Oncotype-DX is a genetic test for breast cancer that predicts the risk of recurrence. It costs about $3,500 and is covered by the majority of U.S. health plans. The test determines, based on the patient's genes and the genes of the cancer, the likelihood of the cancer coming back within 10 years. It can help the patient and oncologist figure out if they should treat the cancer with chemotherapy.
"Before this test existed, it really was a crapshoot. You had no idea," says Dr. Robert Epstein, chief medical officer and president of Medco Health Solutions. "You knew you had to remove the tumor and have a little radiation. But do you really need chemotherapy for a little lumpectomy in stage one, or don't you? For a lot of women, if you've got a 3% chance [of the cancer recurring] in 10 years, would you really go through the horrors of chemotherapy?"
Oncotype-DX is just one of up to 50 genetic tests on the market today that look at drugs and how they can react differently in a patient based on the patient's genes. It's an area of study called pharmacogenomics - how genes affect a person's response to drugs - and it's exploded in recent years.
Last summer, two new cancer drugs that target specific genes, along with the accompanying genetic tests, were approved by the Food and Drug Administration - Zelboraf (manufactured by Genentech and generically known as vemurafenib) for patients with melanoma tumors that express a gene mutation called BRAF V600E, and Xalkori (manufactured by Pfizer and generically know as crizotinib) for patients with a type of lung cancer that expresses an abnormal variation of the ALK gene.
"Many employers have a good sense of what drugs they're paying for but, in my experience, many don't have great visibility into tests - what they're paying for or not paying for," says Epstein. "They look at tests and diagnostics in one bucket and drugs in another. Typically, when most employers look at a line item for diagnostics, it's just one row; it's not broken out. Now that things are changing so fast, this is a good time to be thinking about it."
Pharmacogenomic tests differ from other types of genetic tests in that they relate only to pharmaceuticals and how drugs work in people's bodies. Other genetic tests - carrier tests for diseases such as cystic fibrosis or Huntington's, and predisposition tests to determine how likely someone is to get a particular disease - are far more controversial. It's an important distinction to make because with pharmacogenomics, the disease has already been diagnosed.
Pharmacogenomics was aided by the Genetic Information Non-Discrimination Act, passed in 2008. The law prohibits health insurers from using a person's genetic information to determine eligibility or premiums, and from requiring that a person have a genetic test. GINA also prevents employers from using a person's genetic information in making decisions regarding hiring, firing and job assignments, and prohibits them from requesting, requiring or purchasing genetic information about employees.
"That was really good for the field, employers and patients," Epstein says. "It put a safe harbor in there for these tests to be done."
Protected tests under GINA include classifications of genetic properties of an existing tumor (in the case of cancer, for example) to help determine therapy.
"In our experience, fewer than 3% of people say no [to genetic tests] based on fears about the use of their genetic tests," says Epstein. "People talk about this all the time but if you really ask people, they want to get the right drug for themselves for whatever it is. They're more interested in that than being worried about how someone is collecting data on how fast they metabolize warfarin [a commonly used blood thinner]."
Still, pharmacogenomics is not an area on most employers' radar screens, says Michael Jacobs, a pharmacy benefits consultant with Buck Consultants.
"At this point in time, the risk-reward and cost ratios aren't where the employer needs them to be," he says. He suggests some PBMs are pursuing the genetic testing agenda because it has the potential to generate revenue for them.
Mark Morse, director, pharmacy and clinical sales and segment support with Humana Pharmacy Solutions, says most employers don't have the expertise to know which tests should be covered and so they are counting on their PBM to do what's right. "However, there can be conflicting goals. If one of these tests will do no harm - it might tell you something, it might not - but will drive revenue, that's important to know," he says.
Precision prescribing
Epstein maintains genetic testing enables physicians to bring more precision to their medication prescribing. "The upside for employers is that this brings potentially more precision to prescription drug use," he says. "Instead of just giving 100 people the same drug and hoping for the best, now you'd have a way of knowing that for these 20 people it's never going to work and that these 15 people need a higher dose. If you could be more precise about the money you're spending, you should get better outcomes at a lower cost."
Pharmacogenomic testing isn't just used to determine whether or not a drug will work on a particular patient. It's also used to determine dosing and duration of therapy. In Hepatitis C, for example, certain gene types of the virus only need 24 weeks of therapy, while other gene types of the virus need 48 weeks of therapy.
Cancer is the most prevalent disease area where genetic inroads are being made, but other areas are seeing breakthroughs. "A lot of the viruses now - Hepatitis C, HIV - have multiple therapies where genetics can tell you which drug to use, which one not to use or which dose to use," says Epstein.
Another area that's heating up is cardiovascular disease. There's already a genetic test for warfarin, a commonly used blood thinner. Warfarin can be dangerous for some people - if they get too much they're at risk for a bleed, too little and they're at risk for a clot. A 2010 study from the Medco Research Institute and Mayo Clinic showed that hospitalization rates for heart patients taking warfarin dropped by approximately 30% when genetic information was available to doctors when prescribing the drug.
"And there are four or five other markers in cardiovascular disease that are being looked at," Epstein says. "One of them is for statins and lipid-lowering drugs. There seems to be a marker to predict who gets the muscle pains and liver problems. So, in the future, there could be a genetic test before people go on the drug to make sure they're not going to cause a problem."
There are also ongoing studies on rheumatoid arthritis and other inflammatory diseases. Drugs in this category can be very expensive - up to $30,000 a year - and "you don't know for a year if you're going to respond to them, and they're not without risk," Epstein says. "So there are some big ongoing studies to see if there's a genetic panel - a group of genes - to determine if you're going to respond to these drugs or if you should move on to something else."
Still, Jacobs maintains genetic tests, in many cases, offer little value. "When it comes to some of the tests out there, there are really very few definitive answers you can get from these tests," he says. "If I'm going to give some kind of test before I give some kind of therapy, I need to know: Are the results of that test going to change what I do in any way, shape or form?"
Epstein argues that pharmacogenomic testing does change patient and physician behavior, citing the Oncotype-DX test. "There have been studies looking at whether, after having this test, women avoid the chemotherapy or do they get it anyway?" he says. "And they really do avoid the chemo, so it does change the behavior of the woman and the practitioner."
He notes that in about 50% of clinical drug trials today, researchers collect DNA from patients to better understand why some respond to the drug and others don't. "Ten years ago it was zero percent," he says. "Now we have half of all patients in clinical trials getting tested. Going forward, we're going to see more and more of this."
Genetic testing increases adherencePremliminary results of a Medco and Quest Diagnostics study show that patients who received a gene test adhered better to their statin therapy over a six-month period than those who did not receive the test. The AKROBATS trial is also revealing new information on patient attitudes and perceptions about personalized medicine, including the factors that influence acceptance of gene testing for statin therapy. Full data results are expected at an upcoming scientific congress this spring.
Among the 682 patients who participated in the clinical trial, 53.8% did so because they believe genetic testing is useful. Among the 773 patients who declined to participate in the genetic testing, 6.7% cited privacy concerns, while 8.8% cited anxiety about the results.
