ARCHITECT Syphilis TP Reagent, ARCHITECT Syphilis TP Calibrator, ARCHITECT Syphilis TP Control

{0} 1 # 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY A. 510(k) Number: K153730 B. Purpose for Submission: To establish substantial equivalence to a predicate device and to obtain market clearance for a new assay designed to detect antibodies to *T. pallidum* in human serum and plasma. C. Measurand: Antibodies to *T. pallidum* (IgM and IgG) D. Type of Test: A qualitative immunoassay based on chemiluminescent microparticle immunoassay (CMIA) technology for use on the ARCHITECT iSystem. E. Applicant: Abbott Laboratories Diagnostics Division F. Proprietary and Established Names: ARCHITECT Syphilis TP Reagents, Calibrator, and Controls G. Regulatory Information: 1. Regulation section: 866.3830 *Treponema pallidum* treponemal test reagents 862.1150 Calibrator, secondary 862.1660 Quality control material (assayed and unassayed) 2. Classification: Class II (performance standards) {1} 3. Product code: LIP enzyme linked immunoabsorption assay, Treponema pallidum JIT calibrator, secondary JJX single (specified) analyte controls (assayed and unassayed) 4. Panel: Microbiology (83) H. Intended Use: 1. Intended use(s): Assay: The ARCHITECT Syphilis TP assay is a chemiluminescent microparticle immunoassay (CMIA) for the qualitative detection of antibodies (IgG and IgM) directed against Treponema pallidum (TP) in human serum and plasma. The ARCHITECT Syphilis TP assay is intended to be used as an initial diagnostic test or in conjunction with a nontreponemal laboratory test and clinical findings to aid in the diagnosis of syphilis infection. Warning: The ARCHITECT Syphilis TP assay is not intended for use in screening blood, plasma, or tissue donors. The effectiveness of the ARCHITECT Syphilis TP assay for use in screening blood, plasma, or tissue donors has not been established. Calibrator: The ARCHITECT Syphilis TP Calibrator is for the calibration of the ARCHITECT iSystem when used for the qualitative detection of antibody to Treponema pallidum (TP) in human serum and plasma. Controls: The ARCHITECT Syphilis TP Controls are for the estimation of test precision and the detection of systematic analytical deviations of the ARCHITECT i System when used for the qualitative detection of antibody to Treponema pallidum (TP) in human serum and plasma. 2. Indication(s) for use: The test system, when used in conjunction with non-treponemal based assays, provides serological evidence of infection with T. pallidum. 2 {2} 3. Special conditions for use statement(s): For Prescription Use The ARCHITECT Syphilis TP assay is not intended for use in screening blood, plasma, or tissue donors. The effectiveness of the ARCHITECT Syphilis TP assay for use in screening blood, plasma, or tissue donors has not been established. 4. Special instrument requirements: ARCHITECT iSystem I. Device Description: Assay The ARCHITECT Syphilis TP assay is a two-step immunoassay for the qualitative detection of antibodies (IgG and IgM) directed against TP in human serum or plasma based on CMIA technology for use with the ARCHITECT iSystem. The ARCHITECT iSystem is a fully-automated immunoassay system that allows random and continuous access as well as priority and automated retest processing. The following describes the processes that take place during the reaction: 1. Sample, assay diluent, and recombinant TP antigen (TpN15, TpN17 and TpN47) coated microparticles are combined. Anti-TP antibodies present in the sample bind to the TP coated microparticles. 2. After washing, anti-human IgG and IgM acridinium-labeled conjugate is added to create a reaction mixture. 3. Following another wash cycle, Pre-Trigger and Trigger Solutions are added to the reaction mixture. 4. The resulting chemiluminescent reaction is measured as relative light units (RLUs). There is a direct relationship between the amount of anti-TP antibodies in the sample and the RLUs detected by the ARCHITECT iSystem optics. The presence or absence of anti-TP antibodies in the specimen is determined by comparing the chemiluminescent signal in the reaction to the cutoff signal (S/CO) determined from an active calibration. If the chemiluminescent signal in the reaction is greater than or equal to the cutoff signal, the specimen is considered reactive for anti-TP antibodies. If the chemiluminescent signal is below the cutoff signal, the specimen is considered nonreactive for the anti-TP antibodies. The results are printed out as follows: S/CO ≥ 1.00 Reactive S/CO &lt; 1.00 Nonreactive 3 {3} 4 Interpretation of results: Reactive Reactive for treponemal antibodies Nonreactive Nonreactive for treponemal antibodies Test results are intended to aid in diagnosis only. As with all serological tests for syphilis, results should always be interpreted in conjunction with additional treponemal or nontreponemal serologic test results (as appropriate), the patient’s clinical symptoms, medical history, and other clinical and/or laboratory findings to produce a diagnosis of syphilis by disease stage. The ARCHITECT Syphilis TP Assay Reagents are for use with the ARCHITECT Syphilis TP Calibrator and Controls. ARCHITECT Syphilis TP Calibrator Sold as an accessory; intended for the calibration of the ARCHITECT Syphilis TP assay. Provided as 1 Bottle (4.0 mL) of ARCHITECT Syphilis TP Calibrator and is prepared in recalcified human plasma (inactivated); reactive for anti-TP. Contains preservatives: sodium azide and other antimicrobial agent ARCHITECT Syphilis TP Controls Sold as an accessory; intended for the estimation of test precision and the detection of systematic analytical deviations of the ARCHITECT iSystem when used for the qualitative detection of antibodies to Treponema pallidum (TP) in human serum and plasma. Consists of two vials of controls, one Negative and one Positive; prepared in recalcified human plasma. The Positive Control (inactivated) is reactive for anti-TP. Contains preservatives: sodium azide and other antimicrobial agents. J. Substantial Equivalence Information: 1. Predicate device name(s): DiaSorin LIAISON® Treponema Assay 2. Predicate 510(k) number(s): K061247 3. Comparison with predicate: | Characteristics | ARCHITECT Syphilis TP | DiaSorin LIAISON® Treponema Assay (k061247) | | --- | --- | --- | | Similarities | | | | Intended Use | The ARCHITECT Syphilis TP assay is a chemiluminescent microparticle immunoassay | The LIAISON Treponema Assay uses chemiluminescent immunoassay (CLIA) technology on the LIAISON