According to the 2011 National Diabetes Fact Sheet, diabetes affects nearly 26 million Americans, 95% of whom suffer from type 2 diabetes mellitus (T2DM).

 A 2014 report published by the Pharmaceutical Research and Manufacturers of America (PhRMA) documented the development of 180 medications to treat diabetes or diabetes-related conditions, a majority of which are to treat T2DM.

The drugs being developed are intended to improve on the current therapies to combat the health toll and the healthcare costs associated with the disease. Among the drugs under development is a human peptide, a bioactive part of a gene that regenerates pancreatic islets; additionally, there are novel inhibitors of the protein dipeptidyl peptidase-4 (DPP-4) being developed, as well as a drug that targets sorbitol, a sugar alcohol determined to be responsible for diabetic neuropathy.

 These breakthrough advances are based on the research conducted by scientists to understand disease mechanisms, which include gene sequencing and protein structure elucidation.

GenBank, an all-inclusive, open-source database initiated by the National Center for Biotechnology Information (NCBI), has a very important role to play in this process. GenBank includes nucleotide sequences for more than 280,000 species and the supporting bibliographies, with submissions from individual laboratories as well as large-scale sequencing projects. Additionally, sequences from issued patents are submitted by the US Patent and Trademark Office.

 Despite the open access to this database, researchers all over the world have actively contributed to building up the resource, realizing the vast potential of this knowledge-sharing database. The information either goes to GenBank or is submitted through its European counterpart, the European Bioinformatics Institute (EBI), or its Japanese counterpart, the DNA Data Bank of Japan (DDJB).

All the leading journals need researchers to submit their sequences to GenBank and cite the corresponding access number in the published article. The new sequences can be directly submitted to EBI, DDJB, or GenBank, and the 3 databases are synchronized daily for easy access to all the information on all 3 databases. The data are virtually in real time, with minimal delay in access to the latest data, free of cost.

Other commonly used nucleotide databases include the European Molecular Biology Laboratory (EMBL; EBI is run by EMBL), SwissProt, PROSITE, and Human Genome Database (GDB).

 Taken together, these databases are essentially a bioinformatics tool that helps integrate biological information with computational software. The information gained can be applied to understand disease etiology (in terms of mutations in genes and proteins) and individual variables, and ultimately aid drug development.

According to the National Institutes of Health Biomedical Information Science and Technology Initiative, bioinformatics is defined as “research, development, or application of computational tools and approaches for expanding the use of biological, medical, behavioral, or health data, including those to acquire, store, organize, archive, analyze, or visualize such data.”

Initially called the Los Alamos Sequence Database, this resource was conceptualized in 1979 by Walter Goad, a nuclear physicist and a pioneer in bioinformatics at Los Alamos National Laboratory (LANL).

GenBank followed in 1982 with funding from the National Institutes of Health, the National Science Foundation, and the Departments of Energy and Defense. LANL collaborated with various bioinformatics and technology companies for sequence data management and to promote open access communications. By 1992, GenBank transitioned to being managed by the National Center for Biotechnology information (NCBI).

Submissions to the database include original mRNA sequences, prokaryotic and eukaryotic genes, rRNA, viral sequences, transposons, microsatellite sequences, pseudogenes, cloning vectors, noncoding RNAs, and microbial genome sequences. Following a submission (using the Web-based BankIt or Sequin programs), the GenBank staff reviews the documents for originality and then assigns an accession number to the sequence, followed by quality assurance checks (vector contamination, adequate translation of coding regions, correct taxonomy, correct bibliographic citation) and release to the public database.

How Are Researchers Utilizing This Database?

BLAST (Basic Local Alignment Search Tool) software, a product of GenBank, allows for querying sequence similarities by directly entering their sequence of interest, without the need for the gene name or its synonyms.

 An orphan (unknown) or de novo nucleotide sequence, which may have been cloned in a laboratory, can gain perspective following a BLAST search and a match with another, better-characterized sequence in the database. Further, by adding restrictions to the BLAST search, only specific regions of the genome (such as gene-coding regions) can be examined instead of the 3 billion bases.

 BLAST can also translate a DNA sequence to a protein, which can then be used to search a protein database.

BLAST, which was developed at NCBI, works only with big chunks of nucleotide sequences, and not with shorter reads, according to Santosh Mishra, PhD, director of bioinformatics and codirector of the Collaborative Genomics Center at the Vaccine and Gene Therapy Institute (VGTI) of Florida. Mishra, who worked as a postdoctoral research associate with Goad at LANL, was actively involved in developing GenBank. His work contributed to the generation of the “flat file” format, and he also worked on improving the query-response time of the search engine.

Additionally, he initiated the “feature table” in GenBank—the documentation within that helps GenBank, EMBL, and DDJB exchange data on a daily basis. According to Mishra, the STAR aligner, developed at Cold Spring Harbor, works better with reference sequences, while Trinity, developed at the Broad Institute in Cambridge, Massachusetts, is useful for de novo sequences. (The Broad Institute made news last month with its work on identifying gene mutations that prevent diabetes in adults who have known risk factors, such as obesity.)

Advantages and Disadvantages of the GenBank Platform

The biggest single advantage of GenBank is the open-access format, which allows for a centralized repository in a uniform format. The tremendous amount of data generated by laboratories (such as from microarrays and microRNA arrays) cannot be published in a research article. However, the data, tagged and uploaded on GenBank, can be linked to the journals’ websites and the links can be provided in the print versions of the articles as well.

