Democratizing data at Novartis through clinical trial data access.

Enabling broad access and usage of clinical trial data within biopharmaceutical companies has historically been impeded by technical, cultural, and policy hurdles. Novartis has attempted to address this comprehensively through a program called data42; here, we explore how a diverse set of enterprise-wide stakeholders formulated a risk-based data access approach to streamline access to anonymized clinical trial data and vastly improved its use by authorized research and development (R&D) associates within the company. The result is that most Novartis clinical trial data requests, from internal associates, can now be automatically approved. The process of developing this framework and its impact on Novartis and the broader industry are explored and discussed.

Correction of: Exploring Concordance of Patient-Reported Information on PatientsLikeMe and Medical Claims Data at the Patient Level.

[This corrects the article DOI: 10.2196/jmir.5130.].

Exploring Concordance of Patient-Reported Information on PatientsLikeMe and Medical Claims Data at the Patient Level.

BACKGROUND: With the emergence of data generated by patient-powered research networks, it is informative to characterize their correspondence with health care system-generated data. OBJECTIVES: This study explored the linking of 2 disparate sources of real-world data: patient-reported data from a patient-powered research network (PatientsLikeMe) and insurance claims. METHODS: Active patients within the PatientsLikeMe community, residing in the United States, aged 18 years or older, with a self-reported diagnosis of multiple sclerosis or Parkinson's disease (PD) were invited to participate during a 2-week period in December 2014. Patient-reported data were anonymously matched and compared to IMS Health medical and pharmacy claims data with dates of service between December 2009 and December 2014. Patient-level match (identity), diagnosis, and usage of disease-modifying therapies (DMTs) were compared between data sources. RESULTS: Among 603 consenting patients, 94% had at least 1 record in the IMS Health dataset; of these, there was 93% agreement rate for multiple sclerosis diagnosis. Concordance on the use of any treatment was 59%, and agreement on reports of specific treatment usage (within an imputed 5-year period) ranged from 73.5% to 100%. CONCLUSIONS: It is possible to match patient identities between the 2 data sources, and the high concordance at multiple levels suggests that the matching process was accurate. Likewise, the high degree of concordance suggests that these patients were able to accurately self-report their diagnosis and, to a lesser degree, their treatment usage. Further studies of linked data types are warranted to evaluate the use of enriched datasets to generate novel insights.

Bioinformatics/biostatistics: microarray analysis.

The quantity and complexity of the molecular-level data generated in both research and clinical settings require the use of sophisticated, powerful computational interpretation techniques. It is for this reason that bioinformatic analysis of complex molecular profiling data has become a fundamental technology in the development of personalized medicine. This chapter provides a high-level overview of the field of bioinformatics and outlines several, classic bioinformatic approaches. The highlighted approaches can be aptly applied to nearly any sort of high-dimensional genomic, proteomic, or metabolomic experiments. Reviewed technologies in this chapter include traditional clustering analysis, the Gene Expression Dynamics Inspector (GEDI), GoMiner (GoMiner), Gene Set Enrichment Analysis (GSEA), and the Learner of Functional Enrichment (LeFE).

Metabolomics reveals attenuation of the SLC6A20 kidney transporter in nonhuman primate and mouse models of type 2 diabetes mellitus.