Analyzer | {4} | | (CMIA) for the qualitative detection of antibodies (IgG and IgM) directed against Treponema pallidum (TP) in human serum and plasma. The ARCHITECT Syphilis TP assay is intended to be used as an initial diagnostic test or in conjunction with a nontreponemal laboratory test and clinical findings to aid in the diagnosis of syphilis infection. | family for the qualitative determination of total antibodies directed against Treponema pallidum in human serum. The presence of antibodies to Treponema pallidum specific antigen, in conjunction with non-treponemal laboratory tests and clinical findings may aid in the diagnosis of syphilis infection. | | --- | --- | --- | | Methodology | Chemiluminescent microparticle immunoassay (CMIA) | Sandwich chemiluminescence immunoassay (CLIA) | | Cut-off Index | 1.00 S/CO | Index 1.0 | | Standardization | The calibrator is referenced to an internal reference standard. | The calibrator concentrations are referenced to an in-house antibody preparation. | | Controls | 2 (Negative and Positive) | 2 (Negative and Positive) | | **Differences** | | | | Antigens Used | Recombinant TP antigens: TpN15, TpN17 and TpN47 (obtained in E.coli) | DNA-Tp17 Recombinant antigen (obtained in E. coli) | | Instrument | ARCHITECT i2000SR System | DiaSorin LIASON | | Assay protocol | 2-step | 1-step | | Specimen Type | Human serum or plasma (Potassium EDTA, Lithium heparin, Sodium heparin) | Human serum | | Equivocal Zone | No equivocal zone | Index value 0.9-1.1 | | Components | 1. Microparticles – TP antigen (E.coli, recombinant) coated microparticles in HEPES buffer with detergent. 2. Conjugate – Murine anti-IgG/anti-IgM acridinium-labeled conjugate in MES buffer with protein (bovine) stabilizer. 3. Assay Diluent – Syphilis TP Assay Diluent containing MES buffer with detergent. Preservatives: ProClin 950 and other antimicrobial agents. | 1. Magnetic Particles – Magnetic particles coated with Tp17 DNA recombinant protein (obtained in E. coli), BSA, phosphate buffer. 2. Conjugate – Tp17 DNA recombinant protein (obtained in E. coli) conjugated to an isoluminol derivative, BSA, phosphate buffer. Preservatives: ProClin 300, Gentamycin Sulfate. 3. Specimen Diluent – EDTA, phosphate buffer, bacterial proteins, an inert blue dye. Preservative: ProClin 300. | | Calibrators | 1 positive calibrator, stable up to 30 days. | 2 positive calibrators, stable up to 2 weeks. | 5 {5} 6 K. Standard/Guidance Document Referenced (if applicable): Not applicable. L. Test Principle: Chemiluminescent microparticle immunoassay (CMIA) M. Performance Characteristics (if/when applicable): 1. Analytical performance: a. Precision/Reproducibility: Internal Precision The precision of the ARCHITECT Syphilis TP assay was evaluated in an internal study using three reagent lots and two calibrator lots. The test panel consisted of four samples contrived in recalcified human plasma matrix at the following target antibody concentrations: | Sample | Target S/CO | | --- | --- | | High Non-reactive | 0.55 | | Low Reactive | 1.25 | | High Reactive | 3.55 | | Non-reactive | 0.08 | Positive and negative controls, with target S/CO values of 2.71 and $\leq 0.40$, respectively, were included in each run. The samples were tested in two replicates, twice a day, over 22 testing days, using two ARCHITECT $i2000_{\mathrm{SR}}$ instruments. The mean S/CO value, within-run Standard Deviation (SD) and percent coefficient of variation (%CV), between-run SD and %CV, and between-day SD and %CV were calculated for each sample stratified by instrument and reagent lot. The estimates of total imprecision were calculated by summation of the variance components. The results from the precision study across reagent lots, by instrument are shown below. {6} Instrument: iSR01260 (ARCHITECT i 2000SR) | Sample | N | Mean S/CO | Within-Run | | Between-Run | | Between-Day | | Within-Laboratory^{a} (Total) | | Between-Lot | | Total Imprecision Including Lot-to-Lot^{b} (Overall) | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | | SD | %CV | SD | %CV | SD | %CV | SD | %CV | SD | %CV | SD | %CV | | Negative Control | 392 | 0.04 | 0.004 | 9.4 | 0.000 | 0.0 | 0.002 | 5.6 | 0.004 | 11.0 | 0.002 | 6.2 | 0.005 | 12.6 | | Positive Control | 395 | 2.69 | 0.035 | 1.3 | 0.017 | 0.6 | 0.040 | 1.5 | 0.056 | 2.1 | 0.012 | 0.4 | 0.057 | 2.1 | | Panel B | 396 | 0.56 | 0.010 | 1.8 | 0.005 | 0.9 | 0.006 | 1.1 | 0.013 | 2.4 | 0.005 | 0.8 | 0.014 | 2.5 | | Panel D | 395 | 1.27 | 0.018 | 1.5 | 0.013 | 1.0 | 0.012 | 0.9 | 0.025 | 2.0 | 0.012 | 1.0 | 0.028 | 2.2 | | Panel F | 396 | 3.61 | 0.051 | 1.4 | 0.019 | 0.5 | 0.030 | 0.8 | 0.062 | 1.7 | 0.015 | 0.4 | 0.064 | 1.8 | | Panel G | 396 | 0.08 | 0.004 | 4.9 | 0.001 | 1.6 | 0.002 | 3.1 | 0.005 | 6.0 | 0.004 | 4.7 | 0.006 | 7.6 | a Within-Laboratory variability contains within-run, between-run, and between-day variance components b Overall variability contains within-run, between-run, between-day, and between-lot variance components Instrument: iSR02399 (ARCHITECT i 2000SR) | Sample | N | Mean S/CO | Within-Run | | Between-Run | | Between-Day | | Within-Laboratory^{a} (Total) | | Between-Lot | | Total Imprecision Including Lot-to-Lot^{b} (Overall) | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | | SD | %CV | SD | %CV | SD | %CV | SD | %CV | SD | %CV | SD | %CV | | Negative Control | 392 | 0.04 | 0.004 | 11.1 | 0.000 | 0.0 | 0.001 | 2.4 | 0.004 | 11.4 | 0.004 | 10.6 | 0.006 | 15.5 | | Positive Control | 394 | 2.66 | 0.054 | 2.0 | 0.014 | 0.5 | 0.050 | 1.9 | 0.075 | 2.8 | 0.000 | 0.0 | 0.075 | 2.8 | | Panel B | 394 | 0.54 | 0.013 | 2.3 | 0.006 | 1.1 | 0.006 | 1.2 | 0.015 | 2.8 | 0.003 | 0.5 | 0.015 | 2.9 | | Panel D | 394 | 1.23 | 0.024 | 2.0 | 0.013 | 1.1 | 0.013 | 1.1 | 0.031 | 2.5 | 0.009 | 0.7 | 0.032 | 2.6 | | Panel F | 396 | 3.54 | 0.100 | 2.8 | 0.037 | 1.0 | 0.035 | 1.0 | 0.112 | 3.2 | 0.000 | 0.0 | 0.112 | 3.2 | | Panel G | 394 | 0.08 | 0.004 | 5.5 | 0.000 | 0.6 | 0.002 | 2.2 | 0.005 | 6.0 | 0.004 | 5.2 | 0.006 | 7.9 | a Within-Laboratory variability contains within-run, between-run, and between-day variance components b Overall variability contains within-run, between-run, between-day, and between-lot variance components ## Multi-site Reproducibility A multicenter reproducibility study was performed for the ARCHITECT Syphilis TP assay at 3 clinical sites using 3 lots each of ARCHITECT Syphilis TP Reagents, 2 lots of ARCHITECT Syphilis TP Calibrator, and 1 lot of ARCHITECT Syphilis TP Controls. Two levels of controls and 4 serum panels