Screening assays (2024)

3.1. TYPES OF ASSAY

Various types of assay have been developed for use in blood screening over the past three decades. The assays most commonly in use are designed to detect antibodies, antigens or the nucleic acid of the infectious agent. However, not all assays are suitable in all situations and each assay has its limitations which need to be understood and taken into consideration when selecting assays.

The main types of assay used for blood screening are:

• Immunoassays (IAs):

  —

  Enzyme immunoassays (EIAs)

  —

  Chemiluminescent immunoassays (CLIAs)

  —

  Haemagglutination (HA)/particle agglutination (PA) assays

  —

  Rapid/simple single-use assays (rapid tests)

• Nucleic acid amplification technology (NAT) assays.

In the context of blood screening, appropriate evaluation is required in selecting the type of assay for each TTI, based on critical assay characteristics, such as sensitivity and specificity, as well as cost and ease of use.

3.2. SELECTION OF ASSAYS

The selection of appropriate assays is a critical part of the screening programme. Reliable results depend on the consistent use of well-validated and effective assays. A number of factors need to be considered in selecting the most appropriate assays. In general, a balance has to be found between screening needs and the resources available, including finances, staff and their expertise, equipment, consumables and disposables.

Each screening system has its advantages and limitations that should be taken into consideration when selecting assays. Some limitations include:

• The length of time following infection before the screening test becomes reactive (window period)

• Rates of biological false positives which may result in the wastage of donations and unnecessary deferral of donors

• The complexity of some systems that require automation.

In most situations, EIAs, CLIAs and particle agglutination assays developed specifically for blood screening are the assays of choice as they are suited to screening from relatively small to large numbers of samples. In addition, the formats allow more objective recording and analysis of the results than rapid tests. However, a rigorous scientific evaluation of all assays prior to use is needed to determine their suitability in terms of sensitivity, and where possible, specificity in the situations in which they are to be used. While immunoassays may most often be microplate-based EIAs or specific system-based CLIAs, the use of simple/rapid disposable devices may be appropriate in some situations.

Most EIAs and CLIAs have greater sensitivity and specificity than particle agglutination assays or rapid tests. Their manufacture and performance are generally more reliable and consistent and have better outcomes for blood screening. High quality particle agglutination assays are not available commercially for all the routine markers for which blood is screened.

The use of rapid/simple assays is generally not recommended for blood screening as they are designed for the immediate and rapid testing of small numbers of samples, mainly for diagnostic purposes. These assays are performed using manual techniques; the results therefore have to be transcribed by staff and there is a lack of permanent records and traceability. As a result they may have limited use in laboratories where through-put is medium to high. They may, however, be considered for use in small laboratories that have limited resources and perform only a small number of tests daily as they provide flexibility and no major items of equipment are needed. They may also be appropriate when a laboratory needs to screen specific donations on an emergency basis for immediate release of products due to a critically low blood inventory or when rare blood is required urgently. In such emergency situations, the use of the rapid/simple assay should be followed up with repeat testing using an EIA, CLIA or particle agglutination assay if these assays are routinely used.

The introduction of NAT should be considered only when an efficient and effective programme based on antibody/antigen testing is in place (20) and there is a clear, evidenced, additional benefit. Although NAT reduces the window period of infection, in countries with a low incidence of infection, the incremental gain is minimal as the number of donors in the window period at the point of donation is generally very low. However, in countries with a high incidence of infection there are likely to be significant numbers of window period donations that can be identified by NAT (21). Thus although the risk of transfusing a blood unit collected during the window period may be decreased using NAT, the actual benefit in most populations has first to be determined and should be balanced against the complexity and high cost of performing NAT, including the infrastructure required (22-24).

For countries with sufficient resources, NAT offers certain benefits when combined with antibody/antigen testing. However, the potential benefit of detecting early infections and preventing possible transmissions of infection should be assessed in relation to such factors as the incidence and prevalence of infection in the blood donor population, the effectiveness of the blood donor selection process, the sensitivity of the serological screening currently undertaken and the ability to enhance this through, for example, the use of more sensitive serological assays such as combination antigen-antibody assays.

