Clinical perspectives on the use of composite endpoints

Abstract

Although mortality is the most important endpoint in evaluating new regimens, insistence on its use as the only endpoint in clinical trials can require that thousands of patients be studied. Accordingly, composite endpoints have been increasingly used to increase the overall event rate and thereby reduce the number of patients needed for the trial. For use as part of composite endpoint, nonfatal endpoints should be clinically meaningful, i.e., related to an adverse subsequent prognosis. In acute myocardial infarction (MI), several intermediate endpoints in the well-described pathophysiology of acute MI have been correlated with an adverse long-term outcome: recurrent MI, new onset congestive heart failure or cardiogenic shock, left ventricular dysfunction, large infarct size, and failure to achieve early patency of the infarct-related artery. Furthermore, in acute MI, new therapies that improve these nonfatal endpoints also improve mortality, thereby validating this approach. Once this link is established, such nonfatal endpoints can be validly used in evaluating new therapies. Note, however, if this link has not been made (that mortality is reduced when there is a reduction in the nonfatal endpoint), as in the case of suppression of ventricular premature complexes with antiarrhythmic therapy, the nonfatal endpoint cannot be used validly. Thus, appropriately designed and validated composite endpoints can provide a valid means of testing new treatments in a smaller trial than one using mortality alone. Their use should allow testing of a greater number of new regimens, thereby allowing more rapid progress toward improving the clinical outcome of patients.