Time filter

Source Type

Cekaite L.,University of Oslo | Rantala J.K.,Oregon Health And Science University | Bruun J.,University of Oslo | Guriby M.,University of Oslo | And 8 more authors.
Neoplasia (United States) | Year: 2012

Several microRNAs (miRNAs) are known to be deregulated in colon cancer, but the mechanisms behind their potential involvement on proliferation and tumor cell survival are unclear. The present study aimed to identify miRNAs with functional implications for development of colon cancer. The cell proliferation and apoptosis were examined following perturbations of miRNA levels by employing a comprehensive miRNA library screen. miRNAs nominated for relevance to colon cancer were validated on expression and functional levels. By integrating the effect of miRNA up-regulation with the endogenous miRNA expression levels within the HT29, HCT116, and SW480 colon cancer cell lines, we identified miRNAs controlling cell proliferation (n = 53) and apoptosis (n = 93). From these functionally nominatedmiRNAs, we narrowed the list to 10 oncogene- and 20 tumor suppressor-like miRNAs that were also differentially expressed between colon cancer (n = 80) and normal colonic mucosa (n = 20). The differential expressions of miR-9, miR-31, and miR-182 were successfully validated in a series of colon carcinomas (n = 30) and polyps (n = 10) versus normal colonic mucosa (n = 10), whereas the functional effect was confirmed in an in-depth validation using different cell viability and apoptotic markers. Several transcription factors and genes regulating cell proliferation were identified as putative target genes by integrative miRNA/mRNA expression analysis obtained from the same colon cancer patient samples. This study suggests that deregulated expression of miR-9, miR-31, and miR-182 during carcinogenesis plays a significant role in the development of colon cancer by promoting proliferation and tumor cell survival. © 2012 Neoplasia Press, Inc. All rights reserved.


