CLINICAL TRIAL PROTOCOL SPACE 2 Stent ... - SPACE-2 Studie
CLINICAL TRIAL PROTOCOL SPACE 2 Stent ... - SPACE-2 Studie
CLINICAL TRIAL PROTOCOL SPACE 2 Stent ... - SPACE-2 Studie
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Clinical Trial Code: <strong>SPACE</strong> 2 Version: 10. October 2008 Final<br />
9.4 Statistical methods<br />
Biometric analysis will be defined in the statistical analysis plan which has to be<br />
authorized before closing the database by the biometrician, the sponsor, and the LKP.<br />
Safety: The one-sided 97.5% confidence interval under H0 for the difference of the<br />
event rates (rate in the CAS group – the rate in the CEA group) will be calculated. Noninferiority<br />
is proven if the upper margin of this interval is less than 1.5%. This test has<br />
power of 90% if CAS shows a slight safety advantage compared to CEA (2.5% CAS<br />
versus 3% CEA). The power of the non-inferiority test is 80% in case of a slightly<br />
reduced risk in both procedures (CAS and CEA 2%).<br />
Efficacy: In order to control the overall Type I error on the 5% level, a hierarchical<br />
testing procedure will be used. The first step of the procedure consists in a closed<br />
testing procedure for the difference between the three treatments. The tests are<br />
performed on a 5% level. If there are simultaneously significant differences between the<br />
pairs BMT and CEA as well as BMT and CAS on a 5% level, the hierarchical testing<br />
procedure enters step 2, otherwise there is no further statistical testing. Since<br />
superiority for CEA and CAS over BMT is not yet established in the studied population<br />
both tests need to be significant to allow the interpretation of step 2 in a clinically<br />
sensible way. Step 2 of the procedure consists in a non-inferiority test: The one-sided<br />
97.5% confidence interval for the difference of the event rates (rate in the CAS group –<br />
the rate in the CEA group) will be calculated. Non-inferiority is proven if the upper<br />
margin of this interval is less than 2.5%.<br />
In case of losses to follow up, the event rates and their standard errors will be<br />
calculated by Kaplan-Meier-techniques.<br />
9.5 Interim analyses<br />
No interim analysis is planned with respect to the primary endpoint. An interim analysis<br />
of the safety data is planned after three years. This will not influence the sample size<br />
calculation because this is based on the efficacy endpoint. In addition the safety<br />
committee will check the safety regularly.<br />
_______________________________________________________________________________________________________<br />
Page 31 of 49