Predicting Non-Sentinel Node Metastasis in Breast Cancer Patients

描述（由申请人提供）：前哨淋巴结（SLN）活检是乳腺癌患者准确分期腋窝的微创方法。虽然发现阴性SLN可以避免腋窝淋巴结清扫（ALND），但在SLN中发现微转移则需要完全性ALND。高达80%的sln阳性患者在腋窝没有进一步的疾病。在预测哪些SLN阳性患者发生非SLN转移的可能性较低，从而可以避免ALND方面，我们目前的知识有很大的差距。一种新的基于术中RT-PCR的检测方法（GeneSearch(r)）已被证明与乳腺癌患者的SLN肿瘤负荷相关。这项新技术在预测哪些非sln患者不会进一步发病方面的潜在效用尚未得到研究。利用这些数据开发一个强大的非sln状态的临床预测规则将填补我们的知识空白，并提供及时的结论，这将对患者护理产生直接影响。为达致这个整体目标，我们提出以下具体目标：(1)使用双变量和多变量逻辑回归分析确定GeneSearch(r)检测的定量周期时间值对非sln状态的影响；(2)使用术前和术中可用的临床病理因素，包括从GeneSearch(r)检测获得的数据，创建并验证（使用bootstrap技术）临床预测规则，以预测sln阳性患者的非sln状态。我们假设，在双变量分析中，乳房x线球蛋白（MMG）和细胞角蛋白19 （CK19）的定量RT-PCR周期时间与非sln转移显著相关，使用GeneSearch(r)试验评估为连续值。此外，这些标志物的周期时间将独立于其他因素预测非sln转移。我们假设可以使用术前和术中可用的临床病理因素创建有效的临床预测规则，包括从GeneSearch(r)检测中获得的数据，该数据将识别非sln转移可能性为5%的患者亚群。预计GeneSearch(r)检测数据的增加将显著提高预测这组患者的能力，而不是单独使用临床病理因素。建立和验证一个强大的临床预测规则，结合新的分子数据来识别非sln转移低风险患者亚组，将填补我们目前知识的重大空白，并将减少ALND的不必要发病率，帮助满足NCI的挑战，即到2015年消除癌症带来的痛苦。公共卫生相关性：高达80%的乳腺癌患者在其第一个引流淋巴结（前哨淋巴结）有轻微疾病，在其腋窝没有进一步的疾病。然而，目前还没有准确的方法来预测哪些患者不会有残留的疾病，因此所有前哨淋巴结有肿瘤沉积物= 0.2 mm的患者都将切除剩余的淋巴结——这一手术与相当高的发病率有关。我们试图建立并验证一种临床预测规则，结合新的分子数据，以确定非sln转移风险低的患者亚组，从而填补我们目前知识的重大空白，并减少乳腺癌患者不必要的发病率。

DESCRIPTION (provided by applicant): Sentinel lymph node (SLN) biopsy is a minimally invasive means of accurately staging the axilla in breast cancer patients. While finding negative SLNs avoids axillary lymph node dissection (ALND), the finding of micrometastasis in the SLN mandates a completion ALND. Up to 80% of all SLN-positive patients will have no further disease in the axilla. There is a significant gap in our current knowledge in predicting which SLN- positive patients have a low likelihood of non-SLN metastasis and can therefore avoid ALND. A novel intra- operative RT-PCR based assay (GeneSearch(r)) for detecting SLN micrometastases has been shown to correlate with SLN tumor burden in breast cancer patients. The potential utility of this new technology in predicting which patients will not have further disease in their non-SLNs has not been investigated. Developing a robust clinical prediction rule for non-SLN status using these data will fill this gap in our knowledge and provide timely conclusions, which will have a direct impact on patient care. In order to achieve this over-arching objective, we propose the following specific aims: (1) To determine the impact of quantitative cycle time values of the GeneSearch(r) assay on non-SLN status using both bivariate and multivariable logistic regression analyses, and (2) to create and validate (using bootstrap techniques) a clinical prediction rule to predict non-SLN status in SLN-positive patients using preoperatively- and intraoperatively-available clinicopathologic factors, including data obtained from the GeneSearch(r) assay. We hypothesize that the quantitative RT-PCR cycle time for both mammoglobin (MMG) and cytokeratin 19 (CK19), as evaluated as continuous values using the GeneSearch(r) assay, will be significantly associated with non-SLN metastasis on bivariate analyses. In addition, the cycle time for these markers will predict non-SLN metastasis independent of other factors. We hypothesize that a valid clinical prediction rule can be created using preoperatively- and intraoperatively-available clinicopathologic factors, including data obtained from the GeneSearch(r) assay that will identify a subset of patients in whom the likelihood of non-SLN metastasis is = 5%. It is anticipated that the addition of data from the GeneSearch(r) assay will significantly improve the ability to predict this group of patients over the use of clinicopathologic factors alone. The creation and validation of a robust clinical prediction rule that incorporates novel molecular data to identify a subgroup of patients at low risk of having non-SLN metastasis will fill a significant gap in our current knowledge and will reduce unnecessary morbidity of ALND, helping to meet the NCI's Challenge of eliminating suffering due to cancer by 2015. PUBLIC HEALTH RELEVANCE: Up to 80% of breast cancer patients who have minimal disease in their first draining (sentinel) lymph nodes will have no further disease in their axilla. However, currently there is no accurate means of predicting which patients will not have residual disease, and therefore all patients who have a tumor deposit = 0.2 mm in their sentinel lymph nodes will have the remaining lymph nodes removed -- a procedure that is associated with considerable morbidity. We seek to create and validate a clinical prediction rule, incorporating novel molecular data, to identify a subgroup of patients at low risk of having non-SLN metastasis, thereby filling a significant gap in our current knowledge and reducing unnecessary morbidity for breast cancer patients.