News Article | November 16, 2016
Site: www.prweb.com

Knowing the likely course of cancer can influence treatment decisions. Now a new prediction model published today in Lancet Oncology offers a more accurate prognosis for a patient's metastatic castration-resistant prostate cancer. The approach was as novel as the result – while researchers commonly work in small groups, intentionally isolating their data, the current study embraces the call in Joe Biden's "Cancer Moonshot" to open their question and their data, collecting previously published clinical trial data and calling for worldwide collaboration to evaluate its predictive power. That is, researchers crowdsourced the question of prostate cancer prognosis, eventually involving over 550 international researchers and resulting in 50 computational models from 50 different teams. The approach was intentionally controversial. "Scientists like me who mine open data have been called 'research parasites'. While not the most flattering name, the idea of leveraging existing data to gain new insights is a very important part of modern biomedical research. This project shows the power of the parasites," says James Costello, PhD, senior author of the paper, investigator at the University of Colorado Cancer Center, assistant professor in the Department of Pharmacology at the CU School of Medicine, and director of Computational and Systems Biology Challenges within the Sage Bionetworks/DREAM organization. The project was overseen as a collaborative effort between 16 institutions, led by academic research institutions including CU Cancer Center, open-data initiatives including Project Data Sphere, Sage Bionetworks, and the National Cancer Institute's DREAM Challenges, and industry and research partners including Sanofi, AstraZeneca, and the Prostate Cancer Foundation. Challenge organizers made available the results from five completed clinical trials. Teams were challenged to connect a deep set of clinical measurements to overall patient survival, organizing their insights into novel computational models to better predict patient survival based on clinical data. "The idea is that if a patient comes into the clinic and has these measurements and test results, can we put this data in a model to say if this patient will progress slowly or quickly. If we know the features of patients at the greatest risk, we can know who should receive standard treatment and who might benefit more from a clinical trial," Costello says. The most successful of the 50 models was submitted by a team led by Tero Aittokallio, PhD, from the Institute for Molecular Medicine Finland, FIMM, at University of Helsinki, and professor in the Department of Mathematics and Statistics at University of Turku, Finland. "My group has a long-term expertise in developing multivariate machine learning models for various biomedical applications, but this Challenge provided the unique opportunity to work on clinical trial data, with the eventual aim to help patients with metastatic castration-resistant prostate cancer," Aittokallio says. Basically, the model depended on not only groups of single patient measurements to predict outcomes, but on exploring which interactions between measurements were most predictive – for example, data describing a patient's blood system composition and immune function were only weakly predictive of survival on their own, but when combined became an important part of the winning model. The model used a computational learning strategy technically referred to as an ensemble of penalized Cox regression models, hence the model's name ePCR. This model then competed with 49 other entries, submitted by other teams working independently around the world. "Having 50 independent models allowed us to do two very important things. First when a single clinical feature known to be predictive of patient survival is picked out by 40 of the 50 teams, this greatly strengthens our overall confidence. Second, we were able to discover important clinical features we hadn't fully appreciated before," Costello says. In this case, many models found that in addition to factors like prostate-specific antigen (PSA) and lactate dehydrogenase (LDH) that have long been known to predict prostate cancer performance, blood levels of an enzyme called asparate aminotransferease (AST) is an important predictor of patient survival. This AST is an indirect measure of liver function and the fact that disturbed levels of AST are associated with poor patient performance implies that studies could evaluate the role of AST in prostate cancer. "The benefits of a DREAM Challenge are the ability to attract talented individuals and teams from around the world, and a rigorous framework for the assessment of methods. These two ingredients came together for our Challenge, leading to a new benchmark in metastatic prostate cancer," says paper first author, Justin Guinney, PhD, director of Computational Oncology for Sage Bionetworks located at Fred Hutchinson Cancer Research Center. “A goal of the Project Data Sphere initiative is to spark innovation – to unlock the potential of valuable data by generating new insights and opening up a new world of research possibilities. Prostate Cancer DREAM Challenge did just that. To witness cancer clinical trial data from Project Data Sphere be used in research collaboration and ultimately help improve patient care in the future is extremely rewarding!” says Liz Zhou, MD, MS, director of Global Health Outcome Research at Sanofi. The goal now is to make the ePCR model publicly accessible through an online tool with an eye towards clinical application. In fact, the National Cancer Institute (NCI) has contracted the winning team to do exactly this. Soon, when patients face difficult decisions about the best treatment for metastatic castration-resistant prostate cancer, ePCR tool could be an important piece of the decision-making process. Challenge winners and results can be found on the Prostate Cancer DREAM Challenge homepage. The clinical trial data can be found at Project Data Sphere. The research article describing this work can be found at The Lancet Oncology. Additional papers that describe individual team methods can be found in the DREAM Channel at F1000Research. About University of Colorado Cancer Center The University of Colorado Cancer Center, located at the Anschutz Medical Campus, is Colorado’s only National Cancer Institute-designated comprehensive cancer center, a distinction recognizing its outstanding contributions to research, clinical trials, prevention and cancer control. CU Cancer Center’s clinical partner University of Colorado Hospital is ranked 15th by US News and World Report for Cancer and the CU Cancer Center is a member of the prestigious National Comprehensive Cancer Network®, an alliance of the nation’s leading cancer centers working to establish and deliver the gold standard in cancer clinical guidelines. CU Cancer Center is a consortium of more than 400 researchers and physicians at three state universities and three institutions, all working toward one goal: Translating science into life. For more information visit Coloradocancercenter.org and follow CU Cancer Center on Facebook and Twitter. About the DREAM Challenges Initiative Founded in 2006 by A. Califano (Columbia University) and Gustavo Stolovitzky (IBM Research) the Dialogue on Reverse Engineering Assessment and Methods (DREAM) Challenges Initiative poses fundamental questions about systems biology and translational medicine. Designed and run by a community of researchers from a variety of organizations, the DREAM challenges invite participants to propose solutions — fostering collaboration and building communities in the process. Expertise and institutional support are provided by Sage Bionetworks, along with the infrastructure to host challenges via their Synapse platform. Together, the leaders of the DREAM Challenges Initiative share a vision allowing individuals and groups to collaborate openly so that the “wisdom of the crowd” provides the greatest impact on science and human health. More information is available at: http://dreamchallenges.org/. About the Project Data Sphere Initiative Project Data Sphere, LLC, an independent, not-for-profit initiative of the CEO Roundtable on Cancer's Life Sciences Consortium (LSC), operates the Project Data Sphere® platform (http://www.ProjectDataSphere.org). Launched in April 2014, the Project Data Sphere platform provides one place where the cancer community can broadly share, integrate, analyze and discuss historical patient-level comparator arm data sets (historical patient-level cancer phase III) from multiple providers, with the goal of advancing research. With its broad-access approach, the initiative brings diverse minds and technology together to help unleash the full potential of existing clinical trial data and speed innovation by generating collective insights that may lead to improved trial design, disease modeling and beyond. The platform currently contains 27,600 patient lives of data; 9,400 of those are across a wide spectrum of prostate cancer populations. In order to ensure that researchers can realize the full potential of this data, PDS teamed with CEO Roundtable on Cancer Member, SAS Institute Inc. SAS, a leader in data and health analytics, developed and hosts the site and provides free state-of-the-art analytic tools to authorized users within the Project Data Sphere environment. About Sage Bionetworks Sage Bionetworks is a nonprofit biomedical research organization, founded in 2009, with a vision to promote innovations in personalized medicine by enabling a community-based approach to scientific inquiries and discoveries. Sage Bionetworks strives to activate patients and to incentivize scientists, funders and researchers to work in fundamentally new ways in order to shape research, accelerate access to knowledge and transform human health. It is located on the campus of the Fred Hutchinson Cancer Research Center in Seattle, Washington and is supported through a portfolio of philanthropic donations, competitive research grants, and commercial partnerships. More information is available at http://www.sagebase.org.


