pubs.acs.org/ac Article Label-Free Plasmonic Biosensor for Rapid, Quantitative, and Highly Sensitive COVID-19 Serology: Implementation and Clinical Validation Olalla Calvo-Lozano, Miquel Sierra, Maria Soler,* Maria Carmen Estévez,* Luis Chiscano-Camón, Adolfo Ruiz-Sanmartin, Juan Carlos Ruiz-Rodriguez, Ricard Ferrer, Juan José González-López, Juliana Esperalba, Candela Fernández-Naval, Leticia Bueno, Ruben López-Aladid, Antoni Torres, Laia Fernández-Barat, Sarah Attoumani, Rémi Charrel, Bruno Coutard, and Laura M. Lechuga Cite This: Anal. Chem. 2022, 94, 975−984 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Serological tests are essential for the control and management of COVID-19 pandemic (diagnostics and surveil- lance, and epidemiological and immunity studies). We introduce a direct serological biosensor assay employing proprietary technol- ogy based on plasmonics, which offers rapid (<15 min) identification and quantification of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies in clinical samples, without signal amplification. The portable plasmonic device employs a custom-designed multiantigen (RBD peptide and N protein) sensor biochip and reaches detection limits in the low ng mL−1 range employing polyclonal antibodies. It has also been implemented employing the WHO-approved anti-SARS-CoV-2 immunoglobulin standard. A clinical validation with COVID-19 positive and negative samples (n = 120) demonstrates its excellent diagnostic sensitivity (99%) and specificity (100%). This positions our biosensor as an accurate and easy-to-use diagnostics tool for rapid and reliable COVID-19 serology to be employed both at laboratory and decentralized settings for the disease management and for the evaluation of immunological status during vaccination or treatment. I t has been over a year since the World Health Organization (WHO) declared COVID-19 as a pandemic. The outbreak of this infectious disease, which likely originated in the Hubei related to long turnaround times, rapid antigen tests have already been developed and are being employed in many countries as the point-of-care test, although their sensitivity region (China) in December 2019 caused by the SARS-CoV-2 and reliability do not reach yet those achieved with genomic virus (severe acute respiratory syndrome coronavirus 2), has molecular assays.7,8 Complementary to the detection of the rapidly spread worldwide, and it is generating unprecedented active infection, serological tests, which detect the presence of and devastating consequences at health, social, and economic immunoglobulins (Ig) in blood generated by the infected host, levels. To date, COVID-19 has affected more than 276 million play an important role in infectious disease surveillance and people, with more than 5 million deaths.1 The emergence of an pandemic management, providing relevant information to unknown virus with a lack of population’s immunity and estimate the prevalence of the virus and to better understand accurate diagnostic methods, together with the disease the dynamics of acquired immunity. In the case of SARS-CoV- peculiarities (i.e., varied symptomatology or asymptomatology 2, the immune response is soon triggered, and antibodies are in a significant percentage of the infected people, long detectable after a few days postinfection. First, IgMs appear incubation times, high transmission rate, and so forth), have during the acute infection phase, which decline with time after undoubtedly contributed to ease its unnoticeable spread and a few days or even months. Then, long-lasting IgGs are hinder a fast and early detection of many cases.2−4 generated, as well as IgA antibodies. IgGs are expected to Current standard diagnosis for the detection of an active infection relies on the detection of the SARS-CoV-2 viral Received: September 6, 2021 genetic material from respiratory samples, mainly by RT-PCR Accepted: December 13, 2021 (reverse transcription polymerase chain reaction),5,6 which Published: December 31, 2021 provides excellent levels of sensitivity and specificity, but requiring centralized and specialized laboratories, and between 3 and 48 h to deliver results. To overcome its limitations © 2021 The Authors. Published by American Chemical Society https://doi.org/10.1021/acs.analchem.1c03850 975 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article Figure 1. (A) SARS-CoV-2 virus structure and details of the spike (S) and nucleocapsid (N) proteins; (B) scheme of the two-antigen coimmobilized sensor biochips employed in the SPR biosensor for COVID-19 serology. remain in the blood stream at significant concentrations for at isolation, understanding the relationship between antibody least months after infection, conferring immunity to the levels and the severity of the symptoms, to carry out large-scale virus.9,10 Although serology assays are not suited for systematic epidemiology studies for COVID-19 incidence determination, detection of the virus, they are very helpful for the diagnosis of or helping in vaccine development.11,18,19 past infections (indirect testing) of suspected patients with The ongoing pandemic situation, thus, demands advanced negative PCR results, in the identification of asymptomatic analytical tools that overcome aforementioned sensitivity patients, and also during the development of new vaccines or limitations in serology testing, while still facilitating fast, treatments.9,11,12 In addition, serological tests are extremely quantitative, and reliable detection at the POC. Optical useful in hospitals for ICU bed management and the deisolation of post COVID-19 patients (i.e., PCR-positive biosensors are well-positioned to fulfill these needs as they patients with a positive serological test). Finally, the emergence are sensitive techniques capable of performing label-free, of SARS-CoV-2 variants with increased resistance to sero- direct, and quantitative analysis. Plasmon-based technologies, neutralization by antibodies induced after vaccination or as the surface plasmon resonance (SPR) biosensor, offer primary infection makes serological tests a key component remarkable performance and versatility, and they have become for the response to these variants. The serological assays one of the most consolidated biosensor technologies for developed for COVID-19 are based on the identification of biomolecular interactions and clinical diagnostics,20,21 with IgMs and IgGs (and, to a less extent, IgAs), which are specific potential for compactness and miniaturization. Moreover, SPR for most abundant viral antigens, including the spike protein biosensing have been applied for multiple clinical applications [S1 and S2 subunits, and the receptor-binding domain (RBD)] in virology, including serological assays related to dengue and the nucleocapsid N protein. Traditional microplate-format virus,22,23 Salmonella,24 Epstein−Barr virus,25 and also for the immunoassays, such as ELISA (enzyme-linked immunosorbent first SARS-CoV.26 A few preliminary studies and perspectives assay) and CLIA (chemiluminescence immunoassay), are have been recently reported