Search

Current therapies for treating Glioblastoma (GB), and brain tumours in general, are inefficient and represent numerous challenges. In addition to surgical resection, chemotherapy and radiotherapy are presently used as standards of care. However, treated patients still face a dismal prognosis with a median survival below 15–18 months. Temozolomide (TMZ) is the main chemotherapeutic agent administered; however, intrinsic or acquired resistance to TMZ contributes to the limited efficacy of this drug. To circumvent the current drawbacks in GB treatment, a large number of classical and non-classical platinum complexes have been prepared and tested for anticancer activity, especially platinum (IV)-based prodrugs. Platinum complexes, used as alkylating agents in the anticancer chemotherapy of some malignancies, are though often associated with severe systemic toxicity (i.e., neurotoxicity), especially after long-term treatments. The objective of the current developments is to produce novel nanoformulations with improved lipophilicity and passive diffusion, promoting intracellular accumulation, while reducing toxicity and optimizing the concomitant treatment of chemo-/radiotherapy. Moreover, the blood–brain barrier (BBB) prevents the access of the drugs to the brain and accumulation in tumour cells, so it represents a key challenge for GB management. The development of novel nanomedicines with the ability to (i) encapsulate Pt-based drugs and pro-drugs, (ii) cross the BBB, and (iii) specifically target cancer cells represents a promising approach to increase the therapeutic effect of the anticancer drugs and reduce undesired side effects. In this review, a critical discussion is presented concerning different families of nanoparticles able to encapsulate platinum anticancer drugs and their application for GB treatment, emphasizing their potential for increasing the effectiveness of platinum-based drugs.

Proton Magnetic Resonance Spectroscopy (1H MRS) has proven its diagnostic potential in a variety of conditions. However, MRS is not yet widely used in clinical routine because of the lack of experts on its diagnostic interpretation. Although data-based decision support systems exist to aid diagnosis, they often take for granted that the data is of good quality, which is not always the case in a real application context. Systems based on models built with bad quality data are likely to underperform in their decision support tasks. In this study, we propose a system to filter out such bad quality data. It is based on convex Non-Negative Matrix Factorization models, used as a dimensionality reduction procedure, and on the use of several classifiers to discriminate between good and bad quality data., Peer Reviewed, Postprint (author's final draft)

Machine learning has provided, over the last decades, tools for knowledge extraction in complex medical domains. Most of these tools, though, are ad hoc solutions and lack the systematic approach that would be required to become mainstream in medical practice. In this brief paper, we define a machine learning-based analysis pipeline for helping in a difficult problem in the field of neuro-oncology, namely the discrimination of brain glioblastomas from single brain metastases. This pipeline involves source extraction using k-Meansinitialized Convex Non-negative Matrix Factorization and a collection of classifiers, including Logistic Regression, Linear Discriminant Analysis, AdaBoost, and Random Forests., Peer Reviewed, Postprint (published version)

[Background]: Magnetic resonance imaging (MRI) plays an important role in tumor detection/diagnosis. The use of exogenous contrast agents (CAs) helps to improve the discrimination between lesion and neighbouring tissue, but most of the currently available CAs are non-specific. Assessing the performance of new, selective CAs requires exhaustive assays and large amounts of material. Accordingly, in a preliminary screening of new CAs, it is important to choose candidate compounds with good potential for in vivo efficiency. This screening method should reproduce as close as possible the in vivo environment. In this sense, a fast and reliable method to select the best candidate CAs for in vivo studies would minimize time and investment cost, and would benefit the development of better CAs., [Results]: The post-mortem ex vivo relative contrast enhancement (RCE) was evaluated as a method to screen different types of CAs, including paramagnetic and superparamagnetic agents. In detail, sugar/gadolinium-loaded gold nanoparticles (Gd-GNPs) and iron nanoparticles (SPIONs) were tested. Our results indicate that the post-mortem ex vivo RCE of evaluated CAs, did not correlate well with their respective in vitro relaxivities. The results obtained with different Gd-GNPs suggest that the linker length of the sugar conjugate could modulate the interactions with cellular receptors and therefore the relaxivity value. A paramagnetic CA (GNP (E_2)), which performed best among a series of Gd-GNPs, was evaluated both ex vivo and in vivo. The ex vivo RCE was slightly worst than gadoterate meglumine (201.9 ± 9.3% versus 237 ± 14%, respectively), while the in vivo RCE, measured at the time-to-maximum enhancement for both compounds, pointed to GNP E_2 being a better CA in vivo than gadoterate meglumine. This is suggested to be related to the nanoparticule characteristics of the evaluated GNP., [Conclusion]: We have developed a simple, cost-effective relatively high-throughput method for selecting CAs for in vivo experiments. This method requires approximately 800 times less quantity of material than the amount used for in vivo administrations.

