In this blog series, we examine the impact of COVID-19 on the conduct, analysis, and interpretation of ongoing clinical trials. The safety of trial participants and all staff involved remains the primary concern in all clinical trials. The key message has always been (and remains) that clinical trials are critical and should continue during the COVID-19 pandemic, if it was safe to do so.
If it is safe for a trial to continue, it is important to understand and mitigate against the impact of the pandemic. For all clinical trials affected by the COVID-19 pandemic, the first step should be to perform a risk assessment to assess the impact to the trial and the ability of the trial to still meet its planned objectives. In blinded trials, the risk assessment should be performed on blinded data. This is an ongoing process and should be conducted on a regular basis throughout the duration of the trial. Once the risks are understood, contingency measures/mitigation can then be looked at. It will be necessary to document any changes made to the trial in the protocol and the statistical analysis plan (SAP) and the impact of COVID-19 will need to be addressed in the clinical study report (CSR). It is recommended to discuss and interact with regulatory agencies as early as possible.
Assess the Impact
This will depend on the type of trial, the trial design, the trial phase and stage the trial is at. The main questions to focus the assessment on should include, but are not limited to:
- Does the trial design need to be modified in any way?
- How are trial participants affected by the pandemic?
- Does additional data need to be collected?
- Are there any treatment or study discontinuations and if so, what are the reasons for these?
- How much missing data is there and what are the reasons for it?
- Has the trial population changed?
- Are the original assumptions and objectives of the trial still viable?
- Are alternative methods of data collection required and if so, how will this impact the trial?
- Do the statistical analyses need to be modified?
- Are sensitivity analyses required?
One of the key questions is, does the COVID-19 pandemic change the research question? For most clinical trials designed prior to the pandemic, the primary interest is simply Drug A versus Drug B. However, there may now also be secondary interest in Drug A versus Drug B in the presence of individual COVID-19 infections. Therefore, it is necessary to think about the impact on the treatment effects, which leads to consideration of estimands. The estimands framework is based on linking the trial objectives to the treatment effects of interest (which is specified through the estimand) which helps inform the trial design, the data to be collected, and the planned methods of estimation.
It is also important to understand the extent and pattern of missing data, potential new intercurrent events that may have occurred, how/if enrolment has changed, if the original protocol assumptions need to be reassessed and what the probability of success is for the trial to achieve its pre-specified goals. This should ideally be done via a blinded review of the data, but there may be rare instances where it will be necessary to review unblinded data for the risk assessment and hence, will need to be done using an IDMC in order to avoid introducing additional biases.
Define the Risks
Risk assessment should be started as early as possible and will be a continual process throughout the duration of the trial. It is necessary to understand the extent of participants affected by COVID-19, the number of missed visits and reasons why, study drug interruptions/discontinuations, protocol deviations, as well the impact all of this has on the key endpoints of the trial.
The risk assessment will depend on the type of trial, the trial design, the trial phase, and stage the trial is at. Once the risks are understood, the next step is to think about the implications to the trial in terms of the estimands.
Contingency Measures and Mitigation
Early stopping of a trial may be a possible contingency measure, but it is important to consider the impact of this approach fully. Stopping a trial early may actually do more harm than good, particularly if participants are benefitting from treatment, or if there are no other treatment options. Consideration should also be given to the impact of other contingency measures such as stopping treatment, as well as to issues of treatment supply and distribution, collection of lab samples, etc. The key goal throughout is to minimise the impact on the trial outcome.
Interim analyses are also an important consideration. If interim analyses were planned to be conducted during the trial, it is important to ensure that the required data will be available to conduct the interim analysis and that the data will be of an acceptable quality. Otherwise, it may be necessary to consider the risk of proceeding with the interim analysis with a large amount of missing data versus delaying or not proceeding with the planned interim analysis. The same considerations are also true for a final analysis. If there is a lot of missing data, then there should be discussions around the possibility of increasing the sample size or extending the trial follow-up period in order to counteract the potential effects of the missing data.
Any changes to the planned statistical analyses should be pre-specified and updated in the protocol and the SAP.
in summary, we have looked at general considerations for identifying risks and associated contingency measures and mitigations.
To learn more about two topics that are of particular importance in more detail, check out the other blogs in our series on estimands and missing data.