Personalized medicine is partially premised on the ability of a specific drug to effect change in a specific biomarker. One example of the importance of biomarkers in personalized drug development comes from Millennium: The Takeda Oncology Company, Cambridge, Mass. “In order for any of our drugs to move into clinical development, there must be a biomarker plan … to determine which biomarkers can be used in clinical development to aid in decision making or to help in patient selection,” says Nancy Simonian, MD, chief medical officer of Millennium. “In the case of some of the markers we are trying to develop, one has to develop appropriate assays and evaluate those assays in vitro and in vivo prior to going into the clinic.”
These assays typically come in the form of genetic strategies, which also make up the basis of the personalized medicine approach, a fact that Millennium illustrates quite well. “As we are preparing the [Investigational New Drug] application, we try to test our compound on a relevant cell type and then go ahead and do a global RNAi screen to ask for genes that increase the sensitivity of that particular cancer cell to that drug, or conversely other genes that confer resistance to our compound,” says Joseph B. Bolen, PhD, chief scientific officer of Millennium, who adds that this genetic screen produces a list of candidate genes that are then tested against a broad array of cell types using the same type of genetic strategy.

click to enlarge Source: Interleukin Genetics

Another major pharmaceutical company playing in the personalized medicine space is Johnson & Johnson (J&J). “We apply the pharmacogenomics approach to almost every compound we develop in a variety of therapeutic areas of interest for J&JPR&D [Johnson & Johnson’s Pharmaceutical Research & Development],” says Nadine Cohen, PhD, head of pharmacogenomics and senior research fellow for the J&JPR&D division, Raritan, N.J. In typical personalized medicine fashion, Cohen’s group creates subsets of patient populations to gauge how their genetic makeup might affect their response to specific drugs in various therapeutic areas including metabolic, psychiatric, and neurologic disorders. Over the last nine years, the group has collected DNA samples from clinical trials in most of these areas. “These samples are analyzed using various approaches depending on the disease or drug under study and the clinical issue to be addressed (efficacy, safety, or dosing), and the number of samples available and whether or not we have a hypothesis to test,” says Cohen, who adds that these samples have enabled the group to generate a mass of information on J&J compounds, having been genotyped for single-nucleotide polymorphisms (SNPs) covering several hundred genes relevant to the pathophysiology of the disease or drug under study. These data are used to correlate individual SNPs or combinations of SNPs with clinical end points (efficacy, safety, disease) using statistical analysis, which can be performed prospectively (hypothesis-testing) or retrospectively (hypothesis-generating).

The J&JPR&D division has collected and stored approximately 10,000 DNA samples from patients participating in clinical trials for two of their psychiatric drugs (Risperdal and Pariperidone), in order to identify markers of efficacy and adverse events. All of this helps Cohen’s team move toward their goal of better understanding their current compounds and improving the effectiveness of future compounds. They have also done extensive work in the area of Alzheimer’s disease and drugs such as galantamine (Razadyne). “Our data confirmed a strong association between ApoE4 homozygotes and age at onset of AD, but did not support an effect of ApoE4 allele copy number on rate of cognitive and functional decline, nor on the efficacy of galantamine in patients with the disease,” says Cohen.

J&JR&D employs many approaches to identify these disease-specific genes. “Candidate gene approaches are used to test hypotheses in which genetic polymorphisms within only a selected group of genes are related to drug mode of action, pharmacokinetics/pharmacodynamics, and disease pathophysiology,” says Cohen. In contrast, genome-wide screening approaches are performed to generate hypotheses using large, multiplexing SNP chips from banked samples for retrospective studies. Other methods employed include next-generation sequencing for identification of rare variants, copy number variation, DNA methylation studies.

Beyond the drug 
Another personalized medicine strategy is to develop diagnostic tools for the purpose of predicting success of a therapeutic. Genomic Health (Redwood City, Calif.) is a developer of personalized diagnostics for personalized medicine strategies. “From our point of view the business strategies of personalized medicine should emphasize the strength of the diagnostic as well as the strength of the drug,” says Steven Shak, MD, chief medical officer of Genomic Health. Their latest product—the Oncotype DX—is a laboratory-based diagnostic assay based on a panel of breast cancer genes. “The diagnostic predicts whether the standard chemotherapy regimens would work in newly diagnosed breast cancer patients.”

The personalized medicine strategy employed by Interleukin Genetics (Waltham, Mass.) involves collaboration with pharmaceutical and biotechnology companies in the late stages of development, or even post-marketing, of a compound. The main goal of this strategy is to identify non-responders to a drug and then subtype it using genetic tests, thereby increasing the probability of a drug’s effectiveness. “We can help a company’s compounds become first- or second-line therapies to identify an effective drug for a targeted population much earlier in the process, thereby saving the insurance companies money and getting the right drug to the patient in a much more effective manner,” says Lewis H. Bender, MS, MBA, chief executive officer of Interleukin Genetics, who adds that becoming the higher order therapy, in terms of approval, can significantly increase the commercial success of a drug.

Forward-looking statements
Although there have been many strategies employed in the world of personalized medicine, will they be successful in creating a bright future for the approach?

“I think personalized medicine will become the way people think of discovering and developing new drugs,” says Simonian. “That doesn’t mean that for every drug, we’ll be able to identify the optimal patient population because the success of this strategy depends a lot on the targets and pathways that you are impacting.”

“We think that the next wave of important drugs will come in the form of treatment for polygenic, complex diseases of aging,” says Bender. “Complex diseases are polygenic and many different affected pathways can lead to the same phenotype.” For example, inflammation from osteoarthritis can be due to over-activity of multiple cytokines such as tumor necrosis factor alpha or interleukin-1.

“People always ask: ‘has the promise of personalized medicine been fulfilled?’ And I think five to 10 years ago, there was irrational exuberance over the pace by which this was all going to happen,” says Simonian. “They figured that since we can sequence genes, we’ve solved all the problems. I think there is no question that the promise is there, and that there are a growing number of examples where one is able to segment patients and optimize the way the drugs are used, and I think in the future we are just going to see more and more of these examples.”

About the Author
James Netterwald is president and CEO of BioPharmaComm LLC, a provider of writing, editing, and consulting services to the life science, pharma-biotech, and public relations industries.

This article was published in Drug Discovery & Development magazine: Vol. 12, No. 9, October, 2009, pp. 10-12.