A novel non-parametric method for uncertainty evaluation of correlation-based molecular signatures: its application on PAM50 algorithm.

Motivation: The PAM50 classifier is used to assign patients to the highest correlated breast cancer subtype irrespectively of the obtained value. Nonetheless, all subtype correlations are required to build the risk of recurrence (ROR) score, currently used in therapeutic decisions. Present subtype uncertainty estimations are not accurate, seldom considered or require a population-based approach for this context. Results: Here we present a novel single-subject non-parametric uncertainty estimation based on PAM50's gene label permutations. Simulations results ( n = 5228) showed that only 61% subjects can be reliably 'Assigned' to the PAM50 subtype, whereas 33% should be 'Not Assigned' (NA), leaving the rest to tight 'Ambiguous' correlations between subtypes. The NA subjects exclusion from the analysis improved survival subtype curves discrimination yielding a higher proportion of low and high ROR values. Conversely, all NA subjects showed similar survival behaviour regardless of the original PAM50 assignment. We propose to incorporate our PAM50 uncertainty estimation to support therapeutic decisions. Availability and Implementation: Source code can be found in 'pbcmc' R package at Bioconductor. Contacts: cristobalfresno@gmail.com or efernandez@bdmg.com.ar. Supplementary information: Supplementary data are available at Bioinformatics online.

The tumor microenvironment: characterization, redox considerations, and novel approaches for reactive oxygen species-targeted gene therapy.

The tumor microenvironment is a complex system that involves the interaction between malignant and neighbor stromal cells embedded in a mesh of extracellular matrix (ECM) components. Stromal cells (fibroblasts, endothelial, and inflammatory cells) are co-opted at different stages to help malignant cells invade the surrounding ECM and disseminate. Malignant cells have developed adaptive mechanisms to survive under the extreme conditions of the tumor microenvironment such as restricted oxygen supply (hypoxia), nutrient deprivation, and a prooxidant state among others. These conditions could be eventually used to target drugs that will be activated specifically in this microenvironment. Preclinical studies have shown that modulating cellular/tissue redox state by different gene therapy (GT) approaches was able to control tumor growth. In this review, we describe the most relevant features of the tumor microenvironment, addressing reactive oxygen species-generating sources that promote a prooxidative microenvironment inside the tumor mass. We describe different GT approaches that promote either a decreased or exacerbated prooxidative microenvironment, and those that make use of the differential levels of ROS between cancer and normal cells to achieve tumor growth inhibition.

Matricellular proteins and inflammatory cells: a task force to promote or defeat cancer?

In the last years it became clear that the tumor microenvironment plays a major role in neoplastic growth. Proteins secreted either by the malignant cells or by the tumor-associated stromal cells act as extracellular signal transductors, orchestrating tumor progression. Sentinel cells of the innate immune system patrol the different organs and have proven either to promote tumor growth or induce tumor suppression. In recent years, members of the matricellular family of extracellular proteins were shown to be involved in different aspects of the inflammatory response during tumor development, although in contradictory ways. In this review we discuss the evidence available up to date that relates matricellular proteins with the regulation of the inflammatory response and tumor progression.

Proteomic analysis identified N-cadherin, clusterin, and HSP27 as mediators of SPARC (secreted protein, acidic and rich in cysteines) activity in melanoma cells.

Secreted protein, acidic and rich in cysteines (SPARC) is a secreted protein associated with increased aggressiveness of different human cancer types. In order to identify downstream mediators of SPARC activity, we performed a 2-DE proteomic analysis of human melanoma cells following antisense-mediated downregulation of SPARC expression. We found 23/504 differential spots, 15 of which were identified by peptide fingerprinting analysis. Three of the differential proteins (N-cadherin (N-CAD), clusterin (CLU), and HSP27) were validated by immunoblotting, confirming decreased levels of N-CAD and CLU and increased amounts of HSP27 in conditioned media of cells with diminished SPARC expression. Furthermore, transient knock down of SPARC expression in melanoma cells following adenoviral-mediated transfer of antisense RNA confirmed these changes. We next developed two different RNAs against SPARC that were able to inhibit in vivo melanoma cell growth. Immunoblotting of the secreted fraction of RNAi-transfected melanoma cells confirmed that downregulation of SPARC expression promoted decreased levels of N-CAD and CLU and increased secretion of HSP27. Transient re-expression of SPARC in SPARC-downregulated cells reverted extracellular N-CAD, CLU, and HSP27 to levels similar to those in the control. These results constitute the first evidence that SPARC, N-CAD, CLU, and HSP27 converge in a unique molecular network in melanoma cells.