Modest differences in study designs can have major impact on study outcomes

Beyer-Westendorf J, Lensing AW, Arya R, et al.
Thromb Res. 2016;published online ahead of print

Background

Direct oral anticoagulants (DOACs) have not been studied in head-to-head comparison trials. The existing indirect comparisons between DOACs may be misleading because of differences in study design and patient selection criteria [1,2]. For example, the EINSTEIN-DVT/PE and AMPLIFY trials differed in the duration of therapy and in patient selection. In EINSTEIN-DVT/PE there were few exclusion criteria and anticoagulation therapy could be given for 3, 6 or 12 months [3,4]. On the other hand, the AMPLIFY study exclusion criteria were more stringent, and all patients were treated for 6 months [5].
In this analysis, the impact of the different study designs on treatment outcomes were evaluated. For this purpose, patients enrolled in EINSTEIN-DVT/PE were divided into two cohorts: those who matched the exclusion criteria for AMPLIFY (cohort 1) and those who did not (cohort 2). 

Main results

• Compared with patients in EINSTEIN-DVT/PE cohort 2, those in cohort 1 were older and more often male, had more often unprovoked VTE, a history of prior VTE, or known thrombophilia.
• Compared with patients in AMPLIFY, patients in cohort 1 of EINSTEIN-DVT/PE were slightly older and had more often PE, prior history of VTE, known thrombophilia, or active cancer.
• Recurrent VTE occurred in 1.9% of patients in cohort 1 and in 2.5% of patients in cohort 2 (RR: 0.76; 95% CI: 0.55–1.06).
• Major bleeding occurred in 1.3% of patients in cohort 1 and in 1.1% of patients in cohort 2 (RR: 1.19; 95% CI: 0.75–1.90).
• In EINSTEIN-DVT/PE cohort 1, the rate of recurrent VTE was significantly lower with rivaroxaban than with enoxaparin/VKA (1.5% and 2.3%, respectively; RR: 0.64; 95% CI: 0.43–0.95; P = 0.027).
• The rates of recurrent VTE with rivaroxaban and enoxaparin/VKA were similar in cohort 2 (2.6% and 2.3%, respectively; RR: 1.08; 95% CI: 0.65–1.79; P = 0.77).
• The rate of major bleeding was significantly lower with rivaroxaban than with enoxaparin/VKA in EINSTEIN-DVT/PE cohort 1 (0.8% and 1.7%, respectively; RR: 0.50; 95% CI: 0.30–0.82; P = 0.0068) but not in cohort 2 (1.2% and 1.1%, respectively; RR: 1.03; 95% CI: 0.48–2.18; P = 0.95).
• The first recurrent VTE or major bleeding event in the intention-to-treat population occurred significantly less frequently with rivaroxaban compared with enoxaparin/VKA in cohort 1 (2.4% and 4.0%, respectively; RR: 0.60; 95% CI: 0.44–0.81; P = 0.0011) indicating a superior net clinical benefit with rivaroxaban.
• The frequency of the first recurrent VTE or major bleeding event was similar with rivaroxaban and enoxaparin/VKA in cohort 2 (3.9% and 3.6%, respectively; RR: 1.06; 95% CI: 0.70–1.59; P = 0.79).

Conclusion

This analysis shows that modest differences in study designs can have a major impact on study outcomes. Indirect comparisons may be misleading, if they do not fully adjust for trial differences. These data suggest that caution should be taken when making treatment decisions on the basis of cross study comparisons, in the absence of head-to-head trials.

Find this article online at Thromb Res

References

1.  van Es N, Coppens M, Schulman S, et al. Direct oral anticoagulants compared with vitamin K antagonists for acute venous thromboembolism: evidence from phase 3 trials, Blood 124 (129) (2014) 1968–1975.
2. Castellucci LA, Cameron C, Le Gal G, et al. Clinical and safety outcomes associated with treatment of acute venous thromboembolism: a systematic review and meta-analysis, JAMA 312 (11) (2014) 1122–1135.
3. Bauersachs R, Berkowitz SD, Brenner B, et al. Oral rivaroxaban for symptomatic venous thromboembolism, N. Engl. J. Med. 363 (26) (2010) 2499–2510.
4. Einstein-PE-Investigators, Buller HR, Prins MH, Lensin AW, et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism, N. Engl. J. Med. 366 (14) (2012) 1287–1297.
5. Agnelli G, Buller HR, Cohen A, et al. Oral Apixaban for the Treatment of Acute Venous Thromboembolism, N. Engl. J. Med. 369 (9) (2013) 799–808.