On the flip side, the biggest advantage of being an open-access platform is also the biggest disadvantage of the software. There’s always the probability of scientists registering faulty genetic sequences on the website, which will not be caught unless they are peer reviewed.

Despite the incorporation of several quality control mechanisms into the system, reuse of the data by other scientists alone can help discover glitches in the existing data. Additionally, GenBank encourages its users to submit feedback and update records, which unfortunately is not a very proactive process.

Bioinformatics and Pharmacogenomics in Drug Discovery/Development

Accelerating the drug development process saves costs for the pharmaceutical industry, especially with the way the industry functions today. The company that discovers or invents a new chemical entity, which could metamorphose into a new drug candidate, can squeeze the maximum profit out of the drug before the patent expires and competitors catch on. Essentially, companies jump at every opportunity to accelerate any aspect of the discovery/development process. Resources like the GenBank and EBI are data mines that can speed up the entire process in the following ways:

: Drug candidates can be identified (following a highthroughput screen of chemical libraries) and developed only after a “druggable target” is discovered for a disease condition. Typically, about 1 in 1000 synthesized compounds will progress to the clinic, and only 1 in 10 drugs undergoing  clinical trials reaches the market.

Optimizing/validating a target is essential due to the prohibitively high cost of conducting trials, and the potential targets for drug discovery are increasing exponentially.

information from huge data sets, like the human genome sequence, the nucleotide sequence of the target proteins has become readily available, as has the potential to identify new targets. This can exponentially increase the content of the drug pipelines of pharmaceutical companies.

According to Arathi Krishnakumar, PhD, a protein biochemist and a senior research investigator with the department of Exploratory Biology and Genomics, Bristol-Myers Squibb (BMS), “For compounds that have no obvious targets from a typical phenotypic screening, proteomics offers tools for target identification or target deconvolution. Monitoring the global phosphorylation

status of proteins that are downstream of tyrosine kinase inhibitors—also termed phosphoproteomics&mdash;is a very attractive tool that can also be used for target as well as biomarker identification. These events can be used as reporters (biomarkers) for specific upstream kinase(s).”

 reported on the identification of a mutation in the gene SLC30A8 that protected individuals from developing T2DM. The mutation was identified by genetic tests conducted in more than 150,000 individuals, a multi-collaborative effort between the Broad Institute, Massachusetts General Hospital, Pfizer, and Amgen.

 Establishing a robust association between a likely target and the disease, to confirm that target modulation translates into a beneficial therapeutic outcome, would not only validate the drug development process but also help absorb the risks associated with clinical trial failure of the molecule being developed.

Says Krishnakumar, “Target validation is typically done with knock-out or knock-down of the proposed target using RNAi and then monitoring the disease phenotype in relevant cellular models. Proteomics tools are also highly valuable in monitoring specific events on proteins like post translational modifications, including phosphorylation, methylation, oxidation, etc, new product

generation, degradation products, protein-protein interaction, etc, all of which could be direct or indirect consequences of target activation or engagement.”

 The drug development process is not just lengthy (product development can take 10 to 15 years

), but prohibitively expensive as well. Averaging $140 million in the 1970s, the cost of developing a drug was estimated at a whopping $1.2 billion in the early 2000s,

 and a recent Forbes analysis estimated the cost at $5 billion.

Worth noting is that the final cost of any drug, which includes the total costs from discovery to approval, includes the cost of absorbing all the clinical trial failures.

 Clearly, bioinformatics tools improve the efficiency of target discovery and validation processes, reduce the time spent on the discovery phase, and make the entire process more cost-effective.

Mishra believes GenBank is a good starting point in the drug discovery process. When a new sequence (of known or unknown function) is identified/isolated in the laboratory, a GenBank search will help identify homologues (human or in other organisms) with a 70% to 80% match. Functional studies would then ensue, along with cell and tissue distribution studies.

With the value of personalized medication gaining acceptance, the study of pharmacogenomics (genetic variants that determine a person’s drug response; one size does not fit all) is extremely helpful to tailor the optimal drug, dose, and treatment options for a patient to improve efficacy as well as avoid adverse events (AEs).

 According to the Agency for Healthcare Research and Quality of the HHS, AEs annually result in more than 770,000 injuries and deaths and may cost up to $5.6 million per hospital.

To this end, EMBL-EBI is actively involved in industry partnerships (the partnerships were initiated in 1996), which include Astellas, Merck Serono, AstraZeneca, Novartis, GlaxoSmithKline, BMS, and several others.

 With the high-throughput data that research and development (R&D) activities generate, open-source software and informatics developed by organizations like the GenBank and EBI could greatly improve efficiency and reduce the cost of drug discovery and development.

Translational Bioinformatics and Precision Medicine

Healthcare today is primarily symptom driven, and intervention usually occurs late in the pathological process, when the treatment may not be as effective. Identifying predisease states that could provide a window into the forthcoming risk of developing a disease, identifying reliable markers, and developing useful therapies would be the key to managing disease treatment

—not just to improve efficiency but also to reduce healthcare costs, which it is estimated will steadily increase and by 2022 account for 19.9% of the gross domestic product (GDP).

