Multiple Analytical Platforms Research Articles

Investigating the impacts of solid phase extraction on dissolved organic matter optical signatures and the pairing with high‐resolution mass spectrometry data across a freshwater stream network

AbstractAdvancing our understanding of dissolved organic matter (DOM) chemistry in aquatic systems necessitates the integration of data streams from multiple analytical platforms. Some measurements require pretreatment with solid phase extraction (SPE), while others are performed directly on whole water samples. Evidence has suggested that SPE will be biased against select DOM fractions, leading to concerns over the ability to establish data linkages across platforms with variable needs for SPE pretreatment, such as those from optical measurements and those that provide high‐resolution molecular information. Here, we directly addressed this concern by assessing the impact of SPE on DOM optical properties through excitation–emission matrices with parallel factor analysis (PARAFAC) for 47 samples across a stream network within a single watershed reflective of variable DOM sources. PARAFAC data was further paired with molecular information obtained by Fourier transform ion cyclotron resonance mass spectrometry (FTICR‐MS). A comparison of PARAFAC models first revealed no systematic qualitative differences in major components between whole water DOM and DOM isolated by SPE (SPE‐DOM); however, quantitative biases against select components were observed. Further linkages with FTICR‐MS data revealed that the molecular fingerprint associated with each PARAFAC component was consistent between the whole water DOM and SPE‐DOM. Our results suggest that bulk scale linkages across these analytical platforms could be inferred irrespective of the observed quantitative biases resulting from SPE for samples within this example watershed. This work represents a key step toward the systematic evaluation of linkages between optical and high‐resolution mass spectrometry datasets in freshwater lotic environments.

A biomarker assay validation approach tailored to thecontext of use and bioanalytical platform.

It is widely acknowledged by the bioanalytical and biomarker community that biomarker assay validations should be fit-for-purpose depending on the context of use. The challenge is how to consistently apply these principles in teams responsible for measuring a disparate array of biomarkers, often on multiple analytical platforms, at various stages of the drug discovery and development pipeline and across diverse biology focus areas. To drive consistency, while maintaining the necessary flexibility to allow validations to be driven by scientific rationale and taking into consideration the context of use and associated biological and (pre)analytical factors, a framework applicable across biomarker assayswas developed. Herein the authorsshare their perspective to engage in the ongoing conversation around fit-for-purpose biomarker assay validation.

The GSK Bioanalytical Hub model in bioanalysis: maximizing the synergies of co-locating various platforms in support of regulated and nonregulated study end points.

The term "bioanalytical" encompasses a much greater breadth of analytical deliverables than ever before. Circulating drug concentration data arecomplemented by experimental evidence of drug in biophase, immunogenicity, target engagement and subsequent pathway modulation. Many bioanalytical assays bridge the traditional divide across discovery and development. Our approach is the Bioanalytical Hub model bringing together a wide breadth of bioanalytical support (GxP and non-GxP), multiple end points (pharmacokinetics, anti-drug antibodies and biomarkers)and analytical platforms (LC/MS, immunoassay, flow cytometry, genomics, immunohistochemistry) onto a common lab footprint. This maximizes instrument utilization, facilitates workforce agilityand enhances data interpretation capability while reducing the number of hand-offs as assays evolve from their origins as exploratory end points to fully characterized to support primary and secondary end points.

Abstract 2322: Multiplatform metabolomics studies of human cancers with NMR and mass spectrometry imaging

