Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Mar 5;65(16):e8996615. doi: 10.1002/anie.8996615

A Real‐Time Antifouling Multivalent Aptamer Platform for Wash‐Free Electrochemical Detection of Low‐Abundance Biomarkers in Human Plasma

Payel Sen 1,2, Survanshu Saxena 3, Zijie Zhang 4, Navreen Sandhu 1, Pranati Das 1, Jimmy Gu 4, Todd Hoare 3,5,✉, Yingfu Li 3,4,6,✉, Leyla Soleymani 1,3,4,6,✉
PMCID: PMC13080413  PMID: 41787840

ABSTRACT

Sensitive and specific detection of low‐abundance proteins in complex biofluids is essential for early disease diagnosis and real‐time health monitoring. Electrochemical aptamer‐based biosensors offer rapid, point‐of‐care potential, but their clinical translation has been limited by biofouling, matrix variability, and signal instability in samples such as human plasma. Here, we introduce the Real‐Time Magnetic Multivalent Aptamer (RT‐MagMAp) assay, a one‐pot, wash‐free electrochemical platform that detects the low‐abundance biomarker VEGF165 directly in diluted human plasma. The RT‐MagMAp system integrates three enabling chemical designs: (i) a hierarchical multivalent aptamer architecture combining bead‐immobilized monomeric aptamers with electrode‐bound trimeric aptamers to form highly stable electroactive sandwich assemblies; (ii) antifouling zwitterionic polymer coatings that house trimeric aptamers while suppressing nonspecific adsorption; and (iii) a dynamic internal calibration mechanism using nonfunctional mutant aptamers to correct for plasma‐dependent variability. Together, these elements enable femtomolar VEGF165 detection (32–354 fM, depending on calibration method) and quantitative performance across 124 blinded plasma samples, achieving a Pearson correlation coefficient of 1.00 and a concordance correlation coefficient of 0.996 relative to a commercial ELISA. Together, these results establish RT‐MagMAp as a robust, clinically relevant electrochemical platform capable of quantitative, wash‐free protein detection directly in complex biological fluids.

Keywords: antifouling, aptamers, electrochemical biosensing, protein biosensing, real‐time sensing

Detection of low‐abundance proteins at the point‐of‐care is key for early detection and real‐time monitoring. The clinical translation of such assays is hindered by biofouling, matrix variability, and signal instability. RT‐MagMAp assay overcomes these challenges and enables precise protein quantification in human plasma using a hierarchical assembly of aptamers, an antifouling coating, and a calibration strategy employing mutant aptamers.

graphic file with name ANIE-65-e8996615-g006.jpg

1. Introduction

Reliable quantification of low‐abundance protein biomarkers in biological fluids—a task requiring both high sensitivity and specificity—is crucial for early disease diagnosis and continuous health monitoring [1]. For such assays to be applicable in scenarios requiring frequent health monitoring for patients with chronic conditions [2, 3, 4], following clinical interventions [5, 6], or during therapy [5, 6], they must be both rapid and suitable for point‐of‐care use. This has driven growing interest in developing electrochemical biosensors that function analogously to glucose monitors but are capable of analyzing a broader range of biomarkers beyond just glucose, such as proteins. Developing such capability is particularly important because proteins constitute a major class of clinically relevant biomarkers—reflecting diverse physiological and pathological states—and are key analytes in disease diagnostics and patient monitoring [7].

Whole blood and its derivatives (plasma and serum) are the most common matrices used for analyzing protein biomarkers in clinical settings, as most systemic targets such as cytokines, growth factors, and disease indicators circulate in these fluids [8, 9]. As such, much of the clinical knowledge representing the link between biomarker concentration and health and disease status is tied to biomarker levels in blood or plasma. Despite lacking the cellular components of whole blood, plasma remains a challenging sample for electrochemical biosensors due to its variable and high protein content [10, 11, 12], the presence of clotting factors [11, 13], and variable ionic strength [14, 15, 16].

Numerous studies have sought to develop electrochemical sensors, suitable for rapid detection of proteins in plasma at the point‐of‐care [17, 18, 19, 20, 21]. Nevertheless, a key obstacle in the translation of such assays from the laboratory to practice is the ability to obtain quantitative analytical results at the point‐of‐care, without conducting sample processing, calibration, and washing that is performed in lab‐based gold standard assays such as antibody‐based enzyme‐linked immunosorbent assay (ELISA).

Aptamers have emerged as an alternative to antibodies for specific recognition of target proteins in immunoassays due to their lower production costs [22], ease of chemical modification [23], improved stability [24], and the capacity to be generated on demand through in vitro selection in response to new and emerging targets [25, 26]. Importantly, multivalent aptamers, constructs created by linking multiple aptamers into a single molecule, have previously been shown to enhance binding affinity for large viral targets [27, 28] and enable highly sensitive real‐time electrochemical assays [29, 30, 31]. However, it remains unclear whether such constructs can be adapted to build analogous assays for quantitative analysis of protein biomarkers directly in plasma, without processing steps such as target extraction and enrichment, as well as washes.

Compared to viral particles, protein targets are generally smaller, structurally more complex, and often present at lower concentrations in complex biological fluids [1, 32, 33], which creates unique challenges for their point‐of‐care detection with minimal sample processing. Moreover, in contrast to detecting viruses and other pathogenic agents, simply confirming the presence of a human protein is not sufficient, making accurate quantification within a physiologically relevant range critical. Finally, the limited translation success of aptamer‐based protein assays underscores the need to rigorously demonstrate their performance in native, clinically relevant matrices such as plasma.

