Comparison of etanercept and infliximab in RA

[Guest guest]

Dec 5, 2002

Comparison of etanercept and infliximab in RA

New Orleans, LA - Several comparisons of the 2 TNF blockers available for

the treatment of rheumatoid arthritis, etanercept (Enbrel®, Immunex/Wyeth)

and infliximab (Remicade®, Centocor/Schering-Plough), have come up with some

different conclusions.

At the recent meeting of the American College of Rheumatology, a poster

presentation of a meta-analysis of clinical trial data showed a significant

advantage for etanercept in the early response to treatment, suggesting that

more patients would continue with treatment on it than on infliximab [1].

The researchers, Dr Bacon (University of Birmingham, UK) and Dr Alan

Reynolds (Wyeth, UK), noted that their conclusions tie in with the findings

from clinical practice in Sweden, reported recently in the ls of

Rheumatic Diseases [2]. But another poster at the ACR meeting by Dr

Yocum and colleagues at the University of Arizona [3] reported on US data

and found exactly the oppositethat patients on infliximab remain on therapy

significantly longer than those on etanercept.

The 2 results are quite the reverse of each other, Reynolds commented to

rheumawire; in fact, the curves for survival on therapy from Sweden and the

US are very similar, but the curves are labeled etanercept and infliximab in

one, and vice versa in the other.

Meta-analysis of clinical trial data favors etanercept

The meta-analysis reported by Bacon and Reynolds was supported by a grant

from Wyeth and was based on clinical trial data.

For etanercept, data were taken from 4 trials [4-7]. For infliximab, data

were taken from 2 trials [8,9], with additional information from other

material [10, 11].

The analysis was performed 3 times, once at the drug doses approved in the

US, then at doses approved in Europe, and third at the doses commonly used

to begin therapy in clinical practice (ie, etanercept subcutaneously 25 mg

twice a week, and infliximab intravenously 3 mg/kg every 8 weeks). Efficacy

was assessed on ACR responses, and in 1 study where us responses were

reported these were considered equivalent to the corresponding ACR

responses.

The results showed a highly significant difference in the ACR20 response

rate in favor of etanercept, Bacon and Reynolds reported, and the highly

significant advantage for etanercept remained even when they considered the

higher doses of infliximab. The ACR50 response rate also showed a

significant difference in favor of etanercept, but this advantage was lost

when compared with the higher doses of infliximab. There was no significant

difference between the 2 drugs in ACR70 responses, but the number of

patients achieving this response was relatively small, the researchers

commented.

The difference in the response rate at the usual starting doses of the 2

drugs needs particular consideration, Bacon and Reynolds reasoned, as the

initial response will determine whether or not the drug will be continued

after the initial 3 months. (For instance, current UK guidelines for

anti-TNF therapy recommend that patients who have not responded in 3 months

should discontinue treatment, they note.) " The evidence shows that

significantly more patients would be expected to achieve clinical responses

to etanercept than to infliximab in that time frame, " they say.

Backed by results from Swedish observational study

Bacon and Reynolds say their findings are confirmed by an observational

study based on clinical practice in Sweden [2]. This analysis, reported by

Dr Pierre Geborek and colleagues at Lund University, followed 166 patients

on etanercept, 135 patients on infliximab, and 103 on leflunomide (Arava®,

Aventis). Comparison of the 2 TNF blockers showed that significantly more

patients on etanercept than on infliximab reached an ACR20 response at 3

months (p<0.02) and at 6 months (p<0.05), but for the ACR50 responses only

the difference at 3 months was significant (p<0.05), again in favor of

etanercept. No differences in either ACR20 or ACR50 responses between the 2

were seen at 0.5, 1.5, 9, and 12 months. Both the TNF blockers were found to

be more effective than leflunomide. With etanercept, there was a significant

difference in the ACR20 and ACR50 responses at both 3 and 6 months compared

with leflunomide, but with infliximab only at the 3-month time point.

Geborek et al also reported on " survival while receiving a drug, " ie, the

continuation of the respective treatment. There were no significant

differences between etanercept and infliximab, with 79% and 75% of patients,

respectively, continuing to receive the drug after 20 months. In contrast,

only 22% of patients on leflunomide were still on the drug after 20 months

(p<0.001 vs both etanercept and infliximab).

Bacon and Reynold say that the evidence from their meta-analysis, together

with this data from Sweden, suggest that " etanercept should be the treatment

of choice as significantly more patients achieve a response either at the

starting doses or the approved doses of these drugs. "

But infliximab comes out better in US analysis

But a different story emerges from the other analysis presented at the ACR

meeting. Yocum and colleagues explain that they " wanted to investigate the

duration of therapy for both infliximab and etanercept in a context

different from clinical trials. " Hence they reviewed the medical records of

patients attending their university rheumatology clinic (staffed by 5

rheumatologists and 1 nurse practitioner) and found complete data sets for a

group of 208 patients who began anti-TNF therapy (90 on etanercept and 118

on infliximab). Most of these patients (82%) had rheumatoid arthritis, but

16 patients (7.2%) had psoriatic arthritis, 17 (7.6%) had juvenile

rheumatoid arthritis, and 7 (3.1%) were classified as other. Mean disease

duration was 15.51 years (range 0-56).

Kaplan-Meier estimates for discontinuation of any anti-TNF therapy showed

that the mean time on medication was 768 days (95% CI 693-843), with a

maximum follow-up time of 1260 days. Among the 90 patients on etanercept, 58

stopped using the drug, with a mean time to discontinuation of 595 days (95%

CI 491-700). Of the 118 patients on infliximab, 32 patients stopped, with a

mean time to discontinuation of 931 days (95% CI 844-1018). After 15 months,

there were no discontinuations from infliximab therapy. Statistical analysis

of the curves for survival on therapy showed that the difference was

significant (log rank test coefficient 20.03, p<0.01, and Gehan-Breslow

coefficient 16.01, p<0.001).

" The survival analysis demonstrates that patients on infliximab remain on

therapy significantly longer that those patients on etanercept, " the

researchers concluded. " The analysis also showed a low probability of

discontinuations of infliximab after 1 year of therapy. "