The First Department of Breast Cancer, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center for Cancer, Huan-Hu-Xi Road, He-Xi District, Tianjin, 300060, China.
Key Laboratory of Cancer Prevention and Therapy, Tianjin, 300060, China.
Breast Cancer. 2018 May;25(3):315-324. doi: 10.1007/s12282-018-0833-0. Epub 2018 Jan 20.
pN stage and breast cancer subtypes (BCS) are both well-recognized prognostic indicators. Our previous work has highlighted that patients even with the same pN stage exhibited a significant survival difference in different BCS. Given this achievement, we hypothesized that a statistical interaction might exist between pN stage and BCS. The aim of this retrospective cohort study was to compare the prognostic value of the combined pN stage and BCS (pN stage) with either pN stage or BCS alone, and to determine if this combined new stage could serve as an alternative discriminator of outcome.
We combined pN stage and BCS to create a new variable named pN stage and then divided it into four groups: pN0, pN1, pN2, and pN3. Survival analysis was performed with the use of the Kaplan-Meier method and the log-rank test was used for univariate analysis. For multivariate analysis, cox proportional hazard models were applied, allowing for the estimation of disease-free survival (DFS). To assess discriminatory accuracy of the models, we compared the area under the receiver-operating characteristic curve (AUROC), the Akaike information criterion (AIC), and the Bayesian information criterion (BIC) values. Then, we used this pN stage to generate a TNM staging system according to the 7th AJCC staging system.
A statistical interaction between pN stage and BCS was found. In multivariate survival analysis, the pN stage has been confirmed as an independent prognostic variable of 5-year DFS. The pN stage, with a smaller AIC or BIC value and larger AUROC, was a more powerful predictor of DFS than either pN stage or BCS alone. Results were validated in a separate cohort of patients. The TNM stage proposed in our present study was found comparable to the new 8th AJCC edition which includes anatomic T, N, and M plus tumor grade and the status of the biomarkers Her-2, ER, and PR with respect to prognostic value for breast cancer patients.
The pN stage (combined pN stage and BCS) appears to be a more powerful predictor and discriminator for the outcome of breast cancer, as compared to pN stage or BCS alone, and the TNM stage may serve as a simple, easy-to-use alternative to the 8th AJCC edition staging manual.
pN 分期和乳腺癌亚型（BCS）都是公认的预后指标。我们之前的工作已经表明，即使具有相同的 pN 分期，不同 BCS 的患者的生存差异也很大。鉴于这一发现，我们假设 pN 分期和 BCS 之间可能存在统计学相互作用。本回顾性队列研究的目的是比较联合 pN 分期和 BCS（pN 分期）与单独 pN 分期或 BCS 对预后的预测价值，并确定这种联合新分期是否可以作为预后的替代判别指标。
我们将 pN 分期和 BCS 相结合，创建了一个新的变量，命名为 pN 分期，并将其分为四组：pN0、pN1、pN2 和 pN3。使用 Kaplan-Meier 方法进行生存分析，对数秩检验进行单因素分析。对于多因素分析，应用 cox 比例风险模型，估计无病生存率（DFS）。为了评估模型的判别准确性，我们比较了受试者工作特征曲线下面积（AUROC）、Akaike 信息准则（AIC）和贝叶斯信息准则（BIC）的值。然后，我们根据第 7 版 AJCC 分期系统，使用该 pN 分期生成 TNM 分期系统。
发现 pN 分期和 BCS 之间存在统计学相互作用。在多因素生存分析中，pN 分期已被确认为 5 年 DFS 的独立预后因素。与单独的 pN 分期或 BCS 相比，pN 分期具有更小的 AIC 或 BIC 值和更大的 AUROC，是 DFS 的更有力预测因子。结果在另一组患者中得到验证。本研究提出的 TNM 分期与包括解剖学 T、N 和 M 以及肿瘤分级以及 Her-2、ER 和 PR 状态的新第 8 版 AJCC 版相比，对乳腺癌患者的预后具有可比性。
与单独的 pN 分期或 BCS 相比，pN 分期（联合 pN 分期和 BCS）似乎是乳腺癌预后的更有力预测因子和判别因子，TNM 分期可能是第 8 版 AJCC 分期手册的简单易用的替代方案。
