The billion-dollar model

FRAMINGHAM (03/22/2004) - Life is good at GenPath Pharmaceuticals Inc. right now. Last October, the company raised US$42.7 million in a Series B round and was awarded a patent (#6,639,121) for its inducible cancer technology. In December, it landed a mega deal with Merck that could ultimately be worth $100 million. The company's star-studded advisory board includes MIT Nobel laureate Bob Horvitz; Richard Klausner, former head of the National Cancer Institute; Tyler Jacks, head of the MIT Center for Cancer Research; and Ed Scolnick, president emeritus of Merck. And its VC lineup is no less stellar.

The cause for all the attention is GenPath's novel platform and approach to cancer therapeutics. In 1999, company co-founders Lynda Chin and Ronald DePinho, both of the Dana-Farber Cancer Institute, reported in Nature a system to reliably induce and control tumor growth in mice (Chin, L. et al. Nature 400, 468-72; 1999). This newly patented technology has been licensed exclusively to GenPath, which uses it to screen mouse tumors for genes associated with tumor maintenance, and identify targets and develop therapeutics.

"We believe our platform more faithfully represents tumor biology in vivo," says Tuan Ha-Ngoc, GenPath's president and CEO, "and we focus on tumor maintenance, which we believe is most clinically relevant. By the time a cancer patient presents to the oncologist, the induction phase is over. The genes that are essential for induction may not be the same genes essential for growth and maintenance."

The patented technique is clever yet straightforward: A tissue-specific oncogene -- the ras gene, for example -- controlled by a tetracycline promoter, is engineered into mice genetically engineered to lack the Ink-4 tumor suppressor gene (involved in melanoma). Adding doxycycline to the mice's drinking water turns on ras. Although this isn't enough to trigger cancer, it speeds cell proliferation and the effective rate of mutation. Exposure to UV light also accelerates tumor progression.

As the mice lack Ink-4, they are more susceptible to accumulated mutations, and develop tumors quickly. About 50 percent of mice develop tumors in two months, versus a year using more traditional methods. (Tumors also shrink rapidly when doxycycline is withdrawn, indicating the importance of ras in tumor maintenance.)

The result, GenPath says, is a mouse model of cancer that more closely mimics human cancer, not only because it highlights the importance of genes in tumor maintenance rather than induction, but also because of the natural interplay with the host immune system.

Explains Steven Clark, chief scientific officer: "A lot of people have taken human cancer cell lines and planted them in (immune-deficient) mice, and then studied their biology. But cancer is a disease of mutations and natural selection. Every tumor derived from a person -- although you may have breast cancer or prostate cancer -- is actually different because it goes through a different process of natural selection and mutation. There's also critical interaction between the host and the tumor as it evolves."

"(Chin) designed mice (in which) tumors could develop in a spontaneous way in a melanoma model. Tumors are being pushed to grow, but the activated ras isn't enough. You need multiple other mutations, probably at least a half-dozen to a dozen in one cell (for it) to become a tumor," Clark explains.

Clark and Ha-Ngoc believe this platform not only mimics cancer better than standard mouse models, but also will produce more novel targets and allow faster validation of targets. The company already has a promising lead compound that targets a metabolic enzyme called GP1 that it hopes will be ready for critical toxicology testing by the end of this year. Another 32 "tractable candidate genes" have been identified and are waiting in the wings.

Man With a Plan

Clark and Ha-Ngoc were lured to GenPath following lengthy careers at Genetics Institute (Clark was employee number two at GI). Both stayed on for stints as senior managers when Genetics Institute was acquired by American Home Products and reconstituted as Wyeth. They left Wyeth at different times. "We made Genetics Institute's financial backers very happy," Ha-Ngoc says. Those same backers enticed Clark and Ha-Ngoc out of early retirement in the spring of 2002. GenPath has since grown to 60 employees, and expects to grow to 100 this year.

"We provided the board with a 10-year, US$1-billion plan," Ha-Ngoc says, "and expect to be profitable in year nine, essentially with the launch of the first product." About $300 million to $400 million will come from equity, with the rest from industry deals such as that signed with Merck. GenPath has also struck a partnership with Array BioPharma in Boulder, Colo., to provide X-ray crystallography and medicinal chemistry screening of its drug candidates.

Pressure will undoubtedly increase as the first compound nears critical testing. "You cannot fight against the attrition rate," Ha-Ngoc concedes. "We could be lucky and our first child could go all the way to approval, but chances are you will need more children!" Indeed, Clark and his team tried to kill GenPath's first offspring quickly. "I thought it was a stupid target," admits Clark, whose team ran a battery of "killer tests" to dismiss the target -- but nature didn't cooperate.

"Once you start thinking about it," Clark says, "(GP1) turns out not to be such a crazy idea, because obviously these tumors are rapidly growing. This enzyme turns out to be highly upregulated in a number of different tumors, not just melanoma. We think there certainly will be toxicity associated with blocking it."

While hope is growing for GP1, the company's big bet is on the cancer-inducing technology platform. Chin and DePinho's patent covers the broad use of activated oncogenes to control tumor growth in genetically susceptible mammals.

"I'm always worried about one-product companies," Ha-Ngoc says. "How do you generate corporate partners? If you only have one or two compounds, you can only have (one or two) partners. That's it -- you don't have anything to show investors. We have the Merck deal, but there's still plenty of room to do more ... to fund the company and mitigate risk."

GenPath has developed three other cancer induction models, including an HER-2 breast cancer model. It is also working extensively with RNAi, and developing models in which RNAi expression can be controlled with a linked promoter just as in the ras model. It's easy to imagine such a controllable RNAi system being used to validate specific targets at different stages of disease.

Located in Technology Square in Cambridge, Mass., GenPath hopes the success of its illustrious neighbors, such as Novartis and MIT, rubs off. "We have aggressive goals," Clark says. "We'd like to have a clinical candidate by the end of this year. We're to the point where we are studying a proof-of-principle molecule in our own spontaneous tumor models and in a variety of explant models. I think we'll get there."

SIDEBAR

GenPath Power Points

Focus: Cancer drugs developed using cancer induction model in mice

Scientific founders: Lynda Chin and Ronald DePinho

Key execs: Tuan Ha-Ngoc, CEO; Steve Clark, CSO

Key Investors: MPM Capital, Flagship Ventures, Venrock

Key Collaborators: Merck (drug development), Array BioPharma (medicinal chemistry)

Current staff: 60

Join the newsletter!

Error: Please check your email address.

More about Array BioPharmaDana AustraliaInterplayMimicsMITMPM Capital

Show Comments

Market Place

[]