A multivariate curve resolution analysis of multicenter proton spectroscopic imaging of the prostate for cancer localization and assessment of aggressiveness.

In this study, we investigated the potential of the multivariate curve resolution alternating least squares (MCR-ALS) algorithm for analyzing three-dimensional (3D) 1 H-MRSI data of the prostate in prostate cancer (PCa) patients. MCR-ALS generates relative intensities of components representing spectral profiles derived from a large training set of patients, providing an interpretable model. Our objectives were to classify magnetic resonance (MR) spectra, differentiating tumor lesions from benign tissue, and to assess PCa aggressiveness. We included multicenter 3D 1 H-MRSI data from 106 PCa patients across eight centers. The patient cohort was divided into a training set (N = 63) and an independent test set (N = 43). Singular value decomposition determined that MR spectra were optimally represented by five components. The profiles of these components were extracted from the training set by MCR-ALS and assigned to specific tissue types. Using these components, MCR-ALS was applied to the test set for a quantitative analysis to discriminate tumor lesions from benign tissue and to assess tumor aggressiveness. Relative intensity maps of the components were reconstructed and compared with histopathology reports. The quantitative analysis demonstrated a significant separation between tumor and benign voxels (t-test, p < 0.001). This result was achieved including voxels with low-quality MR spectra. A receiver operating characteristic analysis of the relative intensity of the tumor component revealed that low- and high-risk tumor lesions could be distinguished with an area under the curve of 0.88. Maps of this component properly identified the extent of tumor lesions. Our study demonstrated that MCR-ALS analysis of 1 H-MRSI of the prostate can reliably identify tumor lesions and assess their aggressiveness. It handled multicenter data with minimal preprocessing and without using prior knowledge or quality control. These findings indicate that MCR-ALS can serve as an automated tool to assess the presence, extent, and aggressiveness of tumor lesions in the prostate, enhancing diagnostic capabilities and treatment planning of PCa patients.

Evaluation and comparison of unsupervised methods for the extraction of spatial patterns from mass spectrometry imaging data (MSI).

For the extraction of spatially important regions from mass spectrometry imaging (MSI) data, different clustering methods have been proposed. These clustering methods are based on certain assumptions and use different criteria to assign pixels into different classes. For high-dimensional MSI data, the curse of dimensionality also limits the performance of clustering methods which are usually overcome by pre-processing the data using dimension reduction techniques. In summary, the extraction of spatial patterns from MSI data can be done using different unsupervised methods, but the robust evaluation of clustering results is what is still missing. In this study, we have performed multiple simulations on synthetic and real MSI data to validate the performance of unsupervised methods. The synthetic data were simulated mimicking important spatial and statistical properties of real MSI data. Our simulation results confirmed that K-means clustering with correlation distance and Gaussian Mixture Modeling clustering methods give optimal performance in most of the scenarios. The clustering methods give efficient results together with dimension reduction techniques. From all the dimension techniques considered here, the best results were obtained with the minimum noise fraction (MNF) transform. The results were confirmed on both synthetic and real MSI data. However, for successful implementation of MNF transform the MSI data requires to be of limited dimensions.

Convolutional neural networks to predict brain tumor grades and Alzheimer's disease with MR spectroscopic imaging data.

PURPOSE: To evaluate the value of convolutional neural network (CNN) in the diagnosis of human brain tumor or Alzheimer's disease by MR spectroscopic imaging (MRSI) and to compare its Matthews correlation coefficient (MCC) score against that of other machine learning methods and previous evaluation of the same data. We address two challenges: 1) limited number of cases in MRSI datasets and 2) interpretability of results in the form of relevant spectral regions. METHODS: A shallow CNN with only one hidden layer and an ad-hoc loss function was constructed involving two branches for processing spectral and image features of a brain voxel respectively. Each branch consists of a single convolutional hidden layer. The output of the two convolutional layers is merged and fed to a classification layer that outputs class predictions for the given brain voxel. RESULTS: Our CNN method separated glioma grades 3 and 4 and identified Alzheimer's disease patients using MRSI and complementary MRI data with high MCC score (Area Under the Curve were 0.87 and 0.91 respectively). The results demonstrated superior effectiveness over other popular methods as Partial Least Squares or Support Vector Machines. Also, our method automatically identified the spectral regions most important in the diagnosis process and we show that these are in good agreement with existing biomarkers from the literature. CONCLUSION: Shallow CNNs models integrating image and spectral features improved quantitative and exploration and diagnosis of brain diseases for research and clinical purposes. Software is available at https://bitbucket.org/TeslaH2O/cnn_mrsi.

