FDA And Industry Battle Over Diagnostic Testing
By Cindy Dubin
Personalized medicine has become a significant issue for pharma and diagnostic companies. According to a new report, the $232 billion personalized medicine market in the United States is projected to grow 11% annually over the next several years.
As one of the most significant technology-driven factors of the healthcare revolution, personalized medicine — which tailors treatments to an individual’s genetic makeup — provides the healthcare arena with the ability to take on unmet medical needs and change the industry by turning research discoveries into definitive point-of-care diagnostics as well as safer and more effective therapeutics.
Orphan drugs are a significant part of personalized medicine (see the article in our October issue on this topic). Up to 30 million Americans suffer from nearly 7,000 rare diseases. Typically, big pharma and biotech companies have not dedicated much effort to orphan diseases because they represent small, niche population groups, and treatments do not become blockbusters. However, in recent years the pharma and biotech industries have concentrated on the R&D of medicines for rare diseases. In fact, U.S. applications for orphan drug designations nearly doubled during the past decade with 2009 accounting for the most orphan drug designations by the FDA in one year.
Now more than ever, drugmakers are starting to explore the benefits of personalized medicine. Essential to personalized medicine is genomic testing, which enables physicians to choose the best treatment for a patient. Drug manufacturers, such as Genentech and Amgen, have started to develop companion diagnostic tests which are used on patients before treatment to look for specific proteins or genes — alongside new drugs early in the R&D stage. Companion diagnostics can identify patients who aren’t likely to respond to certain treatments based on genetic information.
Back in June 2010, U.S. health officials laid out a road map to speed drug development and ensure the safety of diagnostic and gene tests based on personalized medicine. In a commentary released online by the New England Journal of Medicine, the U.S. FDA and the National Institutes of Health said they will invest in efforts to provide companies with a clear regulatory path for drugs and their companion diagnostics.
The agencies stated: “The success of personalized medicine depends on having accurate diagnostic tests that identify patients who can benefit from targeted therapies.” The NIH will develop centers for researchers to screen chemicals to find potential drug candidates and foster public-private partnerships to move more drug candidates into commercial development. The FDA is developing scientific standards to establish what genetic information companies must show to prove their devices and drugs are effective. The hope is to develop new drugs and diagnostics so doctors can prescribe “the right drug at the right dose at the right time,” they wrote.
A report by PricewaterhouseCoopers (PWC) last year found that 20% to 75% of patients respond to the drugs they are taking. But with companion diagnostics, that response rate could improve dramatically. About 10% of drugs approved by the FDA already include information about pharmacogenomics — how genes affect the drug’s response. PWC projects that the market for molecular diagnostics, which includes these tests, could double to $5 billion in 2012.
False Positives Are Grounds For Investigation
The FDA and NIH stand firm in their belief that diagnostic tests are not often independently reviewed or validated by the FDA and, as a result, have given wrong results. That was apparent in the findings of a U.S. Government Accountability Office investigation in which the DNA samples of five volunteers were sent to four different “direct-to-consumer” genetic testing companies. Completely contradictory results emerged. For example, three companies informed one donor that he alternatively had a low, average, and high risk of developing prostate cancer. Inaccurate risk predictions can have serious implications.
“While much of the reporting out there is anecdotal, we do know of a situation where a test on the market a few years ago was being used for early detection of ovarian cancer, and data was presented to show a 95% accuracy rate with the test,” says Alberto Gutierrez, director of the Office of In Vitro Diagnostics in the Center for Devices and Radiological Health at the FDA. “What wasn’t known was that for every 15 women who tested positive, 14 were, in reality, negative. Women wound up in surgery having their ovaries removed for no reason.”
Those are the exceptions to the rule, according to the American Clinical Laboratory Association (ACLA) in Washington, D.C. “There are millions of diagnostic tests performed daily, and no problems are associated with an extremely high percentage of them,” says David Mongillo, ACLA’s VP of policy and medical affairs.
Some believe that standardizing the approval process for these tests will prevent patient risk. Currently, there are two regulatory pathways for bringing diagnostic tests to market. One is to develop and validate laboratory-developed tests (LDTs), which are regulated by CMS (Centers for Medicare & Medicaid Services) under CLIA (Clinical Laboratory Improvement Amendments), and the other is to obtain FDA clearance for in vitro diagnostics (IVD) kits that can then be sold to clinical labs, or in some cases, direct to consumers.
Standardizing The Regulatory Process
Standardization is not a new hot button for the FDA. In the 1970s, the agency said it would not enforce approval requirements on lab-developed tests because they were largely developed by hospital-based labs to process small volumes of samples and were generally overseen by pathologists and clinicians. In the 1980s, the FDA realized the difference in regulations between the CLIA and FDA approval process. In the 1990s, a public rule made reagent manufacturers responsible to the FDA to ensure the reagents used in lab-developed tests were made under controlled conditions. In 2000, discussion again arose about the gaps between FDA and CLIA regulations. Fast forward to 2008 when Genentech petitioned the FDA to require all IVDs intended for use in drug or biologic therapeutic decision making be held to the same regulatory standards, regardless of whether the test was developed and sold by device manufacturers as diagnostic test kits or LDTs developed in-house by laboratory-based companies. Genentech’s angle was that if a test dictates that the patient begin a specific drug regimen, the test should be regulated just as the drug is regulated.