as well for SARS-CoV-2,27−29 widely used in clinics, as they provide high sensitivities, can be automated, and offer multiplexed capabilities, but they advocating for the potential of this technology as POC require specific equipment and trained personnel in dedicated diagnostic devices. laboratories and can be time consuming because of sample We have fully implemented an SPR-based serological test manipulation and/or long incubation times.13 For massive combining RBD and N viral antigens for the detection of screening, immunochromatographic lateral flow assays (LFA) SARS-CoV-2-specific antibodies from human sera (Figure 1). have been widely spread because of its facile handling and Our SPR biosensor offers label-free and real-time monitoring rapid time-to-result response, becoming the most commercial- of biomolecular interactions, therefore enabling a one-step ized assays to perform SARS-CoV-2 serology tests. Some of quantitative serological assay performed in less than 15 min, them can differentiate the type of antibody (IgG and/or IgM) including sample injection, signal readout, and result and thus provide information regarding the stage of the interpretation. After an in-depth optimization of the bio- infection (e.g., acute phase or past infection), but only in a recognition interface and bioassay conditions, we have qualitative manner. Although they provide fast results (15 min achieved analytical sensitivity levels in the range of ng mL−1, assay) at the point-of-care (POC), some recent studies show comfortably below the estimated antibody levels in patients, that they are not reliable and accurate enough because of their which appear to be in the μg mL−1 range.30 In order to validate moderate sensitivity (90−94%).14−16 The development of serological assays capable of performing quantitative analysis is our technology, we performed a comprehensive clinical critical for some potentially useful scenarios.17 These scenarios validation with COVID-19-positive and -negative samples include monitoring acquired immunity over time, to be able to collected from patients attended in different hospitals, predict the duration of acquired immunity, evaluate comparing our results to standard and regulated techniques seroconverted patients’ plasma for potential reinfusion in (i.e., ELISA and CLIA), as well as commercial rapid tests based other patients, manage hospital beds and COVID-19 patient on LFA. 976 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article Figure 2. Real-time sensorgrams for different antibody concentrations over a (A) RBD-coated sensor chip and (B) N-coated sensor chip, in standard buffer conditions. Calibration curves in standard buffer for (C) RBD-coated sensors chips and (D) N-coated sensor chips, using the corresponding pAb. Each signal corresponds to the mean ± SD of triplicate measurements. Nonspecific antibodies were measured at a concentration of 2000 ng mL−1. IN (time ∼ 200 s) and OUT (time ∼ 1100 s) arrows indicate the start and end time of the injection, respectively. ■ EXPERIMENTAL SECTION SPR Biosensor Device. The biosensor device employed is were analyzed (ranging from 1.25 to 500 BAU mL−1) in standard buffer (PBST + DS) or in commercial serum diluted to 10% on the RBD/N coimmobilized sensor chip. All the a homemade designed and assembled SPR that incorporates all antibody solutions were injected over the sensor chip at a the optical and microfluidic components in a compact and constant flow of 15 μL min−1. In all the cases, antigen− user-friendly platform (20 × 20 cm2). A description of the antibody interaction was disrupted by injecting a 20 mM device is provided in the Supporting Information (SI) and in NaOH regeneration solution during 1 min at constant flow Figure S1. All the experiments were carried out in appropriate rate. Antigen-biofunctionalized plasmonic sensor chips could safety facilities, with the SPR biosensor located in a laboratory be reused between 15 and 20 times without altering or of biosafety level 2 (BSL-2). modifying the immobilized proteins and the assay perform- Antibody Detection Assays. The experiments were ance. performed with chips immobilized with the viral proteins (N Clinical Sample Collection. A total of 125 clinical protein, RBD peptide, or N + RBD 1:1, prepared as described samples were collected from two hospitals in Barcelona in the Supporting Information) and two different polyclonal (Spain) in three different batches. Two batches were provided antibodies, pAb-N and pAb-RBD, specific for N protein and by Vall d’Hebron University Hospital (VH.1 n = 15, and VH.2 the RBD domain, respectively, and with the first WHO n = 70), and a third batch was provided by the Clinic Hospital international standard anti-SARS-CoV-2 human immunoglo- of Barcelona (CH.1 n = 40). Details on sample and data bulin. collection are summarized in the Supporting Information. Real-time sensorgrams generated during the injection of the Data Analysis. The real-time sensorgrams were processed antibodies (100 μL) were obtained in all the cases, monitoring extracting the final response (Δλ) after signal stabilization once the specific binding in each case (i.e., shift in the position of the whole sample volume has passed through the flow cell. For the resonance peak (Δλ, nm) to higher wavelengths). For the flow rate employed and the sample volume, this single-antigen gold sensor chips, calibration curves were corresponds to approximately 1000 s after injection. Details generated by analyzing different concentrations of the on the fitting curves and to extract detection assay character- corresponding specific antibody (ranging from 100 ng mL−1 istics and statistical analysis are described in the Supporting to 10,000 ng mL−1) in standard buffer (PBST + DS) or in Information. ■ commercial serum diluted to 10%. For the RBD/N coimmobilized sensor chips, several mixtures of pAb-RBD and pAb-N antibodies (1:1) at equal concentrations (from 100 RESULTS AND DISCUSSION ng mL−1 to 10,000 ng mL−1) were prepared in serum and Biosensor Assay Development and Analytical Char- analyzed after diluting at 10%. Calibration curves were also acterization. The in-house-developed SPR biosensor plat- generated employing the first WHO-approved standard for form employed monitors the shift in the position of the serology assays, consisting of freeze-dried pooled plasma from resonance peak of the plasmonic sensor chips, which reflects eleven patients recovered from COVID-19 disease, whose binding (Δλ > 0) or desorption events (Δλ < 0), the signal stock solution has an assigned arbitrary unit of 1000 BAU being proportional to the number of events (i.e., concen- mL−1 (BAU, binding antibody units). Several concentrations tration). Our biosensor device has previously demonstrated its 977 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article potential for clinical diagnostics in several areas,31−36 including The performance of the assays in serum was directly infectious diseases,32 and also for the direct detection of evaluated with serum diluted at 10%, as the detectability range antibodies in human