Companion diagnostics require hitting the jackpot twice
When it comes to personalizing medicine, the pairing of diagnostic tests with drug therapies to select the patients who will benefit from treatment, Roche Diagnostics is an acknowledged leader.
So it was a packed conference room that faced Christian Meisel, the head of Oncology and Translational Medicine for Roche as he announced, “Personalized medicine is not hype anymore, particularly in cancer.”
This workshop session led off BIO-Europe 2010, starting the day for dealmakers with a head-on discussion of the hottest issue on the minds of the 3,000 life science business development executives who gathered in Munich for Europe’s largest partnering event.
At the end of the day it was less the answers than the questions raised during the panel discussion, ‘From Biomarkers to Companion Diagnostics: The Emergence of Personalized Medicine in Clinical Development and Oncology Dealmaking,’ that resonated with business development executives trying to plot strategy as they went into one-to-one meetings with potential partners.
In the session Meisel did not disappoint in his presentation, sharing lessons learned and insights for the double edged process of developing a drug with a companion test.
“The reality of personalized medicine for a pharmaceutical company is that they must hit the jackpot twice, first with an effective drug therapy and then with the companion diagnostic to prove that effectiveness,” said Meisel in what was probably the most repeated comment of the day.
“Roche is seeing the benefits of its long commitment to creating companion diagnostics for its therapies,” he said, citing results from a recent Phase I clinical trial where 80 percent of patients demonstrated a rapid response to the candidate treatment with substantial tumor regression in visceral organs.
Yet the study also clearly showed the patients who did not respond, and here Roche benefited from what Meisel called an in-house capability for integrated and iterative drug development.
“We are able to bring the data on failed patients back into the preclinical development cycle to find a treatment that will work,” he said.
Cecilia Schott, AstraZeneca is partnering with other companies for diagnostics
The Business Development Director for Personalised Healthcare at AstraZeneca, Cecilia Schott, acknowledged that unlike Roche or Abbott Laboratories, “We are not a diagnostics company, so we are partnering with other companies for diagnostics.”
As an example she cited the agreement in January, 2010 between AstraZeneca and the Danish firm Dako to develop companion diagnostic tests for multiple oncology projects in various stages of discovery and development.
Linda Pullan, the President of Pullan Consulting, reviewed a variety of business models that are emerging for meeting the challenge of delivering tests for comparative effectiveness along with therapies.
“Roche’s model of developing tests to promote analysis instruments it sells to clinical laboratories along with the reagents used in the processes, actually delivers the lowest percentage of gross revenue per sale,” she said.
Dako, which sells direct to labs wins a larger percentage of revenue as its assays are included on the label for a specific therapeutic and drug sales, drives diagnostic sales.
Geonomic Health, which deals directly with physicians and has won around 7,500 clinicians to date, runs its own tests in CLIA labs and earns a gross profit of 71 percent from each USD 4,000 test.
Joe Monforte, AltheaDx
Joe Monforte, CSO for Althea Dx, which regularly partners with both pharma and biotech companies to develop assays, said that “genomic testing is far behind the development curve.”
“We are living with classic testing methods using formalin-fixed, paraffin-embedded (FFPE) tissue samples that do not fix RNA and DNA, and which actually degrade it,” he said.
“We are being asked to extract dramatic, high content data from a methodology developed 150 years ago,” he said, adding, that it demands highly specialized knowledge to optimize these widely used tests.
“Because the installed base of this analysis equipment is so broad it represents a critical requirement for developing diagnostics that will meet with ready acceptance,” he said, “and despite the inadequacy of the method such testing can be expected to continue for another 20 years.”
John Freshley with Compendia Bioscience suggested that the next cure for cancer will be developed with information, and not through chemistry.
Hosting a database built on 40,000 patient tissue samples, Compendia’s Oncomine now features tools for predicting effectiveness of drug candidates and tumor segregation panels.
John Freshley, Compendia
The moderator for the panel discussion, Freshley offered commentaries and a long list of unanswered questions essential to guiding business development decisions and strategy formation moving forward into the brave new world of companion diagnostics.
“One of the challenges in doing deals for companion diagnostics is how to share the risk of developing a companion diagnostic for a drug that may fail,” he suggested. “Companies need to be prepared to deal with this situation.”
“Many small biomarker and diagnostic companies, including my own,” he said, talk about retaining commercial rights to biomarkers that they discover in collaborations. We need to consider the challenges for companies that wish to retain these rights.
“Some successful examples of companion diagnostics and personalized medicine involve the measurement of a single biomarker,” he said, “but it seems that many of the newer studies are demonstrating utility of multiple biomarkers being applied together, for example one measuring sensitivity and one measuring resistance.”
“Here, there are unique challenges that come with trying to measure multiple markers from the same patient.”
Finally, he agrees with Monforte that “tissue acquisition remains as a critical challenge to the development of biomarkers at this stage in the evolution of the industry, as well as constraint to readily applying biomarkers in a real world clinical setting.
Related Posts
No related posts.