Background DCE@urLAB is a software application for analysis of dynamic contrast-enhanced magnetic resonance imaging data (DCE-MRI). The tool incorporates a friendly graphical user interface (GUI) to interactively select and analyze a region of interest (ROI) within the image set, taking into account the tissue concentration of the contrast agent (CA) and its effect on pixel intensity. Results Pixel-wise model-based quantitative parameters are estimated by fitting DCE-MRI data to several pharmacokinetic models using the Levenberg-Marquardt algorithm (LMA). DCE@urLAB also includes the semi-quantitative parametric and heuristic analysis approaches commonly used in practice. This software application has been programmed in the Interactive Data Language (IDL) and tested both with publicly available simulated data and preclinical studies from tumor-bearing mouse brains. Conclusions A user-friendly solution for applying pharmacokinetic and non-quantitative analysis DCE-MRI in preclinical studies has been implemented and tested. The proposed tool has been specially designed for easy selection of multi-pixel ROIs. A public release of DCE@urLAB, together with the open source code and sample datasets, is available at http://www.die.upm.es/im/archives/DCEurLAB/ webcite.

Background DCE@urLAB is a software application for analysis of dynamic contrast-enhanced magnetic resonance imaging data (DCE-MRI). The tool incorporates a friendly graphical user interface (GUI) to interactively select and analyze a region of interest (ROI) within the image set, taking into account the tissue concentration of the contrast agent (CA) and its effect on pixel intensity. Results Pixel-wise model-based quantitative parameters are estimated by fitting DCE-MRI data to several pharmacokinetic models using the Levenberg-Marquardt algorithm (LMA). DCE@urLAB also includes the semi-quantitative parametric and heuristic analysis approaches commonly used in practice. This software application has been programmed in the Interactive Data Language (IDL) and tested both with publicly available simulated data and preclinical studies from tumor-bearing mouse brains. Conclusions A user-friendly solution for applying pharmacokinetic and non-quantitative analysis DCE-MRI in preclinical studies has been implemented and tested. The proposed tool has been specially designed for easy selection of multi-pixel ROIs. A public release of DCE@urLAB, together with the open source code and sample datasets, is available at http://www.die.upm.es/im/archives/DCEurLAB/ webcite.

In most plants, sucrose is the primary product of photosynthesis, the transport form of assimilated carbon, and also one of the main factors determining sweetness in fresh fruits. Traditional methods for sugar quantification (mainly sucrose, glucose and fructose) require obtaining crude plant extracts, which sometimes involve substantial sample manipulation, making the process time-consuming and increasing the risk of sample degradation. Here, we describe and validate a fast method to determine sugar content in intact plant tissue by using high-resolution magic angle spinning nuclear magnetic resonance spectroscopy (HR-MAS NMR). The HR-MAS NMR method was used for quantifying sucrose, glucose and fructose in mesocarp tissues from melon fruits (Cucumis melo var. reticulatus and Cucumis melo var. cantalupensis). The resulting sugar content varied among individual melons, ranging from 1.4 to 7.3 g of sucrose, 0.4–2.5 g of glucose; and 0.73–2.83 g of fructose (values per 100 g fw). These values were in agreement with those described in the literature for melon fruit tissue, and no significant differences were found when comparing them with those obtained using the traditional, enzymatic procedure, on melon tissue extracts. The HR-MAS NMR method offers a fast (usually <30 min) and sensitive method for sugar quantification in intact plant tissues, it requires a small amount of tissue (typically 50 mg fw) and avoids the interferences and risks associated with obtaining plant extracts. Furthermore, this method might also allow the quantification of additional metabolites detectable in the plant tissue NMR spectrum.

