Every person’s fingerprint is unique. And so is every person’s cancer fingerprint. This “cancer fingerprint” or “cancer signature” consists of all the genes a cancer is using and can be viewed as a computerized readout.

Researchers are studying cancer fingerprints intensively for their ability to predict the behavior of a cancer. This field of study is called genomics because all the genes in a cell are collectively known as the genome. To understand genomics, we need to define genes and DNA.

DNA is the genetic material that contains all the instructions a cell needs to function. But if a gene is a small portion of DNA, then a cell’s genome is actually a warehouse of genes.

A cell taps into only a small percentage of all the genes in the DNA warehouse at any time. Which genes the cell uses depends on its needs. For example, our skin cells can be thought of as being at rest most of the time, but if we are cut or burned, they become activated and hurl into action to repair the damage.

This repair process is orchestrated by genes that were dormant but became active on injury. Thus, skin cells at rest would be drawing on a different constellation of genes than those performing repairs.

Similarly, a breast cell would use a different set of genes than a lung cell. To relate this to cancer, a breast cancer would use different genes than a lung cancer would because the two arise from different cell types.

But even two breast cancers could employ different genes: a fast-growing breast cancer would use some genes that are distinct from those used by a slow-growing breast cancer (genes that direct the cell to grow fast would be more highly used in the fast-growing cancer). One can continue this line of thought down to the individual. Since no two people are alike, in personality or in DNA, no two cancers are genetically identical.

Each cancer retains the genetic uniqueness of the affected person. This genetic uniqueness or cancer signature can be captured and measured on a “gene chip” that is the size of a dime.

Cancer signatures are revolutionizing cancer research and will soon drastically improve our ability to estimate prognosis for each patient and tailor specific treatments to that patient’s cancer. So it merits repeating.

We can rapidly analyze the many thousands of genes that exist in any cancer today. This incredible technology is enabling researchers to classify cancers by their genetic signatures and to determine which signatures portend a good prognosis and which indicate a significant chance that a cancer will return.

Cancer signatures are also being developed to help guide the best and most effective treatment choices for any cancer. For example, a one-centimeter (one-third-inch) breast cancer with cancer signature A may be easily cured with surgery, whereas a one centimeter (one-third-inch) breast cancer with signature B may have a high likelihood of causing a future cancer relapse if treated only with surgery and radiation.

The first clinically approved genomics test, called Oncotype DX, helps oncologists estimate the chances that an early stage, invasive breast cancer will relapse in another part of the body (form metastases) as well as which patients would benefit most from chemotherapy.

Breast cancers that do not involve axillary lymph nodes and that make the estrogen receptor or progesterone receptor are eligible for Oncotype DX testing. By probing the activity of twenty-one genes in a breast cancer specimen (obtained from a small piece of the stored remainder of the original surgical specimen), a “recurrence score” is generated that places the cancer in a low risk (6.8 percent), intermediate risk (14.3 percent), or high risk (30.5 percent) of developing distant metastases within ten years.

The recurrence score is also used to guide the choice of therapy, as follows: those in the low-risk range derive little benefit from chemotherapy and would typically be treated only with hormone therapy (such as tamoxifen or an aromatose inhibitor, see chapter 7), whereas those in the high-risk category derive great benefit from chemotherapy (a 28 percent reduction in the risk of a cancer recurrence at ten years).

There is uncertainty, however, about how best to treat those with an intermediate risk score, which is why this group of patients is the subject of an ongoing clinical trial called TAILORx. In this study, patients with intermediate recurrence scores are randomized (randomly chosen by a computer) to receive either hormone therapy or chemotherapy plus hormone therapy.

The goal is to determine whether chemotherapy prevents a cancer recurrence and improves survival compared to hormone therapy alone in this group of breast cancer patients.

Gene-based tests like Oncotype DX are being developed for nearly every type of cancer, with cancers of the lung, colon, and prostate furthest along.

The future of cancer diagnosis will increasingly operate likethis: the pathologist will make the diagnosis, and the medical oncologist or surgeon will order genomics testing of the specimen to help determine both prognosis and the optimal treatment for that cancer.

The hope is that genomics testing will allow us to tailor cancer treatments specifically to each individual case.