Prediction of Neoadjuvant Chemo Pathological Response Using MRI Markers

描述(申请人提供)：新辅助或术前化疗正在成为乳腺癌治疗和管理的重要组成部分。同时，重要的是制定可靠的监测方法来评估反应。动态增强磁共振成像(DCE-MRI)已被证明是预测肿瘤残留反应的最准确的成像手段。在这项研究中，我们将应用一系列的DCE-MRI研究来评估治疗前的病变特征，以及在随访MRI研究中确定的早期反应模式，以期准确预测病理完全缓解(PCR)，这是接受新辅助化疗的患者最相关的预后因素。 在我们的机构，我们有一个积极的正在进行的新辅助化疗方案。在以前的患者队列中，我们已经证明MRI非常有助于为及时调整方案提供反应监测信息。基于这些结果，我们已经确定了在当前研究中使用的治疗方案。患者将接受2个周期的AC治疗(阿霉素和环磷酰胺)，然后接受4个周期的紫杉烷方案(NAB-紫杉醇加卡铂，曲妥珠单抗用于HER2阳性癌症，贝伐单抗用于HER2阴性癌症)。在目前的提案中，我们将进一步发挥MRI的作用，调查治疗前的肿瘤特征，包括形态、纹理和血管参数，以及代谢信息，是否可以用来预测患者是否会实现PCR。我们获得的信息将揭示这两组人之间的差异，这些信息可以用来建立一个预测实现聚合酶链式反应的可能性的模型。如果希望患者表现出良好的反应，可以按原样给出方案。如果预计患者在到达PCR时不会表现出良好的反应，则应考虑修改方案。应答监测数据还可以提供用于修改方案以达到改进的疗效的信息。2个周期AC后的早期反应，然后在添加1个周期的紫杉烷后进行评估。总体而言，我们将评估使用治疗前MRI特征参数和治疗过程中早期肿瘤反应模式的准确性，以区分实现和未实现聚合酶链式反应的患者。与单独使用任何一组数据相比，假设联合使用治疗前病变特征和早期肿瘤反应模式可以实现更高的PCR值预测精度。

DESCRIPTION (provided by applicant): Neoadjuvant or preoperative chemotherapy is becoming an important part in breast-cancer treatment and management. In parallel it is important to develop reliable monitoring methods to evaluate the response. Dynamic contrast enhanced MRI (DCE-MRI) has been proven as the most accurate imaging modality to predict residual tumor responses. In this study we will apply serial DCE-MRI studies to evaluate the pre-treatment lesion characteristic features, along with the early response patterns determined in follow-up MRI studies to aim for accurate prediction of pathologic complete remission (pCR), which is the most relevant prognostic factor in patients receiving neoadjuvant chemotherapy. At our institution we have an active on-going neoadjuvant chemotherapy protocol. In previous patient cohorts we have demonstrated that MRI is very helpful to provide response-monitoring information for timely adjustment of regimens. Based on these results we have determined the treatment protocols to be used in the current study. Patients will receive 2 cycles AC treatment (doxorubicin and cyclophosphamide) then followed by 4 cycles Taxane regimen (Nab-paclitaxel with Carboplatin, with trastuzumab for HER2-positive cancer and bevacizumab for HER2-negative cancer). In the current proposal we will take the role of MRI one step further, to investigate whether the pre-treatment tumor characteristics, including morphology, texture, and vascular parameters, as well as metabolic information, can be used to predict whether a patient will achieve pCR. The information that we obtain will reveal the differences between these two groups, and that can be used to build a model to predict the likelihood of achieving pCR. If a patient is expected to show a good response, the protocol can be given as it is. If a patient is expected not to show a good response to reach pCR, a modified protocol should be considered. The response monitoring data may also provide information for modifying the regimen to reach an improved efficacy. The early response after 2 cycles AC, then after addition of 1 cycle of taxane will be evaluated. Overall we will evaluate the accuracy of using pre-treatment MRI characteristic parameters, and early tumor response patterns during the treatment course, to differentiate between patients who achieve pCR vs. those who do not. It is hypothesized that a higher accuracy in prediction of pCR can be achieved using combined pre-treatment lesion characteristics and early tumor response patterns compared to that using either set of data alone.