Abstract Unfortunately, at present, there is no single technique that possesses all the characteristics needed to be considered an ideal global metabolite profiling tool. Thus, the use of multiple analytical platforms, such as combining the strengths of Mass spectrometry (MS) and nuclear magnetic resonance spectroscopy (NMR), for metabolic profiling can maximize coverage and generate more global metabolomic profiles. In this study, we demonstrate the utilities of the combined NMR and MSI multiplatform in our metabolomics results on human prostate and lung cancers. Statistical data on the natural history of prostate cancer (PCa) show that >70% of patients diagnosed by PSA screening will likely experience indolent disease with little impact on well-being. For about 17% of newly PSA-diagnosed patients, however, aggressive PCa proliferation ensues, truncating life expectancy. At present, no reliable clinical test can differentiate between these two groups. Using HRMAS 1HNMR followed by quantitative histology, we showed statistically significant correlations between concentrations of Spm and the amount of histologically-benign epithelial (Hb Epi) prostatic cells and glands in human cancerous prostates. However, as above discussed that using HRMAS NMR alone we cannot prove that Spm was indeed generated or resided in the Hb Epi cells. Nevertheless, using MALDI MSI, we were able to locate Spm (m/z: 203.223 ± 0.001Da) onto Hb Epi, where spermine on the PCa lesions appeared below detection limits. From these maps, for the first time, we could visualize and confirm the differential localizations of Spm in prostates. This proof of Sym relationship to prostate pathologies and its proposed PCa inhibitory effects may support further studies that are critical in differentiating aggressive from indolent PCa for disease evaluations and patient personalized treatment strategies. To search for such screening metabolomics biomarkers in lung cancer, we used HRMAS NMR to analyze 93 pairs of human LuCa tissue and serum samples, and 29 healthy human sera. A number of potential metabolite candidates capable to differentiate LuCa characteristics were identified, including glutamate, lipids, alanine, glycerylphosphorylcholine, glutamine, phosphorylcholine, etc. This list can be further expanded by analyzing metabolite composition in the serum of cancer patients and control healthy subjects using LC-MS, which offers a dramatic increase in sensitivity compared to HRMAS NMR and, therefore, is better suited for the biomarker discovery. In addition to acquiring high-resolution mass data, the high data acquisition rate allows the fragment ion mass spectra (MS/MS) to be generated for the most abundant ionic species in each chromatographic peak. This feature allows specific classes of tumor-attenuated metabolites to be identified based on the presence of unique structurally diagnostic fragment ions in MS/MS spectra. Citation Format: Leo L. Cheng, Anya B. Zhong, Isabella H. Muti, Stephen J. Eyles, Richard W. Vachet, Sylwia A. Stopka, Kristen N. Sikora, Cedric E. Bobst, Jeffrey N. Agar, Mari A. Mino-Kenudson, Chin-Lee Wu, David C. Christiani, Igor A. Kaltashov, Nathalie Y. Agar. Multiplatform metabolomics studies of human cancers with NMR and mass spectrometry imaging [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2322.

Metabolic Effects of ALDH1L1 Knockout in Diethylnitrosamine‐Induced Model of Liver Carcinogenesis

Background Global gene expression profiling identified ALDH1L1 (aldehyde dehydrogenase 1 family member L1, or cytosolic 10-formyltetrahydrofolate dehydrogenase) as one of the most under expressed genes in HCCs compared to normal liver (1). ALDH1L1 is an abundant enzyme in folate metabolism representing about 1% of total soluble protein in hepatocytes. Expression of this protein is strongly and ubiquitously downregulated in human cancers through the promoter methylation. Aim The goal of this study was to analyze changes in metabolic profiles of mouse livers depending on ALDH1L1 status in a diethyl nitrosamine (DEN)-induced model of liver carcinogenesis. Methods Fifteen-days-old male Aldh1l1-/-mice and their wild-type littermate controls (Aldh1l1+/+) were injected intraperitoneally with DEN at the dose of 15 mg/kg body weight. Individual liver samples were collected after 10- and 20-weeks following treatment with DEN and subjected to metabolomics analysis in equivalent manner across multiple analytical platforms. Results Principal component analysis (PCA) showed that the differences in metabotypes between KO and WT mice were more prominent at 20 weeks then at 10 weeks. Several major cellular pathways contributed to the separation between Aldh1l1+/+and Aldh1l1-/- mice in their metabolic responses to DEN including pathways of folate, glycine, lipids and carbohydrates. Additionally, changes in redox-active compounds showed increased oxidative stress/decreased antioxidant capacity in KO livers. Conclusion Metabolic changes in KO livers support enhanced cell proliferation and reveal how the ALDH1L1 loss provides metabolic advantage for rapidly proliferating cells. References Tackels-Horne, D.; Goodman, M.D.; Williams, A.J.; Wilson, D.J.; Eskandari, T.; Vogt, L.M.; Boland, J.F.; Scherf, U.; Vockley, J.G. Identification of differentially expressed genes in hepatocellular carcinoma and metastatic liver tumors by oligonucleotide expression profiling. Cancer 2001, 92, 395-405.

