Many readers may agree that Brazil is the country of soccer (until its recent loss to Germany by 7 to 1 in the 2014 World Cup). Brazil's famous association with soccer galvanized after it won the 1958 World Cup in Sweden and introduced 2 phenomenal players: Pelé and Garrincha. The latter did not become as internationally famous as Pelé, but he was a soccer genius, too, in spite of several significant physical and intellectual limitations. True or not, the story says that just before the match between Brazil and Russia in 1958, the coach gave Garrincha some complicated instructions on how he wanted the player to dribble several adversaries, go deep, then turn to the middle, and shoot on goal. Garrincha listened with concern and asked naively, "Have you set this up with the Russians already?"
I think we are acting like the Brazilian coach as far as the risk-based monitoring (RBM) of clinical trials is concerned. The industry is deciding the moves and plays, but has not set them up with the investigators yet.
RBM is a perfect fit to the current scenario of clinical research for a number of reasons. First, costs of clinical research have risen dramatically in the last years, and the industry is doing whatever is possible to curb them. Second, the study monitoring process, although absolutely necessary, is considered highly inefficient, labor-intensive, and tedious. Third, technology has matured to the point that any clinical information collected from a study participant may be entered into the study database in almost real time, be checked for consistency and protocol adequacy, and statistically analyzed, allowing a protocol manager sitting in headquarters to detect failures and trends before everyone else. Fourth, investigators and research sites are currently treated as service providers, and quality services are mandatory.
The cost of developing a new pharmaceutical product has risen to 2.5 billion dollars, according to the last calculations published by the Tufts Center for the Study of Drug Development (available at http://www.csdd.tufts.edu/). Fifty-six percent are spent in clinical development. In a report from the same group published in 2002, the number was 800 million.1 So, since RBM was proposed, R&D finance saw the chance for big savings in monitoring labor, travels, cycle time reduction, etc. As a matter of fact, initial analysis of projects using RBM have found reductions in monitoring expenses around 25%,2 although further confirmation is pending.
The work of the clinical study monitor is important in many aspects. Monitors identify suitable investigators, they prepare the field for the study initiation, train the team involved, visit the site regularly to ensure that all goes well, and correct things when it does not. Monitors also maintain an open channel between the sponsor and the investigator and make all arrangements for the site to be ready for an inspection by a regulatory agency. The day-to-day activities, however, may be quite tedious, with more than 40% of monitors’ time spent at the site engaged in patient record review for source document verification (SDV). Most projects still require SDV of 100% of the data points. All that effort result in the correction of only 2.4% of critical study information.3 So, again, RBM is more than welcome to point monitors to where the risk is; otherwise monitors may spend hours reviewing non-critical data points.
Information technology has transformed clinical studies more than anything else. Thirty years ago, clinical information was recorded on paper and would be entered into the study database weeks or months after it was generated, and analyzed only at the end of the study. Modern tools of electronic data capture and clinical trials management systems have enabled data review to occur continuously and in almost real time. More recently, robust study analytical tools may detect statistical anomalies which may result in safety signals or other red-flag alerts while patients are being observed. As a result, a new role has gained importance: the centralized monitor. Working directly with the study database, the centralized monitor can identify issues and call the attention of field monitors, and trigger targeted site visits.4
It is easy to conclude, from the standpoint of the industry, that RBM is the way to go: faster, better, less costly. It is also clear what is needed to standardize procedures for RBM: define the risks, create preventive and corrective measures, train the monitors, apply the measures, and collect and analyze results. Mission accomplished? No. We still have to work more with the other side of the counter—the investigator. After all, the investigator is the main person responsible for data integrity and patient safety, which are the main objectives of RBM.
Sponsors and contract research organizations (CROs) wish to work with only the best investigators; however, we still have a significant proportion of sites that do not perform as they should, and a review of 483 forms still shows that most of the inspection findings are related to investigator faults. The obvious conclusions are that, first, there are not enough "best" investigators out there, and second, the training processes of the investigators and sites are ready for an overhaul. Industry and CROs consider investigators as ready and capable of doing their studies, but the fact is that studies have gained complexity and increased the burden over the sites.5 Most research sites do not have the capacity to keep up with new knowledge and regulations by themselves; they need help from outside. That used to come from sponsors, but in the last years, training funds have dried up, or turned difficult to justify in many compliance systems. The situation is aggravated by the transfer of responsibilities from sponsors to CROs. Sponsors usually require CROs to select only the "best" investigators for them, and are reluctant to pay CROs to help developing new investigators. The current approach to investigative sites has created many new pressures: more data points, more difficult patients to recruit, stricter budgets, contracts with performance targets and penalties, and study stop decisions entirely in the hands of the sponsor. As a result, it is not surprising that an increasing proportion of investigators do not want to repeat the experience of conducting a drug trial.6
Studies continue to detect issues only after the fact so many groups are now experimenting with a new and more holistic approach to quality, called Quality By Design (QBD), where quality matters are embedded in the project from the phase of the protocol design until the report of results. The Clinical Trials Transformation Initiative (CTTI) is at the forefront of that (see http://www.ctti-clinicaltrials.org/what-we-do/investigational-plan/qbdqrm). But even in their reports, the investigator’s piece of the process has not yet been thoroughly addressed.
How can we engage investigators' participation and importance?
- Investigators should be an integral part of QBD and RBM initiatives.
- Involve investigators and experienced CRAs in the detection of the operational risks and preparation of mitigation strategies. They know where problems are and how to fix them.
- Train investigators and teams about the perceived risks and customize training activities according to site characteristics and needs.
- The site feasibility process should carefully and thoroughly consider site motivation and training needs.
- CROs must be adequately compensated for investigator training efforts, and develop new sites.
- CRAs should return to the days when they were responsible for site training and performance. Yes, this means they will also need better training as well.
- CRAs should spend more time at the sites during the preparation phase of studies (before site initiation), and document site readiness through direct observation, not by collecting training certificates.
- Savings in CRA on-site time and travels should not be a primary driver for RBM initiatives.
We are definitely on the right path, but there still is a long way to go.
References
- DiMasi JA, Hansen RW. The price of innovation: new estimates of drug development costs. Journal of Heath Economics. 2003;(22):151- 185.
- Coopers PW. Risk based monitoring. Reduce trial costs while protecting safety and quality. 2013; www.pwc.com/us/medtech.
- Sheetz N, Benedict J, Huffman E, et al. Evaluating Source Data Verification as a Quality Control Measure in Clinical Trials. Therapeutic Innovation & Regulatory Science. 48(6):671-680.
- Alsumidaie M. The emergence of the centralized monitor. Applied Clinical Trials Online. 2013.
- Getz K. Assessing the downstream impact of protocol design complexity. Drug Development (Touch Briefings). 2009:93-95.
- CSDD. High Turnover and Protocol Noncompliance Continue to Plague the Global Investigative Site Landscape. 2015 January 15th, 2015; Available from: [email protected].
Eduardo F. Motti is the managing partner of Trials & Training Consult (http://www.trialsntraning.com). He has over 25 years of experience in big pharma and CROs, working in clinical development and clinical operations roles in Brazil and Latin America.