NEW YORK (GenomeWeb) – At the San Antonio Breast Cancer Symposium last week, researchers presented data from many studies exploring the use of molecular technologies to make patient care more precise.
They included studies exploring the ability of breast cancer recurrence molecular diagnostics to predict which patients are likely to experience a relapse and benefit from chemotherapy, as well as the cost-effectiveness of using these tools. Researchers also employed circulating tumor DNA technologies to try to identify which patients are likely to have a poor outcome and see their disease progress, and they conducted whole-genome sequencing studies to track the distinct genomic characteristics of metastatic breast cancer.
Below is a summary of some of the research in these areas featured at the conference.
Recurrence and predicting treatment benefit
The market for breast cancer recurrence testing is a competitive one, where tests that can assess a patient’s risk for relapse, as well as their chance for benefitting from chemotherapy, have the advantage. Since the leading test on the market, Genomic Health’s Oncotype DX, is consistently included in treatment guidelines in both these settings, competing labs are conducting their own studies now to demonstrate these capabilities.
Two studies involving Myriad Genetics’ 12-gene EndoPredict breast cancer recurrence gene expression test explored these features. An analysis of more than 1,700 breast cancer patients demonstrated the test’s ability to predict early distant recurrence (zero to 10 years after initial treatment) and late recurrence (five to 15 years after treatment), according to nodal status. In this study, where some individuals were followed for more than 16 years, patients deemed at low risk by EndoPredict had a 4 percent rate of distant recurrence compared to 16 percent in the high-risk group. According to the study authors, this suggests that EndoPredict can help identify which patients can safely forgo extended endocrine therapy after five years.
Combining data from five large studies involving more than 3,700 breast cancer patients, researchers considered whether EndoPredict can predict chemotherapy benefit in those who have received five years of endocrine therapy or endocrine therapy plus chemotherapy. Patients with a high EndoPredict score and on the combination regimen had significantly lower 10-year distant recurrence risk than those on just endocrine therapy. However, those with low scores had no significant difference in distant recurrence rates on either treatment regimen. According to the researchers, the data suggest that a high EndoPredict score can predict chemotherapy benefit in women with estrogen receptor-positive, HER2 negative breast cancer.
Additionally, researchers independently evaluated the chemo-predictive capabilities of EndoPredict in a prospective study involving 373 breast cancer patients. Approximately 64 percent had low scores and 36 percent had high scores. Of all the enrolled patients, 128 were recommended for endocrine therapy plus chemotherapy, and 92 were compliant in receiving chemotherapy. After three years of median follow up, disease-free survival was 97 percent and distant metastases-free survival was nearly 100 percent in the EndoPredict low-risk group. In the high-risk group, it was 95 percent and 98 percent, respectively.
High-risk patients who received adjuvant chemo as recommended had a three-year disease-free survival rate of 96 percent and were at lower risk of death or recurrence than those with high-risk scores who went without chemo. “These first prospective outcome results show that [EndoPredict] in clinical routine is a valid clinico-molecular marker to predict disease-free survival and guide decision of adjuvant chemotherapy use in hormone receptor-positive, HER2-negative early breast cancer patients with zero to three positive lymph nodes,” the authors wrote.
Meanwhile, for Genomic Health’s Oncotype DX, researchers presented additional results from the large, multi-institutional TAILORx trial, which showed earlier this year that women with early-stage breast cancer over 50 years old with Oncotype DX recurrence scores from 11 to 25 benefit equally when treated with just endocrine therapy, as opposed to hormonal therapy and chemo. At SABCS , researchers presented data exploring whether chemotherapy benefit differed based on race and ethnicity in those with scores from 11 to 25, and showed results for chemotherapy-related quality of life.
In the MINDACT study, researchers previously showed that Agendia’s 70-gene breast cancer recurrence gene expression test MammaPrint can identify early-stage breast cancer patients who can avoid chemotherapy. At SABCS, researchers from Agendia and elsewhere used Blue Cross Blue Shield’s claims data on 6,000 patients tested on MammaPrint to model the cost-saving potential of this test over a decade, compared to an older tool called Adjuvant!Online. The researchers looked at two scenarios, one where all women classified as high risk received chemo plus endocrine therapy and where low-risk patients received endocrine therapy; and another scenario where treatment was modeled based on treatment rates seen in the MINDACT study, with 91 percent of low-risk patients receiving endocrine therapy, while 83 percent of high-risk patients received adjuvant chemo.
The study showed MammaPrint was more cost saving and had a better quality-adjusted life year gain than Adjuvant!Online for both scenarios. “Given the new cases of breast cancer that are expected to be diagnosed in women in the US annually, patients covered in this population by BCBS (~30 percent) and MammaPrint eligible cohort (~36,000 [patients]), projected cost savings are estimated at $529.2 million to $644.3 million per annum for the health insurer,” the authors wrote.
