Tom O’Leary

	This month marks a landmark in the quest for personalized medicine and the growing role of big data in health sciences – from clinical development through tracking patient outcomes long after a therapy has reached the market.

	Fifteen years ago, scientists unveiled the sequencing of the human genome - the genetic map that details the DNA code that defines every human being. The Human Genome Project, which spanned nearly two decades from inception to completion, charted new frontiers in human biology.  The historic project opened the door to a new era of precision medicine that will enable patient treatments to be individualized based upon one’s inherited traits, potentially offering new, more effective treatments, and potentially cures, for some of our most complex diseases and conditions.

	Much work remains to realize the true potential of precision medicine and big data, and the ability to effectively manage and act on it, plays an essential role.  Reverse-engineering the genome requires the rigorous use of data-management, computational and observational tools that only recently came into use - and are still improving.

Biological Passport: Gateway to Precision Medicine

	Genetic testing now gives us the opportunity to pre-emptively identify patients at risk and track incidence of disease onset and degree of progression over a patient’s lifetime.  Health sciences organizations are now increasingly focused on the creation of individual personal health records inclusive of genomic data (e.g. biological passports) and the technologies needed to effectively utilize these data as the vehicle for accelerating the journey to precision medicine.  Just as a travel passport records every country to which you have ever traveled, a “biological passport” records a patient’s health and his/her reaction to environmental conditions and various drugs. 

	The ability to collect unstructured data from social media, along with structured data from genetic sequencing, personal medical devices and the wider internet of things (IoT) has accelerated the variety, volume and velocity of available data.  Often referred to as the “intelligent economy,” this access to large amounts of multi-structured information, combined with the ability to analyze and act upon this information, enhances the ability to deliver individualized patient care, optimizes clinical outcomes, helps reduce the overall cost of care and creates competitive advantage in commercial transactions.  

	These new data sources can provide access to real-time data perspectives on individuals, and populations as well as cohort and system behavior.  A deeper understanding of the genotypic and phenotypic expression of how drugs work coupled with real-time information on individuals’ reaction to drugs is fundamental to personalized medicine.  In clinical trial settings specifically, this type of real-time information on individual participants also provides the ability to monitor drug effects more effectively, supports adaptive study designs, helps improve the overall probability of technical success, reduces cost and helps identify targeted therapies for the right patients.

	Since human physiology remains largely constant throughout a person’s lifetime, the biological passport/personal health record could provide a valuable individualized and granular view of what is normal for each person versus what is normal for a particular population based on a variety of demographics such as age or gender.  As such, with clinical and genomic patient characterization via biological passports, new treatments could be more effectively tested for targeted subpopulations.  This is in contrast to today’s drug testing approach to determine safety and efficacy, in which data are aggregated and analyzed across all patients exposed to an experimental treatment, regardless of genetic or clinical characterization or pre-selection.

	The aggregation of biological passport data across populations creates “Big Data”.  These data provide a treasure trove of information to help define metrics of what does and does not work helping to develop greater precision in defining the best approach to creating an individualized patient treatment plan based on their individual genomic and clinical characteristics and presentation. And as drug development moves from a linear, step-wise pre-clinical to commercialization process to an iterative, bed-side to bench feedback-loop process, more real-world patient care data can be utilized earlier in development to identify better drug targets in disease metabolic pathways for enhanced clinical outcomes. 

	We are seeing significant technological advances – specifically innovation in effectively collecting, storing, normalizing, analyzing, interpreting, and using data from many disparate sources. The challenges do not end there, however, as significant work and hurdles remain, including on the policy front.

	First, privacy policy must effectively evolve with technology. The Health Insurance Portability and Accountability Act (HIPAA) and subsequent federal rules and statutes, require that large-scale genomic studies can be conducted only in tightly controlled “need to know” environments, using encrypted, statistically de-identified data sets.  Each use of data requires separate clearance, and in some cases, separate consent by each individual whose health data is being re-examined. This environment has made it more difficult to conduct collaborative, multidisciplinary research that is vital to advancing precision medicine.

	Interoperability issues continue to impede progress, but are surmountable. The status quo encourages data silos, not cooperation.  The Precision Medicine Initiative, launched earlier this year by President Obama, has the potential to empower policymakers to reboot the system, with universally accessible electronic health records.  It is also important to strengthen “bottom-up” market signals, such as giving patients co-ownership of their health care records, including their genetic results.

	Finally, we must do more to engage and empower health care consumers - the ultimate sources of data. This might include giving data donors access to basic information on the health implications of their genes along with periodic reviews to understand the new findings.

	While the industry has made great strides since the first introduction of the Human Genome Project fifteen years ago, there is still much to be done to truly realize the potential of precision medicine.  Initiatives such as the biological passport, expanded interoperability, and new generations of technology that enable the effective use of big data from sources unimagined only a few years ago, are aligning to position the industry for new levels of success in this life-altering mission.

MaryAnne Rizk, PhD, Global Head, CRO Business Partnerships, Oracle Health Sciences; Tom O’Leary, Chief Information Officer, ICON 

Articles

This month marks a landmark in the quest for personalized medicine and the growing role of big data in health sciences – from clinical development through tracking patient outcomes long after a therapy has reached the market.

