Project Management Evolves with Technology

The past two years have seen a significant increase in the adoption of electronic data capture (EDC) and e-clinical platforms into the clinical trials process, as financial pressures for increased efficiency have converged with an increasing acceptance of technology in clinical development. This has not been matched, however, with major changes in the clinical trials operating process or a focus on the skills required by the project team to maximize the technology employed.

Paul Bleicher was correct when he stated in 2005 that it is the creativity of people that really moves things forward.¹ Role adaptation within the project team must occur in order to support fundamental process change if the true benefits of technology are to be realized in clinical trials. The traditional processes and "functional silo" approaches will not allow project teams to optimize the benefits of technology implementation.

PHOTOGRAPHY: COMSTOCK, DYNAMIC GRAPHICS, EYEWIRE ILLUSTRATION: PAUL A. BELCI

For the first time, the evolution of technology has made the potential for access to near real-time data a possibility in clinical trials. The real benefit of this technology adoption in terms of increased efficiency of the clinical trial process can only be realized when the teams are provided with knowledge and tools to interpret and act upon those data as they become available. This requires the development of new skills and roles by the project team, particularly the project manager, who must have a thorough understanding of how to utilize the e-clinical information in order to optimize the study and gain maximum benefit from the technologies.

Current status

The various reasons given for declining productivity in pharmaceutical R&D in the recent U.S. Government Accountability Office (GAO) report are said to have affected the length of the drug development process as well as the number and types of drugs being developed.² This report directly reinforces the need for productivity improvements at all stages of the drug development process.

The increasing uptake in the use of e-clinical technologies in all phases of clinical trials is also well documented: An estimated 40% of all new trials are expected to use EDC in 2007.³ Several recent articles looking at the role of different e-clinical technologies support this trend.^4,5 Topics include:

• interactive voice response systems (IVRS) for data capture, central randomization, and drug supply management⁶

• electronic patient reported outcomes (ePRO) to replace and enhance the traditional paper-based patient diaries⁷

• imaging technologies to enhance endpoint assessment (e.g., in measurement of tumor size)

• case studies that provide an overall assessment of the impact of technology on the efficiency of clinical trials.⁸

Much of the reported focus, though, is still on individual point solutions to particular aspects of an outdated trial process with less attention to a potential integrated e-clinical platform and even less to the human dimension. Many organizations running clinical trials are still managed as a series of functional silos rather than a series of multidisciplinary project teams with shared project goals. The functional silos are groups such as clinical monitoring, data management, biostatistics, and medical writing and often communicate poorly and operate with variable levels of cooperation between the groups.

The e-clinical model

Interesting parallels can be drawn between the evolution of technology-based clinical trials and the development of the central laboratory business. The October 2006 ACT article by Jean-Marc Leroux⁹ describes how the creativity of the individuals involved was the vital ingredient in developing a successful technology-based process. The goal was to provide a service that could collect specimens from investigator sites and deliver them to a central laboratory within 48 hours while providing study sponsors with an error rate of less than 2%.

At the time (i.e., 1986), testing for clinical trials involved the use of many local and regional laboratories—all using different testing methodologies, reference ranges, and standard operating procedures. In other words, many point solutions were difficult and time consuming to combine. By early 2006, centralized laboratory services had not only evolved to meet the needs of the industry but the 20-year-old business became 100% outsourced. The four founders of the world's first central laboratory had focused on the required outcome, combining several process changes to reach it, including optimizing the use of available technology.

It could be argued that the e-clinical model is more complex, but the key features are identical since the goal in clinical trials is to collect data from investigator sites and patients as quickly and accurately as possible (see Figure 1).

Figure 1. The key feature of the e-clinical model is to integrate all the data of a trial into actionable information.

Integration is essential to turn data into fully actionable information. A series of independent point solutions result in time being wasted reconciling data between the systems and duplicating specific serial tasks. Reporting from separate systems can also be cumbersome and error prone, as different sets of reports are often compared manually and may not be taken at the same time point as the trial progresses.

The growing acceptance of the Clinical Data Interchange Standards Consortium (CDISC)¹⁰ standards is providing a framework to improve the integration process, but management of the data flow through the project and the different systems being used is also critical. For example, the owning system of each data item must be identified and the timing of data uploads into the integrated database from the different systems determined. There is also still considerable scope for improvement in the mechanisms for preparing and accessing reports that utilize the information after the integration process has been successfully performed.

