By Turna Ray
Medco's observational studies — looking at the clinical utility of pharmacogenetic testing prior to treating patients with the anticoagulant warfarin and oncologic tamoxifen — promise an alternative to randomized controlled trials that is more suitable for assessing the benefits of diagnostic tests, a former FDA official said at a recent conference.
"I can go on and on and on about the silliness of randomized controlled trials that are often required to demonstrate the value of diagnostics," Felix Frueh, vice president of research and development of personalized medicine at the pharmacy-benefit manager, said during a conference last week hosted by Cambridge Healthtech Institute in Washington, DC.
"I think [requiring RCTs] is really an over-reach, something that really came out of the drug environment, and it is not necessarily designed for the diagnostics environment," added Frueh, who prior to joining Medco was the associated director of the pharmacology division at the US Food and Drug Administration's Center for Drug Evaluation and Research.
Industry observers have similarly criticized that the evidentiary standards demanded by the FDA and payors are crafted for a drug-centric healthcare system, and may be overly stringent when evaluating the safety and efficacy of diagnostics [see PGx Reporter 04-25-2007].
There is some indication that this may be changing.
Recently the FDA updated the label for colorectal cancer drugs Vectibix and Erbitux to indicate that patients with certain KRAS mutations would not respond to these epidermal growth factor receptor-inhibiting monoclonal antibodies, and therefore, should not receive the drugs. In order to make this change to the label, the FDA, despite its proclivity towards prospective clinical trials, looked at retrospective analyses from drugmakers Amgen and from Bristol-Myers Squibb/ImClone [see PGx Reporter 07-22-2009].
Last week, a multidisciplinary panel led by the National Institutes of Health and the Centers for Disease Control and Prevention published a set of recommendations for the personal genomics industry, in which the group extensively discussed when randomized-controlled trials are and are not appropriate in the evaluation of personal genome tests. "For most genomic applications (and many other diagnostic tests), direct evidence about the effectiveness and value of testing is rarely available from RCTs," the group wrote in a paper published in Genetics in Medicine (see related story, in this issue).
However, the group also outlined in the paper instances when RCTs are useful, such as developing direct evidence for the clinical utility of personal genetic tests compared to behavioral and pharmacological interventions.
For instance, RCTs could be used to identify subgroups of individuals based on the PG profiles where interventions are most effective and to avoid adverse drug reactions. "Even if no differences in the effects of interventions exist by genotype, RCTs can be used to assess whether genotype-based interventions can be more effective overall if they improve adherence to available interventions that are designed for the general population," the group wrote.
At the CHI meeting last week, Frueh said that going forward, there will be "many more studies in the future that are observational in nature, cohort designed, retrospective analysis, et cetera, that will demonstrate the value of diagnostics and will also give good reason for the uptake of the diagnostics in the market.
"If you think about it, medicine has not evolved around randomized controlled trials. Much of what we practice today has evolved, because we observe, we adjust, we readjust … and I think diagnostics will be no different," he added.
Although there are signs that FDA may consider alternative clinical trial designs for pharmacogenetic products on a case-by-case basis, there has been little indication that payors would be flexible under similar conditions.
In May, the Centers for Medicare & Medicaid Services decided not to cover PGx-guided warfarin dosing for its Medicare beneficiaries. Instead, CMS proposed a more "appropriate" alternative, employing a "coverage with evidence development" strategy, in which it would pay for PGx-based warfarin dosing only for Medicare beneficiaries who are part of a prospectively designed RCT showing pharmacogenomics-guided dosing strategies improve health outcomes over standard dosing methods [see PGx Reporter 05-06-2009].
At the time, researchers and industry observers pointed to Medco's observational studies as an example of how CMS should alter its CED criteria for warfarin genetic testing. Some felt CMS's CED criteria were too stringent and out-of step with scientific advancements.
While Frueh didn't say that Medco's observational studies — which have so far looked at the clinical utility of PGx-based warfarin dosing and tamoxifen administration — would definitively change the way payors look at evidentiary standards for these genetic tests, he did suggest that Medco's efforts offer an alternative model for gathering evidence beyond the RCT paradigm.
Medco is collaborating with the Mayo Clinic to look at whether PGx-guided dosing information on warfarin will change doctors' treatment practices and change patient outcomes [see PGx Reporter 12-06-2006].
The pharmacy-benefit manager is also working with the Laboratory Corporation of America to study if Roche’s AmpliChip CYP450 test can help physicians prescribe a safer and more efficacious dose of tamoxifen to breast cancer patients most likely to benefit from the drug [see PGx Reporter 10-31-2007].
"These observational studies are setting a new standard for these types of studies by looking at outcomes in real-life settings," Frueh said.
[ pagebreak ]
By comparison, Frueh noted that the large RCT study being conducted by the National Heart, Lung, and Blood Institute is still looking at traditional parameters in a specialized setting, which artificially inflates the baseline characteristics of trial. "As a result, the genetic test doesn't show to have a large effect on outcomes," Frueh said.
The Agency for Healthcare Research and Quality and NHLBI are conducting a randomized-controlled trial, called Clarification of Optimal Anticoagulation through Genetics or COAG, that will enroll more than 1,500 patients starting on warfarin and compare outcomes when warfarin is dosed using genetic and clinical information versus clinical information alone.
Separately, the University of Washington in Seattle, the University of Utah, the US Food and Drug Administration, the principal investigators of the Harvard Creating an Optimal Warfarin Nomogram Trial, and the Warfarin Pharmacogenomics Consortium are enrolling 5,000 patients in a study to develop a PGx-based dosing algorithm for warfarin.
Medco is still analyzing the results from its warfarin and tamoxifen PGx studies, and intends to publish the results in the coming months. Frueh did not provide a more specific timeline for when the study results would be made public.
However, he did note that preliminary analysis of the data in the warfarin study shows that based on pharmacogenetic data, physicians kept the original starting dose in 30 percent of patients. "That means 70 percent of patients got the wrong starting dose" of warfarin when pharmacogenetic data was unavailable, Frueh pointed out.
In the tamoxifen study, 67 percent of doctors agreed to order the AmpliChip test and 80 percent of patients agreed to be tested. Most doctors who did not order the test said their patients had been already tested (12 percent) or that their patients did not fit the criteria for testing (11 percent). Very few doctors didn't give the test because they were "not interested" (6 percent) or unsure about testing (2 percent).
Preliminary data from these observational studies also suggest that at the very minimum most physicians do not need to be convinced that genetics plays an integral role in how people respond to certain medications. However, study data show there is a critical need to educate the physician community about genetics and make them more comfortable assessing genetic information alongside their patients' medical history.
At the same conference, a representative from the national insurance company UnitedHealthcare said that inaccuracies made by doctors in administering genetic tests often reduce the cost-savings insurance companies would have seen from administering certain pharmacogenetic tests [see PGx Reporter 08-12-2009]