NEW YORK – For breast cancer patients with ductal carcinoma in situ, clinicians generally have used tumor size and grade to determine treatment options, but PreludeDx has worked to provide a more precise and comprehensive option for these patients.
Radiomics firm PreludeDx's DCISionRT test can determine the benefits of radiation therapy in patients with DCIS and is intended to help radiology oncologists and surgeons pinpoint the best course of action for a patient and better personalize treatment.
The test uses prognostic markers to group women based on their risk of 10-year local recurrence of ducal carcinoma in situ. The test is based on original research from the University of California, San Francisco to find biomarkers related to DCIS recurrence, which PreludeDx licensed.
According to PreludeDx President and CEO Dan Forche, approximately 75 percent to 85 percent of women with DCIS receive surgery and radiation therapy, but for many, adding radiation doesn't significantly change outcomes. Using the DCISionRT test, which rates patients' risk on a scale from 0 to 10, clinicians can determine whether only surgery is a better option. The test can also break out invasive risk, in addition to local recurrence risk.
To get the DCISionRT score, PreludeDx uses a nonlinear algorithm that analyzes specific precision markers related to DCIS recurrence or progression and the interactions between them, as well as clinical factors. Those markers are PR, FOXA1, HER2, Ki-67, COX-2, p16, and SIAH2 and the clinical factors include age, palpable lesion extent, and margin, Forche said. The COX2, P16, and Ki-67 markers were originally used by UCSF researchers, and PreludeDx added the others to round out the test.
Forche emphasized the importance of the nonlinear algorithm as a reason PreludeDx's test works better than comparable DCIS tests, including Genomic Health's Oncotype DX Breast DCIS Score. Rather than looking only at the biomarkers and their presence to measure risk, the algorithm monitors all interactions between different biomarkers and clinical factors to maximize risk prediction. "A linear algorithm is sort of A plus B equals C," Forche said. A nonlinear algorithm, meanwhile "is not just A plus B equals C where you add those up."
In a study published in Clinical Cancer Research in 2018, the researchers noted that the nonlinear algorithm allowed "for risk derived from one biomarker to depend on another biomarker, whereas previous development efforts focused linear weighting for each factor, which is unable to account for complex interactions."
As an example, researchers noted that when the tumor-suppressor protein p16 expression is elevated, cell-cycle progression should be decelerated as indicated by cellular proliferation marker Ki-67 having low expression. However, the researchers wrote, "if p16 and Ki-67 expression levels are both elevated, the cell cycle is not being appropriately regulated, leading to increased risk."
Although there are "only 60,000 cases of DCIS per year," it's the fastest growing breast cancer, and determining best treatments can help stop the disease before it progresses, Forche said.
So far, the company has conducted studies with five cohorts of patients and has been able to reclassify patients to the correct risk group 40 to 50 percent of the time, Forche said. It expects to come out with another four to five publications in 2021, he added.
In an external validation study with 455 Kaiser Permanente patients published last year in the American Association for Cancer Research journal, the decision score determined by DCISionRT was prognostic for the risk of later breast events for women diagnosed with DCIS, following breast-conserving surgery.
Previously, clinical factors were the main way clinicians determined treatment options. In the Kaiser study, approximately 50 percent of patients with clinical factors considered low risk, such as negative surgical margins, lower tumor grade, smaller size, no necrosis, and negative family history, were classified into the elevated risk group using the decision score. According to the study's researchers, "defining a low-risk group with a small benefit of radiotherapy could possibly help reduce overtreatment in patients with DCIS."
The study also found a greater than 70 percent risk reduction from radiation therapy in the elevated risk group for invasive breast events. Meanwhile, the low-risk group had a 2 percent benefit from radiation therapy.
Uptake for the test among physicians has been solid, he continued, and almost 1,000 physicians have ordered the test. Among those that have, 90 percent have ordered it again, Forche said.
Chirag Shah, a radiation oncologist at the Cleveland Clinic, said the test is able to "hone in" on patient benefits from radiation and added he discusses it with every DCIS patient he sees. Although many patients already have their minds made up about what treatment they want, some are surprised to hear they likely won't see a benefit from radiation therapy.
"The overarching challenge it meets is better understanding of who to treat and who not to treat … in the same way that I think genomic tests in the systemic therapy realm have really kind of helped patients feel that they have an individual assessment of their benefits from treatment, that's what this is providing in the radiation space for DCIS," Shah said. "There really wasn't anything before that that offered this."
Shah presented a poster and abstract at the 2020 San Antonio Breast Cancer Symposium for PreludeDx's Predict study that found the test impacted radiation therapy recommendations in 45 percent of women overall.
Although Shah noted that he hasn't used the Oncotype DX test in many years, he said the low risk group determined by that test still had high disease recurrence.
PreludeDx's main focus for the test has been the US market, but a recently announced distribution agreement with Australian cancer care network GenesisCare will open the test up to patients in Australia, the United Kingdom, and Spain. Forche said the company is looking at launching the test in other countries, but that it's not a "major focus."
The company is currently working through the reimbursement process for the Centers for Medicare and Medicaid Services to get the test covered, but Forche said it is billing to and working with all payors. The cost is "typical," around the $5,000 mark and comparable to the Oncotype test.
A cost-effectiveness study published last year in JNCI Cancer Spectrum found that testing with the DCISionRT score was more cost effective than not testing with the score. Three strategies were tested: one where no one was tested and no one underwent radiation therapy; one where all patients were tested and elevated risk patients had radiation therapy; and one where no patients were tested and all underwent radiation therapy. Although the first strategy was the most cost effective overall, the second was determined to be more cost effective than the first when the cost of the test was less than $4,588.
The same researchers performed a cost-effectiveness study in 2016 for the Oncotype DX DCIS Score and found that no strategy using the score was cost-effective.
Although DCIS has had the company's full focus, it is developing a test for stage I and II breast cancer to measure radiation benefit and 10-year risk that would be similar to the DCISionRT test. The stage I test is expected to come to market in the next two years. It is also looking at prostate cancer and other diseases, Forche added.
PreludeDx is helping with the tendency in the US to overtreat patients and puts the decision in the hands of the patient, Forche said. According to the Kaiser study's researchers, "defining a low-risk group with a small benefit of radiotherapy could possibly help reduce overtreatment in patients with DCIS." Forche added that the test "allows the women to make up their mind along with the clinician on what is the best treatment strategy for each patient."