3.3. CRITICAL ASSAY CHARACTERISTICS

Sensitivity and specificity are the key factors to be considered in selecting a specific assay. For the screening of blood donations, both sensitivity and specificity should be the highest possible or available. Each assay should be evaluated within the country or region to confirm the technical data provided with regard to assay performance and, where possible, data from other studies should be analysed. The performance actually achieved in routine screening situations may not always meet the expected performance because assays are conducted by a range of staff under differing conditions. The reliability and consistency of the assay will be determined by a number of factors related both to the assay and the specific laboratory in which it is used. Each assay should be validated in its place of use to assure that the performance is as expected according to the results of evaluation.

Assay specific factors include:

• Assay presentation

• Clarity of instructions

• Ease of use

• Assay characteristics, including sensitivity and specificity

• Sample volume

• Sample and reagent addition monitoring

  See Also

  NCI Dictionary of Cancer Terms

• Robustness

• Assay reproducibility and precision

• Number of tests per assay

• Kit size

• Total assay time.

Laboratory specific factors include:

• Number of samples to be tested

• Staff levels

• Staff competence

• Available equipment

• Level of laboratory quality system.

Logistics that need to be taken into consideration include:

• Vendor selection and validation

• Price

• Procurement system

• Availability and reliability of the supply of test kits and reagents

• Shelf-life of test kits and reagents

• Infrastructure: e.g. controlled storage conditions and uninterrupted power supply

• Technical support for trouble-shooting

• Equipment maintenance, servicing and repair.

3.4. EVALUATION OF ASSAYS

Assays produced by the major international diagnostics companies are generally well-designed and are normally evaluated scientifically, both by the manufacturers themselves and by independent laboratories, prior to release onto the market (seewww.who.int/diagnostics_laboratory/evaluations; www.who.int/diagnostics_laboratory/publications/evaluations/en/index.html and www.who.int/bloodproducts/ref_materials/en/). Data published in kit package inserts and the scientific literature also provide useful information guiding selection of vendors, testing platforms and specific assays. However, well-planned and documented assay evaluations prior to their procurement are essential to ensure that the most appropriate selections are made from the available options. Assay evaluations are required to determine scientifically the most suitable assays for use in particular situations.

Evaluations should be carried out in at least one major facility, but some blood transfusion services may not have the necessary resources, expertise, experience and, importantly, panels of samples required. In such situations, the evaluations should be undertaken on behalf of the blood transfusion service, and in close conjunction with it, by an appropriate laboratory, such as the national reference laboratory. If none is available, the evaluation data required should be obtained from a blood transfusion service or reference laboratory in another country with similar demography, infection incidence and prevalence and BTS requirements, preferably in the same region. Reference should also be made to information available from laboratories elsewhere in the region or globally.

The evaluation process normally consists of performing each assay under consideration against selected panels of samples that will challenge the assay and deliver statistically valid results. The panels are generally comprised of:

• True positive samples and true negative samples in which the sensitivity and specificity respectively are determined

• Samples collected during seroconversion

• Low-level positive samples: for example, samples from very early or very late in the course of infection

• Samples covering a range of different genotypes and/or serotypes with emphasis on local samples

• Known non-specifically reacting samples or potentially cross-reactive samples: i.e. samples from patients not infected with the target infection, but with a range of clinically relevant conditions such as hypergammaglobulinaemia, other infections or autoimmune disease.

The overall size of the panels will be determined by local availability but, generally, the more samples tested, the more useful and reliable is the information generated. It is particularly important to include as many examples of locally-acquired infections as possible, especially samples from blood donors found previously reactive and confirmed to be infected. Analysis of the results will identify the assay that gives the best overall performance against all samples tested. It is therefore important that the panels are as broad as possible and that overall performance is assessed in the context of the planned use of the assay.

Each country should determine the minimum sensitivity and specificity levels required for each assay. Evaluation should be conducted on sufficient numbers of known antibody positive and negative samples to ensure that evaluation results are statistically significant. It is recommended that the minimum evaluated sensitivity and specificity levels of all assays used for blood screening should be as high as possible and preferably not less than 99.5%.