News Article | November 15, 2016
Site: www.eurekalert.org

Knowing the likely course of cancer can influence treatment decisions. Now a new prediction model published today in Lancet Oncology offers a more accurate prognosis for a patient's metastatic castration-resistant prostate cancer. The approach was as novel as the result - while researchers commonly work in small groups, intentionally isolating their data, the current study embraces the call in Joe Biden's "Cancer Moonshot" to open their question and their data, collecting previously published clinical trial data and calling for worldwide collaboration to evaluate its predictive power. That is, researchers crowdsourced the question of prostate cancer prognosis, eventually involving over 550 international researchers and resulting in 50 computational models from 50 different teams. The approach was intentionally controversial. "Scientists like me who mine open data have been called 'research parasites'. While not the most flattering name, the idea of leveraging existing data to gain new insights is a very important part of modern biomedical research. This project shows the power of the parasites," says James Costello, PhD, senior author of the paper, investigator at the University of Colorado Cancer Center, assistant professor in the Department of Pharmacology at the CU School of Medicine, and director of Computational and Systems Biology Challenges within the Sage Bionetworks/DREAM organization. The project was overseen as a collaborative effort between 16 institutions, led by academic research institutions including CU Cancer Center, open-data initiatives including Project Data Sphere, Sage Bionetworks, and the National Cancer Institute's DREAM Challenges, and industry and research partners including Sanofi, AstraZeneca, and the Prostate Cancer Foundation. Challenge organizers made available the results from five completed clinical trials. Teams were challenged to connect a deep set of clinical measurements to overall patient survival, organizing their insights into novel computational models to better predict patient survival based on clinical data. "The idea is that if a patient comes into the clinic and has these measurements and test results, can we put this data in a model to say if this patient will progress slowly or quickly. If we know the features of patients at the greatest risk, we can know who should receive standard treatment and who might benefit more from a clinical trial," Costello says. The most successful of the 50 models was submitted by a team led by Tero Aittokallio, PhD, from the Institute for Molecular Medicine Finland, FIMM, at University of Helsinki, and professor in the Department of Mathematics and Statistics at University of Turku, Finland. "My group has a long-term expertise in developing multivariate machine learning models for various biomedical applications, but this Challenge provided the unique opportunity to work on clinical trial data, with the eventual aim to help patients with metastatic castration-resistant prostate cancer," Aittokallio says. Basically, the model depended on not only groups of single patient measurements to predict outcomes, but on exploring which interactions between measurements were most predictive - for example, data describing a patient's blood system composition and immune function were only weakly predictive of survival on their own, but when combined became an important part of the winning model. The model used a computational learning strategy technically referred to as an ensemble of penalized Cox regression models, hence the model's name ePCR. This model then competed with 49 other entries, submitted by other teams working independently around the world. "Having 50 independent models allowed us to do two very important things. First when a single clinical feature known to be predictive of patient survival is picked out by 40 of the 50 teams, this greatly strengthens our overall confidence. Second, we were able to discover important clinical features we hadn't fully appreciated before," Costello says. In this case, many models found that in addition to factors like prostate-specific antigen (PSA) and lactate dehydrogenase (LDH) that have long been known to predict prostate cancer performance, blood levels of an enzyme called asparate aminotransferease (AST) is an important predictor of patient survival. This AST is an indirect measure of liver function and the fact that disturbed levels of AST are associated with poor patient performance implies that studies could evaluate the role of AST in prostate cancer. "The benefits of a DREAM Challenge are the ability to attract talented individuals and teams from around the world, and a rigorous framework for the assessment of methods. These two ingredients came together for our Challenge, leading to a new benchmark in metastatic prostate cancer," says paper first author, Justin Guinney, PhD, director of Computational Oncology for Sage Bionetworks located at Fred Hutchinson Cancer Research Center. "A goal of the Project Data Sphere initiative is to spark innovation - to unlock the potential of valuable data by generating new insights and opening up a new world of research possibilities. Prostate Cancer DREAM Challenge did just that. To witness cancer clinical trial data from Project Data Sphere be used in research collaboration and ultimately help improve patient care in the future is extremely rewarding!" says Liz Zhou, MD, MS, director of Global Health Outcome Research at Sanofi. The goal now is to make the ePCR model publicly accessible through an online tool with an eye towards clinical application. In fact, the National Cancer Institute (NCI) has contracted the winning team to do exactly this. Soon, when patients face difficult decisions about the best treatment for metastatic castration-resistant prostate cancer, ePCR tool could be an important piece of the decision-making process. Challenge winners and results can be found on the Prostate Cancer DREAM Challenge homepage. The clinical trial data can be found at Project Data Sphere. The research article describing this work can be found at The Lancet Oncology. Additional papers that describe individual team methods can be found in the DREAM Channel at F1000Research. The University of Colorado Cancer Center, located at the Anschutz Medical Campus, is Colorado's only National Cancer Institute-designated comprehensive cancer center, a distinction recognizing its outstanding contributions to research, clinical trials, prevention and cancer control. CU Cancer Center's clinical partner University of Colorado Hospital is ranked 15th by US News and World Report for Cancer and the CU Cancer Center is a member of the prestigious National Comprehensive Cancer Network®, an alliance of the nation's leading cancer centers working to establish and deliver the gold standard in cancer clinical guidelines. CU Cancer Center is a consortium of more than 400 researchers and physicians at three state universities and three institutions, all working toward one goal: Translating science into life. For more information visit Coloradocancercenter.org and follow CU Cancer Center on Facebook and Twitter. Founded in 2006 by A. Califano (Columbia University) and Gustavo Stolovitzky (IBM Research) the Dialogue on Reverse Engineering Assessment and Methods (DREAM) Challenges Initiative poses fundamental questions about systems biology and translational medicine. Designed and run by a community of researchers from a variety of organizations, the DREAM challenges invite participants to propose solutions -- fostering collaboration and building communities in the process. Expertise and institutional support are provided by Sage Bionetworks, along with the infrastructure to host challenges via their Synapse platform. Together, the leaders of the DREAM Challenges Initiative share a vision allowing individuals and groups to collaborate openly so that the "wisdom of the crowd" provides the greatest impact on science and human health. More information is available at: http://dreamchallenges. . Project Data Sphere, LLC, an independent, not-for-profit initiative of the CEO Roundtable on Cancer's Life Sciences Consortium (LSC), operates the Project Data Sphere® platform. Launched in April 2014, the Project Data Sphere platform provides one place where the cancer community can broadly share, integrate, analyze and discuss historical patient-level comparator arm data sets (historical patient-level cancer phase III) from multiple providers, with the goal of advancing research. With its broad-access approach, the initiative brings diverse minds and technology together to help unleash the full potential of existing clinical trial data and speed innovation by generating collective insights that may lead to improved trial design, disease modeling and beyond. The platform currently contains 27,600 patient lives of data; 9,400 of those are across a wide spectrum of prostate cancer populations. In order to ensure that researchers can realize the full potential of this data, PDS teamed with CEO Roundtable on Cancer Member, SAS Institute Inc. SAS, a leader in data and health analytics, developed and hosts the site and provides free state-of-the-art analytic tools to authorized users within the Project Data Sphere environment. Sage Bionetworks is a nonprofit biomedical research organization, founded in 2009, with a vision to promote innovations in personalized medicine by enabling a community-based approach to scientific inquiries and discoveries. Sage Bionetworks strives to activate patients and to incentivize scientists, funders and researchers to work in fundamentally new ways in order to shape research, accelerate access to knowledge and transform human health. It is located on the campus of the Fred Hutchinson Cancer Research Center in Seattle, Washington and is supported through a portfolio of philanthropic donations, competitive research grants, and commercial partnerships. More information is available at http://www. .


