External quality assessment (EQA) and alternative assessment procedures (AAPs) in molecular diagnostics: findings of an international survey

Introduction

Molecular diagnostics is one among the fastest developing disciplines in the modern laboratory medicine. Molecular genetic diagnostics has some characteristics which are generally referred to as genetic exceptionalism and which must be taken into account in quality assurance [1], [2]. Additionally and in contrast to other disciplines, molecular diagnostics offers an extremely wide range of detection options for a particular parameter such as identification of sequence variation or the detection of infectious pathogens. For example, a whole range of different molecular genetic genotyping methods can be used to detect a particular sequence variation like FV-Leiden, and these methods can be based on laboratory-developed tests (LDTs) [3].

These different methods require different strategies and procedures for quality control (QC) and must be incorporated routinely into a common quality hierarchy system for the laboratory. This includes a quality management system, standard operating procedures (SOPs), alternative assessment procedures (AAPs), internal quality assurance, validation of test results and external quality assessment (EQA).

EQA, also known as proficiency testing (PT), is a key strategy for the comparison of analytical performance between laboratories [3], [4], [5], [6], [7], [8], [9]. Participation in EQAs is a prerequisite for accreditation according to the International Organization for Standardization (ISO) guidance documents 15189 and 17025 and the Clinical and Laboratory Standards Institute (CLSI) QMS24 documents. Additionally EQAs are mandatory for the performance of human genetic tests in some countries [10], [11], [12]. When EQA schemes are not available to the laboratory, AAPs may be used to provide additional quality assurance for the laboratory and are required in this instance by some accrediting organizations. AAPs may also enhance or augment the quality management program [10], [11], [12].

In order to clarify the standardization for molecular diagnostic tests at an international level and in practice, the Committee for Molecular Diagnostics of the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC C-MD) published previously the results of an online survey addressing the post-analytical phase in molecular laboratories to extend its efforts to provide harmonization and standardization for molecular diagnostic testing [13], [14]. As a follow-up to these efforts, we conducted here a survey among voluntary molecular diagnostic laboratories to identify the quality assurance methods used. The generated data provide basic information on current practices in molecular diagnostic laboratories in more than 30 countries. The results of this study intend to stimulate a broad discussion in laboratories offering molecular genetic testing on current quality procedures and practices in molecular diagnostics and serve as a basis for future queries.

Materials and methods

The survey was designed to capture data on practices in quality assurance for molecular diagnostics. Questions were proposed, discussed and approved by the members of the IFCC C-MD and were deemed important.

The online survey addressed the following sections: (1) laboratory demographics, (2) LDTs, (3) EQA and (4) AAPs. Additionally, participants were asked to designate areas of molecular reporting that are in need of standardization. The survey consisted of 30 questions and had both multiple choice and open answer formats. The survey took approximately 25 min to complete and the data generated were collected using Surveymonkey.com [15]. To increase the number of participating laboratories, we offered a Spanish and an English version of the survey. More than 500 laboratories were invited to participate in a call distributed through the national societies of the IFCC member states. The survey started in November 2018 and was opened for participation for a duration of 8 weeks.

Results

Discussion

QC in a laboratory requires the establishment of a quality management system that includes various procedures including SOPs, AAPs, internal quality assurance, validation of test results and external quality evaluation. In order to investigate which procedures are implemented in the laboratories on an international level, the C-MD of the IFCC has prepared a survey highlighting relevant topics.

The available data show current practices in molecular diagnostic laboratories in more than 30 countries. In addition, the data from the study identified several aspects of quality assurance such as AAP and EQA. For the majority of the laboratories, more than 75% of the parameters/analytes offered are available from EQAs, but 24% of the participating laboratories state that they only perform EQA on 25% of the parameters/analytes performed in their routine diagnostics. There are plenty of opportunities for improvement and may be simply due to non-participation in EQAs or due to the fact that EQA schemes or EQA providers are not available for defined parameters/analytes. In order to mitigate one reason for non-participation, the C-MD has installed a listing of well-known EQA providers on its webpage and specified their offered analytes [17]. Other reasons for omitting EQAs is that EQA providers do not supply to corresponding countries due to logistical issues and/or the lack of adequate financial compensation for schemes. The cost burden to participating laboratories may also be too high for regular participation in the EQA qualifying interlaboratory comparisons.

While participation in the EQAs is now widely accepted in the laboratories, our data show a more heterogeneous picture for participation in AAPs. This may be because, depending on the parameter, AAPs are more complex in their implementation and there is a lack of clear recommendations regarding which kind of AAPs should be employed for the wide variety of molecular testing methods and diagnostic assays currently being offered by molecular laboratories. AAPs as a quality assurance measure are carried out differently by the participating laboratories. While 50.6% of the laboratories perform AAPs only if no EQA is available, 20.6% of the participants always do so and even if an EQA is available for the parameter. Interestingly, 22% of laboratories perform no AAPs.

Although best practice guidelines exist, a lack of clear recommendations for implementation of AAPs may lead to a quite different extent of samples which are included in AAPs. While 17% perform AAPs once after installation of a new diagnostic assay, 32% of the laboratories implemented AAPs twice a year per parameter. The majority of laboratories include less than 10 samples in an AAP, and 18.8% of the laboratories include more than 10 samples. The responsibility for the review and approval of the AAPs is obviously organized differently in laboratories as well: While in 41.4% of the laboratories, the head of department and in 12.6% the clinical pathologist is approving the results of the quality measurement, we found that in 13.8% the technical supervisor is checking the results of the procedure. This demonstrates that AAPs are not clearly defined and need characterization and standardization.

The limitation of this study is certainly the fact that the majority of the participating laboratories are from Europe. There is also an imbalance within Europe, as eight answers came from German laboratories and only one from the UK or Italy. The participation for some countries (i.e. Italy) was lower than in a previous study [18]. A similar imbalance was evident in the results from South America: 10 laboratories answered the questions from Argentina and only one from Brazil, probably due to the languages offered in this survey. In order to increase the number of participating laboratories, consideration should be given to re-inviting them to participate or extending the length of the survey.

Another limitation of the study is that substantial parts of the results may also depend on economic and health policy conditions within the countries of the participating laboratories. Having in mind that participation in EQAs is only mandatory in some countries it is understandable that some parts of the results of the survey can be traced back to this. Additionally, the study does not take into account how financially costly comparable quality assurance measures exist for individual laboratories although these will also have an impact on the results of the survey.

Furthermore, the listing of proficiency test providers is influenced by the origin of the participating laboratories. The C-MD wants to investigate further the topics of quality assurance through AAP and EQAs. It will be interesting to find out how for instances quantitative assays are established in AAP within the laboratories, which reference materials are used therefore or which guidelines are followed in establishing the assays. To this end, the upcoming surveys will also be placed for other countries in order to define the need for quality-improving measures more concretely and to address the current status quo in low-income countries as well.