In this study, we sought to address the question of whether multivalent aptamer designs can be harnessed to construct real‐time electrochemical assays that can detect and quantify proteins in heterogeneous plasma samples. We developed the Real‐Time Magnetic Multivalent Aptamer (RT‐MagMAp) assay, a one‐pot platform that monitors, in real‐time, the formation of aptamer–target–aptamer–magnetic‐bead “sandwiches” on an electrode surface to recreate the multi‐target/particle architecture of viral particles using magnetic beads. To maintain performance in unprocessed plasma, we leveraged two additional strategies: (1) the use of a zwitterionic polymer layer to resist non‐specific adsorption and (2) signal calibration using a non‐functional mutant aptamer as an internal control. Using VEGF165 (vascular endothelial growth factor), a low‐abundance plasma protein, we explored various quantification methods to ensure successful assay performance in heterogeneous plasma samples. While VEGF165 was used as a model target, the approach is anticipated to represent a universal strategy for quantifying a wide range of proteins in heterogeneous samples.

2. Results and Discussion

The RT‐MagMAp assay integrates two types of functional aptamer designs, an antifouling coating, functionalized magnetic beads, and real‐time electrochemical kinetic profiling for protein analysis (Figures 1a–b). This assay is designed to form a sandwich structure on the electrode surface only in the presence of the target (Figure 1b), which leads to a measurable change in electrochemical impedance (Figure 1c). More specifically, bead‐bound monomeric aptamers are used to extract the target from the sample solution; whereas, electrode‐bound trimeric aptamers are harnessed to deliver the target sandwich to the electrode surface for impedance‐based electrochemical readout. The antifouling coating, anchored through Au─S interactions to the gold electrode and attached to thiolated trimeric aptamers via S─S bonds, is used to produce a thin, hydrophilic, and ionically conductive film that can reduce non‐specific protein adsorption [34]. Single frequency impedance measurement is used to characterize the electrode surfaces using real‐time electrochemical kinetic profiling [29]. The formation of the sandwich structure in the presence of the target is expected to inhibit the access of the redox couple ([Fe(CN)6]4‐/3−) to the electrode surface (Figure 1b), increasing the electrochemical impedance compared to the blank solution that lacks the sandwich structure (Figure 1c).

FIGURE 1.

FIGURE 1

Open in a new tab

Design and development of the RT‐MagMAp assay. (a) Assay workflow. (b) Schematic illustration of molecular interactions at the electrode/electrolyte interface over time in the presence (left) and absence (right) of the target. (c) The change in magnitude of electrochemical impedance (|ΔZ/Z|) is recorded at every 4 min for 30 min at a frequency of 5 Hz for the target and blank solutions. T/B indicates the ratio between the |ΔZ/Z|max for the target and blank solutions. (d) Evaluation of different combinations of aptamer/bead hierarchy by measuring T/B on RT‐MagMAp without the antifouling surface coating. For (c) and (d), the target solution includes 250 pM vascular endothelial growth factor (VEGF165), 0.1 mg/mL functionalized beads, 2× binding buffer and 1× readout buffer, while the blank solution lacks VEGF165. Each bar plot is obtained by testing a single sample split into equal volumes on at least three different electrodes. The error bars represent standard deviation from the mean.

To validate the RT‐MagMAp assay, single frequency impedance (|Z|) measurements were conducted every 2 min over a 30‐min period, calculating the rate of change of impedance (|ΔZ/Z|) from the first point of measurements (t = 0) (equation S1), and plotting |ΔZ/Z| every 4 min (Figure 1c). Both target (250 pM VEGF165 spiked in binding buffer) and blank solutions led to an increasing electrochemical impedance profile. However, a significantly larger impedance signal was recorded at every time point for the target versus blank solution, leading to a target‐to‐blank ratio (T/B) of 16.04 such T/B is indicative of the formation of sandwich complexes on the electrodes in the presence of the target. The minimal increase in the signal over time for the blank solution can be attributed to either the photochemical degradation of the redox probe, which would lead to a change in the ionic strength of the electrolyte [35], or non‐specific adsorption of plasma components or magnetic beads on the antifouling polymer‐coated electrode, which can suppress but not entirely prevent protein deposition over time [34].

We suspected aptamer design to be the key to the success of this assay, which requires a balance between effective target extraction by the bead‐based aptamers and sandwich retention on an electrode surface enabled by the surface‐based aptamer. As such, we explored three additional designs leveraging a combination of monomeric and/or homotrimeric aptamers on the beads or electrode surfaces. Interestingly, using the trimeric aptamer (K d = 0.27 ± 0.04 nM (Figure S1)) on the electrode and the monomeric aptamer (K d = 21.3 ± 5.3 nM (Figure S1)) on the bead led to the highest target‐to‐blank ratio (Figure 1d). Given that VEGF165 is a homodimeric protein containing two identical receptor‐binding domains, we hypothesize that the multivalency of the trimeric aptamer enables alternating interactions with both subunits of the VEGF dimer. This multivalent interaction enhances apparent affinity through reduced dissociation rates (Figure S2) and, critically facilitates formation of a stable sandwich complex with the monomeric aptamer immobilized on beads. Whereas, having a trimeric aptamer on both the beads and the surface could disrupt the balance between the binding of targets on beads and their retention on the surface limiting the availability of the target sandwich on the surface. Using the reverse configuration of having trimeric aptamers on beads and monomeric aptamers on the surface could lead to weaker surface retention and limited real‐time electrochemical signal readout. Using monomeric aptamers on both interfaces would likely yield weaker or transient complexes and substantially diminishes assay performance. The ability to engineer the multivalent aptamers is therefore indispensable—not only improving binding strength but also establishing the molecular configuration required for real‐time signal generation.

