Serum lipidomic study reveals potential early biomarkers for predicting response to chemoradiation therapy in advanced rectal cancer: A pilot study

Metabolomics for searching validated biomarkers in cancer studies: a decade in review

INTRODUCTION

In the dynamic landscape of modern healthcare, the ability to anticipate and diagnose diseases, particularly in cases where early treatment significantly impacts outcomes, is paramount. Cancer, a complex and heterogeneous disease, underscores the critical importance of early diagnosis for patient survival. The integration of metabolomics information has emerged as a crucial tool, complementing the genotype-phenotype landscape and providing insights into active metabolic mechanisms and disease-induced dysregulated pathways.

AREAS COVERED

This review explores a decade of developments in the search for biomarkers validated within the realm of cancer studies. By critically assessing a diverse array of research articles, clinical trials, and studies, this review aims to present an overview of the methodologies employed and the progress achieved in identifying and validating biomarkers in metabolomics results for various cancer types.

EXPERT OPINION

Through an exploration of more than 800 studies, this review has allowed to establish a general idea about state-of-art in the search of biomarkers in metabolomics studies involving cancer which include certain level of results validation. The potential for metabolites as diagnostic markers to reach the clinic and make a real difference in patient health is substantial, but challenges remain to be explored.

Multi-omics staging of locally advanced rectal cancer predicts treatment response: a pilot study

Treatment response assessment of rectal cancer patients is a critical component of personalized cancer care and it allows to identify suitable candidates for organ-preserving strategies. This pilot study employed a novel multi-omics approach combining MRI-based radiomic features and untargeted metabolomics to infer treatment response at staging. The metabolic signature highlighted how tumor cell viability is predictively down-regulated, while the response to oxidative stress was up-regulated in responder patients, showing significantly reduced oxoproline values at baseline compared to non-responder patients (p-value < 10-4). Tumors with a high degree of texture homogeneity, as assessed by radiomics, were more likely to achieve a major pathological response (p-value < 10-3). A machine learning classifier was implemented to summarize the multi-omics information and discriminate responders and non-responders. Combining all available radiomic and metabolomic features, the classifier delivered an AUC of 0.864 (± 0.083, p-value < 10-3) with a best-point sensitivity of 90.9% and a specificity of 81.8%. Our results suggest that a multi-omics approach, integrating radiomics and metabolomic data, can enhance the predictive value of standard MRI and could help to avoid unnecessary surgical treatments and their associated long-term complications.

Energy Metabolism Is Altered in Radioresistant Rectal Cancer

Resistance to neoadjuvant chemoradiation therapy is a significant clinical challenge in the management of rectal cancer. There is an unmet need to identify the underlying mechanisms of treatment resistance to enable the development of biomarkers predictive of response and novel treatment strategies to improve therapeutic response. In this study, an in vitro model of inherently radioresistant rectal cancer was identified and characterized to identify mechanisms underlying radioresistance in rectal cancer. Transcriptomic and functional analysis demonstrated significant alterations in multiple molecular pathways, including the cell cycle, DNA repair efficiency and upregulation of oxidative phosphorylation-related genes in radioresistant SW837 rectal cancer cells. Real-time metabolic profiling demonstrated decreased reliance on glycolysis and enhanced mitochondrial spare respiratory capacity in radioresistant SW837 cells when compared to radiosensitive HCT116 cells. Metabolomic profiling of pre-treatment serum samples from rectal cancer patients (n = 52) identified 16 metabolites significantly associated with subsequent pathological response to neoadjuvant chemoradiation therapy. Thirteen of these metabolites were also significantly associated with overall survival. This study demonstrates, for the first time, a role for metabolic reprograming in the radioresistance of rectal cancer in vitro and highlights a potential role for altered metabolites as novel circulating predictive markers of treatment response in rectal cancer patients.

Circulating Proteins Associated with Response and Resistance to Neoadjuvant Chemotherapy in HER2-Positive Breast Cancer

Simple Summary

The goal of this study was to find circulating proteins that can be easily sampled and incorporated into a clinical setting to improve predictive treatment response in HER2-positive breast cancer patients receiving neoadjuvant chemotherapy. We looked for potential biomarkers in serum, which we identified using two proteomics techniques: qualitative LC-MS/MS and a quantitative assay that assessed protein expression between responders and non-responders HER2-positive breast cancer patients to neoadjuvant chemotherapy.

