How would you treat a poorly controlled seronegative RA patient who is steroid dependent and is on azathioprine for autoimmune hepatitis?

I would add a biologic agent, likely a TNF inhibitor or abatacept. As with any patient with active RA, it would be important to add therapy that controls their disease activity, as well as allowing them to reduce or preferentially discontinue steroids. I would avoid other agents with known risk for hepatotoxicity, including IL6 inhibitors, JAK inhibitors, MTX, or leflunomide, both for the potential increased risk in someone with autoimmune hepatitis as well as to avoid possible confusion when trying to attribute increased hepatic enzymes to either the medication or the existing hepatitis. While no longer a first line choice of therapy, azathioprine has been shown to be effective in RA, and I have, on rare occasions, used it as concomitant t...

In the early and mid-1990s, 2 RCTs examined the use of MTX with AZA in RA. Both found that the combination was not more effective than MTX alone, and MTX monotherapy led to better responses than AZA or the combo. AEs were more common with combo and AZA monotherapy; reasons for discontinuation included lack of efficacy, GI AEs, and LFT elevations. These findings are not surprising since both agents are purine synthesis inhibitors. Knowing that we want to avoid further LFT elevations or potential hepatotoxicity, I would agree with Dr. @Dr. First Last that adding a biologic agent would be preferable, and in the case of TNFi's, the use of AZA can help to abrogate immunogenicity with agents such as adalimumab or infliximab.

References:
Willkens e...