were assayed in replicates of 4 at 2 separate times of day for 5 days. | Sample | N | Grand Mean S/CO | Within-Run | | Within-Day | | Within-Laboratory (Total) | | | Precision with Additional Component of Between-Site | | Precision with Additional Component of Between-Lot | | Precision with Additional Components of Site and Lot (Overall) | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | | SD | %CV | SD | %CV | SD | %CV | 95% CI | SD | %CV | SD | %CV | SD | %CV | | Negative Control | 360 | 0.03 | 0.002 | 6.4 | 0.002 | 6.6 | 0.002 | 6.6 | (6.24,7.08) | 0.003 | 9.3 | 0.003 | 9.5 | 0.003 | 9.6 | | Positive Control | 360 | 2.72 | 0.108 | 4.0 | 0.130 | 4.8 | 0.134 | 4.9 | (4.59,5.35) | 0.134 | 4.9 | 0.134 | 4.9 | 0.134 | 4.9 | | Nonreactive Panel | 360 | 0.08 | 0.004 | 5.1 | 0.004 | 5.1 | 0.004 | 5.4 | (5.07,5.78) | 0.006 | 7.5 | 0.006 | 7.4 | 0.007 | 8.8 | | High Nonreactive Panel | 360 | 0.55 | 0.012 | 2.3 | 0.015 | 2.7 | 0.015 | 2.8 | (2.61,3.04) | 0.017 | 3.2 | 0.017 | 3.2 | 0.017 | 3.2 | | Low Reactive Panel | 360 | 1.24 | 0.023 | 1.9 | 0.027 | 2.2 | 0.029 | 2.4 | (2.18,2.56) | 0.034 | 2.7 | 0.035 | 2.8 | 0.035 | 2.8 | | High Reactive Panel | 360 | 3.57 | 0.073 | 2.0 | 0.079 | 2.2 | 0.094 | 2.6 | (2.43,2.91) | 0.107 | 3.0 | 0.100 | 2.8 | 0.107 | 3.0 | a An outlying run was observed at one site for one reagent lot. The total %CV for the Positive Control using the replacement run was 2.6%. {7} b. Linearity/assay reportable range: Not applicable; this is a qualitative assay. c. Traceability, Stability, Expected values (controls, calibrators, or methods): ## Calibrator The ARCHITECT Syphilis TP Calibrator is sold as an accessory to the test. The calibrator is referenced to an Abbott internal reference standard. This internal reference standard is manufactured by dilution of high titer positive plasma with nonreactive human plasma. ## External Controls The ARCHITECT Syphilis TP Controls are sold as an accessory to the test. The Negative Control is made from human plasma shown to be non-reactive for antibodies to TP. The Positive Control is made from TP-non-reactive human plasma spiked with high titer anti-TP reactive human plasma. During the clinical study, the controls were analyzed 140 times; the mean for the Positive control was 2.67 S/CO with a SD of 0.093 and 3.51% CV. ## Sample Stability Studies The stability of TP antibodies in whole blood (serum -no additive and EDTA plasma) was evaluated using paired samples (serum and EDTA plasma) from 18 different donors. Each donor set was screened for non-reactivity with the ARCHITECT Syphilis TP assay. Each donor set was divided into two groups followed by spiking with high reactive TP-Ab plasma to target concentrations: a) High nonreactive ~0.80 S/CO, and b) Low reactive ~1.2 S/CO. Baseline measurements were established by testing from the primary tube (on clot) within 8 hours of collection (time point 1). The stability of TP antibodies in samples was evaluated: - Refrigerated stability (at 2-8°C) ≥ 7 days - Room temperature stability (at 22°C) ≥ 24 hours - Room temperature stability (at 30°C) ≥ 24 hours - Frozen stability (at ≤ -10°C) ≥ 7 days. - Freeze/thaw stability at 3 cycles and 6 cycles (a freeze/thaw cycle was defined as frozen at ≤ -10°C for &gt; 12 hours) Each sample was tested in triplicate at each time point. The evaluation of results was performed separately for each level, storage condition, and time point. - For each donor, concentration level, and tube type, the baseline mean of S/CO 8 {8} was calculated. - For each donor, concentration level, storage condition, and tube type, the mean S/CO was calculated for each subsequent time point. - For each donor, concentration level, storage condition, and tube type, the difference between the mean S/CO at each test time point and the corresponding baseline mean S/CO was calculated. The acceptance criteria were defined as: - For samples within a target S/CO value of $\sim 0.80$ , the difference in measured concentration should be $\leq 0.20$ S/CO; - For samples within a target S/CO value of $\sim 1.20$ , the percent difference in measured concentration should be $\leq 20\%$ . The following results were obtained across tube types (serum and EDTA plasma): | | Storage Condition | | | | | | --- | --- | --- | --- | --- | --- | | | 2 to 8°C | 22°C | 30°C | ≤ -10°C | Freeze/Thaw (6 cycles) | | | Mean Difference in S/CO | | | | | | High Non-reactive | -0.02 to 0.01 | -0.02 to 0.02 | -0.02 to 0.02 | -0.03 to 0.03 | 0.02 to 0.07 | | | Mean % Difference in Mean S/CO | | | | | | Low Reactive | -2% to 1% | -2% to 3% | -5% to 2% | -7% to 1% | 2% to 5% | The results support the following time intervals for storage of samples before testing with no degradation of the analyte observed: | Specimen Type | Storage Temperature | Maximum Storage Time | | --- | --- | --- | | Serum/Plasma (EDTA) | Room temperature | ≤ 72 hours | | | 2-8°C | ≤ 7 days | | | -10°C or colder | ≤ 30 days | ## On-board Stability The stability of TP antibodies in serum samples left on the instrument for 3 hours was evaluated with two levels of antibodies: High non-reactive sample (~ 0.80 S/CO), and Low reactive sample (~ 1.2 S/CO) The samples were prepared in anti-TP negative recalcified plasma spiked with syphilis positive recalcified plasma to the target concentration. Two sample concentrations were prepared: $\sim 0.80$ S/CO and $\sim 1.2$ S/CO. The samples were aliquoted into 13 sample cups and analyzed as a baseline. For the 3 hour time point, the same samples were also aliquoted into 13 sample cups and left on-board of the instrument; the samples were measured after 3 hours elapsed. The results were evaluated by calculating the difference between the mean of the baseline measurements and the mean of the measurements {9} obtained after 3 hours. Percent difference was also calculated. The results showed that samples may be left on board of the instrument up to 3 hours. | Concentration S/CO | N | Mean S/CO at Baseline | SD | Mean S/CO at 3 hours | SD | S/CO Difference | % Difference | | --- | --- | --- | --- | --- | --- | --- | --- | | 0.80 | 13 | 0.78 | 0.020 | 0.83 | 0.008 | 0.05 | 6.4 | | 1.20 | 13 | 1.20 | 0.023 | 1.25 | 0.026 | 0.05 | 4.2 | # Hook Effect Abbott conducted a study to demonstrate that the ARCHITECT Syphilis TP assay is not sensitive to the "Hook" effect, where high levels of antibodies may interfere with the formation of the immune complexes, giving false non-reactive results. The study utilized five unique TP-antibody reactive samples which were