Abstract

Despite the increasing use of neoadjuvant chemotherapy (NAC) in HER2-positive breast cancer (BC) patients, the clinical problem of predicting individual treatment response remains unanswered. Furthermore, the use of ineffective chemotherapeutic regimens should be avoided. Serum biomarker levels are being studied more and more for their ability to predict therapy response and aid in the development of personalized treatment regimens. This study aims to identify effective protein networks and biomarkers to predict response to NAC in HER2-positive BC patients through an exhaustive large-scale LC-MS/MS-based qualitative and quantitative proteomic profiling of serum samples from responders and non-responders. Serum samples from HER2-positive BC patients were collected before NAC and were processed by three methods (with and without nanoparticles). The qualitative analysis revealed differences in the proteomic profiles between responders and non-responders, mainly in proteins implicated in the complement and coagulation cascades and apolipoproteins. Qualitative analysis confirmed that three proteins (AFM, SERPINA1, APOD) were correlated with NAC resistance. In this study, we show that serum biomarker profiles can predict treatment response and outcome in the neoadjuvant setting. If these findings are further developed, they will be of significant clinical utility in the design of treatment regimens for individual BC patients.

Metabolic and lipidomic characterization of radioresistant MDA-MB-231 human breast cancer cells to investigate potential therapeutic targets

To provide preliminary insights into metabolic and lipidomic characteristics in radioresistant triple-negative breast cancer (TNBC) cells and suggest potential therapeutic targets, we performed a comprehensive metabolic and lipidomic profiling of radioresistant MDA-MB-231 (MDA-MB-231/RR) TNBC cells and their parental cells using gas chromatography-mass spectrometry and nano electrospray ionization-mass spectrometry, followed by multivariate statistical analysis. Buthionine sulfoximine (BSO) and radiation were co-treated to radioresistant TNBC cells. The level of glutathione (GSH) was significantly increased, and the levels of GSH synthesis-related metabolites, such as cysteine, glycine, and glutamine were also increased in MDA-MB-231/RR cells. In contrast, the level of lactic acid was significantly reduced. In addition, reactive oxygen species (ROS) level was decreased in MDA-MB-231/RR cells. In the lipidomic profiles of MDA-MB-231/RR cells, the levels of phosphatidylcholine (PC) and phosphatidylethanolamine (PE) were significantly increased, whereas those of most of the phosphatidylinositol species were significantly decreased. BSO sensitized MDA-MB-231/RR cells to radiotherapy, which resulted in decreased GSH level and increased ROS level and apoptosis. Radioresistant TNBC cells showed distinct metabolic and lipidomic characteristics compared to their parental cells. We suggested activated GSH, PC, and PE biosynthesis pathways as potential targets for treating radioresistant TNBC cells. Particularly, enhanced radiosensitivity was achieved by inhibition of GSH biosynthesis in MDA-MB-231/RR cells.

Regulation of EGFR activation and signaling by lipids on the plasma membrane

Lipids on the plasma membrane are not only components of the membrane biophysical structures but also regulators of receptor functions. Recently, the critical roles of lipid-protein interactions have been intensively highlighted. Epidermal growth factor receptor (EGFR) is one of the most extensively studied receptors exhibiting various lipid interactions, including interactions with phosphatidylcholine, phosphatidylserine, phosphatidylinositol phosphate, cholesterol, gangliosides, and palmitate. Here, we review recent findings on how direct interaction with these lipids regulates EGFR activation and signaling, providing unprecedented insight into the comprehensive roles of various lipids in the control of EGFR functions. Finally, the current limitations in investigating lipid-protein interactions and novel technologies to potentially overcome these limitations are discussed.

Serum Apolipoprotein A-I Predicts Response of Rectal Cancer to Neoadjuvant Chemoradiotherapy

Background

Serum lipids have been reported as prognosticators for malignancies, including rectal cancer (RC). Yet, their value in predicting the response of RC to neoadjuvant chemoradiotherapy (NACRT) remains unknown. This study aimed to assess the predictive abilities of serum lipids for a bad response, and to build a serum lipid-based prediction model.