Predicting the performance of handheld near-infrared photonic sensors from a master benchtop device.

Many industries see a shifting focus towards performing on-site analysis using handheld spectroscopic devices. A determining factor for decision-making on the commissioning of these devices is available information on the potential performance of the device for specific applications. By now, myriad handheld solutions with very different specifications and pricing are available on the market. Although specifications are generally available for new devices, this does not directly quantify or predict how available devices will perform for targeted cases. We present a novel chemometric method to estimate the prediction performance of handheld NIR hardware and apply it to estimate the performance of two commercially available handheld NIR technologies in predicting protein content (ranging 120-180 g kg-1) in pig feed from existing data of a benchtop device. Adjusting benchtop data to the wavelength range and resolution of the handheld device lead to over-optimistic estimates of the handheld performances. Our method additionally utilizes information on the error structure of the handheld devices for the estimation. It yielded performance estimates differing less than 1 g kg-1 from the experimentally determined handheld performances and similar model parameters. Our method was effective for linear and nonlinear calibration algorithms, also when estimating performance after averaging multiple scans. Replicate spectra of twenty samples recorded using the handheld were required for replication error estimation to obtain an accurate performance estimation. The error structure could be reported by manufacturers in the future for this approach to be universally employed for predictive quantitative technology assessment. Overall, our method provides estimates of the performance of a handheld device for a specific task with minimal testing required and can thus be used as a device or application screening tool before committing to develop calibrations.

High-throughput single cell data analysis - A tutorial.

White blood cells protect the body against disease but may also cause chronic inflammation, auto-immune diseases or leukemia. There are many different white blood cell types whose identity and function can be studied by measuring their protein expression. Therefore, high-throughput analytical instruments were developed to measure multiple proteins on millions of single cells. The information-rich biochemistry information may only be fully extracted using multivariate statistics. Here we show an overview of the most essential steps for multivariate data analysis of single cell data. We used white blood cells (immunology) as a case study, but a similar approach may be used in environment or biotech research. The first step is analyzing the study design and subsequently formulating a research question. The three main designs are immunophenotyping (finding different cell types), cell activation and rare cell discovery. When preparing the data it is essential to consider the design and focus on the cell type of interest by removing all unwanted events. After pre-processing, the ten-thousands to millions of single cells per sample need to be converted into a cellular distribution. For immunophenotyping a clustering method such as Self-Organizing Maps is useful and for cell activation a model that describes the covariance such as Principal Component Analysis is useful. In rare cell discovery it is useful to first model all common cells and remove them to find the rare cells. Finally discriminant analysis based on the cellular distribution may highlight which cell (sub)types are different between groups.

Cost-benefit analysis of calibration model maintenance strategies for process monitoring.

The long-term prediction performance of spectroscopic calibration models is a critical factor to monitor or control many production processes. Over time, new variations may emerge that deteriorate prediction performance. Therefore, models have to be maintained to retain or improve their prediction performance through time, requiring considerable resources and data. Maintenance should improve relevant predictions but also needs to be resource and cost efficient. Current approaches do not consider these trade-offs. We propose a new method to quantify the effectiveness and cost of model maintenance strategies based on historical data. Model performance over time for past, imminent and future samples is evaluated as these may react differently to maintenance. The model performance and required updating resources are translated into relative cost and benefit to compare strategies and determine optimal maintenance parameters. We used this method to evaluate a maintenance strategy that combines adding incoming samples to the calibration data with re-optimization of spectral preprocessing and modelling parameters. Continuously adding samples to the calibration data is shown to improve prediction performance and leads to more robust and generic models for emerging variations in all investigated data streams. Selectively adding incoming sample variations showed a reduced prediction performance but saves considerably in resources. Comparing model performance on the different sampling windows can also be used to determine an optimal updating frequency. This novel strategy to evaluate the expected performance and determine an optimal maintenance strategy is generally applicable and should lead to robust and consistently high prospective and/or retrospective model performance through time, which can be crucial for optimal operation and fault detection in industrial processes.

Quantitative vibrational spectroscopy on liquid mixtures: concentration units matter.

