Data on the Relationships of Signal-To-Cutoff Ratios of Elecsys HIV Antigen/Antibody and Elecsys Syphilis Assays to Subsequent Confirmatory Testing at an Academic Medical Center

Immunoassays are widely used as screening tests for HIV and syphilis in clinical, public health, and research settings. For syphilis, immunoassays are used in 'reverse syphilis algorithms' that start with treponemal tests such as syphilis IgG or syphilis total antibodies for the initial screen. Many screening immunoassays for HIV and syphilis use signal-to-cutoff (S/CO) values for determining positivity, with a cutoff value such as 1.0 differentiating positive from negative. Published literature indicates that the S/CO value often correlates with the likelihood of subsequent confirmation of HIV and syphilis infections, with low S/CO values barely exceeding the cutoff more likely to represent false positive screens. The data in this article present results from the Roche Diagnostics Elecsys HIV combi PT Assay and the Elecsys Syphilis Assay. The data include 19,368 syphilis total antibody screening results on 15,764 unique patients and 28,952 HIV screening results on 24,556 unique patients, S/CO values, clinical area where testing was ordered, sex, and age. For samples with positive syphilis total antibody screens, the data also include results of RPR (the immediate next step in the reverse algorithm), Treponema pallidum particle agglutination (TP-PA; for those samples in which RPR was non-reactive in the testing cascade), and clinical information and other testing related to diagnosis of syphilis. For positive HIV screens, the data also include HIV antibody differentiation results, HIV-1 PCR or HIV-2 results results (if performed), and clinical information related to diagnosis of HIV. The distributions of S/CO values relative to confirmation status were analyzed.


a b s t r a c t
Immunoassays are widely used as screening tests for HIV and syphilis in clinical, public health, and research settings. For syphilis, immunoassays are used in 'reverse syphilis algorithms' that start with treponemal tests such as syphilis IgG or syphilis total antibodies for the initial screen. Many screening immunoassays for HIV and syphilis use signal-tocutoff (S/CO) values for determining positivity, with a cutoff value such as 1.0 differentiating positive from negative. Published literature indicates that the S/CO value often correlates with the likelihood of subsequent confirmation of HIV and syphilis infections, with low S/CO values barely exceeding the cutoff more likely to represent false positive screens. The data in this article present results from the Roche Diagnostics Elecsys HIV combi PT Assay and the Elecsys Syphilis Assay. The data include 19,368 syphilis total antibody screening results on 15,764 unique patients and 28,952 HIV screening results on 24,556 unique patients, S/CO values, clinical area where testing was ordered, sex, and age. For samples with positive syphilis total antibody screens, the data also include results of RPR (the immediate next step in the reverse algorithm), Treponema pallidum particle agglutination (TP-PA; for those samples in which RPR was non-reactive in the testing cascade), and clinical information and other testing related to diagnosis of syphilis. For positive HIV screens, the data also include HIV antibody differentiation results, HIV-1 PCR or HIV-2 results results (if performed), and clinical information related to diagnosis of HIV. The distributions of S/CO values relative to confirmation status were analyzed.

Value of the Data
• The data provided are of value as HIV and syphilis testing is widely performed for clinical, public health, and research purposes. • Clinicians, other researchers, or personnel in clinical laboratories might find this data useful as a reference for comparison. • Our data set would serve as a starting point for researchers interested in future investigations investigating the reverse syphilis algorithm. • The data provide information on false positive HIV and syphilis testing in a population with overall low prevalence for HIV and syphilis. • The data include 19,368 syphilis total antibody screening results on 15,764 unique patients and 28,952 HIV screening results on 24,556 unique patients.

