BOSTON – Tapering but perhaps not completely stopping tumor necrosis factor inhibitors can result in good control of rheumatoid arthritis with considerable cost savings, according to investigators in two studies.

Dr. Noortje van Herwaarden and colleagues from Sint Maartenskliniek in Nijmegen, the Netherlands, conducted a randomized, open-label study of dose-reduction vs. usual care in 180 patients with stable, well-controlled rheumatoid arthritis (RA) on a tumor necrosis factor inhibitor (TNFi) – either adalimumab (Humira) or etanercept (Enbrel) – plus in some cases a stable disease-modifying antirheumatic drug (DMARD) or glucocorticoid.

“This is the first study to demonstrate that disease activity–guided tapering of adalimumab and etanercept in RA patients doing well is noninferior to usual care regarding major flare,” Dr. van Herwaarden said at the annual meeting of the American College of Rheumatology.

She and her colleagues randomly assigned the patients on a 2:1 basis to TNFi dose reduction or usual care. Dose reduction was accomplished with a stepwise increase in the interval between injections every 3 months until flare. Flare was defined as an increase greater than 1.2 in the 28-joint Disease Activity Score calculated using C-reactive protein (DAS28-CRP) or a DAS28-CRP increase greater than 0.6 with a current DAS28-CRP score of 3.2 or greater, compared with baseline. Follow-up was for 18 months with visits every 3 months or any time a patient reported more symptoms.

Among 169 patients available for a per-protocol analysis, the cumulative incidence of major DAS28-CRP flares lasting more than 3 months (the primary endpoint) was 12% in patients randomized to dose reduction, compared with 10% in the usual-care group, a difference that was not statistically significant. The absolute difference was 2% and the 95% confidence interval of –12% to 12% fell below the prespecified margin of 20% for noninferiority of dose reduction, Dr. van Herwaarden said.

However, there was a significant difference in the cumulative incidence of short-term flares lasting 3 months or less, which occurred in 73% of patients on dose reduction, compared with 27% of those on usual care (P < .001).

Measures of disease activity, functioning, and quality of life were generally similar throughout the study, the investigators found.

In the dose-reduction group, 43% of patients successfully tapered the TNFi, and 20% stopped altogether, but for the remaining 37% of patients, no dose reduction was possible. There were no between-group differences in the use of DMARDs or glucocorticoids.

Minor radiographic progression

Looking at radiographic progression, the investigators found no differences in minimal clinically important change or smallest detectable change between the dose-reduction and usual-care groups. But significantly more patients in the dose-reduction group met stricter cutoff for progression of 0.5%, compared with the usual care group (32 patients vs. 15 patients, respectively). The total modified Sharp-van der Heijde score and joint-space narrowing scores also favored usual care, but the erosion score was not different between the groups.

Adverse events were similar between the groups, but there were more serious adverse events in the dose-reduction group, most of which were attributable to planned surgery among these patients, Dr. van Herwaarden said.

Mean total per-patient costs for the 18-month follow-up were almost $11,000 lower in the dose-reduction group ($15,728 vs. $26,576). The decremental cost-effectiveness ratio was calculated to be $488,116 – the amount that is saved when 1 quality-adjusted life-year (QALY) is lost. The difference is about five times higher than the widely accepted amount that society is willing to pay for 1 QALY gained, Dr. van Herwaarden said.

Reduce yes, stop no

In a separate study, Japanese researchers found that stopping a biological agent altogether in patients with RA in remission was associated with a high rate of failure.

The investigators looked at a Japanese rheumatic diseases registry data and found that among 31 patients who discontinued biological DMARDs while in remission in typical clinical practice, 71% by 1 year and 76% by 2 years had a treatment failure, defined as restart of the biological agent, glucocorticoid dose escalation, nonbiological DMARD addition, or loss of remission on the clinical disease activity index.

“Maintenance of disease activity after discontinuation is relatively difficult in established RA patients in the typical clinical practice setting. This may suggest that a reduction in dose intensity – either by prolongation of treatment intervals or by dose induction intervals – may be a more feasible option in established RA patients,” said Dr. Kazuki Yoshida, currently a research fellow in the section of clinical sciences in the division of rheumatology, immunology, and allergy at Brigham and Women’s Hospital in Boston.

Dr. van Herwaarden’s study was internally funded. Dr. Yoshida’s study was supported by a grant to Kameda Medical Center, Kamogawa City, Japan. Both physicians reported having no relevant disclosures.

Tapering TNFi’s safe and much cheaper in RA