Methotrexate-Based Background Therapy Prolonged Anti-TNF Persistence in Rheumatoid Arthritis
In a large British registry analysis, background methotrexate emerged as the anchor drug most closely linked to sustained use of first-line anti-TNF therapy in rheumatoid arthritis. Patients receiving methotrexate alone, or methotrexate combined with other conventional DMARDs, stayed on treatment longer than those receiving no DMARD, leflunomide, or sulfasalazine monotherapy. The signal was clinically meaningful because persistence with biologic therapy often tracks with real-world effectiveness, tolerability, and the ability to avoid early discontinuation in routine care.
The cohort included 10,396 patients with rheumatoid arthritis enrolled in the British Society for Rheumatology Biologics Register, all starting their first anti-TNF agent. Baseline background therapy varied widely: 3,339 received no DMARD, 4,418 received methotrexate, 610 received leflunomide, 308 received sulfasalazine, 902 received methotrexate plus sulfasalazine, 401 received methotrexate plus hydroxychloroquine, and 418 received methotrexate plus sulfasalazine plus hydroxychloroquine. This spread allowed a direct comparison of how different conventional DMARD backbones influenced biologic durability in daily practice.
One-year drug survival for the first anti-TNF therapy was 71% overall, with a narrow 95% confidence interval of 71% to 72%. By 5 years, persistence had fallen to 42%, with a 95% confidence interval of 41% to 43%. The drop was steep. In long-term inflammatory disease care, this kind of attrition usually reflects a mix of loss of efficacy, adverse effects, treatment burden, and switching decisions made when the initial regimen no longer holds the disease in check.
Background DMARD Choice Changed the Odds of Staying on Treatment
Compared with methotrexate, patients receiving no DMARD, leflunomide, or sulfasalazine were more likely to discontinue their first anti-TNF agent. In contrast, patients receiving methotrexate in combination with other DMARDs showed better treatment persistence. The direction of effect fits with a simple pharmacologic and immunologic logic: methotrexate can reduce anti-drug antibody formation and may help preserve biologic activity, while combination conventional therapy may improve control of baseline inflammation enough to make the biologic regimen more durable.
The clinical relevance is immediate at the bedside. Rheumatoid arthritis treatment failure is rarely caused by a single mechanism, and biologic discontinuation often reflects both inadequate disease suppression and the practical pressures of tolerability. A background regimen that strengthens biologic exposure or blunts immunogenicity can shift a patient from early discontinuation toward sustained benefit.
Patients on leflunomide or sulfasalazine alone did worse than those on methotrexate. In registry terms, that pattern argues against assuming all conventional DMARDs are interchangeable when paired with anti-TNF therapy. The choice of companion drug appears to matter, and methotrexate remained the most reliable background agent in this cohort.
Registry Data Point Toward a Class Effect, with One Combination Exception
The analysis used Kaplan-Meier survival methods to follow persistence with anti-TNF therapy for up to 5 years, and multivariate Cox proportional hazards models to compare treatment survival across DMARD groups. Those models were stratified by anti-TNF used and start year and adjusted for multiple potential confounders, which strengthens the inference that the observed differences were not simply the result of calendar-time drift or biologic selection effects.
The conclusion was not that every combination regimen worked equally well. Instead, methotrexate-containing combinations performed better overall, supporting ongoing use of background DMARD combinations that include methotrexate when anti-TNF treatment begins. A specific caveat remained for sulfasalazine combined with etanercept, where a possible exception was raised. In a field where treatment continuation is often used as a surrogate for usefulness in routine practice, even one combination exception deserves attention because it can affect prescriber behavior at the first biologic decision point.
A biologic that is discontinued early can never show its full value. That reality makes persistence more than a bookkeeping exercise; it is a proxy for whether a regimen can hold inflammatory disease down over months and years rather than days and weeks.
Why the Methotrexate Signal Still Dominates Clinical Practice
Methotrexate likely retains its central role because it does more than add another anti-inflammatory layer. It can dampen immune activation, improve overall disease control, and reduce immunogenicity against biologic agents. In rheumatoid arthritis, those effects matter because anti-TNF therapies are vulnerable to loss of effectiveness when the immune system begins treating the drug itself as a target. A lower rate of discontinuation in methotrexate-treated patients is therefore biologically plausible and consistent with wider clinical experience.
The finding also helps explain why clinicians often keep methotrexate in place when a biologic is introduced, even when the patient has already cycled through conventional therapy. Registry evidence like this supports that habit with durable outcome data rather than short-term symptom control alone. It also argues for caution when stopping background methotrexate without a strong clinical reason.
By contrast, starting anti-TNF therapy on leflunomide or sulfasalazine alone may leave some patients at a disadvantage. The study does not prove that these agents are ineffective in every setting, but it does show that they were associated with poorer persistence than methotrexate-based regimens in this large real-world cohort.
Real-World Persistence Matters More Than First Impressions
The study’s strength lies in its size and its observational reach across routine rheumatology practice. Ten thousand patients provide a broad view of how anti-TNF therapy behaves outside the controlled environment of a registration trial. The trade-off is familiar: observational data can identify durable patterns, but it cannot fully eliminate confounding by indication, disease severity, or prescribing preference.
Even with that limitation, the practical message is consistent. When anti-TNF therapy is started in rheumatoid arthritis, methotrexate remains the safest background assumption, and methotrexate-containing combinations appear to improve persistence further. Discontinuation risk rises when no DMARD is used or when leflunomide or sulfasalazine are used alone. For clinicians choosing a companion regimen, the first months after biologic initiation may be the most revealing window, because early persistence often foreshadows the longer-term treatment course.
The data leave one narrow but important uncertainty: how much of the persistence advantage reflects improved pharmacology versus better overall disease selection for methotrexate-based regimens. Even so, the register offers a strong real-world argument for preserving methotrexate when anti-TNF therapy is introduced, and for reconsidering sulfasalazine or leflunomide monotherapy in that setting.