To enhance understanding of the metabolic indicators of type 2 diabetes mellitus (T2DM) disease pathogenesis and progression, the urinary metabolomes of well characterized rhesus macaques (normal or spontaneously and naturally diabetic) were examined. High-resolution ultra-performance liquid chromatography coupled with the accurate mass determination of time-of-flight mass spectrometry was used to analyze spot urine samples from normal (n = 10) and T2DM (n = 11) male monkeys. The machine-learning algorithm random forests classified urine samples as either from normal or T2DM monkeys. The metabolites important for developing the classifier were further examined for their biological significance. Random forests models had a misclassification error of less than 5%. Metabolites were identified based on accurate masses (<10 ppm) and confirmed by tandem mass spectrometry of authentic compounds. Urinary compounds significantly increased (p < 0.05) in the T2DM when compared with the normal group included glycine betaine (9-fold), citric acid (2.8-fold), kynurenic acid (1.8-fold), glucose (68-fold), and pipecolic acid (6.5-fold). When compared with the conventional definition of T2DM, the metabolites were also useful in defining the T2DM condition, and the urinary elevations in glycine betaine and pipecolic acid (as well as proline) indicated defective re-absorption in the kidney proximal tubules by SLC6A20, a Na(+)-dependent transporter. The mRNA levels of SLC6A20 were significantly reduced in the kidneys of monkeys with T2DM. These observations were validated in the db/db mouse model of T2DM. This study provides convincing evidence of the power of metabolomics for identifying functional changes at many levels in the omics pipeline.

Inhibition of histone deacetylase in cancer cells slows down replication forks, activates dormant origins, and induces DNA damage.

Protein acetylation is a reversible process regulated by histone deacetylases (HDAC) that is often altered in human cancers. Suberoylanilide hydroxamic acid (SAHA) is the first HDAC inhibitor to be approved for clinical use as an anticancer agent. Given that histone acetylation is a key determinant of chromatin structure, we investigated how SAHA may affect DNA replication and integrity to gain deeper insights into the basis for its anticancer activity. Nuclear replication factories were visualized with confocal immunofluorescence microscopy and single-replicon analyses were conducted by genome-wide molecular combing after pulse labeling with two thymidine analogues. We found that pharmacologic concentrations of SAHA induce replication-mediated DNA damage with activation of histone gammaH2AX. Single DNA molecule analyses indicated slowdown in replication speed along with activation of dormant replication origins in response to SAHA. Similar results were obtained using siRNA-mediated depletion of HDAC3 expression, implicating this HDAC member as a likely target in the SAHA response. Activation of dormant origins was confirmed by molecular analyses of the beta-globin locus control region. Our findings demonstrate that SAHA produces profound alterations in DNA replication that cause DNA damage, establishing a critical link between robust chromatin acetylation and DNA replication in human cancer cells.

UPLC-ESI-TOFMS-based metabolomics and gene expression dynamics inspector self-organizing metabolomic maps as tools for understanding the cellular response to ionizing radiation.

Global transcriptomic and proteomic profiling platforms have yielded important insights into the complex response to ionizing radiation (IR). Nonetheless, little is known about the ways in which small cellular metabolite concentrations change in response to IR. Here, a metabolomics approach using ultraperformance liquid chromatography coupled with electrospray time-of-flight mass spectrometry was used to profile, over time, the hydrophilic metabolome of TK6 cells exposed to IR doses ranging from 0.5 to 8.0 Gy. Multivariate data analysis of the positive ions revealed dose- and time-dependent clustering of the irradiated cells and identified certain constituents of the water-soluble metabolome as being significantly depleted as early as 1 h after IR. Tandem mass spectrometry was used to confirm metabolite identity. Many of the depleted metabolites are associated with oxidative stress and DNA repair pathways. Included are reduced glutathione, adenosine monophosphate, nicotinamide adenine dinucleotide, and spermine. Similar measurements were performed with a transformed fibroblast cell line, BJ, and it was found that a subset of the identified TK6 metabolites were effective in IR dose discrimination. The GEDI (Gene Expression Dynamics Inspector) algorithm, which is based on self-organizing maps, was used to visualize dynamic global changes in the TK6 metabolome that resulted from IR. It revealed dose-dependent clustering of ions sharing the same trends in concentration change across radiation doses. "Radiation metabolomics," the application of metabolomic analysis to the field of radiobiology, promises to increase our understanding of cellular responses to stressors such as radiation.

The LeFE algorithm: embracing the complexity of gene expression in the interpretation of microarray data.

