In-house IVD FCM panel or not: it depends

IVDR really has become a buzzword in the medical laboratory, sometimes creating confusion and anxiety over the continued availability of in vitro diagnostic tests for laboratory professionals and hence, for patients. The field of clinical flow cytometry (FCM) is certainly one area where IVDR could prove to be disruptive, as medical laboratories fear that they will no longer be free to deploy their own clinical expertise to build, validate and implement their FCM panels of choice. But what is IVDR – or Regulation (EU) 2017/746 on in vitro diagnostic medical devices¹ – exactly about, and what does it mean for the future of (user-defined) clinical flow cytometry assays?

Introduction

Introduction: scope of the IVDR

Let’s first take a look at what exactly is in scope of the IVDR, and what is not. Simply put, the IVDR governs all products made available on the market of the European Union (EU) that are intended to be used for in vitro diagnostic (IVD) purposes – whereas a ‘product’ may be understood as a reagent, kit, calibrator, control material, instrument, specimen receptacle, software, … Under the IVDR, ‘all products’ means both those IVD products that are commercially available (identifiable by their CE mark) as well as those that are made in-house by health institutions, whereas before IVDR only the first type of products was in scope of the legislation. In the context of this paper, we’ll take medical laboratories (both hospital and private laboratories) as the key example of health institutions. So non-commercial, in-house laboratory tests are now being regulated for the first time ever. The idea behind this is that IVDR is raising the bar significantly for commercial manufacturers, and it would only be fair that non-commercial IVD tests follow suit.

Figure 1. IVDR – What is ‘in’ and what is ‘out’ of scope.

Does this then mean that everything you’ll find and do in a medical laboratory is regulated by the IVDR? As shown in Figure 1, not quite. As part of the CE-marking process, the manufacturer needs to generate evidence that the IVD product is safe and effective when used as intended, in the intended use environment by the intended user. In the context of clinical FCM assays, this last part translates as ‘in the medical laboratory by laboratory professionals’. But the IVDR does not regulate how a medical laboratory should implement IVD tests or qualify employees as being competent to perform the test – this is governed by national accreditation requirements. What is also not in scope of the IVDR are general laboratory products and equipment that cannot be attributed to one or more specific IVD tests or procedures (such as pipets, centrifuges, common buffers like PBS, …), international certified reference materials, and materials for external quality assessment schemes (EQAS). Lastly, the IVDR also does not deal with products for research use only (RUO) with the important caveat that by definition, an RUO product cannot have any medical objective.² Should a medical laboratory use an RUO product for in vitro diagnostic purposes, then this product – sold as RUO – changes into an IVD product made in-house by the laboratory, and as such it is in the scope of the IVDR. So, to conclude this introduction: the IVDR regulates how products for specific in vitro diagnostic purposes are being manufactured and provided to the user, regardless of whether these products are commercially available or made in-house by the laboratory.

IVDR and the ‘manufacturing’ of non-commercial, in-house IVD tests

How to define a laboratory-made, in-house IVD test? Such a test is often referred to as a laboratory-developed test or ‘LDT’ but since this is a strictly defined concept that is regulated by the FDA,³ it is better not to use this term in the context of IVDR. Basically, an in-house IVD test is an IVD assay that is used by the laboratory while the test is not CE-marked or used in a way that is not compliant with the conditions for which CE-marking was obtained (i.e., ‘off-label’ use of CE-marked products – discussed further in this text). Typically, such an assay is (partly) composed of RUO reagents, CE-marked IVD reagents that are used outside of their intended purpose, and/or laboratory-developed critical reagents. Under this umbrella, a significant share of current laboratory tests qualifies as in-house IVD tests.⁴

It is important to underscore that the IVDR explicitly recognizes that not all patient needs may be met by commercially available CE-marked IVD products and that there still is a need for in-house IVD tests. However, the use of such tests is now both restricted and regulated by Article 5.5 of the IVDR which specifies the requirements that laboratories must fulfil in order to have their in-house test(s) partially exempted from the IVDR. Importantly, all of the requirements listed in Article 5.5 and shown in Figure 2 must be met in order for an in-house IVD test to receive exemption from IVDR. Fulfilling these requirements is certainly not a free pass for such tests, and the continued use of in-house IVD tests will require a significant effort for any laboratory. If one or more of these requirements cannot be met, the laboratory must cease to use such an in-house IVD test and switch to a commercially available alternative. That is, unless the laboratory chooses to obtain a CE-mark for their in-house IVD test fulfilling the exact same requirements and conformity assessment procedures as commercial manufacturers of IVD products – an unlikely option for most laboratories, which will not be further discussed in this paper.

