Molecular Profiling in Early Invasive Breast Cancer

March 2013, Highlights

In making decisions about adjuvant chemotherapy, biologic subtype has replaced nodal status and tumor size as parameters to consider, although subtype—and tests that define them—are still insufficient for the provision of optimal care, said Antonio C. Wolff, MD, Professor of Oncology at Johns Hopkins University School of Medicine, Baltimore. Dr Wolff spoke at an educational session devoted to molecular profiling.

Phenotype, based on estrogen receptor (ER) and HER2 status, has a strong negative predictive value, but only a modest positive predictive value for treatment benefit. “We still need traditional measures such as nodal status and tumor size,” he maintained.

In the past decade, microarray expression has revealed 4 intrinsic breast cancer subtypes—basal-type (largely triple-negative); HER2-positive; luminal A (mostly ER-positive and low grade by Ki67 index); and luminal B (mostly ER-positive and high grade). Much of the prognostic variation occurs within these 4 groups, and treatment decisions are now largely based on the characteristics associated with them, Dr Wolff said.

It is becoming clear that tumor-cell proliferation is critical in these subtypes, especially in the luminal groups, and that proliferation is associated with response to chemotherapy.

Gene-Expression Profiling

Studies suggest that the adoption of gene-expression profiling has altered the use of chemotherapy. Based on 2012 findings from 7375 patients, the use of these assays grew from 15% in 2006 to 27% in 2008 and the use of chemotherapy declined from 54% to 47%.

Patients with large node-negative or any node-positive cancer were 11-fold less likely to receive chemotherapy; those with small node-negative cancers were 11-fold more likely to receive chemotherapy.

“I think that clinicians are mostly ordering tests for tumors they are not so sure about. They have a sense of what they are going to do with a given patient when they order the test, and, in most cases, the test results will be used to confirm their decision,” Dr Wolff said. “Molecular assays must not yet override results with standard assays.”

Future Profiles

Efforts are under way to molecularly refine the classification system for tumors. The Cancer Genome Atlas (TCGA) of molecular “portraits” is elucidating phenotype, ge­notype, epigenetics, and proteomic features (eg, copy number variations). “The question is what to do with this information,” Dr Wolff said. “One of the challenges is that the TCGA dataset is still not correlated with clinical outcomes. As we move toward the concept of precision medicine, we need this.”

Because of the heterogeneity of tumors, many questions are yet to be answered. He predicted that the field will move from qualitative, descriptive prognosis toward a quantitative prognosis based on individualized therapies.

Related Articles