Sekeres, Mikkael A. MD, MS
MIKKAEL A. SEKERES, MD, MS
Even the word sounds vaguely technical, like the research focus of some myopic scientist whose femur and tibia have progressively bowed over years due to lack of sun exposure and resultant vitamin D deficiency.
Then why did the Journal of Clinical Oncology—notice that third word, Clinical—devote an entire issue (5/20/13) to the topic? And why are major medical centers (ours included) devoting large amounts of resources towards developing strategies around genomics?
Let's start by defining what genomics means. In an overview to the special issue of JCO, Garraway, Verweij, and Ballman write, “We have chosen the term genomics-driven cancer medicine in recognition of the fact that knowledge that emanates specifically from the cancer genome will likely continue to direct the opening act of precision oncology as it plays out in clinical and translational studies over the next several years” (JCO 2013;31:1803-1805).
Let's put that into plain words. All cancers arise from genetic errors—mistakes in genetic coding that, for example, cause a cell to grow and reproduce out of proportion to other normal cells, or cause a cell to ignore signals that try to stop it from growing.
Some people are born with genetic errors that predispose them to cancer (those with Down Syndrome are a notable example, as are those with BRCA1 and BRCA2 abnormalities); some acquire these genetic defects, either from exposure to chemotherapy or radiation therapy administered to treat other cancers or from the environment (I have had patients who worked in the rubber industry who describe standing in vats of benzene for years and then develop leukemia); and some have a combination of innate and acquired genetic abnormalities that result in cancer.
Using classic techniques, such as metaphase karyotyping, we can find those abnormalities in approximately 50 percent of patients with, for example, acute myeloid leukemia. Yet, we know that the other 50 percent have alterations in their genetic code that caused their cancer, too.
We can now find these genetic alterations using next-generation or massively parallel sequencing technologies. These allow sequencing of the entire human genome, identifying abnormalities that may be associated with cancer development using sophisticated statistical algorithms to distinguish normal human genetic variability (the genes that code my facial features distinctly from those of George Clooney, unfortunately) from genetic drivers of cancer. Sounds complicated, and it is—but the price of performing these analyses has plummeted to less than $5,000, and the time to perform them has shrunk so that 10 human genomes can be sequenced in a day.
Okay, so you find a bunch of genes that are associated with a cancer. Then what?
Well, some of these genes have recently been identified as being disease-defining. It's no secret that the BCR-ABL (Philadelphia chromosome) translocation is the sine qua non of chronic myeloid leukemia (CML)—we've known that since 1960. But did you know that the SF3B1 mutation appears necessary for the ring sideroblast phenotype in myelodysplastic syndromes, or that the SETBP1 mutation defines atypical CML? The more rare disease we are able to sequence, the more we will move towards molecular definitions of disease, and we can make diagnoses in borderline cases.
Other genes have an impact on disease prognosis, and may help dictate future therapies. It is now standard, in a patient with acute myeloid leukemia who has “normal” cytogenetics as assessed by metaphase karyotyping, to further test that patient's bone marrow sample for the molecular abnormalities FLT3 and NPM1. Those who have the NPM1 mutation but are wild type for the FLT3 abnormality have a cure rate with chemotherapy alone that is almost the highest I can quote. Patients who have the opposite—a FLT3 mutation but who are NPM1 wild type—are almost incurable with chemotherapy alone, and are sent for a bone marrow transplant as soon as possible.
Still others—and here's the really exciting part—are what we call actionable, or druggable targets: genetic abnormalities for which specific therapies that target the abnormality (or the results of that abnormality) have been or could be developed.
The classic, of course, is the BCR-ABL mutation with CML, for which four drugs with a mechanism of action specific to the tyrosine kinase domain affected by the mutation now exist. Other, well-known actionable abnormalities include BRAF in melanoma and ALK in lung cancer, both of which can be treated with specific drugs that have already changed the therapeutic algorithms and natural courses of these diseases.
This is an exciting time, in the intersection of molecular biology and clinical care of cancer patients. It begs the question, though: what if we find a molecular abnormality we normally associate with lung cancer in a different organ, such as the pancreas? Do we treat it with standard pancreatic cancer therapies? Or try the targeted lung cancer drug?
In his book Race Matters, Cornel West explores America's obsession with race, encouraging his readers to acknowledge the role race plays in economics and politics in the U.S., and to see beyond racial barriers in recognizing that there is an interdependent humanness and Americanness in all of us.
Genomics has started to break down organ and tissue barriers as we recognize that, though cancer may start in one organ, its genetic basis may be similar to other cancers that occur in the body, regardless of geography, and our treatment should no longer be confined by tissue boundaries.
To put it simply, if we aren't focusing on genomics in cancer, starting now, we will not be treating cancers effectively five years from now.
More ‘Second Thoughts’!
Check out all the previous articles in Mikkael Sekeres' award-winning column in this collection on the OT website: http://bit.ly/OT-SekeresCollection