With the exception of the relevant General Safety and Performance Requirements (GSPR) set out in Annex I of the IVDR, in-house ID tests that are made and used by medical laboratories established within the EU shall be exempted from the requirements of the IVDR provided that all of the following conditions are met:

1. The tests are not transferred to another legal entity and they are not manufactured on an industrial scale;
2. The manufacture and use of the tests occur under appropriate quality management systems;
3. The laboratory is compliant with standard EN ISO 15189 or where applicable national provisions;
4. The laboratory provides information upon request to its competent authority on the use of such tests, which shall include a justification of their manufacturing, modification and use;
5. The laboratory draws up a declaration which it shall make publicly available, including: its the name and address, the details necessary to identify the tests, a declaration that the tests meet the GSPR and, where applicable, information on which GSPR are not fully met with a reasoned justification therefor;
6. For Class D tests in accordance with the rules set out in Annex VIII of the IVDR, the laboratory draws up documentation that makes it possible to have an understanding of the manufacturing facility, the manufacturing process, the design and performance data of the tests, including the intended purpose, and that is sufficiently detailed to enable the competent authority to ascertain that the GSPR are met. Member States may apply this provision also to class A, B or C tests in accordance with the rules set out in Annex VIII;
7. The laboratory takes all necessary measures to ensure that all tests are manufactured in accordance with the documentation referred to in point f.;
8. The laboratory reviews experience gained from clinical use of the tests and takes all necessary corrective actions;
9. The laboratory justifies in its documentation that the target patient group's specific needs cannot be met, or cannot be met at the appropriate level of performance by an equivalent device available on the market.

Member States may require that such laboratories submit to the competent authority any further relevant information about such tests which have been manufactured and used on their territory. Member States shall retain the right to restrict the manufacture and use of any specific type of such tests and shall be permitted access to inspect the activities of the laboratories.

Figure 2. IVDR Article 5.5 requirements to be met for in-house IVD tests, when made by medical laboratories.

IVDR timelines, applied to flow cytometry

Originally, the IVDR was set to come into full effect on 26 May 2022 – but the COVID pandemic disrupted its operational roll-out and now there is a staggered roll-out for different product categories.⁵ Indeed, the IVDR categorizes IVD products into different risk classes and this classification was used to set timelines for commercially available products. What does this mean for flow cytometry-related products? An overview is provided in Figure 3.

New flow cytometers, FCM instrument setup and calibration products, and buffers or solutions for specific FCM sample processing procedures (e.g., red blood cell lysis buffers or fixative reagents) are typically non-sterile Class A (low individual risk and low public health risk) products; and the timeline did not change for those products; IVDR came into effect as of 26 May 2022, with a sell-off provision (allowing manufacturers to ‘empty the warehouse’ of IVDD-compliant products) until 26 May 2025. FCM kits and reagents are typically Class B (moderate individual risk and/or low public health risk) or Class C (high individual risk and/or moderate public health risk) products, depending on their intended purpose. For Class C and Class B products, although certain requirements such as post-market surveillance were also activated on 26 May 2022, IVDR comes into full effect on 26 May 2026 and 26 May 2027 respectively, each with a sell-off provision of one year. There are also Class D and sterile Class A products, but these seem not to comprise any currently available FCM products.

The timelines for the roll-out of Article 5.5 have also been changed, and if a currently used clinical FCM assay qualifies as an in-house IVD test, these timelines must be followed. Most Article 5.5 requirements will have to be met by 26 May 2024, and the justification that ‘the target patient group's specific needs cannot be met or cannot be met at the appropriate level of performance by an equivalent device available on the market’ is only due by 26 May 2028. However, some requirements have already come into effect on 26 May 2022: the requirements that the tests cannot be transferred to another legal entity and cannot be made on an industrial scale, but also the requirement that the tests must meet the Annex I General Safety and Performance Requirements (GSPR). This is a tough requirement to comply with, requiring significant updates to the laboratory quality management system (QMS) and accompanying extra procedures and records to be generated. So, it is very important to determine if a clinical FCM assay that is currently used in the laboratory can meet Article 5.5 requirements, or not.

Figure 3. Current timelines for the implementation of the IVDR.

