C. H. Lin1, W. Kuo1, W. Kuo1 1National Taiwan University Hospital,Surgery,Taipei City, SELECT A STATE/PROVINCE, Taiwan
Introduction:
Previous studies suggested that early luminal breast cancer had favorable prognosis. Nevertheless, some developed distant metastasis as their first recurrence soon after completing loco-regional and systemic therapy. Our objective is to identify those of high risk for early distant metastasis and disease-related mortality.
Methods:
From January 2009 to February 2014, 408 clinically T1-T2N0 breast cancer patients receiving sentinel-node biopsy(SNB) followed by axillary lymph-node dissection(ALND) were reviewed. Based on anatomical and immunohistochemical(IHC) findings, we used Bayesian-Poisson models to predict the numbers of metastatic axillary nodes. We reviewed another cohort composed of 112 consecutive cT1-2N0 estrogen-receptor positive patients with one to two positive sentinel nodes(SN) recovered by breast-conserving surgery and SNB with no further ALND between 2011 and 2015. We stratified their risks of recurrence by estimating their predicted axillary status, and compared the prognosis between different risk groups.
Results:
The number of predicted metastatic axillary node increased with advanced anatomical findings, negative hormonal receptor reactivity, Her2-amplification, highergrade, and Luminal-B1/B2 or Her2-type. At validation, the deviation between prediction and observation was within one node in 75% of cases. In outcome comparison cohort, when predicted score≥5, the risk of developing distant metastasis as the first recurrence was significantly higher(Log-rank p=0.0016) along with poorer disease-free survival (Log-rank p=0.0043) and overall survival(Log-rank p=0.0525) after a median of 4 years follow-up(ranged 1-7.42 years).
Conclusion:
Using integrated clinical pathological data, our models predict advanced nodal status which is associated with risk for distant metastasis as the first recurrence and poor survival in early luminal breast cancer. Among patients with more favorable tumor biology, disease stage at diagnosis may dominate the prediction of patient outcome.