Methotrexate to Prevent RA, Clear as Mud Save

Intervention in individuals predisposed to develop rheumatoid arthritis (RA), with a holy grail of prevention of RA, has long been a hot topic. It is well known that seropositivity for rheumatoid factor and antibodies to cyclic citrullinated peptides (ACPA) can precede RA disease by many years. Many, but by no means all, patients also seem to go through a pre-RA arthralgia phase of varying duration characterised by the presence of increasing joint pain without frank arthritis.

I will admit to being a naysayer on the possibility of RA prevention with any of our existing therapeutic agents. I approached the ARIAA and PRARIE studies, of abatacept and rituximab respectively, with a healthy dose of scepticism. My personal view is that these trials were essentially treating early RA with a high-level medication and that any evidence of efficacy was really just a mirror of the efficacy of these drugs in later stage RA. The APIPPRA study of abatacept has many fans, but I don’t count myself amongst them. A picture paints a thousand words, and the survival curve here was striking – a wide difference at 52 weeks followed by increasingly rapid convergence as we approach 104 weeks. I am a firm believer that given sufficient time these will coalesce and there will no longer be a difference – that all we are seeing here is a delayed biologic withdrawal effect making it appear that there is a causative preventative role.

My (genuine) favourite of these prevention studies has always been the TREAT EARLIER study of methotrexate. All study participants had arthralgia judged at high risk of developing RA and MRI evidence of synovitis, but no clinically evident arthritis. The intervention was methotrexate at maximum tolerated dose (up to 25mg weekly) for 1 year plus one shot of IM methylprednisolone 120mg versus placebo. The investigators recruited 236 participants. Methotrexate did not prevent the development of overt arthritis up to 1 year. In a high-risk subgroup and in ACPA positive patients it delayed arthritis development. There were, however, sustained benefits to methotrexate up to 24 months in physical function (HAQ), MRI inflammation scores, as well as pain and morning stiffness. This suggested the interesting hypothesis that methotrexate does not prevent arthritis development but may modulate the disease course, with early initiation having sustained long-lasting benefits. This may be the correlate of the “window-of-opportunity” with which we are familiar in RA; treatment earlier in the disease course leads to better long-term outcomes than later initiation. It appears that there may be an equally important “window-of-opportunity” in pre-RA that affords the opportunity to initiate long term benefits that will be a lost opportunity if we wait for the development of the arthritis.

We now however have further confusion sown. The 4-year results of the TREAT EARLIER study, presented at Tuesday’s oral abstract session, abstract 2672, show that methotrexate appears to prevent the development of RA in high risk ACPA- patients. 8.6% in the treatment-arm developed RA, compared to 29.0% in the placebo-arm (HR 0.27; 95%CI 0.07-0.99). There was no significant effect in ACPA+ patients (58.1% vs 52.2%, HR 0.93; 95%CI 0.45-1.93) or low risk ACPA- (7.5% vs 9.5%, HR 0.79; 95%CI 0.22-2.80). Subclinical joint inflammation, physical functioning and grip strength persistently improved upon treatment in ACPA-negative participants with increased-risk.

I honestly don’t know what to do with this collection of results now. Is this real? Is it just a chance finding? It certainly needs replication. I struggle to find a hypothesis to explain how methotrexate could prevent seronegative RA in high risk individuals but not prevent seropositive RA, or seronegative RA in low risk individuals. The data is what it is, and it certainly is provocative and in need of further study.