Psoriatic Arthritis

Presenter: Tim Nguyen

This real-world retrospective analysis compared CS injection use among patients initiating secukinumab (SEC) vs IL‑23 inhibitors (RIS or GUS). Before matching, 2,070 SEC-treated and 2,577 IL‑23i-treated patients were identified. After propensity-score matching, 1,515 patients remained in each cohort; within the IL‑23i group, 675 (44.6%) received risankizumab and 840 (55.5%) received guselkumab. Baseline characteristics were well balanced (standardized mean difference ≤0.1), mean age was approximately 56 years, and about 85% and 65% of patients had previously received traditional and advanced therapies, respectively. During 12 months of follow-up, the CS injection rate was lower with SEC than with IL‑23i therapy (0.052 vs 0.067 injections per person-month), corresponding to approximately 62 vs 80 injections per 100 patient-years and an estimated 23% lower injection rate with SEC. These findings indicate that SEC was associated with approximately 23% lower CS injection use over 12 months compared with IL‑23 inhibitors, suggesting reduced need for rescue therapy & potentially better disease control in routine clinical practice.

ACR Convergence 2026, November 6–11, Orlando, Florida.