Human serum amyloid A (SAA) is a liver-produced acute-phase protein associated with the pathogenesis of various inflammatory diseases(Sack Jr., 2018). In healthy individuals, the SAA blood concentration is approximately 1–10 μg/mL. However, it can rapidly increase up to 1000-fold within the initial 24 h after the onset of acute phase response(De Buck et al., 2016; Zhang et al., 2019). Recently, SAA concentrations were found to be significantly and positively associated with increased coronavirus disease (COVID-19) severity and mortality (Zinellu et al., 2021). SAA may be considered as a promising biomarker for predicting the severity and prognosis of COVID-19.

Rapid and accurate detection of SAA is critical for disease prevention (Yuan et al., 2019). A variety of methods are available for the detection of SAA, including immunofluorescence, enzyme-linked immunofluorescence, chemiluminescence, and electrochemiluminescence assays. Although these methods provide accurate and quantitative results, they also have disadvantages such as complex protocols, time-consuming, and the need for specialized equipment and personnel. In resource-limited medical settings, there is a limited availability of methods that are efficient and effective for SAA detection. Therefore, there is an urgent need to develop a user-friendly, rapid, accurate, and point-of-care testing (POCT) method for the detection of SAA that can be adapted to such testing scenarios.

The lateral flow immunoassay (LFIA) for SAA is probably the most rapid and cost-effective, requiring no or minimal equipment. In low-resource settings, qualitative results can be obtained with only the naked eye. Several LFIA methods have been developed for the quantification of SAA. Wang et al. reported a LFIA based on a biotin-streptavidin-phycoerythrin signal amplification system for the quantification of SAA and two other inflammatory biomarkers. SAA was quantified in the range of 3–200 μg/mL(Wang et al., 2022). Liu et al. developed a surface-enhanced Raman scattering (SERS)-based LFIA for the quantification of SAA(Liu et al., 2020). The working range of the SAA calibration curve is 0.1–500 ng/mL. Song et al. developed a fluorescence LFIA using up-conversion nanoparticles as fluorescent labels to detect SAA(Song et al., 2021). The quantification range is from 0.1 to 50 ng/mL.

These LFIA methods for the quantification of SAA appear to be reliable and highly sensitive. However, serum SAA concentrations can rise rapidly to 1–10 mg/mL or higher within the initial 24 h after the onset of acute phase response. For example, SAA concentrations may be slightly elevated (10-100 μg/mL) in viral infections, but moderately to significantly elevated (10–1000 μg/mL) in bacterial and fungal infections(King et al., 2011; de Seny et al., 2013; Thompson et al., 2015; Vietri et al., 2020). Difficulties arise when attempting to quantify SAA at high concentrations, as a hook effect may occur, resulting in an artificially low or negative value. On the other hand, multiple manual dilutions must be performed if an LFIA with a quantification range of ng/mL is used to quantify SAA concentrations at mg/mL concentration. Such dilutions increase reagent and labor costs and may decrease the reliability of the assay. Detection of SAA by LFIA is typically performed immediately at the site of sample collection, where access to specialized quantitative equipment may be limited. Recently, methods for equipment-free semi-quantification have been developed using LFIAs with a multiple test line mode(Leung et al., 2008; Oh et al., 2014; Hu et al., 2017; Gasperino et al., 2018; Serebrennikova et al., 2019). In this mode, increasing the analyte concentration results in the appearance of more test lines due to the saturation of the preceding test lines. Semi-quantitative results can be obtained based on the number of test lines present. This type of analysis reports a concentration range rather than an exact value. However, it remains a challenge to obtain more accurate quantitative results without specialized equipment.

Smartphones with high-resolution rear cameras have emerged as a promising platform for LFIA readout in recent years. Without the need for specialized equipment, the portability and accessibility of smartphones make them a convenient tool for quantitative LFIA analysis. By capturing images of the test strip with a smartphone, software algorithms can analyze the color intensity of the test and provide an objective, quantitative result. Smartphone-based LFIA readout has great potential, especially in low-resource settings with limited access to laboratory and professional equipment. This method can improve diagnostics in these settings by providing accurate, reliable results at a lower cost.

In this study, we present a gradient LFIA test strips for the detection of SAA. This gradient LFIA test strip has three test lines (T-lines) designed to detect SAA at high (>100 μg/mL), inter-mediate (50–100 μg/mL), and low (10–50 μg/mL) concentrations, allowing qualitative, semi-quantitative, and quantitative assessment of SAA concentrations. The gradient LFIA test strip has three readout methods, each with varying concentrations of sensitivity and accuracy. The first method involves visual inspection of the number of T-lines on the strip, providing a rapid and semi-quantitative analysis without any specialized equipment or software. The second method employs a smartphone equipped with a specialized immunoassay app created by our team, which provides a more precise and quantitative analysis of the SAA concentrations. The third method uses a flatbed scanner and ImageJ software, providing the most accurate and precise quantification of SAA concentrations. In addition, the LFIA test strip developed in this study has a significantly broader detection range (5–1000 μg/mL) than conventional LFIAs and effectively reduces the hook effect.