The Food and Drug Administration’s approval of leflunomide in September 1998 as a treatment for rheumatoid arthritis was sandwiched between the debuts of infliximab (Remicade and biosimilars) and etanercept (Enbrel) in August and November of that year, the latter of which was so exciting that “within 2 months you couldn’t get [it],” recalled Eric M. Ruderman, MD. And “like every middle child, [leflunomide] was underloved, underappreciated, and largely dismissed.”
Yet should it have been? Is it worth another look today?
Courtesy Michael Pollard
Dr. Eric M. Ruderman
At the 2024 Rheumatology Winter Clinical Symposium, Dr. Ruderman reflected on some of the clinical trial data published after leflunomide’s approval that “got lost in the shuffle” of the rightful embrace of biologics in United States practice, and urged reconsideration of the loading strategy still advised in the drug’s labeling.
“I’m not telling you that you should be using [leflunomide] in place of biologics, instead of biologics, or before biologics … but it should be in your toolkit,” said Dr. Ruderman, professor of medicine and associate chief of clinical affairs in the division of rheumatology at Northwestern University Feinberg School of Medicine, Chicago. The drug “still has a role in RA, including in combination with methotrexate, and a potential role in other rheumatic diseases.”
“In our PsA clinic,” he noted, “we’ve actually not infrequently added leflunomide to some of the other agents we’ve been using.”
Key Findings Over the Years in RA
Leflunomide showed efficacy similar to that of sulfasalazine in a randomized trial published in 1999 that used primary endpoints of tender/swollen joints and physician and patient global scores. Then, against methotrexate, it proved just as efficacious in achieving at least 20% improvement in American College of Rheumatology composite response criteria (ACR20) over 52 weeks, and in meeting endpoints similar to those of the sulfasalazine trial, in two trials, one published in 1999 and another in 2000.
“So here were two big trials [comparing it with methotrexate] that suggested the drug was just as good as what had become our standard of care by that point,” Dr. Ruderman said.
Each of these three trials used a loading dose of 100 mg leflunomide for 3 days, followed by 20 mg daily. Sulfasalazine was initiated at 2 g and escalated over 4 weeks. Methotrexate was initiated in one of the trials at a dose of 7.5 mg, then increased to 15 mg in almost two-thirds of patients; in the other methotrexate trial the initial dose was 15 mg escalated over 3 months.
Side effects of leflunomide — GI issues, rash, alopecia (reversible), and elevated liver function tests — were similar across the trials, and represented “about the same toxicities as methotrexate,” he said.
Researchers then tested leflunomide as an add-on to methotrexate in patients who had inadequate response, which “was a little bit daunting since we were still concerned about the toxicity of methotrexate at this point,” Dr. Ruderman said. “The idea that we’d take another drug with similar toxicities and add it on to the methotrexate was a little scary.”
But it worked. Patients on a mean background dose of 16.5 mg methotrexate were randomized to placebo or to a 2-day leflunomide loading dose followed by 10 mg/day that could be escalated at 8 weeks to 20 mg if needed. At 6 months, 19.5% and 46.2%, respectively, met ACR20 (P < .001), and “interestingly,” he said, “adverse events were pretty similar” between combination therapy and methotrexate monotherapy.
“This was very much like all the studies we’ve seen over the years with new biologics — they were all added to background methotrexate,” he said. “And the truth is, the [46%] response seen when adding leflunomide to background methotrexate wasn’t very different from the 50% [ACR20] response you tend to see when you add a biologic.”
However, despite the study’s conclusion that combination therapy provided significant benefit to patients with inadequate response to methotrexate alone, “the drug got lost, because everyone was prescribing the biologics,” Dr. Ruderman said.
He said he found only one study comparing leflunomide with a biologic. In a notably small but well-designed study from Sri Lanka published in 2017, 40 patients with an inadequate response to methotrexate were randomized to low-dose rituximab (500 mg x 2) or 20 mg/day leflunomide (no loading dose). At week 24, ACR20 was nearly identical (85% vs 84%), with a similar rate of adverse events.
The researchers pointed out “that there’s a potential cost benefit in developing countries where biologics aren’t as accessible,” he said, agreeing that “the big opportunity for a drug like leflunomide is outside the US, where you don’t have access to the drugs we take advantage of all the time.”
A meeting participant from Canada pointed out that rheumatologists there are “mandated to use it for PsA in combination with methotrexate before we can get a biologic, and for RA we can use it with Plaquenil [hydroxychloroquine] and methotrexate before we get a biologic, so we’re using it all the time.”
Asked about efficacy, the physician said the combination with methotrexate is “absolutely” efficacious. “It works really well” he said. “The problem is, you really have to watch the white cell count and liver function … and the half-life is long.”
Indeed, Dr. Ruderman said during his talk, the plasma half-life of teriflunomide, its active metabolite, is 15.5 days, which is challenging when adverse events occur. “And it’s a terrible drug in young women thinking about pregnancy because it’s teratogenic and stays around,” he said.
Leflunomide, which, notably, was “developed specifically for RA from the get-go” and not borrowed from another specialty, works by blocking de novo pyrimidine synthesis, Dr. Ruderman said. T-cell activation requires the upregulation of pyrimidine production (salvage pathways are insufficient); the “drug prevents that” by inhibiting an enzyme that catalyzes conversion of dihydroorotate to orotate, which, in turn, is converted to pyrimidine ribonucleotides, he explained.
Other potential mechanisms of action have been proposed — mainly, inhibition of tumor necrosis factor signaling and inhibition of kinase activity, including the JAK/STAT pathway — but “there’s not great data for any of them,” he said.