Application of affibody-based biosensors in cancer biomarker detection

Electrochemical biosensors

In the study by Ravalli et al. [83], a label-free impedimetric affisensor for the determination of the cancer biomarker HER2 was developed, as illustrated in Figure 3 [83]. The procedure involved immobilizing anti-HER2 affibody molecules with terminal cysteine modifications on the surfaces of a screen-printed graphite electrode (SPGE) and a gold nanoparticle-modified SPGE (AuNPs-GSPE) via Au-SH bonding. After the formation of a self-assembled monolayer (SAM) using 6-mercapto-1-hexanol (MCH), non-specific binding sites were blocked with bovine serum albumin (BSA).

Figure 3.: Schematic representation of the affisensor for HER2 detection (Reproduced from [84] with copyright permission)

Although the use of BSA may form a thicker layer that can interfere with signal transmission, it effectively reduces non-specific protein adsorption by distancing inactive electrode areas from the sensitive region. Subsequently, affinity interaction with HER2 was evaluated using EIS. The developed biosensor exhibited a linear response range of 0 to 40 μg L-1 for HER2, with an LOD of 6 μg L-1. A good analytical response was also observed in serum samples spiked with HER2 protein. The kinetic and thermodynamic parameters of the affibody-HER2 affinity interaction were further analysed by SPR.

In comparison with the conventional HER2 detection methods typically based on antibodies, the impedimetric affisensor showed a fast, sensitive and specific detection of HER2 in real samples with a very low LOD (6.0 μg L-1), yielding promising results for the use of newly-engineered proteins in biosensor technology for clinical applications. Furthermore, the instrumentation of this biosensor has been miniaturized to pocket-sized dimensions, making it ideal for use in POC devices. The developed affisensor is label-free and designed for single-use, simplifying the detection process and preventing cross-contamination. However, this study also has limitations, namely the lack of direct comparison with standard clinical methods (ELISA/CLIA) on real patient samples to assess clinical relevance. Despite the simplicity of the label-free system, its use can make the signal highly dependent on the surface conditions of the sensor and sample, compared to labelled methods, whose signals are “stronger”.

On the other hand, Baydemir et al. [69] developed a detection method for tumour necrosis factor-alpha (TNF-α) using affibody bioreceptors immobilized on magnetic beads (MBs), which served as capture agents in a sandwich assay configuration. Specific antibodies and secondary antibodies conjugated with alkaline phosphatase were employed, as illustrated in Figure 4.

Figure 4.: Electrochemical test scheme based on magnetic beads and disposable screen-printed sensors (Reproduced from [69] with copyright permission)

The performance of this system was evaluated on standard solutions and human serum samples supplemented with TNF-α by DPV. A LOD of 0.038 ng mL-1 was obtained with a quantification range of 0.076 to 5.000 ng mL-1. Meanwhile, the measurement reproducibility, with a relative standard deviation of 7 %, indicated good performance. Unlike the study by Ravalli et al. [83], which employed a label-free system, this research utilized alkaline phosphatase (AP) to enhance the detection signal. MBs were also employed to facilitate both bioreceptor immobilization and immunocomplex separation. In addition, a comparison with a commercial ELISA method was performed to demonstrate the clinical relevance of the developed assay. The obtained Pearson correlation coefficient (r = 0.979) indicated that the method exhibited high accuracy and reliability in detecting TNF-α. The sandwich assay system used in this study also provided high sensitivity (LOD of 0.038 ng mL-1), making this study superior in sensitivity compared to other related studies.

On the other hand, although affibodies were used as bioreceptors, the antigen-capture step for TNF-α still relied on conventional antibodies, so the full advantages of using affibodies were not yet fully realized. Moreover, the enzyme label used in the system is chemically reactive, which may limit the assay's overall stability and practical applicability under real-world or field conditions.

Given the similarity of the test used, detection method and supporting elements, Ilkhani et al. [84] used the HER2 biomarker for breast cancer detection. They used a double-affibody (Af/Af) sandwich assay, in which the HER2 protein is captured between a primary bioreceptor immobilized on MBs (Biot-Af (Strept--MB/Af)) and a labelled secondary bioreceptor (Biot-Af). Detection was carried out by enzyme amplification by combining a streptavidin-alkaline phosphatase conjugate on Biot-Af, which converts the electro-inactive substrate into an electro-active product, as shown in Figure 5. The signal measured using DPV showed a LOD of 1.8 ng mL-1and a linear range of 0 to 20 ng mL-1. These results indicate that the Baydemir et al. [69] method is superior in sensitivity, whereas the approach of Ilkhani et al. [85] offers a broader range of target concentrations. Furthermore, the second configuration, the antibody/affibody sandwich assay (Ab/Af), utilizes an antibody (Ab1) immobilized on protein A-modified magnetic beads (ProtA-MBs). This method yields an LOD of 3.4 ng mL-1, an average RSD of 11 %, and maintains a linear response over the concentration range of 0 to 20 ng mL-1. The third configuration is the affibody/antibody sandwich assay (Af/Ab), which uses a biotinylated affibody as the capture receptor on Strept-MBs and a biotinylated antibody (Biot-Ab2) as the signal receptor. This assay has an LOD of 2.6 ng mL-1, an average RSD of 10 %, and a linear response in the range of 0 to 20 ng mL-1. Overall, the double-affibody assay (Af/Af) proved superior in sensitivity and reproducibility compared with the other configurations.