News Article | November 30, 2016
Site: www.nature.com

The accompanying Comment1 by Mpindi et al. is an important contribution to the discussion of pharmacogenomic consistency for several reasons. Mpindi et al.1 were able to reproduce our initial finding2 of a substantial inconsistency between the pharmacological profiles generated within the Cancer Genome Project (CGP)3 and the Cancer Cell Line Encyclopedia (CCLE)4, and explored potential reasons behind the problem and possible solutions. To do this, they compared the CGP and CCLE to a new dataset generated by the Institute for Molecular Medicine Finland (FIMM) that includes 308 drugs that were tested across 106 cancer cell lines5. The authors shared the subset of the FIMM data overlapping with CGP and CCLE, including 52 drugs tested in up to 50 cell lines (drug dose–response curves and their comparison with CGP and CCLE curves are available in Supplementary Data). Overall, their comparative analysis1 of this newly released dataset supports our published finding2 of greater consistency between studies in which there is similarity in experimental methods. We agree with Mpindi et al.1 that harmonizing the readout, drug concentration range, and statistical estimator makes it possible to achieve greater consistency across pharmacogenomic studies. Here we provide specific responses to the main results reported by Mpindi et al.1 The FIMM and CCLE studies used a similar experimental protocol that included the CellTiter-Glo pharmacological assay, as opposed to the Syto60 assay used in CGP. As pointed out by the authors1, there were also parts of the experimental protocols that were different between all the three studies, effectively preventing perfect replication of in vitro molecular and pharmacological profiles. In our initial analysis2, we showed that drug sensitivity measures for paclitaxel and lapatinib were more consistent between the GlaxoSmithKline (GSK) dataset and CCLE, than between GSK and CGP or CCLE and CGP. Given that both GSK and CCLE used the CellTiter-Glo assay, we concluded that the use of different pharmacological readouts has a substantial effect on the consistency of drug sensitivity measurements. The results from Mpindi et al.1 further confirm this observation. However, a recent study from Genentech suggested that laboratory-specific effects might induce even greater biases than the use of different readouts5. Indeed, Haverty et al.5 showed that, although their CellTiter-Glo screen was more concordant with CCLE, their new drug sensitivity data generated using the Syto60 assay were more consistent to their previous screen than CGP, despite the use of the same pharmacological assay5. It therefore remains unclear whether there are other experimental factors that drive the observed inconsistencies between large-scale pharmacogenomic studies and further argues for a detailed analysis of experimental protocols. Similar to Pozdeyev et al.7 and the Comment by Bouhaddou et al.8, the authors investigated whether sensitivity metrics computed from the drug concentration range shared between studies yield higher consistency than the published metrics computed on the full (only partially overlapping) concentration range using different curve fitting algorithms. Concurring with previous results, Mpindi et al.1 showed that the modified area under the curve (AUC) statistic (referred to as the drug sensitivity score (DSS)) computed on the shared concentration range (harmonized) was better correlated between studies than published AUC values (unharmonized). To test whether this marginal but significant improvement was due to the use of the same drug dose–response curve modelling or the choice of concentration range, we reproduced the authors’ analysis using our PharmacoGx package9 and used the same curve-fitting algorithm for the FIMM CGP and CCLE studies. We observed significantly higher correlations, across and between cell lines, for AUC values computed using a shared concentration range for the CGP and CCLE comparison (P < 0.05, Wilcoxon signed-rank test; Fig. 1). Although this observation held true for the correlations between cell lines for all comparisons, restricting AUC computation to the common concentration range did not yield significantly higher correlation across cell lines for CCLE versus FIMM and CGP versus. FIMM (Fig. 1). Our results confirm that restricting the analysis to the common concentration range improves consistency between CGP and CCLE, and to a lesser extent with the FIMM dataset. In our original report2, we computed correlation coefficients for each individual drug across cell lines, which was relevant to the overall goal of the CCLE and CGP studies to discover new genomic biomarkers of drug responses in order to increase the emergence of ‘personalized’ treatment regimens. Mpindi et al.1 also computed the correlation of cell line sensitivity data across drugs (referred to as between cell lines in their presentation; see Supplementary Information). While the application of the across cell line correlation analysis is more relevant for identifying biomarkers predictive of response to individual drugs, we agree with Mpindi et al.1 that correlations across and between cell lines should be compared in a consistent manner, as was done in our published in-depth reanalysis of CCLE and CGP10. Consistent with Mpindi et al.1, our results clearly demonstrate that the overall correlation across cell lines is lower than the correlation between cell lines (Supplementary Fig. 1). However, gene expression data are significantly more concordant between studies than the drug response summary statistics (half-maximum inhibitory concentration (IC ) and AUC) values in all comparisons (P < 0.002, Wilcoxon rank-sum test). Consequently, our observation that gene expression data are significantly more correlated than pharmacological response still holds. As we argued previously2, we agree with Mpindi et al.1 that there is a need for harmonization of experimental protocols and cross-validation of large pharmacogenomic studies, and that doing so will improve robustness and reproducibility of the associated data. Author A. C. Jin was a student in A.H.B.’s laboratory and left shortly after publication of the initial study, and did not participate in in the writing of this Reply. Authors Z.S., P.S. and M.F. developed the PharmacoGx software package, which enabled the analyses presented here; A.G. helped with the comparison of the different drug sensitivity metrics, and participated in the interpretation of the results and writing of this Reply.