To demonstrate the need for the magnetic beads for improving the analytical sensitivity of the assay, we compared the VEGF165 detection performance of the RT‐MAp assay reported in our previous study [29] (Figure 2a) with the RT‐MagMAp assay without (Figure 2b) and with (Figure 2c) the antifouling polymer coating. We monitored |ΔZ/Z| over time using the RT‐MAp assay for VEGF165 in diluted plasma (Figure 2a), using an optimized amount of biotinylated trimeric aptamers (Figure S3). Despite positive results for detecting SARS‐CoV‐2 in our previous study [29], the RT‐MAp assay showed poor target‐to‐blank ratio (T/B of 1.08 for 25 pM and 2.17 for 2500 pM VEGF165) with a high and variable relative standard deviation (14%–633%) at each timepoint. Our first attempt at developing the RT‐MagMAp assay without the antifouling polymer coating yielded even lower T/B (0.78 for 25 pM and 1.00 for 2500 pM of VEGF165) and again a high and variable relative standard deviation (58%–308%) (Figure 2b), even using optimized aptamer type and concentration (Figures S4–S6), as well as optimized bead concentrations (Figure S7). This is likely caused by the non‐specific binding of the magnetic beads on the electrode surface regardless of the target concentration as evident by the large |ΔZ/Z| measured for the blank. Interestingly, incorporating the antifouling polymer coating significantly improved the T/B (3.6 for 25 pM and 4.2 for 2500 pM VEGF165) and relative standard deviation of the assay (3%–19%) under optimized trimeric aptamer concentration (Figure S8) for both the high and low target concentrations investigated (Figure 2c). We hypothesize this result directly relates to reduced non‐specific binding of the magnetic beads to the high water‐binding and antifouling zwitterionic polymer coated electrode surface, as evident by the smaller values and variations observed in blank signals (Figure 2c).

FIGURE 2.

FIGURE 2

Open in a new tab

Evolution of the RT‐MagMAp assay. Comparison of the schematic representation of the blank (top) and target (middle) conditions, as well as the measured electrochemical signals (bottom) for the (a) RT‐MAp, (b) RT‐MagMAp without the antifouling polymer and the (c) RT‐MagMAp with the antifouling polymer assays. Biotin‐labelled trimeric aptamers are used in (a) and (b); whereas thiolated aptamers are used in (c). For RT‐MAp, the blank solution contained 0.01% bovine serum albumin (BSA), 20% plasma, 2× binding buffer and 1× readout buffer; whereas the target solution also contained 25 pM or 2500 pM VEGF165. For RT‐MagMAp without the antifouling polymer, the same formulation was used with the addition of 0.1 mg/mL functionalized beads. For RT‐MagMAp, the composition matched RT‐MagMAp without the antifouling polymer but excluded the 0.01% BSA. Each bar plot is obtained by testing a single sample split into equal volumes on at least three different electrodes. The error bars represent standard deviation from the mean.

The RT‐MagMAp and RT‐MAp assay are both wash‐free assays that rely on electrochemical kinetic profiling, the readout of |ΔZ/Z| over time, for reporting on aptamer‐target binding on an electrode surface. Even though the RT‐MAp assay demonstrates an LOD of 701 copies/mL [29] for detecting viral particles in saliva, it yields an LOD (>25 pM versus 32 fM) and a T/B (2.17 versus 4.2) for VEGF in plasma, which are inferior to that of the RT‐MagMAp assay. This is because beyond trimeric aptamer‐VEGF interactions on the electrode surface, the aptamer‐functionalized magnetic beads introduce a second level of multivalency: each bead presents numerous monomeric aptamers capable of binding multiple VEGF molecules simultaneously. This hierarchical multivalent architecture (Figures 1, 2) is similar to the structure present in viruses with multiple proteins present on a single particle. As such, it likely enhances local target concentration near the electrode and contributes to the strong impedance response observed. This architecture, coupled with antifouling electrodes that are designed to reduce bead‐surface interactions of the blank sample, lead to a robust T/B ratio. The antifouling coating developed here is multi‐functional: it enables the direct attachment of trimeric aptamers to the electrode surface via gold‐thiol and thiol‐thiol chemistry, promotes antifouling properties via the zwitterionic moieties, and promotes charge transfer via both zwitterionic and carboxylic acid functionalities. This method is advantageous over previous methods that use nanocomposites as antifouling electrode coatings. Particularly, the method used herein allows the deposition of thiolated aptamers, the most widely used functionalized form of aptamers, rather than carbodiimide chemistry. Furthermore, the zwitterionic chains directly used in our antifouling coating eliminate the need for backfilling with gold nanomaterials, which is needed in the previous studies to enhance the conductivity of the coating [20, 36].

We then studied the ability of the RT‐MagMAp assay to analyze VEGF165 spiked in plasma over a range of concentrations. For this study, we used the VEGF165 functional trimeric aptamer alongside a trimeric mutant aptamer (Figure 3) that was designed to exhibit minimal interaction with VEGF165. Previous studies demonstrated that employing a mutant aptamer sequence as a control for each sample is an effective way of accounting for the sample‐to‐sample variability of physiological samples [37]. For each sample at a specific concentration, we measured |ΔZ/Z| over time on a chip that employed the functional aptamer in parallel to another chip that used the mutant aptamer (Figure 3a‐b). We calculated the target‐to‐mutant (T/M) ratio at each timepoint by dividing |ΔZ/Z| obtained with the functional aptamer at each timepoint with the average |ΔZ/Z| obtained with the mutant aptamer (Figure 3c and Equation S2). Using T/M ratio as a metric, we also performed an experiment to find the plasma dilution factor that achieved the highest T/M ratio (Figure S9). Such analysis demonstrates that the maximum achievable plasma concentration of 40%, considering the dilution caused by the addition of beads, significantly reduces T/M ratio (1.47) compared to the previously used 20% (3.29); whereas the lower concentration of 10% leads to negligible changes in T/M ratio but also lowers overall signal values. This behaviour is caused by an increased concentration of non‐target proteins that contribute to nonspecific binding [38, 39] and/or crowding [40, 41] in samples containing a higher plasma concentration, and a balance between both lower target and background protein concentrations at samples with lower dilutions.

FIGURE 3.

FIGURE 3

Open in a new tab

Evaluating various calibration methods for the RT‐MagMAp assay. (a) Schematic illustration showing the use of both a functional and a mutant aptamer sequence for calibration. (b) |ΔZ/Z| signals obtained using chips modified with functional or mutant trimeric aptamers at various VEGF165 concentrations spiked in 20% plasma. (c) Target‐to‐mutant ratio (T/M) obtained from (b) by dividing the |ΔZ/Z| values from the functional and mutant aptamers at each timepoint and concentration. LOD obtained by calculating (d) T/M at the time of maximum signal (T/M)max, (e) the slope of the linear fit to the T/M curve (T/M)’, (f) the maximum signal change for the chip with the functional aptamer |ΔZ/Z|max and (g) the slope of the linear fit to the |ΔZ/Z| curve for the functional aptamer |ΔZ/Z|’. Each bar plot is obtained by testing a single sample split into equal volumes on three different electrodes modified with the functional or mutant aptamer. The error bars represent standard deviation from the mean of the three replicates.

