Lundbeck Alzheimer’s data justify Otsuka’s exuberance

It was a long time coming, but the phase II Alzheimer’s disease study of Lundbeck’s Lu AE58054 has generated full data at last, explaining Otsuka’s willingness to hand over $150m up front for rights to the project in March.

A Lundbeck spokesman said today that Otsuka had already seen the full results before their presentation at this week’s Alzheimer’s Association International Conference in Boston, and had struck the deal largely on the strength of the phase II dataset. The plan now is to proceed straight into the first study of a 3,000-patient phase III programme later this year.

Lu AE58054 was highlighted as a development priority last year by Lundbeck’s president and chief executive, Ulf Wiinberg, who hinted that many big companies were interesting in licensing it. Sure enough, the Otsuka deal, with a biodollar value of $825m, was struck months later (Lundbeck nets $150m as Otsuka delves further into Alzheimer’s, March 26, 2013).

Two years on

The trial had actually been completed in December 2011, and topline data were revealed in May 2012. The spokesman said it was not unusual for so much time to elapse before full results as the company has to clean the data and comply with conference deadlines.

Not only has Lu AE58054 hit statistical significance in the primary endpoint of cognition improvement, but the trial had enrolled patients with moderate disease, who represent a high bar to success. Lundbeck confirmed, however, that the start of phase III would not trigger a milestone payment.

In the 24-week study in 278 patients, 90mg/day of Lu AE58054 was added on top of Eisai’s Aricept, an established drug that offers borderline cognitive benefits. It resulted in a 2.16-point improvement in the Adas-Cog scale over Aricept alone, with a p value of 0.04.

Although the threshold for clinical importance over six months is usually held to be four points, Lundbeck said the 2.16-point change was “absolutely clinically relevant”, as well as statistically significant. Secondary endpoints including functional improvement were missed – hardly surprising given that it has proved near impossible for other projects to affect this measure with statistical significance.

Lack of functional improvement need not prove fatal, of course, and recent US FDA draft guidance proposed the use of a single cognition endpoint in phase III trials in very mild forms of the disease, which rarely show signs of functional impairment. However, Lu AE58054 is destined to be tested in phase III in mild, as well as the much harder to treat moderate dementia.

Recently, Lilly pressed ahead with development of its amyloid beta antibody solanezumab in mild Alzheimer’s (Once more, with feeling: Lilly sings sola again, July 15, 2013). The Lundbeck spokesman would not comment on the design of the Lu AE58054 phase III study, which will test several doses of the compound plus Aricept.

Different mechanism

Lilly’s determination to plough on with the beta-amyloid approach despite the failure of solanezumab and related agents shows the attraction that Alzheimer’s still holds.

Lu AE58054 uses a different mechanism, having antagonistic activity at the 5-HT6 receptor, which is expressed in brain regions involved in cognition and is thought to modulate the activity of neurotransmitter systems. It had been studied in schizophrenia, but this indication was discontinued after the Otsuka deal.

EvaluatePharma reveals several other 5-HT6 antagonists in development for Alzheimer’s, including GlaxoSmithKline’s GSK742457, but Lu AE58054 looks like the first to advance into phase III.

Lundbeck and Otsuka might be years from knowing whether their bet has been worth it, but they can take some comfort in having a first-in-class compound.

Study Design Trial ID
Phase II 278 moderate Alzheimer's patients, addition to Aricept NCT01019421

To contact the writer of this story email Jacob Plieth in London at or follow @JacobEPVantage on Twitter

Share This Article