Background/Purpose:
ACT-iON was a global, multicenter, observational,
52-wk, clinical practice study of the effectiveness of tocilizumab (TCZ) vs anti–tumor
necrosis factor (aTNF) agents prescribed as first biologic therapy after inadequate
response to DMARDs (DMARD-IR).

Methods:
Eligible patients were DMARD-IR with moderate to severe RA (1987 ACR criteria) for
≥24 wks who were prescribed TCZ or aTNF as their first biologic, in accordance
with local label, within 6 weeks of enrollment. The study comprised a 52-wk
observation period. No study-specific medication or dosing regimen was
stipulated. The primary effectiveness measure was change in DAS28_ESR
from baseline to wk 24. Data were analyzed by treatment group based on the
first administered biologic. Safety was assessed in patients receiving ≥1
dose of biologic. Effectiveness was assessed in patients with a disease
activity assessment <60 days before first biologic dose (effectiveness
population). Missing data were not imputed, and no correction for multiple
testing was applied.

Results:
Of 1250 patients screened, 1225 were
enrolled; 1216 received ≥1 dose of biologic (safety population: TCZ, 423 [34.8%];
aTNF, 793 [65.2%]), and 1083 were included in the effectiveness population (TCZ,
390 [36.0%]; aTNF, 693 [64.0%]). Overall, 158/1216
(13.0%) patients discontinued the study (lack of efficacy: TCZ, 0.9%; aTNF, 2.0%;
adverse events [AEs]: TCZ, 2.1%; aTNF, 1.6%). Drug survival was higher with TCZ
than aTNFs (p<0.001; Table). TCZ vs
aTNF patients had shorter disease duration and started their biologic more
often as monotherapy (28.1% vs 16.0%; Table). Oral corticosteroids were used in
a higher proportion of TCZ patients. Although baseline
DAS28_ESR was similar between groups,
TCZ patients had a significantly larger change from baseline in DAS28_ESR
than aTNF patients (difference in adjusted means
[95% CI]: wk 24, –0.9 [–1.1, –0.6]; wk 52, –0.9 [–1.2, –0.6]; p<0.001
at wks 24 and 52). Changes from baseline at wks 24
and 52 for other effectiveness parameters were also significantly better with
TCZ (Table). Sensitivity analysis restricting validated baseline assessments to
2 weeks before first biologic dose confirmed these results. AEs and serious AEs
(SAEs) occurred in 49.2% and 5.2% of TCZ patients and 56.6% and 8.1% of aTNF patients.
Infections were the most common AEs (TCZ, 20.8%; aTNF, 25.9%) and SAEs (TCZ,
1.9%; aTNF, 3.3%). Three (0.7%) TCZ patients
and 6 (0.8%) aTNF patients died; 1 death in each group (both due to pneumonia) was
deemed treatment related.

Conclusion:
In DMARD-IR patients starting a biologic
for the first time in real-life settings, TCZ was initiated as monotherapy more
often than aTNFs. Safety profiles of TCZ and aTNFs were comparable and similar
to the safety profiles from clinical trials. TCZ was associated with better drug
survival and was possibly associated with better improvements in DAS28, CDAI,
and HAQ-DI than aTNFs.