Following these measurements, we introduced two new metrics, both calculated based on |ΔZ/Z| measurements, to calculate the limit of detection (LOD) of the assay: (1) the maximum T/M value, (T/M)max (Figure 3d and Equation S3) and (2) the slope of T/M, (T/M)’ (Figure 3e and Equation S4). Using (T/M)max, the LOD was 69 fM; using (T/M)’, the LOD was 354 fM. For comparison, we also evaluated the LOD using mutant‐independent approaches based on the maximum |ΔZ/Z|, |ΔZ/Z|max (Figure 3f and Equation S5), and its slope, |ΔZ/Z|’ (Figure 3g and Equation S6), which gave values of 32 fM and 183 fM, respectively. These results highlight that the calculated LOD strongly depends on the chosen method, following the trend: |ΔZ/Z|max < (T/M)max < |ΔZ/Z|’ < (T/M)’.

We further evaluated the specificity of the assay (Figure 4) by testing it against viral protein biomarkers (SARS‐CoV‐2 BA.5 spike protein, Influenza A H1N1 hemagglutinin, and H3N2 hemagglutinin) as well as a cancer protein biomarker (prostate specific antigen, PSA). All non‐specific signals were below the upper (solid) and lower (dotted) thresholds defined by the 2.5 nM VEGF165 and blank signals, respectively, confirming the specificity of the assay towards VEGF165.

FIGURE 4.

FIGURE 4

Open in a new tab

Specificity of the RT‐MagMAp Assay. The |ΔZ/Z| signals are measured for VEGF165 and other non‐specific proteins in 20% plasma. The solid black line represents upper threshold derived from |ΔZ/Z|max for the target – 3 × standard deviation of the target signal) while the dotted black line represents the lower threshold from |ΔZ/Z|max for the blank + 3 × standard deviation of the blank signal. Each sample concentration is tested in triplicates (n = 3) measured on three different electrodes. Each bar plot is obtained by testing a single sample split into equal volumes on three different electrodes and calculating the mean, with the error bars representing the standard deviation from the mean.

Next, we applied each of the four calibration methods developed in Figure 3 to determine the VEGF165 concentrations in double‐blinded analysis of plasma samples that included native VEGF165 as well as VEGF165 spiked at different concentrations (Figure 5). These samples were mixed with a solution containing bead‐modified monomeric aptamers, as well as other assay reagents, and were then analyzed on chips modified with functional or mutant trimeric aptamers (Figure 5a). We analyzed 124 samples (Figure S10), and compared the values obtained from each of the four calibration methods with the true reference values obtained using enzyme‐linked immunosorbent assay (ELISA, Figure S11). The evaluated VEGF165 concentrations were then plotted against the true VEGF165 concentrations (Figure 5b) to assess their correlation, using the Pearson's correlation coefficient (PCC indicated by the dotted black line in Figure 5b and Equation S7). All calibration methods show a strong linear correlation with PCC values ranging from 0.95 to 1.0, with the (T/M)’ calibration demonstrating the highest correlation (PCC = 1.0; Figure 5b).

FIGURE 5.

FIGURE 5

Open in a new tab

Evaluation of the quantification capability of the RT‐MagMAp using different calibration methods. (a) Schematic illustration of the assay operation. (b) Correlation of log‐transformed evaluated (via the RT‐MagMAp Assay) and true (via ELISA) VEGF165 concentrations. Each plot includes the linear trendline (black dotted line), line of identity (blue line), and Pearson's correlation coefficient (PCC). (c) log‐transformed Bland‐Altman plots comparing evaluated (via the RT‐MagMAp Assay) and true (via ELISA) VEGF165 concentrations showing the mean difference (dotted red line), the limit of agreement (LOA) as denoted by the top and bottom blue lines, and concordance correlation coefficient (CCC). A total of 12 different plasma samples were spiked with varying VEGF165 concentrations to generate a total of 124 test samples. Each samples was tested on a single electrode.

Despite the high PCC values, the linear fit (dotted black line) does not perfectly align with the line of identity y = x (dotted blue line in Figure 5b) for the |ΔZ/Z|’ and (T/M)max methods. This deviation suggests a systematic bias, i.e. a consistent, linearly correlated difference between the true and evaluated VEGF165 concentrations. To further analyze this bias between the evaluated and true values, log‐transformed Bland‐Altman plots [42, 43, 44] were employed (Figure 5c). In these plots, the closeness of the mean difference between the true and estimated concentrations to zero (Equation S8) reflects quantification accuracy, while the spread of the limits of agreement (Equations S9 and S10) represents the model's consistency. The (T/M)’ method exhibits the smallest and least variable mean difference, the narrowest limits of agreement (Equation S11), and the fewest outliers, indicating superior quantification accuracy and consistency. The concordance correlation coefficient (CCC) (Equation S12), calculated to understand the combined effects of both linear correlation (from PCC) and quantification accuracy (from the Bland‐Altman analysis), further confirmed that the (T/M)’ method enables the most accurate quantification, making it the most suitable for use in real‐world applications that rely on determining the target concentration in plasma.

