June 13, 2007 | PHILADELPHIA — Experts in the field of biomarkers highlighted numerous ways that the growing deployment of biomarkers is boosting discovery and clinical research. A key message was the need for the concurrent use of target engagement and disease-related biomarkers.
Speaking on a panel of big pharma experts to wrap up CHI’s Biomarker World Congress*, John Wagner (senior director, clinical pharmacology, Merck Research Laboratories) noted that biomarkers are an “increasingly common” aspect of drug development programs. “In fact, it’s becoming rare not to [use them],” Wagner said. “But the strategic use of both of those types of biomarkers is really critical for high-quality decision making.
“If you achieve proof-of-concept with a disease-related biomarker, but you don’t know where you are in target engagement, you have not fully informed the dose range or fully understood the mechanism. Probably even more importantly, if you don’t achieve proof-of-concept with a new mechanism, and you don’t have a target engagement biomarker, you do not know whether the compound has failed or the mechanism has failed fundamentally, unless you make some other assumptions.” Such lessons, he said, were in many ways more important than finding ways to accelerate individual drug development programs.
One of the best examples Wagner highlighted was the ability of his colleagues to shave more than one year off the development of Sitagliptin, the hypoglycemia diabetes drug — a DPP4 (dipeptidyl peptidase-4) inhibitor — approved by the FDA last year. Wagner acknowledged that not all of those time savings could be attributed to strategic use of biomarkers — the program was “frontloaded in terms of other resources,” he said. But it did enable Merck to dispense with a phase IIA proof-of-concept study. “Skipping the 6 or 8 or 9 or 12 months it takes to run a Phase II study is obviously enormously beneficial,” said Wagner.
DPP4 biomarkers increased efficiency of early development for the diabetes drug Sitagliptin.
Jean Lee (scientific director, PKDM, Amgen) said, “We don’t actually wait until Phase II, we start with Phase 0. The preclinical model is very helpful. We run a lot of pharmacological studies in the preclinical model, surprisingly we find that if you have the right panel of biomarker, you can do quantitative pharmacology.”
Frank Dieterle (biomarker project manager, Novartis) said its, “amazing how technology has developed. Every year we come here, we see new methods or ideas becoming reality.” His enthusiasm for multiplex assays was echoed by the other panelists. “We don’t use single biomarkers, we use multivariate approaches.” said Russell Weiner (director, clinical biomarker development, Bristol-Myers Squibb).
But Amgen’s Lee stressed that, “It’s very important to understand the [underlying] biology. Timing is very important.” Expression markers can be transient, and while “some protein biomarkers have staying power, some don’t, so it’s really important to understand the biology.”
Ole Vesterqvist (director, biomarker laboratory, clinical translational medicine, Wyeth Research) said that the community has come a long way in learning how to validate biomarker assays. “But I see the biggest challenge still is still the interaction with the clinical teams... Today with the new technologies we can produce tons of data. How do you present it to the clinical teams that it makes sense?”
Panelists were asked to offer examples of best-case scenarios using biomarkers. Weiner noted that far from the “fail fast” mantra so beloved of biopharma, early biomarker data were also playing a key role in keeping certain drug development programs alive, in the midst of multiple programs competing for funding to enter the clinic. “Will the drug die before you get information? Biomarker data early on is keeping programs alive, and management is enthusiastically behind it, which wouldn’t have happened before,” he said.
Wyeth’s Vesterqvist offered a “fail fast” example that he said, “will never be published.” Two biomarkers used in discovery played a major role in terminating a once-promising program. As the drug candidate entered clinical development, the biomarkers went through rigorous analytical validation, but ultimately resulted in “flatlines for all dose curves.”
Despite the overall enthusiasm for multiplexing, Vesterqvist sounded a note of caution: “We do a lot of [internal] evaluation to choose which platform for which assay... In early development, what we need is precision and sensitivity, and sometimes those are jeopardized when we go to multiplexing.”
CHI will be holding three additional meetings on Biomarkers this fall:
Biomarker Discovery Summit; Philadelphia, Sept. 17-19, 2007
Implementation of Biomarkers in Clinical Trial Design; Toronto, Oct. 15-16, 2007
Biomarkers Europe; Vienna, Nov. 5-6, 2007
Subscribe to Bio-IT World magazine.