With precision medicine or personalized medicine, molecular profiles generated from a patient’s genomic (coupled with other “-omics” such as epigenomics, proteomics, and metabolomics) information could help accurately drive the diagnostic, prognostic, and therapeutic plans, tailored to the patient’s physiological status. Predictive models can also be developed for different biological contexts, such as disease, populations, and tissues.

 However, the deluge of data generated by bioinformatics tools requires a framework to regulate, compile, and interpret the information.

Most importantly, the key stakeholders (government, research industry, biological community, pharmaceutical industry, insurance companies, patient groups, and regulatory bodies

) that would drive the widespread acceptance and implementation of precision medicine need to be brought up to speed with the enormous progress made in the field and the promise it brings. There would also be a revolutionary change in the approach to conducting clinical trials—the phase 3 studies conducted in the target population could focus on a more select patient group, which could improve both clinical and economic efficacy.

At BMS, Krishnakumar’s group actively provides support to clinical trials by developing assays for clinical samples. When it comes to administration of biologics such as antibodies, individual variations such as expression levels of various proteins and their affinity for an antibody necessitate dose-titration in order to personalize treatment to improve efficacy. The developing field of translational bioinformatics creates a platform to bring all the data together, which can then be used to generate a treatment plan personalized to a patient (Figure 1). It has been defined as “the development of storage, analytic, and interpretive methods to optimize the transformation of increasingly voluminous biomedical data into proactive, predictive, preventative, and participatory health.”

The primary goal of translational bioinformatics is to connect the dots and develop disease networks that can be used as predictive models. In other words, harmonization of the data from different sources (genome, proteome, transcriptome, metabolome, and patient’s pathological data) could help in making better-informed treatment decisions. Within medical R&D, a commonly held belief is that cures for diseases could be found residing within existing data, if only the data could be made to give up their secrets.

 The current status of the scientific, medical, and healthcare fields is that experts in each field have set their minds on developing the best technologies; unfortunately, the technologies are compartmentalized and they work in parallel.

The great need, which has been recognized and implemented in limited areas, is to create platforms where the data can be merged to produce meaningful outcomes.

Data Integration Platforms to Boost Evidence-Based Decisions

Implementing these huge changes would necessitate that physicians and providers be more adept at interpreting molecular data, which essentially requires improved education models that include relevant courses during graduate training. Also, development of software that can interpret the data would provide a tremendous advantage to researchers, clinicians, scientists, pathologists, and maybe patients as well.

An application developed by Remedy Informatics, TIMe, boosts the process further. TIMe merges data, registries, applications, analyses, and any other relevant content. TIMe promises to enable faster, more informed decisions in clinical practice, research, and business operations. It also is expected to improve treatment effectiveness, quality of care, and patient outcomes.

In Europe, a collaborative project (DIRECT) was initiated in 2012 by 4 pharmaceutical companies and 21 academic institutions, with the objective of stratifying diabetic patients based on biomarker identification, which would allow targeted intervention, monitor treatment response, conduct stratified trials, and identify nonresponders or those who might be intolerant to treatment.

 The pharmaceutical companies participating in this effort include sanofi-aventis, Eli Lilly, Servier, and Novo Nordisk.

DIRECT, in turn, constitutes 1 of 3 consortia under development by the Innovative Medicines Initiative (IMI). IMI, a partnership between the European Commission and the European Federation of Pharmaceutical Industries and Associations, includes DIRECT, IMIDIA (to slow disease progression by improving beta-cell function), and SUMMIT (developing surrogate markers for late-stage micro- and macro-vascular complications), collaborative efforts aimed to develop novel individualized therapies with improved efficacy and safety.

Applications of Translational Bioinformatics

Once the genomic and/or proteomic data have been generated, what next? How are providers employing these data to their advantage and to guide treatment? There are several reports on clinical studies that are being successfully conducted on the foundation of precision as well as evidence-based medicine. Researchers at the University of Southampton have developed a blood test that identifies young children at risk for developing obesity. The test, to be conducted in children as young as 5 years, differentiates between those with high body fat and those with low body fat when they grow older. This test, which identifies epigenetic changes (DNA methylation), showed that a 10% increase in DNA methylation at age 5 years translates into 12% more body fat by the time the children are 14 years old, independent of gender, physical activity, or their timing of puberty.

The principal investigators on the study believe that identifying at-risk children could help make lifestyle modifications early on to help disease management.

 Genetic analysis using  genotyping and sequencing techniques in a family of type 1 diabetes mellitus (T1DM) patients resulted in the identification of a mutation in the SIRT1 gene, which produces a defunct SIRT-L107P protein.

Essentially, the study identified that the mutant protein was responsible for an autoimmune disorder in the family: 4 members suffered from T1DM and 1 developed ulcerative colitis. Overexpression of SIRT-L107P in beta cells in vitro resulted in an increased expression of nitric oxide, the cytokine TNF-α, and the chemokine KC, compared with the controls. Additionally, SIRT1 knockout mice were more susceptible to islet destruction and hyperglycemia following induction of pancreatic insulitis.

The study provides a foundation for the application of SIRT1 activators, already under development for aging and other metabolic disorders, in T1DM therapy.

Bioinformatics studies have also yielded microRNAs, which are small (~22 nucleotides), noncoding RNA molecules that can repress the transcription of messenger RNA (mRNA) or promote its degradation, thereby silencing gene expression.