serum,31,34 enabling a one-step, label-free, of the assay might certainly tolerate this dilution (i.e., LOD in and sensitive and reliable detection. These features are crucial the ng mL−1 range and presumably expected Ig concentrations to develop a fast test with response times below 15−20 min in the μg mL−1 level) and still ensure a reliable semi- (thanks to the no-need of secondary reagents or further signal quantitative and quantitative detection. This dilution factor is amplification steps), and with the potential of providing considerably lower than the one commonly employed in quantitative information (i.e., the concentration range of ELISA or CLIA tests, which is around 40−200 times.30,39 In antibodies in serum). fact, under these conditions, no undesired effects were In the case of COVID-19 serological assays, a key aspect to observed for commercial serum, with negligible nonspecific maximize both specificity and sensitivity is the viral antigen adsorptions and with a similarly wide dynamic range (see used for the detection of the antibodies. The N protein and the Figure 3A). The limit of detection achieved for the N RBD peptide contained in the spike protein appear to be both specially highly specific targets.9,13,37,38 Thus, we developed two different biofunctionalized sensor chips employing the N and RBD antigens in order to capture the antibodies generated by the host. To evaluate the performance of the biosensor- based assay, we employed commercial polyclonal antibodies for both N and RBD antigens, which can mimic the pool of antibodies with different antigen affinities produced by a host individual after infection. Figure 2A,B show representative detection signals (i.e., real-time sensorgrams) obtained for the two different biofunctionalized surfaces (i.e., N and RBD) with different pAb concentrations in buffer, gradually increasing as the concentrations were higher (i.e., Δλ obtained after signal stabilization). A direct and linear relationship between the antibody concentration and the signal was observed (see Figure 2C,D) for the range of antibodies analyzed (i.e., from 100 to 10,000 ng mL−1), being possible to determine the limit of detection (LOD, defined as the concentration corresponding to a blank signal plus three times its standard deviation) in both cases: 19.9 ng mL−1 for anti-RBD immunoassay (slope = 0.2511 nm mL μg−1, R2 = 0.992) and 45.6 ng mL−1 for anti-N (slope = 0.1536 nm mL μg−1, R2 = 0.994). Some studies suggest that the antibody concentrations in COVID-19 patients’ serum might lie in the range of μg mL−1.30 Figure 3. (A) Calibration curves with pAb-N and pAb-RBD in 10% According to these values, the performance of our biosensor diluted commercial serum using three different biofunctionalized provides enough analytical sensitivity for COVID-19 sero- surfaces (N, RBD and RBD + N). Sensor response represents the logical testing with both RBD and N-coated sensor chips. The mean ± SD of three measurements. (B) Statistical comparison specificity was also evaluated in order to assure the absence of between the positive (PS) and negative (NG) clinical samples: (i) N- nonspecific interactions of antibodies with the sensor chip coated sensor chips; (ii) RBD-coated sensor chips; (iii) RBD + N- coated sensor chip. Kurskal−Wallis test (p = 0.05). Total Ig surface. As Figure 2A,B shows, neither pAb-N nor pAb-RBD concentration calculated from the WHO standard anti-SARS-CoV-2 interacted with the opposite antigen surface (i.e., net sensor immunoglobulin calibration curve is shown in the right axis. response after signal stabilization Δλ = 0 nm), proving that no cross-reactivity between the antigen−antibody pairs was taking biofunctionalized surface was twice higher than that under place. Similarly, a SARS-CoV-2 nonrelated antibody (i.e., anti- standard buffer conditions (from 45.6 to 86 ng mL−1), which CRP) did not result in any signal, overall, confirming that the might be related to a possible hindrance of the antibody− signals come exclusively from specific antigen−antibody antigen interaction because of the serum matrix. However, the interactions. RBD-biofunctionalized surface exhibited a LOD of 21.1 ng In order to apply the described methodology in serological mL−1, very similar to that obtained under standard buffer assays and therefore in patient’s sera samples, we had to take conditions. Under these conditions, both assays were further into account the influence of the serum matrix on the sensor evaluated with real clinical samples. surface and the recognition event, as undiluted serum contains In order to study the reproducibility of the assays in serum high amounts of proteins and other compounds that could dilution, the interassay variability (replicates within different generate nonspecific interactions or hinder the protein− sensor chips) expressed as CV % (i.e., the ratio of the standard antibody interaction. For this reason, we decided to employ deviation to the mean value in percentage) was studied. The a combination of blocking agents, including poly-L-lysine- values obtained for the N-protein and RBD domain were grafted poly(ethylene glycol) (PLL-g-PEG), detergent Tween below the maximum variability recommended for clinical 20, and dextran sodium sulfate (DS), all of which have analysis (15%)40 (Table S1), overall confirming a good successfully reduced nonspecific interactions in previous reproducibility of the assays (i.e., very low variability coming studies.34,35 from chip biofunctionalization and/or sample handling and 978 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article Figure 4. Performance comparison of different COVID-19 serological assays. SPR-biosensor assay, LFA tests, and ELISA tests are shown for positive and negative serum samples. LFA tests were considered as positive after the appearance of a colored band with regular (2) or strong (3) intensity in the IgG and/or IgM line, and negative for very weak (1) or noncolored bands (0). ELISA tests were considered positive for numeric values of IgG and/or IgA cutoff index (COI) > 1.1. SPR biosensor assays were considered positive for samples above the set threshold (red dotted line) calculated, as described in the Experimental section. Detection result rows show the numbers of positive (+) and negative (−) samples for each serological methodology. preparation) and thus, the suitability of these viral antigens for To improve the discrimination between negative and polyclonal antibody detection. positive samples, we assessed the performance of a serological Preliminary Assessment of Clinical Samples. We first assay employing a mixed sensor chip combining both the RBD evaluated a set of 15 clinical serum samples from 15 different and N antigens to capture antibodies targeting both proteins. patients (VH.1 collection consisting of 10 COVID-19 positive Figure 3A shows that the multiantigen sensor surface samples and 5 negative samples, collected in 2016 and stored significantly increased the detection signals of a mixture of in the Sepsis Bank of the Vall d’Hebron University Hospital both pAbs, reaching a