Peer Reviewed, Postprint (published version)

The final publication is available at Springer via http://dx.doi.org/10.1007/s10334-010-0241-8, Object: This study demonstrates that 3T SV-MRS data can be used with the currently available automatic brain tumour diagnostic classifiers which were trained on databases of 1.5T spectra. This will allow the existing large databases of 1.5T MRS data to be used for diagnostic classification of 3T spectra, and perhaps also the combination of 1.5T and 3T databases. Materials and methods: Brain tumour classifiers trained with 154 1.5T spectra to discriminate among high grade malignant tumours and common grade II glial tumours were evaluated with a subsequently-acquired set of 155 1.5T and 37 3T spectra. A similarity study between spectra and main brain tumour metabolite ratios for both field strengths (1.5T and 3T) was also performed. Results: Our results showed that classifiers trained with 1.5T samples had similar accuracy for both test datasets (0.87 ± 0.03 for 1.5T and 0.88 ± 0.03 for 3.0T). Moreover, non-significant differences were observed with most metabolite ratios and spectral patterns. Conclusion: These results encourage the use of existing classifiers based on 1.5T datasets for diagnosis with 3T 1H SV-MRS. The large 1.5T databases compiled throughout many years and the prediction models based on 1.5T acquisitions can therefore continue to be used with data from the new 3T instruments. © 2011 ESMRMB.

The final publication is available at Springer via http://dx.doi.org/10.1007/s10334-010-0241-8, Object: This study demonstrates that 3T SV-MRS data can be used with the currently available automatic brain tumour diagnostic classifiers which were trained on databases of 1.5T spectra. This will allow the existing large databases of 1.5T MRS data to be used for diagnostic classification of 3T spectra, and perhaps also the combination of 1.5T and 3T databases. Materials and methods: Brain tumour classifiers trained with 154 1.5T spectra to discriminate among high grade malignant tumours and common grade II glial tumours were evaluated with a subsequently-acquired set of 155 1.5T and 37 3T spectra. A similarity study between spectra and main brain tumour metabolite ratios for both field strengths (1.5T and 3T) was also performed. Results: Our results showed that classifiers trained with 1.5T samples had similar accuracy for both test datasets (0.87 ± 0.03 for 1.5T and 0.88 ± 0.03 for 3.0T). Moreover, non-significant differences were observed with most metabolite ratios and spectral patterns. Conclusion: These results encourage the use of existing classifiers based on 1.5T datasets for diagnosis with 3T 1H SV-MRS. The large 1.5T databases compiled throughout many years and the prediction models based on 1.5T acquisitions can therefore continue to be used with data from the new 3T instruments. © 2011 ESMRMB.

In the last decade, machine learning (ML) techniques have been used for developing classifiers for automatic brain tumour diagnosis. However, the development of these ML models rely on a unique training set and learning stops once this set has been processed. Training these classifiers requires a representative amount of data, but the gathering, preprocess, and validation of samples is expensive and time-consuming. Therefore, for a classical, non-incremental approach to ML, it is necessary to wait long enough to collect all the required data. In contrast, an incremental learning approach may allow us to build an initial classifier with a smaller number of samples and update it incrementally when new data are collected. In this study, an incremental learning algorithm for Gaussian Discriminant Analysis (iGDA) based on the Graybill and Deal weighted combination of estimators is introduced. Each time a new set of data becomes available, a new estimation is carried out and a combination with a previous estimation is performed. iGDA does not require access to the previously used data and is able to include new classes that were not in the original analysis, thus allowing the customization of the models to the distribution of data at a particular clinical center. An evaluation using five benchmark databases has been used to evaluate the behaviour of the iGDA algorithm in terms of stability-plasticity, class inclusion and order effect. Finally, the iGDA algorithm has been applied to automatic brain tumour classification with magnetic resonance spectroscopy, and compared with two state-of-the-art incremental algorithms. The empirical results obtained show the ability of the algorithm to learn in an incremental fashion, improving the performance of the models when new information is available, and converging in the course of time. Furthermore, the algorithm shows a negligible instance and concept order effect, avoiding the bias that such effects could introduce. © 2011 Elsev

