By Charles Swanton

Over the past two decades, dramatic progress has been made in many areas of medicine where hope once seemed out of reach. An infectious disease like HIV, considered intractable in the 1990s, is now controllable with anti-retroviral therapy; and the right combination of antibiotics can similarly control many difficult bacterial infections.

When it comes to cancer, however, progress in managing solid tumors that have spread beyond their primary site (metastatic disease) has been rather limited. Survival gains from new therapies are often measured in weeks or months, rather than years, and this oncological care comes with a disproportionate rise in costs.

Cancer has justifiably been referred to as the "Emperor of All Maladies." And indeed, cures at the metastatic stage of disease despite some notable exceptions, like testicular cancers are still rare, with tumors inevitably acquiring resistance to multiple drugs over the course of the disease. But today, the science of genomics is showing great promise for cancer research and treatment. New genomic sequencing technologies, which allow us to look at the genetic makeup of tumors, are shining a bright light on the way cancers grow and spread, helping oncologists to better understand why progress in metastatic disease has been so slow; how tumors manage to evade drugs and develop resistance to therapies so rapidly; and ultimately, how we might best treat them.

What's more, sequencing of single cancer cells is beginning to reveal that no two cancer cells share identical genetic codes. Even a small cancer mass measuring 1 to 2 cubic centimeters may contain billions of cells. The potential for genomic diversity is therefore extraordinary. We don't yet understand exactly how this diversity affects drug resistance and treatment outcome, but advanced sequencing technologies hold the key to these critical questions.

Cancer as Evolutionary Process

Why is this relevant? From a reductionist standpoint the approach or belief that the complex mechanisms of life can be understood by simple chemistry the internationally renowned cancer biologists Peter Nowell, Carlo Maley, and Mel Greaves have proposed that cancer growth follows the laws of evolution and selection. These scientists reference the "I think" branched tree diagram that Charles Darwin drew in 1837, in which the genetic diversity of species is represented by branches that grow off a shared evolutionary trunk.

The thinking goes that genetic diversity with cancer subclones following a branched evolutionary path that creates distinctions between them helps cancer survive, providing the necessary substrate for its evolutionary fitness. While many cancer subclones will be eliminated during therapy, only a few need to survive to result in the rapid acquisition of drug resistance over weeks or months that oncologists witness in clinical practice. Indeed, genetic diversity within glioblastoma, the most common and insidious malignant brain tumor, has been shown to result in distinct populations of cells in the same tumor, with sensitivities to different drugs. This may begin to reveal why treating solid tumors can be so difficult, and why resistance to treatment in metastatic disease seems so inevitable.

In the near term, it may not be possible to cure most advanced metastatic tumors using traditional targeted therapy approaches, because of both the genetic distinctions between patients (intertumor heterogeneity) and within individual tumors (intratumor heterogeneity). The range of targeted drugs needed to treat the multitude of genetic dependencies in many advanced tumors simply does not exist and even if it did, combining these targeted drugs at a safe and effective dose would be extremely challenging.

Using Evolutionary Principles to Achieve Results

So, new approaches are required. One such approach harnesses the body's own immune system to recognize tumor diversity, targeting the genetic abnormalities in cancer cells. Tumors appear to dampen the body's own immune system by "stealth," preventing the body from recognizing the abnormal protein signals cancer cells display on their surface, and therefore identifying them as rogue cells. New cancer treatments are increasing the sensitivity of the body's own immune system so it can detect these rogue cells. Such developments are already showing impressive benefits in clinical trials, with tantalizing evidence of long-term disease control, something we rarely see with traditional chemotherapy or targeted drugs.

Another logical goal is preventing tumor diversity. One way to achieve this will be by improving traditional modern oncology through early diagnosis, tumor screening, and prompt and aggressive surgery and adjuvant (chemotherapy or radiotherapy) treatments following surgery. The mantra: hit the tumor hard and early before it has grown, spread, and diversified. Genome sequencing reveals a scientific basis for why this approach which has been so beneficial to patients over the last three decades is resulting in many more cancer cures than ever before.

Understanding the Genetic Complexity of Tumors

Current genomic sequencing technologies enable us to decipher the genetic code of human cancers at an unprecedented rate: overnight in one laboratory, compared to the several years it took in the late 1990s. Genome sequencing reveals that every tumor is distinct from patient to patient, with limited but important genetic changes shared between patients who have the same pathological tumor type.

