Sponsors are encouraged to take into account the limited capacity of assessors, and submit only high quality, complete applications containing only the necessary changes. Over-reporting should be avoided.
Sponsors should be mindful of the current pressure on the medical profession and should carefully assess the pertinence of adding new subjects in ongoing clinical trials. Absolute priority should be given to clinical trials on treatments for COVID-19 and COVID-19-related illnesses, or trials on serious diseases with no satisfactory treatment option.
The Member States support the submission of large, multinational trial protocols for the investigation of new treatments for COVID-19. In addition, sponsors are encouraged to consider the submission of such applications for an accelerated Voluntary Harmonisation Procedure (VHP) assessment when possible.
Guidance on methodological considerations is being prepared by the CHMP Biostats working party and will be linked to [the EMA Guidance] document once finalized.
Selected EMA Guidance
Consider the following measures in your risk assessment:
Conversion of physical visits into phone or video visits, postponement or complete cancellation of visits to ensure that only strictly necessary visits are performed at sites
A temporary halt of the trial at some or all trial sites
Suspension or slowing down of recruitment of new trial participants
Extension of the duration of the trial
Postponement of trials or activation of sites that have not yet been initiated
Interpretations and Recommendations
The impact of protocol changes on clinical data interpretability needs to be assessed and discussed with regulatory authorities:
Consider alternative endpoints for evaluation
Evaluate the comparability of data obtained digitally or virtually vs with “traditional” approaches.
Implement secure digital data elicitation methods.
Amend data management plans for new data sources.
Trials in advanced stages of recruitment could consider unplanned interim analyses to evaluate predictive power and opportunities for early stopping instead of modification or prolongation.
Consider Bayesian borrowing from historical data to counter trial delays, dropouts and missing data. As far as possible, the borrowed information should come from the same population as that sampled for the trial, and analyses should be performed to demonstrate the similarity of important characteristics between the trial data and the borrowed information.
Use blinded forecasting techniques to predict delays to enrolment, and subsequent impact on trial timelines.
Assessing the Impact of Protocol Modifications
Compliance with the trial protocol should be ensured to such an extent that an ongoing benefit-risk assessment for the clinical trial and its participants is still possible. The impact of protocol changes on clinical data interpretability needs to be properly assessed by the sponsor and the overall evidence generation package could be subsequently discussed within scientific advice with regulatory authorities.
Any protocol changes need to be assessed with regards to the objectives, subsequent planned analyses, and interpretation of the clinical trial results. As per ICH E9 (R1) addendum on estimands and sensitivity analysis in clinical trials, clear descriptions of the benefits and risks of a treatment (medicine) for a given medical condition should be made available. Assess whether any impacts due to COVID-19 need to be considered with regards to what treatment effects are being estimated by the trial.
Consult relevant authorities to agree on protocol amendments and revise the statistical analysis plan accordingly. Where possible, all protocol modifications should be documented prior to any unblinding of trial data.
Localized Data Collection
In case the trial participant cannot reach the site to have critical tests performed, it is acceptable that laboratory, imaging or other diagnostic tests are done at a local laboratory (or relevant clinical facility for other tests) authorised/certified (as legally required nationally) to perform such tests routinely, if this can be done within local restrictions on social distancing. Local analysis can be used for safety decisions.
Collect SOPs and instruments from all new laboratories as well as all measurement validation reports. Consider the feasibility of measurement comparison experiments for sensitive tests.
The measurement methods should be explained, assessed and reported in the clinical study report following ICH E3.
Amend data management plans for new data and new data sources.
Consider design strategies and ways to handle potentially higher measurement variability and missing visits. Discuss with the regulatory authorities.
Temporary Solutions for Remote Access
So-called remote source data verification (e.g. providing sponsor with copies of medical records or remote access to electronic medical records) is currently not allowed in most member states as it might infringe trial participants’ rights. In addition, provision of redacted/ de-identified pdfs files will not be acceptable as it puts disproportionate burden on site staff.
Nevertheless, since the coronavirus emergency situation and containment measures are likely to last for a prolonged period, several NCAs have started to look into possible, temporary solutions related to remote access and conditions for such, providing that methods can be used that restrict access to trial participant records, in line with the principles of necessity and proportionality.
Re-evaluate your current risk-based monitoring plan for increased use of centralized monitoring.
Evaluate technologies for updating site / subject interaction methods.
Consider applying simple validation methods on the effect of changes.