Bioinformatics for precision medicine in oncology: principles and application to the SHIVA clinical trial.

Precision medicine (PM) requires the delivery of individually adapted medical care based on the genetic characteristics of each patient and his/her tumor. The last decade witnessed the development of high-throughput technologies such as microarrays and next-generation sequencing which paved the way to PM in the field of oncology. While the cost of these technologies decreases, we are facing an exponential increase in the amount of data produced. Our ability to use this information in daily practice relies strongly on the availability of an efficient bioinformatics system that assists in the translation of knowledge from the bench towards molecular targeting and diagnosis. Clinical trials and routine diagnoses constitute different approaches, both requiring a strong bioinformatics environment capable of (i) warranting the integration and the traceability of data, (ii) ensuring the correct processing and analyses of genomic data, and (iii) applying well-defined and reproducible procedures for workflow management and decision-making. To address the issues, a seamless information system was developed at Institut Curie which facilitates the data integration and tracks in real-time the processing of individual samples. Moreover, computational pipelines were developed to identify reliably genomic alterations and mutations from the molecular profiles of each patient. After a rigorous quality control, a meaningful report is delivered to the clinicians and biologists for the therapeutic decision. The complete bioinformatics environment and the key points of its implementation are presented in the context of the SHIVA clinical trial, a multicentric randomized phase II trial comparing targeted therapy based on tumor molecular profiling versus conventional therapy in patients with refractory cancer. The numerous challenges faced in practice during the setting up and the conduct of this trial are discussed as an illustration of PM application.

Streamlined ion torrent PGM-based diagnostics: BRCA1 and BRCA2 genes as a model.

To meet challenges in terms of throughput and turnaround time, many diagnostic laboratories are shifting from Sanger sequencing to higher throughput next-generation sequencing (NGS) platforms. Bearing in mind that the performance and quality criteria expected from NGS in diagnostic or research settings are strikingly different, we have developed an Ion Torrent's PGM-based routine diagnostic procedure for BRCA1/2 sequencing. The procedure was first tested on a training set of 62 control samples, and then blindly validated on 77 samples in parallel with our routine technique. The training set was composed of difficult cases, for example, insertions and/or deletions of various sizes, large-scale rearrangements and, obviously, mutations occurring in homopolymer regions. We also compared two bioinformatic solutions in this diagnostic context, an in-house academic pipeline and the commercially available NextGene software (Softgenetics). NextGene analysis provided higher sensitivity, as four previously undetected single-nucleotide variations were found. Regarding specificity, an average of 1.5 confirmatory Sanger sequencings per patient was needed for complete BRCA1/2 screening. Large-scale rearrangements were identified by two distinct analyses, that is, bioinformatics and fragment analysis with electrophoresis profile comparison. Turnaround time was enhanced, as a series of 30 patients were sequenced by one technician, making the results available for the clinician in 10 working days following blood sampling. BRCA1/2 genes are a good model, representative of the difficulties commonly encountered in diagnostic settings, which is why we believe our findings are of interest for the whole community, and the pipeline described can be adapted by any user of PGM for diagnostic purposes.

Beta-catenin status in paediatric medulloblastomas: correlation of immunohistochemical expression with mutational status, genetic profiles, and clinical characteristics.