better LOD than using the antigens Biobank). Positive serum samples were collected from patients individually. The limit of detection was of 12.75 ng mL−1 with previously diagnosed with COVID-19 by PCR and who had a a slope of 0.475 nm mL μg−1 (R2 = 0.997), and the calibration positive result of specific IgG and IgM class antibodies against curve still shows a broad dynamic range (i.e., at high the S1 subunit of SARS-CoV-2, as described for the ELISA- concentrations such as 10,000 ng mL−1), enabling the IgG-S and ELISA-IgA-S. All the samples were analyzed with N- detection and quantification of antibodies even at high coated and RBD-coated sensor chips, and a statistical concentrations. Interestingly, we can observe in Figure 3B comparison of both N-based and RBD-based serological assays that the analysis of real samples with the combined serological assay reveals higher responses, derived from the capture of was carried out. As can be seen in Figure 3B, the N-based both N and RBD antibodies. Moreover, the negative samples assays showed poor differentiation between both sample also exhibit lower signals than the single N- and RBD-based groups, not being statistically significant (p > 0.9999). RBD- assays, reflecting more specificity, which significantly reduces based assay performed better as the p-value (p = 0.0208), the threshold and its standard deviation (see Figures 3B and below 0.05, indicates that the discrimination between positive 4). From these factors, multianalyte surface is able to and negative samples does reach statistical significance. discriminate anti-SARS-CoV-2 positive samples from negative This result is in concordance with the respective calibration samples with the most relevant statistical significance (p = curves (Figure 3A) and the better sensitivity and detectability 0.0005), notably improving the performance of the individual levels reached with the RBD-based assays. In addition, RBD- antigen assays. The reproducibility study, which was performed based assay shows less dispersion of the negative sample values for viral antigens individually, was also performed for the compared to the N-based assay, resulting in the absence of multianalyte combination of RBD and N antigens (Table S1). false negative (or indeterminate) values. Table S1 compares As isolated antigen conditions, CV values related to the RBD- and N-based assays analytical parameters, where the multianalyte assay were also below 15%. These data support RBD sensor shows better sensitivity (slope = 0.261 nm mL the good reproducibility of the assay and the aptness of μg−1, R2 = 0.999, and LOD = 23.9 ng mL−1) than the N-sensor multianalyte conditions for the SARS-CoV-2 serological test. (slope = 0.0502 nm mL μg−1, R2 = 0.9016, and LOD = 80.3 ng The RBD/N-based serological biosensor assay was qual- mL−1). itatively compared to standard ELISA performed in clinics as 979 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article well as to two different commercial lateral flow serological Table 1. Clinical Sample Classification/Characterization tests: Wondfo, which detects the total Igs (against S protein) total positivea negative characterizationb and FaStep, which detects both IgG and IgM against N and S1 proteins. We employed both LFA to analyze the 15 clinical Vall d’Hebron 80 60 20 VH.1 (n = 10)ELISA (10/ Hospital (VH) 10)LFA (10/10) samples. Results are summarized in Figure 4. The results VH.2 (n = 50)CLIA (50/ obtained with ELISA, which detects IgG and IgA antibodies 50) against S1 protein, are also included. As can be observed, the Clinic Hospital 40 40 0 CH.1 (n = 40)LFA (40/ SPR serological assay result precisely concurs to the (CH) 40) commercial microplate-based assay, achieving promising mild (n = 14)moderate (n = 14)severe (n = 12) sensitivity and specificity. Despite the ELISA not providing total 120 100 20 quantitative information, a significant correlation between the a relative numeric index obtained using this method (COI, Samples from patients with a positive PCR. bCharacterization of cutoff index, extracted from the relative signal of the sample positive samples is summarized in Table S2 (Supporting Informa- tion). and a control calibrator), and the signal obtained with the SPR assay was observed for most of the samples, which might reflect the good accuracy of the biosensor assay. Interestingly, when analyzing the SPR quantitative detection results for COVID-19-positive samples, the values reveal a clear differ- ence between two groups of samples, 1−5 and 6−10. The first set (1−5) corresponded to patients with mild symptomatol- ogy, while samples 6−10 were obtained from ICU-admitted patients. The SPR signals evidence higher levels of Ig for those patients with severe symptomatology compared with the ones with mild conditions. Moreover, LFA tests failed to identify some of those positive samples (i.e., FaStep 1, 3, and 4 and Wondfo 1) and wrongly identified as positive one negative Figure 5. Sensor signal distribution of 100 COVID-19 positive (PS) sample (FaStep 1). These LFA results are in concordance with and 20 negative (NG) clinical samples. Total Ig concentration several systematic analysis, which evidence deficient sensitivity calculated from the WHO standard calibration curve is shown in the and specificity of some LFA assays.14,15 right axis. Sensitivity, specificity, PPV, NPV, and threshold are also Finally, in order to prove the quantification performance of shown. the SPR biosensor and eventually facilitate its comparison with other serology assays detecting the same class of immunoglo- study employing confirmed negative serum samples obtained bulins, we carried out a calibration curve with the first WHO before the COVID-19 pandemic (Table S3). All the 20 international standard for anti-SARS-CoV-2 immunoglobulin negative samples studied gave signals below the threshold, with the concentration expressed in the arbitrary unit of BAU while only one of the PCR positive samples was considered not mL−1. Calibration curves were generated in the same positive (indeterminate). The SPR-based serological test shows conditions, as previously described for commercial pAb, in a sensitivity and PPV of 99 and 100%, respectively. On the both standard buffer conditions (PBST+DS) and 10% diluted other hand, it was able to discriminate negative cases, with a serum. Figure S2 in the Supporting Information shows no specificity and NPV of 100% both. differences between the calibration curves depending on To evaluate and validate the accuracy of our SPR serological conditions, achieving similar limits of detection, 0.098 BAU· biosensor, we compared our results with the methods and mL−1 for PBST+DS and 0.137 BAU·mL−1 for diluted serum. techniques employed in the two hospitals. Commercial LFAs According to this, patients’ samples were analyzed, and its employed in this study report sensitivities between 90 and immunoglobulin concentration was expressed in this stand- 95%. On the other hand, for standard clinical techniques such ardized units. as CLIA and ELISA, sensitivities usually exceed 95%, reaching Clinical Validation of SPR-Based COVID-19 Serology. 