How to manage the adoption of new assays for serological blood donation screening from a new test provider

Over the last several decades, the safety of donated blood has dramatically increased by introducing serological blood donor screening assays for infectious diseases, which has reduced the risk of transmission of infections from donors to recipients.^1-3 In particular, increased assay sensitivity, which is a test’s ability to identify an individual with a disease as positive, has dramatically improved and led to earlier, more robust detection of infected donors.

However, increasing sensitivity can often be associated with decreased specificity, the test’s ability to identify donors without disease as negative. This interaction can lead to an increased rate of false-positive test results, which refers to a blood test result indicating the disease is present in a donor when the individual does not have the disease.^3,4 

If this occurrence continues, blood donation centers may consider using a different screening assay. However, implementing serological assays from a new test provider can be challenging and blood banks need to be aware of the dynamics associated with switching to a new screening test and take unbiased approaches when comparing screening methods.

When blood centers decide to switch to a new screening assay, the donor pool has to undergo a new culling process, as the new assay might display different reactivities within that same donor pool. As a result, there will be an initial decrease in donors called the “switch effect,” followed by an increase in specificity known as the “culling effect.”

To better understand these consequences and assist decision-making, researchers have developed a mathematical model that can simulate the culling effect with different input factors and predict the stabilization of specificity.⁷

Implemented into a software tool, this model estimates the number of additional false-positive results expected after the switch to a new blood screening assay ​​and which results can be removed through the culling effect. It can also approximate the time frame for the specificity and donor pool to stabilize. Furthermore, the model can be individualized depending on the needs of each blood bank by using different key parameters as input variables. Amongst the tailorable parameters are the assay specificity, the time interval between donations, and the deferral policy of the specific blood center.

How do these variables affect the results? The study showed that a higher assay specificity is associated with reduced specificity improvement through the culling effect. A longer time interval between donations increases the time needed for the donor pool to stabilize and the number of false-positive results. Additionally, waiting until the second or third false-positive result before deferring donors delays the culling process but, on the other hand, may favor donor retention. Hence, blood donation centers may need to strike a balance between the desire to expedite specificity recovery and encouraging donor retention.⁷

The strength of this model was demonstrated using real-world data from a blood donation center in South Africa, with the model predicting in the first 15.5 months post-switch an initial increase of 118 additional false positives compared with the subsequent 15.5 months vs. the actual 117 recorded false-positive results.⁷

Taken together, the model can inform the donation center that a culling effect might occur upon switching to a new assay while estimating the degree and duration of this effect. Such knowledge could facilitate the decision to change to a new assay, alleviate potential concerns, and assist with planning any necessary mitigating measures, such as donor reinstatement programs.⁷

As several assays, including molecular tests like polymerase chain reaction (PCR) or nucleic-acid testing (NAT), have been developed over the years for use in blood screening, assay performance has increased considerably.¹

Different options are currently available from various manufacturers, and performance comparisons are often carried out to evaluate and possibly adopt a new assay panel. However, using an established donor pool for one assay and comparing its performance with another assay might not be the best approach, as bias can arise towards a specific assay during routine screening. To avoid this issue, the assay’s specificity should be conducted on unselected, first-time blood donors. This approach would constitute an unbiased yet more challenging comparison.

A recently published study compared the clinical performance amongst different assays at a French blood donation center ​​and relied on a large number of first-time donors for specificity analyses.⁸ The results demonstrated that the specificities for novel screening tests which use fully automated analyzers ranged from 99.81–100.00% compared to 99.71–99.98% and 99.79–99.98% for commercially available tests in an unbiased setting for screening blood donations.⁸