The study chairman, John Dormandy, M.D., estimated that adding Actos to other diabetic medications in 1,000 people would avoid 21 first MIs, strokes, or deaths. Looked at in another way, 48 patients would need to be treated for three years to avoid one first major cardiovascular event.

Nonetheless, the trial failed to meet its primary endpoint. This was a 20% improvement in time from randomization to next cardiovascular event (defined as composite of all cause mortality, non-fatal MI, stroke, acute coronary syndrome, coronary revascularization, limb revascularization, or amputation above the ankle).

Actos was associated with only a 10% improvement in this composite endpoint, which was not significant (21.0% in the Actos arm versus 23.5% in the placebo group, P=0.095). The components of the endpoint all trended toward an advantage for Actos, with one exception: Actos was associated with a slightly increased number of leg bypasses.

But the significant benefits seen in the secondary endpoints prompted Dr. Dormandy to declare the PROACTIVE findings a "breakthrough for patients who are at high risk from heart attacks, strokes or premature death because it is the first time an oral (diabetes) agent has been shown to have this benefit in a prospective study."

PROACTIVE, which was conducted in 19 European countries, recruited 5,238 diabetics who had prior MI. They were randomized to the highest tolerated dose of Actos (up to 45 mg) versus matched placebo. Both study drugs were given on top of standard therapy, including insulin when indicated. At baseline, 75.4% of patients had hypertension, 57.5% angina, 23.6% claudication, 23.2% retinopathy, 14.1% nephropathy, and 5.7% had a history of transient ischemic attack. Patients were followed for an average of 2.8 years.

Other findings:
· Actos reduced blood glucose (HbA1c) by 0.5% more than placebo.
· Actos increased HDL by 19% versus a 9% increase with placebo.
· Actos reduced triglycerides by 11% versus an increase of 2% with placebo.
· Patients in the Actos arm decreased systolic blood pressure by a median of 3 mm Hg more than placebo (P=0.03).
· Actos decreased progression to permanent insulin use by 50% versus placebo.
At least 10% of Actos patients developed edema.
The EASD discussant and independent reviewer, Hannele Yki-Jarvinen, M.D., of the University of Helsinki in Finland, said the good news is that the primary endpoint was reduced by 10%, but the bad news is that the incidence of heart failure with Actos in the trial was twice as high as the reduction in cardiovascular events. However, she conceded that heart failure may have been either over- or under-diagnosed in the trial.