In-house IVD FCM panel or not: it depends

What impact does the IVDR have on FCM assays in the medical laboratory? Even though CE-marked FCM kits are gradually becoming more available, most clinical FCM assays are still panels composed out of user-defined single monoclonal antibody cocktails. What are the options for a laboratory that is currently running such clinical FCM panels? Can it keep on running these panels using the same antibody cocktails? Or should those cocktails be modified and if yes, how? Or should the laboratory switch to completely different antibody panels – or maybe to kits? And is the situation any different if the laboratory is already running FCM kits? The answer is a favourite for any person working in regulatory affairs: ‘it depends.’ More precisely, it depends on two questions as shown in the simplified decision tree, Figure 4. The first question is whether the kit that is being used, or the antibodies used for the cocktail, all bear the CE mark. The second question is whether the assay adheres to the intended purpose, provided by the manufacturer of the product(s).

Figure 4. Simplified decision tree to check the regulatory status of clinical FCM assays.

According to the IVDR, ‘intended purpose’ means ‘the use for which a product is intended according to the data supplied by the manufacturer on the label, in the instructions for use (IFU) or in promotional or sales materials or statements or as specified by the manufacturer in the performance evaluation.’ In other words, the intended purpose provides the laboratory the specific context in which the product can be used, the most important elements being:

• For what medical purpose: (aid in) diagnosis, prognosis, prediction, …
• For which clinical or physiological condition: for which target patient groups
• With which specimens: peripheral blood, bone marrow, tissue, …
• With which instruments and other products: this can be strictly defined or more agnostic
• By which user: laboratory professional, required level of training or expertise

Based on the decision tree depicted in Figure 4, there are three possible scenarios that define the regulatory status of a clinical FCM assay.

The antibody cocktail composition of many current FCM panels is defined by the individual laboratory that runs the assay i.e., it is a user-defined FCM panel. If that FCM panel consists only of CE-marked reagents that are used entirely in accordance with their intended purpose as described in the IFU leaflet, then there really is no reason to change anything. Although it is a rumour still being spread, it is a mistake to think that user-defined FCM panels would only be allowed to operate under the Article 5.5 exemption. The laboratory must however check (and maybe doublecheck) whether all antibodies in the cocktail are indeed used as described in the IFU. Are they intended for the same diagnostic purpose? For the same target patient groups? Are they intended to be used for the same specimen type? With the same instrument? Is the assay’s operating principle the same? If the answers to all of these questions is yes, then those CE-marked antibodies have been approved for use in exactly these kinds of user-defined FCM assays. From a regulatory point, the FCM panel is a valid CE-marked IVD test.

Typically, the IFU of such antibodies will contain a limited set of performance characteristics i.e., those that apply on each individual antibody. Indeed, the manufacturer cannot verify/validate the performance of the overall assay since the products are specifically intended for user-defined cocktails. The performance of the assay itself is left up to the laboratory. For example, precision and detection capability characteristics can be determined by the manufacturer for each individual antibody, but this doesn’t provide any information on these same characteristics when applied to the whole antibody panel. It is also not possible to provide relevant clinical performance claims for single FCM antibodies. All assay-specific performance characteristics have to be validated by the laboratory. On the other hand, information on analytical specificity of the antibody, specimen stability, stain stability and potentially interfering substances provided by the manufacturer are also applicable to the antibody panel and require less (if any) verification efforts by the laboratory. So, even though the implementation of a user-defined FCM assay requires more verification and validation activities from the laboratory than does the implementation of a ready-made kit, situation no different from today and has nothing to do with the IVDR. Remember: the IVDR does not meddle with the user of IVD products, provided that the product is being used according to its intended purpose.

If the currently used FCM assay consists of a CE-marked kit, the same set of questions from Scenario 1 must be asked because it is the intended purpose that decides whether the IVDR comes into play for the laboratory or not. If the kit is used entirely according to its intended purpose, the regulatory situation is the same as in Scenario 1: the FCM kit is a valid CE-marked IVD test.

As the kit components of the assay are defined by the kit manufacturer, it is the manufacturer that must establish all relevant assay performance characteristics, which includes analytical performance claims but also clinical performance claims such as diagnostic sensitivity and specificity or expected values in patient populations. This reduces the assay implementation burden for the laboratory to a mere verification exercise. As more and more ready-to-use FCM kits become available, using such clinical FCM kits will become more and more attractive because they should be easier to implement in the laboratory. But again, this is related to quality requirements for laboratories that have long existed and that have nothing to do with the IVDR. So also in this second scenario, the laboratory would not be affected by the IVDR – as long as it uses the kit exactly according to its intended purpose.