 Initially thought of as “junk” sequences on the DNA since they are non-coding nucleotides, miRNAs (about 24,521 listed in miRBase, a database maintained by the University of Manchester

) have now found their place in clinical trials as biomarkers (cancer,

) and are also being developed as “drugs” by companies like Mirna Therapeutics Inc.

Articles

Evidence-Based Diabetes Management

The Role of Bioinformatics in Diabetes Drug Development-and Precision Medicine

Normal analysis of discrete health data, like blood pressure and cholesterol, identified 5000 patients at risk of developing congestive heart failure.

Automated analysis of doctors’ notes and other unstructured information from Carilion Clinic’s 8 hospitals turned up 3500 more.

Early treatment should prevent many of those extra cases from ever developing and save the communities of western Virginia hundreds of lives and millions of dollars.

Such efforts to avert the chronic diseases that kill the majority of Americans and consume the majority of their healthcare dollars rank among the most promising medical applications for “Big Data” analysis.

An unbelievably large amount of medically useful information is available for study—a century of published studies, decades of insurance claims&mdash; and that stock expands every time a doctor completes an electronic medical record or a runner dons a heart-rate monitor.

Software that can find, interpret, and analyze it all may eventually revolutionize healthcare.

A McKinsey & Company analysis, for example, predicts such programs will soon save at least $300 billion a year in American healthcare spending—and possibly much more.

“Why is Big Data emerging in healthcare now? There are really 3 reasons,” said McKinsey director Nicolaus Henke. “The first is availability. We have so much more captured, machine-readable data available to us than we did just a few years ago. The second reason is that it’s much cheaper and easier to link these data. The third reason is a big imperative to understand population health better…it’s important both for outcomes and costs.”

Indeed, even with all the limitations in both data and software, would-be innovators are already finding significant ways to use the existing data to improve patient health.

The most famous applications to date lie in taking some of the data that modern life automatically generates about individual activity and reusing it to benefit that individual.

Smartphone applications that tap GPS and clock functionality to track runs have millions of users. Diet applications that use phone cameras and Internet connectivity to help users track what they eat have even more.

Medical practices are using similar tools to help patients.

Billing records have always recorded when patients come in for Papanicolaou tests (Pap smears), but now software sold to gynecologists can automatically look through those records, infer which patients are overdue for another Pap smear, and send reminders to those patients.

Such programs can also look through medical records to see which women began receiving the sequence of shots needed for human papilloma virus vaccination and call to remind them about the next shot.

Applications like that, which use relevant data from each individual to help that same individual, may provide substantial health benefits, but experts see even more promise in tools that use both individual data and collective data, such as the tools that IBM used to predict congestive heart failure (CHF) at Carilion.

“Traditional models use a handful of medical measurements to predict CHF,” said Ed Macko, IBM’s chief technology officer for healthcare & life sciences.

“Our systems—after scanning not only structured data but also free-written material from doctors’ notes, journals, and other sources&mdash;found dozens of relevant factors, including stuff that has rarely been considered before, like whether the patient has a job or someone at home that can provide care during illness.”

Such deep analysis allowed IBM’s technology to identify 70% more atrisk patients than traditional tools, all while maintaining an estimated 85% accuracy rate that matches prior standards.

And each passing month increases both the number of patients identified and the accuracy of the prediction.

IBM has plenty of competitors, big and small, that want to use Big Data to improve healthcare. The McKinsey report estimates that 200 new companies have already entered the space. Older companies, universities, government agencies, and various nonprofits are also getting into the act.

Much of their work resembles the project at Carilion. Its purpose is to predict which people will become chronically ill—be it from CHF, diabetes, chronic obstructive pulmonary disease, drug addiction, or a handful of other problems&mdash;and prevent the downward spiral.

“It makes sense to focus here because a relatively small number of very ill people account for a huge percentage of both the suffering and the cost,” said Erica Mobley, senior manager at a hospitalmonitoring nonprofit called The Leapfrog Group.

“Hospital systems have also focused on using Big Data to expand and improve upon data-driven decision making,” said Mobley, who noted that the real analytical pioneers among hospitals tend to be self-insuring university systems that can get a full picture of patients by using complete medical records, drug records, and insurance records.

The University of Pittsburgh Medical Center, for example, announced in 2012 that it was working with outside companies to create an enterprise data warehouse that would draw on more than 200 data sources to provide doctors with individualized care recommendations for particular patients.

“Ever more data, sometimes right down to the genetic level, give hospitals the ability to help staff determine the correct decision in ever more specific situations. These data-driven decisions replace instinct or gut feeling, which studies have generally shown to be little better than raw guesswork.”

Big Data is also helping groups like Leapfrog improve their hospital  rankings.

When Leapfrog was founded in 2000, hospitals reported so little data on safety that sophisticated analysis was unnecessary. Now, thanks to efforts by Leapfrog and other groups to increase transparency, patients can compare hospitals on issues as specific as the likelihood that the doctors will leave something inside them after surgery or that the staff will give them the wrong type of blood.

To help patients understand how to value all those extra data, Leapfrog (which still wants way more data) now uses sophisticated analytics to weigh the different factors and compile a single letter grade for each facility.

Government agencies have also begun using Big Data to improve healthcare.