Arsenic and Heavy Metals in Vietnamese Rice: Assessment of Human Exposure to These Elements through Rice Consumption.

In this work, twelve heavy metals and arsenic, As, Cd, Co, Cr, Cu, Fe, Hg, Mn, Ni, Pb, Se, and Zn, in a rice sample collected from some areas of Vietnam have been quantified and implemented by using multiple analytical platforms such as ICP-MS, AAS, and mercury analyser. Seventy rice samples collected from the Red River Delta and mining zone activity were analysed. Concentration of heavy metals and arsenic in rice was analysed after appropriated sample digestion using internal or external calibration curves. The mean concentration (mg kg−1 dried weight) of the analysed elements in rice samples decreased on the order of Mn (19.268) > Fe (13.624) > Zn (8.163) > Cu (3.138) > Ni (0.384) > Cr (0.296) > Co (0.279) > As (0.115) > Cd (0.111) > Pb (0.075) > Hg (0.007) > Se (<LOD). Mercury, a highly toxic element, has been only found in rice samples collected in the mining activity zone (frequency detection 14.5% of total samples). The experimental results indicated that the heavy metals and arsenic found in rice collected from mining activity zone were higher than those in rice harvested from normal cultivated areas like the Red River Delta. The heavy metals and arsenic content in Vietnamese rice samples were also compared with the concentration of heavy metals in other foreign rice samples in some recent publications. The estimated daily intake through rice consumption was calculated and compared with the level proposed by the Food and Agriculture Organization of the United Nations. The results indicated that the provisional daily intake of Cd was higher than the level proposed by FAO, while the intake of other heavy metals was in an acceptable range of CODEX standard.

Analysis of food samples made easy by microextraction technologies directly coupled to mass spectrometry.

Because of the complexity and diversity of food matrices, their chemical analysis often entails several analytical challenges to attain accurate and reliable results, especially for multiresidue analysis and ultratrace quantification. Nonetheless, microextraction technology, such as solid-phase microextraction (SPME), has revolutionized the concept of sample preparation for complex matrices because of its nonexhaustive, yet quantitative extraction approach and its amenability to coupling to multiple analytical platforms. In recent years, microextraction devices directly interfaced with mass spectrometry (MS) have redefined the analytical workflow by providing faster screening and quantitative methods for complex matrices. This review will discuss the latest developments in the field of food analysis by means of microextraction approaches directly coupled to MS. One key feature that differentiates SPME-MS approaches from other ambient MS techniques is the use of matrix compatible extraction phases that prevent biofouling, which could drastically affect the ionization process and are still capable of selective extraction of the targeted analytes from the food matrix. Furthermore, the review examines the most significant applications of SPME-MS for various ionization techniques such as direct analysis in real time, dielectric barrier desorption ionization, and some unique SPME geometries, for example, transmission mode SPME and coated blade spray, that facilitate the interface to MS instrumentation.

Identifying unknown metabolites using NMR-based metabolic profiling techniques.

Metabolic profiling of biological samples provides important insights into multiple physiological and pathological processes but is hindered by a lack of automated annotation and standardized methods for structure elucidation of candidate disease biomarkers. Here we describe a system for identifying molecular species derived from nuclear magnetic resonance (NMR) spectroscopy-based metabolic phenotyping studies, with detailed information on sample preparation, data acquisition and data modeling. We provide eight different modular workflows to be followed in a recommended sequential order according to their level of difficulty. This multi-platform system involves the use of statistical spectroscopic tools such as Statistical Total Correlation Spectroscopy (STOCSY), Subset Optimization by Reference Matching (STORM) and Resolution-Enhanced (RED)-STORM to identify other signals in the NMR spectra relating to the same molecule. It also uses two-dimensional NMR spectroscopic analysis, separation and pre-concentration techniques, multiple hyphenated analytical platforms and data extraction from existing databases. The complete system, using all eight workflows, would take up to a month, as it includes multi-dimensional NMR experiments that require prolonged experiment times. However, easier identification cases using fewer steps would take 2 or 3 days. This approach to biomarker discovery is efficient and cost-effective and offers increased chemical space coverage of the metabolome, resulting in faster and more accurate assignment of NMR-generated biomarkers arising from metabolic phenotyping studies. It requires a basic understanding of MATLAB to use the statistical spectroscopic tools and analytical skills to perform solid phase extraction (SPE), liquid chromatography (LC) fraction collection, LC-NMR-mass spectroscopy and one-dimensional and two-dimensional NMR experiments.