The situation only changes when neither of the two previous scenarios are being followed exactly. Even when a kit or a single antibody is CE-marked, deviating from the IFU is off-label use and effectively makes the test an in-house IVD test. From a regulatory point of view, the CE-mark becomes invalid in such cases, and the same is true for RUO-marked products if they are used for diagnostic purposes. Under IVDR, this practice is only allowed if all Article 5.5 requirements are being met. Together with all documentary and regulatory requirements that come with Article 5.5, a major consideration must be given to one specific requirement, notably the last one that will come into effect in 2028, being the justification that ‘the target patient group’s specific needs cannot be met or cannot be met at the appropriate level of performance by an equivalent device available on the market.’ The importance of this requirement cannot be underestimated: it means that the in-house IVD test must prove to outperform any commercially available CE-marked IVD test for the specific purpose that the in-house test is being used for.

Off-label use of reagents comes with more uncertainties regarding their performance, and typically, implementing an in-house IVD FCM assay already requires a lot of verification and validation activities from the lab because of requirements that are driven by the QMS or accreditation provisions. Under IVDR, laboratories will now also have to spend more time on the design (and design control) and manufacturing aspects of the in-house IVD test. These aspects are generally not covered by a laboratory QMS, and this is where IVDR indeed will have a profound impact on laboratories that seek partial IVDR exemption under Article 5.5.

Performance of an equivalent test on the market

The claimed performance of a CE-marked IVD product is – once again – provided by the manufacturer to the user through the IFU. Under IVDR, much more scrutiny is given to the performance claims in the IFU, and the manufacturer must address all applicable performance characteristics, allowing the user to anticipate the expected performance of the IVD product. This encompasses both analytical performance (such as precision, detection capability, analytical specificity, interference, measuring range, specimen stability, …) and clinical performance characteristics (such as diagnostic sensitivity and specificity, positive and negative predictive value, expected values, etc). So, it seems reasonable that a laboratory – in order to seek exemption under Article 5.5 – must compare the performance of its in-house IVD test with the performance claims found in the IFU of an equivalent CE-marked test. The search and/or comparison should be repeated at defined intervals to check continuous compliance with Article 5.5. A decision tree such as Figure 5 can be used for this.

Figure 5. Decision tree to check the presence and performance of equivalent tests on the market. (Image adapted from Spitzenberger F, Patel J, Gebuhr I, Kruttwig K, Safi A, Meisel C. Laboratory-Developed Tests: Design of a Regulatory Strategy in Compliance with the International State-of-the-Art and the Regulation (EU) 2017/746 (EU IVDR [In Vitro Diagnostic Medical Device Regulation]). Ther Innov Regul Sci. Jan 2022;56(1):47-64. doi:10.1007/s43441-021-00323-7; reproduced pursuant to Creative Commons 4.0 https://creativecommons.org/licenses/by/4.0/).

Now, what should be understood by ‘equivalent test’? Even if not entirely defined, two factors are driving the identification of equivalent tests:⁷ they should share the same technology (in this case, flow cytometry) and the same main elements of their intended purpose. Although in-house IVD tests have not yet been put to the test by national competent authorities, it is important to zoom in on intended purpose when it relates to the handling of equivalence. It is reasonable to assume that for FCM assays, equivalence will be determined primarily by the medical purpose and the target patient group but not by the exact antibody composition of the panel, or the instrument with which it is intended to be run.

Take for example an in-house IVD FCM panel that is intended to analyse bone marrow aspirate as an aid in the diagnosis of a specific type of hematologic neoplasia, and that panel makes use of one or more antibodies that are not available as CE-marked product for this specific intended purpose. To find out if this assay can seek IVDR exemption under Article 5.5, the laboratory must check whether there are CE-marked FCM alternatives for the analysis of bone marrow aspirate as an aid in the diagnosis of this type of hematologic neoplasia. The antibody combination wouldn’t matter, nor would the intended flow cytometer with which to use such an alternative – but it should be an FCM alternative. If such an alternative exists, its claimed performance would have to be compared with the in-house IVD test, and clinical performance characteristics like diagnostic sensitivity and specificity would be key in this method comparison.

However, things can get complicated quickly in flow cytometry. Take again the example from the paragraph above. If CE-marked single antibodies are available for the same intended purpose, how can the laboratory justify that none of the possible CE-marked – but user-defined – antibody cocktail combinations have the ‘appropriate level of performance’ when compared to the in-house IVD test? A possible solution could be scientific literature that supports the use of antibodies against those markers for which there are currently no CE-marked options. And it can get even murkier: what if CE-marked single antibodies are available for the same medical purpose but for a different specimen type. Will it be sufficient that the laboratory justifies the use of its in-house IVD test in such cases, purely based on scientific literature? Only time will tell.

