基于基线CT放射组学的肺亚实性结节生长模式预测及分子病理基础研究

The widespread availability of high resolution computed tomography (CT) and the promotion of low-dose chest CT screening programs have increased the detection of pulmonary subsolid nodules (SSN). The persistent SSN has a high likelihood of representing part of the pathologic spectrum of lung adenocarcinoma. Nodular growth or not is the important basis to make follow-up plan and choose surgical timing. CT review is the only way to monitor SSN growth patterns at present. Due to individual differences in SSN growth, there are problems with over-examination and over-treatment in clinical practice. How to predict SSN growth patterns through baseline data is the key to reducing over-examination and over-treatment. Based on the preliminary study of the subtypes prediction of lung adenocarcinoma by CT quantitative analysis, this study proposed to predict the growth pattern of SSN. The research group intends to analyze the radiomics data, regular CT features and clinical features of the baseline examination of persistent SSN patients to construct the predictive model of SSN growth pattern. Moreover, this study also analyzes the differences in the expression of lung adenocarcinoma-driven genes between increased SSN and stable SSN, and the correlation between tumor doubling time of SSN and the expression of lung adenocarcinoma-driven genes, which reveal the role of driving genes in SSN growth. The completion of this study will provides a scientific basis for the development of individualized follow-up plan and the choice of the surgical timing of SSN to reduce over-examination and over-treatment in SSN patients, and the theoretical basis for further elucidating the evolution factors of lung adenocarcinoma.

高分辨率CT的应用及低剂量CT筛查的推广使肺亚实性结节（Subsolid Nodule,SSN）的检出越来越多，持续存在的SSN多为肺腺癌发展过程中的某一亚型，其生长与否决定了手术时机。CT复查是监测SSN生长的唯一方式，SSN生长过程中的个体化差异导致了临床中过度检查和过度治疗的问题。如何通过基线检查预测SSN生长模式是减少该问题的关键。本课题组在前期通过CT定量分析技术横向区分肺腺癌各生长阶段的基础上，拟纵向预测SSN的生长模式。通过对SSN基线CT的放射组学、常规CT及临床特征的综合分析，构建SSN生长模式的预测模型；并从分子病理水平分析不同生长模式SSN驱动基因表达的差异及其与SSN倍增时间的关系，揭示驱动基因在SSN生长过程中的作用。本研究的完成，将为SSN患者个体化随诊方案的制定和手术时机的选择提供科学依据并有望减少过度检查和过度治疗，同时为阐明肺腺癌的演进因素提供理论基础。

随着胸部低剂量CT筛查的推广，肺亚实性结节（Subsolid Nodule, SSN）的检出越来越多，持续存在的SSN多为肺腺癌发展过程中的某一亚型，其生长与否决定了随诊方案和手术时机。CT复查是目前监测SSN生长的唯一方式，由于SSN生长过程中个体化的差异，临床存在着过度检查和过度治疗的问题。如何通过基线检查资料预测SSN生长模式是减少该问题的关键。本课题组在前期通过CT定量分析技术横向区分肺腺癌各生长阶段研究的基础上，对SSN基线CT的放射组学、常规CT及临床特征进行了综合分析，构建了SSN生长模式的预测模型，从而达到纵向预测SSN的生长模式的目的；并从分子病理水平分析不同生长模式SSN驱动基因表达的差异及其与SSN倍增时间的关系，揭示驱动基因在SSN生长过程中的作用。本研究的完成，为SSN患者个体化随诊方案的制定和手术时机的选择提供了科学依据并有望减少过度检查和过度治疗，同时为进一步阐明肺腺癌的演进因素提供理论基础。在课题实施过程中，还衍生出了以下研究：（1）表现为多灶磨玻璃结节的特殊类型肺腺癌患者，在术前行PET-CT和脑MRI检查的必要性研究；（2）影响影像组学特征稳定性的相关参数研究。这两项研究的完成，为CT图像表现为多灶磨玻璃结节的特殊类型肺腺癌患者术前检查的选择提供了依据，避免过度检查和医疗资源的浪费；在影像组学研究过程中，病例选择上尽量避免影响因素对最终结果的干扰，以得到高普适性和高临床应用价值的研究结果。