315P - Long-term tamoxifen benefit in pre- and postmenopausal patients of high and low risk with luminal A and B breast cancer

Background

Breast cancer is a heterogeneous disease with some patients developing distant recurrence early and others decades later, or never. Patients with ER-positive tumors have a considerable long-term risk and can be subdivided into two main groups: luminal A and luminal B subtype. Although tamoxifen therapy is an essential treatment, the long-term benefit for patients with luminal A and B subtype is not well understood.

Methods

Secondary analysis of 952 ER-positive/HER2-negative patients with luminal A or B molecular subtype from the Stockholm Tamoxifen 2, 3 and 5 trials randomized to receive tamoxifen therapy (tamoxifen alone or with goserelin) or control. Agilent microarray gene expression data was used to classify tumors into PAM50 molecular subtypes. The primary endpoint, distant recurrence free interval (DRFI), was assessed using Kaplan Meier analysis, and multivariable Cox proportional hazards regression.

Results

Multivariable analyses showed a significant benefit from tamoxifen in both luminal A (HR=0.58, 95% CI 0.43-0.79) and luminal B patients (HR=0.67, 95% CI 0.46-0.99). Stratifying by tumor and patient characteristics revealed that most favorable prognostic markers were associated with benefit from tamoxifen in both subtypes. However, for premenopausal women a significant benefit from tamoxifen was seen for luminal B patients (HR=0.46, 95% CI 0.22-0.95) but not for luminal A (HR=0.66, 95% CI 0.34-1.29). For postmenopausal patients the reverse was noted (luminal A: HR=0.53, 95% CI 0.37-0.75, luminal B: HR=0.77, 95% CI 0.49-1.21). When comparing the subtypes, patients with luminal B subtype had increased risk compared to luminal A patients for PR positive (HR=1.61, 95% CI 1.16-2.23), lymph node-negative (HR=1.99, 95% CI, 1.23-3.24) and low genomic risk (HR=1.67, 95% CI, 1.15-2.43) tumors.

Conclusions

Our findings suggest a long-term benefit from tamoxifen for most patients with less aggressive tumor characteristics irrespective of molecular subtype, but the benefit differed by menopausal status. Luminal B patients with PR-positive, lymph node-negative or low genomic risk tumors had a higher risk as compared to patients with luminal A tumors.

Clinical trial identification

The trial was initiated in 1976 which was before trial registration was done in Sweden.

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Swedish Research Council (Vetenskapsrådet), Swedish Research Council for Health, Working life and Welfare (FORTE), ALF medicine, Gösta Milton Donation Fund (Stiftelsen Gösta Miltons donationsfond), Swedish Cancer Society (Cancerfonden), Stockholm Cancer Society (Cancerföreningen iStockholm).

Disclosure

L.J. Van't Veer: Financial Interests, Personal, Stocks or ownership, Laura is a cofounder, stockholder, and part-time employee of Agendia.: Agendia. All other authors have declared no conflicts of interest.