serially diluted with negative (TP-Ab non-reactive) serum. Each of the diluted samples was tested with the ARCHITECT Syphilis TP assay in duplicate. The mean, SD, and the $\% \mathrm{CV}$ were calculated for the S/CO values of each dilution level. The serial dilution levels were plotted (X-axis) versus the corresponding mean S/CO values of each dilution level (Y-axis). The data were considered acceptable if there was no increase in S/CO value when comparing the S/CO value of a dilution level to the S/CO value of the preceding dilution level. In all cases, the results showed no increase in S/CO values when compared to the results from the dilution immediately preceding. The data demonstrate that the ARCHITECT Syphilis TP assay is not susceptible to interference from specimens with high levels of anti-syphilis TP. # Carryover Potential sample carryover within the ARCHITECT Syphilis TP assay from a sample containing high levels of TP- antibodies was evaluated by comparing the results of a "protected" negative sample to an "unprotected" negative sample. The negative sample was the ARCHITECT Syphilis TP Negative Control (made of recalcified human plasma) and the high positive patient sample had an anti-syphilis TP antibody concentration greater than or equal to $20.0~\mathrm{S / CO}$ , representing the reactivity level of samples from late acute syphilis infection. The protected negative sample was tested before a high positive patient sample, and the unprotected negative sample was tested after the high positive patient sample. Each carryover run consisted of the following: 1) Wash buffer 2) Protected negative sample (not exposed to potential carryover from a high reactive sample) 3) High reactive sample 4) Unprotected negative sample (exposed to potential carryover from a high reactive sample) {10} The above was repeated 15 times across 5 runs. All protected and unprotected non-reactive sample replicates produced the same S/CO value of 0.03, demonstrating that there is no observable within assay sample carryover. ## Reactivity with IgG and IgM Treponemal Antibodies The ability of the ARCHITECT Syphilis TP assay to detect both IgG and IgM type treponemal antibodies was evaluated utilizing three conjugate preparations: (a) anti-human TP IgG conjugate (experimental EIA), (b) anti-human TP IgM conjugate concentrate (experimental EIA), and (c) a conjugate preparation containing both anti-human TP IgM and anti-human TP IgG- final device composition). Specimens from commercially available syphilis seroconversion panel were tested with these preparations. Specimens that demonstrated high levels of both anti-syphilis TP IgG and anti-syphilis TP IgM with a predicate assay were treated with protein G columns (2-3 cycles) to remove the IgG present in the specimens. The specimens were then assayed using the three conjugate preparations. The results demonstrated a significant decrease in the IgG-specific signal when tested using the IgG-specific experimental ARCHITECT assay and therefore, confirm that IgG is specifically detected by the ARCHITECT Syphilis TP assay. The results also demonstrate that the IgM-specific signal of the high positive samples increases by more than 2-fold (S/N = ~70 to 90) when tested using the IgM-specific experimental ARCHITECT assay and therefore, confirms that IgM is specifically detected. ## d. Detection limit: Not applicable. ## e. Analytical specificity: ### Interference by Endogenous Substances Potential interference with bilirubin, hemoglobin, triglycerides, cholesterol, protein (albumin), and gamma globulins was evaluated in an analytical study by spiking a low-reactive syphilis sample (targeted at 1.2 S/CO) and a high negative sample (targeted at 0.8 S/CO), prepared in recalcified plasma, with each of the selected interfering substances. The samples were tested (in 12 replicates) on the ARCHITECT Syphilis TP assay before and after the addition of the interferent. The mean (for the normally distributed results) or median (for the non-normally distributed results) and SD of the S/CO values of the spiked samples were calculated. The difference between the mean (or median) S/CO values between the spiked and unspiked samples was calculated for each interferent. The acceptance criteria for non-interference were set at ≤ 0.20 S/CO for negative samples and no more than 20% decrease in S/CO values for samples with S/CO ≥ 1.0. In all samples tested, the difference for samples with S/CO values &lt; 1.0 ranged from -0.06 to 0.02 S/CO and the % difference for samples with S/CO values &gt; 1.0 ranged from -8.1% to 2.1%. 11 {11} There was no interference observed for the listed substances up to the levels indicated below. | Interferent | Interferent Level | | --- | --- | | Conjugated Bilirubin | ≤ 20 mg/dL | | Unconjugated Bilirubin | ≤ 20 mg/dL | | Hemoglobin | ≤ 500 mg/dL | | Triglycerides | ≤ 3000 mg/dL | | Total Protein | ≤ 12 g/dL | | Cholesterol | ≤ 500 mg/dL | | Gamma Globulin | ≤ 6 g/dL | # Cross-reactivity in Specimens with Medical Conditions not Related to Syphilis Specimens from individuals diagnosed with other diseases were obtained from commercial vendors. The presence of the given analyte (microorganism or antibody) was confirmed with an FDA-cleared (when available) assay. The specimens were tested by the ARCHITECT Syphilis TP assay and also by the DiaSorin LIASON Treponema Assay. Discordant results were investigated by additional testing with TP-PA and RPR methods at an external laboratory. If discordance was not resolved, an FTA-ABS assay was employed for further testing. Among the samples tested, the following conditions had one sample (each) that was reactive by the ARCHITECT TP assay and non-reactive with other treponemal or nontreponemal assays: CMV IgG, HIV, hyper IgG, and $E.