The FDA has launched a number of projects that mine and analyze data, including a program called Mini-Sentinel that automatically combs medical databases for signs of drug safety issues that were not detected before approval.

The numbers involved are vast. An FDA report from January revealed that as of July 2012, the Mini-Sentinel system had already collected records of some 3.8 billion medical visits and 3.5 billion dispensations of medication for 160 million Americans.

For all those records, however, questions remain about the reliability of the analysis performed by the current system. A research letter published in January’s edition of the Journal of the American Medical Association, for example, noted that traditional studies comparing the bleeding risk of warfarin (Coumadin) and dabigatran (Pradaxa) have all found substantially more risk with dabigatran, while Mini-Sentinel found more with warfarin.

Looking forward, the FDA reportedly plans to expand its automatic monitoring system to read sources like Facebook and Twitter for signs of drug safety issues.

Similar techniques have already shown some usefulness. Google, for example, has demonstrated that it can often spot a regional outbreak of flu earlier than health authorities simply by noting the prevalence of flu-related Web searches.

Researchers at Stanford and Columbia, moreover, were able to find a drug interaction the FDA had missed—the tendency of paroxetine (Paxil) and pravastatin (Pravachol) to raise blood sugar when used together&mdash;by analyzing tens of millions of search queries.8

Amid its efforts to use Big Data to monitor the safety of marketed drugs, the FDA also hopes its collection of drug trial information can help it develop software to better predict the behavior of experimental drugs in the human body.

Agency officials are using their vast archives of data to help build physiologically based pharmacokinetic models to predict drug absorption. Such models may spot potential problems with new drugs and improve the FDA’s ability to evaluate them.

Of course, the FDA’s mountain of trial data could prove useful to many health-related analyses, so the agency plans to throw much of it open to outside researchers. FDA officials have launched a resource called the Janus Clinical Trials Repository, designed not only to release terabytes of information but also to make it user friendly.

The FDA is also tapping outside organizations for help.

It funded a Center for Excellence in Regulatory Science and Innovation (CERSI) at the University of Maryland to help it use Big Data (and many other tools) to modernize and improve the review and evaluation of drugs and medical devices.

The CERSI’s work on Big Data and healthcare, which includes a recent conference on the subject, nicely complements other efforts by Maryland to collect and harness records, efforts like its Research HARBOR (Helping Advance Research By Organizing Resources) project.

“Assembling databases in useful ways has been very hard work. Claims databases lack the detail that researchers want. Medical records are only just going electronic—and even the electronic records we have are often incomplete and sometimes inaccurate,” said Eleanor M. Perfetto, PhD, MS, professor of pharmaceutical health services research at the University of Maryland’s School of Pharmacy.

“Still, while there is much work to be done, not only with traditional data sources but also completely new ones such as social media, we are making progress.”

Indeed, researchers can access data from insurers such as UnitedHealthcare, government entities such as the United Kingdom’s National Health Service, or the companies that make electronic medical record software. There are also data sellers like Humedica that try to link data from several sources to give researchers more holistic views of patient health.

Such data have many uses, but the most valuable, commercially speaking, may be the development of new treatments.

Many device and drug makers think Big Data can significantly improve the success rates of their laboratories and help them bring drugs to market faster, and more of them. Their projects vary widely. Some are monitoring social media, analyzing what people say about their products, and considering that feedback in new designs. Others are using archived medical records to determine the characteristics of target populations and thus improve enrollment criteria for drug trials.

Most of these projects have yet to advance beyond pilot programs and other early-stage initiatives.

The same could be said about virtually all efforts to better healthcare with Big Data. The successes of these efforts, while sometimes impressive, have generally been limited in scope, and many obstacles will hinder attempts to expand them to the system as a whole.

Patient data, as Perfetto said, is sometimes wrong, often sketchy, and almost always stored in dozens of different databases that must be accessed separately, if they can be accessed at all. Territorialism, privacy concerns, and other issues will hinder adequate data assembly. What’s more, computer software suffers real limitations in its ability to interpret and analyze the available material.

Big Data failures still outnumber successes, and some very easy sounding analyses still lie outside the realm of possibility.

That said, each week brings news of another promising application for data-parsing software, applications such as ones that help drug development by “reading up” on the nearly endless supply of peer-reviewed articles that have been published over decades of time.

No 1 person—no team of people&mdash;could ever read all the relevant studies before choosing a drug target or a promising design, but programmers are “teaching” their computers to understand subject areas such as biology and chemistry and to “read” far more research than humans ever could.

One research hospital, in collaboration with IBM, used software IBM to analyze decades’ worth of literature about p53, a protein involved in both normal cell growth and many types of cancer. Using information in those papers about kinases that are known to act on p53, the software created a general understanding of p53-kinase interaction. It then made a list of other proteins mentioned in the literature that were probably kinases that would interact with p53.

Most of the computer’s predictions proved accurate.

	1. IBM predictive analytics to detect patients at risk for heart failure [press release]. Armonk, NY: IBM Newsroom; February 19, 2014. http:// www-03.ibm.com/press/us/en/pressrelease/ 43231.wss. Accessed April 30, 2014.

	2. Kayyali B, Knott D, Van Kuiken S. The big data revolution in US healthcare: accelerating value and innovation. McKinsey & Company website. http://www.mckinsey.com/insights/health_ systems_and_services/the_big-data_revolution_in_us_health_care. Published April 2013. Accessed April 30, 2014.

