Tofacitinib vs bDMARDs in RA: What is the risk of serious infections?

A recent US-based study showed that the targeted synthetic (ts)DMARD* tofacitinib may not have a higher risk of serious infection compared with all biologic DMARDs (bDMARDs).

“[W]e were able to conclude that tofacitinib is associated with a moderately increased risk of hospitalization for serious infection compared with etanercept and a modestly increased risk of hospitalization for serious infection versus abatacept, golimumab, and tocilizumab,” said the researchers. However, the risk of infection was comparable between tofacitinib, certolizumab, and adalimumab. 

Using three health insurance claims databases in the US, the researchers identified eight mutually exclusive groups of adult patients with rheumatoid arthritis (RA; n=130,718, 78 percent female) who were initiating tofacitinib (4.8 percent) or bDMARDs (95.2 percent).

A total of 3,140 serious infections (requiring hospitalization) were recorded over 100,790 person-years of follow-up, corresponding to a crude incidence rate of 3.12 per 100 person-years.

The incidence rate of serious infections requiring hospitalization be it bacterial, viral, or opportunistic infections, among tofacitinib users was 3.52 per 100 person-years, while among bDMARD users, it ranged from 2.32 to 5.75 per 100 person-years in etanercept and infliximab users, respectively.

The risk of serious infection was higher among patients who were treated with tofacitinib compared with those treated with etanercept (adjusted hazard ratio [adjHR], 1.41, 95 percent confidence interval [CI], 1.15–1.73). The risk of serious infection was also higher with tofacitinib than abatacept (adjHR, 1.20, 95 percent CI, 0.97–1.49), golimumab (adjHR, 1.23, 95 percent CI, 0.94–1.62), and tocilizumab (adjHR, 1.17, 95 percent CI, 0.89–1.53), but these results were not statistically significant. [Lancet Rheumatol 2020;2:e84-e98]

The risk of serious infection was comparable between patients on tofacitinib and adalimumab (adjHR, 1.06, 95 percent CI, 0.87–1.30) or certolizumab (adjHR, 1.02, 95 percent CI, 0.80–1.29). However, the risk of serious infection with tofacitinib was lower, albeit not significantly so, than that with infliximab (adjHR, 0.81, 95 percent CI, 0.65–1.00).

The risk of skin or soft tissue infections was higher in tofacitinib compared with abatacept (adjHR, 2.36) or etanercept (adjHR, 2.21) users, while pneumonia or upper respiratory tract infection risk was higher in tofacitinib compared with etanercept users (adjHR, 1.53). The risk of septicaemia or bacteraemia was less common in tofacitinib than in infliximab users (adjHR, 0.68), but comparable to that of other bDMARDs, while the risk of bacterial infection was lower, though not significantly, in tofacitinib compared with infliximab (adjHR, 0.78) and tocilizumab (adjHR, 0.76) users.

The risk of herpes zoster was about twofold higher with tofacitinib compared with all bDMARDs (adjHR, 1.94 [abatacept], adjHR, 1.99 [adalimumab], adjHR, 2.24 [certolizumab], adjHR, 2.12 [etanercept], adjHR, 1.84 [golimumab], adjHR, 1.94 [infliximab], and adjHR, 2.14 [tocilizumab]).

“[P]hysicians prescribing JAK** inhibitors to patients with RA should remain watchful for the possibility of development of this infection,” noted the researchers. Pre-therapy herpes zoster vaccination could help reduce this risk in patients initiating tofacitinib, though the risk of zoster should be weighed against the benefits of tofacitinib, they said.

“The serious infection risk is a major concern for both bDMARDs and tsDMARDs, because the high disease activity in RA itself is a risk factor for infection, and immunosuppressive effects of these medications might further increase this risk,” said the researchers. However, the type or extent of infection may differ from one medication to another.

“[T]his comprehensive assessment derived from data collected during routine clinical care is important in helping patients with RA and rheumatologists make appropriate treatment choices,” they said.

Seeing as how new DMARDs are often introduced after failure of TNF*** inhibitors, the exclusion of patients with prior use of tofacitinib and bDMARDs in this study may reduce the generalizability of the results, they noted. Additionally, the median follow-up period was less than a year and as such, long-term risk of serious infections has yet to be ascertained.