IVD LIFE CYCLE: HOW TO INTEGRATE PERFORMANCE, RISK AND PMS - THE EXAMPLE
Regulation (EU) 2017/746 (IVDR) introduced a change in the way in-vitro diagnostic medical devices (IVDs) must be designed, documented and maintained on the market. One of the most sensitive, and often underestimated, issues is the ability to consistently link performance evaluation, risk management and post-market surveillance activities (PMS and PMPF) throughout the device lifecycle. Often these elements are treated as separate documents and developed at different times creating misalignment, when, instead, consistency, traceability and continuous updating between them is required. The IVDR does not formally introduce asingle 'life cycle model', but distributes the requirements throughout the device life cycle, from the definition of intended use to post-market surveillance. We can schematise the life cycle of an IVD into five mutually dependent phases; in fact, decisions made at the beginning have direct effects on PMS, PMPF, PSUR and technical documentation updates:
Before we look at integration, a brief introduction on performance, risk and post-market surveillance (PMS).
Performance evaluation is not a chapter of the technical documentation, but the scientific basis of the device. Performance has three dimensions: scientific validity, analytical performance and clinical performance. These elements must be consistent with the intended use, justified with adequate data and updateable over time.
Common mistake: treating performance assessment as a 'one-shot' activity, to be completed before CE marking. In reality, performance must be confirmed and monitored even after marketing.
Risk management should not be solely in the service of patient safety. Risk is also closely linked to the risk of underperformance, the risk of false-positive or false-negative results, and the clinical risk of misuse of diagnostic information. This means that the hazards identified in the risk management document must be reflected in the stated performance, that risk control measures must be verified with performance data, and that performance must be shown to reduce risk to an acceptable level. It is immediately apparent how performance and risk are sides of the same coin.
Post-market surveillance (PMS) is not merely a passive collection of complaints, but is a 'system within a management system' that must be:
The post-marketing surveillance system must allow answering questions such as:
The link between pre-market and post-market is the PMPF (Post-Market Performance Follow-up). The PMPF serves to confirm clinical performance in the real-world setting, collect additional data on different populations and verify assumptions made in the pre-market phase. It is not mandatory for all IVDs, but its non-applicability must always be assessed and justified. Typical mistake: declaring 'PMPF not applicable' without a robust scientific justification.
This brings us to practical integration. During a document analysis, also conducted independently, one must look for both the documents as objective evidence and the existence of their cross-consistency. Examples of coherence sought may be:
Let's look at a practical example of integration between performance, risk and PMS. The example given is for illustrative purposes only and was developed as a fictitious case for educational purposes. It does not refer to actual devices, marketed products, specific manufacturers or actual clinical or regulatory situations. Any references to performance, risks, clinical data or post-marketing surveillance activities are for illustrative purposes only and do not represent technical or regulatory evaluations of existing devices.
IVD: Immunological test for the qualitative detection of an infectious biomarker
Declared performance
Intended use: Qualitative immunoassay for the detection of biomarker X in human serum samples to support the diagnosis of infection Y. Reported clinical performance: Clinical sensitivity: ≥ 95% andClinical specificity: ≥ 97%. These values are supported by pre-market studies, scientific literature and analytical validation data.
Direct clinical risks derive from the stated performance, let's look at an example of an identified hazard:
|
Hazard |
Dangerous situation |
Potential harm |
Link with performance |
Risk control measures |
|
false negative result |
unidentified infected patient |
diagnostic and therapeutic delay |
A false negative is directly relatedto clinical sensitivity The risk is only acceptable if the stated sensitivity is maintained over time |
Definition of a minimum acceptable sensitivity value Clear IFU instructions on test limits User training |
Thus,risk "depends" on performance.
Integration into the PMS Plan
Now Post-Market Surveillancecomes into play . In the PMS Plan, the manufacturer defines specific indicators related to that risk and targeted activities to measure the indicators. PMS indicators (KPIs):
targeted activities:
The post-marketing surveillance system (PMS) does not collect data at random, but checks whether the actual performance remains consistent with the claimed performance.
PMPF as a bridge
Assuming the device is Class C or D or uses emerging biomarkers or is based on limited clinical assumptions, the manufacturer plans a targeted PMPF for:
We can find ourselves in a scenario where everything is consistent and therefore the PMS confirms sensitivity ≥ 95% and we do not have a negative trend, so no specific corrective action is required and the claim is confirmed. Or, our indicators inform us of an increase in false negatives in a specific population. That is when integrated system management is triggered. Actions such as, for example, risk management update, intended use limitation, IFU revision or performance assessment update need to be introduced, ensuring consistency between claimed performance, risk and PMS results.
In conclusion, implementing a 'system' and not creating 'documents':
We make it simple!
Contact us: together we will find the simplest solution for your needs.