3.5. MONITORING ASSAY PERFORMANCE

In blood screening, assay performance should be continually monitored in order to identify any changes in performance that are occurring and that, without correction, might ultimately lead to a failure in either the assay runs or the detection of low-level true positive samples. Performance is usually assured by monitoring one or more parameters that can reasonably be expected to change relatively quickly as a result of any change in the performance or use of the assay (the assay or the operator/system performing the assay). These parameters include:

• Quality control sample results

• Assay control values

• Repeat reactivity.

The use of appropriate quality control (QC) samples included with every batch of tests performed will rapidly generate useful and reliable data for monitoring. In this context, a batch of tests can be any defined block of tests; for example, a single microplate is a batch of tests and at least one external QC sample could be included on each plate. External quality controls do not substitute for internal (kit) controls.

QC samples are normally well-characterized samples, individual or pooled, that are calibrated against international standards, where possible, and are diluted in an appropriate matrix. These samples may be used as external go–no–go controls, in which case the QC sample(s) has to be reactive for the assay run to be valid. If QC samples are not available, tracking the assay control values may be used as an alternative for assessing the consistency of performance.

In all cases where quantitative values are used, such as EIA optical density (OD) values, the results should be normalized to allow comparison between different runs and, to a certain degree, between different assays. The normalized OD value is calculated as follows:

• Non-competitive EIAs: divide the sample OD value by the cut-off OD value

• Competitive assays: divide the cut-off OD value by the sample OD value.

The ratio generated can then be directly compared to the ratios generated by any other runs of the assay, including different manufacturers' lots. The analysis is less objective in situations where assay results are qualitative, such as in the use of particle agglutination assays. However, the QC sample can be used to determine whether the results of the assay run are valid. Where it is not, the assay run should be repeated.

3.6. USE OF AUTOMATION FOR PERFORMING ASSAYS

The use of automation is a major consideration for blood transfusion services that perform a large number of screening tests. While all EIAs need a basic level of automation (automated plate washers and readers), highly sophisticated automated screening systems are available that can perform all aspects of an immunoassay from sampling through to the final analysis of the results. These systems perform immunoassays from any major manufacturer and are referred to as “open” systems; they are generally microplate-based and the equipment and assays are not linked. Dedicated systems, known as “closed” systems, are fully automated and use only specific, dedicated assays with all the necessary reagents and controls produced by or in collaboration with the equipment manufacturer.

Depending on the number of donation samples to be screened each day and the resources available, the use of a fully automated system can offer substantial advantages in terms of quality, especially if the system handles the samples as well as performing all the steps of the assay. Automated systems generally offer a high level of consistency and reproducibility in assay performance and can also help to reduce operator errors. However, they have specific additional requirements, including special staff training needs, regular and effective maintenance and calibration and may involve higher capital and running costs. Open and closed systems each have their advantages and disadvantages but, in general, an open system offers greater flexibility and may be more cost-effective, although the technical input and skill required from the user is often greater.

As in the selection of assay types, the overall workload is a major factor in determining whether automation is appropriate. Automated systems are particularly useful where large numbers of samples are screened regularly. At lower workload levels, where EIAs are performed, at the very least automated plate washers and plate readers are essential.

3.7. NEW ASSAYS AND TECHNOLOGIES

New blood safety technologies are constantly becoming available which may offer new opportunities to blood screening programmes. Although it is important to be aware of scientific and technological developments, these may or may not offer any advantages or significant improvements over current practice. In the context of screening donated blood, the use of a new technology is generally an advantage only if the technology currently in use is failing to identify infected donations or if the new technology offers significant cost savings and efficiency benefits without reducing the overall effectiveness of the current screening programme.

Before any new technology is introduced into a blood screening programme, it should be fully investigated and systematically evaluated. Even if there is a potential advantage, the feasibility of implementing a new technology should be fully considered, including the requirements for infrastructure, financing, staffing levels, training and quality systems. Since the overall costs of implementation may far outweigh any potential benefit in terms of increased blood safety, a cost-benefit analysis should be performed and found to be favourable.