基于CT影像组学量化肿瘤免疫微环境及预测早期非小细胞肺癌术后复发风险的研究

Early stage non-small cell lung cancer (NSCLC) accounts for approximately one quarter of all lung cancers. Although surgical resection is the preferred treatment option for the early stage NSCLC patients, nevertheless 30-55%of resected early stage NSCLC patients will have disease recurrence. Many studies confirm that early stage NSCLC patients who are at high risk for disease recurrence may show benefit of long-term survival from adjuvant chemotherapy after surgery. However, there remains an unmet need for the accurate identification of resected early stage NSCLC patients who are at high risk for disease recurrence. Our previous research and other researchers’ results indicated that tumour immune microenvironment is an independent predictor of prognosis, and radiomics, through mining of data from images, can help for the characterization of tumour heterogeneity. Our study hypothesizes that the tumour immune microenvironment can be quantified comprehensively using radiomics analysis, which offers new perspective for accurate prediction of disease recurrence of resected early stage NSCLC. This project is to: (1) based on the immunohistochemical detection of pathological tissues of NSCLC after surgery, the score of immune microenvironment was established to quantify the immune status of tumours; (2) extract and select the key radiomics features highly related to the status of tumour immune based on preoperative CT images and develop an immune-based radiomics signature; (3) integrate the immune-based radiomics signature and clinicopathological variables to build the prognostic prediction model for predicting risk of disease recurrence of resected early stage NSCLC, and the prediction model will be validate. The result of this project will develop an immune-based radiomics prediction model facilitating precise quantification of the risk of disease recurrence of resected early stage NSCLC, which enables the identification of individuals who are at risk of disease recurrence and aids in clinical decision-making in precise medicine for the patient management.

早期非小细胞肺癌（NSCLC）约占NSCLC总数的25%，首选手术切除治疗，但术后复发率高达30-55%；研究证实高复发风险者可从术后辅助化疗获益，但目前缺乏量化复发风险的方法。申请人及其他团队研究证实，肿瘤免疫微环境是预后的独立预测指标，且通过影像组学挖掘肿瘤影像特征可评估肿瘤异质性，我们研究假设基于影像组学可全面量化肿瘤免疫微环境，为早期NSCLC术后复发风险精准预测提供新思路。本项目拟：（1）基于NSCLC术后病理组织免疫组化检测，建立免疫微环境评分，量化肿瘤免疫状态；（2）基于影像组学提取并筛选与肿瘤免疫状态相关的CT影像组学特征，构建免疫相关影像组学标签；（3）融合免疫相关影像组学标签和临床病理信息，构建复发风险预测模型并验证。本研究结果，将构建量化早期NSCLC免疫微环境的免疫相关影像组学模型，实现复发风险精准量化预测，指导临床筛选高复发风险个体，辅助临床决策。

针对早期非小细胞肺癌（NSCLC）术后复发风险难以精准预测的临床挑战，本项目利用影像组学方法，提取可量化早期NSCLC肿瘤免疫微环境的影像组学特征，融合患者的临床病理信息构建个体化预测模型，实现早期NSCLC复发风险的精准预后预测研究。包括：（1）开发免疫组化染色的数字病理图像分析方法，实现对肿瘤免疫微环境风险评分的精准量化；（2）开发完善影像组学特征提取技术及方法，充分挖掘肿瘤影像图像的高通量特征数据，为NSCLC复发风险分层提供影像依据；（3）成功建立可全面量化NSCLC患者的肿瘤免疫生态系统多样性的影像学标签，构建预后预测模型，可实现早期NSCLC患者的肿瘤免疫状态及预后风险的个体化预测；（4）积极拓展本项目开发的技术方法在其他肿瘤领域的应用范畴。. 本项目共发表课题相关SCI文章8篇，合计影响因子42.036；获得授权国家发明专利3项；获得广东省科技进步奖1项。研究成果为NSCLC基于CT图像的影像组学研究提供了分析范例，探讨了影像组学在NSCLC的肿瘤免疫微环境量化及预后预测方面的临床应用价值，为早期NSCLC术后复发风险分层精准量化提供影像依据。本项目研发的基于影像组学分析方法技术在其他肿瘤领域的积极拓展，证明了相关技术方法跨肿瘤领域的推广应用价值。在本项目资助下，项目组积极开发了针对医学图像的信息数据挖掘技术流程，可充分挖掘蕴含于肿瘤影像图像及数字病理图像中的高通量图像特征，构建预后预测模型，辅助指导临床决策，推动了医学影像数据挖掘技术的发展及其在临床实践中的应用，具有重要的临床应用价值。