100% in some cases. All cited methodologies have reported Based on the results achieved with the preliminary clinical specificities between 97 and 99.8%. Table 2 summarizes the evaluation, a larger clinical validation study was initiated. A diagnostic results obtained for the whole collection of COVID- total of 120 clinical samples were analyzed, including 100 19 positive samples (VH.1, VH.2, and CH.1) when analyzed COVID-19-positive clinical samples collected during the by ELISA, CLIA, LFAs, and our SPR biosensor. LFA results, pandemic, with confirmed SARS-CoV-2 PCR test and 20 classified according to the intensity scale described in the negative samples collected prior the outbreak (Tables 1 and Experimental section (i.e., 0-no visible color change, 1-weak, 2- S2). The serum sample collection was carried out between >10 regular, and 3-strong), were categorized as negative (intensity days to months after the PCR results, and they were assessed 0), indeterminate (intensity 1), and positive (intensity 2 and using the SPR biosensor as well as different commercial 3). ELISA, CLIA, and SPR results were categorized depending techniques such as ELISA, CLIA, and LFA (details on the on the determined threshold for each technique (Table S2 different tests employed in the experimental section and the provides all data obtained with each technique). Supporting Information). In view of the results of Table 2, we can affirm that our SPR Figure 5 shows the distribution of the sensor signal obtained biosensor equalizes and even outperforms the different employing our biosensor assay for each of the samples. Positive approved diagnostic techniques employed in this study in samples showed a variable distribution of Ig levels which might terms of sensitivity and specificity for the number of samples go from a few BAU·mL−1 to thousands of BAU·mL−1. The tested. Considering decentralized technologies, the SPR threshold value was determined from the previous assessment biosensor can provide highly accurate detection of COVID- 980 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article Table 2. Summary of COVID-19 Clinical Samples than moderate cases (Table S2). Contrarily, no immunoglo- Validation bulins were detected in mild patients. Thus, LFA assays showed a possible association between the humoral immunol- ELISA LFA VH.1 PCR SPR Euroimmun LFA Wondfo FaStep ogy response and clinical severity because of SARS-CoV-2 positive 10 10 10 9 7 infection. indeterminate 0 0 0 0 2 In contrast to LFA, for all patients, regardless of the severity negative 0 0 0 1 1 group, we detected immunoglobulin levels with the plasmonic VH.2 PCR SPR CLIA CLIA Elecsys biosensor. The antibody levels were more elevated in the Liaison moderate and severe groups versus the mild one. However, the positive 50 49 46 48 levels of immunoglobulins did not differ statistically between indeterminate 0 1 4a 0 groups (0.87[0.36−3.02], 1.44[0.50−1.83], and 1.07[0.92− Negative 0 0 0 2 1.80]]) p = 0.548 (Figure 6). In addition, we did not find the CH.1 PCR SPR LFA LFA Quick correlation between the levels of antibodies and COVID-19 Vazyme Profile symptomatology and severity (r = 0.175 and p = 0.279). positive 40 40 18 18 indeterminate 0 0 4 4 negative 0 0 18 18 a ELISA (Euroimmun) was performed to confirm indeterminate results. 19 antibodies in a 15 min assay time (same as LFA), with a diagnostic reliability equivalent to ELISA and CLIA. There- fore, we herein have demonstrated the benefit promised by label-free plasmonic biosensor technology: simple, rapid, and reliable diagnostics. Relationship between Humoral Immunity in SARS- Figure 6. Correlation outcome severity vs antibody concentration. CoV-2 Infection and Clinical Severity. To test the Sensor signal of 40 COVID-19 positive samples from individuals with capabilities of the SPR biosensor for quantitative assessment different degrees of severity (mild, moderate, and severe symptoms). of acquired immunity, a preliminary study was carried out to Spearman test (p = 0.05). Total Ig concentration calculated from the ascertain the existence of a possible correlation between the WHO standard calibration curve is shown in the right axis. severity outcome and the levels of SARS-CoV-2 antibodies in sera (as seemingly observed in the preliminary assessment with From this preliminary study, and with the limited pool of VH.1 collection). The study was performed with a set of samples analyzed, we can assert that the SPR biosensor samples after a daily screening in the Clinic Hospital of technology shows high sensitivity for identifying total SARS- Barcelona (collection CH.1, n = 40). Patients with SARS-CoV- CoV-2 immunoglobulins, but so far, it has not provided 2 antibodies were confirmed by LFA, identifying the presence conclusive information regarding a possible correlation of IgG and IgM anti-SARS-CoV-2. To stratify patients between the severity degree and immunity response reflected according to severity and the symptomatology, the date of as SARS-CoV-2 antibody concentration in serum. Although symptoms onset, symptoms description, hospital or ICU others publications cannot confirm this relationship admission, and the length of stay were analyzed. Finally, 40 either,42−44 most studies reported in bibliography state that serum samples from convalescent COVID patients with as in the case of Middle East respiratory syndrome (MERS) diverse severity [mild (n = 14), moderate (n = 14), and virus infection, there is a strong association between humoral severe (n = 12)] were included on the validation assay with the immunity response and severity outcome after SARS-CoV-2 plasmonic biosensor. All included patients were symptomatic infection.45−48 The study differences related to the number of without statistical difference on symptoms between groups samples, the time since the onset of symptoms until sample (see the Supporting Information). The time since symptoms collection, and other factors as the limitations of each study, onset until sample collection differed between groups (in days) could hinder the corroboration of this association. Although 52.00[44.75−63.25], 76.00 [67.00−88.00], and we acknowledge the necessity of completing the study with an 118.50[73.50−123.75], p < 0.0001 (Figure S3). This was extended number of samples and a longitudinal study, this attributed to a more deteriorated health status in severe pilot study exemplifies the convenience a serological patients who needed more recovery time from symptoms onset quantitative assay may provide to monitor immune response to the inclusion visit.41 evolution, even for early samples. Although LFA showed limited sensitivity for the detection of anti-SARS-CoV-2 antibodies in some cases, especially in mild severity patients, a correlation was detected between the ■ CONCLUSIONS We have demonstrated and fully validated our biosensor intensity of the LFA and COVID severity. Related to IgG, technology, based on SPR, for rapid, less than 15 min, 85.7% of moderate and 83.3% of severe patients had positive identification and quantification