Medulloblastoma is the most frequent malignant paediatric brain tumour. The activation of the Wnt/beta-catenin pathway occurs in 10-15% of medulloblastomas and has been recently described as a marker for favourable patient outcome. We report a series of 72 paediatric medulloblastomas evaluated for beta-catenin protein expression, CTNNB1 mutations, and comparative genomic hybridization. Gene expression profiles were also available in a subset of 40 cases. Immunostaining of beta-catenin showed extensive nuclear staining (>50% of the tumour cells) in six cases and focal nuclear staining (<10% of cells) in three cases. The other cases either exhibited a signal strictly limited to the cytoplasm (58 cases) or were negative (five cases). CTNNB1 mutations were detected in all beta-catenin extensively nucleopositive cases. The expression profiles of these cases documented strong activation of the Wnt/beta-catenin pathway. Remarkably, five out of these six tumours showed a complete loss of chromosome 6. In contrast, cases with focal nuclear beta-catenin staining, as well as tumours with negative or cytoplasmic staining, never demonstrated CTNNB1 mutation, Wnt/beta-catenin pathway activation or chromosome 6 loss. Patients with extensive nuclear staining were significantly older at diagnosis and were in continuous complete remission after a mean follow-up of 75.7 months (range 27.5-121.2 months) from diagnosis. All three patients with focal nuclear staining of beta-catenin died within 36 months from diagnosis. Altogether, these data confirm and extend previous observations that CTNNB1-mutated tumours represent a distinct molecular subgroup of medulloblastomas with favourable outcome, indicating that therapy de-escalation should be considered. International consensus on the definition criteria of this distinct medulloblastoma subgroup should be achieved.

Anaplastic oligodendrogliomas with 1p19q codeletion have a proneural gene expression profile.

BACKGROUND: In high grade gliomas, 1p19q codeletion and EGFR amplification are mutually exclusive and predictive of dramatically different outcomes. We performed a microarray gene expression study of four high grade gliomas with 1p19q codeletion and nine with EGFR amplification, identified by CGH-array. RESULTS: The two groups of gliomas exhibited very different gene expression profiles and were consistently distinguished by unsupervised clustering analysis. One of the most striking differences was the expression of normal brain genes by oligodendrogliomas with 1p19q codeletion. These gliomas harbored a gene expression profile that partially resembled the gene expression of normal brain samples, whereas gliomas with EGFR amplification expressed many genes in common with glioblastoma cancer stem cells. The differences between the two types of gliomas and the expression of neuronal genes in gliomas with 1p19q codeletion were both validated in an independent series of 16 gliomas using real-time RT-PCR with a set of 22 genes differentiating the two groups of gliomas (AKR1C3, ATOH8, BMP2, C20orf42, CCNB1, CDK2, CHI3L1, CTTNBP2, DCX, EGFR, GALNT13, GBP1, IGFBP2, IQGAP1, L1CAM, NCAM1, NOG, OLIG2, PDPN, PLAT, POSTN, RNF135). Immunohistochemical study of the most differentially expressed neuronal gene, alpha-internexin, clearly differentiated the two groups of gliomas, with 1p19q codeletion gliomas showing specific staining in tumor cells. CONCLUSION: These findings provide evidence for neuronal differentiation in oligodendrogliomas with 1p19q codeletion and support the hypothesis that the cell of origin for gliomas with 1p19q codeletion could be a bi-potential progenitor cell, able to give rise to both neurons and oligodendrocytes.

VAMP: visualization and analysis of array-CGH, transcriptome and other molecular profiles.

MOTIVATION: Microarray-based CGH (Comparative Genomic Hybridization), transcriptome arrays and other large-scale genomic technologies are now routinely used to generate a vast amount of genomic profiles. Exploratory analysis of this data is crucial in helping to understand the data and to help form biological hypotheses. This step requires visualization of the data in a meaningful way to visualize the results and to perform first level analyses. RESULTS: We have developed a graphical user interface for visualization and first level analysis of molecular profiles. It is currently in use at the Institut Curie for cancer research projects involving CGH arrays, transcriptome arrays, SNP (single nucleotide polymorphism) arrays, loss of heterozygosity results (LOH), and Chromatin ImmunoPrecipitation arrays (ChIP chips). The interface offers the possibility of studying these different types of information in a consistent way. Several views are proposed, such as the classical CGH karyotype view or genome-wide multi-tumor comparison. Many functionalities for analyzing CGH data are provided by the interface, including looking for recurrent regions of alterations, confrontation to transcriptome data or clinical information, and clustering. Our tool consists of PHP scripts and of an applet written in Java. It can be run on public datasets at http://bioinfo.curie.fr/vamp AVAILABILITY: The VAMP software (Visualization and Analysis of array-CGH,transcriptome and other Molecular Profiles) is available upon request. It can be tested on public datasets at http://bioinfo.curie.fr/vamp. The documentation is available at http://bioinfo.curie.fr/vamp/doc.

