NEW YORK (GenomeWeb) – Results from a survey involving 14 oncologists in Ontario who have ordered Genomic Health's Oncotype DX test suggest some doctors use the diagnostic in breast cancer patients who don't fit the recommended criteria.
The results from this regional Canadian study coincide with a large US-based survey of Medicare patients tested with Oncotype DX, in which researchers found that the test was predominantly used according to treatment guidelines.
In The Oncologist, researchers led by Yvonne Bombard from St. Michael's Hospital in Toronto published a paper in which they described the use patterns and perceptions of 10 academic and four community cancer doctors who used Genomic Health's Oncotype DX test to gauge their patients' risk of breast cancer recurrence and the likelihood that they'd benefit from chemotherapy.
The oncologists interviewed for the study were all based in Ontario. Oncotype DX has been funded within Ontario's healthcare system since 2010. Five other Canadian provinces – Quebec, Alberta, British Columbia, Saskatchewan, and Newfoundland – also now fund the test, before which some women were reportedly paying out-of-pocket for the nearly C$4,000 test.
"We found that oncologists generally valued [gene-expression profiling] as a decision support tool that provided additional confidence in situations of uncertainty and allowed them to order chemotherapy for the right patients on a more consistent basis," Bombard said. "However, the test's perceived value was tempered by oncologists' reservations about its use."
The study cohort was small, Bombard acknowledged, and the survey is not intended to represent views on the test across the Canadian population or in other jurisdictions. Still, she told GenomeWeb, a study of this size is typical of qualitative research experiments and provides insights into the impact that gene-expression profiling is having on clinical practice, about doctors' reservations regarding Oncotype DX, and about the inherent challenges of communicating information about the test to patients.
Overall, the survey found that some oncologists find Oncotype DX to be a "critical" tool to help them garner more confidence about forgoing chemotherapy in breast cancer patients with low recurrence risk. Doctors find value in using the test as a "tie breaker" when patients are at intermediate risk by standard assessment tools or when there are questions about the accuracy of standard risk assessments.
Still, the oncologists had reservations about the high cost of the test, about over-relying on the test results, and about inappropriate use of the test.
One doctor recounted trying to be "very clear" with patients that if the docotr and patient decide to perform a $4,000 test and the results suggest the need for chemotherapy, then patients need to take chemotherapy according to the Oncotype DX results. Another doctor noted encountering colleagues who use Oncotype DX as "just kind of second nature" and order "it on everybody," which could lead to overuse. While oncologists generally agreed that they should only order Oncotype DX if it adds value to their clinical decision making, Bombard and her colleagues noted in the paper that gauging patients' willingness to act on the results as a requisite for ordering the test might be "unintentionally coercive."
Over-reliance on one test over classical pathology indicators was another point of concern among some doctors. One oncologist noted that Oncotype DX is "still a test, it's still not infallible, and it really is only one factor in the big decision that we're making." Another doctor would be very concerned if the initial pathology report for a patient deemed her to be at high risk for recurrence, and the Oncotype DX test result said she was at low risk. This oncologist cautioned against putting more weight on the Oncotype DX results.
Still, according to Genomic Health's own data from a study conducted among British Columbia breast cancer patients and reported in 2012, oncologists changed their decision to treat early-stage, estrogen receptor-positive, lymph node-negative patients with chemotherapy in 30 percent of cases. For 20 percent of these patients, doctors decided to forgo chemo and only give hormonal therapy; and for 10 percent they decided chemo was needed.
However, according to the surveyed oncologists in Ontario, the test did not reduce how much chemotherapy they gave in their practices, but allowed them to more precisely avoid chemo in those that didn't need it. "I think on balance I'm probably prescribing just as much chemotherapy because … there are patients that I may have recommended chemotherapy to that then come back lower risk and don't need it, and then there are those patients that I would not have recommended chemotherapy to … who then come back with a test where I'm more inclined to offer them chemotherapy," one oncologist said.
In a statement, Genomic Health spokesperson Emily Faucette pointed out that cost-effectiveness estimates for Oncotype DX factors in the test's impact on reducing chemo for those who won't benefit and the test's ability to increase chemo for those likely to derive a benefit. The Province of Ontario conducted a 17-month field evaluation of 1,000 patients between 2012 and 2013, and found that Oncotype DX was used by doctors to avoid chemo in 38 percent of patients and to upstage 15 percent of patients to receive chemo.
Oncotype DX has been clinically validated in early-stage breast cancer patients with estrogen receptor-positive, lymph node-negative and node-positive patients. However, because the test was first validated in the node-negative population, most healthcare providers tend to use the test largely in this subpopulation according to treatment guidelines, such as those from the National Comprehensive Cancer Network.
