Carbon-based biosensor, preparation and noninvasive detection method in PD-1

Abstract

The invention relates to a carbon-based biosensor, a preparation method thereof and a non-invasive detection method in PD-1, and solves the problems of direct fixation of an antibody and rapid detection of human-mouse source PD-1. And the channel material terminating electrode forms low ohmic contact which is directly used as a sensing interface. The preparation comprises design; performing cleaning; preparing an electrode; and depositing the s-SWCNTs to finish the preparation. In the detection, quantitative detection of PD-1 is realized mainly by fixing a PD-1 monoclonal antibody site and specifically binding with PD-1 protein. According to the invention, the CNT-FET is used for non-invasive and rapid real-time detection of human and mouse source PD-1 for the first time, and s-SWCNTs ends are connected to the upper surfaces of the source electrode and the drain electrode; the preparation process of the first electrode and the second CNT directly fixes the antibody through electrostatic adsorption without modification, so that the activity of the antibody is ensured. Sensitivity is high, process is simple and cost is low; the kit is used for PD-1 detection in NSCLC immunotherapy, and can also be used for multi-index synchronous detection and bedside detection.

Description

Technical Field

[0001] The present invention belongs to the field of electronic technology and mainly relates to a biosensor and its application in the rapid, non-invasive, real-time detection of PD-1 in non-small cell lung cancer (NSCLC). Specifically, it relates to a carbon-based biosensor and its preparation and non-invasive detection method in PD-1. The carbon-based biosensor is applied to the detection of PD-1 content in NSCLC immunotherapy and can also be used for multi-index detection and the development of handheld portable detection equipment. Background Art

[0002] Carbon-based biosensors, also known as carbon nanotube field-effect transistors (CNT-FETs), have demonstrated advantages in the field of biosensing due to their ultrahigh carrier mobility and remarkable surface charge sensitivity. The one-dimensional electronic structure of single-walled carbon nanotubes (SWCNTs) enables quantum confinement, enabling direct conversion of biomolecule binding events into detectable conductance changes, enhancing sensitivity. The nanoscale channel dimensions of CNT-FETs provide an extremely high surface-to-volume ratio, improving biomolecule capture efficiency. Their direct electron transduction mechanism avoids interference from labeling steps in traditional optical detection. In theory, SWCNTs can achieve femtomolar or even single-molecule detection limits. Appropriately surface-modified CNT-FETs maintain excellent performance in complex matrices such as serum. For example, CNT-FETs eliminate the need for expensive enzyme-labeled antibodies and chromogenic substrates, resulting in lower costs compared to enzyme-linked immunosorbent assays (ELISAs). Furthermore, their label-free direct electron detection mechanism eliminates complex washing and incubation steps, streamlining the process. While ELISAs typically require approximately 4-6 hours, CNT-FETs require much shorter detection times, better meeting clinical point-of-care testing needs.

[0003] In recent years, biosensors based on advanced sensing technologies, such as CNT-FETs, have demonstrated promising applications in disease diagnosis and therapeutic efficacy monitoring. Their high sensitivity, rapid response, and miniaturization make them particularly useful in detecting biomarkers during cancer immunotherapy. These biosensors can monitor dynamic changes in key biomarker levels in real time, providing important empirical evidence for clinical decision-making. While CNT-FET biosensors have been studied for biomarker detection in some diseases, there are no reports on their use for PD-1 detection. Furthermore, other CNT-FET assays still face challenges, such as insufficient sensitivity, complex procedures, and poor cost-effectiveness.