Notably, the (T/M)’ method demonstrated the most effective performance for quantifying blinded plasma samples despite previously exhibiting the poorest LOD among the four methods. This discrepancy likely arises from experimental differences between calibration and blinded analyses experiments. During calibration experiments, a single plasma sample verified by ELISA to contain no detectable VEGF165 was spiked with known concentrations of VEGF165; in contrast, blinded quantification involved diverse plasma samples that were variable not only in VEGF165 concentration but also in overall plasma composition. In cases in which the target and blank solutions share similar background signals (calibration), |ΔZ/Z|max demonstrates the best LOD since such calibration experiments are not affected by the patient‐to‐patient changes in plasma composition and they do not undergo the more stringent background normalization imposed by the mutant signal. In contrast, in the real‐life case of each patient demonstrating a different plasma‐induced background signal (blinded quantification), it is necessary to normalize the signals to accommodate for such variations, which is achieved by the mutant signal in (T/M)max and (T/M)’ methods. Furthermore, using multiple timepoint measurements overcomes transient variations that can lead to repeatability issues, justifying the improvements introduced by the slope methods. As such, under heterogeneous conditions, the (T/M)’ method's incorporation of an internal control through the mutant aptamer combined with its use of multiple timepoint data through the slope enables more robust and reliable quantification.

Importantly, the linear dynamic range achieved with the (T/M)’ method for VEGF (25 fM‐250 pM or 0.95 pg/mL‐9.5 ng/mL) covers the range required for quantification of plasma VEGF levels in patients with cancer [45, 46], diabetic retinopathy [47, 48], and angiogenic disorders [49] (Table S1). Compared to commercial assays used for VEGF analysis (Table S2) including ELISA [50], chemiluminescent immunoassay (CLIA) [51], and flow luminescence immunoassay (FLIA) [52], our assay is significantly simpler; as such, the full assay requires only two steps (mixing and measurement) rather than several steps including washing, and faster, requiring operation time of 40 min compared to >3 h. Furthermore, even though the RT‐MagMAp assay does not offer the lowest reported LOD amongst aptamer‐based electrochemical assays for VEGF [53, 54, 55, 56, 57, 58, 59] (Table S3), it is the only assay that is both wash‐free and is thoroughly evaluated with a large number of human samples against ELISA.

3. Conclusion

In this study, we developed the Real‐Time Magnetic Multivalent Aptamer (RT‐MagMAp) assay for quantitative electrochemical detection of VEGF165 directly in human plasma. Unlike many existing aptamer‐based methods that focus primarily on detection, the RT‐MagMAp platform enables precise and reproducible quantification of a clinically relevant low‐abundance protein across diverse plasma samples from multiple donors. This capability arises from a suite of mutually reinforcing chemical and architectural innovations:

  1. Hierarchical multivalent aptamer architecture: The combination of monomeric aptamers on magnetic beads and trimeric aptamers on electrode surfaces enables the formation of highly stable, multivalent aptamer‐target “sandwich” complexes that remain bound to the electrode during real‐time impedance monitoring.

  2. Bead‐induced multivalency amplification: Aptamer‐modified beads introduce an additional layer of multivalency that strongly enhances the target‐to‐blank ratio by increasing the local impedance upon sandwich assembly.

  3. Dual‐function antifouling polymer interface: Zwitterionic polymer coatings both house the trimeric aptamers and suppress nonspecific protein adsorption, ensuring that bead‐bound targets—not plasma fouling—dominate the electrochemical response.

  4. Built‐in internal calibration: Each assay incorporates both functional and mutant aptamers, enabling internal referencing to correct for sample‐to‐sample matrix variation without external calibration curves.

  5. Time‐resolved electrochemical profiling: Real‐time monitoring generates rich kinetic data that can be leveraged for multi‐parameter calibration and improved quantitative accuracy in complex clinical samples.

Together, these design elements allow the RT‐MagMAp assay to achieve a femtomolar limit of detection (32‐354 fM) and deliver exceptional agreement with ELISA across a blinded cohort of 124 human plasma samples (with a Pearson correlation coefficient of 1.00 and a concordance correlation coefficient of 0.996), establishing the method as a powerful alternative to gold‐standard immunoassays.

While the present assay design takes advantage of the dimeric architecture of VEGF, the underlying principles of hierarchical multivalency and real‐time electrochemical profiling are far more general. In principle, this framework can be readily extended to monomeric proteins provided that two distinct aptamers—engineered to recognize different epitopes with appropriately tuned affinities—are available. Such modularity opens the door to a universalizable, plug‐and‐play platform for the quantitative measurement of a wide range of clinically relevant biomarkers.

More broadly, by integrating hierarchical multivalency with antifouling surface chemistry and internal calibration, the RT‐MagMAp assay provides a general design framework for real‐time electrochemical biosensing in complex media. Its wash‐free, one‐pot workflow and compatibility with human plasma suggest potential utility in streamlined diagnostic formats. With further refinement in aptamer discovery and multivalent assembly, this approach may be adapted to other protein targets and contribute to the development of next‐generation quantitative biosensing platforms.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: The experimental methods and data supporting this study are presented in the Supporting Information.

ANIE-65-e8996615-s001.docx (1.9MB, docx)

Acknowledgments

The research received financial support from MITACS Accelerate Internship (Rapid Diagnostics for Viral and Bacterial Pathogens: IT30008), Natural Sciences and Engineering Research Council, as well as Zentek. L. S. is funded by the Canada Research Chair program.

Contributor Information

Todd Hoare, Email: hoaretr@mcmaster.ca.

Yingfu Li, Email: liying@mcmaster.ca.

