By Kate O'Rourke

San Antonio—The Breast Cancer Index (BCI, bioTheranostics) outperforms the Oncotype DX Recurrence Score (RS; Genomic Health) and the immunohistochemical (IHC)4 in accurately predicting breast cancer recurrence risk five to 10 years after a disease-free period, according to the TransATAC (Arimidex, Tamoxifen, Alone or in Combination) study, presented by Dennis Sgroi, MD, director of breast pathology at Massachusetts General Hospital in Boston, at the San Antonio Breast Cancer Symposium (abstract S1-9).

The ATAC trial 10-year follow-up data demonstrated that late disease recurrence is a hallmark of estrogen receptor (ER)-positive breast cancer, with greater than 50% of recurrences occurring after five years of adjuvant tamoxifen or anastrozole therapy. TransATAC collected tumor blocks for biomarker assessment. Previous analyses have shown that RS and IHC4 (ER, progesterone receptor, HER2 and Ki67) predict overall recurrence risk in the TransATAC cohort, beyond the Clinical Treatment Score (CTS), an algorithm consisting of nodal status, tumor size and grade, age and treatment.

In the new TransATAC study, researchers evaluated whether BCI adds prognostic information to clinical variables in predicting distant recurrence in ER-positive, lymph node–negative patients with primary breast cancer. The test stratifies patients into three risk groups and combines two independently developed biomarkers: HOXB13:IL17BR gene expression ratio, which is both prognostic and predictive for extended adjuvant hormone therapy, and the molecular grade index, a set of cell cycle–related genes that predicts distant recurrence beyond tumor grade. The study cohort included 665 primary tumor samples.

At 10 years of follow-up, BCI distinguished three risk groups with a 10-year rate of distant recurrence of 4.2% in the low-risk group, 18.3% in intermediate-risk patients and 30% in the high-risk group. “In analyzing the comparative prognostic performance over the same time frame [0 to 10 years], one sees that BCI, IHC4 and the recurrence score demonstrate highly significant prognostic performance,” said Dr. Sgroi. “BCI and IHC4 provided equivalent prognostic information and both biomarkers provide greater prognostic information than the recurrence score.”

Dr. Sgroi said recurrence information is valuable in two time frames. The early recurrence time frame is at diagnosis, when one is considering using adjuvant hormone therapy alone or in combination with other systemic therapy. The late recurrence time frame is at five years postdiagnosis, when one is considering extending adjuvant therapy for patients who are disease-free after five years of hormonal therapy.

BCI identified two early recurrence risk groups. The first group, 83% of the cohort, consisted of BCI low- and intermediate-risk patients who had an average five-year rate of distant recurrence of less than 4%. The second group consisted of BCI high-risk patients with a five-year rate of distant recurrence of 18.1%.

BCI identified two late recurrence risk groups. The first, 61% of the cohort, consisted of BCI low-risk patients who had a five-year rate of distant recurrence of 3.5%. The second group consisted of intermediate-/high-risk patients who had an average rate of distant recurrence of 13.5%. BCI was a significant prognostic factor beyond CTS for prediction of late distant recurrences (Tables 1 and 2).

Table 1. Multivariate Analysis of Prognostic Performance: Early Recurrence (Years 0-5)
Likelihood Ratio Statistic P Value
Breast Cancer Index 15.4 <0.0001
IHC4 28.8 <0.0001
Recurrence Score 18.2 <0.0001
Table 2. Multivariate Analysis of Prognostic Performance: Late Recurrence (Years 5-10)
Likelihood Ratio Statistic P Value
Breast Cancer Index 8 0.0005
IHC4 1.6 0.2
Recurrence Score 0.5 0.5

“In the early recurrence time frame, all three biomarkers demonstrate highly significant prognostic performance in a multivariate analysis adjusted for the Clinical Treatment Score,” said Dr. Sgroi. “However, when we compared the performance in the late recurrence time frame. … BCI demonstrates sustained prognostic performance in a multivariate analysis, whereas IHC4 and the RS lose their prognostic ability.”

A patient with an intermediate BCI score will do well in the first five years with adjuvant hormone therapy, Dr. Sgroi said, but should be considered for extended adjuvant therapy at that point.

Mathew Goetz, MD, a medical oncologist and an associate professor of oncology and pharmacology at Mayo Clinic in Rochester, Minn., said the three tests provide different information, with BCI providing “better information in years 5 through 10.” He said the test has been validated in multiple data sets, including the MA.17 trial. From a clinician standpoint, he continued, the lymph node–negative, ER-positive patients are the ones in whom clinicians are least likely to use extended adjuvant hormone therapy, and identifying a high-risk subset in this group in years five through 10 is novel.

“I would consider using this test in lymph node–negative, ER-positive patients that have completed five years of tamoxifen,” said Dr. Goetz. He pointed out, however, that many community oncologists are using aromatase inhibitors earlier on, and studies have not defined the benefit of extending aromatase inhibitor therapy in this group. “If we have additional data demonstrating that 10 versus five years of aromatase inhibitors is beneficial, this will really extend the potential ability of this test at that point,” he said.