Monitoring the Integrity of Clinical Serology Data

26th July 2016 (Last Updated July 18th, 2018 09:15)

Shekema Hodge, Sanofi Pasteur, explains ways to minimize data integrity risks associated with outsourcing clinical serology testing services

Monitoring the Integrity of Clinical Serology Data

As the number of clinical trials increase, the number of clinical specimens that need to be analyzed increases in kind. The ability to test this increased number of specimens within an organization is not always feasible due to internal capacity and resource limitations. Therefore, clinical testing activities tend to be outsourced to Contract Research Organizations (CROs). A Sponsor may transfer any or all of their trial-related duties and functions to a CRO, but the ultimate responsibility for the quality and integrity of the trial data resides with the Sponsor. Additionally, the outsourcing of clinical serology testing adds another layer of separation between the activities required to generate clinical test data and the Sponsor's ability to ensure the generation of quality and consistent data. This situation makes monitoring the integrity of clinical test data generated at a CRO paramount.

In establishing a data integrity monitoring program for implementation at a CRO, critical components that impact data integrity must be identified and monitored at an established frequency, with pre-determined acceptance criteria. Critical components identified for your consideration include: analyst, reagent, and method qualification processes, as well as method performance and data monitoring procedures. It is also of importance that the aspects of the data integrity monitoring program be communicated to the CRO and that Sponsors provide the necessary training so expectations are understood.

For Sponsor validated methods, whenever possible, CRO analysts should participate in onsite training and qualification on the method. Onsite training is encouraged as it gives the CRO the opportunity to see laboratory setups, testing plans and workflows, and the Sponsor's best practices. These opportunities increase the probability of the CRO completing a successful system suitability check at its facility prior to execution offormal method transfer activities. For CRO validated methods, it is advised that Sponsors review the CRO analyst's training records to determine whether training is being provided and documented as per the CRO's training procedure. In addition, a Sponsor should also review the CRO's validation package and perform a gap assessment against current method validation standards. Any gaps identified should be addressed via supplemental validation activities that have been approved by the Sponsor’s quality unit and, in some instances, by regulatory authorities.

A similar process is also applicable to critical reagents whereby the Sponsor qualifies a reagent and the qualification is subsequently verified at the CRO. This is an important process since environmental factors may influence the performance of a reagent resulting in non-concordant results. Any critical reagent qualification performed at the CRO should be reviewed by the Sponsor to determine adequacy, and in some cases verified at the Sponsor's laboratory. In all instances, the activities need to be documented in a compliant manner via a technical or similar report thatwill allow traceability of reagent qualification status, as well as easy accessibility for auditing purposes.

Furthermore, method performance monitoring is an important process to ensure the assay at the CRO continues to perform as expected since the time of the transfer or validation. Performance monitoring is based on three key points: 

  1. Method appropriate randomized sample panels,
  2. Pre-established acceptance criteria, and
  3. Monitoring frequency.

It is recommended that prior to implementing a performance monitoring program at a CRO, a Sponsor should assemble appropriate performance sample panels along with documentation outlining the origin, storage conditions, and characteristics of the samples within the panel. This documentation should include statistical analyses used in establishing the target sample values. Additionally, there must be pre-defined criteria used to determine whether the results generated at the CRO are acceptable or notand a pre-defined monitoring schedule. In our experience, it is best to have a process flow algorithm mapping out the process and decision making points as a guide as it removes subjectivity from the process and allows adherence to a process approved by the Sponsor’s quality unit. Pre-defining the performance monitoring frequency is key as it reduces the risk associated with foregoing or delaying monitoring activities due to testing demands with shortened timelines.

It is important to actively monitorthe CRO generated serology data as well. Raw data should be audited to ensure that correct results are being reported into the Laboratory Information Systems (LIS) housed at the CRO and that there are no transcription errors or misaligned results.  Raw data should also be monitored from a technical point of view verifying that the method was performed in accordance with the Standard Operating Procedure (SOP) by qualified analysts using qualified reagents and calibrated equipment. For multi-analyte methods or multi-method studies, it is recommended that a scientific review be performed on the blinded dataset to identify any abnormalities within the dataset. 

Any data integrity monitoring program implemented by a Sponsor should be a continuous process and mustbe revisited and revised based on opportunities for increased efficiencies, innovation implementation, and general feedback from regulatory authorities and industry peers.

 

*Shekema Hodge, Ph.D, is the Deputy Director, Sanofi Pasteur