In the last decade, machine learning (ML) techniques have been used for developing classifiers for automatic brain tumour diagnosis. However, the development of these ML models rely on a unique training set and learning stops once this set has been processed. Training these classifiers requires a representative amount of data, but the gathering, preprocess, and validation of samples is expensive and time-consuming. Therefore, for a classical, non-incremental approach to ML, it is necessary to wait long enough to collect all the required data. In contrast, an incremental learning approach may allow us to build an initial classifier with a smaller number of samples and update it incrementally when new data are collected. In this study, an incremental learning algorithm for Gaussian Discriminant Analysis (iGDA) based on the Graybill and Deal weighted combination of estimators is introduced. Each time a new set of data becomes available, a new estimation is carried out and a combination with a previous estimation is performed. iGDA does not require access to the previously used data and is able to include new classes that were not in the original analysis, thus allowing the customization of the models to the distribution of data at a particular clinical center. An evaluation using five benchmark databases has been used to evaluate the behaviour of the iGDA algorithm in terms of stability-plasticity, class inclusion and order effect. Finally, the iGDA algorithm has been applied to automatic brain tumour classification with magnetic resonance spectroscopy, and compared with two state-of-the-art incremental algorithms. The empirical results obtained show the ability of the algorithm to learn in an incremental fashion, improving the performance of the models when new information is available, and converging in the course of time. Furthermore, the algorithm shows a negligible instance and concept order effect, avoiding the bias that such effects could introduce. © 2011 Elsev

Development of molecular diagnostics that can reliably differentiate amongst different subtypes of brain tumors is an important unmet clinical need in postgenomics medicine and clinical oncology. A simple linear formula derived from gene expression values of four genes (GFAP, PTPRZ1, GPM6B, and PRELP) measured from cDNA microarrays (n=35) have distinguished glioblastoma and meningioma cases in a previous study. We herein extend this work further and report that the above predictor formula showed its robustness when applied to Affymetrix microarray data acquired prospectively in our laboratory (n=80) as well as publicly available data (n=98). Importantly, GFAP and GPM6B were both retained as being significant in the predictive model upon using the Affymetrix data obtained in our laboratory, whereas the other two predictor genes were SFRP2 and SLC6A2. These results collectively indicate the importance of the expression values of GFAP and GPM6B genes sampled from the two types of microarray technologies tested. The high prediction accuracy obtained in these instances demonstrates the robustness of the predictors across microarray platforms used. This result would require further validation with a larger population of meningioma and glioblastoma cases. At any rate, this study paves the way for further application of gene signatures to more stringent biopsy discrimination challenges. © 2010, Mary Ann Liebert, Inc.

[EN] Aims: Gene signatures obtained from microarray experiments may be of use to improve the prediction of brain tumor diagnosis. Nevertheless, automated and objective prediction with accuracy comparable to or better than the gold standard should be convincingly demonstrated for possible clinician uptake of the new methodology. Herewith, we demonstrate that primary brain tumor types can be discriminated using microarray data in an automated and objective way. Methods: Postsurgical biopsies from 35 patients [17 glioblastoma multiforme (Gbm) and 18 meningothelial meningioma (Mm)] were stored in liquid nitrogen, total RNA was extracted, and cDNA was labeled with Cy3 fluorochrome and hybridized onto a cDNA-based microarray containing 11,500 cDNA clones representing 9300 loci. Scanned data were preprocessed, normalized, and used for predictor development. The predictive functions were fitted to a subset of samples and their performance evaluated with an independent subset. Expression results were validated by means of real time-polymerase chain reaction. Results: Some gene expression-based predictors achieved 100% accuracy both in training resampling validation and independent testing. One of them, composed of GFAP, PTPRZ1, GPM6B and PRELP, produced a 100% prediction accuracy for both training and independent test datasets. Furthermore, the gene signatures obtained, increased cell detoxification, motility and intracellular transport in Gbm, and increased cell adhesion and cytochrome-family genes in Mm, agree well with the expected biologic and pathologic characteristics of the studied tumors. Conclusions: The ability of gene signatures to automate prediction of brain tumors through a fully objective approach has been demonstrated. A comparison of gene expression profiles between Gbm and Mm may provide additional clues about patterns associated with each tumor type.

