Abstract
Since the birth of Louise Brown in 1978, IVF has evolved from a research-based environment to an established and regulated clinical treatment with around 2 million treatment cycles performed annually worldwide.
3.1 Introduction
Since the birth of Louise Brown in 1978, IVF has evolved from a research-based environment to an established and regulated clinical treatment with around 2 million treatment cycles performed annually worldwide.
In the early days of IVF, success rates were relatively limited. Culture media and materials used within IVF clinics were homemade or developed for other purposes. While research on media for embryo culture increased from the 1950s, it is clear that media used in the infancy of IVF did not fully meet the needs of embryos to develop in vitro.1 Before the implementation of disposable devices, sterilized glassware was used. For handling of oocytes or embryos as well as any micromanipulation technique applied, homemade pipettes were used. No information on endotoxin levels or other possible toxins present in these sterilized and homemade tools was available. Institutes with animal facilities were able to run a Mouse Embryo Assay on products while some others were using the Human Sperm Survival Assay as quality control. Many clinics, however, did not have the possibility to properly quality control the products used. There is evidence of effects on development from that period.2, 3
While the role of a clinical IVF lab is to maintain viability of human gametes and embryos, it soon became evident that gametes and embryos have specific needs or sensitivities and that the products used did not always support requirements.
Development and commercialization of material and products for clinical ART started in the 1980s. At that time, regulation of ART was limited and classification of materials and products lacking.
While standard IVF is a relatively simple procedure, initially performed in a natural cycle, many new techniques have been implemented during the past four decades, adding increasing complexity to the laboratory and increasing the need for specific equipment and tools. Today, IVF labs are usually cleanroom-like environments where equipment and consumables, such as all types of media products, labware, or other single use devices, are fit for purpose, supporting success rates. Consumables used in the IVF laboratory are classified and approved by competent authorities before clinical use. This evolution has certainly contributed to today’s success rates. The aim of this chapter is to describe specific aspects related to development and quality control of products used in clinical IVF on a daily basis, providing information to users to more critically assess product information provided by suppliers but also to challenge suppliers to provide more details on their efforts to provide products with maximum performance and safety.
3.2 Product Development for ART
Products used for treatment of patients are classified as medicinal products or medical devices. For many years it was unclear how products used in the human ART laboratory should be classified. Today, products used when handling human gametes and embryos are in many countries classified as medical device For the user, this means they should look for and use devices with the appropriate medical device approval. For the industry supplying products for IVF it means products must be developed, produced, and approved according to relevant market region regulations before they can be released on that specific market.
In Europe, medical device regulation has been harmonized since the early 1990s in the Medical Device Directive (MDD; Council Directive 93/42/EEC). The uncertainty regarding classification of products used in clinical ART resulted in use of products approved as in vitro diagnostic devices or without approval. To reduce confusion or misunderstanding regarding classification of products used in clinical ART, an EU guidance document (MEDDEV 2.2/4) was developed. Practically it means that products used when handling gametes or embryos are classified as medical devices.
To align with the developments in the medical field, the EU has recently updated the regulation for medical devices. This new Medical Device Regulation (MDR; Regulation (EU) 2017/745) will be enforced as of May 2020 and will replace the MDD and related guidance documents. For products currently approved under the MDD, manufacturers will have to comply with the MDR when the currently valid product CE certificate expires. (CE or Conformité Européenne means the product conforms to European legislation.) For their products to be CE approved, manufacturers of medical devices should also be ISO 13485 certified. This standard describes the quality system requirements for compliance of medical devices for regulatory approval. This means that manufacturers are audited for compliance with the ISO standard as well as for compliance of products with the general safety and performance requirements stipulated in the MDR.
Regulations on medical devices aim to guarantee safety and performance of products used. Authorities approve the products in the higher risk classes before they can be placed on the market and manufacturers are obliged to show that every batch of products released fulfils specifications. This is important for users as they rely on the suppliers of their products.
Development of medical devices is strictly regulated and MDR will further increase requirements and control from authorities on safety and performance of devices. Requirements for approval depend on the classification of the device. In the clinical IVF lab all classes of devices may be used (MDR Chapter V: Classification and Conformity Assessment). Even more than under MDD, post market surveillance will be important. Follow-up of products will be an important requirement and approvals will not be prolonged when evidence on safety and performance of products is not available.
