EVENTS | VIEW CALENDAR
Guest commentary: The clinical trial landscape and its data collection challenges
Clinical trials are one of the most important components of the drug development process. They are research studies that involve human volunteers and are used to test new ways to prevent, detect, diagnose or treat diseases.
The average drug-to-market time is about 15 years and costs a staggering $1.8 billion.1,2 Clinical trials account for 60 to 70 percent of the time and more than 90 percent of the cost incurred in drug development.3,4
During the last 20 years, the complexity of conducting clinical trials has grown significantly. From 1999 to 2005, the average length of a clinical trial increased by 70 percent and the average clinical trial staff work burden increased by 67 percent.5 About 86 percent of all clinical trials experience delays, thereby prolonging the entire drug development process.6 On average, it costs $8 million to the sponsor for each day delay to market.7 Therefore, improving the pace of clinical trials could not only translate into considerable cost savings for drug developers but, could also shorten the bench to bedside time for effective therapeutic drugs.
A major driver of the high cost and long duration of clinical trials is data collection. Traditionally, data was collected using paper case report forms (pCRFs). An estimated $150,000 to several million dollars is spent to manage pCRFs in a typical late-phase trial.2 In a large clinical trial, thousands of pCRFs are collected from hundreds of trial sites. The data is then manually entered into the database. The entire process takes about eight to 12 weeks and, until then, the study sponsor usually has no access to the data.2
Once entered into the database, the data is reviewed, queries are sent to the site and the responses are reentered and rechecked. The data management process is clumsy, time-consuming and error-prone. It requires extensive human intervention starting from interpretation of investigator’s handwriting, double data entry and multiple cycles of manual data review.
With the advent of electronic data capture (EDC) systems, it was expected that data collection would be cleaner and faster. However, most of the currently available EDC systems are web-based and, in most instances, it is not convenient to enter subject data directly into a web-based EDC system while conducting subject evaluation. Therefore, the sponsors print and supply pCRFs to the study sites to provide a uniform source document. This enables the investigator to complete the form while evaluating a subject during a visit and to then transcribe the information into the EDC system at a later time.
This apparently convenient way of data collection has its downsides too. In a recently conducted survey among 598 investigator site personnel, it was found that there was a significant degree of inefficiency in the clinical trial data collection process.8 Almost 75 percent of respondents reported that they still receive documents by overnight courier, fax and email attachment. Fifty-nine percent of respondents track project due dates, status reports and other milestones manually using paper or a whiteboard. Two-thirds of respondents spend two to nine hours per week searching for lost documents. At least once per week, 75 percent of respondents resend documents to CRO or sponsor a second time, because the original document was misfiled, misrouted or otherwise lost in the first attempt.8 Added to these is the fact that sites are usually not reimbursed for this work, thereby further reinforcing negative emotions from study sites, related to participating in clinical trials.
In the past few decades, the clinical trial industry has seen a paradigm shift of data collection activities from the United States and other “first-world” countries towards the developing countries. Factors facilitating this change include the presence of a large heterogeneous patient base, faster recruitment of patients, availability of low-cost CROs, increasing government support and shortage of trial volunteers in North American and European countries.9
As a result of this trend, the global market for EDC systems, which was so far dominated by North America and Europe (60 percent and 24 percent of market share, respectively), will show the highest compound annual growth rate of 19.5 percent in the Asia-Pacific region from 2013 to 2018 as per a recent forecast.9
Unfortunately, in spite of this fast adoption of EDC systems in developing countries, maintaining data integrity in these resource-poor settings is still an issue. As most of the currently available EDC systems are web-based, they are of limited use in developing countries where access to internet is a constant challenge. Therefore, the data capture process in these settings still remains very labor-intensive and manual, resulting in excessive queries, high time to data lock and delayed access to raw data for the investigators, eventually leading to delay in analysis.
Another challenge faced by the current EDC systems is lack of integration with multiple upstream and downstream systems, i.e., electronic health records (EHR), interactive web/voice response systems, clinical trial management systems, corporate safety systems, clinical coding applications, etc. This systems-interoperability challenge forces sponsors to collect data separately in eCRF or pCRF. Achieving effective interoperability between these systems, especially EHR and EDC, can greatly reduce the burden of duplicate data entry, and therefore save time and cost. However, this connectivity remains largely a vision at present because of a number of technical barriers.
Unfortunately, this widely practiced system of capturing data in pCRF (due to convenience, unavailability of internet or lack of systems integration) and later transcribing the information into the EDC system leads to a high reliance on source data verification (SDV). Individuals called “monitors,” whose job it is to verify the accuracy of the entered data against original data sources (such as paper medical records), must travel to study sites around the world to establish data accuracy and maintain regulatory compliance. Currently this accounts for more than 30 percent of clinical trial time and costs.10
Every subject visit is captured and transcribed, resulting in thousands of pages of information. For years, the industry practice was to conduct SDV for all the documents of all the subjects in the study. In some instances, it can take a full day or more to complete SDV of a single subject’s data. This has been driven partly by the overcautious approach to link quality of data to the extent of monitoring and SDV and partly to be on the safer side of regulations. Therefore, the greater the number of source documents, the greater the time and resources necessary for monitoring, which dramatically increases the overall cost and duration of a clinical trial.