[EN] Aims: Gene signatures obtained from microarray experiments may be of use to improve the prediction of brain tumor diagnosis. Nevertheless, automated and objective prediction with accuracy comparable to or better than the gold standard should be convincingly demonstrated for possible clinician uptake of the new methodology. Herewith, we demonstrate that primary brain tumor types can be discriminated using microarray data in an automated and objective way. Methods: Postsurgical biopsies from 35 patients [17 glioblastoma multiforme (Gbm) and 18 meningothelial meningioma (Mm)] were stored in liquid nitrogen, total RNA was extracted, and cDNA was labeled with Cy3 fluorochrome and hybridized onto a cDNA-based microarray containing 11,500 cDNA clones representing 9300 loci. Scanned data were preprocessed, normalized, and used for predictor development. The predictive functions were fitted to a subset of samples and their performance evaluated with an independent subset. Expression results were validated by means of real time-polymerase chain reaction. Results: Some gene expression-based predictors achieved 100% accuracy both in training resampling validation and independent testing. One of them, composed of GFAP, PTPRZ1, GPM6B and PRELP, produced a 100% prediction accuracy for both training and independent test datasets. Furthermore, the gene signatures obtained, increased cell detoxification, motility and intracellular transport in Gbm, and increased cell adhesion and cytochrome-family genes in Mm, agree well with the expected biologic and pathologic characteristics of the studied tumors. Conclusions: The ability of gene signatures to automate prediction of brain tumors through a fully objective approach has been demonstrated. A comparison of gene expression profiles between Gbm and Mm may provide additional clues about patterns associated with each tumor type.

A Feature Selection (FS) process with a simple Machine Learning method, namely the Single-Layer Perceptron (SLP), is shown to discriminate metastases from glioblastomas with high accuracy using single voxel H-MRS from an international, multi-centre database of brain tumors. The method has low computational cost and its results are intuitively interpretable., Postprint (author’s final draft)

This paper describes a process of rule-extraction from a multi-centre brain tumour database consisting of nuclear magnetic res- onance spectroscopic signals. The expert diagnosis of human brain tumours can benefit from computer-aided assistance, which has to be readily interpretable by clinicians. Interpretation can be achieved through rule extraction, which is here performed using the LR-FIR algorithm, a method based on fuzzy logic. The experimental results of the classification of three groups of tumours indicate in this study that just three spectral frequencies, out of the 195 from a range pre-selected by experts, are enough to represent, in a simple and intuitive manner, most of the knowledge required to discriminate these groups., Postprint (published version)

Non-invasive techniques such asMagnetic Resonance Imaging (MRI) and Magnetic Resonance Spectroscopy (MRS) are often required for the diagnosis of tumours for which conclusive biopsies are not commonly available.While radiologists are used to interpretingMRI, many of them are not accustomed to make sense of the biochemical information provided by MRS. In this situation, oncology radiologists may benefit from the use of computer-based support in their decision making. As part of the AIDTumour research project, the analysis of MRS data corresponding to various tumour pathologies is used to assist expert diagnosis. The high dimensionality of the MR spectra might obscure atypical aspects of the data that would jeopardize their automated classification and, as a result, the process of computerbased diagnostic assistance. In this study, we put forward a method to overcome this potential problem that combines methods of visualization through non-linear dimensionality reduction, automatic outlier detection, and radiologists’ expert opinion., Postprint (published version)

[EN] We present an agent-based distributed decision support system for the diagnosis and prognosis of brain tumors developed by the HealthAgents project. HealthAgents is a European Union funded research project, which aims to enhance the classification of brain tumors using such a decision support system based on intelligent agents to securely connect a network of clinical centers. The HealthAgents system is implementing novel pattern recognition discrimination methods, in order to analyze in vivo Magnetic Resonance Spectroscopy (MRS) and ex vivo/in vitro High Resolution Magic Angle Spinning Nuclear Magnetic Resonance (HR-MAS) and DNA micro-array data. HealthAgents intends not only to apply forefront agent technology to the biomedical field, but also develop the HealthAgents network, a globally distributed information and knowledge repository for brain tumor diagnosis and prognosis.