News Article | November 30, 2016
Site: www.nature.com

The comparative analysis by Haibe-Kains et al.1 concluded that data from two large-scale studies of cancer cell lines2, 3 showed highly discordant results for drug sensitivity measurements, whereas gene expression data were reasonably concordant. Here, we cross-compared the two original datasets2, 3 against our own data of drug response profiles in overlapping cancer cell line panels. Our results indicate that it is possible to achieve concordance between different laboratories for drug response measurements by paying attention to the harmonization of assays and experimental procedures. There is a Reply to this Comment by Safikhani, Z. et al. Nature 540, http://dx.doi.org/10.1038/nature20172 (2016). Haibe-Kains et al.1 reported on a comparative evaluation of two drug sensitivity and molecular profiling datasets, one from the Cancer Genome Project (CGP)2 and the other from the Cancer Cell Line Encyclopedia (CCLE)3. In their analyses, gene expression profiles between hundreds of common cancer cell lines across all genes showed high consistency between the two studies (median rank correlation (MRC) = 0.85), whereas the drug response data for 15 common compounds were highly discordant (MRC = 0.28 for half-maximum inhibitory concentration (IC ) values). This report1 and the accompanying commentary4 suggested that differences in laboratory protocols, compounds and their tested concentration ranges, and computational methods may account for the differences, but these reports did not elaborate which of these factors are important and whether they can be controlled for. Here, we reanalysed the dose–response data from both CGP and CCLE using a standardized area under the curve (AUC) response metric, which we call the drug sensitivity score (DSS)5. We then compared the CGP and CCLE data with a new dataset of drug responses profiled using the Institute for Molecular Medicine Finland (FIMM) compound testing assay5, covering 308 drugs across 106 cancer cell lines. The FIMM data included 45 compounds in common with CGP and 14 with the CCLE in 50 cell lines (Supplementary Data 1). In the AUC calculation, we unified the drug concentration ranges across the CGP, CCLE and FIMM assays. We observed a significantly higher level of consistency (P = 4.2 × 10−5), especially between the CCLE and FIMM drug response data (MRC = 0.74), as compared to the consistency between FIMM and CGP data (MRC = 0.54) (Fig. 1a). Similar experimental protocols were applied at FIMM and CCLE, including the same readout (CellTiter-Glo, Promega), similar controls (vehicle as negative control and positive controls of toxic compounds 100 μM benzethonium chloride or 1 μM MG132). However, there were also differences, such as the plate format used (1,536 versus 384 wells). Importantly, there was no effort made to standardize cell numbers used or any other parameters between the three laboratories, such as the source, passage number and media used for cells, nor the origin and handling of drugs. Therefore, this observed level of drug response agreement could be substantially improved by further standardization of the laboratory protocols. The CGP experimental protocol differed from the two others in terms of the readout (fluorescent nucleic acid stain Syto 60, Life Technologies), in the use of controls (drug-free cells as negative and no cells as positive controls), and the plate format used (96- or 384-well plates). We compared the drug response profiles between the same cell lines from different laboratories, in line with the approach of Haibe-Kains et al.1, in which they showed consistency in gene expression profiles from CGP and CCLE (MRC = 0.85)1. The Haibe-Kains et al.1 approach, in which the correlation is calculated for each drug separately across the cell lines, showed more variability (Fig. 1b), owing to the fact that some drugs show minimal efficacy in all the tested cell lines. Analogously, gene expression correlations vary more widely when analysed at the level of genes across cell lines (MRC = 0.58 between CGP and CCLE), as certain genes are not expressed above technical noise. Although both ways to compare the data are relevant to the overall goal of personalized therapy, emphasized in the original publications2, 3, the same evaluation approach should ideally be used when comparing the consistency of gene expression and drug response measurements. In summary, we show that standardization of assay methods and laboratory conditions will help to improve the inter-laboratory agreements in drug response profiling. Global standards, similar to the minimum information about a microarray experiment (MIAME) standard for the microarray data7, should be developed.


News Article | November 17, 2016
Site: www.sciencedaily.com