	3. Tracey JK. Big Data and the research HARBOR. Presented at: Leveraging Big Data II: What Does It Mean for Improving Product Development and Health Care? February 11, 2014; University of Maryland School of Pharmacy, Baltimore, MD. http://www.pharmacy.umaryland.edu/centers/cersievents/pdf/presentations/TracyNotes.pdf. Accessed April 30, 2014.

	4. UPMC fosters ‘personalized medicine’ with $100 million investment in sophisticated data warehouse and analytics [press release]. Pittsburgh, PA: University of Pittsburgh Medical Center; October 1, 2012. http://www.upmc.com/media/newsreleases/2012/Pages/upmc-personalized-medicine-investment.aspx. Accessed April 30, 2014.

	5. Mini-Sentinel Data Score: Mini-Sentinel Distributed Database Year 3 Summary Report. Mini-Sentinel website. http://www.minisentinel.org/work_products/Data_Activities/Mini-Sentinel_Year-3-Distributed-Database-Summary-Report.pdf. Published July 2012. Accessed April 30, 2014.

	6. Sipahi I, Celik S, Tozun N. A comparison of results of the US Food and Drug Administration’s Mini-Sentinel program with randomized clinical trials: the case of gastrointestinal tract bleeding with dabigatran. 

	7. Karlin S. Adverse events in social media: FDA expects Signal Detection Revolution. The Pink Sheet. http://www.pharmamedtechbi.com/publications/the-pink-sheet/76/4/adverseevents-in-social-media-fda-expects-signal-detection-revolution. Published January 27, 2014. Accessed April 30, 2014.

	8. Markhoff J. Unreported side effects of drugs found using data, study finds. The New York Times. March 7, 2013. http://www.nytimes.com/2013/03/07/science/unreported-sideeffects-of-drugs-found-using-internet-datastudy-finds.html. Accessed April 30, 2014.

	9. Florian J. FDA use of Big Data in modeling and simulations. Presented at: Leveraging Big Data II: What Does It Mean for Improving Product Development and Health Care? February 11, 2014; University of Maryland School of Pharmacy, Baltimore, MD. http://www.pharmacy.umaryland.edu/centers/cersievents/pdf/presentations/FlorianNotes.pdf. Accessed April 30, 2014.

	10. Rosario L. Janus Clinical Trials repository: an update and insights into future directions. Presented at: Leveraging Big Data II: What Does It Mean for Improving Product Development and

	Health Care? February 11, 2014; University of Maryland School of Pharmacy, Baltimore, MD. http://www.pharmacy.umaryland.edu/centers/cersievents/pdf/presentations/RosarioNotes.pdf/. Accessed April 30, 2014.

	11. Center of Excellence in Regulatory Science and Innovation website. University of Maryland. http://cersi.umd.edu/. Accessed April 30, 2014.

	12. Clinical and Translational Sciences Institute website. University of Maryland. http://medschool.umaryland.edu/ctsi/harbor.asp. Accessed April 30, 2014.

“This software isn’t going to cure cancer yet, but it did make significant new discoveries about a very heavily studied protein, and there is a significant possibility that some of these proteins could be medically useful,” said Ying Chen, a research staff member from IBM’s Watson Group. “This technology is ready to make real contributions.” 

The Big Data Revolution: From Drug Development to Better Health Outcomes?

The interaction between stress and diabetes mellitus, though long established, has flown under the radar in a flurry of new pharmacologic therapies in recent years. 

 revisited this important finding and its implications for patient care with 2 pioneers of the field from Duke University who have worked together to elucidate much of what we know today: Mark Feinglos, MD, professor of medicine, endocrinology, metabolism, and nutrition; and Richard Surwit, PhD, vice chairman and head professor, department of psychiatry and behavioral sciences.

A Relationship Between Levels of Stress and Blood Glucose

You’ve been working on the effects of stress and blood sugar levels for 3 decades, but it has not received the kind of spotlight that pharmacologic

treatments routinely receive. Why do you think that is?

 Speaking as the psychologist on the team, interest in many psychological issues, including stress, come and go in the popular press. People attend to them and then forget about them. We were on the cover of Newsweek magazine in 2004 regarding the role of stress and disease and there was a big to-do about it, but that faded.

: As an endocrinologist, I believe physicians sometimes have a different approach to cognitive function. Some ignore it. Some tend to view their patients from the neck down, so to speak. They worry about the physiology and the anatomy but they don’t think about the brain’s involvement in disease.

Then there are other groups of physicians and care providers—and I believe endocrinologists are part of this group&mdash;who more often consider the role that brain and behavior play in diabetes care. You see evidence of that in the steady stream of articles on the topic appearing in the specialty journals and in the journals related to nursing care or the ancillary health professions. Unfortunately, most physicians tend to not think quite as much about behavior issues.

I haven’t seen too many articles recently relating levels of stress to diabetes. 

 The reasons for this are: (1) It’s a complex area—what does stress mean? Everybody lives with some kind of stress; (2) It’s time consuming to delve into how much stress a patient

has and how that is interacting with someone’s diabetes. As you know, most physicians don’t have a great deal of time in today’s office environment; (3) Once you’ve even taken that time, how do you treat it? Richard and I have thought about that for a long time and have looked at both behavioral and pharmacological ways of thinking about those issues. But it’s very complicated. We’re frequently asked by attorneys whether a stressful event, like an auto accident, can trigger the onset of diabetes. What is more likely is that a stressful situation will not cause diabetes, but the event in the presence of underlying diabetes increases stress hormone levels, which raises glucose levels, perhaps more so in someone with diabetes.