Figure 5.: Schematic illustration of the streptavidin-modified MB-based affibody assay for HER2 detection (Reproduced from [85] with copyright permission)

Among other studies, Ilkhani et al. [85] is the most relevant for HER2 (0-20 ng mL-1). This confirms the high potential of this method for real-world applications in the detection of HER2 cancer biomarkers. Furthermore, the affibody-only sandwich configuration offers higher selectivity and sensitivity than sandwiches incorporating antibodies. However, a limitation of this study is the lack of data on the long-term stability of the biosensor. This study also reported that affibody-modified magnetic beads could be stored at 4°C for approximately one week, without further testing for long-term bioreceptor durability.

Optical biosensors

Pham et al. [86] explored the use of anti-HER2 affibody molecules conjugated with fluorescein isothiocyanate (FITC) for detecting the HER2 biomarker on HER2-positive extracellular vesicles (EVs) and compared their performance with conventional antibodies using the fluorescence polarization (FP) method.Anti-HER2 affibody molecules utilized in this work are head-to-tail affibody dimers linked through the peptide backbone for a higher binding affinity. In addition, the three anti-HER2 antibodies were derived from three different clones, namely 2G11 and 24D2, which were conjugated with FITC.

In this study, EVs were immobilized on MBs via antibody capture and subsequently detected with fluorescently labelled ligands. The results showed that FITC-bound dimeric anti-HER2 affibody could detect SKBR3 EVs and HT-29 EVs at low concentrations of 8.1×10⁶ and 7.0×10⁶ EV/mL, respectively. This LOD is much lower than the LOD of ELISA for EVs and many other published assays [87,88]. As shown in Figure 6, the two FITC-labelled anti-HER2 antibody clones did not display significant differences in binding to EVs from SKBR3 or HT-29 cells. In contrast, the dimeric anti-HER2 affibody demonstrated significantly superior HER2 receptor access, characterized by approximately 4.4-fold higher fluorescence signal on SKBR3 EVs and 3.8-fold higher on HT-29 EVs compared to conventional antibodies. Due to its small size (~6.5 kDa), the Affibody molecule may outperform monoclonal antibodies in accessing and detecting its targets on EVs.

Figure 6.: Superiority of FITC-conjugated dimeric anti-HER2 Affibody molecules over FITC-conjugated anti-HER2 antibodies in binding to SKBR3 EVs (Reproduced from [86] with copyright permission)

On the other hand, the study by Sayyadi et al. [89] developed a simpler detection system that does not require two separate binding steps, unlike Pham et al. [86]. They directly used affibody-functionalized beads (affiBeads) as a novel strategy for high-sensitivity cancer exosome detection, using anti-EGFR affiBeads as bioreceptors. Detection was performed using a fluorescent detector with FITC labelling, which enhanced both the sensitivity and quantification of exosome detection.

During the fabrication stage, carboxyl-functionalized polystyrene microbeads were modified with anti-EGFR affibodies. The performance of the affiBeads in detecting exosomes via flow cytometry is shown in Figure 7a. A linear relationship was observed between the number of captured exosomes and the resulting fluorescence signal. Remarkably, even at very low concentrations (15.6 ng mL-1, approximately 12 exosomes per microbead), the affiBeads could still generate a significant signal.

Figure 7.: (A) cytometry histogram of affiBeads (Reproduced from [89] with copyright permission) (B) Biopanning rounds of phage ELISA (Reproduced from [90] with copyright permission)

Unlike commonly used conventional ELISA methods, Liu et al. [90] developed an affibody-based ELISA method. The affibodies used were synthesized using phage display technology. These affibody bioreceptors were used as capture reagents along with polyclonal antibodies to detect and quantify human serum proteins. This strategy was based on the affibody’s ability to bind to the solid phase at a higher molar density than antibodies due to its smaller molecular size.

After three rounds of biopanning (stepwise selection), the phage’s ability to bind AFP increased significantly, with the phage ELISA signal rising tenfold, as shown in Figure 7b. From the selected phage clones, ZAFP D2 was identified as an affibody with high specificity and strong affinity for AFP. To further enhance affinity, the researchers constructed a dimer form (ZAFP D₂)₂ by linking two ZAFP D₂ units using a G4SG4S linker.

The resulting dimer interacted with AFP twice as strongly as the monomer. The affibody covers specific regions of AFP, indicating shape complementarity that allows specific interactions. In this assay, ZAFP D2 or (ZAFP D₂)₂ served as the capture reagent binding to one side of AFP, while a polyclonal anti-AFP antibody acted as the detection reagent binding to the opposite side. An HRP-conjugated secondary antibody then produced the detection signal. This (ZAFP D₂)₂-based method demonstrated more sensitive results with a LOD of 2 ng mL-1 and a wide linear range of 6 to 100 ng mL-1, as shown in Figure 8a. Furthermore, selectivity tests, as shown in Figure 8b, demonstrated that even in the presence of various potential interfering agents (CEA, BSA, EGF, glucose, glycine, folic acid, PBS) at high concentrations, AFP detection remained stable, demonstrating the specificity of the test. Sample recovery using human serum ranged from 90 to 112 %, with good precision.

