Patricia F. Fitzpatrick Dimond Ph.D. Technical Editor of Clinical OMICs President of BioInsight Communications
New analytical tools are advancing research efforts.
Instead of one gene or protein, the new paradigm in cancer research focuses on the analysis of entire sets of genes and proteins to identify drug targets. Multiple experimental and data analysis methodologies are being used to achieve this goal, and new bioinformatics tools will advance successful research efforts.
The availability of array-based gene profiling to yield cancer gene-expression signatures has already impacted clinical decision making in several cancers. Known types and subtypes of cancers can be distinguished by gene-expression patterns, and new molecular subtypes of cancer have been discovered that are associated with a propensity to metastasize and sensitivity or resistance to particular therapies.
While researchers have been using DNA microarrays to yield information about the molecular heterogeneity of cancer, analyses that evaluate cancer transcriptome information alongside other data will be able to extract deeper biological insights. The idea is to look at cancer interaction networks and understand regulatory mechanisms encoded in cancer gene expression.
Pursuing this new line of attack requires multiple, complex, and integrated technologies and, in particular, an array of bioinformatics tools allowing organization and analysis of the data. And companies are building significant franchises to offer scientists the requisite tools to help them stay closer to their own data, analyzing it across multiple databases without needing a team of bioinformaticians.
PerkinElmer acquired Geospiza in 2011, adding to its franchise in bioinformatics and cloud computing. The company reportedly has an ongoing program of software and automated workflow development, tied to its Sequencing Services program. The software is based on industry-accepted and supported tools, put together in a robust, user-friendly, automated LIMS and analysis pipeline.
Geospiza software systems include its GeneSifter statistical, visualization, and annotation tools for microarray and next-generation sequencing. They are available to PerkinElmer customers to use through an access fee for the company’s cloud environment. The software is capable of performing three-tiered analyses on DNA and RNA for arrays, sanger, and next-generation sequencing platforms.
“There are a lot of ways to put tools and data together that allow scientists to explore their data,” Geospiza founder and CSO, Todd Smith, Ph.D., told GEN. “But what’s often missing for scientists is a way to work with their data directly without having to engage a team of bioinfomaticians for every question.
“After I see my data, the next thing I want to do is to look at it in a different context. But I have to keep rerunning the programs with different data plugged in. For example, one of our clients studying head and neck cancer wanted to compare cohorts of smokers, ex-smokers, and people who have never smoked. But throughout the study if I want to bring in different samples or look at different combinations of samples together or filter data in different ways, I have to rerun everything.
“Really what I want to do,” Dr. Smith explained, “is click on an interface and set up my comparisons and test ideas for analysis and data filtering. That’s what GeneSifter does for you. It allows you to set up comparisons and run standard algorithms and convert raw data to forms that can be statistically compared in a single environment to generate lists of genes that are linked to ontologies and pathway databases to create new knowledge.
“One of the reasons PerkinElmer acquired Geospiza was to provide customers with a web-based interface connected to a high-powered computer system that could tap into databases and run several standard bio programs like BWA, Bowtie, GATK, and other alignment and variant detection tools along with an integrated statistical analysis platform for measuring gene expression, detecting alternative splicing, and identifying nucleotide differences.
“And most groups have just started to think about combining the transcriptome with the genome and layering in analysis of noncoding RNAs. For the past quarters we’ve been adding these capabilities to GeneSifter so we can give scientists the freedom to explore the full transcriptome when they are ready.”
Scientists have made enormous progress in characterizing the molecular aberrations in cancer, developing and refining their own tools for molecular analysis, and have begun to apply this knowledge to patient care. Research in the laboratory of Kornelia Polyak, M.D., Ph.D., at Dana-Farber Cancer Institute has focused on the molecular analysis of human breast cancer. The goal is to improve the clinical management of the disease by identifying the differences between normal and cancerous breast tissue and determining their consequences.
Her laboratory uses multiple tools to characterize molecular alterations that occur during breast tumor progression. These include SAGE (serial analysis of gene expression) for gene expression profiling, SNP arrays and array CGH for genetic changes, MSDK (methylation-specific digital karyotyping) for the characterization of global DNA methylation profiles, and ChIP-Seq for the analysis of histone modification patterns.
Dr. Polyak and her colleagues, in a paper published last month in Nature Reviews Cancer, said that while recent technological advances have improved the molecular understanding of cancers and the identification of targets for therapeutic interventions, “it has become exceedingly apparent that the utility of profiles based on the analysis of tumors en masse is limited by intratumor genetic and epigenetic heterogeneity.” This means, she says, that characteristics of the most abundant cell type might not predict the properties of mixed populations.
Intratumor heterogeneity continues to present a major clinical problem because tumor cell subtypes display variable sensitivity to therapeutics and may play different roles in progression, Dr. Polyak explained. Her team has provided major insights into potential targets that address intratumor heterogeneity. Her laboratory characterized two cell populations in human breast tumors with distinct properties: CD44+/CD24- cells that have stem cell-like characteristics and CD44-/CD24+ cells that resemble more differentiated breast cancer cells.
The investigators performed a large-scale shRNA loss-of-function screen to define genes upon which each of these distinct cell types specifically depend. They identified 15 genes required for cell growth or proliferation in CD44+/CD24- human breast cancer cells, finding that inhibition of several of these (IL6, PTGIS, HAS1, CXCL3, and PFKFB3) reduced Stat3 (signal transducer and activator of transcription 3) activation. Constitutive STAT3 activation is associated with multiple human cancers. They also found that the IL6/JAK2/Stat3 pathway was preferentially active in CD44+/CD24- breast cancer cells compared to other tumor cell types, and inhibition of JAK2 decreased their number and blocked growth of xenografts.
These results, Dr. Polyak noted, highlight the differences between distinct breast cancer cell types and identify targets such as JAK2 and Stat3 that may lead to more specific and effective breast cancer therapies.
GEN asked Dr. Polyak whether intratumor heterogeneity played a part in tumor treatment yet. Dr. Polyak said, “intertumoral heterogeneity, yes. Breast cancer is commonly tested for hormone receptors and Her2 neu and treated accordingly.” But as for intratumor heterogeneity, she noted, “not yet.” But, she pointed out, strategies to decrease intratumor diversity and combined therapies based on diversity are more likely to lead to more effective eradication of tumors.
In another recent application of multilayer analyses to discover cancer molecular mechanisms, Yon Hui S. Kim and colleagues at MD Anderson gastric cancer center identified the molecular underpinnings of gastric cancer using an RNA-sequencing approach to compare gastric tumor and noncancerous speciments, generating 680 million informative short reads. These short reads were then applied to quantitative characterization of the entire transcriptome of gastric cancer, including mRNAs and miRNAs.
They then developed a multilayer analysis to identify various types of transcriptional aberrations associated with different stages of gastric cancer, including differentially expressed mRNAs, recurrent somatic mutations, and key differentially expressed miRNAs. Through this approach, the scientists said, they identified the central metabolic regulator AMPK-α as a potential functional target in Asian gastric cancer. They also experimentally demonstrated the translational relevance of this gene as a potential therapeutic target for early-stage gastric cancer in Asian patients.
Apart from providing a valuable information resource for identifying and elucidating the molecular mechanisms of Asian gastric cancer, the authors said that their work also represents a general integrative framework to develop more effective therapeutic targets. And as more multiple, complex, and integrated technologies become accessible to scientists and in particular, an array of bioinformatics tools, new potential drug treatments may emerge.
Patricia F. Dimond, Ph.D. ([email protected]), is a principal at BioInsight Consulting.