The World Health Organization's (WHO) Global Tuberculosis Report for 2024 shows that the global tuberculosis epidemic remains severe, and the incidence and mortality rates still face great challenges. In 2023, the number of new tuberculosis (TB) patients worldwide was 10.8 million, and a total of 1.25 million people died of TB. Three years after the COVID-19 pandemic, TB may once again become the leading cause of death caused by a single infectious pathogen worldwide [1]. TB is an infectious disease caused by Mycobacterium tuberculosis (Mtb), which can be spread through the air when patients with an active infection cough, sneeze or spit. After infection, the bacteria primarily affect the lungs and then other organs of the body, such as the bones, brain, and spine [2]. Although several antibiotics have been developed to significantly relieve the TB epidemics, there are still several important scientific issues regarding TB. As one of the most critical issues for the control of TB and drug-resistant TB epidemics, the diagnosis of TB is critical for the timely therapy of TB [3].

Traditionally, the diagnosis of TB has relied on sputum smear microscopy or culture methods, which are time-consuming (two to three weeks for positive results and up to six weeks for negative results) but cost-effective due to the special growth characteristics of Mtb [[4], [5], [6]]. And it shows restricted sensitivity when the bacterial load of Mtb is lower than 10,000 organisms/mL in sputum samples [3,7,8], which is not conducive to the timely diagnosis and treatment of TB. Tuberculin skin test (TST) and interferon-γ release assay analysis (IGRA) are also commonly used to identify people who may have been infected with M. tuberculosis. IGRA is used as a surrogate test to screen larger populations at risk for tuberculosis depending on risk factors, and preferably not in the elderly [9]. TST is also used for screening larger populations [10]. Both IGRA and TST are not used as a primary test to screen for active infection [11,12]. TST is inexpensive and widely used, but it takes 48-72 hours to obtain results, and the results are easily interfered with by some non-tuberculosis mycobacteria. IGRA has high sensitivity, but it decreases significantly in immunocompromised individuals. Its specificity is also very high, unaffected by Bacillus Calmette–Guérin (BCG) vaccination, but interfered with by infection with certain nontuberculous mycobacteria [11,13]. High cost and the requirement for a special laboratory environment and trained staff also partially limit the application of IGRA [2,3,8,14].

WHO recommends the use of rapid molecular methods such as Xpert Mtb/ Rifampicin (RIF) assay, line probe assay, and loop-mediated isothermal amplification (LMAP), which can diagnose Mtb very effectively within 2 hours [1,3,15,16]. The Xpert Mtb/RIF test is a molecular diagnostic method based on a semi-nested real-time polymerase chain reaction (PCR) that determines the presence of tuberculosis infection by detecting specific gene sequences in Mtb DNA and simultaneously detects drug resistance to RIF, a cornerstone drug in Mtb treatment [17]. Line-probe assays (LPAs) amplifies Mtb DNA using PCR, then applies the output to oligonucleotides immobilized on a strip for colorimetric detection. Labs usually first detect Mtb and then look for resistance using varying methodologies depending on the laboratory. LPAs can also identify fluoroquinolone and aminoglycoside resistance [18]. LAMP directly amplifies Mtb DNA at a constant temperature, without the need for thermal cycling. This simplifies the procedure and shortens the detection time, but its sensitivity in smear-negative samples is questionable [4,19].

However, molecular diagnostic methods are expensive and require advanced instrumentation with special centers and expert technicians [20]. The use of rapid molecular tests is still very limited, especially in resource-limited areas, low-income countries, or special populations [21]. Only 48 % of reported cases in 2023 received WHO-recommended rapid molecular diagnostic tests at the time of initial diagnosis [1]. Therefore, it is urgent to develop new rapid, cost-effective, and sensitive diagnostic tests for the detection of Mtb through rigorous evaluation of potential biomarkers and new detection methods.

Unlike traditional laboratory-based detection methods, point-of-care (POC) tests offer a potential solution for TB diagnosis by enabling immediate on-site TB diagnosis, even in resource-limited settings [22]. Biosensors targeting Mtb-related biological elements are capable of providing high specificity, increased sensitivity, and direct and on-site detection of Mtb [2]. The development of electrochemical biosensors based on nanomaterials has attracted significant attention for potential applications in medical diagnostics, food safety, and environmental monitoring [23]. Among different nanomaterials, three-dimensional (3D) nanostructures are more suitable for making biosensors because they have a higher surface area than their corresponding planar electrodes [24]. The high active surface area of nanostructures can allow for the immobilization of large numbers of antibody molecules within the biosensor, thereby enhancing sensitivity [25].

Paper-based electrochemical sensors are a novel electrochemical sensing technology that combines the inherent advantages of paper substrates and electrochemical detection to create simple, economical, portable, and disposable analytical devices [26,27]. Paper is considered a "green" material that is well-suited for infectious disease detection due to its low cost, natural abundance, lightness, and safe disposal through incineration after use [28,29]. Our research team previously proposed a paper-based electrochemical biosensor modified with electrospun cellulose acetate (CA) nanofibers (NFs), which performed well in the detection of glucose, Ag85B protein, and Escherichia coli O157:H7 [30]. 3D NFs have the characteristics of large specific surface area, good biocompatibility, high porosity, and good hydrophilicity, which can provide more binding sites for biorecognition elements and thus enhance the detection ability of the sensor [31]. The TB antigen Ag85B is the most abundant secretory protein of Mtb, and can be detected in blood, urine, sputum, or cerebrospinal fluid [32]. Ma et al. compared the specificity of different targets and recognition elements of Mtb and found that the Ag85B protein achieved good detection results and has high diagnostic potential in TB diagnosis [33]. Therefore, antigen Ag85B can be selected as a specific recognition substance of Mtb for the diagnosis of TB. Previous work has also demonstrated the potential of Ag85B protein as a biomarker in the diagnosis of TB [30,32], but its clinical application has not yet been verified.

Here, this study is based on a 3D NFs paper-based electrochemical biosensor, by immobilizing Ag85B antibody on the surface of the 3D NFs electrode, to specifically recognize Mtb secretory protein Ag85B. Clinical trials have verified its feasibility in POC detection, providing an alternative method for the rapid diagnosis of TB.