Interpretation of microarray data remains a challenge, and most methods fail to consider the complex, nonlinear regulation of gene expression. To address that limitation, we introduce Learner of Functional Enrichment (LeFE), a statistical/machine learning algorithm based on Random Forest, and demonstrate it on several diverse datasets: smoker/never smoker, breast cancer classification, and cancer drug sensitivity. We also compare it with previously published algorithms, including Gene Set Enrichment Analysis. LeFE regularly identifies statistically significant functional themes consistent with known biology.

Phosphoprotein pathway mapping: Akt/mammalian target of rapamycin activation is negatively associated with childhood rhabdomyosarcoma survival.

Mapping of protein signaling networks within tumors can identify new targets for therapy and provide a means to stratify patients for individualized therapy. Despite advances in combination chemotherapy, the overall survival for childhood rhabdomyosarcoma remains approximately 60%. A critical goal is to identify functionally important protein signaling defects associated with treatment failure for the 40% nonresponder cohort. Here, we show, by phosphoproteomic network analysis of microdissected tumor cells, that interlinked components of the Akt/mammalian target of rapamycin (mTOR) pathway exhibited increased levels of phosphorylation for tumors of patients with short-term survival. Specimens (n = 59) were obtained from the Children's Oncology Group Intergroup Rhabdomyosarcoma Study (IRS) IV, D9502 and D9803, with 12-year follow-up. High phosphorylation levels were associated with poor overall and poor disease-free survival: Akt Ser(473) (overall survival P < 0.001, recurrence-free survival P < 0.0009), 4EBP1 Thr(37/46) (overall survival P < 0.0110, recurrence-free survival P < 0.0106), eIF4G Ser(1108) (overall survival P < 0.0017, recurrence-free survival P < 0.0072), and p70S6 Thr(389) (overall survival P < 0.0085, recurrence-free survival P < 0.0296). Moreover, the findings support an altered interrelationship between the insulin receptor substrate (IRS-1) and Akt/mTOR pathway proteins (P < 0.0027) for tumors from patients with poor survival. The functional significance of this pathway was tested using CCI-779 in a mouse xenograft model. CCI-779 suppressed phosphorylation of mTOR downstream proteins and greatly reduced the growth of two different rhabdomyosarcoma (RD embryonal P = 0.00008; Rh30 alveolar P = 0.0002) cell lines compared with controls. These results suggest that phosphoprotein mapping of the Akt/mTOR pathway should be studied further as a means to select patients to receive mTOR/IRS pathway inhibitors before administration of chemotherapy.

Asparagine synthetase as a causal, predictive biomarker for L-asparaginase activity in ovarian cancer cells.

L-Asparaginase (l-ASP), a bacterial enzyme used since the 1970s to treat acute lymphoblastic leukemia, selectively starves cells that cannot synthesize sufficient asparagine for their own needs. Molecular profiling of the NCI-60 cancer cell lines using five different microarray platforms showed strong negative correlations of asparagine synthetase (ASNS) expression and DNA copy number with sensitivity to l-ASP in the leukemia and ovarian cancer cell subsets. To assess whether the ovarian relationship is causal, we used RNA interference to silence ASNS in three ovarian lines and observed 4- to 5-fold potentiation of sensitivity to l-ASP with two of the lines. For OVCAR-8, the line that expresses the least ASNS, the potentiation was >500-fold. Significantly, that potentiation was >700-fold in the multidrug-resistant derivative OVCAR-8/ADR, showing that the causal relationship between ASNS expression and l-ASP activity survives development of classical multidrug resistance. Tissue microarrays confirmed low ASNS expression in a subset of clinical ovarian cancers as well as other tumor types. Overall, this pharmacogenomic/pharmacoproteomic study suggests the use of l-ASP for treatment of a subset of ovarian cancers (and perhaps other tumor types), with ASNS as a biomarker for patient selection.

Analysis of fracture healing by large-scale transcriptional profile identified temporal relationships between metalloproteinase and ADAMTS mRNA expression.

