Effectiveness of Switch to a Second Anti-TNF-α in Primary Nonresponders, Secondary Nonresponders and Failure Due to Adverse Events

Purpose: It is known from previous studies that switching to a second antiTNFα may be effective after failure to the first. However, it is unclear whether switching is as effective for patients who did not show any response to their first treatment (primary nonresponders), as compared to patients who had initial response but failed later (secondary nonresponders) and patients who failed due to adverse events.

Methods: Patients with RA who switched to a second antiTNFα before January 2006 after failure to a first were selected from a Dutch antiTNFα register. DAS28 scores were available at start, 3 months and stop of both treatments. For each treatment the best DAS28 was selected. Based on the reason of failure to the first antiTNFα 3 groups were defined: primary nonresponders, secondary nonresponders and adverse events. Response was defined as good or moderate EULAR response based on the best DAS28. Groups were compared using ANOVA for continuous data, Pearson Chi-squared test for categorical data and nonparametric tests for data without normal distribution.

Results: A total of 126 patients were included: 35 primary nonresponders; 41 secondary nonresponders; and 50 patients who failed due to adverse events. Patientcharacteristics and baseline DAS28 were comparable at start of first treatment. After switch, all groups including the primary nonresponders showed significant improvement of the best DAS28 compared to baseline: -1,77 for primary nonresponders, -1,18 for secondary nonresponders and -1,54 for patients failed due to adverse events (p = 0,189). The percentage of responders was 70,6% for primary nonresponders, 65% for secondary nonresponders and 64,2% for patients failed due to adverse events (p = 0,142). However, the absolute DAS28 scores during the second treatment were significantly higher for primary nonresponders compared to the other groups and lowest for patients who failed due to adverse events (shown in figure).

Conclusion: In patients who were primary nonresponder to their first antiTNFα treatment, a switch to a second antiTNFα may be beneficial. However, a second treatment is more successful in patients who showed an initial response to the first antiTNFα.