In neuroscience swap, Amgen buys into Novartis' BACE drug program for Alzheimer's

Amgen is buying into Novartis' early-stage BACE program for Alzheimer's, paying an unspecified upfront and milestones in exchange for a cut of the program, which looks to slash the production of a key toxic protein widely viewed as a prime suspect in the development of the memory-wasting ailment.

The risk-sharing deal focuses closely on Novartis' Phase I/IIa BACE inhibitor (CNP520), which will be the lead molecule in a collaboration that will also pool the two giants' preclinical BACE drugs.

Amgen is willing to cover the lion's share of the research costs for an initial period in the quid pro quo partnership, followed by a 50/50 split. And the big biotech says it was encouraged to enter into the collaboration after the BACE target's genetic authenticity was confirmed by Amgen subsidiary deCODE Genetics.

As part of the neuroscience tie-up Novartis is also in turn taking on part of the burden of Amgen's migraine drug development. Amgen has been moving AMG 334 along in Phase III and AMG 301 in Phase I in what is a crowded field. And Novartis gains an option to commercialize an additional early-stage Amgen molecule.

Like all the BACE drugs, this lead program seeks to prevent the upstream production of amyloid beta, which is frequently found in toxic clusters in the brains of Alzheimer's patients. (Though not frequently enough to convince regulators at this stage.) Merck ($MRK) is widely considered the leader in this field, with Eli Lilly ($LLY) following up with an in-licensed program from AstraZeneca ($AZN) after its own therapy foundered in the clinic due to toxicity. Biogen ($BIIB) and Roche ($RHHBY) have had their own separate efforts in this field, though Roche quietly scuttled a BACE drug in 2013 without ever explaining what had gone wrong.

Alzheimer's remains one of the biggest commercial prospects the industry has ever considered, with megablockbuster potential for any drug that makes it through the clinic. The failure rate over the last decade has run at 99%, which also qualifies the disease as one of the most frustrating ailments the industry has ever tackled.

'334, meanwhile, is one of a number of CGRP therapies in the clinic. Allergan ($AGN) just bought out Merck's position in that race. Amgen recently reported that 52 weeks of therapy with 70 mg of AMG 334 cut the number of migraine days per month by 4.9 in a Phase II study, working off a baseline average of 8.7 mean monthly migraines. Teva ($TEVA) had already noted that its migraine drug TEV-48125 had achieved a nearly 6-day reduction, working with a patient group that had a much higher baseline and an average placebo response of 3.34 days. At 225 mg, Teva's drug registered a 2.8-day improvement for migraine sufferers. As Evercore ISI analyst Umer Raffat noted, that rate easily beat out Amgen's earlier number, along with improvements over a rival drug from Alder Biopharmaceuticals ($ALDR) as well as Eli Lilly.

Amgen R&D chief Sean Harper

"We are very pleased to be joining forces with Novartis on two important neuroscience programs where there remains high unmet medical need," said Dr. Sean Harper, executive vice president of research and development at Amgen, in a statement. "Our collaboration on BACE inhibition reflects Amgen's strategic focus on genetically validated drug candidates while our collaboration in migraine creates an opportunity to more rapidly advance AMG 334 on a global scale."

- here's the release