BIRMINGHAM, ENGLAND – Data from two early RA inception cohorts provide reassurance that methotrexate does not cause interstitial lung disease and suggest that treatment with methotrexate might even be protective.
In the Early RA Study (ERAS) and Early RA Network (ERAN), which together include 2,701 patients with RA, 101 (3.7%) had interstitial lung disease (ILD). There were 92 patients with RA-ILD who had information available on exposure to any conventional synthetic disease-modifying antirheumatic drug (csDMARD); of these, 39 (2.5%) had been exposed to methotrexate (n = 1,578) and 53 (4.8%) to other csDMARDs (n = 1,114).
Multivariate analysis showed that methotrexate exposure was associated with a reduced risk of developing ILD, with an odds ratio of 0.48 (P = .004). In a separate analysis that excluded 25 patients who had ILD before they received any csDMARD therapy (n = 67), there was no association between methotrexate use and ILD (OR, 0.85; P = .578). In fact, there was a nonsignificant trend for a delayed onset of ILD in patients who had been treated with methotrexate (OR, 0.54; P = .072).
Methotrexate use is associated with an acute hypersensitivity pneumonitis in patients with RA, explained, of St. George’s University Hospitals NHS Foundation Trust in London at the British Society for Rheumatology annual conference. “This is well recognized, it’s very rare [0.43%-1.00%], it’s easy to spot, and usually goes away if you stop methotrexate,” said Dr. Kiely, adding that “it’s not benign, and severe cases can be life threatening.”
Because of the association between methotrexate and pneumonitis, there has been concern that methotrexate may exacerbate or even cause ILD in RA but there are sparse data available to confirm this. The bottom line is that you should not start someone on methotrexate if you think their existing lung capacity is not up to treatment with methotrexate, Dr. Kiely said.
ILD is not always symptomatic in RA, but when it is, it is associated with very poor survival. The lung disease can be present before joint symptoms, Dr. Kiely said. Although less than 10% of cases may be symptomatic, this “is a big deal, because it has a high mortality, with death within 5 years. It’s the second-commonest cause of excess mortality in RA after cardiovascular disease.”
To look at the association between incident RA-ILD and the use of methotrexate, Dr. Kiely and associates analyzed data from ERAS (1986-2001) and ERAN (2002-2013), that together have more than 25 years of follow-up data on patients who were recruited at the first sign of RA symptoms. Patients within these cohorts have been treated according to best practice, and a range of outcomes – including RA-ILD – have been assessed at annual intervals.
In the patients who developed ILD after any csDMARD exposure, older age at RA onset (OR, 1.04; P less than .001) and having ever smoked (OR, 1.91; P = .016) were associated with the development of the lung disease. Incident ILD was also associated with being positive for rheumatoid factor (OR, 2.02; P = .029) at baseline. Being male was also associated with a higher risk for developing ILD, Dr. Kiely reported, as was a longer duration of time between the onset of first RA symptoms and the first secondary care visit. Conversely, the presence of nonrespiratory, major comorbidities at baseline appeared to be protective (OR, 0.62; P = .027).
“We found no association between methotrexate treatment and incident RA-ILD and a possibility that it may be protective,” Dr. Kiely concluded, noting that these data were now published in BMJ Open (2019;9:e028466..
Following Dr. Kiely’s presentation, an audience member asked if the protective effect seen with methotrexate could have been caused by better disease control overall.
Dr. Kiely answered that, up until 2001, the time when ERAS was ongoing, standard practice in the United Kingdom was to use sulfasalazine, but then methotrexate started to be used in higher and higher doses, as seen in ERAN.
The interesting thing is that in ERAN more methotrexate was used in higher doses, but less RA-ILD was seen, Dr. Kiely observed. The overall prevalence of RA-ILD in the later early RA cohort was 3.2% and the median dose of methotrexate used was 20 mg. In ERAS, the prevalence was 4.2% and the median dose of methotrexate used was 10 mg.
There was a suggestion that disease control was slightly better in ERAN than ERAS, but that wasn’t statistically significant, Dr. Kiely said.
So, should a patient with RA and ILD be given methotrexate? There’s no reason not to, Dr. Kiely suggested, based on the evidence shown. Part of the challenge will now be convincing chest physician colleagues that methotrexate is not problematic in terms of causing ILD.
These findings are completely on board with the ILD group’s findings that methotrexate doesn’t cause pulmonary fibrosis in patients with RA, commented, of St. Helens and Knowsley Teaching Hospitals NHS Trust, St. Helens, England. Her own research, which includes a 10-year follow-up of patients with inflammatory arthritis, has shown that methotrexate does not appear to increase the risk of pulmonary fibrosis.
The study had no specific outside funding. Dr. Kiely reported having no conflicts of interest.
SOURCE: Kiely P et al. Rheumatology. 2019;58(suppl 3),