Quantitative vibrational absorption spectroscopies rely on Beer's law relating spectroscopic intensities in a linear fashion to chemical concentrations. To address and clarify contrasting results in the literature about the difference between volume- and mass-based concentrations units used for quantitative spectroscopy on liquid solutions, we performed near-infrared, mid-infrared, and Raman spectroscopy measurements on four different binary solvent mixtures. Using classical least squares (CLS) and partial least squares (PLS) as multivariate analysis methods, we demonstrate that spectroscopic intensities are linearly related to volume-based concentration units rather than more widely used mass-based concentration units such as weight percent. The CLS results show that the difference in root mean square error of prediction (RMSEP) values between CLS models based on mass and volume fractions correlates strongly with the density difference between the two solvents in each binary mixture. This is explained by the fact that density differences are the source of non-linearity between mass and volume fractions in such mixtures. We also show that PLS calibration handles the non-linearity in mass-based models by the inclusion of additional latent variables that describe residual spectroscopic variation beyond the first latent variable (e.g., due to small peak shifts), as observed in the experimental data of all binary solvent mixtures. Using simulation studies, we have quantified the relative errors (up to 10-15%) that are made in PLS modeling when using mass fractions instead of volume fractions. Overall, our results provide conclusive evidence that concentration units based on volume should be preferred for optimal spectroscopic calibration results in academic and industrial practice.

A methodological approach to correlate tumor heterogeneity with drug distribution profile in mass spectrometry imaging data.

BACKGROUND: Drug mass spectrometry imaging (MSI) data contain knowledge about drug and several other molecular ions present in a biological sample. However, a proper approach to fully explore the potential of such type of data is still missing. Therefore, a computational pipeline that combines different spatial and non-spatial methods is proposed to link the observed drug distribution profile with tumor heterogeneity in solid tumor. Our data analysis steps include pre-processing of MSI data, cluster analysis, drug local indicators of spatial association (LISA) map, and ions selection. RESULTS: The number of clusters identified from different tumor tissues. The spatial homogeneity of the individual cluster was measured using a modified version of our drug homogeneity method. The clustered image and drug LISA map were simultaneously analyzed to link identified clusters with observed drug distribution profile. Finally, ions selection was performed using the spatially aware method. CONCLUSIONS: In this paper, we have shown an approach to correlate the drug distribution with spatial heterogeneity in untargeted MSI data. Our approach is freely available in an R package 'CorrDrugTumorMSI'.

Multi-set Pre-processing of Multicolor Flow Cytometry Data.

Flow Cytometry is an analytical technology to simultaneously measure multiple markers per single cell. Ten thousands to millions of single cells can be measured per sample and each sample may contain a different number of cells. All samples may be bundled together, leading to a 'multi-set' structure. Many multivariate methods have been developed for Flow Cytometry data but none of them considers this structure in their quantitative handling of the data. The standard pre-processing used by existing multivariate methods provides models mainly influenced by the samples with more cells, while such a model should provide a balanced view of the biomedical information within all measurements. We propose an alternative 'multi-set' preprocessing that corrects for the difference in number of cells measured, balancing the relative importance of each multi-cell sample in the data while using all data collected from these expensive analyses. Moreover, one case example shows how multi-set pre-processing may benefit removal of undesired measurement-to-measurement variability and another where class-based multi-set pre-processing enhances the studied response upon comparison to the control reference samples. Our results show that adjusting data analysis algorithms to consider this multi-set structure may greatly benefit immunological insight and classification performance of Flow Cytometry data.

Angle Distribution of Loading Subspace (ADLS) for estimating chemical rank in multivariate analysis: Applications in spectroscopy and chromatography.

Multivariate analyses are increasingly popular to explore the underlying structure of multivariate datasets, which are more and more prevalent in analytical chemistry. However, difficulties can be associated with estimating the number of components for the data with considerable coherence and noise. The method of Angle Distribution of Loading Subspace (ADLS) has been proposed to estimate the number of components for Principal Component Analysis (PCA) and PARAllel FACtor analysis (PARAFAC), which showed some advantages, in particular in the case of datasets with high coherence, over the commonly used methods (scree plot and cross-validation in PCA, and core consistency diagnostics (CORCONDIA) in PARAFAC). In this paper, we systematically improved and applied ADLS to estimate the number of components in different multivariate methods including, Multivariate Curve Resolution (MCR), PARAFAC and four-way PARAFAC. Firstly, we showed that ADLS performed better when estimating the chemical rank for MCR analysis, compared with scree plots. As well as this, we improved ADLS in multi-way analysis (three- and four-way PARAFAC) by calculating the loading subspace in advance using the Khatri-Rao product. The improved ADLS in multi-way analysis provided the correct result for the simulated three-way fluorescence datasets with unevenly distributed coherence at different dimensions, while the previous version of ADLS showed biased results and CORCONDIA / split-half analysis provided relatively unstable results. Moreover, ADLS was used to estimate the chemical rank for a four-way real-life fluorescence dataset analyzed by four-way PARAFAC. In this case the result of chemical rank results from ADLS was more precise and informative compared with CORCONDIA /split-half analysis in four-way analysis.