Large-scale cancer sequencing efforts conducted by organizations like The Cancer Genome Atlas (colorectal cancers) and the Wellcome Trust's Cancer Genome Project (breast cancer) have already led to breakthroughs in our understanding of human cancers. For example, Dr. Andy Futreal, professor of genomic medicine at the University of Texas's MD Anderson Cancer Center, together with his colleagues, identified a cancer mutation in a gene called BRAF that has led to one of the most exciting and effective treatments "targeted" against this mutation in melanoma, a disease against which chemotherapy was almost always ineffective. (Mutations of the BRAF gene are often found in melanoma. Approximately 160,000 new cases are diagnosed worldwide each year, and the disease is responsible for 75 percent of all skin cancer-related deaths.)

Genomic sequencing is revealing unexpected complexities in tumor development. Cancer is a clonal disease, meaning that one precursor cell spawns daughter progeny that grow in an uncontrolled manner, and resist destruction by the body thanks to multiple gene mutations (or other modifications). Sequencing is revealing that these "subclone" cells may, in some cases, dominate the metastatic site of disease, and resist treatment through ways we have yet to understand.

Genetic sequencing is also challenging traditional concepts of tumor growth and evolution. We now know that tumors evolve and change over time, and that different subclones of tumor cells, with shared genetic mutations but distinct genetic makeups, may reside in different parts of the same primary tumor. In fact, these distinctions between subclones may result in more differences in cancer DNA sequences than similarities.

Finding New Targets for Drugs

By harnessing the laws of evolution and thinking of tumors like the tree of Darwin's "I think" diagram, with a shared genetic trunk and differing branches oncologists may be able to develop better drugs for patients with metastatic disease. Genomic sequencing can give us a more intricate understanding of the clonal origins of cancer cells, and help us to define the primary drivers of cancer growth particularly the core genetic dependencies of cancer cells, and the early genetic events that will be present in every tumor subclone at every site of disease. These shared drivers from the "trunk" may be better drug targets to control disease everywhere. Indeed, from sequencing we may learn that the highly effective drug targets identified over the last decade are targeting early "trunk" events in the tumor, as in the cases of Her2, the targeted gene of herceptin in breast cancer, and the BRAF mutation in melanoma.

In this regard, genomics is already being implemented in cancer treatments, through the sequencing of specific mutations in the cancer-signaling proteins of some sarcomas (c-KIT and gastrointestinal stromal tumor), melanoma (BRAF), and lung cancer (epidermal growth factor receptor). In clinical trials, Next-Generation Sequencing technologies are looking at tumors before treatment and after the tumors acquire drug resistance, helping us to understand not only how drugs work, but also why they stop working.

Understanding the "branches"analogous to the differences between cancer cells will give oncologists a better idea of how cancer drug resistance may occur and how we can prevent it. For example, we already know, in lung cancer and other solid and haematological (blood, and blood-forming tissue) tumors, that drug treatment can select out a resistant subclone that may exist as a small subpopulation of genetically distinct cells in the tumor before therapy. This is akin to pruning the tree branches, leaving one or two branches of the tumor tree left to grow and dominate the disease bulk.

What is now emerging through sequencing studies is just how complex this process can be, and how many branches the tumor tree may have. In one tumor, the data reveals multiple drug resistance mechanisms existing in minor but distinct cancer subclones that emerge during therapy. Clearly, understanding how such diversity develops is critical in order to try to limit its emergence in the first place. Continuing the tree analogy, we must understand processes within the cancer cell that initiate the branching of the tree, and the number of branches the tree bears.

Diversity could theoretically be the result of mutations that occur every day in normal cells. However, many cancers may generate diversity through elevated gene mutation rates or by failing to accurately separate the genes contained within chromosomes every time a cancer cell divides, termed chromosomal instability. Genomic research is already giving scientists a better understanding of these processes. And once they are clarified, we may actually be able to exploit and harness the very mechanisms that tumors depend upon for survival in order to find new targets for drugs to limit tumor evolution, stop them in their tracks, and destroy them.

Charles Swanton is the head of Translational Cancer Therapeutics