Chris Anderson

As therapeutics become more precision targeted, the practice of clinical medicine is moving away from one size fits all, to one size fits one.

As recently as ten years ago, just after completion of the Human Genome Project, the concept of a personalized approach to cancer treatment was largely just that—an idea of what could be. And companion diagnostics were limited to predicting patient response to Herceptin for breast cancer and Gleevec for chronic myelogenous leukemia.

But it was the introduction of these therapies along with the nascent field of gene sequencing that began to move the treatment of cancer from a broad-based, one-size-fits-all approach to one that seeks to better understand the unique underlying molecular pathology of each patient’s cancer.

“In the 1990s the pharmaceutical companies argued that because the cancer population is small the only way you could develop a drug and profit from that was by using a broad spectrum chemo approach,” said Richard Ding, CEO of bioTheranostics, a provider of prognostic and diagnostic cancer tests. “Because then you can hit on multiple tumor types and have a broad cluster.

“I think the poster child for changing this was Herceptin. It was able to demonstrate that even though the target population for breast cancer of 50,000 or 60,000 was once considered too narrow, that even with the complexity of making monoclonal antibodies, it showed you can make a business out this.”

Precision Medicine Hits Its Stride

Based on the excitement generated by the first human genome sequencing and the development of a small handful of cancer therapeutics with companion diagnostics, there existed a naïve optimism in the early 2000s that the industry would be able to rapidly develop and commercialize highly targeted therapeutics based on the newfound knowledge of our DNA. In reality, researchers found that they hadn’t received the key to the precision medicine treasure chest, but instead had merely cracked open the first of a series of Russian nesting dolls and that more hard work lay ahead.

“There were a few examples and discoveries that were intriguing, but they were the exception,” said Brad Gray, president and CEO of NanoString Technologies. “I was involved with the Personalized Medicine Coalition and over and over again at meetings we would talk about the same three examples and, quite frankly, it got boring. People stopped going to meetings because not much was happening.”

The same can’t be said today, as research in those early days is now yielding an increasing flow of cancer therapeutics and diagnostics that allow clinicians to more accurately target therapies to treat a broad range of cancer subtypes.

As Simon Stevens, the new chief executive of England’s National Health Service (NHS), noted recently in The Guardian: “As we are discovering with cancer, what we once thought of as a single condition may be dozens of distinct conditions. So common diseases may in fact be extended families of quite rare diseases.” As Steven sees it, the advances in our understanding of each person’s individual genetic makeup—and the mutations that play a role in cancer—are moving treatment “from carpet-bombing to precision targeting. From one size fits all, to one size fits one.”

Anthony Mato, M.D., director of Center for Chronic Lymphocytic Leukemia (CLL) at the University of Pennsylvania, is living this rapid transformation each day. He says rapid improvements in cytogenetic, sequencing, and immunohistochemistry technologies have helped create new models of the hematologic cancers he treats. It is this better understanding of the biology of individual tumors that shows how one patient might fail a conventional therapy and allows him to alter therapy based on disease-related biology.

Dr. Mato notes that ten years ago traditional cytogenetics might have identified a small percentage of acute myelogenous leukemia patients with poor risk while the majority of patients would show a normal karyotype. With the technology and prognostic and diagnostic tests available today, he can see with much greater specificity the patient’s risk. “Today we are able to use molecular studies, mutation analyses, for example looking at mutations in FLT3 and MTM1, and we are able to risk stratify the patients who are so-called intermediate risk by conventional cytogenetics into a favorable or unfavorable risk group and base treatment on that information.”

Next-generation sequencing is also playing a growing role in helping clinicians provide highly targeted therapies. While still relatively expensive, cost has come down dramatically in the past five years and should continue to drop.

“Even just two years ago the ability to take someone’s tumor to look at driver mutations by sequencing was an almost unheard of technique as a part of general practice,” Dr. Mato said. “Now anyone can take a tumor sample and call up one of the companies that are able to analyze the tumor and look for hundreds of driver mutations.”

