The third, in a blog series about clinical trials for medical device innovators. Read part I: Designing Clinical Trials and part II: Regulatory Considerations for Clinical Studies, here.
Except for small, clinical feasibility studies, unless you are an experienced clinical trialist, do not try to manage a clinical trial on your own. The larger the trial, the more help you will need. Hiring a contract research organization (CRO) to manage the trial may be your best, or only, option, but be prepared for the expense. Similarly, if FDA submissions are needed, you may be wise to seek the help of seasoned regulatory affairs professionals.
Conducting a clinical trial entails a lot of individual tasks. Your study design will have determined the number of patients you need to enroll. How many investigative sites will you need? This decision involves a trade-off between cost and speed of enrollment. Each site costs on the order of $75,000 (2017 USD) to bring up. You may wish to set a minimum number of patients to be enrolled per site, but that is no guarantee that each site will come through for you. Select sites judiciously, which may require on-site screening visits. Be careful, because the inclusion of a site that enrolls no, or only 1 or 2, patients is expensive and demoralizing. Remember, you will need a contract with each site, and each site will need local IRB (Institutional Review Board) review and approval unless you have contracted with a central IRB to oversee the entire study. Currently each IRB submission also usually costs money. You will need to perform an initial site visit, often with some level of training on the protocol and data entry for the investigators and the personnel who will actually be doing the study and collecting the data. Interim monitoring visits are required to ensure compliance with the protocol and to compare source data with data entered in the database according to a predetermined monitoring plan. Closeout visits are also needed at the end of a trial. As data come in, it is rarely perfect. Someone will be needed to review incoming data and prepare queries to the sites as necessary; we cannot stress how important this activity is to ensure the integrity and quality of the data.
Larger, more complex, multicenter clinical trials may also require (or benefit from) one or more of the following management tools.
Clinical Events Committee (CEC)
A group of unbiased physicians knowledgeable in the field of interest, typically charged with adjudicating whether or not individual serious adverse events (SAEs) are related to the use of the device or to the procedure under study.
Data Safety Monitoring Board (DSMB)
A group of unbiased physicians not involved in the trial but knowledgeable in the area. They are charged with developing rules for halting the trial based on anticipated aggregate rates of SAEs.
A central laboratory set up to provide unbiased analysis of specific results (e.g., imaging, measurements, or analysis of all clinical specimens); such a laboratory provides consistency in the analyses. Smaller studies, particularly early feasibility trials, will need this level of oversight only rarely.
ANALYZING THE RESULTS
Once again, smaller trials or observational studies that are simply reporting means and standard deviations and perhaps simple t-tests for comparisons may not require an independent biostatistician. In contrast, more complicated studies requiring complicated statistical analyses, studies with multiple endpoints, and studies requiring computational modeling, are all likely to need a biostatistician with particular expertise in the analysis of complex study designs.
For more information, please visit our website at https://medinstitute.com/.
Continue to follow our blog to view the upcoming Part IV: Cost Drivers for Clinical Trials.