of total SARS-CoV-2 anti- LFA for IgG (p = 000.1), whereas 100% of mild patients had a bodies in blood serum. Different strategies were explored negative LFA result for IgG. With regard to IgM, 100% of mild depending on the antigen selected to identify SARS-CoV-2 and 93% of moderate patients had negative LFA results, but antibodies (N protein and RBD peptide), achieving the most 42% of severe patients had positive LFA for IgM (p = 0.007). sensitive and specific results when combining RBD and N Both IgG and IgM were present only in severe patients, antigens onto the SPR sensor chip surface. The multianalyte- showing in IgG case a higher intensity of the line LFA result based biosensor reached excellent limits of detection in serum 981 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article (low ng mL−1 range) that enable direct one-step detection and Dysfunction and Resuscitation Research Group, Vall quantification of SARS-CoV-2 antibodies in COVID-19 d’Hebron Research Institute (VHIR), Vall d’Hebron Hospital patients, providing an excellent discrimination between Universitari, Vall d’Hebron Barcelona Hospital Campus, positive and negative samples (p = 0.0005). Moreover, we Barcelona 08035, Spain have implemented the biosensor assay with the first approved Adolfo Ruiz-Sanmartin − Intensive Care Department, Vall anti-SARS-CoV-2 immunoglobulin standard that will allow for d’Hebron Hospital Universitari, Vall d’Hebron Barcelona further comparison with other serological assays. We have Hospital Campus, Barcelona 08035, Spain; Shock, Organ completed an extended clinical validation with COVID-19 Dysfunction and Resuscitation Research Group, Vall positive and negative samples (n = 120) that demonstrate a d’Hebron Research Institute (VHIR), Vall d’Hebron Hospital diagnostic sensitivity of 99% and diagnostic specificity of 100% Universitari, Vall d’Hebron Barcelona Hospital Campus, of our biosensor, outperforming current available techniques Barcelona 08035, Spain like immunoassays and rapid tests. We have also conducted a Juan Carlos Ruiz-Rodriguez − Intensive Care Department, preliminary study of correlation between the humoral immune Vall d’Hebron Hospital Universitari, Vall d’Hebron response and the clinical severity outcome, although a larger Barcelona Hospital Campus, Barcelona 08035, Spain; Shock, cohort would be necessary to generate more conclusive Organ Dysfunction and Resuscitation Research Group, Vall information. d’Hebron Research Institute (VHIR), Vall d’Hebron Hospital Overall, the results obtained position our biosensor device as Universitari, Vall d’Hebron Barcelona Hospital Campus, an accurate and robust tool for rapid and reliable COVID-19 Barcelona 08035, Spain serology to be employed both at laboratory and eventually in Ricard Ferrer − Intensive Care Department, Vall d’Hebron decentralized settings. In addition, the biosensor compact and Hospital Universitari, Vall d’Hebron Barcelona Hospital user-friendly platform design may pave the way to a smooth Campus, Barcelona 08035, Spain; Shock, Organ Dysfunction technological transfer. This work further illustrates the large and Resuscitation Research Group, Vall d’Hebron Research versatility that SPR biosensors account to be readily adapted to Institute (VHIR), Vall d’Hebron Hospital Universitari, Vall the detection of different types of target biomarkers, thereby d’Hebron Barcelona Hospital Campus, Barcelona 08035, becoming a potential alternative tool for rapid diagnostics with Spain great perspectives in clinical practice implementation. Juan José González-López − Clinical Microbiology ■ * ASSOCIATED CONTENT sı Supporting Information Department, Vall d’Hebron Hospital Universitari, Vall d’Hebron Barcelona Hospital Campus, Barcelona 08035, Spain; Vall d’Hebron Institut de Recerca (VHIR), Vall The Supporting Information is available free of charge at d’Hebron Barcelona Hospital Campus, Barcelona 08035, https://pubs.acs.org/doi/10.1021/acs.analchem.1c03850. Spain; Department of Genetics and Microbiology, Universitat Autònoma de Barcelona, Barcelona 08193, Spain Description of the biosensor SPR device and details of Juliana Esperalba − Clinical Microbiology Department, Vall reagents and buffers, biosensor chip biofunctionalization, d’Hebron Hospital Universitari, Vall d’Hebron Barcelona data analysis, clinical sample collections, and additional Hospital Campus, Barcelona 08035, Spain; Vall d’Hebron results for the validation of the methodology (PDF) Institut de Recerca (VHIR), Vall d’Hebron Barcelona ■ Hospital Campus, Barcelona 08035, Spain AUTHOR INFORMATION Candela Fernández-Naval − Clinical Microbiology Department, Vall d’Hebron Hospital Universitari, Vall Corresponding Authors d’Hebron Barcelona Hospital Campus, Barcelona 08035, Maria Soler − Nanobiosensors and Bioanalytical Applications Spain; Vall d’Hebron Institut de Recerca (VHIR), Vall Group (NanoB2A), Catalan Institute of Nanoscience and d’Hebron Barcelona Hospital Campus, Barcelona 08035, Nanotechnology (ICN2), CSIC, CIBER-BBN and BIST, Spain Barcelona 08193, Spain; orcid.org/0000-0001-7232- Leticia Bueno − Cellex Laboratory, CiberRes (Centro de 2277; Email: maria.soler@icn2 Investigación Biomédica en Red de Enfermedades Maria Carmen Estévez − Nanobiosensors and Bioanalytical Respiratorias, 06/06/0028), Institut d’Investigacions Applications Group (NanoB2A), Catalan Institute of Biomèdiques August Pi I Sunyer (IDIBAPS), Barcelona Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- 08036, Spain; School of Medicine, University of Barcelona, BBN and BIST, Barcelona 08193, Spain; orcid.org/ Carrer de Casanova, 143, Barcelona 08036, Spain; 0000-0003-3694-7186; Email: mcarmen.estevez@icn2.cat Department of Pneumology, Thorax Institute, Hospital Clinic Authors of Barcelona, Barcelona 08036, Spain Olalla Calvo-Lozano − Nanobiosensors and Bioanalytical Ruben López-Aladid − Cellex Laboratory, CiberRes (Centro Applications Group (NanoB2A), Catalan Institute of de Investigación Biomédica en Red de Enfermedades Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- Respiratorias, 06/06/0028), Institut d’Investigacions BBN and BIST, Barcelona 08193, Spain; orcid.org/ Biomèdiques August Pi I Sunyer (IDIBAPS), Barcelona 0000-0001-5486-2237 08036, Spain; School of Medicine, University of Barcelona, Miquel Sierra − Nanobiosensors and Bioanalytical Carrer de Casanova, 143, Barcelona 08036, Spain; Applications Group (NanoB2A), Catalan Institute of Department of Pneumology, Thorax Institute, Hospital Clinic Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- of Barcelona, Barcelona 08036, Spain BBN and BIST, Barcelona 08193, Spain Antoni Torres − Cellex Laboratory, CiberRes (Centro de Luis Chiscano-Camón − Intensive Care Department, Vall Investigación Biomédica en Red de Enfermedades d’Hebron Hospital Universitari, Vall d’Hebron Barcelona Respiratorias, 06/06/0028), Institut d’Investigacions Hospital Campus, Barcelona 08035, Spain; Shock, Organ Biomèdiques August Pi I Sunyer (IDIBAPS), Barcelona 982 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article 08036, Spain; School of Medicine, University