The Path to Unified Clinical and Beyond

Jennifer Goldsmith, Senior Vice President, Veeva Systems

Clinical Researcher—January 2018 (Volume 32, Issue 1)

PEER REVIEWED

Jennifer Goldsmith

[DOI: 10.14524/CR-17-0021]

In 1992, the U.S. Food and Drug Administration (FDA) approved just 26 drugs.1 Fast-forward more than two decades, and there have been more new product launches than ever before, reaching a 66-year high with 51 drug approvals in 2015.2 This growth is being fueled by the shift to specialty medicines and precise, impactful treatments that are customized for rare diseases and individual patients.

More than 40% of all drugs in the pipeline are personalized medicines.3 The life sciences industry is flourishing, as researchers and scientists push the frontiers of genomics, microbiology, and diagnostics.

With this accelerated innovation, the number of trials has risen dramatically.4 Meanwhile, the time required to bring new medicines to market remains long, and the effort to develop them has become increasingly complex.5 Sponsor companies are collecting larger volumes of clinical research data, where endpoints are difficult and time-consuming to measure, and patient populations have become more targeted.6 Paradoxically, researchers are making breakthrough scientific discoveries, yet running trials as slow as ever.

The volume, global execution, and rising complexity of studies are creating industrywide challenges. Finding patients and investigators to participate in studies is getting harder, with competition intensifying among sponsors seeking to collect more data and to differentiate their products in the marketplace.6

As data are collected from a variety of sources—and in many different formats—there is a heavier burden on sites and sponsors to manage the collection and review of those data. At the same time, the cost to develop and approve new drugs rises and product revenue growth declines, forcing sponsors to find ways to do more with less.

Adding More Cooks to the Kitchen

In order to drive greater efficiency, sponsors shifted to a clinical trial model that would allow them to add services with greater specialization, as needed. Because of this, outsourcing has continued to rise as a way for sponsors to augment their available resources and leverage the best trial practices from contract research organizations (CROs) to complete activities faster, with greater efficiency, and more predictable costs.

CROs have helped make trials efficient, but adding partners to the mix has also introduced intricacies. Not only are there more stakeholders, but also the use of different processes and systems across sponsors and partners make collaboration and real-time decision-making difficult.

In addition, each time a new trial starts, there is often a completely new trial team. Research shows that sponsors’ simultaneous and inconsistent use of CROs to support project- and program-specific responsibilities is causing measurable operating friction and inefficiency.7

The industry has reached a tipping point; the pressure to innovate specialized products and get them to market quickly on a global scale is not going away. The ecosystem of stakeholders will continue to expand and outsourcing will remain a core part of trial strategies—in fact, sponsors will increase their reliance on CROs to perform essential functions.8

However, the current model will not accommodate the growing number of trials and rise in complexity. A next-generation model is needed to drive new levels of efficiency and effectiveness across the entire clinical ecosystem.

The Era of Unified Clinical is Under Way

A significant transition has started; the clinical research enterprise now recognizes the value and need for its sponsor and site stakeholders to unify their clinical operations to address pressing challenges. The philosophy of “unified clinical” may be described as a single work environment where all teams across the end-to-end clinical ecosystem can execute their workflows in tandem, rather than in isolation with targeted technologies only being exploited late in the game to produce reports.

Unified clinical not only centralizes related documents and data for full visibility, but also makes it easier to align all teams—internal and external. Ultimately, this enables seamless collaboration between the many stakeholders involved in trials today, less redundant effort, and real-time decision-making throughout the study.

A unified clinical environment should include the following five major characteristics:

  1. Common clinical language
  2. Universal and flexible operating model
  3. Collaborative clinical ecosystem
  4. Insights from measurement
  5. Modern and unified information systems

There is growing and industrywide recognition that this type of transformation is needed. In fact, the transition is already under way. As described in the next section, standards organizations have formed as life sciences organizations are more willing to adopt common operational processes, and technology has evolved to the point that it can now support a unified clinical environment.

While companies are taking steps toward developing a common clinical operational platform for all key stakeholders, there is still a long way to go. From sponsors to CROs, from investigators to institutional review boards (IRBs), there is significant opportunity for the industry to continue to come together, collaborate, set universal and flexible operating processes, and break down silos in trials.

A Common Language for Clinical Trial Information

The first step toward a unified clinical environment is harmonizing trial information and operational data definitions.

