CEO SUMMARY: Clinical laboratories of all sizes are poised to become the source of much of a hospital or health system’s “big data.” At many academic center labs, greater use of genetic and molecular testing requires that more space and more staff be devoted to data management. At the same time, the latest generation of gene sequencing instruments and molecular analyzers are cheaper, faster, and more automated. These systems make it feasible for even smaller labs to offer sophisticated genetic tests.
DATA IS POISED TO BE A DISRUPTIVE FORCE to the existing model of clinical laboratory testing. That’s just one prediction from a lab scientist who has been at the forefront of molecular diagnostics over the past 25 years.
“Not only are clinical laboratories starting to produce much greater volumes of important clinical data than ever before, but the demand from physicians and payers for this data is growing at an equally significant pace,” stated Gregory J. Tsongalis, Ph.D., Professor of Pathology and Director of Molecular Pathology at the Theodor Geisel School of Medicine at Dartmouth College, in Hanover, New Hampshire.
“Going forward, laboratories of all sizes will be generating essential data in volumes unimagined a decade ago,” he noted. “This will be true of both small labs and large labs. It is easy to see the signs of this trend.
“For example, we just opened a new clinical lab facility of 11,000 square feet at the Geisel School,” said Tsongalis. “Over 25% of that space will be devoted to data management. The same is true at the Jackson Laboratory in Farmington, Connecticut, [where he serves as Director of the Clinical Genomics Laboratory]. Because of the increasing volume of data generated at this site, there are more computational technologists than lab technicians.”
According to Tsongalis, laboratories will make significant data contributions in healthcare big data for multiple reasons. “First, larger numbers of clinical labsboth large and small-are using more automated lab testing systems. Lab automation makes it easier and cheaper to perform greater lab test volume, thus producing more data,” he noted. “Second, new diagnostic technologies are giving labs and physicians additional tools to diagnose disease while providing more precise guidance on therapies, and helping them monitor patients with greater accuracy.
Huge Volumes of Lab Data
“Third, new diagnostic technologies are producing huge volumes of data in the clinical lab,” he continued. “This will not be limited to genetic testing and molecular diagnostics, where analysis of exomes and genomes generates hundreds of millions, even billions of data points. But it appears likely that clinical assays for testing the human metabolome and human microbiome will become common.
“All of these elements are why medical labs will be generating huge quantities of data,” stated Tsongalis. “We already have evidence in many academic centers that all of this clinical laboratory test data will be the foundation for the big data efforts of hospitals and health systems.”
Tsongalis advised pathologists and lab administrators not to underestimate how the ongoing improvements to automation solutions-including automated lab analyzers with small footprints and lower costswill benefit even small hospital labs. “Some of this new equipment may even change how reference labs operate,” he speculated.
“As laboratory equipment evolves to produce more data on patients’ conditions, the role of clinical labs will change,” stated Tsongalis. “This will happen concurrent with how labs are paid. Current trends indicate that the role of the clinical lab is poised to change from the entity that keeps specimens to the entity that is the keeper of data. And by that I mean a keeper of big data.
“In many respects our new lab space at Dartmouth-with up to 25% of the space dedicated to data analysis-looks like central command with computer monitors and servers to monitor all of the data being generated,” he said. “That data comes from our gene sequencers, from the other molecular analyses we do, and, as we move forward, from the personal health monitoring or wearable devices being developed.
“Another factor that will transform clinical labs is the need for new skills that all the new diagnostic and information technology require,” Tsongalis continued. “Labs will need professionals who can do data analysis and who can look at the clinical lab test results from new technologies to flag results of clinical concern or to point out variations and trends.
“At the Jackson Lab, the majority of the staff are computational biologists and curation scientists who process data produced by a small number of laboratory technicians,” he said. “Consider this trend as evidence of the game-changing disruptive technology that data represents for all of us working in clinical labs today. It will be essential for every lab to have individuals with the skills to manage and analyze data.”
New Uses for EHR, Lab Data
Tsongalis believes that most hospital administrators have yet to recognize that their electronic health records systems will not be able to store the large volumes of data generated by clinical labs. “This is probably one aspect of the big data trend where pathologists and lab scientists are ahead of health system administrators,” he observed.
“Should an EHR contain the raw data from a patient’s whole genome?” he asked. “Probably not. So where will that data be stored? It will be kept by the clinical lab that performed the gene sequencing. This is just one example of how and why clinical labs will be the repositories for a large proportion of every hospital’s ‘big data.’
“Take this one step further,” added Tsongalis. “The time and expense to sequence a whole human genome is shrinking at a rapid rate. At the same time, the analyzers used in clinical labs are becoming smaller, faster, and more automated. In our lab, we are producing large amounts of genetic and molecular data in one day or less.
Clinical Exome Sequencing Has the Potential To Replace Some Multiple Gene Tests
CLINICAL EXOME SEQUENCING (CES) will be the next game-changing technology in clinical laboratories, stated Gregory J. Tsongalis, Ph.D., Professor of Pathology and Director of Molecular Pathology at the Theodor Geisel School of Medicine at Dartmouth College. That’s because CES has the potential to replace a long list of tests in every clinical lab.
“Consider that clinical exome sequencing allows us to sequence 4,800 genes, all of which are known to be or have mutations that are causative for some type of human disease,” he said. “In other words, it is a rapid way to provide a lot of information quickly.
