With each passing decade since the drug’s “re-discovery” by rheumatology, evidence for a positive trade-off between its benefits and harms has continued to accrue
Like most rheumatologists, I retain a deep and abiding love for methotrexate. And with each passing decade since the drug’s “re-discovery” by rheumatology, evidence for a positive trade-off between its benefits and harms has continued to accrue, ensuring its role as the centrepiece of our treatment strategies for rheumatoid arthritis, even as an avalanche of biologics and targeted synthetic DMARDs has swept over us.
The role of methotrexate in the treatment of inflammatory rheumatic diseases other than rheumatoid arthritis, however, has often been relegated to bystander status – if it works so well for RA, it stands to reason that it probably works as well for other inflammatory arthropathies.
Psoriatic arthritis (PsA) is a prime example of this situation – MTX is widely used for PsA (and indeed is mandated by the Australian pharmaceutical benefits scheme before the use of bDMARDs). The anecdotal experience for most of us fits well with the hypothesis that the efficacy of MTX in PsA is analogous to RA. But evidence, as we well know, is not the plural of anecdote.
So what is the evidence for the efficacy of MTX in PsA?
We are soon to publish the Cochrane review of MTX for PsA – it would be premature to reveal our conclusions but it is no surprise that the review results are strongly influenced by the MIPA trial published by Kingsley et al in 2012, which found no significant benefit of MTX 15mg/week over placebo (measured by PsARC) at six months.
It is now highly unlikely that we will see another placebo-controlled trial of MTX in PsA, so any residual questions about its true efficacy may need to be approached tangentially via circumstantial evidence.
One important clue is provided by a fascinating late-breaking poster (#L11) presented by James Chung and colleagues: a phase III double-blind RCT of Etanercept and Methotrexate as monotherapy or in combination in patients with psoriatic arthritis. A total of 851 patients who were naive to MTX or bDMARDs and who had active PsA (at least three tender and swollen joints) were randomised to either oral MTX (target 20mg/week), etanercept 50mg/week, or the combination of MTX plus etanercept. Major outcome measures were ACR20 response and minimal disease activity (MDA) at 24 weeks.
Participants had active disease (approximately 12 swollen and 20 tender joints at baseline). Importantly, participants in the MTX arms received a ‘modern’ dose of MTX (median 20mg/week, mean 18.8mg/week).
At week 24, the ACR20 and MDA outcomes were equivalent for the etanercept monotherapy and combination therapy arms, both of which were superior to the MTX monotherapy arm (36% MDA in the etanercept arms versus 23% MDA in the MTX arm). This difference was maintained out to 48 weeks.
While improvements were observed in the MTX monotherapy group (note the 23% MDA response), it is possible that some or all of this is explained by regression to the mean – without a placebo arm, of course, we cannot know this. Furthermore, while radiographic progression at 48 weeks was minor, the etanercept arms were again superior to MTX monotherapy. Adverse events were similarly distributed across the three arms.
Therefore, in contrast to RA, addition of MTX to etanercept does not appear to lead to an incremental benefit.
These results suggest two things. Firstly, it would seem reasonable for our patients commencing a TNFi for PsA to discontinue MTX if this is their preference. Secondly, and perhaps most importantly, if MTX were truly an important DMARD in PsA, one might reasonably expect it to enhance the benefit of TNF inhibition. Should we re-consider our conviction that MTX is a disease-modifying drug in PsA?
Dr Samuel Whittle works in both clinical practice, as a senior consultant rheumatologist at The Queen Elizabeth Hospital in South Australia, and research, as an Australia and New Zealand Musculoskeletal Clinical Trials Network (ANZMUSC) Practitioner Fellow.
This article first appeared in RheumNow and is reproduced with permission