$ coli. There were three additional specimens (one from a patient with gonorrhea and two from patients with HTLV-II) that were reactive by the ARCHITECT Syphilis TP but could not be tested by the other tests (DiaSorin LIAISON, RPR, TP-PA, FTA-ABS) because these were plasma specimens and paired serum samples from these patients were not available. The results of the specificity study in samples from individuals with other diseases are shown below. | Clinical Category | N | Number of ARCHITECT Syphilis TP Reactive Results | | --- | --- | --- | | Chlamydia | 15 | 1a | | Cytomegalovirus (CMV) IgG | 10 | 1b | | Cytomegalovirus (CMV) IgM | 10 | 1a | | Anti-dsDNA Autoantibodies | 10 | 0 | | Epstein-Barr Virus (EBV) IgG | 10 | 1a | | Epstein-Barr Virus (EBV) IgM | 24 | 0 | | Anti-Escherichia coli (E. coli) | 10 | 1b | | Gonorrhea | 11 | 1c | | HAVAB IgG | 10 | 0 | {12} | HBc IgM | 4 | 0 | | --- | --- | --- | | Hemodialysis | 10 | 0 | | Hepatitis A Virus (HAV) | 10 | 1a | | Hepatitis B Virus (HBV) | 10 | 0 | | Hepatitis C Virus (HCV) | 10 | 1a | | Herpes Simplex Virus (HSV) | 10 | 1d | | Human Anti-Mouse Antibodies (HAMA) | 10 | 0 | | Human Immunodeficiency Virus (HIV) | 10 | 1b | | Human T-Lymphotropic Virus-I (HTLV-I) | 10 | 0 | | Human T-Lymphotropic Virus-II (HTLV-II) | 6 | 2c | | Influenza Vaccine Recipient | 20 | 0 | | Leptospirosis | 6 | 0 | | Leptospirosis IgM | 5 | 0 | | Lyme Disease | 10 | 0 | | Monoclonal Hyper IgG | 7 | 1b | | Anti-Nuclear Antibody (ANA) | 10 | 0 | | Polyclonal Hyper IgG | 3 | 0 | | Pregnant | 90 | 0 | | Rheumatoid Factor | 10 | 0 | | Rubella IgG | 10 | 0 | | Systemic Lupus Erythematosus | 10 | 0 | | Toxoplasma gondii IgG | 12 | 0 | | Toxoplasma gondii IgM | 3 | 0 | | Transplant Recipient | 10 | 0 | | Varicella Zoster Virus | 10 | 0 | | Total | 416 | 13 | ${}^{a}$ Sample was confirmed to be positive by other tests (DiaSorin LIAISON, RPR, TP-PA, FTA-ABS). ${}^{b}$ Sample was not confirmed positive by other tests (DiaSorin LIAISON, RPR, TP-PA, FTA-ABS. ${}^{c}$ Sample was reactive by ARCHITECT Syphilis TP but was not confirmed because the only available sample type (plasma) was not suitable for the other tests (DiaSorin LIAISON, RPR, TP-PA, FTA-ABS). ${}^{d}$ Sample was positive by DiaSorin LIAISON and indeterminate by TP-PA and FTA-ABS. # f. Assay cut-off: The ARCHITECT Syphilis TP assay was initially developed as a qualitative assay with a target performance of high agreement with a syphilis Treponema pallidum particle agglutination assay (TP-PA). The assay cutoff was evaluated by ROC analysis based on the ARCHITECT Syphilis TP results from a total of 6,845 specimens. Assay sensitivity was calculated based on the 409 antibody-positive specimens; assay specificity was evaluated based on the results from the "normal" donor and diagnostic population consisting of 6,436 specimens. Specimen data were analyzed using fractions/multiples of the calibrator to determine a cutoff multiplier value (a numeric value with which the {13} mean Relative Light Unit (RLU) signal of the positive calibrator is multiplied to calculate the cutoff that exhibited optimal agreement to TPPA. The multiplier of 0.2 was selected to minimize the frequency of both false negative and false positive results versus TPPA while maximizing the differentiation between true negative and true positive patients. The analytically selected cutoff was further validated in a clinical study in the intended use population. The results confirmed that the selected cutoff multiplier of 0.2 resulted in high positive and negative % agreements when comparing to the reference method algorithm result without the need to assign an equivocal zone to the candidate assay. The ARCHITECT Syphilis TP assay results are expressed as the ratio of the sample RLU to the cutoff RLU (S/CO). The S/CO is calculated using the equation: $$ \text{S/CO} = \frac{\text{Sample RLU}}{\text{Cutoff RLU}}, \text{ where } \text{Cutoff RLU} = \text{Calibrator Mean RLU} \times 0.2 $$ 2. Comparison studies: a. Method comparison with predicate device: The results of the ARCHITECT Syphilis TP assay were compared to a composite comparator based on an algorithm of results obtained from three commercially available syphilis assays: (a) treponemal chemiluminescent immunoassay (TP-CLIA), a nontreponemal assay (Rapid Plasma Reagin [RPR]), and a second treponemal assay (Treponema Pallidum Particle Agglutination [TP-PA]). Additional details of how the final comparator result was determined are provided below in the "Clinical Studies" section. b. Matrix comparison: The suitability of various blood collection tube types for use with the ARCHITECT Syphilis TP assay was evaluated in an analytical study. Samples were obtained from multiple donors and venous blood was collected in the following vacutainer tube types: - serum (plastic)-Used as the control - dipotassium EDTA - lithium heparin - sodium heparin - lithium heparin plasma separator - serum separator (plastic) - tripotassium EDTA A total of 28 donor sets, found to be nonreactive using the ARCHITECT Syphilis TP assay were spiked with a highly reactive syphilis sample to targeted concentrations, leaving one sample in each set not spiked. A total of four levels of reactivity were evaluated: 14 {14} Group 1: Nonreactive Target S/CO: $\leq 0.40$ Group 2: High Nonreactive Target S/CO: $\sim 0.80$ Group 3: Low Reactive Target S/CO: $\sim 1.20$ Group 4: High Reactive Target S/CO: $\sim 6.00$ Each sample was tested in triplicate on one ARCHITECT $i$ 2000sR instrument and the mean of the results for each matrix type was compared to the mean of the respective serum control tube. The data was tabulated for each matrix type and for each sample. The matrix types under evaluation were considered acceptable if, when compared to the control tube type (serum, plastic): - the difference was less than or equal to $0.20\mathrm{S} / \mathrm{CO}$ for samples with S/CO values lower than 1.00, and - the $\%$ difference was greater than or equal to $-20\%$ for samples with S/CO values greater than 1.00 The ARCHITECT Syphilis TP assay showed the following mean/median S/CO difference and distribution of S/CO difference for nonreactive samples when compared to the control tube type (serum). | Tube Type | Nonreactive Samples (Native Samples, Unspiked) | | | | | High Nonreactive Samples (Target 0.8 S/CO) | | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | N | Mean/ Median S/CO Differencea | Distribution of S/CO Differences | | | N | Mean/ Median S/CO Differencea | Distribution of S/CO Differences | | | | | | | < 0.10 S/CO | 0.10–0.20 S/CO | > 0.20 S/CO | | | < 0.10 S/CO | 0.10–0.20 S/CO | > 0.20 S/CO | | Serum Separator, Plastic | 28 | 0.00 | 100.0% | 0.0% | 0.0% | 27 | 0.02 | 81.5% | 18.5% | 0.0% | | | | | (28/28) | (0/28) | (0/28) | | | (22/27) | (5/27) | (0/27) | | Dipotassium EDTA | 28 | -0.00 | 100.0% | 0.0% | 0.0% | 27 | 0.02 | 81.5% | 18.5% | 0.0% | | | | | (28/28) | (0/28) | (0/28) | | | (22/27) | (5/27) | (0/27) | | Tripotassium EDTA | 27 | -0.00 | 100.0% | 0.0% | 0.0% | 27 | 0.02 | 85.2% | 11.1% | 3.7% | | | | | (27/27) | (0/27) | (0/27) | | | (23/27) | (3/27) | (1/27) | | Lithium Heparin Plasma Separator | 28 | -0.01 | 100.0% | 0.0% | 0.0% | 27 | 0.01 | 88.9% | 11.1% | 