Annex I GSPR compliance

Regardless of how the laboratory justification to continue using in-house IVD tests could or should look like, an equally important question is whether it is worth the effort. First and foremost, that particular test will have to comply with the GSPR set out in Annex I of the IVDR. Basically, the GSPR represents a regulatory checklist. Not all GSPR clauses are applicable to each test and some clauses will likely never apply to in-house IVD tests, but Chapter 1 (‘General Requirements’) and several clauses of Chapter 2 (‘Requirements regarding Performance, Design and Manufacture’) are always applicable, as they deal with design and development as well as risk management for IVD products. Just looking at the GSPR, it is already clear that compliance with EN ISO 15189:2013 (Medical laboratories – Requirements for quality and competence) alone will not be sufficient, and the laboratory QMS is recommended to integrate significant portions on design control and manufacture from industry standards such as EN ISO 13485:2016 (Medical devices – Quality management systems – Requirements for regulatory purposes) and EN ISO 14971:2019 (Medical devices – Application of risk management to medical devices). Becoming fully compliant with the relevant GSPR will require a major effort for the laboratory Quality Assurance (QA) department, have a profound impact on staff involved in making and using in-house IVD tests and, therefore, also have an impact on the profitability of such tests. Laboratories may ask themselves first if there is a sustainable alternative available to deliver the same diagnostic service, rather than to ask if the right justifications can be provided to keep running in-house IVD tests. The exemption provided by Article 5.5 will be a last-resort solution for many laboratories – if deemed feasible at all.

Conclusion

All clinical FCM assays are in scope of the IVDR when it comes to how they are being made, but the IVDR does not necessarily have an impact on all laboratories where clinical FCM assays are being run. The IVDR only impacts the deliberate use by the laboratory of products that are not CE-marked, or the use of CE-marked products outside of their intended purpose. When the laboratory uses CE-marked FCM products (kits or single antibodies) in accordance with their intended purpose, these products constitute an IVDR responsibility for the manufacturer and there are no IVDR-related actions required for the laboratory. It is important for any laboratory to check whether their current FCM assays can indeed be considered as CE-IVD tests, especially when the assay is a user-defined panel rather than a kit.

If, on the other hand, the FCM assay qualifies as an in-house IVD test, all requirements from IVDR Article 5.5 must be met by the laboratory that makes and uses such assay. If one or more requirements cannot be met, the laboratory should cease to make and use the assay. A heavy requirement regarding the assay performance isn’t due until 2028, but an equally heavy (if not heavier) requirement is already in effect: compliance with the applicable IVDR Annex I General Safety and Performance Requirements has been mandatory since 26 May of 2022. Meeting this particular requirement will have a profound impact on the laboratory QMS and resource occupation. The time to act is now.

References

1. Regulation (EU) 2017/746 of the European Parliament and of the Council of 5 April 2017 on in vitro diagnostic medical devices and repealing Directive 98/79/EC and Commission Decision 2010/227/EU. Official Journal of the European Union L117, Volume 60, p 176-332
2. MEDDEV. 2.14/2 rev.1. IVD Guidance : Research Use Only products
3. Framework for Regulatory Oversight of Laboratory Developed Tests (LDTs). Guidance for Industry, Food and Drug Administration Staff, and Clinical Laboratories. Draft. (2014).
4. Vermeersch P, Van Aelst T, Dequeker EMC. The new IVD Regulation 2017/746: a case study at a large university hospital laboratory in Belgium demonstrates the need for clarification on the degrees of freedom laboratories have to use lab-developed tests to improve patient care. Clin Chem Lab Med. Jul 21 2020;59(1):101-106. doi:10.1515/cclm-2020-0804
5. Regulation (EU) 2022/112 of the European Parliament and of the Council of 25 January 2022 amending Regulation (EU) 2017/746 as regards transitional provisions for certain in vitro diagnostic medical devices and the deferred application of conditions for in-house devices. Official Journal of the European Union L019, Volume 65, p3-6
6. Spitzenberger F, Patel J, Gebuhr I, Kruttwig K, Safi A, Meisel C. Laboratory-Developed Tests: Design of a Regulatory Strategy in Compliance with the International State-of-the-Art and the Regulation (EU) 2017/746 (EU IVDR [In Vitro Diagnostic Medical Device Regulation]). Ther Innov Regul Sci. Jan 2022;56(1):47-64. doi:10.1007/s43441-021-00323-7
7. MEDDEV 2.7/1 rev.4. Guidance on Medical devices. Clinical Evaluation: A Guide for Manufacturers and Notified Bodies