Prognostic Value of the Nodal Ratio and Ki-67 Expression in Breast Cancer Patients Treated with Postmastectomy Radiotherapy

Abstract

Introduction: We performed this study to evaluate prognostic factors of postmastectomy radiotherapy for breast cancer patients undergoing systemic therapy in either preoperative or postoperative setting, in order to identify patients at high risk of disease relapse and survival. Methods: Between 2003 and 2009, 113 patients received postmastectomy radiotherapy in Seoul National University Bundang Hospital: 61 underwent preoperative systemic therapy and 52 received postoperative systemic therapy. The most common chemotherapy regimen was six cycles of docetaxel and doxorubicin in patients with preoperative systemic therapy

and four cycles of doxorubicin, cyclophospha¬mide, and paclitaxel in patients with postoperative systemic therapy. Hormonal therapy was administered in patients with a positive hormone receptor status

and trastuzumab was recommended for patients with a tumor exhibiting c-erbB-2 overexpression (3+) or HER2 gene amplification. For radiotherapy, the chest wall and supraclavicular fossa were irradi¬ated with up to 50.4 Gy at 1.8 Gy per fraction with 5 fractions per week. Following histopathologic parameters were evaluated by immunohistochemical analysis: the status of hormone receptor and the expression of c-erbB-2, p53, Ki-67, and COX-2. The positive cut-off values were immuno¬histochemical staining in ≥1% for hormone receptor, in >10% for p53, in >20% for Ki-67, and a 3+ staining score for COX-2 and c-erbB-2. The analysis of HER2 gene amplification was performed with Fluorescence in situ hybridization. Lymph node status was evaluated by hematoxylin and eosin staining. The nodal ratio was defined as the number of axillary lymph nodes with cancer involvement divided by the total number of excised axillary lymph nodes. The cut-off value was 0.2, after comparing survival rates by using the maximal chi-square method in the R program version 2.13.0. Results: The median follow-up time was 72.3 months (range, 34.0-109.4 months) for surviving patients. In univariate analysis of all patients, disease-free survival (DFS) was associated with age, nodal ratio, and Ki-67 expression

overall survival (OS) was associated with nodal ratio and Ki-67 expression. Pathologic N stage and HER2 expression were marginally associated with DFS and OS. In patients with postoperative systemic therapy, DFS was associated with age, nodal ratio, venous invasion, and Ki-67 expression

OS was associated with age. In patients with preoperative systemic therapy, DFS was associated with ypN stage and nodal ratio

OS was associated with ypN stage, histologic grade, HER2 expression, and p53 expression. In multivariate analysis of all patients, DFS and OS were significantly associated with nodal ratio (p = 0.003 and p = 0.019, respectively) and Ki-67 expression (p = 0.002 and p = 0.015, respectively). Patients were classified into low-risk (nodal ratio ≤0.2 and Ki-67 ≤20%

n=34), intermediate-risk (nodal ratio >0.2 or Ki-67 >20%

n=63), and high-risk (nodal ratio >0.2 and Ki-67 >20%

n=16) subgroups. All low-risk patients were alive at the time of analysis. High-risk (p < 0.001 and p = 0.001, respectively) and intermediate-risk (p = 0.022 and p = 0.008, respectively) patients had significantly shorter DFS and OS than low-risk patients. This prognostic model was statistically significant for DFS when applied to patients with preoperative systemic therapy (p = 0.001) and with postoperative systemic therapy (p = 0.016) separately. We classified patients into three intrinsic subtypes: luminal A (hormone receptor positive and HER2 negative

n=55), luminal B (hormone receptor positive and HER2 positive

n=12), HER2 overexpression (hormone receptor negative and HER2 positive

n=16), and basal-like (hormone receptor negative and HER2 negative

n=30). DFS and OS had no association with intrinsic subtypes (p = 0.249 and p = 0.202, respectively). When our prognostic model was applied to luminal A subtype, there was a marginal association in DFS (p = 0.078), while not in OS (p = 0.173). Conclusions: For breast can¬cer patients undergoing postmastectomy radiotherapy, nodal ratio and Ki-67 are potential prog¬nostic factors. A model using these factors might help predict a poor prognosis. Whether nodal ratio and Ki-67 are also prognostic for different setting of systemic therapy, preoperative or postopera¬tive, warrants further study to develop a more sophisticated prognostic model.