Leyla Soleymani, Email: soleyml@mcmaster.ca.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  • 1. Cohen L. and Walt D. R., “Highly Sensitive and Multiplexed Protein Measurements,” Chemical Reviews 119 (2019): 293–321, 10.1021/acs.chemrev.8b00257. [DOI] [PubMed] [Google Scholar]
  • 2. Leo D. G., Buckley B. J. R., Chowdhury M., et al., “Interactive Remote Patient Monitoring Devices for Managing Chronic Health Conditions: Systematic Review and Meta‐analysis,” Journal of Medical Internet Research [Electronic Resource] 24 (2022): e35508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Prakashan D., P R R., and Gandhi S., “A Systematic Review on the Advanced Techniques of Wearable Point‐of‐Care Devices and Their Futuristic Applications,” Diagnostics (Basel) 13 (2023): 916, 10.3390/diagnostics13050916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Lima Chagas G. C., Teixeira L., Clemente M. R. C., et al., “Use of Continuous Glucose Monitoring and Point‐of‐care Glucose Testing in Hospitalized Patients With Diabetes Mellitus in Non‐intensive Care Unit Settings: A Systematic Review and Meta‐analysis of Randomized Controlled Trials,” Diabetes Research and Clinical Practice 220 (2025): 111986, 10.1016/j.diabres.2024.111986. [DOI] [PubMed] [Google Scholar]
  • 5. Iqbal M. J., Javed Z., Herrera‐Bravo J., et al., “Biosensing Chips for Cancer Diagnosis and Treatment: A New Wave Towards Clinical Innovation,” Cancer cell international 22 (2022), 10.1186/s12935-022-02777-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Singh S., Podder P. S., Russo M., Henry C., and Cinti S., “Tailored Point‐of‐care Biosensors for Liquid Biopsy in the Field of Oncology,” Lab on A Chip 23 (2023): 44–61, 10.1039/D2LC00666A. [DOI] [PubMed] [Google Scholar]
  • 7. Steffen P., Kwiatkowski M., Robertson W. D., et al., “Protein Species as Diagnostic Markers,” Journal of Proteomics 134 (2016): 5–18, 10.1016/j.jprot.2015.12.015. [DOI] [PubMed] [Google Scholar]
  • 8. Whittaker K., Burgess R., Jones V., et al., “Quantitative Proteomic Analyses in Blood: A Window to human Health and Disease,” Journal of Leukocyte Biology 106 (2019): 759–775, 10.1002/JLB.MR1118-440R. [DOI] [PubMed] [Google Scholar]
  • 9. Geyer P. E., Holdt L. M., Teupser D., and Mann M., “Revisiting Biomarker Discovery by Plasma Proteomics,” Molecular Systems Biology 13 (2017): 942, 10.15252/msb.20156297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Hortin G. L. and Sviridov D., “The Dynamic Range Problem in the Analysis of the Plasma Proteome,” Journal of Proteomics 73 (2010): 629–636, 10.1016/j.jprot.2009.07.001. [DOI] [PubMed] [Google Scholar]
  • 11. Anderson N. L. and Anderson N. G., “The Human Plasma Proteome,” Molecular & Cellular Proteomics 1 (2002): 845–867, 10.1074/mcp.R200007-MCP200. [DOI] [PubMed] [Google Scholar]
  • 12. Leeman M., Choi J., Hansson S., Storm M. U., and Nilsson L., “Proteins and Antibodies in Serum, Plasma, and Whole Blood—size Characterization Using Asymmetrical Flow Field‐flow Fractionation (AF4),” Analytical and Bioanalytical Chemistry 410 (2018): 4867–4873, 10.1007/s00216-018-1127-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Wilkirson E. C., Singampalli K. L., Li J., et al., “Affinity‐based Electrochemical Sensors for Biomolecular Detection in Whole Blood,” Analytical and Bioanalytical Chemistry 415 (2023): 3983–4002, 10.1007/s00216-023-04627-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Dimeski G. and Barnett R. J., “Effects of Total Plasma Protein Concentration on Plasma Sodium, Potassium and Chloride Measurements by an Indirect Ion Selective Electrode Measuring System,” Critical Care and Resuscitation 7 (2005): 12–15, 10.1016/S1441-2772(23)01564-8. [DOI] [PubMed] [Google Scholar]
  • 15. Newhouse I. J., Johnson K. P., Montelpare W. J., and McAuliffe J. E., “Variability Within Individuals of Plasma Ionic Magnesium Concentrations,” BMC Physiology [Electronic Resource] 2 (2002): 6, 10.1186/1472-6793-2-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Feldman M. and Dickson B., “Plasma Electrolyte Distributions in Humans—Normal or Skewed?,” The American Journal of the Medical Sciences 354 (2017): 453–457, 10.1016/j.amjms.2017.07.012. [DOI] [PubMed] [Google Scholar]
  • 17. Zamay G. S., Zamay T. N., Kolovskii V. A., et al., “Electrochemical Aptasensor for Lung Cancer‐related Protein Detection in Crude Blood Plasma Samples,” Scientific Reports 6 (2016): 34350, 10.1038/srep34350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Centi S., Tombelli S., Minunni M., and Mascini M., “Aptamer‐Based Detection of Plasma Proteins by an Electrochemical Assay Coupled to Magnetic Beads,” Analytical Chemistry 79 (2007): 1466–1473, 10.1021/ac061879p. [DOI] [PubMed] [Google Scholar]
  • 19. Kikkeri K., Wu D., and Voldman J., “A Sample‐to‐answer Electrochemical Biosensor System for Biomarker Detection,” Lab on A Chip 22 (2022): 100–107, 10.1039/D1LC00910A. [DOI] [PubMed] [Google Scholar]
  • 20. Del Río J. S., Henry O. Y. F., Jolly P., and Ingber D. E., “An Antifouling Coating That Enables Affinity‐based Electrochemical Biosensing in Complex Biological Fluids,” Nature Nanotechnology 14 (2019): 1143–1149. [DOI] [PubMed] [Google Scholar]