0.0% | | | | | (28/28) | (0/28) | (0/28) | | | (24/27) | (3/27) | (0/27) | | Lithium Heparin | 28 | -0.01 | 100.0% | 0.0% | 0.0% | 27 | 0.02 | 88.9% | 7.4% | 3.7% | | | | | (28/28) | (0/28) | (0/28) | | | (24/27) | (2/27) | (1/27) | | Sodium Heparin | 28 | -0.01 | 100.0% | 0.0% | 0.0% | 26 | 0.00 | 88.5% | 11.5% | 0.0% | | | | | (28/28) | (0/28) | (0/28) | | | (23/26) | (3/26) | (0/26) | a If the Shapiro-Wilk p-value is $\leq 0.0100$ , then the value displayed is the median. The ARCHITECT Syphilis TP assay showed the following mean/median percent difference and distribution of percent difference for reactive samples when compared to the control tube type (serum). {15} | Tube Type | Low Reactive Samples (Target 1.2 S/CO) | | | | | High Reactive Samples (Target 6.0 S/CO) | | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | N | Mean/Median % Differencea | Distribution of % Differences | | | N | Mean/Median % Differencea | Distribution of % Differences | | | | | | | < 10% | 10–20% | > 20% | | | < 10% | 10–20% | > 20% | | Serum Separator, Plastic | 28 | 0.3 | 96.4% (27/28) | 3.6% (1/28) | 0.0% (0/28) | 28 | -0.1 | 96.4% (27/28) | 3.6% (1/28) | 0.0% (0/28) | | Dipotassium EDTA | 28 | 0.9 | 100.0% (28/28) | 0.0% (0/28) | 0.0% (0/28) | 28 | -0.7 | 100.0% (28/28) | 0.0% (0/28) | 0.0% (0/28) | | Tripotassium EDTA | 28 | -0.0 | 96.4% (27/28) | 3.6% (1/28) | 0.0% (0/28) | 27 | -0.3 | 100.0% (27/27) | 0.0% (0/27) | 0.0% (0/27) | | Lithium Heparin Plasma Separator | 27 | 3.7 | 63.0% (17/27) | 33.3% (9/27) | 3.7% (1/27) | 28 | -0.1 | 100.0% (28/28) | 0.0% (0/28) | 0.0% (0/28) | | Lithium Heparin | 28 | -0.3 | 96.4% (27/28) | 3.6% (1/28) | 0.0% (0/28) | 28 | -0.0 | 100.0% (28/28) | 0.0% (0/28) | 0.0% (0/28) | | Sodium Heparin | 28 | 0.8 | 96.4% (27/28) | 3.6% (1/28) | 0.0% (0/28) | 28 | -0.3 | 100.0% (28/28) | 0.0% (0/28) | 0.0% (0/28) | a If the Shapiro-Wilk p-value is $\leq 0.0100$, then the value displayed is the median. The results showed that the following blood collection tube types are acceptable for use with the ARCHITECT Syphilis TP assay: - serum (plastic) - serum with separator (plastic) - dipotassium EDTA - tripotassium EDTA - lithium heparin - sodium heparin - lithium heparin plasma with separator ## 3. Clinical Studies: A multicenter study was conducted on the ARCHITECT iSystem between July and December 2015 to evaluate the ability of the ARCHITECT Syphilis TP assay to detect antibodies (IgG and IgM) directed against *Treponema pallidum* (TP). A total of 2222 specimens were tested in the ARCHITECT Syphilis TP clinical study. Two specimens were excluded due to insufficient sample volume. The remaining 2220 specimens included 1145 prospectively collected from the intended use population; 406 pre-selected samples presumed to be positive for antibodies directed against TP based on previous laboratory testing (including 20 pregnant women known to be reactive for syphilis antibodies); 480 from apparently healthy individuals; 179 from medically diagnosed individuals with primary, secondary, or latent syphilis; and 10 specimens from pregnant females each spiked with individual high antibody-positive syphilis TP specimens The 1145 prospective specimens analyzed in the ARCHITECT Syphilis TP clinical study were collected at clinical sites in the following locations: {16} - Baltimore, Maryland: $12.4\%$ - Colton, California: $5.9\%$ Fort Lauderdale, Florida: $7.8\%$ Hyannis, Massachusetts: $0.1\%$ Los Angeles, California: $19.7\%$ Miami, Florida: $27.2\%$ San Antonio, Texas: $11.5\%$ - Temple, Texas: $14.8\%$ - Location Unknown: $0.4\%$ The clinical performance of the ARCHITECT Syphilis TP assay was evaluated by calculating positive percent agreement and negative percent agreement of the assay with the final comparator result based on an algorithm of results from three commercially available syphilis assays: a treponemal chemiluminescent immunoassay (TP-CLIA), a nontreponemal assay (Rapid Plasma Reagin [RPR]), and a second treponemal assay (Treponema Pallidum Particle Agglutination [TP-PA]). Because the clinical diagnosis of syphilis must be supported by two reactive laboratory tests, consisting of a treponemal assay and a nontreponemal assay, or at least two treponemal assays, employing an algorithm of three syphilis tests to determine the comparator result presents a picture closest to the serological "truth." The final comparator result was determined using a 2 out of 3 rule (TP-CLIA, RPR, and TP-PA). In cases where the TP-CLIA result was "equivocal" (as per the device labeling), and the TP-PA result was "inconclusive", the final comparator result could not be determined; those results would be excluded from the final analysis (there were no "indeterminate" comparator results in this study). The table below shows how the final comparator result was interpreted. | Treponemal (CLIA) (Predicate) | Non-treponemal (Predicate) | 2ndTreponemal (TPPA) | Final Comparator Result | | --- | --- | --- | --- | | Negative | Negative | Positive | Negative | | | | Negative | Negative | | | | Inconclusive | Negative | | Negative | Positive | Positive | Positive | | | | Negative | Negative | | | | Inconclusive | \(Negative^1\) | | Positive | Positive | Positive | Positive | | | | Negative | Positive | | | | Inconclusive | Positive | | Positive | Negative | Positive | Positive | | | | Negative | Negative | | | | Inconclusive | Positive | | Equivocal | Negative | Positive | Positive | | | | Negative | Negative | | | | Inconclusive | Indeterminate | | Equivocal | Positive | Positive | Positive | | | | Negative | Negative | | | | Inconclusive | Indeterminate | {17} 18 # Clinical Performance in Prospectively Collected Specimens in Intended Use Population The 1145 prospectively collected specimens in the intended use population, analyzed in the ARCHITECT Syphilis TP clinical study consisted of 442 specimens sent for routine syphilis testing (325 female and 117 male, 6-91 years old), 304 pregnant females (16-43 years old), and 399 HIV positive (44 female and 355 male, 18-72 years old). Of the 1145 prospectively-collected specimens, 136 were from pediatric subjects (6-21 years old). Each of those specimens was analyzed with the ARCHITECT Syphilis TP assay and with the three reference assays. A summary of the serological test profile for all prospectively-collected specimens in the intended use population is presented in the following table. | TP-CLIA | RPR | TP-PA | Final Comparator Result | ARCHITECT | Number of Subjects | | --- | --- | --- | --- | --- | --- | | + | + | + | + | + | 113 | | + | + | I | + | + | 1 | | + | - | + | + | + | 37 | | + | - | - | - | - | 4 | | + | - | - | - | + | 2 | | - | + | + | + | - | 6 | | - | + | + | + | + | 1 | | - | + | - | - | - | 167 | | - | + | - | - | + | 3 | | - | + | I | - | - | 1^{a} | | - | + | I | - | + | 1^{a} | | - | - | + | - | - | 8 | | - | - | + | - | + | 2 | | - | - | - | - | - | 796 | | - | - | - | - | + | 2 | | E | - | + | + | + | 1^{b} | | Total | | | | | 1145 | + = Positive/Reactive, - = Negative/Nonreactive, E = Equivocal, I = Inconclusive a Two specimens that were TP-CLIA nonreactive, RPR reactive, and TP-PA inconclusive were assigned a final comparator result of negative based on the nonreactive treponemal test. b One specimen that was TP-CLIA equivocal, RPR nonreactive, and TP-PA reactive was assigned a final comparator result of positive based on the reactive treponemal test. {18} The comparison between the ARCHITECT Syphilis TP result and the final comparator result for the prospectively-collected specimens in the intended use population is shown in the following table. Percent Agreement in the Prospective Intended Use Population | ARCHITECT Syphilis TP Result Interpretation | Final Comparator Result | | | --- | --- | --- | | | Reactive | Nonreactive | | Reactive | 153 | 10 | | Nonreactive | 6^{a} | 976 | a Six specimens were nonreactive by TP-CLIA and reactive by TP-PA and RPR. Positive percent agreement was 96.2% (153/159) with a 95% confidence interval of 92.0% to 98.3%. Negative percent agreement was 99.0% (976/986) with a 95% confidence interval of 98.1% to 99.4%. Percent Agreement by Category | Category | Positive Percent Agreement % (x/n) | 95% Confidence Interval (%) | Negative Percent Agreement % (x/n) | 95% Confidence Interval (%) | | --- | --- | --- | --- | --- | | Routine Syphilis | 97.3 (36/37) | 86.2–99.5 | 99.5 (403/405) | 98.2–99.9 | | Pregnant | NA | NA | 99.7 (303/304) | 98.2–99.9 | | HIV Positive | 95.9 (117/122) | 90.8–98.2 | 97.5 (270/277) | 94.9–98.8 | NA = not applicable ## Clinical Performance in Pre-selected Retrospective Samples Of the total 2220 specimens analyzed in the ARCHITECT Syphilis TP clinical study, 406 specimens were pre-selected and included 386 (106 female and 278 male, 16–78 years old; gender was not reported for 2 specimens) presumed positive for antibodies directed against TP based on previous laboratory testing (RPR and/or TP-PA). Pre-selected specimens also included 20 reactive pregnant female specimens. Each of these specimens was analyzed with the ARCHITECT Syphilis TP assay and with the three reference assays. A summary of the serological test profile for the preselected samples in the study is shown below. {19} | TP-CLIA | RPR | TP-PA | Final Comparator Result | ARCHITECT | Number of Subjects | | --- | --- | --- | --- | --- | --- | | + | + | + | + | + | 259 | | + | + | I | + | + | 1 | | + | - | + | + | - | 2 | | + | - | + | + | + | 116 | | + | - | - | - | - | 1 | | + | - | - | - | + | 2 | | + | - | I | + | - | 1^{a} | | - | + | - | - | - | 3 | | - | - | - | - | - | 20 | | E | - | + | + | - | 1^{b} | | Total | | | | | 406 | + = Positive/Reactive, - = Negative/Nonreactive, E = Equivocal, I = Inconclusive ^aOne specimen that was TP-CLIA reactive, RPR nonreactive, and TP-PA inconclusive was assigned a final comparator result of positive based on the reactive treponemal test. ^bOne specimen that was TP-CLIA equivocal, RPR nonreactive, and TP-PA reactive was assigned a final comparator result of positive based on the reactive treponemal test. The comparison between the ARCHITECT Syphilis TP result and the final comparator result for the pre-selected presumed positive specimens is shown below. **Percent Agreement in Pre-selected Retrospective Samples** | ARCHITECT Syphilis TP Result Interpretation | Final Comparator Result | | | --- | --- | --- | | | Reactive | Nonreactive | | Reactive | 376 | 2 | | Nonreactive | 4 | 24 | Positive percent agreement was 98.9% (376/380) with a 95% confidence interval of 97.3% to 99.6%. Negative percent agreement was 92.3% (24/26) with a 95% confidence interval of 75.9% to 97.9%. **Percent Agreement by Category** | Category | Positive Percent Agreement % (x/n) | 95% Confidence Interval (%) | Negative Percent Agreement % (x/n) | 95% Confidence Interval (%) | | --- | --- | --- | --- | --- | | Presumed Positive | 98.9 (356/360) | 97.2–99.6 | 92.3 (24/26) | 75.9–97.9 | | Reactive Pregnant | 100.0 (20/20) | 83.9–100.0 | NA | NA | NA = not applicable {20} Clinical Performance in Medically Diagnosed Individuals Of the total 2220 specimens analyzed in the ARCHITECT Syphilis TP clinical study, 179 were from individuals medically diagnosed with primary, secondary, or latent syphilis. The diagnosis of syphilis and the stage of the disease were made by a licensed physician based on the patient's clinical symptoms, medical history and laboratory test results at the time of diagnosis. These specimens included 9 females and 170 males. Reactivity of the ARCHITECT Syphilis TP Assay in Subjects Medically Diagnosed with Syphilis | Medically Diagnosed Individuals | | | ARCHITECT Syphilis TP Result | | | --- | --- | --- | --- | --- | | Syphilis Stage | Treatment Status | N | Reactive | Nonreactive | | Primary | Treated | 44 | 33 | 11^{a} | | | Untreated | 25 | 25 | 0 | | Secondary | Treated | 29 | 29 | 0 | | | Untreated | 27 | 27 | 0 | | Latent | Treated | 25 | 25 | 0 | | | Untreated | 29 | 29 | 0 | *Nine specimens were nonreactive by TP-CLIA. ## Clinical Performance in Pregnant Females A total of 334 pregnant female samples were analyzed in the ARCHITECT Syphilis TP clinical study; 304 samples were prospectively collected from women coming to the healthcare facilities at the collection sites and 20 specimens were sourced from pregnant women known to be reactive for antibodies to TP by previous laboratory testing. The percent agreement between the ARCHITECT Syphilis TP results and the final comparator results for the prospectively-collected and the pre-selected positive specimens from pregnant females is shown below, stratified by the pregnancy trimester. Percent Agreement in Pregnant Women | Category | Positive Percent Agreement % (x/n) | 95% Confidence Interval (%) | Negative Percent Agreement % (x/n) | 95% Confidence Interval (%) | | --- | --- | --- | --- | --- | | Prospectively-Collected | NA | NA | 99.7 (303/304) | 98.2–99.9 | | First Trimester | NA | NA | 100.0 (13/13) | 77.2–100.0 | | Second Trimester | NA | NA | 99.2 (126/127)^{a} | 95.7–99.9 | | Third Trimester | NA | NA | 100.0 (161/161) | 