NEW YORK (GenomeWeb) – New research has helped illuminate how physicians are using Genomic Health's Oncotype DX to guide treatment decision-making for women with early-stage breast cancer: whether they tend to order the test in the correct circumstances, and how they then act on the results.
The results, published today in the journal Cancer, suggest that when Oncotype DX is administered, physicians' decisions and patient treatment largely reflect the test's results. However, the data also show that a percentage of women who should not receive the test based on current guidelines did receive testing, and depending on their clinical risk, a certain percentage of women were also treated not in accordance with their genomic risk scores.
Oncotype Dx is an algorithmic assay that gauges the expression of 21 genes, with the aim of distinguishing women with the lowest risk of recurrence, who can avoid chemotherapy, from women with higher genetics-based risk who should be more confident in their need for treatment.
Previous research has suggested that the test can reduce overall use of chemotherapy because up to half of tested patients are classified as low-risk, while only about 10 percent have high risk scores.
The remaining patients fall into an intermediate category for whom the benefit of chemo is still not clear. However, data from Genomic Health's TAILORx trial are expected to yield new clarity on the utility of Oncotype DX for these patients.
In the study published today, a group led by investigators from the University of Michigan   examined surveys from a cohort of 3,880 women treated for breast cancer in 2013 and 2014. These women were part of the Los Angeles County and Georgia Surveillance, Epidemiology, and End Results (SEER) registries.
These registries included women's reports of chemotherapy recommendations, their actual receipt of chemotherapy, testing experiences, and their decision satisfaction. This was linked with their Oncotype DX risk score (RS) and other tumor molecular data.
According to the authors, their goal was to try to add to what is currently limited data on how Oncotype DX is actually used by oncologists, and whether they and their patients follow the recommendations generated by the test.
"Decisions about testing and treatment are less understood in the United States, where [test] use is more common and treatment occurs in more diverse settings," the authors wrote.
Within the larger cohort from the LA County and Georgia SEER participants, the researchers identified 1,527 women with stage I or II, ER- and PR-positive, HER2-negative disease, of which 778 received an Oncotype DX test results.
The researchers then divided the cohort into three clinical categories — patients with node-negative, clinically favorable disease; those with node-negative but clinically unfavorable disease; and those with node-positive disease.
Broken down by disease state, 62.6 percent of patients who had node-negative, favorable disease got the test while 24.3 percent who had node-negative but unfavorable disease got it.
Interestingly, 13 percent of patients with node-positive disease in the registries also got tested. Current guidelines typically recommend that these women should always receive chemotherapy, and therefore do not need Oncotype DX or similar tests to guide treatment. However, Genomic Health said that an ongoing clinical trial is looking at the value of the test for this group.
According to the study authors, results for the node-positive group underscore "growing support of wider RS use for tailoring treatment recommendations despite guidelines recommending chemotherapy … for these patients."
Overall, doctors recommended that about 47 percent of women in the study could avoid chemotherapy, 40 percent should have it, and 12 percent of the women reported that their oncologist left the decision up to them. About 31 percent actually received chemo.
The Oncotype DX RS results correlated well with these recommendations — nearly all the patients with high risk scores (RS of 31-100) received a chemotherapy recommendation (between 87 percent and 96 percent, depending on their clinical subgroup).
In contrast, a majority, though not all, of patients with low-risk results (RS of 0-18) received a recommendation against chemo. Depending on clinical factors, between 66 percent and 78 percent of women with a low RS were told that they could avoid treatment.
In terms of the treatment patients actually did receive, the correlation with test results was similar. Most patients with high RS got chemo — 87 percent to 100 percent across the subgroups — while few of those with low scores had this extra treatment. The percent of low-RS patients who got chemo ranged from 3 percent amongst those with the lowest clinical risk up to 27 percent for those with node-positive disease.
According to the authors, intermediate scores were associated with chemo use at rates between the low- and high-score groups across the three clinical subgroups.
The researchers also compared chemo use among women who did get tested, and those who did not. More than four times as many clinically low-risk women who did not get Oncotype DX testing received chemotherapy (13 percent compared to only 3 percent amongst low-risk women who received a low RS), suggesting that use of the test had a significant impact in reducing potentially unnecessary chemo use.
The impact of testing also appeared high for the node-positive patients — 83 percent of untested node-positive women received chemo, compared to only 27 percent of node-positive women who had a low RS.
"Results served to identify node-positive patients with low scores for whom chemotherapy might logically be omitted. However, definitive evidence for the benefit of RS testing among node-positive patients awaits the results of clinical trials," the study authors wrote.
Beyond these results, the study also found no significant differences in Oncotype DX use across education, income, and race/ethnicity categories.
The majority of patients — about 60 percent — accurately recalled receiving the test, and nearly two-thirds thought the test was impactful, either shifting their opinion toward or away from chemotherapy.
However, the fact that no more than 60 percent of patients accurately recalled their test result "suggests that while precision medicine for breast cancer has left the station, we have left a few women behind," first author Christopher Friese said in a statement. "There's opportunity in the oncology community to improve how we explain to women the purpose of these tests, how to interpret the results, and what the results mean for their breast cancer treatment," he added.