  • 21. Timilsina S. S., Durr N., Yafia M., Sallum H., Jolly P., and Ingber D. E., “Ultrarapid Method for Coating Electrochemical Sensors With Antifouling Conductive Nanomaterials Enables Highly Sensitive Multiplexed Detection in Whole Blood,” Advanced Healthcare Materials 11 (2022): 2102244, 10.1002/adhm.202102244. [DOI] [PubMed] [Google Scholar]
  • 22. Zhou J. and Rossi J., “Aptamers as Targeted Therapeutics: Current Potential and Challenges,” Nature Reviews Drug Discovery 16 (2017): 181–202, 10.1038/nrd.2016.199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Jayasena S. D., “Aptamers: An Emerging Class of Molecules That Rival Antibodies in Diagnostics,” Clinical Chemistry 45 (1999): 1628–1650, 10.1093/clinchem/45.9.1628. [DOI] [PubMed] [Google Scholar]
  • 24. Liu S., Xu Y., Jiang X., Tan H., and Ying B., “Translation of Aptamers Toward Clinical Diagnosis and Commercialization,” Biosensors and Bioelectronics 208 (2022): 114168, 10.1016/j.bios.2022.114168. [DOI] [PubMed] [Google Scholar]
  • 25. Yüce M., Ullah N., and Budak H., “Trends in Aptamer Selection Methods and Applications,” Analyst 140 (2015): 5379–5399. [DOI] [PubMed] [Google Scholar]
  • 26. Stoltenburg R., Reinemann C., and Strehlitz B., “SELEX—A (r)Evolutionary Method to Generate High‐affinity Nucleic Acid Ligands,” Biomolecular Engineering 24 (2007): 381–403, 10.1016/j.bioeng.2007.06.001. [DOI] [PubMed] [Google Scholar]
  • 27. Li J., Zhang Z., Gu J., et al., “Three on Three: Universal and High‐Affinity Molecular Recognition of the Symmetric Homotrimeric Spike Protein of SARS‐CoV‐2 With a Symmetric Homotrimeric Aptamer,” Journal of the American Chemical Society 144 (2022): 23465–23473, 10.1021/jacs.2c09870. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Liu R., Li J., Gu J., Salena B. J., and Li Y., “Higher Affinity Enables More Accurate Detection of SARS‐CoV‐2 in Human Saliva Using Aptamer‐Based Litmus Test,” Angewandte Chemie International Edition 63 (2024): e202407049, 10.1002/anie.202407049. [DOI] [PubMed] [Google Scholar]
  • 29. Sen P., Zhang Z., Sakib S., et al., “High‐Precision Viral Detection Using Electrochemical Kinetic Profiling of Aptamer‐Antigen Recognition in Clinical Samples and Machine Learning,” Angewandte Chemie International Edition 136 (2024): e202400413. [DOI] [PubMed] [Google Scholar]
  • 30. Zhang Z., Pandey R., Li J., et al., “High‐Affinity Dimeric Aptamers Enable the Rapid Electrochemical Detection of Wild‐Type and B.1.1.7 SARS‐CoV‐2 in Unprocessed Saliva,” Angewandte Chemie (International ed in English) 60 (2021): 24266–24274, 10.1002/anie.202110819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Hamidi S. V., Jahromi A. K., Hosseini I. I., et al., “Surface‐Based Multimeric Aptamer Generation and Bio‐Functionalization for Electrochemical Biosensing Applications,” Angewandte Chemie 136 (2024): e202402808, 10.1002/ange.202402808. [DOI] [PubMed] [Google Scholar]
  • 32. Lošdorfer Božič A., Šiber A., and Podgornik R., “Statistical Analysis of Sizes and Shapes of Virus Capsids and Their Resulting Elastic Properties,” Journal of Biological Physics 39 (2013): 215–228, 10.1007/s10867-013-9302-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Guo L., Zhao Y., Huang Q., et al., “Electrochemical Protein Biosensors for Disease Marker Detection: Progress and Opportunities,” Microsyst Nanoeng 10 (2024): 65, 10.1038/s41378-024-00700-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Saxena S., Lu Y., Zhang Z., Li Y., Soleymani L., and Hoare T., “Zwitter‐repel: An Anti‐fouling Coating Promoting Electrochemical Biosensing in Biological Fluids,” Chemical Engineering Journal 495 (2024): 153522, 10.1016/j.cej.2024.153522. [DOI] [Google Scholar]
  • 35. Vogt S., Su Q., Gutiérrez‐Sánchez C., and Nöll G., “Critical View on Electrochemical Impedance Spectroscopy Using the Ferri/Ferrocyanide Redox Couple at Gold Electrodes,” Analytical Chemistry 88 (2016): 4383–4390, 10.1021/acs.analchem.5b04814. [DOI] [PubMed] [Google Scholar]
  • 36. Saxena S., Sen P., Soleymani L., and Hoare T., “Anti‐Fouling Polymer or Peptide‐Modified Electrochemical Biosensors for Improved Biosensing in Complex Media,” Advanced Sensor Research 3 (2024): 2300170, 10.1002/adsr.202300170. [DOI] [Google Scholar]
  • 37. Adhikari B. R., Wang Q., Li J., et al., “A Real‐Time Enzyme‐Linked Aptamer Assay (RT‐eELAA) for Rapid and Sensitive Detection of Viral Antigens in Saliva Using Electrochemical Readout,” Analytical Chemistry 97 (2025): 13698–13706, 10.1021/acs.analchem.5c02555. [DOI] [PubMed] [Google Scholar]
  • 38. Szunerits S., Pagneux Q., M'Barek Y. B., Vassal S., and Boukherroub R., “Do Not Let Electrode Fouling be the Enemy of Bioanalysis,” Bioelectrochemistry 153 (2023): 108479, 10.1016/j.bioelechem.2023.108479. [DOI] [PubMed] [Google Scholar]
  • 39. Gerszten R. E., Accurso F., Bernard G. R., et al., “Challenges in Translating Plasma Proteomics From Bench to Bedside: Update From the NHLBI Clinical Proteomics Programs,” American Journal of Physiology‐Lung Cellular and Molecular Physiology 295 (2008): L16–L22, 10.1152/ajplung.00044.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Ricci F., Lai R. Y., Heeger A. J., Plaxco K. W., and Sumner J. J., “Effect of Molecular Crowding on the Response of an Electrochemical DNA Sensor,” Langmuir 23 (2007): 6827–6834, 10.1021/la700328r. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Demek E. and Arroyo‐Currás N., “General Approach to Achieving Electrochemical Aptamer‐based Sensor Sensitivity of Buffer in Blood Plasma,” Nature Partner Journal Biosensing 3 (2026), 10.1038/s44328-025-00066-7. [DOI] [Google Scholar]