Methods

In total, 751 patients diagnosed with stage cII–III RC and treated with NACRT plus surgery from January 2007 to August 2018 were retrospectively reviewed and randomly divided into two data sets, in a ratio of 1:1. Receiver operating characteristics (ROC) analysis was conducted in the development set to select possible predictors of bad NACRT response from pathoclinical factors, including serum lipids. Multivariate logistic regression was conducted to further determine independent predictors, which were then used to develop a prediction index (PI). Finally, the PI was verified in the validation set, through ROC analysis and chi-squared test.

Results

Five independent predictors were identified: tumor length ≥4 cm, cT4 stage, carcinoembryonic antigen ≥5.0 ng/mL, irradiation with three-dimensional conformal radiotherapy technique, and apolipoprotein A-I ≤1.20 g/L. Each of them was assigned a number of points. In the validation set, the area under the curve of PI appeared as 0.642 (95% confidence interval 0.586–0.697). The sensitivity, specificity, positive and negative predictive values, and concordance were 72.3%, 52.3%, 63.8%, 61.9%, and 63.0%, respectively.

Conclusion

Serum apolipoprotein A-I was found to correlate negatively with the RC response to NACRT. It could serve as a biomarker for guiding individualized treatment and a potential target for improving sensitivity to chemoradiation.

Transplantation of telomerase/myocardin-co-expressing mesenchymal cells in the mouse promotes myocardial revascularization and tissue repair

AIM

Cell therapies are hampered by poor survival and growth of grafts. We tested whether forced co-expression of telomerase reverse transcriptase (TERT) and myocardin (MYOCD) improves post-infarct revascularization and tissue repair by adipose tissue-derived mesenchymal stromal cells (AT-MSCs).

METHODS AND RESULTS

We transplanted AT-MSCs overexpressing MYOCD and TERT in a murine model of acute myocardial infarction (AMI). We characterized paracrine effects of AT-MSCs. When transplanted into infarcted hearts of C57BL/6 mice, AT-MSCs overexpressing TERT and MYOCD decreased scar tissue and the intra-scar CD3 and B220 lymphocyte infiltration; and increased arteriolar density as well as ejection fraction compared with saline or mock-transduced AT-MSCs. These effects were accompanied by higher persistence of the injected cells in the heart, increased numbers of Ki-67⁺ and CD117⁺ cells, and the expression of cardiac actin and β-myosin heavy chain. Intramyocardial delivery of the secretome and its extracellular vesicle (EV)-enriched fraction also decreased scar tissue formation and increased arteriolar density in the murine AMI model. Proteomic analysis of AT-MSCs-EV-enriched fraction predicted the activation of vascular development and the inhibition of immune cell trafficking. Elevated concentrations of miR-320a, miR-150-5p and miR-126-3p associated with regulation of apoptosis and vasculogenesis were confirmed in the AT-MSCs-EV-enriched fraction.

CONCLUSIONS

AT-MSCs overexpressing TERT and MYOCD promote persistence of transplanted aged AT-MSCs and enhance arteriolar density in a murine model of AMI. EV-enriched fraction is the component of the paracrine secretion by AT-MSCs with pro-angiogenic and anti-fibrotic activities.

In vitro and in vivo lipidomics as a tool for probiotics evaluation

The probiotic bacteria are helpful for nutritional and therapeutic purposes, and they are commercially available in various forms, such as capsules or powders. Increasing pieces of evidence indicate that different growth conditions and variability in manufacturing processes can determine the properties of probiotic products. In recent years, the lipidomic approach has become a useful tool to evaluate the impact that probiotics induce in host physiology. In this work, two probiotic formulations with identical species composition, produced in two different sites, the USA and Italy, were utilized to feed Caenorhabditis elegans, strains and alterations in lipid composition in the host and bacteria were investigated. Indeed, the multicellular organism C. elegans is considered a simple model to study the in vivo effects of probiotics. Nematodes fat metabolism was assessed by gene expression analysis and by mass spectrometry-based lipidomics. Lipid droplet analysis revealed a high accumulation of lipid droplets in worms fed US-made products, correlating with an increased expression of genes involved in the fatty acid synthesis. We also evaluated the lifespan of worms defective in genes involved in the insulin/IGF-1-mediated pathway and monitored the nuclear translocation of DAF-16. These data demonstrated the involvement of the signaling in C. elegans responses to the two diets. Lipidomics analysis of the two formulations was also conducted, and the results indicated differences in phosphatidylglycerol (PG) and phosphatidylcholine (PC) contents that, in turn, could influence nematode host physiology. Results demonstrated that different manufacturing processes could influence probiotics and host properties in terms of lipid composition. KEY POINTS: • Probiotic formulations impact on Caenorhabditis elegans lipid metabolism; • Lipidomic analysis highlighted phospholipid abundance in the two products; • Phosphocholines and phosphatidylglycerols were analyzed in worms fed the two probiotic formulations.