of Barcelona, contributing to science advancement. We are indebted to the Carrer de Casanova, 143, Barcelona 08036, Spain; HCB-IDIBAPS Biobank for the human samples and data Department of Pneumology, Thorax Institute, Hospital Clinic procurement and to the Fundació Glòria Soler for its support of Barcelona, Barcelona 08036, Spain to the COVIDBANK collection. We thank the IDIBAPS Laia Fernández-Barat − Cellex Laboratory, CiberRes (Centro Biobank for its valuable contribution to sample processing and de Investigación Biomédica en Red de Enfermedades storage. The authors acknowledge the EU Horizon 2020 Respiratorias, 06/06/0028), Institut d’Investigacions Program under grant agreement no. 644956 (RAIS project) for Biomèdiques August Pi I Sunyer (IDIBAPS), Barcelona funding the Hospital Vall d’Hebron Biobank. The VHIR- 08036, Spain; School of Medicine, University of Barcelona, HUVH is supported by Plan Nacional de I + D + i 2013-2016 Carrer de Casanova, 143, Barcelona 08036, Spain; and ISCIII-Ministerio de Ciencia e Innovación, and Spanish Department of Pneumology, Thorax Institute, Hospital Clinic Network for Research in Infectious Diseases (REIPI RD16/ of Barcelona, Barcelona 08036, Spain 0016/0003)cofinanced by European Development Regional Sarah Attoumani − Unité Des Virus É mergents (UVE: Aix- Fund “A way to achieve Europe,” Operative program Univ-IRD 190-Inserm 1207), Marseille 13005, France Intelligent Growth 2014. Part of the samples and data from Rémi Charrel − Unité Des Virus É mergents (UVE: Aix-Univ- patients included in this study were provided by the Vall IRD 190-Inserm 1207), Marseille 13005, France d’Hebron University Hospital Biobank (PT17/0015/0047), Bruno Coutard − Unité Des Virus É mergents (UVE: Aix- integrated in the Spanish National Biobanks Network, and they Univ-IRD 190-Inserm 1207), Marseille 13005, France were processed following standard operating procedures with Laura M. Lechuga − Nanobiosensors and Bioanalytical the appropriate approval of the Ethical and Scientific Applications Group (NanoB2A), Catalan Institute of Committee. The authors kindly appreciate the generous Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- donation of samples and clinical data of the donors of the BBN and BIST, Barcelona 08193, Spain; orcid.org/ Sepsis Bank of HUVH Biobank and COVID-19 patients 0000-0001-5187-5358 attended at HUVH. Complete contact information is available at: https://pubs.acs.org/10.1021/acs.analchem.1c03850 ■ REFERENCES (1) COVID-19 MapJohns Hopkins Coronavirus Resource Center Author Contributions https://coronavirus.jhu.edu/map.html (accessed December 15, O.C.-L. and M.S. carried out the optimization of the assay and 2021). the SPR measurements. S.A., R.C., and B.C. carried out the (2) Hu, B.; Guo, H.; Zhou, P.; Shi, Z. L. Nat. Rev. Microbiol. 2021, synthesis of the nucleocapsid viral antigen. L.C.-C., A.R.-S., 141−154. J.C.R.-R., R.F., J.J.G.-L., J.E., and C.F.-N. collected, treated, and (3) Wu, D.; Wu, T.; Liu, Q.; Yang, Z. Int. J. Infect. Dis. 2020, 94, 44. validated the clinical samples from Vall d’Hebron Hospital and (4) Wiersinga, W. J.; Rhodes, A.; Cheng, A. C.; Peacock, S. J.; L.B., R.L.-A., A.T., and L.F.-B. from the Clinic Hospital. O.C.- Prescott, H. C. JAMA 2020, 324, 782−793. (5) Zhang, N.; Wang, L.; Deng, X.; Liang, R.; Su, M.; He, C.; Hu, L.; L., M.S., M.S., and M.-C.E. conceptualized the experiments and Su, Y.; Ren, J.; Yu, F.; Du, L.; Jiang, S. J. Med. Virol. 2020, 92, 408− wrote the presented manuscript. L.F.-B., J.J.G.-L., J.E., J.C.R.- 417. R., B.C., and L.M.L. took a part in the discussion of the (6) Hasell, J.; Mathieu, E.; Beltekian, D.; Macdonald, B.; Giattino, following manuscript and revised it critically for important C.; Ortiz-Ospina, E.; Roser, M.; Ritchie, H. Sci. Data 2020, 7, 1−7. intellectual content. All authors have given approval to the final (7) Nagura-Ikeda, M.; Imai, K.; Tabata, S.; Miyoshi, K.; Murahara, version of the manuscript. N.; Mizuno, T.; Horiuchi, M.; Kato, K.; Imoto, Y.; Iwata, M.; Mimura, Notes S.; Ito, T.; Tamura, K.; Kato, Y. J. Clin. Microbiol. 2020, 58, e01438− e01420. The authors declare no competing financial interest. ■ (8) Hirotsu, Y.; Maejima, M.; Shibusawa, M.; Nagakubo, Y.; Hosaka, K.; Amemiya, K.; Sueki, H.; Hayakawa, M.; Mochizuki, H.; Tsutsui, ACKNOWLEDGMENTS T.; Kakizaki, Y.; Miyashita, Y.; Yagi, S.; Kojima, S.; Omata, M. Int. J. ICN2 and UVE acknowledge financial support from H2020 Infect. Dis. 2020, 99, 397−402. Research and Innovation Programme of the European (9) Long, Q. X.; Liu, B. Z.; Deng, H. J.; Wu, G. C.; Deng, K.; Chen, Y. K.; Liao, P.; Qiu, J. F.; Lin, Y.; Cai, X. F.; Wang, D. Q.; Hu, Y.; Ren, Commission (H202-SC1-PHE-Coronavirus-2020, CONVAT J. H.; Tang, N.; Xu, Y. Y.; Yu, L. H.; Mo, Z.; Gong, F.; Zhang, X. L.; Project, No. 101003544). The ICN2 is funded by the CERCA Tian, W. G.; Hu, L.; Zhang, X. X.; Xiang, J. L.; Du, H. X.; Liu, H. W.; program/Generalitat de Catalunya and supported by the Lang, C. H.; Luo, X. H.; Wu, S. B.; Cui, X. P.; Zhou, Z.; Zhu, M. M.; Severo Ochoa Centres of Excellence program funded by the Wang, J.; Xue, C. J.; Li, X. F.; Wang, L.; Li, Z. J.; Wang, K.; Niu, C. C.; Spanish Research Agency (AEI, grant no. SEV-2017-0706). Yang, Q. J.; Tang, X. J.; Zhang, Y.; Liu, X. M.; Li, J. J.; Zhang, D. C.; ICN2 group is very grateful to EPI Industries (Barcelona, Zhang, F.; Liu, P.; Yuan, J.; Li, Q.; Hu, J. L.; Chen, J.; Huang, A. L. Spain) for its kind donation supporting our research in Nat. Med. 2020, 26, 845−848. COVID-19. O.C.-L. acknowledges the economic support from (10) Qu, J.; Wu, C.; Li, X.; Zhang, G.; Jiang, Z.; Li, X.; Zhu, Q.; Liu, the Spanish Ministry of Science and Innovation and the L. Clin. Infect. Dis. 2020, 71, 2255−2258. Spanish Research Agency and the European Social Fund (11) Krammer, F.; Simon, V. Science 2020, 368, 1060−1061. (ESF) (ref. BES-2017-080527) linked to the TEC 2016- (12) Leung, G. M.; Lim, W. W.; Ho, L. M.; Lam, T. H.; Ghani, A. C.; Donnelly, C. A.; Fraser, C.; Riley, S.; Ferguson, N. M.; Anderson, 78515-R project Predict. A part of the work was supported by R. M.; Hedley, A. J. Epidemiol. Inf. 2006, 134, 211−221. the European Virus Archive GLOBAL (EVA-GLOBAL) (13) Carter, L. J.; Garner, L. V.; Smoot, J. W.; Li, Y.; Zhou, Q.; project that has received funding from the EU Horizon 2020 Saveson, C. J.; Sasso, J. M.; Gregg, A. C.; Soares, D. J.; Beskid, T. R.; (grant agreement No. 871029). A.T. and L.F.-B. acknowledge Jervey, S. R.; Liu, C. ACS Cent. Sci. 2020, 6, 591−605. financial support from GENCAT-DGRIS COVID. We are (14) Whitman, J. D.; Hiatt, J.; Mowery, C. T.; Shy, B. R.; Yu, R.; indebted to all the patients who accepted to participate Yamamoto, T. N.; Rathore, U.; Goldgof, G. M.; Whitty, C.; Woo, J. 983 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984 Analytical Chemistry pubs.acs.org/ac Article M.; Gallman, A. E.; Miller, T. E.; Levine, A. G.; Nguyen, D. N.; Bapat, (33) Peláez, E. C.; Estevez, M. C.; Domínguez, R.; Sousa, C.; S. P.; Balcerek, J.; Bylsma, S. A.; Lyons, A. M.; Li, S.; Wong, A. W. Y.; Cebolla, A.; Lechuga, L. M. Anal. Bioanal. Chem. 2020, 412, 6407− Gillis-Buck, E. M.; Steinhart, Z. B.; Lee, Y.; Apathy, R.; Lipke, M. J.; 6417. Smith, J. A.; Zheng, T.; Boothby, I. C.; Isaza, E.; Chan, J.; Acenas, D. (34) Soler, M.; Estevez, M. C.; Villar-Vazquez, R.; Casal, J. I.; D.; Lee, J.; Macrae, T. A.; Kyaw, T. S.; Wu, D.; Ng, D. L.; Gu, W.; Lechuga, L. M. Anal. Chim. Acta 2016, 930, 31−38. York, V. A.; Eskandarian, H. A.; Callaway, P. C.; Warrier, L.; Moreno, (35) Peláez, E. C.; Estevez, M. C.; Portela, A.; Salvador, J. P.; Marco, M. E.; Levan, J.; Torres, L.; Farrington, L. A.; Loudermilk, R. P.; M. P.; Lechuga, L. M. Biosens. Bioelectron. 