Patients and doctors alike agree on the need for tests that show which patients will benefit from which treatments. For more than 10 years, Genomic Health’s Oncotype DX has been used to try to identify women with hormone positive breast cancer who may not need chemotherapy. The Oncotype DX breast cancer assay was was approved by the FDA in 2004 and wasthe first early stage breast cancer recurrence test to go to market. It uses a 21-gene assay to inform treatment decisions by predicting the benefit of systemic therapy after surgery as well as estimating the risk of recurrence. Oncotype DC is generally considered to be an accurate predictor of recurrence within the first five years of diagnosis for patients without positive lymph nodes.
In the years since, several other companies have introduced competing tests, such as Mammaprint, which provides their own estimate of risk for early stage breast cancer to help guide doctors and patients in making decisions about treatment after surgery. One such test is EndoPredict, developed by Sividon Diagnostics, which aims to identify a subset of patients who have a particularly good prognosis in the decade after diagnosis following five years of hormone therapy.
On Thursday, researchers from the Institute of Cancer Research in the UK presented data on EndoPredict (EPclin), which uses molecular signatures combined with a more traditional assessment (tumor size and lymph node status) to predict the risk of late recurrence, between years 5-10 after five years of treatment with hormone therapy alone, for patients with hormone-positive HER2-negative breast cancer. Although this study needs further validation, initial data suggests the test successfully identifies a subset of patients who have a good prognosis up to ten years after diagnosis, and may not need chemotherapy. What is novel is that EPclin generates a score by integrating gene profile data with nodal status and tumor size.
Over half of all cases were deemed low risk (59%), meaning patients had less than a 10% risk of recurrence over 10 years, while 41% were deemed high risk, meaning they had a greater than 10% risk of recurrence over 10 years. Overall, 6% of the patients deemed to be low risk by EPclin developed distant recurrence compared to 10% of patients with low-risk Oncotype DX recurrent scores. Looking at patients with positive lymph nodes, who are generally considered to be higher stage and higher risk, the EPclin test outperformed Oncotype DX with only 5% of patients going on to develop a recurrence compared to 25% of patients with low Oncotype DX scores.
The very next paper presented on Thursday morning looked at data from the phase III SWOG trial which tries to identify node-positive patients who can avoid chemotherapy. Genomic Health was the industry partner for the SWOG trial. The trial design started in 1988, looking at post-menopausal women with hormone-positive breast cancer with positive lymph nodes, treated either with tamoxifen alone or tamoxifen with chemotherapy. There were two chemotherapy arms, one in which patients started chemo at the same time as tamoxifen and the other where patients were treated sequentially.
Previously, researchers have been able validate the Oncotype DX score for hormone positive, lymph node negative patients. Now researchers used the stored tissues of women collected 25 years ago to assess whether Oncotype Dx’s 21 gene score also holds up in patients with positive lymph nodes.
Researchers looked at 20,000 genes to see if they could find patterns of relapse, separating the first five years and the second five years after diagnosis. Thousands of genes are associated with early relapse, but these researchers found that only nine genes were associated with relapse between years 5-10.
Metagenes are patterns of gene expressions, rather than individual genes. (For example, groups of genes associated with proliferation.) For early relapse, estrogen receptor was the biggest predictor. These researchers reported that there are five metagenes associated with late recurrence and three metagenes associated with benefit from chemo. While lymph node involvement did matter in terms of recurrence, researchers found there was a small group (about 10%) with these five metagenes who had excellent prognosis, even with positive lymph nodes.
While this data is not clinically actionable in the immediate future, the hope is that researchers will develop better tools to predict risk of recurrence and help identify which patients will benefit from which treatment, to reduce the number of women getting chemotherapy unnecessarily.