According to research by the Tufts Center for Drug Development, which set out to assess how CRA workload has been affected by the global expansion of the clinical trials sector over the last 15 years, CRAs spend 41 percent of their time in on-site monitoring and 20 percent of their time in travelling to and from the trial sites.11 This enormous time and cost of SDV has led the industry to seek for process optimization that would help reduce SDV. Recently, many companies have adopted various hybrid approaches to monitoring e.g. targeted and remote monitoring, real time reporting, adaptive monitoring schedules, etc., to reduce cost.
An integral part of data collection in clinical trials is the informed consent procedure. Obtaining informed consent from research participants prior to all study procedures is mandatory in clinical trials. Unfortunately, subject recruitment and consenting procedures are still manual even in developed countries, leading to variation in ethical standards in research practice. As developing countries in Asia, Africa and South America are becoming hubs for clinical research outsourcing activities, conducting clinical trials in these resource-poor settings are posing several challenges. For pharmaceutical companies, the challenge is to adhere to federal regulations for maintaining ethical standards and data quality.
Noncompliance with government standards and best practices may translate into significant liability. Foreign governments, on the other hand, face the challenge of protecting their citizens from being victims of unethical research practices. Therefore, to avoid compliance issues, even pharmaceutical companies undertaking risk-based monitoring strategies recommend 100 percent SDV of consent forms. Instead of the traditional approach of monitors verifying the original signature on the consent form for each subject at the site, many companies have adopted alternative approaches to make the process more effective and efficient. For example, the study site faxes/e-mails or uploads signed consent forms to an internet portal so that the monitor can access them remotely. However, all these processes are manual, labor intensive and time consuming.
The existing EDC systems have done little to reduce the huge cost of monitoring and data collection that eventually leads to high cost of drug development. Therefore, the need of the hour is for an EDC system that would delve deeper into and analyze the issues of source data generation and minimize data verification, through technology innovations. The solution should be mobile, user-friendly, workflow-based and internet-independent. Additionally, features like medical device and software integration, e-signature of consent forms and biometric verification during subject registration can further reduce monitoring costs and improve data integrity and reliability.
1J. Orloff, F. Douglas, J. Pinheiro, S. Levinson, M. Branson, P. Chaturved, E. Ette, P. Gallo, G. Hirsch, C. Mehta and N. Patel, "The future of drug development: advancing clinical trial design," Nature Reviews Drug Discovery, vol. 8, pp. 949-957, 2009.
2Kalorama Information, "Outsourcing in Drug Development," Kalorama Information , New York, 2012.
3Institute of Medicine, "Transforming Clinical Research in the United States: Challenges and Opportunities," The National Academies Press, Washington, D.C., 2010.
4A. Roy, "Stifling New Cures: The True Cost of Lengthy Clinical Drug Trials," Manhatten Institute for Policy Research, New York, 2012.
5K. Kaitin, "Impact Report: Growing Protocol Design Complexity Stresses Investigators, Volunteers," Tufts Center for the Study of Drug Development, Boston, 2008.
6C. Canavan, "Integrating Recruitment into eHealth Patient Records," Applied Clinical Trials, June 2006.
7inVentiv Health Clinical, "Clinical Trial Educators: inVentiv Health Clinical," inVentiv Health Clinical, January 2014. [Online]. Available: http://www.inventivhealthclinical.com/Collateral/Documents/English-US/Clinical-Trial-Educators_Game-Changers-in-Patient-Enrollment.pdf. [Accessed April 2014].
8INTRALINKS SURVEY, "With Paper, Site Pain: Clinpage," June 2011. [Online]. Available: http://www.clinpage.com/article/with_paper_site_pain/C9. [Accessed April 2014].
9MarketsandMarkets, "eClinical Solutions Market by Products (CDMS/EDC, CTMS, Ecoa , Randomization & Trial Supply Management, Safety), Services, Buyers (Pharma/Biopharma, CROS, Healthcare Providers) & Delivery Modes (Web Hosted, on-Premise, Cloud-Based) - Global Forecast to 20," MarketsandMarkets, January 2014. [Online]. Available: http://www.marketsandmarkets.com/Market-Reports/eclinical-solutions-market-553.html. [Accessed 2014].
10V. Tantsyura, I. Grimes, J. Mitchel, K. Fendt, S. Sirichenko, J. Waters and J. Crowe, "Risk-Based Source Data Verification Approaches: Pros and Cons," Therapeutic Innovation & Regulatory Science, vol. 44, no. 6, pp. 745-756, November 2010.
11Tufts Centerforthe Study of Drug Development, "Study monitor workload high & varied with wide disparity by global region," 2012.
Avik Pal is the founder and CEO of CliniOps Inc. Before starting CliniOps, he worked with two successful start-ups for over a decade. He is also a founding board member at iKure, a social entrepreneurship healthcare startup in India.