WASHINGTON, DC—Advances from human genome research have already improved patient care—especially in cancer—but as more genomic technologies are integrated into clinical practice there is a danger they may add to rather than reduce costs. That was the conclusion of speakers here at an Institute of Medicine workshop here, sponsored by the IOM's Roundtable on Translating Genomic-Based Research for Health.
“We're hoping to move the needle about what the economic models for integrating genomics into medicine will be,” said the Roundtable's Co-Chair (with Wylie Burke, MD, PhD), Sharon Terry, MA, President and CEO of the Genetic Alliance. She told OT a report will be prepared from the meeting, which should be published in several months.
The roundtable is assessing the impact—including economically—of the advent of genomic medicine on clinical practice. One major concern: a “cost cascade” of further downstream testing, counseling, and follow-up triggered by the initial genetic testing, a cascade that also exposes a patient to unintended adverse consequences—such as unnecessary medical intervention.
“Even if you have the mutation, it doesn't mean you'll get the disease,” cautioned James P. Evans, MD, PhD, the Bryson Distinguished Professor of Genetics and Medicine at the University of North Carolina at Chapel Hill.
“It's not really about costs, it's about tradeoffs,” said another speaker, David L. Veenstra, PharmD, PhD, Professor in the Pharmaceutical Outcomes Research and Policy Program in the Department of Pharmacy and a member of the Institute for Public Health Genetics at the University of Washington. “Should we spend $50,000 to prevent a heart attack or $50,000 to prevent breast cancer?”
‘Something's Got to Give’
“As these tests provide more and more information, something's got to give,” added Scott Ramsey, MD, PhD, a Full Member in the Cancer Prevention Program at Fred Hutchinson Cancer Research Center, where he directs the Research and Economic Assessment in Cancer and Healthcare (REACH) group.
He pointed out that right now it is unclear what percentage of genetic information is actionable clinically, and most physicians don't have enough time to counsel patients on the often complex implications of genetic testing.
Kenneth Offit, MD, Chief of the Clinical Genetics Service at Memorial Sloan-Kettering Cancer Center and Professor of Medicine and Public Health at Weill College of Medicine at Cornell University, said that at MSKCC, patients receive testing for a targeted panel of genetic mutations.
He noted that $100,000 per life-year saved is a figure generally used when evaluating the economics of a medical intervention—including genetic testing. There are three main models for integrating genomics into clinical oncology, he said:
- Targeted mutation testing;
- Sequencing of about 230 actionable cancer genes; and
- Whole genome sequencing—Speakers cautioned, though, that the era of the $1,000 whole genome sequencing test for cancer is still in the yet-to-be-determined future.
And while genetic testing can be useful and valuable for clinical decision-making, “the effect sizes of most GWAS [genome-wide association] SNPs [single nucleotide polymorphisms] are small and of no clinical utility,” noted Offit, whose research group was among the first to measure prospectively the impact of preventive ovarian surgery in women with BRCA mutations. Targeted mutation testing is the current clinical practice, but gaining access to this testing is an issue for many patients, he said.
Broader application of genomics in medicine could be cost-effective if it leads to improved patient outcomes, Ramsey said. “To me one of the benefits of doing genetic testing is showing us when we don't have to do something”—such as determining, for example, that a man with early-stage, organ-confined prostate cancer does not need treatment.
And, predicted Stephen L. Eck, MD, PhD, Vice President and Global Head of Oncology Medical Science at Astellas Pharma Global Development, in the future, patients who do need treatment will get a series of targeted therapies based on periodic molecular reassessment of their tumors. A major issue right now and for the future, he said, is: “How small can a disease state be and still be attractive to a commercial drug developer?”
He answered, “The market is already segmented; segmenting it for scientific reasons is the best idea.” He pointed out that in the past, a therapy for one cancer has often proved effective in treating several others—making it more attractive to the commercial developer.
“There are driver mutations across cancers,” agreed Timothy J. Ley, MD, holder of the Lewis T. and Rosalind B. Apple Chair in Oncology, Professor of Medicine and of Genetics at Washington University and Associate Director of the Genome Institute (for cancer genetics). Therefore, said Ley—past president of the American Society for Clinical Investigation—restricting the use of a cancer drug to what is on the label “is insane.”
Also at the meeting, several patient cases were presented for discussion, including that of a 50-year-old patient who presented with cough, dyspnea, and chest discomfort. Evaluation revealed a lung mass; bronchoscopy and biopsy revealed a non-small-cell lung cancer (NSCLC). Her tumor was found to have genetic variations that allowed the patient to be treated and go into remission.
Lung cancer patient Richard Heimler, a former executive of non-profits, said he is a real-life example of how genetic testing can be used to extend life.
A non-smoker, Heimler was diagnosed in 2004 at age 44 with NSCLC. In the past eight years, he was diagnosed six times with additional cancerous tumors in his lung, brain, and thorax. After progression of his lung cancer to stage 4, he was tested and found to be positive for the ALK gene. He said that after enrolling in a clinical trial of the targeted drug crizotinib, his tumors have shrunk, his pulmonary function has increased, and he has more energy.
“I would not have been able to participate in this clinical trial if I had not been tested for this gene variation,” he said. “Lung cancer may not be curable, but it is treatable and livable.”
Evans described the case of a 48-year-old patient of his who was adopted, and diagnosed with breast cancer. “This is a case where genetic analysis might be more valuable,” he told OT, since the patient knew nothing about her family history or her ethnicity. Evans said that while her relatively young age put her on the edge of recommended BRCA testing, he decided to offer it to the patient given her unknown family background, and she chose to have it; he is currently awaiting results of the testing.
“There are things that should be acted on, and things that probably shouldn't be,” he said. “I'm in favor of targeting those sections of the genome that it makes sense to target.”
John West, MBA, MS, CEO of Personalis, Inc., a company founded by global leaders in human genome interpretation, described how he decided to have his genome sequenced by a commercial testing source after experiencing two unprovoked pulmonary embolisms six months apart at age 43. The testing revealed a Factor V heterozygous clotting mutation. He said that had this mutation been identified after his first pulmonary embolism, he would have been spared months of testing for an occult cancer, since many patients with his history undergo testing for a suspected underlying malignancy.
Costs aside, Ley told OT, “ultimately in cancer the driving force is going to be bottom-up rather than top-down; patients will drive it.” And while privacy issues have been raised about genetic testing, they “are irrelevant if you've got a lethal disease,” he said. “To me, the major bottleneck is the analysis of the data; we need more data and more people who can interpret it.”
Integrating genomics into general clinical practice will be a “difficult problem,” he added. “We're flexing through a period where the technology is changing so rapidly that people don't want to invest in it. The only thing that's going to win at the end of the day is whole genome sequencing…nothing else will survive. That's how I see it. All these [gene mutation] panels are just interim.”