Discovery of Pancreatic Adenocarcinoma Biomarkers by Untargeted Metabolomics

Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive and lethal cancers, with a 5-year survival rate of less than 5%. In fact, complete surgical resection remains the only curative treatment. However, fewer than 20% of patients are candidates for surgery at the time of presentation. Hence, there is a critical need to identify diagnostic biomarkers with potential clinical utility in this pathology. In this context, metabolomics could be a powerful tool to search for new robust biomarkers. Comparative metabolomic profiling was performed in serum samples from 59 unresectable PDAC patients and 60 healthy controls. Samples were analyzed by using an untargeted metabolomics workflow based on liquid chromatography, coupled to high-resolution mass spectrometry in positive and negative electrospray ionization modes. Univariate and multivariate analysis allowed the identification of potential candidates that were significantly altered in PDAC patients. A panel of nine candidates yielded excellent diagnostic capacities. Pathway analysis revealed four altered pathways in our patients. This study shows the potential of liquid chromatography coupled to high-resolution mass spectrometry as a diagnostic tool for PDAC. Furthermore, it identified novel robust biomarkers with excellent diagnostic capacities.

Integrated Lipidomics and Metabolomics Analysis of Tears in Multiple Sclerosis: An Insight into Diagnostic Potential of Lacrimal Fluid

Metabolomics based on mass spectrometry represents an innovative approach to characterize multifactorial diseases, such as multiple sclerosis (MuS). To date, the most important biomarker source for MuS diagnosis is the cerebrospinal fluid. However, an important goal for research is to identify new molecules in more easily accessible biological fluids. A very interesting biofluid in MuS is represented by tears, considered as an intermediate fluid between the cerebrospinal fluid and serum. In this work, we developed a merged strategy for the analysis of lipids containing choline by Liquid Chromatography coupled to Tandem Mass Spectrometry (LC-MS/MS), as well as for the targeted analysis of free carnitine, acylcarnitines and aminoacids by direct infusion mass spectrometry. Samples for both metabolomics and lipidomics approaches were obtained in a single extraction procedure from tears of patients affected by MuS and healthy controls. Tear lipidomics showed 30 phospholipids significantly modulated and, notably, many sphingomyelins resulted lower in MuS. Moreover, the metabolomics approach carried out both on tears and serum highlighted the diagnostic potential of specific aminoacids and acylcarnitines. In conclusion, the metabolic profiling of tears appears to reflect the pathological conditions of the central nervous system, suggesting that the molecular repository of tears can be considered as a source of potential biomarkers for MuS.

Live single cell mass spectrometry reveals cancer-specific metabolic profiles of circulating tumor cells

Recently, there has been increased attention on the analysis of circulating tumor cells (CTCs), also known as liquid biopsy, owing to its potential benefits in cancer diagnosis and treatment. Circulating tumor cells are released from primary tumor lesions into the blood stream and eventually metastasize to distant body organs. However, a major hurdle with CTC analysis is their natural scarcity. Existing methods lack sensitivity, specificity, or reproducibility required in CTC characterization and detection. Here, we report untargeted molecular profiling of single CTCs obtained from gastric cancer and colorectal cancer patients, using live single cell mass spectrometry integrated with microfluidics‐based cell enrichment techniques. Using this approach, we showed the difference in the metabolomic profile between CTCs originating from different cancer groups. Moreover, potential biomarkers were putatively annotated to be specific to each cancer type.