Data Description
For syphilis testing, the retrospective analysis in the present study includes detailed data on 19,368 samples from 15,764 unique patients who had syphilis testing using the reverse algorithm performed at an academic medical center. Fig. 1 (panel A) shows a flow diagram of the syphilis testing in the retrospective study. Multiple studies have shown an association of the signal-to-cutoff (S/CO) value of syphilis immunoassay tests with subsequent positive confirmatory testing indicating either past or current infection [1][2][3][4] . In the reverse algorithm, confirmation testing typically entails RPR as the next test following a positive syphilis antibodies result [5] . If RPR is non-reactive, then another treponemal test such as TP-PA or fluorescent treponemal antibody absorption (FTA-ABS) is performed. Positive syphilis antibodies in conjunction with reactive RPR are common in active or resolving syphilis infection. Positive syphilis antibodies with non-reactive RPR but a reactive TP-PA or FTA-ABS most often indicates prior treated syphilis or late/latent syphilis. If the RPR and the second treponemal test (TP-PA or FTA-ABS) are both non-reactive, this often represents a "false positive" for actual past or present syphilis infection [6] .
For HIV testing, the retrospective analysis in the present study includes detailed data on 28,952 samples originating from 24,556 unique patients who had HIV screening tests ordered at an academic medical center. Fig. 1 (panel B) shows a flow diagram of the HIV testing in the retrospective study. Similar to syphilis antibodies, multiple studies have shown an association of the S/CO value of HIV screening tests with subsequent verification of HIV diagnosis using confirmatory testing and clinical information [7][8][9][10][11][12][13][14][15] . Confirmation testing commonly used in the United States and some other countries includes antibody differentiation assays (in the present study was the Bio-Rad Geenius assay) and HIV RNA PCR. For the present study, a true (confirmed) positive required positive HIV RNA PCR and clinical documentation of HIV diagnosis. A non-true (false) positive would be indicated by a positive HIV screen with negative confirmatory testing (especially HIV PCR) and lack of a clinical diagnosis of HIV. We did extensive chart review on all cases with a positive HIV screen.
For syphilis testing, Table 1 shows demographics of the population being tested for syphilis by the reverse algorithm, including the clinical locations and specialties associated with orders. Fig. 2 shows the distribution of S/CO values for all negative syphilis screens (defined according to the package insert instructions as an S/CO value less than 1.00), encompassing 19,013 negative syphilis screening tests on 15,518 unique patients.  1 There were 10 unique patients (4 females, 6 males; 13 total results) who had one or more RPR non-reactive and TP-PA inconclusive results after a positive syphilis total antibodies screen. Three of these patients later had a reactive TP-PA result and are included in unique patient total for the "Confirmed Positive" category above. One patient later had a nonreactive TP-PA and is included in the unique patient total for "False Positive Screen" category. for syphilis, representing either past or present syphilis infection. For categories (D) and (E), TP-PA would not be performed, because the RPR is reactive and no further testing is needed based on the reverse algorithm. The raw data for the syphilis study are included in Supplementary files 1 and 2. In Table 1 and Fig. 3 , there are patients with multiple test results that fell into two or more categories. This is explained in more detail in the footnote to Table 1 and the legend for Fig. 3 . In the retrospective timeframe, there were only 24 unique patients who had negative syphilis screening results but then screened positive for syphilis at a later timepoint.
For HIV testing, Table 2 shows demographics of the population being tested for HIV, including the clinical locations and specialties associated with orders. Fig. 4 shows the distribution of S/CO values for all negative HIV screens (defined according to the package insert instructions as an S/CO value less than 1.00), encompassing 28,789 negative HIV screening tests on 24,430 unique patients.

Experimental Design, Materials and Methods
All data was obtained from patient data in the electronic medical record from the University of Iowa Hospitals and Clinics (Iowa City, Iowa, United States). A reporting tool within the electronic medical record, known as Epic Reporting Workbench [16] , was used to identify all syphilis and HIV screening tests performed in the retrospective timeframes. Only data from patients who had syphilis or HIV testing performed at the University of Iowa Hospitals and Clinics were included; no data was obtained from diagnostic vendors of any of the laboratory assays used for clinical testing. During the retrospective analysis period for syphilis testing (April 2, 2018 through May 14, 2021), syphilis screening using the reverse algorithm was performed using the Roche Diagnostics Elecsys Syphilis assay (Roche Diagnostics, Indianapolis, IN). During the retrospective analysis period for HIV testing (July 17, 2018 through May 14, 2021), HIV screening was performed using the Roche Diagnostics Elecsys HIV combi PT (Roche Diagnostics, Indianapolis, IN). In a previous study, we reported the relationship of S/CO ratio to confirmation for two other HIV assays (Abbott Architect HIV Ag/Ab Combo Assay, a 4th generation HIV combination assay; Bio-Rad Bioplex 2200 HIV Ag-Ab Assay, a 5 th generation HIV assay) [12] .
The Elecsys Syphilis assay is an immunoassay for the detection of the total antibodies (especially IgG and IgM) to Treponema pallidum in the human plasma and serum and outputs a single S/CO ratio that translates into a positive or negative qualitative result [17] . An S/CO ratio of 1.00 or greater on the Elecsys assay is considered positive, with repeat testing to verify the result. For samples with a positive syphilis antibodies screen, subsequent confirmatory testing includes first an RPR (Bio-Rad Laboratories Bioplex 220 Syphilis Total & RPR Assay, Redmond, WA). If RPR is reactive, no further testing is done. If RPR is non-reactive, then TP-PA testing (referred to reference laboratory, ARUP Laboratories, Salt Lake City, UT) is then done.
The Elecsys HIV combi PT is a 4 th generation HIV assay that simultaneously detects HIV-1 p24 antigen along with antibodies to HIV-1 and HIV-2 and outputs a single signal-to-cutoff (S/CO) ratio that translates into a reactive/nonreactive (referred to as positive/negative in this manuscript) qualitative result [13 , 18 , 19] . An S/CO ratio of 1.00 or greater on the Elecsys assay is considered positive, with repeat testing to verify the result. For all samples with a positive HIV screen, antibody differentiation was performed with the Bio-Rad Geenius HIV-1/HIV-2 Antibody Differentiation assay (Bio-Rad Laboratories, Redmond, WA), an assay that can confirm HIV infection. Final confirmation of HIV infection utilized HIV RNA PCR, testing which required a new specimen from the patient.

Ethics Statement
The analyses had approval by the University of Iowa Institutional Review Board (protocols # 202001165 and 202105258) as a retrospective project.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.