Individualized surveillance and follow-up based on breast cancer (BC) subtypes and risk of relapse in BC patients who received curative surgery.

Abstract

e11053 Background: Gene expression profiles using DNA microarrays have shown at least 5 distinct breast cancer (BC) subtypes. According to these subtypes, clinical features can vary in relapse patterns, preferential metastatic sites, and the risk of systemic failures. However, the surveillance programs which have been developed to date are uniformly designed, and do not reflect the differences among subtypes. The purpose of this study was to evaluate the need for an individualized surveillance and follow-up based on subtypes and risk of relapse in BC patients after curative surgery. Methods: We retrospectively analyzed the clinicopathologic characteristics and outcomes of patients with invasive BC who received curative surgery between 2000 and 2004 at the Samsung Medical Center. Results: Among 1,879 patients who underwent BC surgery, 316 (17%) showed signs of relapse during a median follow-up of 75 months. HER2 positivity (HR, 2.3; p < 0.0001), triple negativity (HR, 1.6; p = 0.007), high histological grade (HG) (HR, 1.6; p = 0.001), and TNM stage 3 (HR, 3.7; p < 0.0001) were identified as independent risk factors for prediction of relapse. We used scoring system by these risk factors as follows: score 0, no risk factors; score 1, 1 risk factor; score 2, 2 risk factors; and score 3, 3 risk factors. Patients with a score of 0-2 were identified as being at low risk and those with a score 3 at high risk for relapse. This high-risk group was composed of two subsets: (1) HER2 positive BC with a high HG and TNM stage 3 and (2) triple negative BC with a high HG and TNM stage 3. The relapse rates of low- and high-risk group were 16% and 53%, respectively. The distant relapse rates for both groups were 12% and 51%. The median relapse-free survival for high-risk patients were 49 months whereas not reached for low risk (p < 0.0001). Most instances of relapse occurred within 3 years of surgery for high-risk patients. The 5-year survival rate of the high-risk patients was 65% compared to 96% for low- risk. Conclusions: Our study showed that there is a need to implement an individualized surveillance based on subtypes and risk of relapse for postoperative surveillance and it should be focused more on patients with a high risk of relapse. No significant financial relationships to disclose.