The Analytical Performance Report (APR) – Central component of the performance evaluation according to IVDR
Regulation (EU) 2017/746 on in vitro diagnostic medical devices (IVDR) places increased demands on the performance evaluation of IVD products. A central element of the technical documentation is the Analytical Performance Report (APR). It documents the extent to which the product fulfils the analytical performance as defined in Annex I (Section 9.1, letter a) and Annex XIII, Part A, Section 1.2.2 of the IVDR.
Overview and definition of the analytical performance
Analytical performance describes the ability of a device to detect, measure or recognise a specific analyte correctly, accurately and reliably – under defined conditions. It represents the first part of the performance evaluation according to IVDR and forms the basis for a later evaluation of clinical performance. According to Annex XIII, Part A, Section 1.2.2 of the IVDR, the manufacturer is obliged to demonstrate the analytical performance of his device by means of appropriate studies – unless he can conclusively justify why certain evidence is not applicable or possible. This applies in particular to products without direct analytes or with novel biomolecular target structures for which no reference methods or certified reference materials are available.
The aim of the analytical performance evaluation is to use technically validated measurement methods to ensure that false or false-negative or false-positive results due to methodological weaknesses are avoided.
While the clinical performance evaluation makes statements about the validity of the test in a medical context, the analytical performance forms the mandatory basis for the technical reliability of a test result.
Typical parameters for analysis performance
The following parameters according to Annex I, section 9.1 (a) of the IVDR are typical test aspects for analytical performance – however, they only need to be addressed if they apply to the device:
- Analytical sensitivity: determination of the limit of detection (LoB, LoD, LoQ), e.g. for quantitative assays.
- Analytical specificity: Control of possible interferences (e.g. haemolysis, lipaemia, cross-reactions with structurally similar molecules).
- Trueness: Deviation from the true value (reference value).
- Precision: Repeatability under identical (intra-assay) and reproducible (inter-assay) conditions. Sometimes indicated by the standard deviation.
- Accuracy: Result of trueness and precision.
- Measuring range and linearity: Range in which the test provides valid results, including verification of linearity.
- Cut-off value: incl. validation of threshold values and sample stability/sample handling.
Not all products require all detections. For example, a qualitative rapid test will not require a quantitative measurement range. In such cases, the APR must clearly explain why certain parameters are not applicable or not technically relevant. The justification is based on the risk classification, the intended use and the claim.
Differentiation from diagnostic sensitivity/specificity
In contrast to diagnostic sensitivity (proportion of correctly recognised sick people) and diagnostic specificity (proportion of correctly recognised healthy people), analytical sensitivity and specificity exclusively evaluate the technical ability of the test to provide correct measurement results – regardless of the clinical significance. The aim is to minimise false or false-negative or false-positive measurement results at a technical level, e.g. by avoiding interference, saturation effects or incorrectly set cut-off values.
Content and structure of the APR
The APR consolidates all analytical data of a product and answers the question of whether and to what extent the respective analyte is suitable for identifying the underlying disease or physiological condition.
Examples from practice
- Quantitative immunoassay for the determination of C-reactive protein (CRP):
For this test, for example, the limit of detection (LoD), linearity range and precision are crucial. Validation could be carried out using referenced human sera, standardised calibration curves and repeatability and reproducibility studies. - Molecular diagnostic PCR test for the detection of influenza A/B:
Here, for example, the focus is on analytical sensitivity (determination of the lowest detectable viral load) and specificity (e.g. exclusion of cross-reactions with other respiratory pathogens). In addition, trueness could be proven by comparison with an established gold standard PCR method. - Urine test strips for multiparametric analysis:
These tests particularly require testing for analytical specificity, for example by evaluating possible cross-reactions between chemically similar substances such as ketones and glucose. The influence of interfering factors such as pH value or density on the colour change should also be taken into account as part of the robustness test.
 Special case: Class A IVD products without an analytical target value
Products such as centrifuges, reagent containers, buffer solutions or incubators fall under Class A according to IVDR Rule 5. These products do not have their own analytical target value, but contribute to the overall performance of the diagnostic system.
A centrifuge as a class A IVD without direct analytical performance serves as an example to illustrate this:
A centrifuge processes samples pre-analytically, i.e. it has a decisive influence on the quality of the sample, but not directly on the measurement of an analyte. Nevertheless, a technical performance evaluation is necessary, including
- Verification of the speed using a calibrated tachometer
- Timer accuracy with defined running times
- Maximum load under routine conditions
Compliance with these technical parameters indirectly supports the analytical performance of the subsequent test – especially for sensitive analytes (e.g. cellular components that would be destroyed by incorrect centrifugation).
However, a classical analytical performance evaluation is not applicable for these products.
In such cases, the IVDR (Annex XIII Part A 1.2.2) allows a justified exception if analytical parameters are not objectively applicable.
In these cases, conformity can be demonstrated by technical tests, bench tests and risk-based reasoning, analogous to the approaches described for Class I devices in the MDR.
Integration into the technical documentation
The APR is an integral part of the Performance Evaluation Report (PER) and must be harmonised with the Performance Evaluation Plan (PEP). Further references exist to:
- Summary of Safety and Performance (SSP) – for class C and D products.
- Instructions for Use (IFU) – all statements in the IFU must be supported by data in the APR or Clinical Performance Report (CPR).
- Parameters from stability reports (ISO 23640, if applicable under real transport and storage conditions) are included in the APR.
The performance data must be suitable to fully address the safety and performance requirements according to Annex I of the IVDR. A conclusion on the analytical performance must be documented in the APR.
Conclusion
The Analytical Performance Report (APR) is a key element of the IVDR-compliant performance evaluation. It proves whether a product works analytically valid – or, in the case of Class A IVDs, whether it is technically suitable to support the performance of the overall system.
The aim of the analytical performance is to avoid technically incorrect results and thus ensure the reliability of the diagnostics.
A well-founded APR is based on planned, comprehensibly documented and statistically analysed data. It is the central basis for a conclusive, auditable and successful conformity assessment.
seleon can support you in planning and creating an IVDR-compliant performance evaluation!