Performance of Norudia Glycated Albumin Assay on Multiple Analytical Platforms and Comparison to Lucica Assay.

The Norudia glycated albumin (GA) assay was evaluated for analytical performance and assay applicability on multiple analytical platforms. The evaluation included precision, linearity, reference interval, and comparison with Lucica GA assay. A multicenter study was conducted to compare the results of Norudia GA assay on five kinds of widely used automated clinical chemistry analyzers. Within-laboratory imprecisions for GA% presented 1.3 - 3.3% and 0.8 - 2.6% for low- and high-level control materials, respectively, on different analyzers. GA assay was linear from 20.0 to 680.0 µmol/L of GA. The claimed reference range (12 - 16 GA%) was verified. Norudia GA showed a good GA% correlation with Lucica GA (correlation coefficient 0.999). GA% from each analyzer showed good correlation with the consensus mean of the results of five analyzers (correlation coefficient 0.997 - 0.999). The Norudia GA assay can successfully be implemented in all the tested platforms, with good GA% correlation.

Using Multiple Analytical Platforms to Investigate the Androgen Depletion Effects on Fecal Metabolites in a Mouse Model of Systemic Lupus Erythematosus.

Systemic lupus erythematosus (SLE) is an autoimmune disease characterized by circulating autoantibodies that deposit in target organs (e.g., kidneys), resulting in chronic inflammation and eventual destruction of the organ. SLE is much more prevalent in females than males in both humans and spontaneous mouse models of lupus, such as NZBxNZW F1 (BWF1) mice. Depleting androgens by castration dramatically increases the susceptibility of BWF1 male to lupus. We compared fecal metabolite profiles of castratedBWF1 (androgen-depleted) male, intact (androgen-replete) male, and female mice. Four analytical platforms were employed to study the profiles of polar metabolites in mouse feces collected from adult BWF1 mice, and a total of 435 metabolites was identified. Of these, the abundance levels of 72 metabolites were significantly different between castrated and intact male groups, and 63 metabolites were different between female and male groups. Pathway analysis indicated that the pathway differences between castrated and intact male mice closely resembled the pathway differences between female and intact male mice, suggesting that low levels of androgens, whether due to depletion (castrated male) or endogenous (female), are associated with multiple fecal metabolomic alterations, which could potentially affect SLE progression. Our findings demonstrate that analyzing fecal metabolites using multiple analytical platforms holds great promise for detecting metabolomic alterations in complex disease model systems.

Common and distinct variation in data fusion of designed experimental data

IntroductionIntegrative analysis of multiple data sets can provide complementary information about the studied biological system. However, data fusion of multiple biological data sets can be complicated as data sets might contain different sources of variation due to underlying experimental factors. Therefore, taking the experimental design of data sets into account could be of importance in data fusion concept.ObjectivesIn the present work, we aim to incorporate the experimental design information in the integrative analysis of multiple designed data sets.MethodsHere we describe penalized exponential ANOVA simultaneous component analysis (PE-ASCA), a new method for integrative analysis of data sets from multiple compartments or analytical platforms with the same underlying experimental design.ResultsUsing two simulated cases, the result of simultaneous component analysis (SCA), penalized exponential simultaneous component analysis (P-ESCA) and ANOVA-simultaneous component analysis (ASCA) are compared with the proposed method. Furthermore, real metabolomics data obtained from NMR analysis of two different brains tissues (hypothalamus and midbrain) from the same piglets with an underlying experimental design is investigated by PE-ASCA.ConclusionsThis method provides an improved understanding of the common and distinct variation in response to different experimental factors.