Potential of real-time data

Technology that makes real-time data a possibility in clinical trials has created the potential for radical process change. Project teams now have access to a range of technology solutions that were unheard of just a few years ago.

This technology provides almost unlimited possibilities to project teams but relies on the creativity of individuals to design and implement solutions and requires project team members to have a broader skill set than what is required for traditional paper-based clinical trials. Project managers particularly need to understand the capabilities and costs of the solutions offered by the technologies at their disposal and be able to visualize the benefits of both individual technologies and their potential for integration with each other.

Real-time benefits

The availability of real-time data during a trial allows team members to stay focused and implement a metrics system.

Team focus. Technology enables clinical team members to stay focused on study management and trial outcomes, but it does not drive the approach since there may be several technology solutions. There is not a one size fits all when applying technology to clinical trials. For example, a 300 site cardiovascular Phase III trial has differing challenges from a Phase I oncology trial in terms of geographic spread, data to be collected, methods of data collection, endpoints, and types of sites involved. Technology competency in the project team is crucial to the success of an outcomes approach.

An example of an outcome to focus on is online access to site data to track enrollment rates in order to decide at what point an extra monitoring visit or additional training is needed. Who makes the decision and on what criteria? What report from the real-time data is needed to make that assessment? How fast must the data be made available to the project team after the patient visit for this approach to be meaningful?

Metrics. With real-time data available, there is the possibility of managing the trial through metrics on an ongoing basis. The team can, for example, assess:

• cycle times, such as time to clean data

• data quality, such as number of errors on specific data points and number of queries at particular sites

• data completeness

• other issues as they arise.

For the above to be of benefit, the project team needs to have access to and the ability to implement solutions to the issues they identify quickly. There is little value in identifying problems and then not being able to fix them in a time frame that benefits the remainder of the study. Solutions may include:

• protocol amendments

• additional site training on the protocol or the technology

• closing poor enrollment sites

• opening additional sites

• adaptation of the monitoring visit schedules according to the volume and quality of data being collected at different sites

• adding booster measures to improve patient recruitment, such as advertising, newsletters, and more frequent site contact.

Real-time data availability not only enables proactive trial management at a relatively tactical and mundane level it also enables the implementation of true adaptive trial designs.¹¹ These designs allow for changes as the trial proceeds based on results so far. Results are watched closely and changes are guided by a complex plan developed in advance, usually using computer simulations. When appropriately applied, they have the potential to reduce patient numbers in a trial and also save time.

Process change

Although appropriate application of technology creates the potential for radical change in the clinical trial process, it has to be proactively managed and implemented. It is too easy to add technology to an old process only to achieve a more expensive old process.

As the outcome of the clinical trial process is to collect, make available, and report the trial data as quickly and accurately as possible, the trial process can be thought of as a data flow. The emphasis in the technology-based trial is then on providing the correct technology solution to facilitate this data flow and providing the tools to monitor its progress. This leads to the potential to focus on managing trial progress during the conduct of the study, using the information that is available as a result of real-time data capture. Such proactive management can lead to significant time savings in the execution phase of the trial and, most importantly, ensures rapid detection of any safety trends or concerns.

Activities such as clinical trial management reports based on real-time EDC (or IVRS, drug supplies, labs, etc.), real-time data, early development of the statistical report template, and introductory sections of the clinical study report for use as data management guidelines should also be occurring in parallel, as should other traditional activities such as serious adverse event (SAE) reconciliation, coding, and data clarification. The goal is: Only the work on the information collected at the last patient visit needs to be performed after that visit, thus leading to rapid database lock and reporting of the study.

This objective is supported by holding periodic formal data reviews with appropriate team members to ensure consistency in data handling conventions, clarity over protocol violators, and agreement on SAEs as the trial progresses (see Figure 2).

Figure 2. With new technology comes an updated approach to trial management-one that allows for rapid detection of safety trends but requires discipline and good communication among team members.

The above approach sounds intuitive but rarely occurs. It takes discipline, firm management, and the ability of the team members to communicate and operate across the traditional functional areas to succeed. The roles of the data managers and clinical research associates, for example, converge during the data monitoring and cleaning process, as so much more can be done remotely through online data review.