Knowing the likely course of cancer can influence treatment decisions. Now a new prediction model published in Lancet Oncology offers a more accurate prognosis for a patient's metastatic castration-resistant prostate cancer. The approach was as novel as the result -- while researchers commonly work in small groups, intentionally isolating their data, the current study embraces the call in Joe Biden's "Cancer Moonshot" to open their question and their data, collecting previously published clinical trial data and calling for worldwide collaboration to evaluate its predictive power. That is, researchers crowdsourced the question of prostate cancer prognosis, eventually involving over 550 international researchers and resulting in 50 computational models from 50 different teams. The approach was intentionally controversial. "Scientists like me who mine open data have been called 'research parasites'. While not the most flattering name, the idea of leveraging existing data to gain new insights is a very important part of modern biomedical research. This project shows the power of the parasites," says James Costello, PhD, senior author of the paper, investigator at the University of Colorado Cancer Center, assistant professor in the Department of Pharmacology at the CU School of Medicine, and director of Computational and Systems Biology Challenges within the Sage Bionetworks/DREAM organization. The project was overseen as a collaborative effort between 16 institutions, led by academic research institutions including CU Cancer Center, open-data initiatives including Project Data Sphere, Sage Bionetworks, and the National Cancer Institute's DREAM Challenges, and industry and research partners including Sanofi, AstraZeneca, and the Prostate Cancer Foundation. Challenge organizers made available the results from five completed clinical trials. Teams were challenged to connect a deep set of clinical measurements to overall patient survival, organizing their insights into novel computational models to better predict patient survival based on clinical data. "The idea is that if a patient comes into the clinic and has these measurements and test results, can we put this data in a model to say if this patient will progress slowly or quickly. If we know the features of patients at the greatest risk, we can know who should receive standard treatment and who might benefit more from a clinical trial," Costello says. The most successful of the 50 models was submitted by a team led by Tero Aittokallio, PhD, from the Institute for Molecular Medicine Finland, FIMM, at University of Helsinki, and professor in the Department of Mathematics and Statistics at University of Turku, Finland. "My group has a long-term expertise in developing multivariate machine learning models for various biomedical applications, but this Challenge provided the unique opportunity to work on clinical trial data, with the eventual aim to help patients with metastatic castration-resistant prostate cancer," Aittokallio says. Basically, the model depended on not only groups of single patient measurements to predict outcomes, but on exploring which interactions between measurements were most predictive -- for example, data describing a patient's blood system composition and immune function were only weakly predictive of survival on their own, but when combined became an important part of the winning model. The model used a computational learning strategy technically referred to as an ensemble of penalized Cox regression models, hence the model's name ePCR. This model then competed with 49 other entries, submitted by other teams working independently around the world. "Having 50 independent models allowed us to do two very important things. First when a single clinical feature known to be predictive of patient survival is picked out by 40 of the 50 teams, this greatly strengthens our overall confidence. Second, we were able to discover important clinical features we hadn't fully appreciated before," Costello says. In this case, many models found that in addition to factors like prostate-specific antigen (PSA) and lactate dehydrogenase (LDH) that have long been known to predict prostate cancer performance, blood levels of an enzyme called asparate aminotransferease (AST) is an important predictor of patient survival. This AST is an indirect measure of liver function and the fact that disturbed levels of AST are associated with poor patient performance implies that studies could evaluate the role of AST in prostate cancer. "The benefits of a DREAM Challenge are the ability to attract talented individuals and teams from around the world, and a rigorous framework for the assessment of methods. These two ingredients came together for our Challenge, leading to a new benchmark in metastatic prostate cancer," says paper first author, Justin Guinney, PhD, director of Computational Oncology for Sage Bionetworks located at Fred Hutchinson Cancer Research Center. "A goal of the Project Data Sphere initiative is to spark innovation -- to unlock the potential of valuable data by generating new insights and opening up a new world of research possibilities. Prostate Cancer DREAM Challenge did just that. To witness cancer clinical trial data from Project Data Sphere be used in research collaboration and ultimately help improve patient care in the future is extremely rewarding!" says Liz Zhou, MD, MS, director of Global Health Outcome Research at Sanofi. The goal now is to make the ePCR model publicly accessible through an online tool with an eye towards clinical application. In fact, the National Cancer Institute (NCI) has contracted the winning team to do exactly this. Soon, when patients face difficult decisions about the best treatment for metastatic castration-resistant prostate cancer, ePCR tool could be an important piece of the decision-making process.


