West Coast Editor

Almost exactly four months after F. Hoffmann-La Roche Ltd. said it would stop developing on its own a pair of phosphodiesterase-4 inhibitors that were part of an ongoing deal with Memory Pharmaceuticals Inc., the companies are retooling their arrangement - and Memory is taking the compounds back.

Under the amended deal, Basel, Switzerland-based Roche will be granted an option to continue development and commercialization work with MEM 1414 or MEM 1917 after Phase II trials are done. Meanwhile, Roche is keeping its exclusive, worldwide license to develop and commercialize all other drug candidates from Memory's PDE4 inhibitor program, which is "alive and well," said Jzaneen Lalani, vice president of legal affairs for Montvale, N.J.-based Memory.

"Now that the amendment is done, we get access to all the data" related to the work done so far by Roche on MEM 1414 and MEM 1917, and Memory will use that to make a decision on how to proceed, she said, adding that "we haven't set a time line for that, because we haven't got all the data yet."

Memory's stock (NASDAQ:MEMY) closed Friday at $2.26, up 9 cents.

"We haven't got a specific reason [why Roche gave the compounds back], but I think we'll be able to give our reason once we've gone through the data," Lalani told BioWorld Today.

If Roche in the future exercises its option for a license to MEM 1414 and/or MEM 1917 and goes ahead with more work on them, Memory would get milestone payments for specified events after Phase III trials begin, plus royalties on worldwide sales at levels higher than specified in the original deal. Also, under certain circumstances, Memory would have the option to co-promote MEM 1414 and MEM 1917 in the U.S.

In mid-April, Roche decided to quit experimenting with MEM 1414 for Alzheimer's disease and MEM 1917 as the backup candidate for Alzheimer's as well as potentially for depression. The companies said at the time that testing was still under way with PDE4 inhibitor compounds under the deal that began in 2002 and was extended last summer by at least 18 months of funding, or $5.3 million. (See BioWorld Today, July 31, 2002.)

"MEM 1414 was in Phase I when Roche told us they were going to stop clinical development, and 1917 was in preclinical," Lalani said.

In its 2004 annual report, Memory said Roche had paid $45.4 million so far in two separate deals that included the two PDE4 compounds along with certain nicotinic alpha-7 partial agonists, including MEM 3454 and MEM 63908. The former is at the Phase I stage for schizophrenia and might be useful against Alzheimer's disease. Also for Alzheimer's is MEM 63908, still in preclinical study.

Going public in the spring of 2004, Memory priced significantly below its intended range, selling 5 million shares at $7 each to raise $35, after targeting an amount up to $86.25 million. (See BioWorld Today, April 4, 2004.)

Lalani said the company had $25.2 million in cash and cash equivalents as of June 30, enough to operate through the first quarter of 2006.