One medical institute steeped in genomics has pooled the resources of multiple departments to create its own cancer genetics laboratory. Another cancer center has joined a regional alliance to meet its DNA sequencing needs. Other institutions are slowly piecing together a diverse range of tools and personnel as they begin assembling their own diagnostic genomics laboratories.

Fueled in large part by breakthroughs in DNA sequencing technology, many cancer clinicians and lab oratories are making a vigorous push toward molecular diagnostics based on whole-genome sequencing. However, as the promise of personalized medicine begins to rise on the horizon, logistical and practical constraints are forcing pathologists to try a variety of routes at varying speeds on their quest to make genomic pathology a clinical reality.

Some laboratories have benefited from their proximity to major genome sequencing centers. At Baylor College of Medicine in Houston, Texas, multiple departments teamed up in 2011 to create the Cancer Genetics Laboratory to develop and run new genomic tests for cancer. Federico Monzon, MD, the laboratory’s director of molecular pathology, says he and his colleagues have already launched 2 versions of a gene panel targeting mutational hotspots in 50 carcinoma-linked genes and a second panel examining 26 leukemia-linked genes.

In collaboration with Baylor’s Human Genome Sequencing Center, the laboratory has initiated its own cancer exome sequencing effort, a push to sequence the protein-encoding segments of all 20,000 known genes in DNA samples extracted from individual tumors. Dr. Monzon says the new center has likewise piggybacked off the bioinformatics expertise of Baylor’s more established genetics and genomics laboratories, although it has refashioned, tested, and validated many of the research tools to make them more clinically relevant.

In New York, Memorial Sloan-Kettering Cancer Center (MSKCC) has taken a different route by joining 10 other major biomedical institutions as an institutional founding member of a stand-alone entity called the New York Genome Center. The independent, nonprofit genome sequencing and bioinformatics center, formally launched in 2012, will cater in part to research institutes and companies pursuing whole-genome sequencing.

In the meantime, the molecular diagnostics laboratory at MSKCC is using a more modest panel of 30 to 40 cancer- linked genes in its sequence-based clinical testing. Michael Berger, PhD, assistant attending in the department of pathology, is helping to develop a third approach dubbed IMPACT (Integrated Mutation Profiling of Actionable Cancer Targets).

Dr. Berger describes the testing panel, which targets 275 genes implicated in solid tumors, as a middle ground between sequencing the whole genome or exome and profiling individual genes with specific mutations. He and his colleagues selected the IMPACT genes based on their diagnostic and prognostic value and on the presence of approved or in-development treatment options targeting the known mutations. “By doing fewer genes, it simplifies the analysis a little bit and it brings down the cost, and for a large-volume molecular diagnostics lab with today’s technology and today’s costs, I think it’s a sensible approach,” Dr. Berger says.

However, as MSKCC scales up with next-generation sequencing and ever-larger gene panels, the cancer institute is installing additional storage and analytical infrastructure and hiring bioinformatics experts to help parse the growing mounds of data. Dr. Berger says the in-house molecular laboratories are also adapting many of the tools developed by other computational groups for large-scale sequencing projects sponsored by the federally funded Cancer Genome Atlas.

Investing in the Future

  1. Top of page
  2. Investing in the Future

Because the clinical usefulness of genomic medicine may depend heavily on complicated data interpretation (eg, which mutations may drive oncogenesis and which are merely along for the ride), several laboratory directors say they are investing more in bioinformatics and interpretation tools than in the rapidly evolving sequencing technology. In some cases, that means sharing part of the workload with the growing corps of genomics consulting firms.

Investing in the necessary infrastructure also may mean constructing segregated spaces to avoid contamination during the tumor DNA sequencing steps. That consideration is particularly important given the growing consensus that all genes may need to be sequenced hundreds of times to achieve the necessary depth of coverage for clinical diagnosis. Deep coverage is especially critical for identifying mutations that may be present only in a small number of cells due to heterogeneity of the tumor or infiltration by adjacent normal tissue.

For many laboratories, the next-generation sequencing and data refinement processes are being ironed out in the context of clinical trials. Universities are also establishing tumor boards of oncologists, pathologists, surgeons, bioinformatics experts, and other clinicians to help identify the actionable mutations and direct the therapy for individual patients.

Richard L. Haspel, MD, PhD, assistant professor of pathology at Beth Israel Deaconess Medical Center and Harvard Medical School in Boston, Massachusetts, agrees that it may not make sense for every hospital to have its own sequencing machines and bioinformatics analysts. “I think a lot of this is going to be outsourced to companies who want to keep up with it, but I still think a pathologist is going to need to be involved to make sure that the methods they’re using are accurate, precise, and validated and to help interpret the results,” he says. If pathologists do not lead the way in genomic testing and interpretation, he says, “there’s a real risk that it’s going to be done poorly.”

As genomic testing technology matures, some experts foresee a dramatic and exciting expansion of the role of pathology in personalized medicine. Iris Schrijver, MD, professor of pathology at Stanford University Medical Center in Stanford, California, says pathologists often worked in the background, where they were “relatively invisible” to other clinicians. “Pathologists are becoming much more centrally integrated into health care teams,” she says, “with more direct and frequent interactions with other clinical colleagues.” No matter how they get there, Dr. Monzon says pathologists need to start positioning themselves as experts in translating genomic language to clinical language, and in consulting clinicians on the best tests for managing patients and on the implications of those test results. “I think we just need to grab the bull by the horns and start doing it,” he says.