Can Personalized Be Standardized?
In July 2010, the FDA held a public hearing to get industry’s input on the agency stepping in to oversee and regulate LDTs. One company that spoke in favor of agency oversight is Agendia, which manufactures MammaPrint, an FDA-cleared breast cancer recurrence test, performed by analyzing a high-quality sample of RNA (ribonucleic acid) taken directly from the tumor at the time of biopsy or surgery. Using bioinformatics software to read which genes in the tumor tissue sample are switched on or off, a genomic signature is generated — personalized for each patient— which provides physicians with an accurate assessment of the possibility of recurrence.
“The medical diagnostic field has reached a new level, and testing has become more complex, involving many genes,” says Bernhard Sixt, CEO of Agendia. “The results of these tests plot a course of action. While the complexity of the tests aren’t of interest to the FDA, risks associated with the outcomes of those tests are, and it is the risk that should be leading the regulation.” Thus, Agendia believes that only FDA oversight can assure that LDTs are safe and effective for decision making regarding patient treatment.
Like Agendia, Fujirebio Diagnostics, Inc. supports FDA regulations as a means of leveling the playing field and preventing labs from making rogue label claims about their products. Fujirebio is a cancer diagnostics company, which manufactures manual assay products, reagents, and controls.
“There are bad eggs out there, and that is frustrating to companies like Fujirebio that have the scientific background to substantiate their claims, and some of the claims out there are just alarming,” says Diana Dickson, manager of regulatory affairs at Fujirebio. “We go through rigorous screening; why shouldn’t everyone?”
Predictably, there is opposition to that screening because it will stifle innovation. The Coalition for 21st Century Medicine, which is composed of diagnostic technology companies, clinical labs, physicians, patient advocacy groups, and venture capitalists, supports the FDA’s goal of working to assure reliable testing, but stresses that the approach must be “flexible and balanced.”
“Any new regulatory approach has to make a clear distinction between IVD medical device manufacturers and clinical labs that perform laboratory-developed tests, and we’re not certain if the proposed regulations will do that,” agrees ACLA’s Mongillo.
Genomic Health believes it underwent the appropriate validation process in compliance with federal and state laws before introducing its Oncotype DX colon cancer test earlier this year. The multigene expression test assesses the risk of individual recurrence in patients with stage II disease following surgery. “Our approval process was fair, but we are concerned that if the FDA gets involved, the approval process for tests like ours could take up to 18 months or longer, and by then the tests could be obsolete,” says Genomic Health’s CEO Kim Popovits. “These tests are advancing personal medicine, and there is not a one-size-fits-all approach to healthcare.”
Gutierrez says there is no question that regulations could slow down innovation, but that the goal is to prevent dangerous products from reaching the public. ACLA and its members accept that a limited subset of LDTs would benefit by having FDA clearance before reaching the market.
Popovits and others are also concerned that FDA regulations may be redundant or contradict what is currently required by CLIA. For example, the FDA might regulate some portion of the test, but many stakeholders consider how it is used in clinical practice to be the “practice of medicine.” Popovits says CLIA already regulates analytic validity, and that it is the assessment of clinical validity that remains at the center of the conversation. “We need to figure out how best to draw the line on who determines how this type of advanced molecular information is used in the practice of medicine.”
Details Remain Vague
Many are also trying to figure out just how the FDA will go about overseeing molecular diagnostic tests and what exactly will be required of labs. The devil is in the details — details that have yet to be determined or revealed.
For instance, there are rumblings about the FDA creating a separate center, like it did with the Center for Biologics Evaluation & Research (CBER), to regulate the biologics industry. Gutierrez is nipping that theory in the bud. “There would be no gain to having an agency like that.”
The FDA is considering if there would be a gain to grandfathering in certain tests that are already on the market, but remains vague about this option. “Many tests are so ingrained in the healthcare system that not having these tests available to the market would be disruptive and have negative implications on patient access to these tests,” says Mongillo.
In addition to determining which labs will actually require regulatory compliance, industry insiders want to know how far the regulations will go. For example, if a test is approved for diagnosing blood samples, but is found to be more useful with saliva, will the FDA require the test to go through another approval process? “That is a big, looming question that needs an answer,” says Mongillo. Gutierrez responds to that by saying, “We have told ACLA that defining which changes will require a new submission is an open issue for LDTs, and that the agency will have a fair amount of leeway in where it draws that line.”
While the details get ironed out and folks remain varied in their opinions, there is a consensus that something should be done about overseeing molecular diagnostic tests at some level. With the advent of genetic testing and personalized medicine, the diagnostic landscape has changed. ACLA remains steadfast that regulations will be welcomed if they are “risk-based, flexible, and balanced,” while the NIH states that “readily available information about these tests, including whether they were cleared or approved by the FDA, will help clinicians and consumers make informed decisions about using the test to optimize healthcare.” According to Dickson, “The discussion about FDA oversight of LDTs needed to be done five years ago. We want to see good data that supports indications, and the only way to do that is through regulation.”