More Than 50,000 Exposures Included in Analysis
From the pooled registry, data were gathered from 13,374 courses of JAKi, 25,049 courses of TNFi, and 16,482 courses of OMA. There were some differences between these groups, including a significantly lower median age for those in the JAKi pool (57.1 years vs 58.3 and 60.5 years for TNFi and OMA, respectively) and median disease duration (8.3 years vs 11.0 and 11.9 years, respectively).
A greater proportion of patients on TNFi were naive to therapy (44.6%), compared with either JAKi (20.2%) or OMA (16.1%). More patients in the TNFi pool (60.0%) were also on concomitant therapy than those in the JAKi pool (49.5%) or the OMA pool (51.9%).
Other characteristics such as disease activity, body mass index, and percentage of smokers were comparable.
When TNFi was used as the reference, there were no significant differences in the rate of all infections, the rate of all infections excluding herpes zoster, and all serious infections. In all three groups, the incidence rates were numerically but not significantly lower in patients on JAKi vs OMA. With the exception of serious infections, for which the adjusted incidence of JAKi was 0.99 relative to TNFi, both JAKi and OMA had numerically higher incidence rate ratios than TNFi.
Herpes Zoster Risk on JAKi Is > Twofold Higher
Because the CIs overlapped in all cases, none of the differences were significant. The exception was herpes zoster. The 1.07 incidence rate ratio for OMA was not significantly different than the TNFi reference, but the 2.27 rate ratio for JAKi far exceeded either of the other two comparators (95% CI, 1.17-3.02).
In a separate analysis of patients at least 55 years of age with at least one cardiovascular risk factor, the numerical differences between groups were narrower and thus did not reach statistical significance, even for herpes zoster. Although the herpes zoster rate ratio was 1.62 for JAKi vs 1.23 for OMA (TNFi as the reference was 1.0), the CI for JAKi (0.86-3.03) overlapped both.
Based on a Poisson regression analysis, this study took into account a wide variety of variables, including age, disease activity, comorbidities, and tobacco use, Mr. Aymon said. He noted that the analyses were performed on data from each registry as well as with the pooled data, and the data were reasonably consistent.
Initially, the investigators had planned to evaluate differences between therapy groups, if any, for COVID-19 infection, but differences in the availability and use of vaccinations among the countries where the registries were maintained made this analysis too complicated to conduct.
Mr. Aymon reported no potential conflicts of interest, but some coauthors reported financial relationships with manufacturers of both bDMARDs and JAKi. Dr. van Gaalen reported financial relationships with AbbVie, Bristol Myers Squibb, Eli Lilly, Merck Sharp & Dohme, Novartis, and UCB.
A version of this article appeared on Medscape.com.