Drug-Homogeneity Index in Mass-Spectrometry Imaging.

Enhancing drug penetration in solid tumors is an interesting clinical issue of considerable importance. In preclinical research, mass-spectrometry imaging is a promising technique for visualizing drug distribution in tumors under different treatment conditions and its application in this field is rapidly increasing. However, in view of the huge variability among MSI data sets, drug homogeneity is usually manually assessed by an expert, and this approach is biased by interobserver variability and lacks reproducibility. We propose a new texture-based feature, the drug-homogeneity index (DHI), which provides an objective, automated measure of drug homogeneity in MSI data. A simulation study on synthetic data sets showed that previously known texture features do not give an accurate picture of intratumor drug-distribution patterns and are easily influenced by the tumor-tissue morphology. The DHI has been used to study the distribution profile of the anticancer drug paclitaxel in various xenograft models, which were either pretreated or not pretreated with antiangiogenesis compounds. The conclusion is that drug homogeneity is better in the pretreated condition, which is in agreement with previous experimental findings published by our group. This study shows that DHI could be useful in preclinical studies as a new parameter for the evaluation of protocols for better drug penetration.

Automated flow cytometric identification of disease-specific cells by the ECLIPSE algorithm.

Multicolor Flow Cytometry (MFC)-based gating allows the selection of cellular (pheno)types based on their unique marker expression. Current manual gating practice is highly subjective and may remove relevant information to preclude discovery of cell populations with specific co-expression of multiple markers. Only multivariate approaches can extract such aspects of cell variability from multi-dimensional MFC data. We describe the novel method ECLIPSE (Elimination of Cells Lying in Patterns Similar to Endogeneity) to identify and characterize aberrant cells present in individuals out of homeostasis. ECLIPSE combines dimensionality reduction by Simultaneous Component Analysis with Kernel Density Estimates. A Difference between Densities (DbD) is used to eliminate cells in responder samples that overlap in marker expression with cells of controls. Thereby, subsequent data analyses focus on the immune response-specific cells, leading to more informative and focused models. To prove the power of ECLIPSE, we applied the method to study two distinct datasets: the in vivo neutrophil response induced by systemic endotoxin challenge and in studying the heterogeneous immune-response of asthmatics. ECLIPSE described the well-characterized common response in the LPS challenge insightfully, while identifying slight differences between responders. Also, ECLIPSE enabled characterization of the immune response associated to asthma, where the co-expressions between all markers were used to stratify patients according to disease-specific cell profiles.

Chemometrics-assisted microfluidic paper-based analytical device for the determination of uric acid by silver nanoparticle plasmon resonance.

This manuscript reports on the application of chemometric methods for the development of an optimized microfluidic paper-based analytical device (µPAD). As an example, we applied chemometric methods for both device optimization and data processing of results of a colorimetric uric acid assay. Box-Behnken designs (BBD) were utilized for the optimization of the device geometry and the amount of thermal inkjet-deposited assay reagents, which affect the assay performance. Measurement outliers were detected in real time by partial least squares discriminant analysis (PLS-DA) of scanned images. The colorimetric assay mechanism is based on the on-device formation of silver nanoparticles (AgNPs) through the interaction of uric acid, ammonia, and poly(vinyl alcohol) with silver ions under mild basic conditions. The yellow color originating from visible light absorption by localized surface plasmon resonance of AgNPs can be detected by the naked eye or, more quantitatively, with a simple flat-bed scanner. Under optimized conditions, the linearity of the calibration curve ranges from 0.1-5 × 10-3 mol L-1 of uric acid with a limit of detection of 33.9 × 10-6 mol L-1 and a relative standard of deviation 4.5% (n = 3 for determination of 5.0 × 10-3 mol L-1 uric acid). Graphical abstract A chemometrics-assisted microfluidic paper-based analytical device was developed as a low-cost and rapid platform for the determination of uric acid (UA). The detection method is based on the chemical interaction of UA, ammonia, and polyvinyl alcohol under mild basic condition with silver ions inducing formation of yellow silver nanoparticles (AgNPs).

