The methodology of calculation of sample size in “non-inferiority” comparative controlled clinical trials: a comparison of two proportions in parallel group

The absolute majority of clinical trials of generic drugs aimed to compare the therapeutic efficacy of the tested drug and the drug of an active control. It is necessary to estimate that the test drug is not less effective (or non-inferior) than the control drug. The designs of the aforementioned trials are called “non-inferiority” study; often, these are phase III of clinical trials. The primary criterions of effectiveness which are quite often used in the clinical trials are frequencies of signs, as a result of the analysis of binary data. Binary data are analyzed by comparing proportions. One of the most important conditions for a correct interpretation of the data obtained during the clinical trials — to provide necessary and suffi cient sample size. The purpose of this article is to show how to plan the study, how to provide necessary and sufficient sample size to ensure that the results and conclusions would be reliable; how to interpret the data in a comparative controlled study aimed to establish “non­inferiority” using proportions comparing in parallel groups.

1. Schumi J., Wittes J.T. Through the looking glass: understanding non¬inferiority. // Trials. 2011 May 3;12:106.

2. Dunnett C.W., Gent M. Significance testing to establish equivalence between treatments, with special reference to data in the form of 2X2 tables. // Biometrics.1977 Dec;33(4):593-602.

3. International Conference on (ICH) of technical requirements for registration of pharmaceuticals for human use. // Statistical principles for clinical trials, 1998.

4. Food and Drug Administration (FDA) (1992) Points to consider. Clinical evaluation of Anti-infective drug products.

5. Hou Y., Wu X.Y., Li K. Issues on the selection of non-inferiority margin in clinical trials. // Chin Med J (Engl). 2009 Feb 20;122(4):466-70.

6. Committee for proprietary medicinal products (CPMP) (2004). Points to consider on the choice of non-inferiority margin.

7. Committee for proprietary medicinal products (CPMP) (2000) Points to Consider on Switching Between Superiority and Non-Inferiority.

8. Julious S.A. Sample Sizes for Clinical Trials. CRC Press 2010, p 180.

9. ICH Note for Guidance E9 (Statistical Principles for Clinical Trials) (1998), CPMP/ICH/363/96.

10. ICH Note for Guidance E10 (Choice of Control Group) (2001), CPMP/ ICH/364/96