As a clinician, Dr. Mato is also fortunate to have in-house sequencing at UPenn’s Center for Personalized Diagnostics, which has a guided mutation detection system that can look for hundreds of known mutations in both hematological and solid tumor samples. Quite often, he will choose this route for patients who are not responding to traditional therapies. “We can now look genome wide for mutation in our patients’ tumor samples and that helps us to guide clinical decisions. This is definitely the future.”

Leveraging Diagnostics and Prognostic Tools

As pharmaceutical companies discover new targets for oncology therapies it is almost a given today that development of a new drug will go hand-in-hand with the development of a companion diagnostic. “It is exceptional to have targeted therapies developed without some kind of biomarker to select patients for that therapy,” said Gray. He added that the FDA sees the value of companion diagnostics and is encouraging companies to continue biomarker research. “They have worked to clarify the regulatory path for developing these precision medicines, and there are more and more positive examples of these coming to market with diagnostics.”

NanoString has set itself up to take advantage of these favorable conditions, as evidenced by its recent collaboration with Celgene. The agreement allows Celgene to use NanoString’s nCounter analysis system to support the clinical validation and development of a companion diagnostic for Revlimid, for treatment of diffuse large B-cell lymphoma, a potential new application of the drug. In this way, diagnostic technologies are helping to feed a continuous cycle of drug development in addition to improvements in patient care. As noted by Jean-Pierre Bizarri in announcing its collaboration with NanoString, “Biomarker-driven clinical trials are the future of clinical oncology.”

While many diagnostic and prognostic tests are tied directly to a specific drug, other companies like bioTheranostics provide tests designed to more broadly identify cancer subtypes or to stratify patient populations based on the specific biology of their disease. These tests are intended to help doctors chose—from among the array of therapies available—those that should be most effective for each patient. After bioTheranostics analyzes the tumor it produces a report that suggests a treatment path with the understanding that most physicians aren’t trained molecular biologists.

“Actionability, simplicity, and cost efficiency are three elements oncologists want to see: what information can lead to a therapeutic intervention, what drugs to use, and those type of things, we want to make sure it is simple,” Ding noted.

While clinicians want to have this information at their fingertips, the road to commercialization can be more difficult than it is with companion diagnostics. That’s because getting insurers to pay for a companion diagnostic is easy considering it is a relatively small expense in the overall cost of the drug and is hence “baked in”. Stand-alone diagnostics face greater scrutiny since insurers often see them as a cost center outside the cost of treatment. “What diagnostics companies need to do is show themselves as an enabler of the right therapy, a vital part of delivering personalized medicine and that it has value beyond just the diagnostic itself,” Ding said.

For physicians like Dr. Mato having more information from these tools as well as multiple options for treating cancer are what makes delivering precision medicine such an exciting field right now. “In terms of treatment options, it is unbelievable,” he concluded. “A couple of years ago, it was a handful of agents. In the next couple of years there are going to be probably ten agents available for CLL and it will only grow. But (picking the right agent) is only going to get more and more complicated.

“If you take prognosis by itself, whereas before there were few markers that served to prognosticate patients, we are going to go to hundreds in the next couple of years. Clinical medicine is now going to need to figure out how to decide how to use those for individual patients.”


Celgene is using NanoString’s nCounter analysis system to support the clinical validation and development of a companion diagnostic for Revlimid, for treatment of Diffuse Large B-Cell Lymphoma.

Chris Anderson ([email protected]) is the former chief editor of Drug Discovery News, which he helped launch in 2005.

This article was originally published in the June 26 issue of Clinical OMICs. For more content like this and details on how to get a free subscription to this new digital publication, go to www.clinicalomics.com.

Previous article“Most Wanted” Microbes Sent to Solitary Confinement
Next articleNew Technology Traces Entire Life of a Cell