In vitro characterisation and clinical evaluation of the diagnostic accuracy of a new antigen test for SARS-CoV-2 detection.

Background and aims: Quick, user-friendly and sensitive diagnostic tools are the key to controlling the spread of the SARS-CoV-2 pandemic in the new epidemiologic landscape. The aim of this work is to characterise a new Covid-19 antigen test that uses an innovative chromatographic Affimer-based technology designed for the qualitative detection of SARS-CoV-2 antigen. As rapid technology to detect Covid-19, the test was extensively characterised in vitro. Once the analytical parameters of performance were set, the test system was challenged in a test field study. The aim of this study was to evaluate its diagnostic accuracy, as compared by the gold standard RT-PCR and other existing lateral flow tests. The study was approved by MiRNAX corporate review board to ensure i) that the test complied with all the ethical requirements, ii) that the rights of participants were protected, and iii) that donors were fully informed about the likelihood that they would not personally benefit from the research. The tests were completed under the frame of Project SENSORNAS RTC-20176501 in collaboration with MiRNAX Biosens Ltd. and Hospital Carlos III and are currently under submission and review from the Ethics Committee of Universidad Autonoma de Madrid. Keywords: respiratory disease COVID-19; SARS-CoV-2; ELISA; RT-PCR; antigen lateral flow test.


Introduction
In late December, 2019, a number of patients with viral pneumonia were found to be epidemiologically associated with the Huanan seafood market in Wuhan, in the Hubei province of China. A novel, human-infecting coronavirus, provisionally is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

Objectives
The starting point was to select the lead anti SARS-CoV-2 S1 Affimer® candidate for efficient binding to the SARS-CoV-2-S1 protein in anterior nasal swab samples and further migration of the complex, based on the characterization of the binding kinetics of the Affimer® reagents towards SARS-CoV-2 S1 protein in a label-free surface plasmon resonance imaging (SPRi) system 11,12 .
To complete the in vitro characterisation of the selected biotinylated anti SARS-CoV-2 S1 Affimer® within the system, a complete set of tests was performed using purified viral protein. Cross-reactivity and interference tests were completed to generate a specificity profile of the LFD against a) recombinant S1 coronavirus targets and b) common nasal sprays. The analytic Limit of Detection (LoD) was defined, utilizing a) the entire test system from sample preparation to detection and b) spiking inactivated virus (e.g., irradiated virus) into real clinical matrix (e.g., nasopharyngeal (NP) swabs) 13,14,15 .
To complete the estimation of the clinical performance (and roughly compare to the in vitro data) a total of 250 samples from volunteers enrolled in test-field epidemiologic studies was analysed, and the sensitivity and specificity were calculated and compared to the RT-PCR data from nasopharyngeal swabs. The studies were completed under the frame of Project SENSORNAS RTC-20176501 in collaboration with MiRNAX Biosens Ltd. and Hospital Carlos III, including a total of 150 negative samples and 100 positive samples, each test documented internally and deposited in agreement to the ISO 15189 norm.

Materials and Methods
The selection of the lead biotinylated anti SARS-CoV-2 S1 Affimer® candidate(s) was performed attending to the binding kinetics towards SARS-CoV-2 S1 protein.
To this end, a set of Affimer® candidates containing a single cysteine were diluted to 5 µM in 10 mM sodium phosphate, 2.7 mM potassium chloride, 137 mM sodium chloride, pH 7.4 (Sigma Aldrich, P4417) and were printed in triplicate to Epoxyde- is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Affimer® spot were then normalised using an injection of 3 mg/mL Sucrose (Fisher Scientific, S/8600/53). Analyte injections of SARS-CoV-2 S1 protein (ACRO Biosystems, S1N-C52H3), were performed at concentrations ranging from 100 nM to 3 nM. Response data for all SARS-CoV-2 S1 analyte injection concentrations was exported from EZSuite and analysed using Scrubber2 (Ver2.0g, HORIBA). Further referencing was performed against the negative reference Affimer® spots, and the association rate constant (ka), dissociation rate constant (kd) and equilibrium dissociation constant (KD) were then determined by fitting using a 1:1 Langmuir model.
In vitro characterisation of the novel anti SARS-CoV-2 S1 Affimer® technology: cross-reactivity and interference testing.
A specificity profile of the system was completed against recombinant S1 coronavirus targets (Table 3) with cross-reactivity observed for SARS-CoV only. In this system, the biotinylated anti SARS-CoV-2 S1 Affimer® reagent binds to the SARS-CoV-2-S1 protein and the antigen-Affimer® complex is then specifically bound to conjugated microparticles. The antigen-Affimer®-microparticle complex migrates along the lateral flow strip by capillary action until it reaches the test line.
When contact is made, the immobilized poly-streptavidin on the test line binds the complex via the available biotin label on the Affimer® reagent, whilst the unbound microparticles continue along the lateral flow strip until they reach the control line.
Interference testing was also performed with 3 common nasal sprays (Table 4).
In vitro characterisation of the novel anti SARS-CoV-2 S1 Affimer® technology: determination of the limit of detection (LoD). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted October 29, 2021. ; https://doi.org/10.1101/2021.10.28.21265544 doi: medRxiv preprint A pool of (SARS-CoV-2 negative) anterior nasal swab samples inoculated with the inactivated virus (England 02 strain lot 19/60) was diluted to 10 4 , 10 3 , 5X10 2 , 10 2 and 5X101 pfu/mL ( Figure 1). Confirmatory testing was performed at 1581, 500 and 158.1 pfu/mL giving a pass rate of 20/20 for 1581 pfu/mL and 17/20 for 500 pfu/mL. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  TaqPath COVID-19 CE-IVD RT-PCR   Table 0. Summary of consumables used for estimation of the clinical performance from volunteers enrolled in test-field epidemiologic studies.

