Clinical Trial Imaging: How to Manage Blinded Independent Central Reads and Independent Safety Committees

Joseph Pierro and David Raunig | |

Clinical trial imaging may be performed for eligibility confirmation, determining treatment efficacy or monitoring safety endpoints. You may have seen our previous blog discussing the various blinded independent central read (BICR) methodologies that may be appropriate for these assessments.

Over the years, regulatory bodies have placed more emphasis on the use of imaging in the evaluation of safety during clinical development. This information is used to make dosing adjustments/interruptions or study ending decisions with the overall goal of ensuring minimal risk or harm to subjects.

Potential safety issues may be defined based on:

  • Therapeutic drug class (e.g. anticipated safety signals)
  • Non-clinical studies
  • Clinical studies
  • Post-approval marketing experience data

Information from clinical trial imaging is often compiled into the drug safety monitoring plan to define safety signals which may fall into known or unknown categories. With known categories, it’s somewhat easier to define a more narrow or targeted imaging approach for safety monitoring. But, in situations where the safety signal is unknown, a more thorough imaging approach may be required.

Clinical trial safety assessments are performed in real-time, requiring an active project manager and a well-defined imaging core lab providing timely image evaluation in order to identify potential safety issues. When images are determined to be clinically significant these cases need a process for escalation and/or further verification by two readers in a similar manner to other study data review.

This review process is often dictated by the:

  • Potential severity of treatment on targets organs
  • Relationship to the overall safety question
  • Balance of known and unknown safety signals relative to comparator products/drug classes
  • Regulatory bodies

The overall goal is to improve our understanding of the treatment’s safety and to lower long-term exposure risks for study subjects..

Since ongoing safety assessments may be performed by the principal investigator (PI), here we’re focusing on situations where regulators, based on an elevated concern for safety, request to have site safety assessments confirmed independently via BICR. This is fairly common, and in some situations, regulators have requested that this data also be reviewed along with clinical study data by the independent Data Safety Monitoring Committees (DSMC).

Image assessments performed by the PI are not included in the BICR assessments, whereas both data streams would be provided to the DSMC. One operational model for a safety BICR which often uses a 2 reader plus adjudicator model for handling discordant assessments is designed to report subjects meeting protocol defined safety endpoints. These could include image assessments which met or exceeded the threshold for a clinically significant change which would be clinically important (such as deterioration in organ function) and in this example these cases would then be processed for transmittal and assessment by qualified members of the DSMC.

Regulators view this as a rigorous and robust process to understand the impact on clinical decision-making via a sequential progression and comparison of blinded and unblinded imaging incorporating the available clinical information provided by the PI.  This review can be done with all available information or with only a minimal amount of clinical information so that readers can adequately and objectively perform the evaluation.

These parallel data streams can be merged and presented to the DSMB review which serves as another independent body charged to make a final determination on the clinical importance of the safety observations.

It’s important to note that the latter two assessments performed by the BICR and DSMB are considered the most unbiased and objective assessments. Based on the rigorous central reader processes, the assessments provided by independent reviewers typically have extremely low rates of disagreement between BICR and DSMB reviews. When discordant assessments occurred our adjudication process included a consensus cross-committee review for final adjudication which was prospectively defined in imaging charter and DSMB charter documents.

Click here to learn how the right technology can significantly improve your clinical trial results.


Joseph Pierro is the Medical Director of Imaging at ERT and David Raunig is the Senior Principal Imaging Statistician at ERT.