The direction of the work that Mark and I have been doing for over 30 years has led us to look at how the brain and the central nervous system are involved in the etiology of diabetes, and

not just in terms of stress. That’s the thrust of what we’re doing now. The fact that diabetes is a stress-responsive disease, as are other diseases like high blood pressure, suggested to us a long time ago that there may be something wrong in the neurologic control of blood glucose that makes a person susceptible to developing diabetes. Let me give you a specific example from some of our recent work. We have shown that, in some groups, epinephrine (adrenaline), a classic stress hormone circulating in the body, interacts with central adiposity in producing elevated fasting blood sugar levels. Lean people with high levels of epinephrine don’t seem to have this problem.

However, obese people who have low levels of epinephrine (whom you might say are not stress responsive, or are relatively laid back) don’t demonstrate this abnormality in blood sugar. So there may be an interaction between the factors that create obesity—and those are probably numerous&mdash;and the activity of the autonomic nervous system in producing abnormalities in blood glucose concentrations. We don’t really know what fraction of patients with type 2 diabetes mellitus (T2DM) is affected by this interaction.

What’s more interesting to us at this point is not so much that people with diabetes will show exacerbations of diabetes when they’re under stress, but rather, what does that tell us about the

pathophysiology of diabetes in general? That suggests to us that certain drugs commonly used for other things might actually be helpful in the management of people with T2DM, particularly the

What role does insulin resistance play in that interaction, and does the brain, directly or indirectly, regulate insulin resistance?

The simplified explanation is that when adrenalin interacts with fat mass, it elevates free fatty acids, which then stimulate the liver to produce more sugar—essentially creating insulin resistance.

 And this isn’t an abnormal response—it’s the old fight-or-flight response. If you’re running or fighting, you need a source of energy. These hormones help you release energy from the liver and put it just where you need it. In that scenario, insulin resistance isn’t necessarily abnormal, but it’s an exacerbation of what was designed through evolution as a normal response.

: The idea that stress only affects people with a proclivity to the disease—in this case, people with substantial central adiposity&mdash;is not new. In fact, that’s probably true for every disease linked to stress. Stress will not cause hypertension or irritable bowel syndrome in and of itself. If someone has an autoimmune disease affecting the gastrointestinal system, that person will be more susceptible to the effects of stress. Stress should be viewed as working through one of the lesions, if you will, that predisposes a person to disease. Not something that causes disease de novo.

 In your research, you found certain subgroups that seem to have a greater stress response than other groups.

Right. We found this to be the case for African American women. There are 2 possible reasons why African American women have a greater problem with this. One is that they have much higher levels of central adiposity than either Caucasian women and men or African American men. As a group, they have high levels of central adiposity. The other potential contributing factor that could play a role is that they have more sensitive beta-adrenergic receptors, which lead to the release of free fatty acids and are stimulated by adrenalin. So those 2 factors are probably conspiring to make them more susceptible.

: What we’re seeing is not unusual in other diseases: an interaction of a genetic predisposition with factors that will bring out that genetic predisposition.

: In this same research, I found it interesting that you measured hostility, or markers of hostility. Is that a proxy for stress levels?

 I don’t really know. Hostility is defined by a 27-item scale that was de-rived from the Minnesota Multiphasic Personality Inventory. The scale tends to describe people who are relatively

untrusting of what’s going on around them. That more defines suspiciousness than it does hostility, although that’s the name the scale was given.

We don’t really understand why that is related, particularly in African American women, to abnormalities in blood glucose levels. It may be related to the fact that they are more responsive to high

epinephrine levels, that their autonomic nervous systems are more attuned to the fight-or-flight response. But that’s an open question. It’s an observation we and other groups have seen on multiple occasions. Exactly what it means and why it’s more common in African American women is not clear at this point.

Let’s take this link between stress and glucose management into the physician’s office. What does the doctor or nurse educator tell the patient with diabetes about stress in the

management of the disease? What should patients do about it?

I wish there were a simple answer to that. Someone who takes the time to question patients about environmental stressors will often find fertile territory. But everyone has stress. Some

people live in a much more stressful environment than others and you can’t necessarily change the environment.

Nor do you want to put all kinds of people on pharmacologic therapies to alter their cognitive responses to stress. Beyond attempting behavioral approaches to modifying stress, making

people aware of what’s happening is important. If they can’t avoid the stress, they can modify their diabetes treatment to better cope with it. For example, if you know your blood sugar level is rising because of stress, you should try to bring your stress level down, knowing that this will positively affect your glucose concentrations.

If that is not possible, it may be necessary to modify the medication regimen for periods of increased stress.

 The system for payment of services in medicine and psychiatry is really tying our hands, because we can’t see people and bill them for a couple of sessions and teach them stress management for their diabetes unless they carry an outright psychiatric diagnosis. Most people with minor stress problems don’t have that diagnosis.

