Making Money from the Interactome

FEATUREHuman Interactome Project Making Money from the InteractomeBY KATE FODOR© THOM GRAVESAs companies consider their roles in the massive human interactome project, they are keeping in mind some hard-learned lessons. Chief among their case studies is Celera Genomics, which proved that databases cannot, per se, be a firm's main revenue source. "Companies don't talk anymore about the raw data they can amass fo

Mar 1, 2006
Kate Fodor
FEATURE
Human Interactome Project

Making Money from the Interactome

© THOM GRAVES

As companies consider their roles in the massive human interactome project, they are keeping in mind some hard-learned lessons. Chief among their case studies is Celera Genomics, which proved that databases cannot, per se, be a firm's main revenue source.

"Companies don't talk anymore about the raw data they can amass for protein-protein interactions," says Douglas Loe, a healthcare analyst at Versant Partners. "Investors realized that any drug developer that acquired that information would still be somewhere between 10 and 15 years away from commercializing a drug based on the information."

As a result, most proteomics firms have "transformed themselves into more conventional drug-development companies," by either developing products in-house or out-licensing them once they have significant supporting data, Loe explains. Geoff Houlton, a healthcare analyst with Octagon Capital, says the new business model means at least the potential for a proteomics firm to strike it big with a blockbuster drug-a scenario that is impossible under a database or fee-for-service model.

Prolexys Pharmaceuticals, headquartered in Salt Lake City, is a case in point. Myriad Genetics and other corporate investors founded Prolexys in 2001 with the aim of selling nonexclusive subscriptions to a human protein-protein interaction dataset. Prolexys quickly revised its business model to focus on using protein interaction data for its own drug-discovery efforts, which include cancer and cardiovascular indications.

Prolexys still licenses access to its Human Interactome Database, which contains more than 258,000 protein-protein interactions, to GlaxoSmithKline and several other firms. However, those deals are "just a minor part of the business," says chief scientific officer Sudhir Sahasrabudhe. "It's an attempt to monetize as much as we can from the technology platform we created, but really, the robust return on our technology investment will come from the drug candidates we are putting forth in the clinic."

A. Donny Strosberg, founder and former CEO of Paris-based Hybrigenics, who recently joined the Scripps Research Institute in Florida, says interactome databases are a hard sell not only because of the long drug-development timeline, but also because of the nonspecific information the databases provide.

"In the beginning, Hybrigenics was trying to sell protein interaction maps for companies' favorite cells ... and they said, 'No, thank you. Keep your information or publish it, but we're not going to pay for it,'" Strosberg recalls. "It was far too much information for what they were interested in." Hybrigenics has shifted its focus to pathway-based drug discovery. The privately held company employs about 44 people and has raised 47 million euros in three rounds from investors, including Advent Management and Alafi Capital.

Another key strategy for industry is to collaborate with academia, says Daniel Figeys, who ran the protein interaction-mapping project at Toronto-based MDS Proteomics before moving to the University of Ottawa. MDS Proteomics was reorganized under the name Protana in 2004. The Toronto-based biopharmaceuticals firm Transition Therapeutics, which acquired its assets, is using Protana's drug-discovery technology to forge partnerships with Big Pharma and to fill its own pipeline.

Industry can offer high-throughput capabilities and other technological advantages, while academia brings to the table a range of knowledge that might not be available within one company, Figeys explains. "You need a lot of intellectual input on how you select the proteins you're going to study, which cell line you're going to do them in, and whether you're going to do it in an over-expression mode or a normal-expression mode or a down-regulation mode," he points out.

Strosberg agrees that industry-academia partnerships make sense. "There is no way academics can reproduce what a company like Hybrigenics has put in place, which is literally a factory for studying protein-protein interactions," he says. "On the other hand, there is no way a company like Hybrigenics can analyze the results the way it can be done by individual academic labs."

Such partnerships can give companies access to significant sums of grant money. Prolexys, for example, recently joined a consortium of investigators working on drugs for latent tuberculosis infection supported by $20 million from the Bill & Melinda Gates Foundation, among other funders.