The aim of this study was to validate the use of transcriptional profiling as a means of characterizing the complex interactions of the thousands of genes that are expressed during fracture healing. Standard mid-diaphyseal tibia fractures were generated in C57/B6 murine tibiae and the transcriptional expression of approximately 13,000 genes was assessed. Three time points after fracture were assessed: day 3, representative of the inflammatory phase; day 10, representative of the peak of cartilage formation; and day 21, representative of the period of primary bone formation and coupled remodeling. A self-organizing mapping approach of the data revealed the temporal relationships between the expression of mRNAs for extracellular matrix proteins and the proteases that degrade the proteoglycan and collagenous matrices. A broad group of extracellular matrix protein mRNAs representative of basement membranes, blood vessels and cartilage all showed elevated expression over the first 21 days of fracture healing. The sorting of the data identified an orderly temporal expression of the metalloproteinases and ADAMTS during the progression of fracture healing with (MMP2/MMP14/TIMP2) and ADAMTS4 and 15 preceding the expression of (MMP9/MMP13). Based on their patterns of expression, relative to the known activities of the encoded proteolytic enzymes, our results suggest that the dissolution of cartilage protoeglycans proceeds before the underlying collagenous components of the matrix are removed. The exclusion of several mRNAs that are normally expressed by osteoclasts in the profiles of mRNAs from days 3 and 10 suggests that osteoclastic activity was largely absent during the early periods of cartilage tissue formation and that proteoglycan and specific collagenase activities, precedes or is prerequisite to later osteoclast infiltration into the remodeling tissues.

Comparison of methods for sequential screening of large compound sets.

Sequential screening is an iterative procedure that can greatly increase hit rates over random screening or non-iterative procedures. We studied the effects of three factors on enrichment rates: the method used to rank compounds, the molecular descriptor set and the selection of initial training set. The primary factor influencing recovery rates was the method of selecting the initial training set. Rates for recovering active compounds were substantially lower with the diverse training sets than they were with training sets selected by other methods. Because structure-activity information is incrementally enhanced in intermediate training sets, sequential screening provides significant improvement in the average rate of recovery of active compounds when compared with non-iterative selection procedures.

Towards a holistic, yet gene-centered analysis of gene expression profiles: a case study of human lung cancers.

Genome-wide gene expression profile studies encompass increasingly large number of samples, posing a challenge to their presentation and interpretation without losing the notion that each transcriptome constitutes a complex biological entity. Much like pathologists who visually analyze information-rich histological sections as a whole, we propose here an integrative approach. We use a self-organizing maps-based software, the gene expression dynamics inspector (GEDI) to analyze gene expression profiles of various lung tumors. GEDI allows the comparison of tumor profiles based on direct visual detection of transcriptome patterns. Such intuitive "gestalt" perception promotes the discovery of interesting relationships in the absence of an existing hypothesis. We uncovered qualitative relationships between squamous cell tumors, small-cell tumors, and carcinoid tumor that would have escaped existing algorithmic classifications. These results suggest that GEDI may be a valuable explorative tool that combines global and gene-centered analyses of molecular profiles from large-scale microarray experiments.

Evidence by molecular profiling for a placental origin of infantile hemangioma.

The origin of the pathogenic endothelial cells in common infantile hemangioma is unknown. We show here that the transcriptomes of human placenta and infantile hemangioma are sufficiently similar to suggest a placental origin for this tumor, expanding on recent immunophenotypical studies that have suggested this possibility [North, P. E., et al. (2001) Arch. Dermatol. 137, 559-570]. The transcriptomes of placenta, hemangioma, and eight normal and diseased tissues were compared by hierarchical and nonhierarchical clustering analysis of >7,800 genes. We found that the level of transcriptome similarity between placenta and hemangioma exceeded that of any other tissue compared and paralleled that observed between a given tissue and its derived tumor, such as normal and cancerous lung. The degree of similarity was even greater when a subset of endothelial cell-specific genes was analyzed. Genes preferentially expressed in both placenta and hemangiomas were identified, including 17-beta hydroxysteroid dehydrogenase type 2 and tissue factor pathway inhibitor 2. These data demonstrate the value of global molecular profiling of tissues as a tool for hypothesis-driven research. Furthermore, it suggests that the unique self-limited growth of infantile hemangioma may, in fact, mirror the lifetime of placental endothelium.

