Lender Yan, Givton Ofer, Bornshten Ruth, Azar Meitar, Moscona Roy, Yarden Yosef, Rubin Eitan
Shraga Segal Department of Microbiology, Immunology & Genetics, Ben-Gurion University in the Negev, Beer Sheba 8410501, Israel.
The Faculty of Medical & Health Sciences, Tel Aviv University, Tel Aviv 6997801, Israel.
Biomedicines. 2025 Apr 2;13(4):849. doi: 10.3390/biomedicines13040849.
Lung adenocarcinoma, the most prevalent type of non-small cell lung cancer, consists of two driver mutations in KRAS or EGFR. These mutations are generally mutually exclusive and biologically and clinically different. In this study, we aimed to test if lung adenocarcinoma tumors could be separated by their immune profiles using an unsupervised machine learning method. The underlying assumption was that differences in the immune response to tumors are characteristic of tumor subtypes. RNA-seq data were projected into inferred immune profiles. Unsupervised learning was used to divide the lung adenocarcinoma population based on their projected immune profiles. The patient population was divided into three subgroups, one of which appeared to contain mostly EGFR patients. The tumors in the different clusters significantly differed in their expression of some of their known immune checkpoints (TIGIT, PD-1/PD-L1, and CTLA4). We argue that EGFR mutations in each subgroup are immunologically different, which implies a distinct tumor microenvironment and might relate to the relatively high resistance of EGFR-positive tumors to immune checkpoint inhibitors. However, we cannot make the same claim about KRAS mutations.
肺腺癌是最常见的非小细胞肺癌类型，由KRAS或EGFR中的两种驱动基因突变组成。这些突变通常相互排斥，在生物学和临床上也有所不同。在本研究中，我们旨在测试是否可以使用无监督机器学习方法根据免疫特征将肺腺癌肿瘤分开。潜在的假设是对肿瘤的免疫反应差异是肿瘤亚型的特征。RNA测序数据被投射到推断的免疫特征中。使用无监督学习根据投射的免疫特征将肺腺癌群体进行划分。患者群体被分为三个亚组，其中一个亚组似乎主要包含EGFR患者。不同簇中的肿瘤在一些已知免疫检查点（TIGIT、PD-1/PD-L1和CTLA4）的表达上有显著差异。我们认为每个亚组中的EGFR突变在免疫方面是不同的，这意味着存在独特的肿瘤微环境，并且可能与EGFR阳性肿瘤对免疫检查点抑制剂的相对较高耐药性有关。然而，对于KRAS突变，我们不能得出同样的结论。