Abstract 4953: Differential expression of ICOS on circulating immune cells and tumor infiltrating lymphocytes in patient with solid tumors

Abstract Background: The Inducible T-cell costimulator (ICOS/CD278) is upregulated during T cell activation and is an important regulator of the immune response against pathogens and cancer cells. When engaged by its ligand, the activation of ICOS signaling improves the survival of ICOS+ effector (TEff) and regulatory T cells (TReg). In addition, depending on the T cell subtypes expressing ICOS, activation of the signaling pathway induces the expression and secretion of either pro- (by TEff) or anti-inflammatory (by TReg) cytokines. ICOS has been reported to be upregulated in the tumor microenvironment (TME) of most tumor indications. However, the variability in ICOS expression levels on ICOS expressing T cells (TEff and TReg) and their relative abundance in different tumor types is still poorly understood. The aim of this study was to take advantage of multiple analytical platforms to assess the expression of ICOS on immune cells in the periphery and on tumor infiltrating lymphocytes (TILs) from cancers known to express high levels of ICOS, based on previous TCGA analyses. Methods: We analysed ICOS expression on peripheral blood mononuclear cell (PBMCs) and TILs by FACS, ChipcytometryTM and by single cell RNAseq. The analysis aimed to compare the expression of ICOS on different immune T cell subtypes. This comparison was performed on samples isolated from treatment naive non-small cell lung cancer (NSCLC) patients. In parallel, we established dual immunohistochemistry staining protocols supplemented by digital pathology to quantify and compare the expression of ICOS in the context of either FOXP3 or CD8 positive cells in NSCLC, gastric, esophageal, cervical, bladder and head and neck tumor samples. Results and conclusion: Using these approaches, we confirmed high ICOS expression in several solid tumor types. ICOS expression was also shown to be higher on immune cells in the TME compared to the periphery. Finally, ICOS expression was found to be more abundant on the surface of intratumoral TReg than on CD8+ T effector cells, confirming the data obtained in preclinical syngeneic tumor models (AACR2018 abstract #2792). Importantly, this work highlighted variability in the abundance and ratio of the different ICOS+ T cells subtypes (TEff and TReg) between different ICOSHigh tumors. Altogether, the variability in the intratumoral ratio of ICOS positive T cells observed in different tumor types could differentially affect their response to antibodies targeting ICOS. This work suggests that understanding the expression of ICOS in the context of different immune T cell subtypes may improve selection of patients with higher potential to respond to anti-ICOS therapies such as KY1044. Citation Format: Richard C. Sainson, Miha Kosmac, Rachael Kimber, Joana Carvalho, Gwenoline Bohris, Anil Thotakura, David Melvin, Matthew McCourt. Differential expression of ICOS on circulating immune cells and tumor infiltrating lymphocytes in patient with solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4953.

An Integrative Multi-Omics Workflow to Address Multifactorial Toxicology Experiments.

Toxicology studies can take advantage of omics approaches to better understand the phenomena underlying the phenotypic alterations induced by different types of exposure to certain toxicants. Nevertheless, in order to analyse the data generated from multifactorial omics studies, dedicated data analysis tools are needed. In this work, we propose a new workflow comprising both factor deconvolution and data integration from multiple analytical platforms. As a case study, 3D neural cell cultures were exposed to trimethyltin (TMT) and the relevance of the culture maturation state, the exposure duration, as well as the TMT concentration were simultaneously studied using a metabolomic approach combining four complementary analytical techniques (reversed-phase LC and hydrophilic interaction LC, hyphenated to mass spectrometry in positive and negative ionization modes). The ANOVA multiblock OPLS (AMOPLS) method allowed us to decompose and quantify the contribution of the different experimental factors on the outcome of the TMT exposure. Results showed that the most important contribution to the overall metabolic variability came from the maturation state and treatment duration. Even though the contribution of TMT effects represented the smallest observed modulation among the three factors, it was highly statistically significant. The MetaCore™ pathway analysis tool revealed TMT-induced alterations in biosynthetic pathways and in neuronal differentiation and signaling processes, with a predominant deleterious effect on GABAergic and glutamatergic neurons. This was confirmed by combining proteomic data, increasing the confidence on the mechanistic understanding of such a toxicant exposure.

Toward a better understanding of chronic kidney disease with complementary chromatographic methods hyphenated with mass spectrometry for improved polar metabolome coverage

The prevalence of chronic kidney disease (CKD) is increasing worldwide. New technical approaches are needed to improve early diagnosis, disease understanding and patient monitoring, and to evaluate new therapies. Metabolomics, as a prime candidate in the field of CKD research, aims to comprehensively analyze the metabolic complexity of biological systems. An extensive analysis of the metabolites contained in biofluids is therefore needed, and the combination of data obtained from multiple analytical platforms constitutes a promising methodological approach. This study presents an original workflow based on complementary chromatographic conditions, reversed-phase and hydrophilic interaction chromatography hyphenated to mass spectrometry to improve the polar metabolome coverage coupled with a univocal metabolite annotation strategy enabling a rapid access to the biological interpretation. This multiplatform workflow was applied in a CKD cohort study to assess plasma metabolic profile modifications related to renal disease. Multivariate analysis of 278 endogenous annotated metabolites enabled patient stratification with respect to CKD stages and helped to generate new biological insights, while also confirming the relevance of tryptophan metabolism pathway in this condition.