[EN] We present an agent-based distributed decision support system for the diagnosis and prognosis of brain tumors developed by the HealthAgents project. HealthAgents is a European Union funded research project, which aims to enhance the classification of brain tumors using such a decision support system based on intelligent agents to securely connect a network of clinical centers. The HealthAgents system is implementing novel pattern recognition discrimination methods, in order to analyze in vivo Magnetic Resonance Spectroscopy (MRS) and ex vivo/in vitro High Resolution Magic Angle Spinning Nuclear Magnetic Resonance (HR-MAS) and DNA micro-array data. HealthAgents intends not only to apply forefront agent technology to the biomedical field, but also develop the HealthAgents network, a globally distributed information and knowledge repository for brain tumor diagnosis and prognosis.

[EN] H-1 MRS is becoming an accurate, non-invasive technique for initial examination of brain masses. We investigated if the combination of single-voxel H-1 MRS at 1.5 T at two different (TEs), short TE (PRESS or STEAM, 20-32 ms) and long TE (PRESS, 135-136 ms), improves the classification of brain tumors over using only one echo TE. A clinically validated dataset of 50 low-grade meningiomas, 105 aggressive tumors (glioblastoma and metastasis). and 30 low-grade glial tumors (astrocytomas grade II, oligodendrogliomas anti oligoastrocytomas) was used to fit predictive models based on the combination of features from short-TEs and long-TE spectra. A new approach that combines the two consecutively was used to produce a single data vector from which relevant features of the two TE spectra could be extracted by means of three algorithms: stepwise, reliefF, and principal components analysis. Least Squares support vector machines and linear discriminant analysis were applied to fit the pairwise and multiclass classifiers, respectively. Significant differences in performance were found when short-TE, long-TE or both spectra combined were used as input. In our dataset, to discriminate meningiomas, the combination of the two TE acquisitions produced optimal performance. To discriminate aggressive tumors from low-grade filial tumours, the use of short-TE acquisition alone was preferable. The classifier development strategy used here lends itself to automated learning and test performance processes, which may, be of use for future web-based multicentric classifier development studies. Copyright (c) 2008 John Wiley & Sons, Ltd.

[EN] H-1 MRS is becoming an accurate, non-invasive technique for initial examination of brain masses. We investigated if the combination of single-voxel H-1 MRS at 1.5 T at two different (TEs), short TE (PRESS or STEAM, 20-32 ms) and long TE (PRESS, 135-136 ms), improves the classification of brain tumors over using only one echo TE. A clinically validated dataset of 50 low-grade meningiomas, 105 aggressive tumors (glioblastoma and metastasis). and 30 low-grade glial tumors (astrocytomas grade II, oligodendrogliomas anti oligoastrocytomas) was used to fit predictive models based on the combination of features from short-TEs and long-TE spectra. A new approach that combines the two consecutively was used to produce a single data vector from which relevant features of the two TE spectra could be extracted by means of three algorithms: stepwise, reliefF, and principal components analysis. Least Squares support vector machines and linear discriminant analysis were applied to fit the pairwise and multiclass classifiers, respectively. Significant differences in performance were found when short-TE, long-TE or both spectra combined were used as input. In our dataset, to discriminate meningiomas, the combination of the two TE acquisitions produced optimal performance. To discriminate aggressive tumors from low-grade filial tumours, the use of short-TE acquisition alone was preferable. The classifier development strategy used here lends itself to automated learning and test performance processes, which may, be of use for future web-based multicentric classifier development studies. Copyright (c) 2008 John Wiley & Sons, Ltd.

As part of the AIDTumour research project, the analysis of MRS data corresponding to various tumour pathologies is used to assist expert diagnosis. The high dimensionality of the MR spectra might obscure atypical aspects of the data that would jeopardize their automated classification and, as a result, the process of computer-based diagnostic assistance. In this paper, we put forward a method to overcome this potential problem that combines automatic outlier detection, visualization through dimensionality reduction, and expert opinion., Postprint (published version)

This paper describes a process of rule-extraction from a multi-centre brain tumour database consisting of nuclear magnetic res- onance spectroscopic signals. The expert diagnosis of human brain tumours can benefit from computer-aided assistance, which has to be readily interpretable by clinicians. Interpretation can be achieved through rule extraction, which is here performed using the LR-FIR algorithm, a method based on fuzzy logic. The experimental results of the classiffication of three groups of tumours indicate in this study that just three spectral frequencies, out of the 195 from a range pre-selected by experts, are enough to represent, in a simple and intuitive manner, most of the knowledge required to discriminate these groups., Postprint (published version)