Results
Affimer® candidate 620_826257 generated the binding kinetics towards SARS-CoV-2 S1 protein as shown in figure 1, where the SPRi sensorgrams show results in triplicate across the series of injections (100 nM to 3 nM SARS-CoV-2 S1 protein). Furthermore, Affimer® candidate 620_826257 also displayed good specificity as shown in figure 2, where its positive response to injection of 50 nM SARS-CoV-2 S1 is compared to the negative response to injections of equally prepared other related human coronavirus S1 proteins (MERS-CoV S1, HCoV-HKU1 S1, SARS-CoV S1 and HCoV-229E S1). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted October 29, 2021. ; Figure 2. SPRi sensorgrams for Affimer® candidate 620_826257 with injections of 50 nM SARS-CoV-2 S1 and other related human coronavirus S1 proteins: MERS-CoV S1, HCoV-HKU1 S1, SARS-CoV S1, HCoV-229E S1.
In order to define the cross-reactivity and interference testing of the aforementioned LFD system (Materials and Methods) with the chosen Affimer® candidate 620_826257, a specificity profile was completed against recombinant S1 coronavirus targets (Table 1) with cross-reactivity observed for SARS-CoV only.
Interference testing was also performed with 3 common nasal sprays ( Table 2).
Nicorette and Vicks were shown to have no effect on performance of the LFD.
Slight interference was observed for Pirinase with reduced intensity of the control and test line. However, no false negatives or positives were observed. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted October 29, 2021. ; Table 2. LFD interference testing. Cross-reactivity observed for SARS-CoV only.
The final analytical LoD (defined in the former section as the lowest concentration resulting in positive detection of 18 out of 20 replicates -90% of all true positive-) was confirmed across a dilution series (10 4 , 10 3 , 5X10 2 , 10 2 and 5X10 1 pfu/mL). Further testing at 600 pfu/mL was performed providing positive detection for 19 out of 20 replicates (95%), setting this final LoD value. Figure 3. The graph shows digital read outs of the test line intensity against viral titre in pfu/mL across a dilution series prepared using pooled (SARS-CoV-2 negative) anterior nasal swab samples, within which the inactivated virus (England02 strain lot 19/60) was inoculated to 10 4 , 10 3 , 5X10 2 , 10 2 and 5X10 1 pfu/mL.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted October 29, 2021. ; https://doi.org/10.1101/2021. 10.28.21265544 doi: medRxiv preprint The clinical performance of the AffiDX® SARS-CoV-2 Antigen Rapid test based on the novel SARS-CoV-2 S1 Affimer® technology was evaluated in the study conducted at different investigational sites in Madrid, Spain as it has been detailed in Materials and Methods. The consenting patients of any age, gender, or race/ethnicity who presented at the test site with a former PCR result for COVID-19 no older than 4 days were tested. RTqPCR data were the standard for comparison of the results from the nasal swab on the LFD with the AffiDX® SARS-CoV-2 Antigen Rapid test based on the novel SARS-CoV-2 S1 Affimer® technology.
150 nasal samples were identified as negative with the AffiDX® SARS-CoV-2 Antigen LFD matching their former 150 negative RTqPCR results. Therefore, there was 100% correlation for the detection of negative samples as shown in table 3, with no false positives observed.

Determination
Number of Negative identifications Specificity PCR 150 AffiDX® Antigen -Nasal 150 100% is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint A parallel study under review for publishing will further delve into comparative analysis versus a complete panel of LFDs.

Discussion
This study is the first to report the journey behind the new AffiDX® SARS-CoV-2 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted October 29, 2021. ;