News Article | April 27, 2016
Site: www.nature.com

One of the world’s largest pharmaceutical companies has launched a massive effort to compile genome sequences and health records from two million people over the next decade. In doing so, AstraZeneca and its collaborators hope to unearth rare genetic sequences that are associated with disease and with responses to treatment. It’s an unprecedented number of participants for this type of study, says Ruth March, vice-president and head of personalized health care and biomarkers at AstraZeneca, which is headquartered in London. “That’s necessary because we’re going to be looking for very rare differences among individuals.” To achieve that ambitious goal, AstraZeneca will partner with research institutions including the Wellcome Trust Sanger Institute in Hinxton, UK, and Human Longevity, a biotechnology company founded in San Diego, California, by genomics pioneer Craig Venter. AstraZeneca also expects to draw on data from 500,000 participants in its own clinical trials, and medical samples that it has accrued over the past 15 years. In doing so, AstraZeneca will be following a burgeoning trend in genetics research. For years, geneticists pursued common variations in human DNA sequences that are linked to complex diseases such as diabetes and heart disease. The approach yielded some important insights, but these common variations often accounted for only a small percentage of the genetic contribution to individual diseases. Researchers are now increasingly focusing on the contribution of unusual genetic variants to disease. Combinations of these variants can hold the key to an individual's traits, says Venter. The hunt for important rare variants has led AstraZeneca to partner with the Institute for Molecular Medicine Finland, says Aarno Palotie, who heads the Human Genomics Program there. Finland’s population was geographically isolated until recently, he notes, which makes for a unique genetic make-up. As a result, some variations that are very rare in other populations may be more common in Finland, making them easier to detect and study. AstraZeneca did not disclose exactly how much it would be investing in the project — “hundreds of millions of dollars” over the course of ten years was all that Menelas Pangalos, executive vice-president of the company's innovative medicines programme, would say. The company intends to use the data to inform drug development in all of its major disease areas, from diabetes to inflammation to cancer, says March. It is not the first time that a large drug company has poured money into genomics in hopes of fuelling drug discovery, notes David Goldstein, who studies human genetics at Columbia University in New York City and is an adviser to AstraZeneca. “Genomicists have for decades now been promising that genomics is going to revolutionize the way that medicines are developed and the way that medicines are used,” he says. “We are now here saying it again.” Those past efforts often disappointed, but the field has turned a corner, Goldstein adds. Genome sequencing is faster and cheaper than ever before, and researchers are armed with better bioinformatics tools to interpret the data. Advances in stem-cell biology and genome-editing methods such as CRISPR–Cas9 are making it much easier for researchers to determine how a particular change in a DNA sequence affects living cells. In all, the project should generate about 5 petabytes of data. “If you put 5 petabytes on DVDs, it would be four times the height of the Shard,” said Pangalos, referring to a nearly 310-metre London skyscraper. “If you wanted to put it on your iPod, it would take about 5,000 years to listen to it all.” Much of that data will come from Human Longevity. The company, which ultimately hopes to accrue 10 million human genomes, already has 26,000 completed and paired with medical records. Its databases also contain additional partial genome sequences. “We’re adding one about every 15 minutes on average,” Venter says. Using DNA sequence alone, Venter says that his company can now predict a person’s height, weight, eye colour and hair colour, and produce an approximate picture of their face. Much of that detail is lurking in rare sequence variations, says Venter, whose own genome has been in public databases for more than a decade. Human Longevity's databases are kept locked behind layers of security. “If I were advising a younger Craig Venter, I’d say, ‘Think carefully before you just dump your genome on the Internet’,” Venter says. “The levels of prediction are getting much more interesting.”