The combination of an Artificial Neural Network classifier, a feature selection process, and a novel linear dimensionality reduction technique that provides a data projection for visualization and which preserves completely the class discrimination achieved by the classifier, is applied in this study to the analysis of an international, multi-centre 1H-MRS database of brain tumors. This combination yields results that are both intuitively interpretable and very accurate. The method as a whole remains simple enough as to allow its easy integration in existing medical decision support systems., Peer Reviewed, Postprint (published version)

[Purpose]: Chemotherapy increases survival in breast cancer patients. Consequently, cerebral metastases have recently become a significant clinical problem, with an incidence of 30-40% among breast carcinoma patients. As this phenomenon cannot be studied longitudinally in humans, models which mimic brain metastasis are needed to investigate its pathogenesis. Such models may later be used in experimental therapeutic approaches. [Material and methods/results]: We report a model in which 69% of the animals (9/13 BALB/c nude mice) developed MR-detectable abnormal masses in the brain parenchyma within a 20 to 62-day time window post intra-carotid injection of 435-Br1 human cells. The masses detected in vivo were either single (7 animals) or multiple (2 animals). Longitudinal MR (MRI/MRS) studies and post-mortem histological data were correlated, revealing a total incidence of experimental brain metastases of 85% in the cases studied (11/13 animals). ADC maps perfectly differentiated edema and/or CSF areas from metastasis. Preliminary MRS data also revealed additional features: decrease in N-acetyl aspartate (NAA) was the first MRS-based marker of metastasis growth in the brain (micrometastasis); choline-containing compounds (Cho) rose and creatine (Cr) levels decreased as these lesions evolved, with mobile lipids and lactate also becoming visible. Furthermore, MRS pattern recognition-based analysis suggested that this approach may help to discriminate different growth stages. [Conclusions]: This study paves the way for further in vivo studies oriented towards detection of different tumor progression states and for improving treatment efficiency. © 2008 ESMRMB.

1H MRS is evolving into an invaluable tool for brain tumor classification in humans based on pattern recognition analysis, but there is still room for improvement. Here we propose a new approach: to challenge tumor metabolism in vivo by a defined perturbation, and study the induced changes in MRS pattern. For this we recorded single voxel 1H MR spectra from mice bearing a stereotactically induced GL261 grade IV brain glioma during a period of induced acute hyperglycemia. A total of 29 C57BL/6 mice were used. Single voxel spectra were acquired at 7 T with point resolved spectroscopy and TE of 12, 30 and 136 ms. Tumors were induced by stereotactic injection of 105 GL261cells in 17 mice. Hyperglycemia (up to 338±36 mg/dL glucose in the blood) was induced by intraperitoneal bolus injection. Maximal increases in glucose resonances of up to 2.4-fold were recorded from tumors in vivo. Our observations are in agreement with extracellular accumulation of glucose, which may suggest that glucose transport and/or metabolism are working close to their maximum capacity in GL261 tumors. The significant and specific MRS pattern changes observed when comparing euglycemia and hyperglycemia may be of use for future pattern-recognition studies of animal and human brain tumors by enhancing MRS-based discrimination between tumor types and grades. Copyright © 2007 John Wiley & Sons, Ltd.

A new iron-based T 1 contrast agent consisting of a complex of iron ions coordinated to phosphate and amine ligands (Fe(phos) in short) has been characterized by spectroscopic and magnetic measurements. NMR relaxation studies showed r 1 values to be dependent on the phosphate salt concentration, K2HPO4, present in the medium. r 1 reaches a maximum value of 2.5 mM-1 s-1 for measurements carried out at 7 T and 298 K. 31P MRS, Mössbauer spectroscopy and magnetic measurements of Fe(phos) solutions suggest paramagnetic Fe3+ ions present in the studied iron-phosphate complex. In vitro and in vivo toxicity experiments with C6 cells and CD1 mice, respectively, demonstrated lack of toxicity for Fe(phos) at the highest dose tested in the MRI experiments (12 mM iron for C6 cells and 0.32 mmol iron/kg for mice). Finally, T 1 weighted images of brain tumours in mice have shown positive contrast enhancement of Fe(phos) for tumour afflicted regions in the brain. © 2007 ESMRMB.