Roles and skills

We have seen that when appropriately implemented, technology-based trials provide an environment in which the trial process can be radically improved through the availability of real-time data. It is also clear that the success of both the technology and the process are dependent on moving away from the traditional silo-based structure of clinical project teams.

New roles are emerging and other more traditional roles, such as that of the clinical data manager, are changing. The roles are still evolving, but some of the additional skills required are clear. These include:

• help desk operators

• clinical research associates (CRAs) able to review and query data online

• CRAs who can perform onsite source data verification and trial troubleshooting (more science and less box checking)

• CRAs able to provide first-line technology support to sites

• project managers with a good appreciation of clinical trial technology and how to apply it

• project managers able to manage their study from the real-time data available (rapid problem identification and resolution)

• data managers with the technical ability to provide telephone support on systems for sites in addition to performing online data review and pro-active problem resolution (less clerical tasks and more expertise required)

• clinical programmers conversant in building databases, workflows, entry screens, and edit checks in the technologies being employed.

Some key roles are beginning to merge, such as the data manager with the CRA, forming a broader class of data reviewer. There is also an increase in technology skills required within data management that overlaps with traditional IT roles. The balance of skills needed varies depending on the type of study. A third dimension to this is the technology skills required of a project manager, depending on the complexity of the systems that might be employed in a study (see Figure 3).

Figure 3. The different balance of management/logistics and scientific/therapeutic skills required to manage a highly scientific Phase I study compared with a large, simple Phase IV.12

Most important of all, however, is the ability of the project team to function as a team with complementary roles rather than a series of discrete functions. Optimal technology implementations will force that change to take place.

Conclusion

Much of the focus of any attention that has been paid to the role of project team members in an EDC or an e-clinical implementation has been on data managers and, to a lesser extent, on CRAs. However, it would seem that if technology enables a trial to become outcomes focused and allows a much more pro-active approach to trial management than the traditional silo approach to the project, team structure will be inadequate. Rapid recognition and resolution of issues identified through the availability of real-time data will require communication across the project team members and, in many cases, a team approach to problem solving.

Real-time data technology is causing a convergence of the roles on the project team as it works together to expedite clinical trials. It is therefore incumbent not only upon data managers¹³ to redefine their own roles in light of the changing nature of their work, but the whole project team.

Anne Wiles is senior VP of Data Systems and Processes at AAIPharma Inc., email: anne.wiles@aaipharma.com

References

1. P. Bleicher, "Designing New Trials with Technology," Applied Clinical Trials, April 2005.

2. GAO-07-49 Report, "New Drug Development: Science, Business, Regulatory and Intellectual Property Issues Cited as Hampering Drug Development Efforts," November 2006.

3. D. Borfitz, "Forecast: EDC Money-Making Shifts to Phase II Trials," Bio-IT World eCliniqua, November 2006.

4. Tufts R&D Management Report, "New Drivers of E-Clinical Technology Adoption," 2005.

5. K. Trainor, "Leveraging Technology is Key to Fulfilling the Promise of Automation and Streamlining Clinical Trials," Applied Clinical Trials, March 2006.

6. D. McEntegart and B. O'Gorman, "The Impact of Supply Logistics of Different Randomization and Medication Management Strategies Using Interactive Voice Response (IVR) Systems," Pharmaceutical Engineering, September/October 2005, 36–46.

7. B. Byrom, "Innovative ePRO: Tapping into the Potential," Applied Clinical Trials, June 2006.

8. J.T. Mitchel, Y. Joong Kim, J. Choi, V. Hays, J. Langendorf, S. Cappi, "Impact of IBCTs on Clinical Trial Efficiency," Applied Clinical Trials, August 2006.

9. J.M. Leroux, "The Central Laboratory: 20 Years of Evolution," Applied Clinical Trials, October 2006.

10. C. Rozwell, R. Kush, E. Helton, "CDISC Standards: Enabling Reuse Without Rework," Applied Clinical Trials, June 2006.

11. Eva R. Miller, "Implementation of Adaptive Randomizations for Clinical Trials," CRFocus, 17 (3), March 2006.

12. J. Potthoff, "On Demand Monitoring," IIR Partnerships in Clinical Trials Conference USA, April 2006.

13. T. Pratt, "Data Management: R.I.P. or Brave New World?" Applied Clinical Trials, October 2006.