The overall risk of serious infection was similar among patients with rheumatoid arthritis being treated with tocilizumab (Actemra) and those receiving tumor necrosis factor (TNF) inhibitors, although risks were higher with tocilizumab for certain types of infections, a large cohort study found.
The incidence rate of composite serious infections requiring hospitalization was 4.68 per 100 person-years in patients given the interleukin 6 receptor blocker tocilizumab and 3.99 per 100 person-years in patients on TNF inhibitors, according to Seoyoung C. Kim, MD, ScD, of Brigham and Women’s Hospital in Boston, and colleagues.
The result was a nonsignificant pooled hazard ratio of 1.05 (95% CI 0.95-1.16), the researchers reported online in Annals of the Rheumatic Diseases.
However, the risks of serious bacterial infections, skin and soft tissue infections, and diverticulitis were higher in the tocilizumab group.
“Our results suggest that we as clinicians/rheumatologists should educate patients about the risk of infection in general with any biologics and be more vigilant for even less common types of infection (i.e., skin and soft tissue infection and diverticulitis) when treating with biologic drugs, particularly with tocilizumab,” Kim told MedPage Today.
Because the individual biologics target different cellular and cytokine pathways, the risks of infection could differ, and previous comparisons have had conflicting results. To help clarify this, the researchers analyzed outcomes in three large U.S. databases: Medicare from 2010-2015, IMS PharMetrics Plus from 2011-2015, and Truven MarketScan from 2011-2015.
Patients included in the analysis were required to have used one or more biologics or tofacitinib (Xeljanz) in the past and to be new users of tocilizumab or a TNF inhibitor. A secondary analysis compared infection rates in patients initiating tocilizumab and those given another biologic (abatacept, Orencia).
Participants were propensity score matched for more than 70 baseline characteristics, including disease severity, demographics, comorbidities, history of surgery, and other medications used.
After matching, the primary analysis included 13,102 patients initiating tocilizumab and 26,727 starting a TNF inhibitor, while the secondary analysis included 10,414 matched pairs of patients starting tocilizumab or abatacept.
Mean patient ages were 72 in the Medicare group, 51 in the PharMetrics group, and 53 in MarketScan. The majority of patients in all three databases were taking methotrexate and corticosteroids at baseline.
When the researchers analyzed the risks according to specific types of infection, greater risks were found with tocilizumab for the following:
- Serious bacterial infections, HR 1.19 (95% CI 1.07-1.33)
- Skin and soft tissue infections, HR 2.38 (95% CI 1.47-3.86)
- Diverticulitis, HR 2.34 (95% CI 1.64-3.34)
But the absolute risks were small, with rate differences of 0.16 and 0.31 per 100 patients for skin and soft tissue infections and diverticulitis, respectively, and “estimates can be imprecise in such analyses,” the researchers pointed out.
In the secondary analysis of tocilizumab versus abatacept, the composite serious infection risk was greater for tocilizumab (HR 1.40, 95% CI 1.20-1.63). Risks also were higher with tocilizumab than with abatacept for serious bacterial infections (HR 1.50, 95% CI 1.27-1.78), diverticulitis (HR 1.79, 95% CI 1.12-2.84), septicemia/bacteremia (HR 1.39, 95% CI 1.08-1.78), and pneumonia/upper respiratory tract infection (HR 1.37, 95% CI 1.04-1.80).
“This head-to-head comparative safety information between tocilizumab and TNF inhibitors or abatacept may help better manage patients with rheumatoid arthritis in clinical practice,” the authors concluded.
“Patients who are at increased risk of infection (e.g., known history of infection, recurrent episode) should be closely monitored while on these drugs. Treatment decisions should be individualized based on not only the efficacy of the drugs, but also the safety of drugs including infection risk,” Kim advised.
A limitation of the study, the researcher said, was the possibility of residual confounding despite the propensity score matching.
The study was funded by Roche.
The authors reported financial relationships with Roche, Pfizer, Bristol-Myers Squibb, Genentech, Amgen, AbbVie, Corrona, Merck, Vertex, Aetion, and Boehringer Ingelheim; several are employees of Genentech.