The small molecules and biologics used to treat rheumatoid arthritis (RA) today have significant efficacy but the potential to cause serious adverse events in some patients. Although direct comparisons of safety are not available for all therapies, trends in the safety profiles of inhibitors of JAK, TNF, and IL-6 have been identified that may aid clinicians in personalizing treatment for patients based on their individual risk factors.

The treatment options that we have now for RA have good safety profiles relative to their efficacy, unlike older drugs, which often had significant safety concerns coupled with efficacy that was sometimes marginal. Comparing the safety of these agents, the results of a postmarketing safety study showed that the JAK inhibitor tofacitinib did not meet noninferiority criteria relative to TNF inhibitors with respect to major adverse cardiovascular events (MACE) and malignancy, so the US Food and Drug Administration (FDA) indication for tofacitinib (and for other JAK inhibitors) is after at least 1 TNF inhibitor. Based on these results, some people have concluded that JAK inhibitors are not safe, but there are nuances to consider when evaluating the data. For example, the study population in this trial was older and at higher risk for MACE at baseline. When applying the results to an individual patient, I am not concerned about using a JAK inhibitor in a younger patient with no known cardiovascular risk factors. However, if an older patient has risk factors such as hypertension and smoking, using a TNF inhibitor is probably the safer move.

We also know that there is an inherently higher risk of MACE in patients with RA, and TNF inhibitors reduce that risk more so than, say, methotrexate, either because of their specific mechanisms of action or because they are more effective at reducing systemic inflammation. So, perhaps neither TNF inhibitors nor JAK inhibitors increase the risk of MACE, but rather the TNF inhibitors reduce that risk more than the JAK inhibitors due to factors related to the mechanism of the molecules. The risk of venous thromboembolism (VTE) is similar to the risk of MACE in that RA itself increases the patient’s risk. There is no clear evidence of an increased risk of VTE with JAK inhibitors, but some data suggest a potentially higher risk with JAK inhibitors than with TNF inhibitors. Again, in practice, if you have somebody who has had multiple VTEs or an unprovoked pulmonary embolism, I might consider avoiding a JAK inhibitor.

Gastrointestinal perforations have been reported with IL-6 inhibitor use, most notably in patients with diverticulitis who are also taking steroids. If a patient with a history of diverticulitis is taking prednisone, using an IL-6 inhibitor would not be my first choice. Further, there have been well-documented reports of the reactivation of latent tuberculosis in patients soon after starting TNF inhibitor therapy, so it is now standard to screen for latent tuberculosis before starting any biologic. Invasive fungal infections have also been reported with TNF inhibitors, but they are very uncommon. The majority of infections in those receiving TNF inhibitors, or any biologic, are routine bacterial infections, not mycobacterial or invasive fungal infections. Finally, most biologics are generally considered safe for use during pregnancy with close monitoring.

Although there are risks of adverse events with every drug, we can risk stratify patients and think about which drug may be the best choice given patients’ individual risk factors. Treatment choices always need to be individualized, and we should avoid making clinical decisions based on blanket assumptions that 1 drug is less safe than another for all patients.