Figure 8.: Results of the two-site ELISA method based on affibody-polyclonal antibody for AFP detection. (a) The linear range of the (ZAFP D2)₂ based two-site ELISA for the AFP concentration from 6 to 100 ng mL−1. (b) Selectivity of the (ZAFP D2) based two-site ELISA for AFP detection. (Reproduced from [91] with copyright permission)

Compared to other studies, Liu et al. [90] highlighted the stability of the dimeric affibody (ZAFP D₂)₂, which remained stable even when heated to 80 °C. Moreover, the dimer exhibited twice the interaction strength with AFP compared to its monomeric form. The ELISA LOD based on (ZAFP D₂)₂ was four times lower than that of the monomeric ZAFP D2. In healthy adults, normal serum AFP levels are typically below 25 ng mL-1, whereas in patients with liver cancer, they can exceed 400 ng mL-1. Therefore, this LOD is well below the diagnostic threshold (10 ng mL-1), making this method suitable for detecting AFP in real serum samples.

Meanwhile, Zhang et al. [91] focused on detecting CEA, a biomarker for lung cancer. They covalently conjugated affibodies with PEG-functionalized gold nanoparticles to create a nanogold-affinity peptide probe as the biological recognition element. This conjugation provided more controlled orientation, better colloidal stability and more consistent performance compared to antibody-based systems.

The detection mechanism of this device is shown in Figure 9a and follows a sandwich immunoassay format, in which the affibody and specific antibody work synergistically to capture the target antigen. The test strip membrane was coated with a CEA-specific antibody to capture the probe-antigen complex. The test results were determined by the colorimetric signal on the test line, allowing quick and easy interpretation. The system achieved a low LOD of 2.5 ng mL-1 and a diagnostic accuracy of 91.7 % in clinical serum samples.

Figure 9.: (A) Detection principle of the affibody nanogold probe test strip (Reproduced by [91]) (B) Illustration of the interaction between HRP-conjugated affibodies and immobilized EVs in microplate wells via CD9, CD63 and CD8 (Reproduced from [86] with copyright 2024 permission)

In contrast, Pham et al. [86] used the enzymatic reaction between HRP and the TMB substrate to produce a colour change. HER2-positive EVs from SKBR3 cells were immobilized on streptavidin-coated wells using biotinylated capture antibodies (anti-CD9, CD63 and CD81), yielding an optical density (OD₄₅₀ nm) signal up to 10 times higher than antibody-based systems, as illustrated in Figure 9b.

This study also produced consistent results in human plasma samples, in which affibodies demonstrated high sensitivity with a LOD of 2.1×109 EV mL-1, with EVs referring to extracellular vesicles. However, the HRP signal amplification system was less stable than inorganic nanoparticle labels, such as the colloidal gold system reported by Zhang et al. [91]. Various developments in cancer biomarker detection using affibodies are shown in Table 3.

Affibody | Bio-marker | Cancer type | Sensing layer | Signalling probe | Detection method | LOD, ng mL-1 | Linear range, ng mL-1 | Sample | Ref.
Anti TNF-α Affibody® | TNF-α | General | SPCEs functionalized HOOC-MBs | Alkaline phosphatase | DPV | 0.038 | 0.076 to 5 | Spiked serum | [69]
Biot-Af (Strept-MB/Af)+Biot-Af | HER2 | Breast | Functionalized with strep-MBs | S-AP | DPV | 1.8 | 0 to 20 | Spiked serum | [85]
Ab1 (ProtA-MB/Ab)+Biot-Af | 3.4
Biot-Af+Biot-Ab2 | 2.6
Anti-HER2 affibody | HER2 | Breast | AuNP-GSPE | Label-free | EIS | 6.0 | 0 to 40 | Human serum | [84]
Anti-EGFR-affiBeads | EGFR | Lung | Functionalized with polystyrene carboxylate microbeads | FITC | Fluorescence (optic) | 15.6 | - | Diluted exosome | [89]
Nanogold-affibody | CEA | Lung | Immobilized CEA-specific antibody | AuNPs | Colorimetric | 2.5 | 0 to 200 | Human serum | [91]
(ZAFP D2)2 | AFP | Liver | Two-site ELISA | HRP | Optic | 2.0 | 6 to 100 | Human serum | [90]
Affibody anti-HER2 | HER2 | Breast | Functionalized with biotinylated anti-CD9/CD63/CD81 antibodies | HRP | Colorimetric | 2.1×109* | 108 to 1010* | Human serum | [86]
Functionalized with biotinylated anti-CD9/CD81 antibodies | FITC | Fluorescence polarization | 8.1×106* | 108 to 10* | Human serum