2018, 119, 149−155. Koshal, K.; Zorn, K. C.; Garcia-Beltran, W. F.; Yang, D.; Astudillo, M. (36) Calvo-Lozano, O.; Aviñó, A.; Friaza, V.; Medina-Escuela, A.; G.; Bernstein, B. E.; Gelfand, J. A.; Ryan, E. T.; Charles, R. C.; Iafrate, Huertas, C. S.; Calderón, E. J.; Eritja, R.; Lechuga, L. M. Nanomaterials 2020, 10, 1−18. A. J.; Lennerz, J. K.; Miller, S.; Chiu, C. Y.; Stramer, S. L.; Wilson, M. (37) Premkumar, L.; Segovia-Chumbez, B.; Jadi, R.; Martinez, D. R.; R.; Manglik, A.; Ye, C. J.; Krogan, N. J.; Anderson, M. S.; Cyster, J. G.; Raut, R.; Markmann, A. J.; Cornaby, C.; Bartelt, L.; Weiss, S.; Park, Ernst, J. D.; Wu, A. H. B.; Lynch, K. L.; Bern, C.; Hsu, P. D.; Marson, Y.; Edwards, C. E.; Weimer, E.; Scherer, E. M.; Rouphael, N.; A. Nat. Biotechnol. 2020, 38, 1174−1183. Edupuganti, S.; Weiskopf, D.; Tse, L. V.; Hou, Y. J.; Margolis, D.; (15) Mekonnen, D.; Mengist, H. M.; Derbie, A.; Nibret, E.; Sette, A.; Collins, M. H.; Schmitz, J.; Baric, R. S.; de Silva, A. M. Sci. Munshea, A.; He, H.; Li, B.; Jin, T. Rev. Med. Virol. 2021, 31, Immunol. 2020, 5, No. eabc8413. No. e2181. (38) Burbelo, P. D.; Riedo, F. X.; Morishima, C.; Rawlings, S.; (16) Rashid, Z. Z.; Othman, S. N.; Samat, M. N. A.; Ali, U. K.; Smith, D.; Das, S.; Strich, J. R.; Chertow, D. S.; Davey, R. T.; Cohen, Wong, K. K. Malays J. Pathol. 2020, 42, 13−21. J. I. J. Infect. Dis. 2020, 222, 206−213. (17) Lisboa Bastos, M.; Tavaziva, G.; Abidi, S. K.; Campbell, J. R.; (39) Hu, X.; Zhang, R.; An, T.; Li, Q.; Situ, B.; Ou, Z.; Wu, C.; Haraoui, L. P.; Johnston, J. C.; Lan, Z.; Law, S.; MacLean, E.; Yang, B.; Tian, P.; Hu, Y.; Ping, B.; Wang, Q.; Zheng, L. Clin. Chim. Trajman, A.; Menzies, D.; Benedetti, A.; Khan, F. A. BMJ 2020, 370, Acta 2020, 509, 288−292. 2516. (40) USFDA. Bioanalytical Method Validation, Guidance for Industry; (18) Grifoni, A.; Weiskopf, D.; Ramirez, S. I.; Mateus, J.; Dan, J. M.; US Food and Drug Administration, 2018. Moderbacher, C. R.; Rawlings, S. A.; Sutherland, A.; Premkumar, L.; (41) Fernández-Barat, L.; López-Aladid, R.; Torres, A. Eur. Respir. J. Jadi, R. S.; Marrama, D.; de Silva, A. M.; Frazier, A.; Carlin, A. F.; 2020, 56, 2002411. Greenbaum, J. A.; Peters, B.; Krammer, F.; Smith, D. M.; Crotty, S.; (42) Yongchen, Z.; Shen, H.; Wang, X.; Shi, X.; Li, Y.; Yan, J.; Chen, Sette, A. Cell 2020, 18, 1489−1501.e15. Y.; Gu, B. Emerg. Microbes Infect. 2020, 9, 833−836. (19) Zhang, X.; Tan, Y.; Ling, Y.; Lu, G.; Liu, F.; Yi, Z.; Jia, X.; Wu, (43) GeurtsvanKessel, C. H.; Okba, N. M. A.; Igloi, Z.; Bogers, S.; M.; Shi, B.; Xu, S.; Chen, J.; Wang, W.; Chen, B.; Jiang, L.; Yu, S.; Lu, Embregts, C. W. E.; Laksono, B. M.; Leijten, L.; Rokx, C.; Rijnders, J.; Wang, J.; Xu, M.; Yuan, Z.; Zhang, Q.; Zhang, X.; Zhao, G.; Wang, B.; Rahamat-Langendoen, J.; van den Akker, J. P. C.; van Kampen, J. J. S.; Chen, S.; Lu, H. Nature 2020, 583, 437−440. A.; van der Eijk, A. A.; van Binnendijk, R. S.; Haagmans, B.; (20) Masson, J. F. ACS Sens. 2017, 2, 16−30. Koopmans, M. Nat. Commun. 2020, 11, 1−5. (21) Soler, M.; Huertas, C. S.; Lechuga, L. M. Expert Rev. Mol. (44) Phipps, W. S.; SoRelle, J. A.; Li, Q.-Z.; Mahimainathan, L.; Araj, Diagn. 2019, 19, 71−81. E.; Markantonis, J.; Lacelle, C.; Balani, J.; Parikh, H.; Solow, E. B.; (22) Kumbhat, S.; Sharma, K.; Gehlot, R.; Solanki, A.; Joshi, V. J. Karp, D. R.; Sarode, R. Am. J. Clin. Pathol. 2020, 154, 459−465. (45) Zhao, J.; Yuan, Q.; Wang, H.; Liu, W.; Liao, X.; Su, Y.; Wang, Pharm. Biomed. Anal. 2010, 52, 255−259. X.; Yuan, J.; Li, T.; Li, J.; Qian, S.; Hong, C.; Wang, F.; Liu, Y.; Wang, (23) Jahanshahi, P.; Zalnezhad, E.; Sekaran, S. D.; Adikan, F. R. M. Z.; He, Q.; Li, Z.; He, B.; Zhang, T.; Fu, Y.; Ge, S.; Liu, L.; Zhang, J.; Sci. Rep. 2014, 4, 1−7. Xia, N.; Zhang, Z. Clin. Infect. Dis. 2020, 71, 2027−2034. (24) Jongerius-Gortemaker, B. G. M.; Goverde, R. L. J.; Van (46) Okba, N. M. A.; Mü ller, M. A.; Li, W.; Wang, C.; Knapen, F.; Bergwerff, A. J. Immunol. Methods 2002, 266, 33−44. Geurtsvankessel, C. H.; Corman, V. M.; Lamers, M. M.; Sikkema, (25) Riedel, T.; Rodriguez-Emmenegger, C.; de los Santos Pereira, R. S.; De Bruin, E.; Chandler, F. D.; Yazdanpanah, Y.; Le Hingrat, Q.; A.; Bědajánková, A.; Jinoch, P.; Boltovets, P. M.; Brynda, E. Biosens. Descamps, D.; Houhou-Fidouh, N.; Reusken, C. B. E. M.; Bosch, B. Bioelectron. 2014, 55, 278−284. J.; Drosten, C.; Koopmans, M. P. G.; Haagmans, B. L. Emerg. Infect. (26) Park, T. J.; Hyun, M. S.; Lee, H. J.; Lee, S. Y.; Ko, S. Talanta Dis. 2020, 26, 1478−1488. 2009, 79, 295−301. (47) Edouard, S.; Colson, P.; Melenotte, C.; Di Pinto, F.; Thomas, (27) Soler, M.; Estevez, M. C.; Cardenosa-Rubio, M.; Astua, A.; L.; La Scola, B.; Million, M.; Tissot-Dupont, H.; Gautret, P.; Stein, A.; Lechuga, L. M. ACS Sens. 2020, 5, 2663−2678. Brouqui, P.; Parola, P.; Lagier, J. C.; Raoult, D.; Drancourt, M. Eur. J. (28) Ruiz-Vega, G.; Soler, M.; Lechuga, L. M. JPhys Photonics 2021, Clin. Microbiol. Infect. Dis. 2021, 40, 361−371. 3, 11002. (48) Garcia-Beltran, W. F.; Lam, E. C.; Astudillo, M. G.; Yang, D.; (29) Djaileb, A.; Charron, B.; Jodaylami, M. H.; Thibault, V.; Coutu, Miller, T. E.; Feldman, J.; Hauser, B. M.; Caradonna, T. M.; Clayton, J.; Stevenson, K.; Forest, S.; Live, L. S.; Boudreau, D.; Pelletier, J. N.; K. L.; Nitido, A. D.; Murali, M. R.; Alter, G.; Charles, R. C.; Dighe, A.; Masson, J. F. A rapid and quantitative serum test for SARS-CoV-2 Branda, J. A.; Lennerz, J. K.; Lingwood, D.; Schmidt, A. G.; Iafrate, A. antibodies with portable surface plasmon resonance sensing. J.; Balazs, A. B. Cell 2021, 184, 476−488.e11. ChemRxiv. 2020 https://doi.org/10.26434/chemrxiv.12118914.v1. (30) Ma, H.; Zeng, W.; He, H.; Zhao, D.; Yang, Y.; Jiang, D.; Zhou, P.; Qi, Y.; He, W.; Zhao, C.; Yi, R.; Wang, X.; Wang, B.; Xu, Y.; Yang, Y.; Kombe, A. J.; Ding, C.; Xie, J.; Gao, Y.; Cheng, L.; Li, Y.; Ma, X.; Jin, T.COVID-19 diagnosis and study of serum SARS-CoV-2 specific IgA, IgM and IgG by chemiluminescence immunoanalysis. medRxiv. 2020 https://doi.org/10.1101/2020.04.17.20064907. (31) Soler, M.; Mesa-Antunez, P.; Estevez, M. C.; Ruiz-Sanchez, A. J.; Otte, M. A.; Sepulveda, B.; Collado, D.; Mayorga, C.; Torres, M. J.; Perez-Inestrosa, E.; Lechuga, L. M. Biosens. Bioelectron. 2015, 66, 115−123. (32) Peláez, E. C.; Estevez, M. C.; Mongui, A.; Menéndez, M. C.; Toro, C.; Herrera-Sandoval, O. L.; Robledo, J.; García, M. J.; Del Portillo, P.; Lechuga, L. M. ACS Infect. Dis. 2020, 6, 1110−1120. 984 https://doi.org/10.1021/acs.analchem.1c03850 Anal. Chem. 2022, 94, 975−984