SatNav for Oncology Developers May Help Celleron Steer AZ's HDAC Inhibitor

Ever get lost? Get one of the many satellite navigation systems for your car or even your phone. But what happens when oncology drug developers are hitting dead ends in the clinic, even with the most promising cancer candidates? Celleron Therapeutics may have just the technology to put drugs on the right path-and now it has a high-profile clinical compound to drive through the clinic. In May 2009 the biotech landed worldwide rights to AstraZeneca's lead histone deacetylase (HDAC) inhibitor, AZD9468, adding to the list of pharmaceutical players who have opted, for now, to exit the promising but tricky HDAC space.

Ever get lost? Get one of the many satellite navigation systems for your car or even your phone. But what happens when oncology drug developers are hitting dead ends in the clinic, even with the most promising cancer candidates?

Oxford spin-out Celleron Therapeutics Ltd. may have just the technology to put drugs on the right path—and now it has a high-profile clinical compound to drive through the clinic. In May 2009 the biotech landed worldwide rights to AstraZeneca PLC’s lead histone deacetylase (HDAC) inhibitor, AZD9468, adding to the list of pharmaceutical players who have opted, for now, to exit the promising but tricky HDAC space. [See Deal] Terms have not been disclosed but include milestones and royalties payable to AZ, as well as "additional share of financial proceeds in the event of successful commercialization by a third party," according to the release announcing the deal. AZ also has a right of negotiation to re-license the compound, but Celleron founder Nicholas La Thangue, PhD, chair of cancer biology at the University of Oxford, describes that as "a quick look. If we want to sell the product, they’re just asking if they can see it in parallel, but we set the terms." Though the biotech has sold equity stakes to other partners, La Thangue and a spokesperson for AstraZeneca would not comment on whether AZ took a stake in Celleron.

The company’s predictive biomarker platform, CancerNav, is designed to determine which tumors will respond to which new or existing drug treatments. The technology identifies genes that regulate the sensitivity of tumor cells in vitro to a particular drug of interest, explains La Thangue. "It’s a genetic loss-of-function screen, and we can show that genes we get out of the screen really do regulate sensitivity to drugs because we can manipulate their expression levels," he says. Unlike microarray expression profiling built on statistical analysis, CancerNav also delivers a functional biomarker for the killing effect of a drug. Once Celleron identifies the appropriate gene, it expresses the protein and makes an antibody to that protein. "And then we use that antibody to screen a repository of tumor biopsies—we have thousands of those here in Oxford," he says. "And that tells us where the biomarker is expressed, and therefore you can wind your clinical development plan around that information."

The platform has helped Celleron put together a small but growing pipeline of preclinical and Phase I oncology candidates. The approach is quite different from the small handful of innovative systems biology-type strategies emerging in cancer—such as the ones undertaken by the newly formed Sage Bionetworks consortium and Nodality Inc.—but La Thangue argues that technologies that rely on data mining don’t always mine relevant data for tumor response mechanisms. When a CancerNav biomarker is expressed in tumors, he says, "one can say with some confidence that the pathway that the drug targets is intact." (See "A Friend-ly New Resource for Biomarker Validation?" START-UP, March 2009 (Also see "A Friend-ly New Resource for Biomarker Validation?" - Scrip, 1 Mar, 2009.).)

The AstraZeneca compound, now called CXD101, was essentially IND-ready, says La Thangue. "This is a robustly generated large pharma product," he says. "They had done absolutely everything except go into patients. They even put the investigators’ brochure together." La Thangue chalks up Celleron’s successful in-licensing to a familiarity between the companies—AZ has ongoing relationships with Oxford in a variety of areas—and Celleron’s technology platform and clinical infrastructure. AZ was repositioning its cancer pipeline, "and they had this product and didn’t know where to put it, we had our biomarker, so eventually they said, ‘take it,’" says La Thangue.

From Celleron’s perspective, HDAC inhibition had been on its radar screen since its inception in 2005. La Thangue previously founded and was CSO at the UK oncology-focused biotech Prolifix, which was acquired in 2002 by the Danish biotech Topotarget AS. [See Deal] That deal was in large part driven by belinostat, an HDAC inhibitor discovered and developed by La Thangue’s team that at the time was about to enter the clinic. Once in Phase I, belinostat was licensed to CuraGen Corp. [See Deal] The 2004 deal included $10 million in up-front payments and equity purchases. But perhaps illustrative of the difficulties of HDAC development, four years later the compound—then in Phase II development—was bought back by TopoTarget for $39 million in up-front cash and equity. [See Deal]

Toxicity issues have sunk a variety of pharma and biotech HDAC projects, others have faced poor solubility or half-life issues. But more frustratingly, "it’s unclear where you apply HDAC inhibitors in a clinical context," says La Thangue. "Which of the 300 types of cancer do you target? Invariably people get that wrong and this is why so many products fall out at the end of Phase II." Celleron had to be content with getting its hands on an earlier-stage asset. "We went searching for a product that had a superior profile but that was a little behind in the race, clinically," says La Thangue. Existing HDAC inhibitors also tend to target multiple HDACs—there are 18 such proteins in three classes—and they are involved in many aspects of developmental biology. (See "A New But Familiar Drug for Treating Memory Impairment," START-UP, May 2009 A#2009900119].)

In fact HDAC inhibition is increasingly a biotech endeavor after several pharma compounds have fallen out of development in the past few years; others, like CXD101 and entinostat—which was licensed by Bayer HealthCare Pharmaceuticals AG to Syndax Pharmaceuticals Inc. in 2007—have been out-licensed. [See Deal] (See "Syndax Leverages Novel IP for Cash, Phase II HDAC," START-UP, May 2007 (Also see "Syndax Leverages Novel IP for Cash, Phase II HDAC" - Scrip, 1 May, 2007.).) Only Merck & Co. Inc.’s vorinostat (Zolinza), acquired in Merck’s $145 million acquisition of Aton Pharma Inc. [See Deal] has reached the market, where it is approved to treat cutaneous T cell lymphoma (CTCL). Gloucester Pharmaceuticals Inc. is next up with romidepsin (which it licensed in from Fujisawa, now Astellas Pharma Inc., in 2004 [See Deal]), which has a pending NDA in CTCL with a November 2009 PDUFA date, and several other companies including Syndax, TopoTarget, and Novartis AG’s Novartis Pharma AG are in Phase II studies in a variety of tumor types. Zolinza hasn’t exactly set the world on fire, either. Yearly sales have increased slowly from $11 million in 2007 to $20 million projected for 2009.

Meanwhile, CXD101 will enter the clinic later this year, says La Thangue, though he declines to say more about the company’s clinical plans for AZ’s castoff, for now. Why give up the edge?

—Christopher Morrison