97.7–100.0 | | Trimester Unknown | NA | NA | 100.0 (3/3) | 43.8–100.0 | | Pre-Selected Positive | 100.0 (20/20) | 83.9–100.0 | NA | NA | | First Trimester | 100.0 (1/1) | 20.7–100.0 | NA | NA | {21} 22 | Second Trimester | NA | NA | NA | NA | | --- | --- | --- | --- | --- | | Third Trimester | 100.0 (5/5) | 56.6–100.0 | NA | NA | | Trimester Unknown | 100.0 (14/14) | 78.5–100.0 | NA | NA | NA = not applicable * One specimen reactive on ARCHITECT was reactive by TP-CLIA and nonreactive by RPR and TP-PA In addition, 10 specimens from nonreactive pregnant females, spiked with samples known to be highly reactive for TP antibodies, were analyzed with the ARCHITECT Syphilis TP assay and with the TP-CLIA assay. All samples tested reactive by both assays for a 100.0% positive agreement with a confidence interval of 72.2% to 100.0%. ## Clinical Performance in Apparently Healthy Individuals Of the total 2220 specimens analyzed in the ARCHITECT Syphilis TP clinical study, 480 specimens were collected from apparently healthy individuals. These included 244 females and 236 males of which 113 were pediatric subjects. The reactivity of the ARCHITECT Syphilis TP assay in apparently healthy population is shown below. | Category | ARCHITECT Syphilis TP Result | | | | --- | --- | --- | --- | | | Number of Reactive (%) | Number of Nonreactive (%) | Total | | Adults | 15 (4.1%)^{a} | 352 (95.9%) | 367 | | Pediatrics | 0 (0.0%) | 113 (100.0%) | 113 | | Total | 15 (3.1%) | 465 (96.9%) | 480 | a Fourteen specimens were reactive and 1 specimen was equivocal by TP-CLIA. a. Clinical Sensitivity: The clinical sensitivity of the assay was expressed as percent agreement of the ARCHITECT Syphilis TP assay results with the final comparator result, as explained above. b. Clinical specificity: The clinical specificity of the assay was expressed as percent agreement of the ARCHITECT Syphilis TP assay results with the final comparator result, as explained above. c. Other clinical supportive data (when a. and b. are not applicable): Of the total number of 2220 samples tested in the clinical study, 2 samples generated exception codes on the initial testing and had to be retested, each producing a final valid result. The calculated rate of invalids in this study was 0.09% with a 95% confidence interval of 0.02% to 0.33%. {22} 4. Clinical cut-off: Not applicable. 5. Expected values/Reference range: In this clinical study, there were a total of 1145 specimens prospectively collected from the intended use population and tested with the ARCHITECT Syphilis TP assay. The ARCHITECT Syphilis TP assay was reactive in 163 subjects, for a 14.2% prevalence of treponemal antibodies in the studied population. The distribution of the ARCHITECT Syphilis TP reactive and nonreactive results by age and gender is summarized below. | Age Range (Years) | Gender | ARCHITECT Syphilis TP Result | | Total | | --- | --- | --- | --- | --- | | | | Number of Reactive (%) | Number of Nonreactive (%) | | | 2 to 12 | Female | 0 (0.0) | 1 (100.0) | 1 | | 13 to 21 | Female | 1 (0.8) | 118 (99.2) | 119 | | | Male | 1 (6.3) | 15 (93.8) | 16 | | 22 to 29 | Female | 6 (2.6) | 229 (97.4) | 235 | | | Male | 12 (23.5) | 39 (76.5) | 51 | | 30 to 39 | Female | 6 (2.9) | 199 (97.1) | 205 | | | Male | 32 (23.7) | 103 (76.3) | 135 | | 40 to 49 | Female | 9 (12.9) | 61 (87.1) | 70 | | | Male | 50 (29.6) | 119 (70.4) | 169 | | 50 to 59 | Female | 5 (21.7) | 18 (78.3) | 23 | | | Male | 29 (40.8) | 42 (59.2) | 71 | | 60 to 64 | Female | 3 (50.0) | 3 (50.0) | 6 | | | Male | 2 (16.7) | 10 (83.3) | 12 | | 65 to 100 | Female | 1 (7.1) | 13 (92.9) | 14 | | | Male | 6 (33.3) | 12 (66.7) | 18 | | Total | | 163 (14.2) | 982 (85.8) | 1145 | {23} 24 N. Instrument Name: ARCHITECT i2000SR System O. System Descriptions: The ARCHITECT i2000SR System uses a System Control Center (SCC) computer system which contains software providing interface with the ARCHITECT System and a host computer. The ARCHITECT System software interprets assay information, calculates results and provides the interface for controlling the system hardware. This chemiluminescent immunoassay contains many automated specimen processing steps prior to the detection of a luminescent signal by the ARCHITECT system optics. The presence or absence of target antigen in the sample is determined by comparing the chemiluminescent signal in the reaction to the cutoff signal determined from an active calibration. 1. Modes of Operation: This is a fully automated immunoassay analyzer. Does the applicant’s device contain the ability to transmit data to a computer, webserver, or mobile device? ☐ Yes ☑ X or ☐ No Does the applicant’s device transmit data to a computer, webserver, or mobile device using wireless transmission? ☐ Yes ☑ X or ☐ No 2. Software: The latest version of the ARCHITECT Syphilis TP assay software is v9.10. The clinical studies were conducted using the ARCHITECT System software v8.10 and v9.00. The modifications made to the software from v8.10 to v9.10 do not impact the decision making process. FDA has reviewed applicant’s Hazard Analysis and software development processes for this line of product types: ☑ Yes ☐ X or ☐ No 3. Specimen Identification: {24} The specimens are identified by an internal barcode reader. 4. Specimen Sampling and Handling: The specimen processing is automated. The operator only needs to place the primary sample tube in the instrument sample carrier and place the carrier onto the loading tray. 5. Calibration: Calibration is required after assay installation and requires generation of an active calibration curve. Assays do not need to be calibrated every time they are run; however, certain variables make recalibration necessary. Calibration must be performed when a new reagent lot is used, when documentation accompanying a new version of an existing assay states that calibration is required, and when the controls are out of range. 6. Quality Control: Quality control is addressed by running external controls for the ARCHITECT Syphilis TP assay which include negative and positive controls prepared in recalcified human plasma. The positive control (inactivated) is reactive for anti-TP. P. Other Supportive Instrument Performance Characteristics Data Not Covered In The "Performance Characteristics" Section above: Not applicable. Q. Proposed Labeling: The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10. R. Conclusion: The submitted information in this premarket notification is complete and supports a substantial equivalence decision. 25