ITTACA: a new database for integrated tumor transcriptome array and clinical data analysis.

Transcriptome microarrays have become one of the tools of choice for investigating the genes involved in tumorigenesis and tumor progression, as well as finding new biomarkers and gene expression signatures for the diagnosis and prognosis of cancer. Here, we describe a new database for Integrated Tumor Transcriptome Array and Clinical data Analysis (ITTACA). ITTACA centralizes public datasets containing both gene expression and clinical data. ITTACA currently focuses on the types of cancer that are of particular interest to research teams at Institut Curie: breast carcinoma, bladder carcinoma and uveal melanoma. A web interface allows users to carry out different class comparison analyses, including the comparison of expression distribution profiles, tests for differential expression and patient survival analyses. ITTACA is complementary to other databases, such as GEO and SMD, because it offers a better integration of clinical data and different functionalities. It also offers more options for class comparison analyses when compared with similar projects such as Oncomine. For example, users can define their own patient groups according to clinical data or gene expression levels. This added flexibility and the user-friendly web interface makes ITTACA especially useful for comparing personal results with the results in the existing literature. ITTACA is accessible online at http://bioinfo.curie.fr/ittaca.

Inhibition of human bladder tumour cell growth by fibroblast growth factor receptor 2b is independent of its kinase activity. Involvement of the carboxy-terminal region of the receptor.

The b isoform of fibroblast growth factor receptor 2, FGFR2b/FGFR2-IIIb/Ksam-IIC1/KGFR, a tyrosine kinase receptor, is expressed in a wide variety of epithelia and is downregulated in several human carcinomas including prostate, salivary and urothelial cell carcinomas. FGFR2b has been shown to inhibit growth in tumour cell lines derived from these carcinomas. Here, we investigated the molecular mechanisms underlying the inhibition of human urothelial carcinoma cell growth following FGFR2b expression. Using a nylon DNA array, we analysed the gene expression profile of the T24 bladder tumour cell line, transfected or not with a construct encoding FGFR2b. The expression of FGFR2b in T24 cells decreased insulin-like growth factor (IGF)-II mRNA levels. This decrease was correlated with a decrease in IGF-II secretion and may have been responsible for the observed inhibition of cell growth because the addition of exogenous IGF-II restored growth rates to normal levels. Using SU5402, an inhibitor of FGFR tyrosine kinase activity, and a kinase dead mutant of the receptor, FGFR2b Y659F/Y660F, we also demonstrated that the growth inhibition and decrease in IGF-II secretion induced by FGFR2b did not require tyrosine kinase activity. Finally, we demonstrated the involvement of the distal carboxy-terminal domain of the receptor in decreasing IGF-II expression and inhibiting T24 cell growth, as Ksam-IIC3, a variant of FGFR2b carrying a short carboxy-terminus, neither downregulated IGF-II nor inhibited cell proliferation. Our data suggest that FGFR2b inhibits the growth of bladder carcinoma cells by reducing IGF-II levels via its carboxy-terminal domain, independent of its tyrosine kinase activity.

Selecting biomedical data sources according to user preferences.

MOTIVATION: Biologists are now faced with the problem of integrating information from multiple heterogeneous public sources with their own experimental data contained in individual sources. The selection of the sources to be considered is thus critically important. RESULTS: Our aim is to support biologists by developing a module based on an algorithm that presents a selection of sources relevant to their query and matched to their own preferences. We approached this task by investigating the characteristics of biomedical data and introducing several preference criteria useful for bioinformaticians. This work was carried out in the framework of a project which aims to develop an integrative platform for the multiple parametric analysis of cancer. We illustrate our study through an elementary biomedical query occurring in a CGH analysis scenario. AVAILABILITY: http://www.lri.fr/~cohen/dss/dss.html