Patrick Stone is president and lead consultant for TradeStone QA LLC in Austin, Texas (www.tradestoneqa.com). A strong advocate for human subject protection, his specialties include bioresearch monitoring in human clinical and pre-clinical trials. The following interview focuses on the FDA and clinical trials.
Question: What is the latest from the FDA on clinical trials guidance?
Stone:The FDA is focusing on monitoring in regards to electronic records and remote monitoring. They are also focused on contract research organizations (CROs) and the qualifications and training of monitors.
In general, what are most monitors lacking, in your opinion?
They lack the scientific background necessary for understanding a device, biologic, or drug protocol. You do not have to be an MD or RPh, but a good basic understanding of chemistry or biochemistry and body systems is helpful. Monitors also need to understand good clinical practice and what the FDA audits during regulatory inspections—this is a must for doing their jobs and it is greatly lacking.
The most common 483s observations issued are for not following the protocol and missing source documents—both of these problems should be easily found by monitors. Monitors must have a tool or checklist every time for checking primary efficacy and SAE data. Protocol adherence should also always be checked.
What are the top issues for the FDA when it comes to clinical trials—what do they flag the most?
Top issues are CRO and sponsor audits to check the monitor plan and qualifications of individuals conducting the monitor visits. Quality assurance (QA) is also a big issue at FDA because the QA departments of sponsors and CROs do not always include QA throughout the process. QA is more of an end-point check rather than a continuing process from start to finish. The FDA is also trying to conduct more inspections with user fee monies that must be spent. FDA only conducted 551 clinical investigator inspections in 2011—a very low number for the amount of on-going clinical trials. FDA must pick up the pace and train more auditors for this job or outsource.
What do companies fail to understand (or misunderstand) regarding clinical trials?
Companies fail to understand FDA language and interpretation of current guidance. Companies also fail to hire qualified workers and think they can be trained on the job. It is very difficult for a business major or English major to make the switch to monitoring clinical trials with no science background.
That sounds like a crazy thing to do—do you see it a lot?
I have observed too many unqualified monitors reviewing site information and it is a big problem out there. A two-week course in clinical trials does not qualify you to monitor a multi-billion dollar clinical trial. Companies need to hire the most qualified candidates to monitor these studies and, yes, it may involve paying someone more for their qualifications. Companies also need to understand the compliance program guides for bioresearch monitoring (BIMO) in general are not just for clinical investigators, CROs and sponsors.
What are some simple things companies can do to make the process easier and more compliant?
Match your standard operating procedures (SOPs) with the compliance guide and expand on current practices to comply with the CFR. Get qualified individuals to conduct QA third-party audits. Companies try to prepare for every possible scenario and that helps, but they do not always focus on a risk-based approach to these preparations. Everything FDA does is now on a risk-based approach model.
What are the most common problems you see with risk-based approaches?
Vulnerable population studies (for example, children, the elderly, or the mentally challenged) have much more risk than a study involving healthy adults. Sometimes it is the protocol that does not take risk into account, and that does not help either. With risky protocols you must review more data and make sure these patients are treated with utmost care, with safety being number one. Risk-based approach means checking all primary and secondary efficacy data points and SAE reporting. If the study is an infused drug or implantable device, it has more risk so the review must be more in depth, but this rarely happens. Many monitors just breeze in and breeze out with no risk-based approach to their work.
What is the one thing every company should avoid doing when it comes to clinical trials?
Handing over all control to a CRO and thinking it will take care of everything. Companies must be a part of their projects and conduct QA audits. Companies need to verify through QA mock FDA audits that their projects keep patient safety in mind, adhere to the protocol, and confirm test article accountability.
You indicate the FDA scrutinizes subject protection, human subject rights, and data integrity—what do companies need to watch out for here?
Data integrity is the most problematic point of the process. The data does not always match up with source data or primary efficacy data points are missing. Little things can mess up the statistical analysis and throw off a protocol proof. Data will then not be acceptable by FDA reviewers. SAE reporting verification is also still a problem these days. I would say the biggest issues (and this is evident from the FDA 483 metrics) are clinical investigators not always following the protocol and/or not making sure their team is following the protocol. Monitors share the blame for this problem as well.
Following a protocol seems like it should be easy to fix—is it?
It is not easy to follow a protocol, plus if the patient does not cooperate, or go to all visits, or write in their diary, or do what is asked, the study goes south quick. You have to get your patients to follow your directions every day, and that can really be a challenge. Clinical trials are collaborations between FDA sponsor clinical investigators and the patients. Sponsors forget they need the patients help and do not always train this at the investigator meeting or on-site initiation. Protocols can change mid-stream and so do study coordinators. This adds to and compounds the problems of protocol adherence. I cannot tell you how many problems I have observed at a clinic site when the monitors and study coordinators are changed during a trial. This problem does not have a quick fix because it is a moving target. Study coordinators are another weak link in this process. So between training study coordinators and monitors to do the lion’s share of the work, many problems can be fixed. What is the first program to get cut when money is tight? Training for study sites and monitors!
How prominent is the trend toward all-electronic data entry, review, and data processing for clinical trials?
Electronic records are becoming the normal process because it provides an easy access point for QA to check the progress of a study. I would estimate that 40 percent of the ongoing trials have an electronic record element; e-case report forms or e-patient records are becoming more common every day.
You have said that companies conducting clinical trials in the all-electronic environment with e-patient source, e-case report forms, and scanned regulatory records may find that costs go down—would you expand on this?
Remote monitoring is becoming a reality and sponsors can now implement monitoring plans that include fewer on-site visits. Fewer on-site visits mean less travel dollars spent and less time away from clinic staff duties. The biggest stumbling block right now is access to the source documentations due to HIPAA compliance. Clinics do not always want to provide monitors access to their data base. This makes remote monitoring impossible. Centralized monitoring is another means to reduce costs by having one CRO conduct all of the monitoring activities. Monitors can do more audits in a week with no travel, thus making the process more cost effective. Air travel has become a real hassle and a major cost element for monitors, plus it limits how much they can get done in a week’s time. Alternative monitoring is doing the work in a non-traditional method. Alternative monitoring means building compliance into your system so the only result is quality work. Having your computer system compare data sets and flag inconsistent data is also an alternative method to aid in monitoring. A computer system cannot currently take the place of a monitor but it is a tool for cross-checking compliance.
What should companies do to make clinical trials go as smoothly as possible and meet all FDA expectations?
Conduct third-party QA at every step of the process. If QA is just an end point, companies will only get a list of 483 observations to work with instead of a passing grade.