Comprehensive three-dimensional ray tracing model for three-mirror cavity-enhanced spectroscopy.

A 3D ray tracing model is used to simulate optical reinjection in a nonresonant optical cavity, for off-axis integrated cavity output spectroscopy. The optical cavities are optimized for maximum intensity enhancement factors via a grid search and a genetic algorithm. Intensity enhancement factors up to 1400 are found for short cavities (3 cm) and up to 101 for long cavities (50 cm). The model predicts that short absorption cells can be used, having a long effective path length and a high throughput power. This opens new opportunities in the field of ultrasensitive absorption spectroscopy and allows the design of compact optical gas sensors.

Combining ANOVA-PCA with POCHEMON to analyse micro-organism development in a polymicrobial environment.

Revealing the biochemistry associated to micro-organismal interspecies interactions is highly relevant for many purposes. Each pathogen has a characteristic metabolic fingerprint that allows identification based on their unique multivariate biochemistry. When pathogen species come into mutual contact, their co-culture will display a chemistry that may be attributed both to mixing of the characteristic chemistries of the mono-cultures and to competition between the pathogens. Therefore, investigating pathogen development in a polymicrobial environment requires dedicated chemometric methods to untangle and focus upon these sources of variation. The multivariate data analysis method Projected Orthogonalised Chemical Encounter Monitoring (POCHEMON) is dedicated to highlight metabolites characteristic for the interaction of two micro-organisms in co-culture. However, this approach is currently limited to a single time-point, while development of polymicrobial interactions may be highly dynamic. A well-known multivariate implementation of Analysis of Variance (ANOVA) uses Principal Component Analysis (ANOVA-PCA). This allows the overall dynamics to be separated from the pathogen-specific chemistry to analyse the contributions of both aspects separately. For this reason, we propose to integrate ANOVA-PCA with the POCHEMON approach to disentangle the pathogen dynamics and the specific biochemistry in interspecies interactions. Two complementary case studies show great potential for both liquid and gas chromatography - mass spectrometry to reveal novel information on chemistry specific to interspecies interaction during pathogen development.

Application of a cocktail approach to screen cytochrome P450 BM3 libraries for metabolic activity and diversity.

In the present study, the validity of using a cocktail screening method in combination with a chemometrical data mining approach to evaluate metabolic activity and diversity of drug-metabolizing bacterial Cytochrome P450 (CYP) BM3 mutants was investigated. In addition, the concept of utilizing an in-house-developed library of CYP BM3 mutants as a unique biocatalytic synthetic tool to support medicinal chemistry was evaluated. Metabolic efficiency of the mutant library towards a selection of CYP model substrates, being amitriptyline (AMI), buspirone (BUS), coumarine (COU), dextromethorphan (DEX), diclofenac (DIC) and norethisterone (NET), was investigated. First, metabolic activity of a selection of CYP BM3 mutants was screened against AMI and BUS. Subsequently, for a single CYP BM3 mutant, the effect of co-administration of multiple drugs on the metabolic activity and diversity towards AMI and BUS was investigated. Finally, a cocktail of AMI, BUS, COU, DEX, DIC and NET was screened against the whole in-house CYP BM3 library. Different validated quantitative and qualitative (U)HPLC-MS/MS-based analytical methods were applied to screen for substrate depletion and targeted product formation, followed by a more in-depth screen for metabolic diversity. A chemometrical approach was used to mine all data to search for unique metabolic properties of the mutants and allow classification of the mutants. The latter would open the possibility of obtaining a more in-depth mechanistic understanding of the metabolites. The presented method is the first MS-based method to screen CYP BM3 mutant libraries for diversity in combination with a chemometrical approach to interpret results and visualize differences between the tested mutants.

Climatic and geographical dependence of the H, C and O stable isotope ratios of Italian wine.

In this study, we investigated the relationship between (D/H)1, (D/H)2 and δ(13)C of ethanol and δ(18)O of water in wine, and variables describing the climate and the geography of the production area, using exploratory visualisation tools, regression analysis and linear modelling. For the first time, a large amount of data (around 4000 wine samples collected over 11 years in Italy) and all the official isotopic parameters, as well as a large number of significant climatic and geographical descriptors (date of harvest, latitude, longitude, elevation, distance from the sea, amount of precipitation, maximum daily temperature, minimum daily temperature, mean daily temperature, δ(18)O and δ(2)H of precipitation) were considered. δ(18)O, followed by (D/H)1, was shown to have the strongest relationship with climate and location. The dominant variables were latitude, with a negative relationship, δ(18)O and δ(2)H of precipitation and temperature, both with positive relationships. The identified correlations and models could be used to predict the isotopic composition of authentic wines, offering increased possibilities for detecting fraud and mislabelling.

