NEW YORK — Oncotype DX, a gene expression-based assay for assessment of early breast cancer recurrence risk, could have value as a tool to guide the use of adjuvant radiation therapy, based on data from two prospective trial datasets shared at the San Antonio Breast Cancer Symposium late last week.
The test, marketed by cancer genomics firm Exact Sciences, uses algorithms to assess the expression of 21 genes in early-stage breast cancer, or a subset of 12 genes for the extremely localized version of these tumors called ductal carcinoma in situ, and already has a well-established presence in the clinic informing patient and physician decisions regarding the use of chemotherapy.
More recently, investigators have become interested in whether these same gene expression profiles might be used to stratify patients for other treatments like radiation and endocrine therapy based on their predicted recurrence risk.
Interim results from both trials highlighted at SABCS suggest the Oncotype test can indeed do this for radiotherapy, though investigators were cautious about clinical implementation prior to full readouts of the studies in another five years, as well as results from randomized controlled follow-ups that are currently in the works.
One study presented at the meeting and published simultaneously in the Journal of Clinical Oncology targeted a subgroup of postmenopausal women who were younger than those studied in earlier trials of radiation de-escalation. Investigators in the trial, called IDEA, used Oncotype DX results to enroll women aged 50 to 69 years with localized disease and successful lumpectomies who also had risk scores of 18 or less according to the assay.
Nearly all of the participants who opted out of adjuvant radiotherapy were disease-free five years after surgery, according to Reshma Jagsi, chair of radiation oncology at Emory University School of Medicine, who presented the results last Thursday.
According to Jagsi, radiation therapy is standardly recommended after lumpectomy surgery because numerous randomized trials and meta-analyses have demonstrated it provides a modest overall survival benefit.
However, not all subgroups of women attain the same absolute benefit from radiation treatment, and the survival benefit "appears specifically to be restricted to women who have a larger absolute reduction in their recurrence risk from radiotherapy," she said.
Previous studies in older patients have already demonstrated that individuals with favorable clinical risk factors can forgo adjuvant radiation, as long as they are comfortable with a slightly higher risk of recurrence. But the same hasn't yet been shown for younger patients.
"We know that radiation treatment gives you a fairly consistent relative risk reduction," Jagsi said during a press conference, with about a 50 percent reduction in overall recurrence risk and closer to two-thirds for risk of a recurrence in the same breast. The issue is that if a patient's standing risk is already very low, a 50 or 60 percent reduction doesn't offer much of an absolute change.
With IDEA, investigators aimed to sift out a low-risk subset for whom this might be true — that the relative impact of radiation would be negligible enough that patients could be comfortable forgoing treatment. Oncotype DX is already clinically established as a tool to guide the use of chemotherapy in these patients, and the Emory team hypothesized that it could do the same for radiation.
"We're trying to empower patients with choices [and] return to them a sense of agency that, of course, can be so deeply meaningful," Jagsi said.
In the trial, 186 stage I HR-positive, HER2-negative breast cancer patients with risk scores of 18 or under agreed to opt out of radiotherapy while still receiving standard-of-care adjuvant endocrine therapy for at least five years.
All the patients were still alive five years after surgery, and 99 percent were breast cancer-free at this time.
Although the results were striking, Jagsi warned against overexcitement. "As much as we investigators like to have really press-worthy, ready-for-application studies, I think we need to walk with a little bit of caution here," she said.
Although there were only two recurrences before the five-year interim analysis mark, there have since been six more in the ongoing 10-year study, Jagsi said. Moreover, recent advances in radiation have made the treatment much less severe than it had been.
"It is extremely well tolerated, so … I don't want us to scare off, first of all, the patients who clearly do need radiation or really anyone, because what we want is rational deliberation about treatment options for patients."
"The takeaway of this is actually that if there is a younger, postmenopausal woman who meets the criteria for inclusion in IDEA, what we really should be doing is encouraging them to enroll on DEBRA, which is a randomized trial of this approach and will actually collect quality of life information," Jagsi said.
The second report at SABCS focused on an even lower-risk group — patients with ductal carcinoma in situ, commonly deemed "stage 0" breast cancer. Similar to the IDEA trial, this study, ECOG-ACRIN 4112, has followed its cohort for five years out of a planned 10.
According to Seema Khan, professor at the Feinberg School of Medicine and Lurie Comprehensive Cancer Center of Northwestern, there is a lot of discussion about optimization of local therapy in DCIS because it's "clear that some patients are being overtreated."
The E4112 trial incorporated both MRI and the 12-gene DCIS-version of Oncotype DX, exploring whether imaging could guide the use of breast-conserving surgery over mastectomy, and whether subsequent genetic testing could pick out whether certain individuals who got the less severe surgery could also forgo radiation.
The MRI results have been previously published, so Khan focused her discussion on the gene-expression findings.
Overall, 171 women in the trial, about half the total cohort, had breast-conserving surgery and met all the criteria for the next step of Oncotype DX testing. Patients with low scores (less than 39) were advised that they could avoid radiation, while those with intermediate or high scores were advised to follow through with radiation. All had the choice whether to follow their doctors' recommendations or not.
According to Khan, the cohort was divided about equally into those with a low score and those with an intermediate or high score, and they were "quite adherent to the radiotherapy recommendations."
"The low score patients generally agreed to omit radiotherapy, and the higher score patients generally agreed to receive it so that the overall adherence rate to the radiation recommendation was actually 92 percent," she said.
Across the board about 5 percent of the cohort ended up with a local recurrence by the five-year mark. That recurrence rate held true in the low-score group who did not receive radiation, while in the high-risk, radiation-treated patients, the recurrence rate was 4.3 percent.
"Analysis of 10-year outcomes is very important because this is something that will evolve over time, and we will expect to see more events during that period. Also, larger prospective trials are needed, which are under consideration at the moment," Khan said.
As the first prospective study using a genetic risk score to advise women as to whether they should pursue radiation, Khan said the data are reassuring in that the recurrence rate for the treated high- and intermediate-risk group was "much lower than observed previously," but because the trial was not randomized, this shouldn't be interpreted as direct evidence of a radiation benefit.
"I think we need to do larger, better trials to really establish this finding. But it's certainly going in the right direction, and women can be informed about this finding in terms of helping them to make their decision," she said.