Today, different data models and vocabularies exist that refer to the same trial documents, content, or information. Language varies from sponsor to sponsor, CRO to CRO, and even functional group to functional group within the same company. Descriptions of protocols, investigator CVs, or data elements use different naming conventions. This challenge makes it difficult to complete work accurately and efficiently or gain insights when the same information is named or categorized differently, and the problem is exacerbated across sponsors’ full portfolio of trials, making it nearly impossible to harmonize the data.

Companies need a consolidated view of all studies to capture insights, regardless of which specialized providers have contributed. To this end, the industry has started to create a common language for how clinical teams talk about information and refer to the same content and data within their own processes.

For example, the Clinical Data Interchange Standards Consortium (CDISC) is defining standards for a common data language to make clinical research more efficient.8 In collaboration with CDISC, TransCelerate BioPharma is also developing clinical data standards to make it easier to exchange and submit clinical research and metadata. TransCelerate is also developing protocol templates with a common structure and language to improve data recording accuracy.9 Likewise, the Drug Information Association created the TMF Reference Model for standardizing descriptions of information that can be adapted to an electronic or paper trial master file (TMF).10

All of this standardization work is critical in continuing to create a common language, so that the puzzle pieces of information and data across trials can fit together.

A Universal and Flexible Operating Model

Equal in importance to standardizing on a common vocabulary is standardizing on a universal operational data model for trials, such as the early days of the CDISC for patient data. In this way, organizations can combine their operational datasets to easily see all of their operational data across portfolios. Organizations want such interoperability, operators expected it, and information technology organizations are frustrated that they cannot easily deliver it.

Fortunately, there is now a tremendous opportunity for sponsors, CROs, sites, IRBs, health authorities, and others to create a common framework that standardizes trial processes, while still enabling flexibility to support individual study needs. An agile operating model can eliminate the rework that takes place with each new trial and reduce the time to full study optimization.

Take, for example, the study start-up process. In general, study start-up has similar, basic steps across the industry. There may be nuances from study to study and country to country, but processes are relatively the same for sponsors and CROs. In contrast, the model for tracking and managing data during study start-up varies, contributing to making it one of the most inefficient activities in a clinical trial.

Imagine if there was not only a common language for clinical information, but also a set of defined milestones during a course of a study and an agreed-upon way to execute activities as part of the study process. A basic industry framework could greatly reduce inefficiencies. Additionally, a consistent data model would better inform sponsors to better plan across their full portfolio of studies.

The key is finding a balance between control and standardization of the processes, while still maintaining a level of flexibility in execution from study to study to accommodate local requirements and procedures, enable the agility to make changes as the study progresses, and address planned or unknown events as they occur.

In another example, a clinical research associate (CRA) usually follows a standard approach for planning, conducting, and following-up a visit during a trial. However, the activities that a CRA executes will vary across trials. In this case, once again, the business process needs to be standardized to a common framework for conducting a monitoring visit, but be flexible enough to manage variability from study to study. The supporting systems also need to be flexible enough to account for changes in study execution, but the data need to be collected and managed in such a way that they can be easily aggregated across the portfolio for historical benchmarking, future planning, and risk mitigation.

Rigid information systems often exacerbate the challenges in adopting a common model, because they limit the flexibility of a business process and force people to find workarounds to complete their tasks at hand. If the systems they work within are not flexible enough, people revert to managing activities and tasks outside the process. Typically, this is when less-than-optimum, manual tools like study trackers in Excel spreadsheets are used.

To achieve a universal and flexible operating model, sponsors, partners, investigators, and others will need to continue to rethink the clinical environment and be willing to adopt common industry processes. Ultimately, the opportunity is to find the most efficient way to conduct trials, and a common process framework would help significantly.

The Collaborative Clinical Ecosystem

Collaboration among sponsors, CROs, IRBs, investigator sites, and others in the trials process becomes easier with a common language and a universal and flexible operating model. However, moving to a unified clinical environment will also require adopting highly collaborative processes and shifting from specialized and siloed processes to ones that fit into a common framework across all trial activities.

Operational silos between functional areas, as well as sponsors and partners, have made collaboration difficult. As clinical trials expanded in scope and complexity, sponsors built dedicated teams internally with specialized functions to handle specific parts of a trial. Over time, these functions developed their own organizational structures, processes, and systems. Teams drove toward efficiency in specialization, with limited visibility into end-to-end trial processes and an inability to conduct effective handoffs between groups.