One of the things that we’re looking at now is using the beta-blocking drug carvedilol as a way of helping people whose glucose levels are more sensitive to stress. The drug may have very positive effects on glucose concentrations in these patients, because it blocks the mobilization of free fatty acids caused by adrenalin release. Mark and I are trying to get this study funded. If carvedilol works as we hope, we would have a very low cost intervention that wouldn’t make people drowsy or affect their cognitive function, but could blunt their glucose response to stress quite significantly.

 Realize that at this point, our ideas are unencumbered by data! We have to be a little careful about that.

 Data do exist on carvedilol’s unique properties, and it has been shown to improve glucose tolerance in people with diabetes being treated for hypertension as well. The problem is that

nobody has taken a look at whether this works particularly well in people with high central adiposity, which is what our hypothesis would suggest. And we don’t suggest using other β-blockers, like metoprolol or propranolol. Carvedilol blocks α-adrenergic receptors as well as β receptors. That prevents the reduction in insulin secretion that occurs when giving another β-blocker. We think this is a promising avenue.

It sounds like this approach might be more appropriately categorized as a treatment for the metabolic syndrome than specifically for T2DM.

 That could possibly be the case. We’ve been trying to get the National Institutes of Health to fund the study for 6 years now. And people either tell us, “It won’t work” or “Everybody knows it will work.” Meanwhile, we haven’t yet obtained grant support. If the drug had still been on patent, we might have gotten the manufacturer (GlaxoSmithKline) to fund it. There’s no commercial interest now that carvedilol is sold as a generic.

Has any research been done on the effect of anti-anxiety drugs to reduce stress in patients with diabetes?

We looked at that ages ago, both in animal models that we developed and in humans. First of all, we can’t put everyone on anti-anxiety drugs.

We evaluated the benzodiazepines. The one that we’ve studied most was alprazolam (Xanax). The first problem is that people develop a tolerance to it. The second problem, as Mark alluded to, is that it does produce sedation…

 …and it has an addictive potential. It wasn’t really a good idea. We tried to demonstrate that using it would modify stress and affect glucose levels, and it did. Which is very nice, but you can’t use that as treatment long term in large numbers of people.

Some Early Clues to the Relationship Between Fat Intake and Glucose Management

Both of you have been evaluating this area for a long time. How did your interest begin?

 In 1977 or 1978, when I first came to Duke, I was brought in to run the stress management unit of the department of psychiatry. Mark sent me a consult for a patient with type 1 diabetes who was in very poor control, wanting to know if we could do anything to help her. I said I’d never worked with people with diabetes before. We tried stress management techniques, and this

woman’s glycemic control responded very dramatically. Then we tried it with another patient, who also responded dramatically. So we decided we might be on to something here.

: In the late 1980s, we decided to look at this effect in animal models, the Black 6 mouse (also called the BL/6J mouse) which, being the strain on which the mutation is based, everybody thought was a control for the ob-ob mouse, a genetic model of obesity and diabetes. We discovered that this mouse in and of itself was susceptible to developing both significant obesity and T2DM if we fed it a diet with American junk food (basically cookie dough, containing lard and sugars). We asked ourselves, how would we make a susceptible mouse diabetic?

We decided to feed it exactly what the American population eats. It worked. We published that in 1988. We learned that other strains of mice were not as responsive. On further study, it was clear that epinephrine was one of the mechanisms involved in this phenomenon, and it interacted with adiposity. Then we tried to identify the genes that were responsible for this particular problem.

 When we fed the mice pure sugar, it didn’t do a thing to their blood sugar. Fat was the culprit. The idea that sugar causes diabetes at least in rodents is clearly false. It doesn’t happen.

We published a study in 1997 in the American Journal of Clinical Nutrition, which showed that if we fed women trying to lose weight a hypocaloric diet that was 50% sucrose, they didn’t have any

problems either. They lost weight just fine. So we’re not big proponents of the sugar theory of obesity and diabetes. We think fat is really the culprit. And when you have a high fat diet, adding

sugar makes it worse. But in the absence of fat, sugar and other simple carbohydrates are relatively benign. So I think people have really been paying too much attention to sugar and not enough to fat intake.

A Visit With Mark Feinglos, MD, and Richard Surwit, PhD

May 2014

The Role of Bioinformatics in Diabetes Drug Development-and Precision Medicine

The Big Data Revolution: From Drug Development to Better Health Outcomes?

The How and Why of Stress, Diabetes, and the Brain

Medicaid Expansion Choices Mean Different Care for Poor Diabetics, Depending on Where They Live

The Yin and the Yang of CV Risks in Patients With Diabetes

Stem Cells Create a Therapeutic Niche

Cleveland Clinic Study: Use of Bariatric Surgery Beats Medical Therapy in Diabetes Management

Diet Drinks Linked to Heart Trouble for Older Women

Aims of ACA Get Better Reviews Than Implementation at Cardiologists' Meeting

To Treat or Not to Treat? Questions, Controversies in Prevention

Use of Rivaroxaban Could Trim Hospital Spending in Treating Pulmonary Embolism

Relationship Between Cardiac Issues, Diabetes Gets Plenty of Attention

Study Answers Long-Standing Question About Metformin After Heart Attack

Evidence Builds on Yoga, but No Reimbursement Yet

Inhaled Insulin's Long Journey to Commercialization

A Patient's Opinion of Unmet Need

The Persistent Complication of Hypoglycemia in Diabetics