“This is what distinguishes clinical exome sequencing from whole exome sequencing,” noted Tsongalis. “Whole exome sequencing involves sequencing the coding regions of every gene. The clinical exome narrows that down to sequence the coding regions only of those genes identified as being responsible for some type of human disease.
“Here’s why CES is important,” he emphasized. “Our lab can currently sequence a clinical exome and do the data analysis for less than
$1,000. That is much cheaper than what our lab pays when it refers out genetic testing, such as for children with autism, developmental delay, or some type of a syndrome phenotype. When done by an outside lab, those tests cost us between $2,000 to $15,000 each.
“But now we can take this one clinical exome test and run it on every one of those patients,” stated Tsongalis. “We don’t need to send those tests out. At the same time, our inhouse CES test allows us to focus the analysis on the genes of interest, meaning those 4,800 genes. For certain patients there are only 50 or 100 genes of interest and that allows us to narrow the data analysis down specifically to those particular areas.
“We believe that CES testing done in-house will reduce the overall cost of this type of testing tremendously,” he continued. “Moreover, the cost savings to the institution will be huge because most payers are reimbursing these genetic and molecular tests at very low rates, if at all. That is why the cost savings and cost avoidance is huge for these cases.
“These are some of the reasons why we are excited about the potential for clinical exome sequencing to allow us to return a faster, more accurate answer that improves patient outcomes while holding down the cost of this testing when compared to using reference labs,” concluded Tsongalis. “Our expectation is today’s send-out genetic disease panel has the potential to be replaced by one test-the clinical exome sequence-because of stillfalling costs, automated systems to produce the sequence, and more robust informatics to store and analyze the data.”
“There is another development that works in favor of smaller clinical labs,” he added. “What many lab professionals have yet to realize about molecular diagnostics is that it is a universal technology. That means we don’t need one instrument to do genetic testing and another instrument to do infectious disease testing and a third instrument to do cancer testing. The tools − the analyzers − are all the same. The only thing that changes is the chemistry, meaning the primers or the probes used to detect the sequences that are our target, regardless of the actual test.
“This is a significant factor for small hospital labs, for reference labs, and in fact for the entire clinical lab industry,” he said. “Look at the test menu when I joined the staff at Dartmouth in 2004. At that time, there were just a handful of molecular tests being done routinely in the clinical lab. There was no automation or high-throughput systems yet because the volume was low and the costs of these tests were high.
“At that time, we had three tests in genetics (FII, FRAX, and FV), three tests in heme-onc (Bcl-2, IgH, and TCR), and two tests in infectious disease (B. pertussis and Parvo B19),” he recalled. “There were no routine molecular tests in oncology or pharmacogenomics.
“Today, of course, the list is much longer,” continued Tsongalis. “There are 14 tests in genetics, 10 tests in heme-onc, 13 tests in ID, 22 tests in oncology, and six tests in PGX. In all five categories, the demand for new clinical tests, markers, genes, and mutation profiles went through the roof.
“In response, we automated the testing and developed high complexity testing,” he said. “Now our lab has a long list of different tests for genetics, heme-onc, infectious diseases, and oncology. This all happened in the past three to five years because of the explosion of new therapeutics and advances in precision medicine.
Multigene Panels Offered
“Notice also that our lab quickly went from single gene and single mutation tests to panels of tests in hematology where we routinely sequence 54 genes for all patients,” he explained. “In oncology, it’s similar. We offer a 50-gene cancer hot spot panel. One hereditary cancer panel, HCP, is in validation now and has 94 genes in it that we will sequence routinely.”
Pathologists and lab administrators working in community hospitals and smaller labs are poised to benefit from the latest generation of sophisticated, automated analyzers now available. “These new analyzers are why we have gone from hands-on, labor-intensive technologies to some automated platforms that can do everything,” observed Tsongalis.
“Today, systems exist where the sample is added into a cartridge, and the cartridge extracts the DNA, does the PCR, does the quantification, then generates a result,” he continued. “These are game-changing technologies for clinical labs, whether the lab is small, medium, or large. Why? Because these instruments can have a big impact on patient care due to faster turnaround time.
“Cepheid has the GeneXpert System, which was one of the first to offer this concept of sample-to-answer with a compartmentalized cartridge,” commented Tsongalis. “In each compartment a different step of the assay takes place and a flywheel moves the reagents from one compartment to the next. When it is time for the PCR reaction, the cartridge is then plugged into the instrument and the results of the assay are available within an hour or so.
“Other companies followed with similar technologies because of the huge market represented by smaller hospital labs,” he stated. “These labs get great benefit from offering more genetic tests that contribute to improved patient care.
“Another example is the BD Mac System, which is all cartridge-based,” he added. “Systems like these provide a way to perform fast, real-time PCR assays even though the through-put is low because it’s one sample at a time. The latest generation of these instruments is fast and − in the right settings − provide test results that have a major impact on how a patient is diagnosed and treated at specific moments in time. It is why these genetic and molecular technologies are changing how laboratories operate.
“These machines are simple to operate and offer rapid turnaround times,” noted Tsongalis. “They use a single-use consumable that’s closed, which is important because it eliminates the potential for contamination. Any size lab can use these machines and there is a huge cost-benefit ratio for this equipment.
“In addition, these new cartridge-based analyzers solve a big problem for our clinicians who are unwilling to wait three or four days for results,” emphasized Tsongalis. “They want a much more rapid turnaround time. These are all reasons why I believe these technologies will become even quicker and cheaper.”
Contact Gregory Tsongalis at 603-650-5498 or Gregory.J.Tsongalis@hitchcock.org.