Commentary by Dario C. Altieri, M.D.
The recent series in the New York Times on personalized medicine is an excellent look at what is doubtlessly the future of cancer medicine. ("Genetic Gamble," see part one, part two, part three.) However, these articles tell only part of the story. Gene sequencing will be a powerful tool for cancer medicine, but it is just a tool. To create real personalized medicine, we need to know more about cancer biology and to design better drugs.
Personalized medicine is an exciting and scientifically valid concept. It involves decoding an individual’s DNA, seeing what genes or gene products (such as RNA or proteins) are involved in the disease, and devising concrete strategies to target that gene product. In theory, it means getting the right medicine to the right patient at the right time. Just as importantly, it means not giving the “wrong” medicine to the “wrong” patients. Personalized medicine allows physicians to determine which medications are unlikely to benefit patients, thereby reducing costs and avoiding the impact of side effects.
Indeed, personalized medicine will become the mainstream. As I write, medical centers around the world are investing millions of dollars in new gene sequencing technologies for this purpose. I think it is realistic to say that, in five years, patient tumor gene sequencing will become as common as a tissue biopsy is today. Have no doubt, this is coming.
But here’s the challenge. Personalized genetic data will amount to little if we cannot personalize a course of therapy. For it to work we need to find actionable mutations—mutations that we can target with existing or yet-to-be devised therapeutics. Unfortunately, many tumors don’t have actionable mutations, or at least none that we can immediately see. And among those that do, only about 10 percent of cancer genes are currently considered “druggable.”
In the last two decades we have seen the rise of incredible targeted therapeutics: drugs designed to specifically target tumors in cases of advanced cancer. Gleevec (imatinib), for example, may restore a normal life expectancy to some chronic myelogenous leukemia patients, provided they remain on the drug indefinitely.
In practice, these drugs are not perfect. That fact remains that cancer is a complex, heterogenous disease, where it is unlikely that a single agent alone can cure disease by shutting down a single chemical pathway. As recent clinical trials have shown, these drugs may merely put the tumor under great selective pressure, killing off some cancerous cells within the tumor, but allowing resistant cells to multiply. The cancer that comes back after such a treatment is usually much more aggressive.
How to Realize Personalized Medicine
The challenge for personalized medicine lies not with gene sequencing technology, but in the gaps in our knowledge. We still do not have a complete understanding of how the chemical signaling pathways work in the context of living tumors. Targeting more than one pathway at a time—so-called combination therapy—may limit the chances that a tumor may escape treatment. The problem is that we do not have a comprehensive understanding of how all of these pathways intersect.
Fortunately, the science is advancing steadily. A good example would be that of BRAF, a gene that is mutated in about half of melanoma patients and functions as a clear disease driver. Wistar’s Meenhard Herlyn, D.V.M., D.Sc., pioneered the research that led to the creation of therapeutic inhibitor of the activated BRAF gene. In clinical trials, this drug worked miraculously well, for a time. In nearly every case, though, the tumors return within a few months.
Herlyn wants to do better than a few months. He and his colleagues determined that tumors treated with BRAF inhibitors re-routed their signaling pathways, relying on another pathway to further their survival. In fact, they have shown that targeting melanoma cells—including melanoma cells taken from patients in the trial—with a second inhibitor overwhelms the tumor.
To make personalized medicine work, we first need to chart all the molecular pathways that promote tumor growth and identify molecules in those pathways that we may target with drugs. This is where basic research will save lives.
Secondly, we need more targeted therapies, as well as new drug technologies, so that we can hit targets currently considered “undruggable.”
There will be no magic bullet for cancer—no single “blockbuster” drug. Indeed, the notion is absurd. Cancer is a complex disease driven by genetic forces that continually change. The tumor seen at diagnosis is guaranteed to be different than the tumor that metastasizes at a later stage.
Instead, we need to broaden the portfolio of available targeted therapies that cut off cancer-promoting molecular pathways. For personalized medicine to work, oncologists need a broad spectrum of drugs at the ready. They need to know how—and when—they can apply combinations of these drugs to defeat the specific tumor of a specific patient.
Then, and only then, can they deliver the right drug to the right patient at the right time.
And only scientific discovery can make that happen.