Genomes, Big Data and the Journey to Precision Medicine

	There is nothing more disappointing for a pharmaceutical company than to place its bets on a specific molecule/compound in its pipeline and invest billions of dollars in development, only to have the drug fail in late phase trials. Or, the compound  be labeled as a “me too” class drug-meaning the drug is characterized as a follow-on drug instead of a drug that is first in its class-because of marginal efficacy and/or safety profiles.  

	Big Data-specifically data sources like electronic medical record (EMR) and electronic health record (EHR) repositories, comprehensive and integrated clinical repositories, prescription information, real world outcomes data and registries, as well as comparator drug pharmacovigilance, pharmacokinetics (PK) and pharmacodynamics (PD) data-can often guide decision making around pipeline priorities and commercial viability, should the drug be approved.  Predictive analytics and modeling tools-specifically for this type of analysis-need to be integrated into the periodic pipeline review activities to ensure commercial viability. 

	While the benefits of integrating big data into pipeline reviews are clear, achieving this objective has been elusive for many organizations. The tide is changing, however, as sponsors rethink both analytics and their relationships with their research partners.

	The rise of social media, biological sequencing, personal medical devices and the wider internet of things (IoT) provide a huge variety, volume and velocity of available data. Often referred to as the intelligent economy, this access to large amounts of multi-structured information, combined with the ability to analyze and act upon it, creates competitive advantage in commercial transactions.  

	While big data is an increasingly essential component of the discovery and clinical trial processes, sponsors are now turning their sights on leveraging big data as a critical element of pipeline management.

	These new data sources can provide access to real-time data perspectives on individual, group, and system behavior. This not only has the potential to unlock the door to wide-scale attainment of personalized medicine, it also has exciting potential for improving the ability to monitor drug effects in pre- and post-market studies and monitoring.

	Fundamental to any clinical program is the right balance of operational intelligence and clinical analysis to drive decision making and operational effectiveness. Deep operational intelligence/analytics are required not only to provide insight into clinical trial progress and performance, but also to steer the trial in recruitment, site management, and drug supplies. There is increased emphasis on optimizing the clinical trial process and enabling maximum use of data, the industry’s key asset. Better analyses of clinical trial data and optimization of operational aspects (e.g. administrative and financial) of each trial can improve both cycle time and efficiency.

	The challenge is to build a flexible platform that can consume, aggregate, transform, and enrich data in a way that its meaning remains true and it can be accessed via a variety of tools for full data exploration-including pipeline management. Increasingly, sponsors are looking to their research partners to establish new levels of collaborations and co-development with CROs.This further illustrates the increasingly vital and expanding relationships between sponsors and their CRO partners and underscores that the future of medicine will have many authors.

More informed decision making through analytics

	CROs have the potential to be especially valuable partners as their core business increasingly depends on the ability to aggregate and analyze big data-clinical and operational. Instead of rigid data silos that may occur across research, development, surveillance, analysis reporting, and financial reporting, empowering real-time predictive analytics that generate business value is at the forefront of how CROs are driving better clinical decision making with sponsors. Establishing data warehouse capabilities provides a single source of truth to support the acquisition and management of data from multiple studies and data sources into a single, compliant infrastructure for data access, transformation, persistence, and distribution. 

	Forward-looking CROs are focused on building out their infrastructures to effectively capture, consolidate, standardize, and visualize both operational and clinical data from multiple sources. With the goal of providing a holistic view of all study information, from start up to database lock, and incorporating important marketing intelligence, creating a cloud-based clinical data  repository  organizations are rapidly creating the sophisticated infrastructures needed to support it. In addition, CROs are increasingly the central repository for operational and clinical data vital to accelerate drug development. 

Improved R&D efficiency by enabling the use of standards

	By blending operational, quality, efficacy, and clinical safety data across multiple studies and various therapeutic areas, these types of infrastructures can enable a new level of proactive clinical development and deliver true transparency across the complete drug development lifecycle.  This, in turn, enables sponsors to turn data into real knowledge and support better and faster decision making at every stage of the research and clinical development process.  The integration and aggregation of data can be used to provide clear business intelligence to drive portfolio decisions and reduce the risks inherent in conducting a clinical research program. Whether making decisions for adaptive clinical trials based on predetermined milestones or comparing financial, safety, efficacy, and progress information on a clinical program with comparator and outcomes data, CRO partners can leverage such infrastructure and tools to support their decision making.  Moreover, this enables organizations to adopt existing and emerging data standards and benefit from tools and methods associated with those standards. Data in multiple standards (such as CDISC CDASH, CDISC SDTM, JANUS, etc.) can coexist and interoperate with companywide and therapeutic area standards.

	The industry is in agreement that it needs a leaner way to drive innovation throughout the clinical development lifecycle.  Effective pipeline management is a critical part of this puzzle.  Big data is poised to transform this process by giving life sciences organizations the power to gaze into the future and predict with greater speed and accuracy, which candidates will be most effective and successful in the market.   

Tom O’Leary, CIO, ICON and MaryAnne Rizk, PhD, Global Head, CRO Business Partnerships, Oracle Health Sciences.

There is nothing more disappointing for a pharmaceutical company than to place its bets on a specific molecule/compound in its pipeline and invest billions of dollars in development, only to have the drug fail in late phase trials.

Tom O’Leary

Articles

Genomes, Big Data and the Journey to Precision Medicine

Big Data, Big Impact on Business