NorthWest Biotherapeutics Inc (NWBO)

[10baggerz]

https://www.nejm.org/doi/full/10.1056/nejmra1510064
If the primary outcome is negative, positive findings for secondary outcomes are usually considered to be hypothesis-generating. Certainly, regulatory approval of a new drug is unlikely to follow. However, in some instances, secondary findings are compelling enough to affect guidelines and practice. For instance, in the ASCOT trial of amlodipine versus atenolol for hypertension,25 the hazard ratio for the composite primary outcome of nonfatal myocardial infarction or fatal coronary heart disease was 0.90 (95% CI, 0.79 to 1.02; P=0.11). However, the data supporting evidence of the superiority of amlodipine with respect to stroke, total cardiovascular events, death from any cause, and new-onset diabetes were overwhelming (P<0.001, P<0.0001, P=0.02, and P<0.0001, respectively) (Figure 1). In hindsight, the primary outcome was an odd choice: the decision not to include stroke in a hypertension trial is unconventional. These results support recommendations against the use of atenolol as a first-line or second-line antihypertensive agent.
Few studies are appropriately powered to assess effects on mortality. Proper interpretation can thus be challenging when a large trial shows a reduction in all-cause mortality, which is plausible but not prespecified — especially if the primary outcome was negative. For example, in the MATRIX trial,26 patients with an acute coronary syndrome who underwent PCI were randomly assigned to receive procedural anticoagulation with bivalirudin or unfractionated heparin. There was no significant difference in the 30-day composite primary outcome of death, myocardial infarction, or stroke (relative risk, 0.94; 95% CI, 0.81 to 1.09; P=0.44). However, bivalirudin was associated with a markedly lower incidence of major bleeding as well as lower all-cause mortality (relative risk 0.71; 95% CI, 0.51 to 0.99; P=0.04), a result also observed in some previous studies.27 This finding of reduced mortality with bivalirudin, although mechanistically plausible, ideally requires an additional adequately powered trial for resolution.