Lipidomic Analysis of Cancer Cell and Tumor Tissues

Due to their role in cellular structure, energetics, and signaling, characterization of changes in cellular and extracellular lipid composition is of key importance to understand cancer biology. In addition, several mass spectrometry-based profiling as well as imaging studies have indicated that lipid molecules may be useful to augment existing biochemical and histopathological methods for diagnosis, staging, and prognosis of cancer. Therefore, analysis of lipidomic changes associated with cancer cells and tumor tissues can be useful for both fundamental and translational studies. Here, we provide a high-throughput single-extraction-based method that can be used for simultaneous lipidomic and metabolomic analysis of cancer cells or healthy or tumor tissue samples. In this chapter, a modified Bligh-Dyer method is described for extraction of lipids followed by analysis of fatty acid composition by gas chromatography-mass spectrometry (GC-MS) or untargeted lipidomics using electrospray ionization mass spectrometry (ESIMS) coupled with reverse-phase (RP) ultraperformance liquid chromatography (UPLC) followed by multivariate data analysis to identify features of interest.

Metabolomic Signature in Sera of Multiple Sclerosis Patients during Pregnancy

Multiple sclerosis (MuS) is an autoimmune disease of the central nervous system characterized by neuroinflammation, neurodegeneration, and degradation of the myelin sheath. Epidemiological studies have shown that the female gender is more susceptible than the male gender to MuS development, with a female-to-male ratio of 2:1. Despite this high onset, women have a better prognosis than men, and the frequency of the relapsing phase decreases during pregnancy, while it increases soon after birth. Therefore, it is interesting to investigate hormonal fluctuations during pregnancy and whether they correlate with metabolic signatures. To gain a deeper inside into the biochemical mechanism of such a multifactorial disease, we adopted targeted metabolomics approaches for the determination of many serum metabolites in 12 pregnant women affected by MuS by mass spectrometry analysis. Our data show a characteristic hormonal fluctuation for estrogens and progesterone, as expected. They also highlight other interesting hormonal alterations for cortisol, corticosterone, 11-deoxycortisol, 4-androstene-3,17-dione, testosterone, and 17α-hydroxyprogesterone. Furthermore, a negative correlation with progesterone levels was observed for amino acids and for acylcarnitines, while an imbalance of different sphingolipids pathways was found during pregnancy. In conclusion, these data are in agreement with the characteristic clinical signs of MuS patients during pregnancy and, if confirmed, they may add an important tessera in the complex mosaic of maternal neuroprotection.

Individualized metabolic profiling stratifies pancreatic and biliary tract cancer: a useful tool for innovative screening programs and predictive strategies in healthcare

BACKGROUND

Pancreatic cancer (PC) and biliary tract cancer (BTC) are highly aggressive cancers, characterized by their rarity, difficulty in diagnosis, and overall poor prognosis. Diagnosis of PC and BTC is complex and is made using a combination of appropriate clinical suspicion, imaging and endoscopic techniques, and cytopathological examination. However, the late-stage detection and poor prognosis of this tumor have led to an urgent need for biomarkers for early and/or predictive diagnosis and improved personalized treatments.

WORKING HYPOTHESIS

There are two hypotheses for focusing on low-mass metabolites in the blood. First, valuable information can be obtained from the masses and relative amounts of such metabolites, which present as low-mass ions (LMIs) in mass spectra. Second, metabolic profiling of individuals may provide important information regarding biological changes in disease states that is useful for the early diagnosis of PC and BTC.

MATERIALS AND METHODS

To assess whether profiling metabolites in serum can serve as a non-invasive screening tool for PC and BTC, 320 serum samples were obtained from patients with PC (n = 51), BTC (n = 39), colorectal cancer (CRC) (n = 100), and ovarian cancer (OVC) (n = 30), and from healthy control subjects (control) (n = 100). We obtained information on the relative amounts of metabolites, as LMIs, via triple time-of-flight mass spectrometry. All data were analyzed according to the peak area ratios of discriminative LMIs.

RESULTS AND CONCLUSIONS

The levels of the 14 discriminative LMIs were higher in the PC and BTC groups than in the control, CRC and OVC groups, but only two LMIs discriminated between PC and BTC: lysophosphatidylcholine (LysoPC) (16:0) and LysoPC(20:4). The levels of these two LysoPCs were also slightly lower in the PC/BTC/CRC/OVC groups compared with the control group. Taken together, the data showed that metabolic profiling can precisely denote the status of cancer, and, thus, could be useful for screening. This study not only details efficient methods to identify discriminative LMIs for cancer screening but also provides an example of metabolic profiling for distinguishing PC from BTC. Furthermore, the two metabolites [LysoPC(16:0), LysoPC(20:4)] shown to discriminate these diseases are potentially useful when combined with other, previously identified protein or metabolic biomarkers for predictive, preventive and personalized medical approach.