Several studies combined circulating tumor cell (CTC) analysis with circulating tumor DNA (ctDNA) testing, using Guardant Health’s Guardant360 NGS test. In one 200-patient study, researchers led by a team at Northwestern University evaluated when HER2 overexpressing circulating tumor cells appeared in advanced breast cancer patients and found that those with HER2-positive CTCs didn’t necessarily have HER2-positive breast cancer and viceversa.
Breast cancer subtype or HER2 status were not associated with patients with HER2-positive CTCs. However, patients who had HER2-positive CTCs were more likely to have aggressive disease characteristics and bone metastases. In the 168 patients tested on Guardant360, 22 percent had ERRB2 alterations and were significantly associated with HER2-positive CTCs. “Further studies will be investigating the role of HER2-positive CTCs in endocrine resistance and the potential of anti-HER2 therapy in this unique CTC-defined setting,” the researchers wrote in their abstract.
Using the same tools, another team involving many of the same researchers attempted to retrospectively characterize the biological characteristics of inflammatory breast cancer in around 250 patients. They were able to identify specific ctDNA and CTC features for inflammatory breast cancer, for example that patients with low ctDNA had lower HER2-positive CTCs and CTC clusters.
Researchers involved in the I-SPY2 trial analyzed ctDNA using Natera’s Signatera research-use only assay in high-risk, early-stage breast cancer patients enrolled in that study who received neoadjuvant treatment and surgery. The scientists hypothesized that ctDNA analysis during early treatment will enhance the ability of other molecular testing and imaging to predict which patients will have a pathologic complete response and that ctDNA levels after neodadjuvant treatment will be associated with residual cancer burden and recurrence. Plasma samples for ctDNA analysis were collected at four time points in 84 high-risk patients from before receiving neoadjuvant treatment to before surgery and tested for 16 ctDNA variants.
The data showed that as patients got on therapy and continued on their treatment, fewer patients had ctDNA variants detected in plasma samples. Before patients received neoadjuvant treatment, ctDNA levels were associated with increased tumor burden and aggressive biology, reflected in MRI examples, higher MammaPrint scores, and cancer subtype analysis. Out of 84 patients, 27 percent achieved a pCR and all five patients who were ctDNA positive after neoadjuvant treatment did not achieve a pCR.
The researchers are now conducting additional analyses to explore the relationship between ctDNA and outcomes at different timepoints in treatment. "Our study provides a platform to evaluate the clinical significance of ctDNA for serial monitoring of response to neoadjuvant treatment,” the study authors wrote in their abstract, and such testing used in conjunction with other tests can fine-tune pCR as a surrogate endpoint for disease recurrence-free survival and event-free survival.
In another study involving 49 breast cancer patients who received adjuvant therapy, Signatera detected ctDNA ahead of metastatic relapse diagnosed by clinical examination andradiological and biochemical (CA15-3) measurements in 16 out of 18 clinically relapsing patients. One of the patients whose relapse was not detected by ctDNA had a local recurrence that was resected and the other had three primary tumors. None of the 31 non-relapsing patients were ctDNA positive at any time point in the treatment continuum. “Metastatic relapse was predicted by Signatera with high accuracy and a lead time of up to two years,” the researchers wrote in their abstract.
Genomic landscape of breast cancer
A number of studies presented at the meeting contained new insights into genomic features of metastatic breast cancer, providing further evidence that advanced cancers are biologically distinct from early-stage disease.
One study conducted whole-genome sequencing of tumor biopsies from 442 patients and found targetable alterations in 42 percent. Among these patients, 11 percent had high tumor mutational burden or microsatellite instability, making them candidates for checkpoint inhibitors; 13 percent had homologous recombination deficiency and could have received PARP inhibitors; and 24 percent had other mutations that could be addressed by targeted therapy. The analysis also showed that metastatic breast cancers tended to have driver mutations in certain genes more frequently than primary breast cancers, tended to have higher tumor mutational burden, and had differential mutational signatures. The researchers also identified GPS2 as a novel driver gene and found distinct mutational signatures in metastatic disease.
Another group of researchers also looked at DNA from tumors and blood of around 630 metastatic breast cancer patients and conducted whole-genome sequencing. The study showed that when tumors metastasize, they evolve and acquire new genomic alterations.
The evaluation revealed 10 driver alterations that are enriched in hormone receptor-positive, HER2-negative metastatic breast cancer. There were three genomic signatures that were also enriched in this population that were associated with very poor outcome. Among metastatic triple-negative breast cancer patients, there was a subset with somatic biallelic loss-of-function mutations in hormone receptor pathway genes. Overall, metastatic breast cancers were genomically more complex than early breast cancers, and among metastatic patients, there was inter-patient heterogeneity in clonal diversity and mutational burden.