Tumors are comprised of both cancer cells and surrounding stromal components. As an essential part of the tumor microenvironment, the tumor stroma is highly dynamic, heterogeneous and commonly tumor-type specific, and it mainly includes noncellular compositions such as the extracellular matrix and the unique cancer-associated vascular system as well as a wide variety of cellular components including activated cancer-associated fibroblasts, mesenchymal stromal cells, pericytes. All these elements operate with each other in a coordinated fashion and collectively promote cancer initiation, progression, metastasis and therapeutic resistance. Over the past few decades, numerous studies have been conducted to study the interaction and crosstalk between stromal components and neoplastic cells. Meanwhile, we have also witnessed an exponential increase in the investigation and recognition of the critical roles of tumor stroma in solid tumors. A series of clinical trials targeting the tumor stroma have been launched continually. In this review, we introduce and discuss current advances in the understanding of various stromal elements and their roles in cancers. We also elaborate on potential novel approaches for tumor-stroma-based therapeutic targeting, with the aim to promote the leap from bench to bedside.
The aim of this study was to assess the diagnostic ability of radiomics combined with multiple machine learning algorithms to differentiate pancreatic ductal adenocarcinoma (PDAC) from pancreatic neuroendocrine tumor (pNET). This retrospective study included a total of 238 patients diagnosed with PDAC or pNET. Using specialized software, radiologists manually mapped regions of interest (ROIs) from computed tomography images and automatically extracted radiomics features. A total of 45 discriminative models were built by five selection algorithms and nine classification algorithms. The performances of the discriminative models were assessed by sensitivity, specificity and the area under receiver operating characteristic curve (AUC) in the training and validation datasets. Using the combination of Gradient Boosting Decision Tree (GBDT) as the selection algorithm and Random Forest (RF) as the classification algorithm, the optimal diagnostic ability with the highest AUC was presented in the training and validation datasets. The sensitivity, specificity and AUC of the model were 0.804, 0.973 and 0.971 in the training dataset and 0.742, 0.934 and 0.930 in the validation dataset, respectively. The combination of radiomics and multiple machine learning algorithms showed the potential ability to discriminate PDAC from pNET. We suggest that multi-algorithm modeling should be considered for similar studies in the future rather than using a single algorithm empirically.