Genomics: gearing up for the laboratory of the future

A giant step forward has been taken, as the first next-generation sequencer for clinical care received FDA marketing authorization this past November. This advance means clinicians can start to look for “an almost unlimited number of genetic changes” in patients’ DNA, wrote NIH director Francis S. Collins and FDA Commissioner Margaret A. Hamburg in the New England Journal of Medicine article that announced the launch.¹ More such instruments will follow.

This remarkable feat has been enabled by decades of research and investment. The cost of sequencing a human genome has plummeted from millions of dollars to less than $5,000. Instead of taking more than a decade and using hundreds of instruments, a genome can be sequenced in about 24 hours on a single machine. Now, doctors will be able to start using targeted sequencing to upgrade their diagnoses of a growing list of conditions.

But is this a case of a technology in search of a purpose? Or are we truly on the cusp of the long-awaited age of personalized medicine? The evidence points to the latter.

For one thing, there has been an explosion in the number of genetic markers that indicate tumor susceptibility to certain treatments. Projects such as the NIH’s own Cancer Genome Atlas are revealing myriad genetic “drivers” of specific types of cancer.²

Recent research on glioblastoma, for example, uncovered new mutations in the LZTR1, ATRX, KEL, and QK1 genes. These genes were added to the list of suspected drivers thanks to work using tools that can read more than 100 million sequences of DNA at a time. Overall, the work revealed that glioblastoma tumors grow and spread thanks to large, complex signaling networks with redundancy built into them that help tumors outmaneuver targeted therapies.³

Treatment of breast cancer, lung cancer, melanoma, and a growing number of other cancers is currently biomarker-driven. As with glioblastoma, we are starting to understand the relationships between many of the pathways in which these markers’ molecules operate. One day, doctors will use such biomarkers to determine exactly when and how to use multiple types of combination therapies against cancers, including traditional chemotherapy, surgery, immunotherapy, targeted therapies, and even newer types of treatments, such as drugs against cancer stem cells. Should the targeted therapy come first? Or should it be given concurrently to the immunotherapy and chemo? And which drugs in specific should be used?

Genomics, and pharmacogenomics, will become vital in other areas besides cancer. Recently the Mississippi-based Jackson Heart Study and Massachusetts-based Framingham Heart Study announced a major collaboration to look at the relationship between an individual’s biology and lifestyle and specific treatments. They want to better predict who is at risk of heart disease or stroke.⁴ Of the 75 million people in the United States with hypertension, for instance, only about half can control their disease with medication.  Genetic research could uncover subsets of such patients who need different types of treatment.

We’re also seeing steady growth in the number of clinical genomics divisions in major hospitals. Millions of dollars are being spent building these clinics, with Mount Sinai in New York alone investing more than $100 million in a new Institute for Genomics and Multiscale Biology.⁵ The hospital sees about 500,000 patients a year and gathers reams of information about their health and treatment outcomes.

One area where these clinical genomics divisions are already seeing great success is the diagnosis of previously unknown diseases. In a breakthrough case reported in 2010, scientists and physicians at the Medical College of Wisconsin and Children’s Hospital in Wisconsin used genetic sequencing to identify a previously unknown mutation that was causing intractable inflammatory bowel disease in a boy.⁶

By the age of 3, this child had already endured more than 100 surgeries. The researchers sequenced his exome, examining the 20,000 coding genes in his genome.  After months of analysis, they found a single gene that seemed like it could be causing the child’s condition. What’s more, they were able to recommend a treatment, cord blood transplant, which essentially cured him.

In addition, at least 120 FDA-approved drugs have pharmacogenomic information in their labeling—information about specific genes that influence the patient’s potential response to the drug, including safety issues.  This area has been controversial, with most insurers demanding firm evidence of utility before they will cover the tests—but this field too has great potential. One study estimated that nearly one-fourth of all outpatients receive one or more drugs with pharmacogenomics prescribing information.⁷

In their NEJM editorial about the new clinic-ready next-gen sequencer, Collins and Hamburg describe the instrument’s approval as a “landmark.” But they add that it is only the beginning. Much work is left to do to truly bring the genomic age into full bloom, and clinical laboratories will need to do a lot of it. Studies linking mutations to specific conditions need to be better validated, doctors need help interpreting this exciting new data, and patients need to be better educated for the genomic age. But once that work is done, huge rewards will follow.

Scientia