Martin Luther University Halle-Wittenberg, Institute of Pathology, Halle, Germany; Robert W. Franz Cancer Research Center, Earle A. Chiles Research Institute, Providence Cancer Center, Portland, OR, United States.
Robert W. Franz Cancer Research Center, Earle A. Chiles Research Institute, Providence Cancer Center, Portland, OR, United States; School of Medicine, Oregon Health & Science University, Portland, OR, United States.
Curr Opin Immunol. 2017 Apr;45:60-72. doi: 10.1016/j.coi.2017.01.005. Epub 2017 Feb 20.
Immune cell infiltration is common to many tumors and has been recognized by pathologists for more than 100 years. The application of digital imaging and objective assessment software allowed a concise determination of the type and quantity of immune cells and their location relative to the tumor and, in the case of colon cancer, characterized overall survival better than AJCC TNM staging. Subsequently, expression of PD-L1, by 50% or more tumor cells, identified NSCLC patients with double the response rate to anti-PD-1. Soon, automated staining methods will improve reproducibility of multiplex staining and allow for CLIA standards so that multiplex staining can be used to make clinical decisions. Ultimately, machine-learning algorithms will help interpret data from tissue images and lead to improved delivery of precision medicine.
免疫细胞浸润在许多肿瘤中很常见，这一现象已被病理学家认识了 100 多年。数字成像和客观评估软件的应用，可以简洁地确定免疫细胞的类型和数量，以及它们相对于肿瘤的位置，在结肠癌的情况下，其对总生存期的判断优于 AJCC TNM 分期。随后，肿瘤细胞表达 PD-L1 达到 50%或以上，可识别出对 PD-1 抗体的反应率增加一倍的非小细胞肺癌患者。不久，自动化染色方法将提高多重染色的可重复性，并达到 CLIA 标准，从而使多重染色可用于做出临床决策。最终，机器学习算法将有助于解释组织图像数据，并改善精准医疗的实施。