Orpana A.K.,University of Helsinki | Ho T.H.,Minerva Foundation Institute for Medical Research | Stenman J.,Minerva Foundation Institute for Medical Research | Stenman J.,Institute for Molecular Medicine Finland | Stenman J.,Karolinska Institutet
Analytical Chemistry | Year: 2012

PCR amplification over GC-rich and/or long repetitive sequences is challenging because of thermo-stable structures resulting from incomplete denaturation, reannealing, and self-annealing of target sequences. These structures block the DNA polymerase during the extension step, leading to formation of incomplete extension products and favoring amplification of nonspecific products rather than specific ones. We have introduced multiple heat pulses in the extension step of a PCR cycling protocol to temporarily destabilize such blocking structures, in order to enhance DNA polymerase extension over GC-rich sequences. With this novel type of protocol, we were able to amplify all expansions of CGG repeats in five Fragile X cell lines, as well as extremely GC-rich nonrepetitive segments of the GNAQ and GP1BB genes. The longest Fragile X expansion contained 940 CGG repeats, corresponding to about 2.8 kilo bases of 100% GC content. For the GNAQ and GP1BB genes, different length PCR products in the range of 700 bases to 2 kilobases could be amplified without addition of cosolvents. As this technique improves the balance of amplification efficiencies between GC-rich target sequences of different length, we were able to amplify all of the allelic expansions even in the presence of the unexpanded allele. © 2012 American Chemical Society.


Khan N.A.,University of Helsinki | Auranen M.,University of Helsinki | Paetau I.,University of Helsinki | Pirinen E.,Ecole Polytechnique Federale de Lausanne | And 8 more authors.
EMBO Molecular Medicine | Year: 2014

Nutrient availability is the major regulator of life and reproduction, and a complex cellular signaling network has evolved to adapt organisms to fasting. These sensor pathways monitor cellular energy metabolism, especially mitochondrial ATP production and NAD+/NADH ratio, as major signals for nutritional state. We hypothesized that these signals would be modified by mitochondrial respiratory chain disease, because of inefficient NADH utilization and ATP production. Oral administration of nicotinamide riboside (NR), a vitamin B3 and NAD+ precursor, was previously shown to boost NAD+ levels in mice and to induce mitochondrial biogenesis. Here, we treated mitochondrial myopathy mice with NR. This vitamin effectively delayed early- and late-stage disease progression, by robustly inducing mitochondrial biogenesis in skeletal muscle and brown adipose tissue, preventing mitochondrial ultrastructure abnormalities and mtDNA deletion formation. NR further stimulated mitochondrial unfolded protein response, suggesting its protective role in mitochondrial disease. These results indicate that NR and strategies boosting NAD+ levels are a promising treatment strategy for mitochondrial myopathy. © 2014 The Authors. Published under the terms of the CC BY license.


Bockelman C.,University of Helsinki | Koskensalo S.,University of Helsinki | Hagstrom J.,University of Helsinki | Lundin M.,Institute for Molecular Medicine Finland | And 2 more authors.
Cancer Biology and Therapy | Year: 2012

Background: To improve the prognostic evaluation of colorectal cancer requires new molecular markers. Cancerous inhibitor of protein phosphatase 2A (CIP2A) serves as an oncoprotein by targeting PP 2A-mediated inhibition of c-Myc. A prognostic role for CIP2A has been demonstrated in gastric, lung and tongue cancers. Results: CIP2A was overexpressed in 661 (87.9%) specimens. CIP2A overexpression was associated with tumor differentiation grade (p = 0.014), p53 immunopositivity (p = 0.042), EGFR immunopositivity (p = 0.007) and c-Myc nuclear immunopositivity (p = 0.018). In survival analysis, CIP2A failed to show any prognostic significance (p = 0.270, log-rank test). Methods: 863 consecutive colorectal cancer patients treated at Helsinki University Central Hospital in 1983-2001 were collected with 752 scored successfully for CIP2A immunohistochemical expression from tumor tissue microarrays. Associations with clinicopathologic variables and molecular markers were explored by the chi-square test, and the Kaplan-Meier method served for survival analysis. Conclusions: Overexpression of CIP2A in colorectal cancer patients may be an important step in colorectal carcinogenesis. Based on our findings, CIP2A shows no association with patient prognosis in colorectal cancer, but is associated with nuclear c-Myc. © 2012 Landes Bioscience.

Loading Institute for Molecular Medicine Finland collaborators
Loading Institute for Molecular Medicine Finland collaborators