The clinical applications of The Cancer Genome Atlas project for bladder cancer

ABSTRACTIntroduction: Knowledge of the molecular subtypes of bladder cancer enables powerful generalizations involving the distinctive biology and pathways driving different disease subsets.Areas covered: In this review, we summarize the findings of a number of published studies exploring the molecular landscape of bladder cancer by analysis of genomic data from The Cancer Genome Atlas (TCGA). TCGA project has provided a comprehensive data resource of 412 muscle-invasive bladder cancers as characterized by multiple molecular analytical platforms. These data have been and will continue to be utilized in numerous subsequent studies aimed at better understanding the molecular basis of bladder cancer. The catalog of DNA-level alterations can greatly inform personalized and precision medicine approaches. Molecular subtypes of bladder cancer include distinct ‘basal/squamous’ and ‘luminal’ subtypes, cancers with papillary histology, disease subsets with prominent leukocyte infiltration and immune checkpoint marker expression, and a ‘neuronal’ subtype lacking small cell or neuroendocrine histology. The gene-level alterations and subtypes as revealed by TCGA data are relevant from the standpoint of both basic biology and clinical trial studies.Expert commentary: Multiple studies analyzing TCGA muscle-invasive bladder cancer cases point to the existence of five major expression-based molecular subtypes of the disease, with these subtypes having therapeutic implications.

Metabolic discrimination of pine resins using multiple analytical platforms

Resins are one of the first sites of interaction between plants and biotic and abiotic factors. Despite their evident morphological and chemical differentiation from other plant organs, the detailed correlation between resins and biological or environmental factors is not yet clear. In this study, 1H nuclear magnetic resonance (NMR), gas chromatography coupled with mass spectrometry (GC-MS) and high-performance thin-layer chromatography (HPTLC)-based profiling techniques were applied to the metabolic characterisation of plant resins of different species and season of collection, using samples from five different species that were collected during early and late spring. The 1H NMR analysis confirmed the main metabolic groups in the resins to be terpenoids and further GC-MS analysis revealed a notable chemical variation between the species and collection periods. Abies grandis displayed a significant differentiation from the other species, showing a higher number of monoterpenes. The HPTLC-based profiling method hyphenated with multivariate data analysis (MVDA) also showed a clear separation confirming the GC-MS terpenoidal profiling results. Additionally, the unknown compounds were obtained by preparative TLC for identification. Based on the results of the three analytical platforms, it was concluded that the major difference in chemical composition of pine species was between species rather than the collection period. Nonetheless, the chemical profiles of resins from different species and collection periods can be well discriminated and correlated to mono- and sesquiterpenes in the case of species and diterpenes for the collection periods.

Comprehensive Molecular Characterization of Muscle-Invasive Bladder Cancer

We report a comprehensive analysis of 412 muscle-invasive bladder cancers characterized by multiple TCGA analytical platforms. Fifty-eight genes were significantly mutated, and the overall mutational load was associated with APOBEC-signature mutagenesis. Clustering by mutation signature identified a high-mutation subset with 75% 5-year survival. mRNA expression clustering refined prior clustering analyses and identified a poor-survival "neuronal" subtype in which the majority of tumors lacked small cell or neuroendocrine histology. Clustering by mRNA, long non-coding RNA (lncRNA), and miRNA expression converged to identify subsets with differential epithelial-mesenchymal transition status, carcinoma insitu scores, histologic features, and survival. Our analyses identified 5 expression subtypes that may stratify response to different treatments.

Combining NMR and LC/MS Using Backward Variable Elimination: Metabolomics Analysis of Colorectal Cancer, Polyps, and Healthy Controls.