Analysis of the Metabolic Characteristics of Serum Samples in Patients With Multiple Myeloma

Aims: This study aimed to identify potential, non-invasive biomarkers for diagnosis and monitoring of the progress in multiple myeloma (MM) patients. Methods: MM patients and age-matched healthy controls (HC) were recruited in Discovery phase and Validation phase, respectively. MM patients were segregated into active group (AG) and responding group (RG). Serum samples were collected were conducted to non-targeted metabolomics analyses. Metabolites which were significantly changed (SCMs) among groups were identified in Discovery phase and was validated in Validation phase. The signaling pathways of these SCMs were enriched. The ability of SCMs to discriminate among groups in Validation phase was analyzed through receiver operating characteristic curve. The correlations between SCMs and clinical features, between SCMs and survival period of MM patients were analyzed. Results: Total of 23 SCMs were identified in AG compared with HC both in Discovery phase and Validation phase. Those SCMs were significantly enriched in arginine and proline metabolism and glycerophospholipid metabolism. 4 SCMs had the discriminatory ability between MM patients and healthy controls in Validation phase. Moreover, 12 SCMs had the ability to discriminate between the AG patients and RG patients in Validation phase. 10 out of 12 SCMs correlated with advanced features of MM. Moreover, 8 out of 12 SCMs had the negative impact on the survival of MM. 5'-Methylthioadenosine may be the only independent prognostic factor in survival period of MM. Conclusion: 10 SCMs identified in our study, which correlated with advanced features of MM, could be potential, novel, non-invasive biomarkers for active disease in MM.

The Peripheral and Intratumoral Immune Cell Landscape in Cancer Patients: A Proxy for Tumor Biology and a Tool for Outcome Prediction

Functional systemic and local immunity is required for effective anti-tumor responses. In addition to an active engagement with cancer cells and tumor stroma, immune cells can be affected and are often found to be dysregulated in cancer patients. The impact of tumors on local and systemic immunity can be assessed using a variety of approaches ranging from low-dimensional analyses that are performed on large patient cohorts to multi-dimensional assays that are technically and logistically challenging and are therefore confined to a limited sample size. Many of these strategies have been established in recent years leading to exciting findings. Not only were analyses of immune cells in tumor patients able to predict the clinical course of the disease and patients' survival, numerous studies also detected changes in the immune landscape that correlated with responses to novel immunotherapies. This review will provide an overview of established and novel tools for assessing immune cells in tumor patients and will discuss exemplary studies that utilized these techniques to predict patient outcomes.

Enhanced release of acid sphingomyelinase-enriched exosomes generates a lipidomics signature in CSF of Multiple Sclerosis patients

Multiple Sclerosis (MuS) is a complex multifactorial neuropathology, resulting in heterogeneous clinical presentation. A very active MuS research field concerns the discovery of biomarkers helpful to make an early and definite diagnosis. The sphingomyelin pathway has emerged as a molecular mechanism involved in MuS, since high levels of ceramides in cerebrospinal fluid (CSF) were related to axonal damage and neuronal dysfunction. Ceramides are the hydrolysis products of sphingomyelins through a reaction catalyzed by a family of enzymes named sphingomyelinases, which were recently related to myelin repair in MuS. Here, using a lipidomic approach, we observed low levels of several sphingomyelins in CSF of MuS patients compared to other inflammatory and non-inflammatory, central or peripheral neurological diseases. Starting by this result, we investigated the sphingomyelinase activity in CSF, showing a significantly higher enzyme activity in MuS. In support of these results we found high number of total exosomes in CSF of MuS patients and a high number of acid sphingomyelinase-enriched exosomes correlated to enzymatic activity and to disease severity. These data are of diagnostic relevance and show, for the first time, high number of acid sphingomyelinase-enriched exosomes in MuS, opening a new window for therapeutic approaches/targets in the treatment of MuS.

Partial least squares-discriminant analysis (PLS-DA) for classification of high-dimensional (HD) data: a review of contemporary practice strategies and knowledge gaps

This review highlights and discusses critically various knowledge gaps in classification modelling using PLS-DA for high dimensional data.