Similar challenges exist between sponsors and partners; each commonly has their own processes and systems in place, which make information more difficult to share and access. Sponsor companies and their partners should focus on breaking down the barriers that functional and operational silos create in effective trial collaboration.

Further, defining a collaborative process among internal and external stakeholders is important in driving a trial’s efficiency and overall success. Part of that process should include establishing shared goals, objectives, and outcomes to foster collaboration and incentivize teams to transfer knowledge and follow end-to-end trial processes. Also, integrating stakeholders into early decision-making on areas such as protocol design, for example, can build trust among stakeholders that they are working toward the same outcomes.

Insights from Metrics and Measurements

Common operational metrics and measurement are important aspects of a unified clinical environment, in that they provide insights to manage risk, take action, and implement process improvements, as well as allow for industry benchmarking, reporting, and feedback.

Performance metrics give a view into the ongoing status and quality of a trial. More importantly, metrics can help identify overall trends to drive process improvements across a portfolio of studies.

Historically, the most common challenge among investigators, sponsors, and CROs has been the lack of visibility into how trials are performing relative to other studies, making it tough to judge what is working or how to improve trial performance.

A set of common industry benchmarks can provide insight into baseline trial efficiency, completeness, and quality, so that internal and external teams can identify where improvements can be made. For example, the time for a U.S. site (central IRB) activation should take no more than 12 weeks. In examining cycle times, the Tufts Center for the Study of Drug Development found that Phases II to IV took 10.4 months on average (ranging from 7.1 to 13.0 months) (see Figure 1).5 If the data show it is consistently taking longer to complete different phases of studies, or even routine processes within those phases, the potential pain points should be uncovered and actions taken to address the delays.

Additionally, clinical teams can identify possible trends and determine whether a problem is isolated to one study, one site, a therapeutic area, or another common denominator. This type of information then becomes a strategic asset to perform predictive analysis across multiple sites and studies, using real-world evidence and historical operational metrics to better inform trials moving forward.

Looking ahead, sponsors and CROs will make greater use of information and drug development management metrics for robust, predictive analytics to drive operating efficiency, improve feasibility, inform portfolio strategy, and support more effective patient recruitment.5

Modern and Unified Information Systems

Technology plays a critical role in supporting the standards, universal and flexible operating model, ecosystem collaboration, and measurement needed to create an end-to-end unified clinical environment.

Limitations in available technology created the silos that companies are now trying to eliminate. Systems were implemented to support specific functional activities, not end-to-end trial processes. As a result, most clinical teams work in many different systems and often without the benefit of direct collaboration between teams, either internally or externally. The systems also have very different purposes; while one system may manage content, another manages the data being produced. Therefore, content and data are collected and managed from multiple sources, even though the information is all associated with the same study.

Study start-up, for example, is a nexus for many different types of systems (e.g., for electronic data capture [EDC], eSource, eTMFs, and clinical trial management systems). These systems are needed to execute an effective study start-up process, but because the systems cannot communicate directly with one another nor handle either content or data, duplicate information is created and it is difficult to get a full picture of the study process. The administrative burden for inputting data, requesting information, and uploading documentation is also significant.

Now life sciences companies are bringing together previously disparate systems in the cloud to support the end-to-end trial process. Open standards and emerging native cloud solutions allow companies to better support a unified clinical environment and collaboration among internal and external partners. In addition, next-generation cloud applications can manage both content and data to eliminate information and process silos. Clinical information systems that are inherently integrated by leveraging a single platform will be essential in propelling the industry toward a unified clinical environment.

Further, groups are establishing standard technology approaches and platforms to facilitate better engagement and collaboration between sponsors and their partners. The flexibility and collaborative nature of next-generation clinical systems creates an environment where people are enabled to work within their processes, as well as ensures information can be traced and viewed across the trial.

Beyond Clinical and Into the Future

Clinical trials are not an end unto themselves; rather, they’re a critical part of the broader product lifecycle, including issues tied to regulatory oversight, quality assurance, and manufacturing concerns, as well as downstream commercial and medical applications. As the next step forward in the industry, information will need to flow through to all of these other parts of the organization and result in a positive impact that ripples across many areas.

Clinical and other functional groups need to access much of the same information and data throughout the drug development lifecycle. Today, manual processes or complex integrations are put in place to ensure the proper handoff of information between groups. Moving forward, different functional teams will be able to access the same piece of information in multiple ways. In other words, there will be a single source of truth for content and data to be consumed in context, by people who need access, regardless of where they sit in the organization or their role.