Both nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS) play important roles in metabolomics. The complementary features of NMR and MS make their combination very attractive; however, currently the vast majority of metabolomics studies use either NMR or MS separately, and variable selection that combines NMR and MS for biomarker identification and statistical modeling is still not well developed. In this study focused on methodology, we developed a backward variable elimination partial least-squares discriminant analysis algorithm embedded with Monte Carlo cross validation (MCCV-BVE-PLSDA), to combine NMR and targeted liquid chromatography (LC)/MS data. Using the metabolomics analysis of serum for the detection of colorectal cancer (CRC) and polyps as an example, we demonstrate that variable selection is vitally important in combining NMR and MS data. The combined approach was better than using NMR or LC/MS data alone in providing significantly improved predictive accuracy in all the pairwise comparisons among CRC, polyps, and healthy controls. Using this approach, we selected a subset of metabolites responsible for the improved separation for each pairwise comparison, and we achieved a comprehensive profile of altered metabolite levels, including those in glycolysis, the TCA cycle, amino acid metabolism, and other pathways that were related to CRC and polyps. MCCV-BVE-PLSDA is straightforward, easy to implement, and highly useful for studying the contribution of each individual variable to multivariate statistical models. On the basis of these results, we recommend using an appropriate variable selection step, such as MCCV-BVE-PLSDA, when analyzing data from multiple analytical platforms to obtain improved statistical performance and a more accurate biological interpretation, especially for biomarker discovery. Importantly, the approach described here is relatively universal and can be easily expanded for combination with other analytical technologies.

Feature Selection Methods for Early Predictive Biomarker Discovery Using Untargeted Metabolomic Data.

Untargeted metabolomics is a powerful phenotyping tool for better understanding biological mechanisms involved in human pathology development and identifying early predictive biomarkers. This approach, based on multiple analytical platforms, such as mass spectrometry (MS), chemometrics and bioinformatics, generates massive and complex data that need appropriate analyses to extract the biologically meaningful information. Despite various tools available, it is still a challenge to handle such large and noisy datasets with limited number of individuals without risking overfitting. Moreover, when the objective is focused on the identification of early predictive markers of clinical outcome, few years before occurrence, it becomes essential to use the appropriate algorithms and workflow to be able to discover subtle effects among this large amount of data. In this context, this work consists in studying a workflow describing the general feature selection process, using knowledge discovery and data mining methodologies to propose advanced solutions for predictive biomarker discovery. The strategy was focused on evaluating a combination of numeric-symbolic approaches for feature selection with the objective of obtaining the best combination of metabolites producing an effective and accurate predictive model. Relying first on numerical approaches, and especially on machine learning methods (SVM-RFE, RF, RF-RFE) and on univariate statistical analyses (ANOVA), a comparative study was performed on an original metabolomic dataset and reduced subsets. As resampling method, LOOCV was applied to minimize the risk of overfitting. The best k-features obtained with different scores of importance from the combination of these different approaches were compared and allowed determining the variable stabilities using Formal Concept Analysis. The results revealed the interest of RF-Gini combined with ANOVA for feature selection as these two complementary methods allowed selecting the 48 best candidates for prediction. Using linear logistic regression on this reduced dataset enabled us to obtain the best performances in terms of prediction accuracy and number of false positive with a model including 5 top variables. Therefore, these results highlighted the interest of feature selection methods and the importance of working on reduced datasets for the identification of predictive biomarkers issued from untargeted metabolomics data.

Antibody-drug conjugate bioanalysis using LB-LC-MS/MS hybrid assays: strategies, methodology and correlation to ligand-binding assays.

Antibody-drug conjugates (ADCs) are complex drug constructs with multiple species in the heterogeneous mixture that contribute to their efficacy and toxicity. The bioanalysis of ADCs involves multiple assays and analytical platforms. A series of ligand binding and LC-MS/MS (LB-LC-MS/MS) hybrid assays, through different combinations of anti-idiotype (anti-Id), anti-payload, or generic capture reagents, and cathepsin-B or trypsin enzyme digestion, were developed and evaluated for the analysis of conjugated-payload as well as for species traditionally measured by ligand-binding assays, total-antibody and conjugated-antibody. Hybrid assays are complementary or viable alternatives to ligand-binding assay for ADC bioanalysis and PK/PD modeling. The fit-for-purpose choice of analytes, assays and platforms and an integrated strategy from Discovery to Development for ADC PK and bioanalysis are recommended.