TGMI is producing practical resources and tools for use in clinical genetic testing laboratories. Such laboratories have to adhere to externally assessed technical and quality management standards that ensure accuracy, consistency and reproducibility.
Several of the TGMI team are involved in clinical testing laboratories in different capacities, which helps us to ensure the applicability and usability of our tools in that setting. The similarities and differences with other aspects of genetic science and medicine are interesting and impactful.
In this blog we highlight what management of uncertainty means in a genetic testing laboratory. In a previous blog we have addressed management of uncertainty in the context of communication in genetic medicine and in using genetic information in disease prediction.
Uncertainty of measurement
Genetic test results are usually considered to be qualitative – i.e. someone has, or does not have, a gene mutation that causes a medical condition. The majority of medical laboratory tests (e.g. lipid levels, blood gases, etc.) generate quantitative results. This means that laboratory accreditation standards have a strong focus on ensuring the correct measurement of quantitative components, with less focus on the qualitative assessment.
This focus can have substantial impacts on laboratories.
Monitoring temperatures in a genetics lab
TGLclinical is a clinical gene testing laboratory, accredited to the ISO 15189 standards, to undertake testing of cancer predisposition genes using the TruSight Cancer Panel. The step-by-step Standard Operating Procedure (SOP) developed to ensure accuracy and consistency in using this method specifies temperature parameters for the storage of reagents and the temperature at which DNA is amplified.
Failure to apply the correct temperature parameters in either of these would inevitably be reflected in qualitative assessments included throughout the testing process. However to achieve and maintain accreditation this is not sufficient. We are required to demonstrate detailed assessment of temperature monitoring. What are the implications for laboratory management?
Defining uncertainty of measurement
Our first task was to write an overarching policy outlining our understanding of the concept of measurement – an interesting task for scientists who not unreasonably believe that this is already embedded in their DNA, so to speak. Once we had “defined” measurement, we then had to specify our understanding of what constitutes uncertainty of measurement. For each of the quantitative measures – in this example, temperature – the degree of uncertainty that could be tolerated for each process where temperature is a specified parameter had to be detailed.
Once we defined the degree of tolerance in the measurement, we needed to ensure the equipment used to make the measurement was able to do so to the level of accuracy required. This is determined through calibration processes. For laboratory management, the complexity of calibration comes at various stages:
- Purchasing equipment. Equipment purchased has to include product specifications indicating the range of measurement offered and the availability of a calibration certificate. This may seem simple and straight-forward. However, you may find calibration has only been assessed over a narrow range. If the temperature level you need to monitor is within the working range of the equipment but outside of the range within which calibration has taken place, then, in the eyes of the accreditation body, it is an uncalibrated piece of equipment.
- Performing calibration. In-house calibration can be designed, but it is complex and time-consuming, so most laboratories use calibration service providers. These providers also need to be accredited to externally assessed standards and it is incumbent on the laboratory management system to ensure that you have documented, up to date, proof of their accreditation status.
- Maintaining calibration. Calibration only remains valid for a specified period of time. In some instances, it is less expensive to purchase new equipment than to have existing equipment recalibrated! In TGLclinical, we have an annual give-away of our freezer thermometers to other teams that can make use of them within less restrictive contexts.
- Documenting calibration. For every item that requires calibration, up to date certification of the calibration has to be kept on record. So a robust document library is needed, to easily locate the necessary record and to monitor when recalibration is due. We also take the extra step of demonstrating, with an electronic stamp, that someone has looked at the certificate and confirmed that the appropriate level of calibration has been carried out.
Once the correctly calibrated item of equipment is in place, the next stage is making sure the temperatures are consistently monitored. A monitoring form is used and needs to reflect the range and tolerances within which temperature needs to be monitored, and be clear about the regularity of monitoring. The laboratory management system needs to ensure that only the latest, approved version of the monitoring form is being used and that all staff are trained on the why, how and when of monitoring.
Recording errors and non-conformities
A further requirement is an unambiguous process for recording any times when temperatures have not been maintained within the correct ranges, e.g. where a freezer has failed. This process needs to accommodate and document the following.
- The exact nature of the non-conformity.
- The implications for tests in progress and in the future.
- The immediate steps taken to ensure results are not impacted by the incident.
- Ongoing evaluation to detect any reoccurrence or need to change equipment or procedures.
- Updates of standard operating procedures to reflect changes necessary to prevent future incidents.
- Tracking and versioning of all changes to documented procedures.
Measurement is ultimately superseded by a qualitative assessment
As you can see, meeting and maintaining external standards requires careful planning and detailed documentation. Although these are meticulously maintained within TGLclinical, the determination of whether the gene test has been successfully completed ultimately relies on qualitative assessments.
Surprisingly there are, as yet, rather limited requirements in laboratory accreditation standards about how to undertake these critical qualitative assessments in genetic testing.
This needs to change. Clinical lab accreditation processes need to focus more attention on the qualitative assessments used in genetic testing, to ensure robust, standardised high-quality gene testing is delivered in genetic medicine.