Classification of brain tumours from MR spectra: the INTERPRET collaboration and its outcomes.

The INTERPRET project was a multicentre European collaboration, carried out from 2000 to 2002, which developed a decision-support system (DSS) for helping neuroradiologists with no experience of MRS to utilize spectroscopic data for the diagnosis and grading of human brain tumours. INTERPRET gathered a large collection of MR spectra of brain tumours and pseudo-tumoural lesions from seven centres. Consensus acquisition protocols, a standard processing pipeline and strict methods for quality control of the aquired data were put in place. Particular emphasis was placed on ensuring the diagnostic certainty of each case, for which all cases were evaluated by a clinical data validation committee. One outcome of the project is a database of 304 fully validated spectra from brain tumours, pseudotumoural lesions and normal brains, along with their associated images and clinical data, which remains available to the scientific and medical community. The second is the INTERPRET DSS, which has continued to be developed and clinically evaluated since the project ended. We also review here the results of the post-INTERPRET period. We evaluate the results of the studies with the INTERPRET database by other consortia or research groups. A summary of the clinical evaluations that have been performed on the post-INTERPRET DSS versions is also presented. Several have shown that diagnostic certainty can be improved for certain tumour types when the INTERPRET DSS is used in conjunction with conventional radiological image interpretation. About 30 papers concerned with the INTERPRET single-voxel dataset have so far been published. We discuss stengths and weaknesses of the DSS and the lessons learned. Finally we speculate on how the INTERPRET concept might be carried into the future.

Feasibility of MR metabolomics for immediate analysis of resection margins during breast cancer surgery.

In this study, the feasibility of high resolution magic angle spinning (HR MAS) magnetic resonance spectroscopy (MRS) of small tissue biopsies to distinguish between tumor and non-involved adjacent tissue was investigated. With the current methods, delineation of the tumor borders during breast cancer surgery is a challenging task for the surgeon, and a significant number of re-surgeries occur. We analyzed 328 tissue samples from 228 breast cancer patients using HR MAS MRS. Partial least squares discriminant analysis (PLS-DA) was applied to discriminate between tumor and non-involved adjacent tissue. Using proper double cross validation, high sensitivity and specificity of 91% and 93%, respectively was achieved. Analysis of the loading profiles from both principal component analysis (PCA) and PLS-DA showed the choline-containing metabolites as main biomarkers for tumor content, with phosphocholine being especially high in tumor tissue. Other indicative metabolites include glycine, taurine and glucose. We conclude that metabolic profiling by HR MAS MRS may be a potential method for on-line analysis of resection margins during breast cancer surgery to reduce the number of re-surgeries and risk of local recurrence.

Lactate and glycine-potential MR biomarkers of prognosis in estrogen receptor-positive breast cancers.

Breast cancer is a heterogeneous disease with a variable prognosis. Clinical factors provide some information about the prognosis of patients with breast cancer; however, there is a need for additional information to stratify patients for improved and more individualized treatment. The aim of this study was to examine the relationship between the metabolite profiles of breast cancer tissue and 5-year survival. Biopsies from breast cancer patients (n=98) were excised during surgery and analyzed by high-resolution magic angle spinning MRS. The data were analyzed by multivariate principal component analysis and partial least-squares discriminant analysis, and the findings of important metabolites were confirmed by spectral integration of the metabolite peaks. Predictions of 5-year survival using metabolite profiles were compared with predictions using clinical parameters. Based on the metabolite profiles, patients with estrogen receptor (ER)-positive breast cancer (n=71) were separated into two groups with significantly different survival rates (p=0.024). Higher levels of glycine and lactate were found to be associated with lower survival rates by both multivariate analyses and spectral integration, and are suggested as biomarkers for breast cancer prognosis. Similar metabolic differences were not observed for ER-negative patients, where survivors could not be separated from nonsurvivors. Predictions of 5-year survival of ER-positive patients using metabolite profiles gave better and more robust results than those using traditional clinical parameters. The results imply that the metabolic state of a tumor may provide additional information concerning breast cancer prognosis. Further studies should be conducted in order to evaluate the role of MR metabolomics as an additional clinical tool for determining the prognosis of patients with breast cancer.