Professionals from the field have suggested a path for development of products and implementation of new technologies.4 The suggested path involves different phases starting with testing on animals followed by preclinical and clinical testing. Interestingly, when looking at the process for development of medical devices, the suggested path is in line with the development path for medical devices described in the relevant ISO standards and regulatory documents.
Development of a new product includes different phases. Once research has provided enough information supporting development of a new product, the design control phase can be initiated. It is this phase that is scrutinized by relevant authorities when a manufacturer applies for regulatory approval. The design control phase includes different steps eventually leading to a final product (Figure 3.1) that can be transferred to the organization for production, testing, release, and distribution. All actions taken to go through this process are documented. Support from users can be required at different stages in this process and clinical investigations for development of new products usually take place during the design validation, which is among the final stages in the product development process (Figure 3.1). When all steps in the development process have been finalized and the documentation work for a product is completed, the approval process starts. Only when a product is approved by the competent authority can a new device become commercially available.
Twenty years ago, product development and product evaluation were very different. Safety and performance are now crucial criteria where evidence is required before a new product can be approved. This is for the benefit of the patient and gamete or embryo exposed to the devices. In practice, however, it also means that it will take several years before an idea can become an approved commercial product. When this path has been followed and a product becomes available, it is then up to the clinic to validate if the device meets clinic requirements or expectations for routine clinical use. When products are commercially available, local validation at the IVF clinic is performed according to the instructions for use. IVF clinics have their own procedures that may deviate from the manufacturer’s instructions for use. Hence, validation by the clinic to collect evidence on the performance of a device is required before clinical implementation.
3.3 Product Validation in the ART Laboratory
As mentioned, the approach during the late 1990s was that new products could be introduced easily. This was also a period where substantial progress in laboratory performance was made and people were quite open to implementing new products. The implementation of regulations, such as the directive on tissues and cells (2004/23/EC), and the increasing introduction of quality management systems mean that laboratories are now more prudent in the introduction of new devices or methods. Today, implementation of new devices involves validation to confirm the performance of the device in the IVF process of the clinic. In practice, this means that the introduction of a new device in clinical routine involves two independent validations, that is during the product development process by the manufacturer and the validation by the user in their own setting.
A clinic’s quality management system normally describes how validation of a new device in clinical routine should be performed. Validation of a new device or method is not the same as a clinical investigation. While a clinical investigation aims to answer a scientific question, a validation aims to examine if performance of a device is according to expectations or requirements. Unlike a scientific study that may investigate a device as part of its development process, a validation aims to implement a product with adequate regulatory approval and therefore has different requirements regarding ethical approval and patient consent.
How to run a validation for implementation of a new device is a frequently asked question. Different aspects need to be considered.5 The type of device affects how a validation can be performed. Validation of a time-lapse system is very different from introduction of a new medium for embryo transfer, a new denudation pipette or another method for oocyte vitrification. Each of them has specific aspects that must be taken into account. Another important criterion is how to run a comparison before introduction of a new product or method. If a device can be tested preclinically first, this should be the preferred track. This may, however, not always be possible. A comparison in the clinic can be made on sibling oocytes, or one can compare between patients. The differences between the two methods are listed in Table 3.1.
| Sibling design | Patient comparison |
|---|---|
| ∎ Comparison on the same cohort | ∎ Comparison on different cohorts |
| ∎ Only patients with sufficient numbers of oocytes/zygotes can be included | ∎ All patients can be included |
| ∎ Avoids effect of patient factors | ∎ Patient factors can have important impact |
| ∎ Not possible to compare outcome except for single embryo transfer cases | ∎ Comparison of outcome possible but effect of sample size important |
| ∎ Affects workload due to duplication of treatment aspects | ∎ More demands on randomization to minimize possible patient effects |
| ∎ Very valuable for assessment of laboratory performance | ∎ Preferred when outcome data are important |
Besides the choice on how to set up a validation, the sample size is another aspect to consider. Ideally, a power calculation should be performed to determine the required number of cases but for validations this is usually not realistic. The possibility to run validations is largely affected by the size of the clinic. In any case, the sample size of a comparison should be large enough to give at least a good indication about the performance of a device. For example, sibling comparisons with fewer than six oocytes or zygotes per patient are difficult to compare. While it is always interesting to perform a statistical evaluation on the findings of a validation, interpretation should always be made with caution when limited sample sizes are involved. Significant differences from underpowered numbers can always be a chance finding. A comparison using a sibling design6 and a patient comparison7 are examples of sufficiently powered scientific studies comparing two products. For a validation, inclusion of at least 50 patients in a sibling design and 80–100 in a patient comparison can provide useful information on performance of a device.
