Event – An Improve-It failure could threaten more than Vytorin

Merck & Co will finally report the full results of the long-running cardiovascular outcomes trial of its controversial cholesterol-lowering therapy Vytorin at the meeting of the American Heart Association, and appropriately for Chicago in November the data are likely to meet with a chilly reception.

The trial is widely expected to fail on the strictest statistical reading of the results; some believe that it might fail even on more generous measures. This outcome could affect more than Merck, which is forecast to collect a respectable $1.6bn in highly profitable sales of the drug this year.

Failure to demonstrate a clear link between lowering LDL-cholesterol and extending life could read through to the development of other projects that rely on this hypothesis, most notably the PCSK9 inhibitors, on which even bigger sales forecasts rest. On the other hand the finger of blame is likely to be pointed largely at Improve-It's design.

The 18,000-patient Improve-It study is testing Vytorin, which combines the statin Zocor with the cholesterol absorption inhibitor Zetia, against Zocor alone. It is 90% powered to detect a 9.4% reduction in cardiovascular events between the two arms.

Improve-It was started in the wake of data that raised questions about the effectiveness of the expensive combination pill, and Merck will be hoping to show that Vytorin is more effective than Zocor alone at helping patients live longer. Improve-It enrolled patients who had recently suffered a heart attack or been diagnosed with acute coronary syndrome, and are therefore at high risk of a secondary event.

Release of the earlier trial data was accompanied by enormous controversy. The Enhance trial measured the effectiveness of Vytorin and Zocor in preventing the build-up of plaque in the arteries of people with an inherited form of heart disease. Its surprise finding, that more plaque built up in the arteries of patients given Vytorin despite a more pronounced impact on LDL, caused serious doubts about the effectiveness and safety of the combination.

Further questions were raised by the SEAS study in patients with the heart valve disease aortic stenosis. This failed on its primary endpoint, finding Vytorin no better than placebo at lowering the risk of cardiac events, and raised a cancer risk signal.

These findings hit both Merck and its then partner Schering-Plough hard in 2008. Subsequent allegations that the companies had suppressed the Enhance data, prompting a congressional enquiry and an investigation into the decision of Carrie Smith Cox, then Schering-Plough’s president, to sell $28m of stock in April and May 2007 – before the Enhance data emerged – was almost the coup de grace.

Perhaps one of the most surprising aspects of the Vytorin story is just how well the drug's sales have held up. Sure, the franchise has been in decline when growth had previously been expected, but Vytorin is still expected to sell $1.6bn this year; revenues peaked at $2.8bn in 2007. Despite questions around Vytorin the LDL hypothesis – that lowering LDL improves outcomes, particularly in a secondary prevention setting – remains intact.

Miss, but by how much?

At the AHA meeting two cuts of the data will be presented. On November 17 the statistically rigorous intent-to-treat (ITT) analysis will be revealed in a late-breaker session; this includes data from all patients that entered the trial. The next day the on-treatment analysis will be detailed, a reading that excludes patients for whom follow-up was not possible, for whatever reason.

The Improve-It trial tracked patients for six to seven years – this was extended from the original plan of two to three years – so the numbers in these two analyses are likely to be markedly different.

Few believe that the ITT read out will be a statistical hit. As well as the change in the length of the trial, analysts point to evidence from other studies that has called into question the drug’s potency and the design of the trial itself. Morgan Stanley analysts, for example, write that Improve-It is unlikely to behave according to the average of a meta-analysis of four trials.

Owing to the perceived shortcomings of the study, many will take comfort should a trend to a survival advantage be seen. And then all eyes will be on the on-treatment analysis. Should this succeed in showing a benefit, damage to Vytorin and to the LDL hypothesis will be lessened.

A failure on both measures will have the broadest ramifications. The PCSK9 inhibitors are widely expected to receive approval in niche patient groups – for example statin-intolerants or those with a genetic susceptibility to high cholesterol – before the readout of outcome studies, on the basis of their ability to lower LDL aggressively.

The fear is that if the Improve-It study weakens the predictive power of this biomarker, those early approvals might be harder to justify. Currently, Amgen and Sanofi/Regeneron expect their PCSK9 candidates to gain FDA approval next year, two years before their own outcomes trials read out.

The Improve-It data will dictate Vytorin’s future sales, currently forecast to be considerable. They will also give an inkling of the likely reception the PCSK9s, and other drug classes, will meet when they go before the regulators.

To contact the writer of this story email Elizabeth Cairns in London at [email protected] or follow @LizEPVantage on Twitter