Some of the world’s biggest selling cancer drugs today work for only a small fraction of patients, and researchers have struggled to say for sure why that is. A Cambridge, MA-based biotech company, Ensemble Discovery, has developed a diagnostic test that its partner Roche, the world’s largest maker of cancer drugs, is advancing to clinical trials early in 2009. The goal is to see if Ensemble has developed a more precise method to predict if a drug is going to work for an individual patient.
Ensemble is announcing today that it has extended a collaboration that started in July 2007 with Switzerland-based Roche to see whether its proprietary diagnostic tests will offer better predictive clues about whether drugs like Genentech and Roche’s trastuzumab (Herceptin) or ImClone Systems’ cetuximab (Erbitux) are going to work. The company didn’t disclose how much it is receiving in milestone payments from Roche, or what the clinical trial timetable looks like, but I was able to get a basic rundown on the technology from CEO Mike Taylor, senior vice president of biology David Livingston, and Laurence Reid, the chief business officer.
Ensemble was founded in 2004 based on a new class of therapeutics developed by David Liu at Harvard University. It has raised $32 million since inception from Flagship Ventures, Arch Venture Partners, Harris & Harris Group, CMEA Ventures, and Boston University. Most of their bet is being placed on Ensemble’s experimental drugs, but the company has also applied its technology to a new form of diagnostic test that could someday help insurers decide whether to give a patient a $10,000-a-month drug like cetuximab, or to try something else because they know it’s doomed to fail.
“We’re very pleased that Roche has perceived the value of it and look forward to working with them in the coming months,” Taylor says. “It’s an important application for us, and we’re pursuing it aggressively.”
The company’s technology is called DNA-programmed chemistry. These molecules can be designed to be nimble enough to detect interactions between proteins in ways that other engineered biotech molecules can’t, Taylor says. It makes proprietary probes with fluorescent tags that will light up under a microscope and tell a researcher whether a protein drug like trastuzumab will bind properly with the special variety of protein, called a dimer, on that individual’s tumor cells.
It’s too early to say yet what this sort of information might mean, Livingston says. It could mean that researchers would suggest fewer patients get a certain drug, or might prompt recommendations that more get them. It might suggest a combination of drugs would work better than one alone, he says.
Roche clearly has a big interest in this field of personalized therapy, given its status as a leading maker of diagnostic tests and the majority owner of South San Francisco-based Genentech, which pioneered personalized therapy by introducing trastuzumab to the market in 1998. That drug is approved for only one-fourth of breast cancer patients who have a mutated form of the HER-2 gene. Animal tests have suggested that tumors with a slight variation of this gene, called a heterodimer, might be better equipped to develop resistance to the drug—which could lead researchers to think ahead about strategies to launch pre-emptive attacks on resistance with another drug, Livingston says.
