In recent years, in-depth studies have shown that extracellular matrix stiffness plays an important role in cell growth, proliferation, migration, immunity, malignant transformation, and apoptosis. Most of these processes entail metabolic reprogramming of cells. However, the exact mechanism through which extracellular matrix stiffness leads to metabolic reprogramming remains unclear. Insights regarding the relationship between extracellular matrix stiffness and metabolism could help unravel novel therapeutic targets and guide development of clinical approaches against a myriad of diseases. This review provides an overview of different pathways of extracellular matrix stiffness involved in regulating glucose, lipid and amino acid metabolism.

Background: The incidence of colorectal cancer in patients younger than 50 years has been increasing in recent years. Objective: Develop and validate prognostic nomograms predicting overall survival (OS) and cancer-specific survival (CSS) for early-onset locally advanced colon cancer (EOLACC) based on the Surveillance, Epidemiology, and End Results (SEER) database. Results: The entire cohort comprised 13,755 patients with EOLACC. The nomogram predicting OS for EOLACC displayed that T stage contributed the most to prognosis, followed by N stage, regional nodes examined (RNE) and surgery. The nomogram predicting CSS for EOLACC demonstrated similar results. Various methods identified the discriminating superiority of the nomograms. X-tile software was used to classify patients into high-risk, medium-risk, and low-risk according to the risk score of the nomograms. The risk stratification effectively avoided the survival paradox. Conclusions: We established and validated nomograms for predicting OS and CSS based on a national cohort of almost 13,000 EOLACC patients. The nomograms could effectively solve the issue of survival paradox of the AJCC staging system and be an excellent tool to integrate the clinical characteristics to guide the therapeutic choice for EOLACC patients. Methods: Nomograms were constructed based on the SEER database and the Cox regression model.

Pancreatic ductal adenocarcinoma (PDAC) is rich in dense fibrotic stroma that are composed of extracellular matrix (ECM) proteins. A disruption of the balance between ECM synthesis and secretion and the altered expression of matrix remodeling enzymes lead to abnormal ECM dynamics in PDAC. This pathological ECM promotes cancer growth, survival, invasion, and alters the behavior of fibroblasts and immune cells leading to metastasis formation and chemotherapy resistance, which contribute to the high lethality of PDAC. Additionally, recent evidence highlights that ECM, as a major structural component of the tumor microenvironment, is a highly dynamic structure in which ECM proteins establish a physical and biochemical niche for cancer stem cells (CSCs). CSCs are characterized by self-renewal, tumor initiation, and resistance to chemotherapeutics. In this review, we will discuss the effects of the ECM on tumor biological behavior and its molecular impact on the fundamental signaling pathways in PDAC. We will also provide an overview of how the different ECM components are able to modulate CSCs properties and finally discuss the current and ongoing therapeutic strategies targeting the ECM. Given the many challenges facing current targeted therapies for PDAC, a better understanding of molecular events involving the interplay of ECM and CSC will be key in identifying more effective therapeutic strategies to eliminate CSCs and ultimately to improve survival in patients that are suffering from this deadly disease.

Generating natural sentences from Knowledge Graph (KG) triples, known as Data-To-Text Generation, is a task with many datasets for which numerous complex systems have been developed. However, no prior work has attempted to perform this generation at scale by converting an entire KG into natural text. In this paper, we verbalize the entire Wikidata KG, and create a KG-Text aligned corpus in the training process 1 . We discuss the challenges in verbalizing an entire KG versus verbalizing smaller datasets. We further show that verbalizing an entire KG can be used to integrate structured and natural language data. In contrast to the many architectures that have been developed to integrate the structural differences between these two sources, our approach converts the KG into the same format as natural text allowing it to be seamlessly plugged into existing natural language systems. We evaluate this approach by augmenting the retrieval corpus in REALM and showing improvements, both on the LAMA knowledge probe and open domain QA.