Evaluate performance of the Abbott chemiluminescent microparticle immunoassay assay for detection of syphilis infection in Chinese blood donors

Abstract Background and Objectives To prevent Treponema Pallidum (TP) transmission from blood transfusion, enzyme‐linked immunosorbent assay (EIA) for anti‐TP has been widely used in routine blood donation screening in China for many years. The aim of this study was to evaluate the performance of the Abbott CMIA assay for detection of anti‐TP in Chinese blood donors. Materials and methods A total of 2420 plasma samples, already routinely screened for anti‐TP by two different EIAs, from four blood Centers were tested for anti‐TP by Abbott CMIA. Subsequently, all samples with positive results by one or both EIAs and/or by Abbott CMIA were subjected to confirmatory testing (CT) using recombinant immunoblot assay (RIBA) or Treponema Pallidum particle agglutination assay (TPPA). TP infection was defined by a RIBA or TPPA positive. Results Compared with two EIAs strategy, Abbott CMIA showed a relatively best sensitivity as 98.80% (95% CI: 97.44%‐100.16%) and a relatively best specificity as 99.58% (95% CI: 99.30%‐99.85%), yielding the best consistency (99.49%) between anti‐TP CT results with the highest κ value of .98. Conclusion This is the first study to evaluate the performance of the Abbott CMIA assays for detection of syphilis in Chinese blood donors. Our results suggested that CMIA performed better than both EIAs, and implementation of CMIA replacing two different EIA reagents might help to further reduce the risk of transfusion‐transmitted TP infection, decrease unnecessary blood waste and loss of blood donors.

new cases were estimated by World Health Organization (WHO) in 2016. 8 Additionally, historical data demonstrate that untreated early syphilis in pregnant women, if acquired during the 4 years before delivery, can lead to infection of the fetus in up to 80% of cases and may result in stillbirth or death of the infant in up to 40% of cases. 9 988 000 pregnant women were estimated being infected with syphilis in 2016, resulting in over 350 000 adverse birth outcomes including 200 000 stillbirths and newborn deaths. 10 In recent years, syphilis infection has become a serious problem in China. The total incidence of syphilis increased from 1.0 to 32.2 per 100 000 between 1995 and 2016 in China. 11  Since 2010, the rising trend of syphilis in most Chinese provinces had reversed that incidence in eastern and southern central China has largely decreased (2.5-25.1/100 000) while remaining high in northwest China (11.5-25.6/100 000). 11

| MATERIAL S AND ME THODS
This study was approved by the ethics committee of the Institute of Blood Transfusion (IBT), Chinese Academy of Medical Sciences.

| Specimens
A total of 2420 donor samples were selected from The Recipient Epidemiology and Donor Evaluation Study-III (REDS-III) sample repository. All samples in REDS-III repository had gone through routine screening for anti-HCV, hepatitis B surface antigen (HBsAg), anti-TP, and HIV-1/2 Ab/Ag by two different licensed EIAs. Samples were selected randomly: These samples were from four different blood centers, and details of sample characteristics are shown in Table 1.

| Anti-TP screening by CMIA
The collected plasma samples were blinded to EIA results and assigned unique new sample identification numbers before tested with ARCHITECT Syphilis TP test (Abbott Diagnostics) on the automated chemiluminescent microparticle immunoassay analyzer ARCHITECT i2000 system (CMIA, Abbott Diagnostic). The ARCHITECT Syphilis TP assay is a two-step immunoassay for qualitative detection of antibody to TP in human serum or plasma using CMIA technology with assay protocols, referred to as Chemiflex. All procedures were carried out according to manufacturer's instruction. 14 Testing results were expressed as signal to cutoff (S/CO), and S/CO >1.0 is considered reactive. The sensitivity and the specificity of this assay are 100% and 99.76% respectively according to manufacturer's instruction.

| Confirmatory testing
As recombinant immune blot assay (RIBA) and Treponema pallidum particle agglutination (TPPA) were considered as the gold standard reference methods, we used Mikrogen ® RIBA or FUJIREBIO ® TPPA to confirm anti-TP screening results. Samples tested reactive by either one or two EIAs or CMIA were subjected to CT. The testing algorithm is shown in Figure 1.

Number of the samples (%)
a All samples were freshly frozen after collection before shipped to the Institute of Blood Transfusion (IBT) and restored at −20℃. b Gray zone: samples tested with borderline results (0.5 < S/CO < 1) by EIA.

| Data analysis
Descriptive statistics were applied using the SPSS software (version 16.0) (SPSS Institute). The κ consistency test was used to compare consistency between screening results and CT results (κ < .40 indicated poor consistency; .40 ≤ κ < .75 indicated average consistency; and κ ≥ .75 indicated good consistency). Categorical variables were compared using the chi-square test.

| Results of the samples tested nonreactive for anti-TP by two ELAs
A total of 1995 samples tested nonreactive for anti-TP by two EIAs were tested by Abbot CMIA, including 10 anti-TP gray zone (0.5 < S/ CO < 1) and 809 screening reactive for other pathogens. Of these, 18 were reactive by Abbot CMIA. 10 of these 18 samples underwent CT: 4 (40%, 4/10) were confirmed reactive, 5 (50%, 5/10) were nonreactive, and 1 (10%, 1/10) were indeterminate. All 10 samples with EIA anti-TP gray zone results were negative by Abbott CMIA. Nine of these 10 samples were subjected to CT and 1 (11.11%, 1/9) was indeterminate in CT. Data are detailed in Table 2.

| Results of the samples tested reactive for anti-TP by ELA
425 samples were screening reactive by anti-TP EIA, including 299 reactive by both EIAs, 40 by EIA reagent 1, and 86 by EIA reagent 2.

| Coincidence rate between CT results and EIAs or Abbott CMIA
In the EIA nonreactive group, 8 of the 18 CMIA reactive samples and 1 gray zone samples were not tested for CT due to insufficient sample volume. These samples were excluded for the coincidence  Table 4, the data are displayed in four groups (CMIA, EIA reagent 1, EIA reagent 2, and two EIAs).

| D ISCUSS I ON
The current practice at Chinese blood centers for serological screening of TP infection is by using two different EIAs in parallel and no routine confirmatory testing. If either of the two EIAs tested reactive, the donor is permanently deferred and the donated unit discarded. CMIA has been very widely used for clinical diagnosis and shown to be at least as sensitive and specific as the traditional techniques such as EIA. 15  There are several limitations: (a) TPPA is recognized the "gold standard" of syphilis antigen serology test by WHO, historically having the best sensitivity of the treponemal tests, but actually had been shown to be less sensitive than EIA and CMIA; (b) the 1967 samples that were negative did not undergo confirmatory testing and were assumed to be anti-TP negative, Therefore, discordances between the algorithms may be underestimated.
We report the first study to evaluate the performance of the Abbott CMIA assays for detection of syphilis in Chinese blood donors. Our results suggested that CMIA performed better than the two EIA approaches. The implementation of CMIA in routine donor may help to further reduce the risk of transfusion-transmitted TP infection and decrease unnecessary blood waste.

D I SCL A I M ER
None.