Integrative analysis to select cancer candidate biomarkers to targeted validation

Abstract

Targeted proteomics has flourished as the method of choice for prospecting for and validating potential candidate biomarkers in many diseases. However, challenges still remain due to the lack of standardized routines that can prioritize a limited number of proteins to be further validated in human samples. To help researchers identify candidate biomarkers that best characterize their samples under study, a well-designed integrative analysis pipeline, comprising MS-based discovery, feature selection methods, clustering techniques, bioinformatic analyses and targeted approaches was performed using discovery-based proteomic data from the secretomes of three classes of human cell lines (carcinoma, melanoma and non-cancerous). Three feature selection algorithms, namely, Beta-binomial, Nearest Shrunken Centroids (NSC), and Support Vector Machine-Recursive Features Elimination (SVM-RFE), indicated a panel of 137 candidate biomarkers for carcinoma and 271 for melanoma, which were differentially abundant between the tumor classes. We further tested the strength of the pipeline in selecting candidate biomarkers by immunoblotting, human tissue microarrays, label-free targeted MS and functional experiments. In conclusion, the proposed integrative analysis was able to pre-qualify and prioritize candidate biomarkers from discovery-based proteomics to targeted MS.

Mass Spectrometry for Cancer Biomarkers

“…The introduction of the Orbitrap mass analyzer allowed routine HRMS to be also used for biomarker discovery studies. Using an LTQ-Orbitrap, Kawahara et al defined the phenotypic differences in the secretome of melanoma, carcinoma, and noncancerous cells [43]. It is interesting to note that the top two candidates from the carcinoma secretome were validated in the saliva of patients using pseudo-selected reaction monitoring (pseudo-SRM) experiments on the same conventional hybrid instrument, usually dedicated only to the discovery approach, which subjects the flexibility and the impact of this platform.…”

Proteomic Findings in Melanoma

“…Although it is possible to perform quantitative gene profiling using formalin-fixed-paraffin-embedded (FFPE) tissues, but the RNA derived from such archived tissues are often degraded and fragmented which sometimes results in “FFPE bias” [3]. Recent work by Kawahara et al [4], showed the potential use of discovery-based proteomics data obtained from the secretion of melanoma cell lines along with other human cell lines. They used a MS-based approach followed by clustering techniques and bioinformatics to determine proteins that were differentially expressed in the melanoma cell lines (A2058 and SK-MEL-28) compared to noncancerous cell lines (HaCaT and HEK293).…”

Computational methods in Biology: cancer biomarkers, protein networks and lateral gene transfer

“…Then, we demonstrate how computational methods select proteins as potential candidate biomarkers by reporting three studies on biomarkers published during this doctorate: KAWAHARA, R.; MEIRELLES, G. V; HEBERLE, H.; et al Integrative analysis to select cancer candidate biomarkers to targeted validation. Oncotarget, v. 6, n. 41, p. 43635-43652, 2015. (KAWAHARA et al, 2015 HEBERLE, H.; MEIRELLES, G.; DA SILVA, F. R.; TELLES, G. P.; MINGHIM, R. InteractiVenn: A web-based tool for the analysis of sets through Venn diagrams.…”

“…(KAWAHARA et al, 2015 HEBERLE, H.; MEIRELLES, G.; DA SILVA, F. R.; TELLES, G. P.; MINGHIM, R. InteractiVenn: A web-based tool for the analysis of sets through Venn diagrams. BMC Bioinformatics, v. 16, n. 1, 2015. (HEBERLE et al, 2015 CARNIELLI, C. M.; MACEDO, C. C. S.; DE ROSSI, T.; et al Combining discovery and targeted proteomics reveals a prognostic signature in oral cancer.…”

“…Entre os resultados estão a avaliação da estabilidade dos métodos de ranqueamento e assinaturas aplicados aos dados proteômicos de descoberta, uma nova abordagem para selecionar proteínas candidatas desde a descoberta até proteômica direcionada, listas de candidatos a biomarcadores para câncer oral e novas técnicas para a visualização de redes biológicas e supertrees filogenéticas. Our second work was published in the journal Oncotarget (KAWAHARA et al, 2015) in collaboration with LNBio, UNICAMP and other institutions, where we applied computational models to mine quantified proteins from healthy, carcinoma, and melanoma cancer cell samples. In this project, we integrated visualization, statistical, and machine learning methods to define a small list of differentially expressed proteins, establishing a pipeline to select candidate biomarkers to targeted validation.…”