乌鸦传媒

Results of the PAM50 gene signature accurately stratified patients with early breast cancer into groups with a high- or low-risk of distant recurrence within 10 years, Danish investigators reported.

After a median follow-up of 9.2 years, patients with low-risk node-positive disease by the Prosigna PAM50 assay had a 10-year recurrence risk of 3.5% as compared with 22.1% for those classified as having a high risk of recurrence. Among patients with node-negative disease, the 10-year risk of recurrence was 5.0% for patients with low-risk disease and 17.8% for high-risk patients.

An analysis of recurrence risk by intrinsic breast cancer subtype showed that patients with luminal B tumors had significantly worse outcomes as compared with those with luminal A, Anne-Vibek Laenkholm, MD, of Zealand University Hospital in Slagelse, Denmark, and coauthors reported in the .

"This study identified a substantial population of patients with one to three positive lymph nodes with sufficiently low risk to safely avoid chemotherapy. By tailoring risk categorization to the number of positive nodes, 37% of patients with a single positive lymph node and 15% of patients with two positive nodes were characterized as low risk, by Prosigna, with very favorable outcomes when treated with adjuvant endocrine therapy alone.

"Although chemotherapy could still provide some relative additional benefit for patients, when the low-risk group carries a residual rate of recurrence of <5% at 10 years, the absolute benefit that can be achieved by still giving chemotherapy could be, at best, very small (and more than counterbalanced by toxicities)."

Acknowledging the availability of several genomic risk-assessment assays, the authors of an emphasized the need for clinicians to understand the differences among the assays in order to use them appropriately. For example, patients classified as low-risk by the PAM50 assay had a lower 10-year risk of distant recurrence as compared with the widely used 21-gene Oncotype Dx assay, which was validated in patients with node-negative disease and primary tumors that were mostly ≤2 cm in size.

"These observations highlight the differences between the assays, particularly with respect to how risk groups are defined and classified," wrote Ricardo L. B. Costa, of Moffitt Cancer Center in Tampa, FL, and William J. Gradishar, MD, of Northwestern University in Chicago.

The diverse offerings among assays to inform decision-making about adjuvant chemotherapy is reflected in recent clinical guidelines. For example, last year the American Society of Clinical Oncology amended an existing guideline to include support for use of the 70-gene MammaPrint Assay for selected patients. The decision was influenced by results of a European study suggesting that a substantial number of women with low-risk breast cancer might be able to avoid chemotherapy.

The clinical guidelines support use of the Oncotype Dx assay in patients with ER-positive breast cancer and one to three positive nodes. In the editorial, Costa and Gradishar stated that "the collective data from several experiences with the 21-gene [risk score, RS] suggest that patients with ER-positive disease, one to three positive nodes, and a low RS can safely avoid adjuvant chemotherapy."

In addition, the European Society for Medical Oncology added a to its clinical guideline for early breast cancer in 2015. In the United States, the PAM50 has FDA clearance for prognostic use.

The study by Laenkholm and co-authors involved analysis of tissue specimens from 2,740 women with newly diagnosed, early breast cancer during 2000 through 2003. All of the patients underwent breast-conserving surgery or mastectomy, followed by adjuvant endocrine therapy without chemotherapy.

After excluding patients with HER2-positive disease, the investigators performed PAM50 analysis of tissue samples from 2,558 patients with ER-positive/HER2-negative disease, including 1,395 patients with nodal involvement. The cohort had a 10-year distant-recurrence rate of 8.9%, and 46 patients (1.8%) died of breast cancer.

The PAM50 gene expression data were combined with clinical variables to generate a risk category and associated score, representing a patient's risk of distant recurrence. The investigators compared the PAM50 results with PAM50 molecular profiles of the four intrinsic breast cancer subtypes (luminal A, luminal b, basal, and HER2 enriched) to determine similarity.

PAM50 results classified 359 patients (26%) with node-positive disease as low risk, and those patients subsequently had a significantly lower 10-year recurrence risk as compared with the 648 patients classified as high risk (95% CI 1.9-6.1% versus 95% CI 18.6-25.8%, P<0.001). The results between high- and low-risk patients differed significantly in both the node-negative and node-positive subgroups (P<0.001).

Intermediate-risk patients with node-positive disease had a 10-year recurrence risk of 7.3%, and the intermediate-risk, node-negative subgroup had a recurrence risk of 11.5% at 10 years. Both were significantly reduced as compared with high-risk patients (P<0.001).