For example, a lot of the same information needed in a regulatory submission is also needed in a TMF. Medical writing teams create protocols within a regulatory submission system and, at some point when the protocol document is approved, a copy is made, transferred to clinical, and placed into an eTMF. From that point forward, the information serves as the basis for much of the work that is done around the trial. Information collected throughout the course of the study will eventually need to be sent back to the regulatory submissions team once the study is complete, so the submission can be sent to health authorities for approval.

In order to get this clinical information into a regulatory submission, regulatory operations will typically go through the process of copying many documents, including investigator CVs and the “Statement of Investigator” (Form FDA 1572), which are already part of the eTMF environment. With all this copying and transferring of content, there is no traceability between the different documents. If a health authority makes an inquiry, the original source of information will need to be tracked down, but that source is several steps removed from the final submission process.

In another example, case report form (CRF) documents that are produced as a result of data capture activities with an EDC system also need to go into the regulatory submission structure. The CRF is put into a PDF format and made available as part of the eTMF, which needs to be linked into multiple submissions to health authorities around the globe.

In both of these examples, collecting and processing the data is a manual process today and grossly inefficient, cumbersome, and time-intensive. The need to create copies and transfer them between clinical and regulatory introduces many risks and potential for human error. These are just two examples, but the challenge is much broader and occurs across drug development, manufacturing, and commercialization.

Conclusion

There is a tremendous opportunity for life sciences companies to transform their operations and processes through a unified clinical environment. Moving to this next-generation clinical model will enable the industry to address the growing complexity of trials, support end-to-end trial processes, and drive new levels of efficiency and effectiveness across the entire clinical ecosystem.

Unified clinical is a significant and essential step toward bringing together the entire life sciences environment and reducing the number of steps involved in drug development. Streamlining the process of how data are used—not only within clinical, but also within other areas of the organization—will empower life sciences to bring innovative new therapies and drugs to market much faster.

References

  1. Pink Sheet. 1993. FDA’s 26 New Molecular Entity approvals in 1992. https://pink.pharmaintelligence.informa.com/PS021990/FDAs-26-new-molecular-entity-approvals-in-1992-are-fewer-than-record-30-nme-clearances-in-1991-but-are-improvement-over-agency-performance-from-19861990
  2. Forbes. 2016. 2015 New Drug Approvals hit 66-year high! https://www.forbes.com/sites/bernardmunos/2016/01/04/2015-new-drug-approvals-hit-66-year-high/#42f614c27874
  3. Tufts Center for the Study of Drug Development. 2015. Biopharmaceutical companies’ personalized medicine research yields innovative treatments for patients. http://phrma-docs.phrma.org/sites/default/files/pdf/pmc-tufts-backgrounder.pdf
  4. ClinicalTrials.gov. 2017. Total Number of Registered Studies. https://clinicaltrials.gov/ct2/resources/trends
  5. Tufts Center for the Study of Drug Development. 2016. Outlook 2016. http://csdd.tufts.edu/reports/outlook_reports; see also www.appliedclinicaltrialsonline.com/assessing-practices-inefficiencies-site-selection-study-start-and-site-activation?pageID=3
  6. Tufts Center for the Study of Drug Development. 2008. Protocol design trends and their effect on clinical trial performance. http://csdd.tufts.edu/_documents/www/2816Getz.pdf
  7. Tufts Center for the Study of Drug Development. 2015. Outlook 2015. http://csdd.tufts.edu/files/uploads/Outlook-2015.pdf
  8. Clinical Data Interchange Standards Consortium. 2017. https://www.cdisc.org/standards
  9. Outsourcing-Pharma. 2017. Updated common protocol templates align clinical trial objectives. https://www.outsourcing-pharma.com/Article/2017/05/08/Common-protocol-templates-align-clinical-trial-objectives-endpoints
  10. Drug Information Association. 2015. TMF Reference Model. https://tmfrefmodel.com/about/

Jennifer Goldsmith (jennifer.goldsmith@veeva.com) is Senior Vice President of, and leads the Veeva Vault initiative for, Veeva Systems. PharmaVOICE named her one of the “Top 100 Most Inspirational Leaders” in life sciences in 2011 and, in 2015, she was invited to serve on the editorial advisory board of RAPS Regulatory Focus.

[BACK to Clinical Researcher]