NEW YORK – At the San Antonio Breast Cancer Symposium on Thursday researchers presented data on a genomic signature that can identify patients with ductal carcinoma in situ who are likely to have poor outcomes and therefore should receive more aggressive treatment.
DCIS is a preinvasive breast cancer precursor, and Siri Strand, a postdoctoral scholar at Stanford University School of Medicine who presented the study data at SABCS, noted that many patients with DCIS are overtreated. Her team hoped to identify a genomic signature that could predict which patients with DCIS needed more aggressive treatment to prevent invasive breast cancer in the future.
"Screening by mammography since the late 1980s has reduced breast cancer specific mortality, but also has led to a more than 500 percent increase in detection of DCIS," Strand explained. "At the same time, we have observed a much lower increase in the detection of invasive breast cancer, which indicates there are many clinically irrelevant DCIS [cases] being detected."
There are several treatment options for patients with DCIS, including monitoring, lumpectomy with or without radiation therapy, hormonal therapy, or mastectomy. "There is a strong need to risk stratify patients to identify which [patients] should get aggressive and less-aggressive treatment," Strand said.
To identify a prognostic genomic signature, the researchers analyzed 774 DCIS samples from two independent longitudinal cohort studies: TBCRC 038, a study from the Translational Breast Cancer Research Consortium, and the Resource of Archival Human Breast Tissue DCIS cohort. There was a median of 7.3 years of follow-up that included patients who had invasive breast cancer recurrence after a DCIS diagnosis and those who did not.
The researchers performed RNA sequencing, DNA sequencing, and multiplex protein imaging on the samples and identified 812 genes that were associated with invasive breast cancer recurrence five years after initial DCIS treatment. The 812 genes were involved in cell cycle and metabolism functions, MEK and mTOR signaling, and the P53 pathway.
The sequencing data also identified three subtypes of DCIS, which the researchers called estrogen receptor (ER)-low, quiescent, and ER-high. These subtypes were largely characterized by expression of ER and HER2. The ER-low cluster had significantly higher levels of ERBB2 and lower levels of ESR1 compared to quiescent and ER-high clusters. The quiescent subtypes, which also had high ER expression and low expression of KI67 and other genes, were less proliferative and less metabolically active than ER-high and ER-low subtypes.
Based on their modeling, the researchers found that the 812-gene signature was able to identify patients with increased breast cancer recurrence risk. The patients with poor outcomes showed increased ER expression, MYC signaling, and oxidative phosphorylation, suggesting that these pathways are important for DCIS recurrence and progression.
"This classifier could be the basis for a future clinical test to assess outcomes in patients with DCIS and stratify patients for treatment," Strand said. "It was found significant in multivariant analysis including across treatment, age, grade, and ER status."
This study is part of the Human Tumor Atlas Network's Breast Pre-Cancer Atlas Center, which Strand said is the largest comprehensively profiled DCIS cohort to date. She also noted that the data from this trial will be made available for the research community. Strand's team also plans to continue exploring the signature in other cohorts and prospective studies, she said.