Extracellular matrix controls myosin light chain phosphorylation and cell contractility through modulation of cell shape and cytoskeletal prestress.

The mechanism by which vascular smooth muscle (VSM) cells modulate their contractility in response to structural cues from extracellular matrix remains poorly understood. When pulmonary VSM cells were cultured on increasing densities of immobilized fibronectin (FN), cell spreading, myosin light chain (MLC) phosphorylation, cytoskeletal prestress (isometric tension in the cell before vasoagonist stimulation), and the active contractile response to the vasoconstrictor endothelin-1 all increased in parallel. In contrast, MLC phosphorylation did not increase when suspended cells were allowed to bind FN-coated microbeads (4.5-microm diameter) or cultured on micrometer-sized (30 x 30 microm) FN islands surrounded by nonadhesive regions that support integrin binding but prevent cell spreading. Cell spreading and MLC phosphorylation also both decreased in parallel when the mechanical compliance of flexible FN substrates was raised. MLC phosphorylation was inhibited independently of cell shape when cytoskeletal prestress was dissipated using a myosin ATPase inhibitor in fully spread cells, whereas it increased to maximal levels when microtubules were disrupted using nocodazole in cells adherent to FN but not in suspended cells. These data demonstrate that changes in cell-extracellular matrix (ECM) interactions modulate smooth muscle cell contractility at the level of biochemical signal transduction and suggest that the mechanism underlying this regulation may involve physical interplay between ECM and the cytoskeleton, such that cell spreading and generation of cytoskeletal tension feed back to promote MLC phosphorylation and further increase tension generation.

Gene Expression Dynamics Inspector (GEDI): for integrative analysis of expression profiles.

UNLABELLED: Genome-wide expression profiles contain global patterns that evade visual detection in current gene clustering analysis. Here, a Gene Expression Dynamics Inspector (GEDI) is described that uses self-organizing maps to translate high-dimensional expression profiles of time courses or sample classes into animated, coherent and robust mosaics images. GEDI facilitates identification of interesting patterns of molecular activity simultaneously across gene, time and sample space without prior assumption of any structure in the data, and then permits the user to retrieve genes of interest. Important changes in genome-wide activities may be quickly identified based on 'Gestalt' recognition and hence, GEDI may be especially useful for non-specialist end users, such as physicians. AVAILABILITY: GEDI v1.0 is written in Matlab, and binary Matlab.dll files which require Matlab to run can be downloaded for free by academic institutions at http://www.chip.org/~ge/gedihome.html SUPPLEMENTARY INFORMATION: http://www.chip.org/~ge/gedihome.html

Caldesmon-dependent switching between capillary endothelial cell growth and apoptosis through modulation of cell shape and contractility.

Caldesmon (CaD), a protein component of the actomyosin filament apparatus, modulates cell shape and cytoskeletal structure when overexpressed. When capillary endothelial cells were infected with an adenoviral vector encoding GFP-CaD under Tet-Off control, progressive inhibition of contractility, loss of actin stress fibers, disassembly of focal adhesions, and cell retraction resulted. This was accompanied by a cell shape (rounding)-dependent increase in apoptosis and concomitant inhibition of cell cycle progression. Cell growth also was inhibited in low expressor cells in which cell tension was suppressed independently of significant changes in cell shape, cytoskeletal structure, or focal adhesions. Thus, changes in both cytoskeletal structure and contractility appear to be central to the mechanism by which extracellular matrix-dependent changes in capillary cell shape influence growth and apoptosis during angiogenesis, and hence the cytoskeleton may represent a potential target for anti-angiogenesis therapy.