  • 42. Euser A. M., Dekker F. W., and Le Cessie S., “A Practical Approach to Bland‐Altman Plots and Variation Coefficients for Log Transformed Variables,” Journal of Clinical Epidemiology 61 (2008): 978–982, 10.1016/j.jclinepi.2007.11.003. [DOI] [PubMed] [Google Scholar]
  • 43. Abbasi R., Imanbekova M., and Wachsmann‐Hogiu S., “On‐chip Bioluminescence Biosensor for the Detection of Microbial Surface Contamination,” Biosensors and Bioelectronics 254 (2024): 116200, 10.1016/j.bios.2024.116200. [DOI] [PubMed] [Google Scholar]
  • 44. Zayat R., Goetzenich A., Lee J.‐Y., et al., “Comparison Between Radial Artery Tonometry Pulse Analyzer and Pulsed‐Doppler Echocardiography Derived Hemodynamic Parameters in Cardiac Surgery Patients: A Pilot Study,” Peer‐Reviewed Journal 5 (2017): e4132, 10.7717/peerj.4132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Kut C., Mac Gabhann F., and Popel A. S., “Where Is VEGF in the Body? A Meta‐analysis of VEGF Distribution in Cancer,” British Journal of Cancer 97 (2007): 978–985, 10.1038/sj.bjc.6603923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Abdallah A., Belal M., El Bastawisy A., and Gaafar R., “Plasma Vascular Endothelial Growth Factor 165 in Advanced Non‐small Cell Lung Cancer,” Oncology Letters 7 (2014): 2121–2129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Tsai D.‐C., Charng M.‐J., Lee F.‐L., Hsu W.‐M., and Chen S.‐J., “Different Plasma Levels of Vascular Endothelial Growth Factor and Nitric Oxide Between Patients With Choroidal and Retinal Neovascularization,” Ophthalmologica 220 (2006): 246–251, 10.1159/000093079. [DOI] [PubMed] [Google Scholar]
  • 48. Wang J., Chen S., Jiang F., et al., “Vitreous and Plasma VEGF Levels as Predictive Factors in the Progression of Proliferative Diabetic Retinopathy After Vitrectomy,” PLoS ONE 9 (2014): e110531, 10.1371/journal.pone.0110531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Sandalcioglu I. E., Wende D., Eggert A., et al., “Vascular Endothelial Growth Factor Plasma Levels Are Significantly Elevated in Patients With Cerebral Arteriovenous Malformations,” Cerebrovascular Diseases 21 (2006): 154–158, 10.1159/000090526. [DOI] [PubMed] [Google Scholar]
  • 50.“Human VEGF‐165 ELISA Kit,” can be found under, https://www.abcam.com/en‐us/products/elisa‐kits/human‐vegf‐165‐elisa‐kit‐ab273164.
  • 51.“Human VEGF165 (Vascular Endothelial Growth Factor165) CLIA Kit (HUES01139),” can be found under, https://www.assaygenie.com/human‐vegf165‐vascular‐endothelial‐growth‐factor165‐clia‐kit‐hues01139/?srsltid=AfmBOoraeekGmtowLbk4k‐sBaZXHxWThJD‐8xh‐DG2GczoxvCWFicdcY.
  • 52.“Multiplex Assay Kit for Vascular Endothelial Growth Factor 165 (VEGF165),etc. by FLIA (Flow Luminescence Immunoassay),” can be found under, https://www.cloud‐clone.com/products/LMB696Hu.html.
  • 53. Liustrovaite V., Ratautaite V., Ramanaviciene A., et al., “Electrochemical Sensor for Vascular Endothelial Growth Factor Based on Self‐assembling DNA Aptamer Structure,” Science of The Total Environment 955 (2024): 177151, 10.1016/j.scitotenv.2024.177151. [DOI] [PubMed] [Google Scholar]
  • 54. Park Y., Hong M.‐S., Lee W.‐H., Kim J.‐G., and Kim K., “Highly Sensitive Electrochemical Aptasensor for Detecting the VEGF165 Tumor Marker With PANI/CNT Nanocomposites,” Biosensors 11 (2021): 114, 10.3390/bios11040114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Liao Y., Hu H., Tang Z., Deng D., and Li C., “Development of a Polydopamine–graphene Nanocomposite Electrochemical Aptasensor for Sensitive Detection of Lung Cancer Biomarker VEGF 165,” Biomedical Materials 20 (2025): 055009, 10.1088/1748-605X/adffd2. [DOI] [PubMed] [Google Scholar]
  • 56. Pourbahram B., Majd S. M., and Shamsipur M., “DPV and EIS‐based Ultrasensitive Aptasensor for VEGF165 Detection Based on Nanoporous Gold Platform Modified With SH‐aptamer,” Sensing and Bio‐Sensing Research 47 (2025): 100766, 10.1016/j.sbsr.2025.100766. [DOI] [Google Scholar]
  • 57. Shamsipur M., Farzin L., Amouzadeh Tabrizi M., and Molaabasi F., “Highly Sensitive Label Free Electrochemical Detection of VGEF165 Tumor Marker Based on ‘Signal off’ and ‘Signal on’ strategies Using an Anti‐VEGF165 Aptamer Immobilized BSA‐gold Nanoclusters/Ionic Liquid/Glassy Carbon Electrode,” Biosensors and Bioelectronics 74 (2015): 369–375, 10.1016/j.bios.2015.06.079. [DOI] [PubMed] [Google Scholar]
  • 58. Ni S., Shen Z., Zhang P., and Liu G., “Enhanced Performance of an Electrochemical Aptasensor for Real‐time Detection of Vascular Endothelial Growth Factor (VEGF) by Nanofabrication and Ratiometric Measurement,” Analytica Chimica Acta 1121 (2020): 74–82, 10.1016/j.aca.2020.05.003. [DOI] [PubMed] [Google Scholar]
  • 59. Amouzadeh Tabrizi M., Shamsipur M., and Farzin L., “A High Sensitive Electrochemical Aptasensor for the Determination of VEGF165 in Serum of Lung Cancer Patient,” Biosensors and Bioelectronics 74 (2015): 764–769, 10.1016/j.bios.2015.07.032. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting File 1: The experimental methods and data supporting this study are presented in the Supporting Information.

ANIE-65-e8996615-s001.docx (1.9MB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Angewandte Chemie (International Ed. in English) are provided here courtesy of Wiley

RESOURCES