Today, implementing a new medical device means a substantial amount of work to prove the performance and safety of a device. First there is the work done by the manufacturer before a product is approved. Proper validation at the clinic before general use adds to the information on performance of the device and will allow general use with limited risks of failure.
3.4 Quality Control on Medical Devices for ART
For the laboratory, an IVF procedure starts with collection of gametes and ends with the transfer of embryos, either fresh or after cryopreservation. Between the start and the end of a treatment, >100 devices may be used. While exposure to some devices may be only a few seconds, exposure to other devices can last several days. Nevertheless, the quality of each device contributes to the success of a treatment. Hence, the strength of a culture system is no stronger than the weakest link in the chain.
An IVF treatment involves a multidisciplinary approach where collaboration between the clinic and laboratory is required. The laboratory aims to create an environment where fertilization can take place and embryos develop in the best possible conditions. Physical and chemical conditions are important aspects in providing an optimal environment.8 Optimization of temperature and the gas phase contributes substantially. Other aspects are the consumables that gametes and embryos are in contact with. Gametes and embryos are in direct contact with media and culture dishes and have the longest contact time with these products. This can partly explain why these products are often suspected first in case of changes in results. Although gametes and embryos are not in direct contact with it, oil can also be harmful and affect results.9 Besides, any device used can potentially exert a negative effect, independent from its exposure time or whether there is direct or indirect contact. Therefore, quality control (QC) should be equally strict for any device used.
Results of IVF have improved continuously over the years. Part of this can certainly be attributed to increased quality of devices used. As mentioned in 3.1 Introduction, in the early days of IVF no products fitting the exact purpose existed and no adequate QC was performed before using devices clinically. Evidence on suboptimal quality of products used in the IVF lab has been published.10, 11
While gametes and embryos have specific needs, they also have specific sensitivities, imposing specific requirements on products designed and produced for use in clinical ART. It is essential that QC tests identify products of suboptimal quality. Depending on the type of product, different QC tests are used. Common tests for most devices are for sterility and endotoxin, as well as bioassays.
Exposure of medical devices to gametes or embryos as part of a QC program is a logical approach for these types of devices, and so the use of bioassays for human ART was suggested in the early days of IVF.12 Sperm cells of different species have been used in bioassays.13 Today, a Human Sperm Survival Assay (HSSA) is most common.10 For tests including embryos, the Mouse Embryo Assay (MEA) is used.11, 14 Other types of cell lines have been suggested in the past but have not proven to be equally effective as MEA.
Users receive a Certificate of Analysis (COA) listing important specifications of the product and the results of QC testing on the final product. The COA lists specifications considered important for the device and results from the QC testing must be within the predefined specifications. These specifications can be dimensions or other features that are relevant for the product. For media products, pH and osmolality is often included. Furthermore, results on sterility, endotoxins, and bioassays are listed whenever applicable.
For most products, the bioassay of choice is an MEA but for products exposed to sperm cells only, it can be an HSSA. The QC tests listed on the COA have been specified during the development process and are thus part of the suppliers’ quality management system. This testing is performed on the final product. A failure in final product testing, that is when any of the parameters listed on the COA is out of specification, does not allow release of the product which, in a worst-case scenario, can have an impact on treatment of patients. In the late 1980s and early 1990s when commercialization of products for IVF started, sometimes a company could not provide product due to failing QC. This can be overcome by making appropriate QC testing measures during the production process, including for incoming raw materials.
This testing on raw materials can also include MEAs.14 A recent inquiry on manufacturer practices for QC testing showed that only a limited number of manufacturers perform MEAs on incoming material.15 Such testing has, however, important benefits. When testing raw materials before release to the production process, raw materials meeting requirements can be selected. Such a selection process, together with a controlled production process and final product testing supports performance and safety of devices (Figure 3.2). Raw materials can be, e.g., metal for production of aspiration needles, glass for production of micromanipulation or denudation pipettes, pellets of plastic raw material used for production of labware, oil, chemicals used for production of media as well as any product in contact with the raw materials during production processes. Including such raw material testing and selection requires extensive effort but is also efficient as it will exclude inferior raw materials, some already being critical for success in IVF, at an early time point, thereby minimizing risks in quality issues.
