BioCentury
ARTICLE | Tools & Techniques

Significance not necessarily predictive

June 21, 1999 7:00 AM UTC

In statistics, showing a significant difference between observed results in two groups is distinct from the predictive power of that observation.

Alexion Inc. maintains that it was appropriate to do a statistical analysis of a handful of patients in a Phase I trial of its 5G1.1 C5 complement inhibitor in lupus, while acknowledging that its ability to predict future results based on those small numbers is limited. But based on the results, ALXN has decided to develop its 5G1.1 C5 complement inhibitor to treat proteinuria...