The survey among Ontario oncologists, however, revealed that some doctors are departing from this practice. One doctor recounted colleagues using the test among hormone receptor-negative patients, and another admitted to using it for some patients with grade 3 tumors (signaling quickly growing cancer cells), node-positive disease, and for patients with micro-metastases.
The Ontario Health Technology Assessment Committee in 2010 reviewed available data on Oncotype DX and decided to fund the test for women who are estrogen receptor-positive and/or progesterone receptor-positive, HER2 negative, and node negative. At the time, the committee did not find evidence supporting the use of the test in node-positive patients. The government in Alberta also funds the test for the same indication. "Other Canadian provinces have undertaken similar reviews and implemented their own guidelines," Genomic Health's Faucetted added.
The Ontario study was published around the same time as another large survey of Oncotype DX use among Medicare patients in the US. Researchers from Duke University gathered data on 70,802 Medicare recipients diagnosed with breast cancer between 2005 and 2009 and noted that use of Oncotype DX jumped from being ordered for 1.1 percent of women in 2005 to 10.1 percent in 2009. Oncotype DX has been commercially available in the US since 2004 and received Medicare coverage in 2006.
In this study, published in JAMA Oncology, researchers found that US doctors were predominantly ordering the test for estrogen receptor-positive, node-negative patients according to guidelines, and use of the test didn't vary according to race or socioeconomic status. Among the women who received testing but didn't meet guidelines, most were borderline patients who had a few lymph nodes implicated or had tumor sizes that were a bit smaller than recommended. Researchers also noted that compared to other regions of the country, in the South there were higher rates of the test being ordered for non-intermediate risk patients who had tumor dimensions of greater than 5 mm but equal to less than 10 mm, which seemed to fall outside the scope of recommended use.
The researchers led by Duke's Michaela Dinan concluded that the test was largely being used according to guidelines and didn't display the usage variability as seen with other emerging technologies, such as PET imaging. The authors attributed this consistency or limited variability to a very specific testing indication for Oncotype DX and to the diagnostic being a "send out" test that didn't require doctors and hospitals to make additional investments on new technology.
Given the difference in the size and design of the studies conducted in US and Ontario, it's difficult to gauge to what extent Oncotype DX use patterns would vary between the geographic regions. Furthermore, "it is difficult to speculate," Bombard said, "since Oncotype DX has been used in the US for much longer than in Canada."
Still, Bombard noted that gene-expression profiling tests are increasingly being used in clinical practice, and the usage variations among oncologists haven't been much explored. Indeed, the study by Duke researchers is the first analysis of how Oncotype DX is being used in the US on a national scale.
Bombard noted that the reservations about overuse of, or over-reliance on, Oncotype DX are in line with general concerns about unnecessary lab testing straining the healthcare system. She cited data suggesting that between 10 percent and 50 percent of lab testing in Canada may be unnecessary and one-third of US healthcare spending is attributable to overuse or misuse of medical interventions.
"Assessments of overuse rest on balancing benefits and harms, [and] benefits and costs with patient preferences," Bombard said. "These tradeoffs are inherent in [gene-expression profile] testing, placing [such] tests in the vanguard for potential overuse," she added. "The results from our study reveal potential overuse and divergent practice patterns, and identify a need for clinical practice guidelines to support standardized ordering, application, and interpretation of the test."
The Ontario survey also revealed challenges associated with communicating to patients about Oncotype DX. One oncologist commented that Oncotype DX results give patients confidence that "they're not going to be forgoing a small potential benefit by avoiding chemotherapy, if they're really, truly low-risk."
However, there was variation as to how doctors explained information about the test to patients. Some emphasized that Oncotype DX provides a more personalized result compared to standard assessment tools, while others emphasized the predictive capabilities of the test. The test quantifies recurrence risk as a continuous variable, but some doctors presented the test results in a dichotomous – chemo or no chemo – framework.
Another oncologist noted that it is challenging to explain concepts of risk to patients. "Even though you try to explain it carefully and you make yourself available to ask questions and go over it again, I'm really not sure that most of [the] patients really understand what the test is all about." Bombard noted that the survey results "identify a need for decision aids to support patients' understanding," of gene-expression profiling tests.
According to Bombard, the latest analysis is part of a mixed methods study to investigate the perceived value of gene-expression profiling tests among breast cancer patients and medical oncologists. "We have also conducted a survey with a discrete choice experiment that estimates the utility of the [gene-expression profiling] test relative to other factors among representative samples of Canadian women, breast cancer patients, and medical oncologists," she said. This study is currently under review.