Anti–tumor necrosis factor α switching in rheumatoid arthritis and juvenile chronic arthritis


To the Editor:

Over the past few years, tumor necrosis factor α (TNFα)–blocking agents have been increasingly used in the treatment of severe refractory rheumatoid arthritis (RA) (1–3) and juvenile RA (JRA) (4). The 2 most widely used approaches to inhibiting the action of TNFα have thus far been the administration of soluble TNF receptor (etanercept) or anti-TNFα chimeric antibodies (infliximab). Several trials have shown that both of these biologic agents are very effective in reducing the signs and symptoms of joint inflammation, inhibiting radiographic progression, combatting disability, and improving the quality of life of patients with RA or JRA (5, 6). However, some patients have to discontinue treatment because of adverse events or a lack of efficacy (5, 6), and, because most of them have been previously treated with all of the nonbiologic disease-modifying antirheumatic drugs (DMARDs), it may be difficult to choose an alternative therapy. In a open-label pilot study, we investigated whether switching from one to the other anti-TNFα agent is a safe and effective procedure in such patients.

The study group included 15 patients (14 women and 1 man; mean ± SD age 46.4 ± 6.92 years). Eight of these patients had RA, 7 had JRA, and all of them had discontinued infliximab or etanercept therapy because of a lack of efficacy or adverse events. None of the patients had contraindications to the use of anti-TNFα agents, and all gave their informed consent. All patients had severe, long-standing, active disease (mean ± SD disease duration 13.2 ± 2.07 years, and a Disease Activity Score in 28 joints [DAS28] >3.7). In all patients disease was refractory to treatment with traditional DMARDs.

Fourteen of the 15 patients were being treated with infliximab (3 intravenous infusions of 3 mg/kg every 45 days) and low-dose methotrexate (median dosage 12.5 mg/week), which was discontinued after a mean (± SD) period of 10.3 ± 2.03 months because of a lack of efficacy (according to the American College of Rheumatology [ACR] 20% improvement criteria) in 8 patients or adverse events (2 cases of hypotension, 1 case of angioedema, 1 case of urticaria, and 2 cases of respiratory distress). The only patient being treated with etanercept (25 mg subcutaneously twice weekly) discontinued the therapy because of a lack of efficacy after 6 months of treatment.

Patients were switched to the alternative treatment 4 weeks after the discontinuation of the previously administered anti-TNFα drug and were then monitored for tolerance and efficacy once a month for 6 months. The efficacy parameters used in this study were the number of tender and swollen joints (28-joint count), pain measured with a visual analog scale (VAS), a VAS general health assessment, the first-hour erythrocyte sedimentation rate, the levels of C-reactive protein and hemoglobin, the DAS28, and the disability index of the Health Assessment Questionnaire. Safety and tolerability was monitored by recording all adverse events and performing a thorough physical examination and standard laboratory investigations at each control visit.

Twelve of the 14 patients treated with etanercept completed the 6-month study period. The other 2 had to discontinue therapy for the same reason they had stopped their previous treatment with infliximab: one patient discontinued treatment after 5 months because of a lack of efficacy, and the other discontinued treatment after 3 months because of the recurrence of the same adverse event (angioedema). During the 6 months of therapy, diffuse urticaria developed in 2 patients, which was easily controlled by antihistamine treatment and did not prevent them from continuing etanercept. The only patient who switched to infliximab completed the 6-month therapy without experiencing any adverse event. After 6 months of therapy, the median values of the efficacy variables measured in all patients of were significantly better than those recorded at baseline (Table 1). By month 6, 13 of the 15 patients were responders according to the ACR 20% improvement criteria.

Table 1. Clinical and laboratory parameters at baseline and after 6 months in patients with RA or JRA*
 Baseline6 months
Mean ± SDMedian (range)Mean ± SDMedian (range)
  • *

    RA = rheumatoid arthritis; JRA = juvenile rheumatoid arthritis; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; VAS = visual analog scale; GH = general health; DAS28 = Disease Activity Score in 28 joints; HAQ = Health Assessment Questionnaire.

  • P < 0.05, RA versus JRA, by Wilcoxon's rank test for all comparisons.

ESR, mm/hour50.5 ± 27.755.4 ± 22.941 (21–100)40 (25–100)34 ± 22.132 ± 20.132 (7–77)31 (6–81)
CRP, mg/dl2.57 ± 2.63.12 ± 2.81.8 (0.3–9.5)1.7 (0.4–9.2)1.1 ± 0.71.2 ± 0.61.25 (0.1–2.8)1.3 (0.1–2.9)
Hemoglobin, gm/dl11.5 ± 1.011.4 ± 1.211.4 (10.3–13)11 (10.7–13)12.3 ± 1.012.6 ± 1.312.3 (11–14)12 (11.5–14)
Tender joint count13.6 ± 7.711.7 ± 6.911 (5–26)11.5 (6–25)6.4 ± 5.56.1 ± 5.25 (0–18)5.5 (0–15)
Swollen joint count12.1 ± 5.512.3 ± 5.413.5 (5–20)13.5 (4–22)7.1 ± 5.36.9 ± 5.15 (1–18)5 (0–16)
Pain VAS61.3 ± 25.965.6 ± 22.862 (16–98)64 (20–96)40.1 ± 21.240 ± 21.844 (8–64)42 (8–61)
GH VAS62.8 ± 21.366.4 ± 20.362.5 (20–90)62.5 (25–90)35.8 ± 18.933.8 ± 17.932.5 (9–65)30.5 (8–62)
DAS286.02 ± 1.26.1 ± 1.16.1 (4.4–7.9)6 (4.3–7.8)4.6 ± 1.44.5 ± 1.54.7 (2.7–6.6)4.6 (2.6–6.5)
HAQ1.85 ± 0.51.88 ± 0.62 (0.37–2.5)2 (0.4–2.5)1.51 ± 0.61.47 ± 0.71.7 (0.2–2.2)1.6 (0.2–2.4)

The results of this study seem to indicate that, as previously reported by other authors (7, 8), switching from one anti-TNFα treatment to the other may be successful in patients with severe and active RA or JRA who fail to respond to all of the nonbiologic DMARDs. Only 2 of our 15 patients failed therapy with both drugs (curiously enough, for the same reason).

In conclusion, our results suggest that, if the first administered anti-TNFα drug (infliximab or etanercept) has to be discontinued because of a lack of efficacy or an adverse event, switching to the other anti-TNFα is likely to be safe and effective.

Ennio Giulio Favalli MD*, Marco Arreghini MD*, Cristina Arnoldi MD*, Benedetta Panni MD*, Antonio Marchesoni MD*, Sergio Tosi MD*, Irene Pontikaki MD†, * Gaetano Pini Institute Milan, Italy, † Center for Rheumatic Children Gaetano Pini Institute Milan, Italy.