[0004] In the process of using CNT-FET to detect biomarkers, Professor Zhang Zhiyong's team at Peking University spin-coated diluted polylysine (PLL) on the CNT-FET. Taking advantage of the reducing property of the large number of amine groups contained in PLL, gold nanoparticles (Au NPs) were reduced and deposited in situ. The thiol aptamer probe was then mixed with tris(2-carboxyethyl)phosphine (TCEP) to activate the thiol group. The aptamer solution was added dropwise to the CNT-FET and incubated overnight to allow the aptamer to bind to the AuNPs surface through the Au-S bond. After repeated rinsing with PBS and ultrapure water, the antibody was blocked with BSA for 1 hour to block unbound active sites, completing the fixation of the antibody on the CNT-FET. This method utilizes the sensing interface of CNT-FET and distinguishes samples from healthy individuals and breast cancer patients by comparing the expression levels of MUC1. In this detection method, the aptamer-modified CNT FET biosensor can identify breast cancer patients and healthy individuals by comparing the expression levels of exosomal MUC1, and is expected to become a new method for early diagnosis of breast cancer. However, the PLL functionalization and Au NPs modification processes in the antibody fixation process are very cumbersome. In particular, the antibody fixation process undergoes multiple complex steps - including PLL spin coating, in situ reduction of Au NPs, aptamer activation (TCEP treatment), overnight incubation and BSA blocking. The entire process takes more than 12 hours, and there are many operational links. Some operations are extremely cumbersome, and the slightest carelessness will affect the detection results.

[0005] The team also developed a floating gate structure sensor based on randomly arranged high-purity semiconducting single-walled carbon nanotubes (s-SWCNTs). A yttrium oxide dielectric layer was grown as a floating gate by atomic layer deposition technology. The antibody was fixed on the device surface through a series of steps including fixing Au NPs on the yttrium oxide surface, activating thiol groups, fixing antibodies through Au-S bonds, PBS washing and BSA blocking. The floating gate structure sensor was applied to the detection of biomarkers such as lung cancer marker CEA, porcine epidemic diarrhea virus (PEDV) surface antigen and SARS-CoV-2 spike protein. The floating gate structure CNT-FET sensor still has problems such as complex preparation process and high cost, which is not conducive to clinical promotion. The design of the yttrium oxide floating gate completely wrapped in carbon nanotubes limits its direct interaction with biological molecules, which may affect the detection sensitivity. In addition, when detecting CEA, Zhang Zhiyong's team used fetal bovine serum (FBS) to simulate a complex biological environment and verify the detection performance of the sensor in serum, and did not directly use human clinical samples; when detecting biomarkers such as swine fever virus and antibiotic residues in milk, the performance of the sensor in actual clinical samples (such as blood) was not tested. The sample detection range is limited, and expensive equipment and complex operating procedures are required, which limits its application in resource-limited areas and makes it difficult to promote it in primary medical institutions or on-site testing.

[0006] A research team from King Abdulaziz University in Saudi Arabia developed a CNT-FET-based biosensor. Ti / Au source and drain electrodes were fabricated on a Si / SiO2 substrate using photolithography and lift-off techniques. A conductive channel was formed by inkjet-printed s-SWCNT networks. The electrodes were passivated with SU-8 photoresist, and the device surface was functionalized by immersing in a methanol solution of PBASE for one hour. During the detection process, an anti-SARS-CoV-2 S1 monoclonal antibody was added dropwise to the PBASE-modified channel region for antibody immobilization. The antibody was incubated overnight at 4°C to bind via amide bonds. The device was then rinsed with PBST and deionized water, and nonspecific sites were blocked with 1% BSA before protein detection. This method simplifies the antibody immobilization process compared to Zhang Zhiyong's team, reducing costs and enabling rapid detection of the SARS-CoV-2 surface spike protein S1. However, this study only used artificially spiked saliva as the test sample, and its anti-interference ability in real human samples has not been verified. Its applicability to complex biological matrices requires further investigation.

[0007] Currently, PD-1 is a key biomarker for evaluating the efficacy of immunotherapy for non-small cell lung cancer. Medical detection technologies such as ELISA include multiple steps: coating → blocking → sample addition → incubation → washing → addition of enzyme-labeled antibody → incubation → washing → addition of substrate for color development → termination of reaction → measurement of OD value. The entire process takes 4-6 hours and cannot meet the needs of clinical rapid detection. It also relies on expensive biochemical reagents such as enzyme-labeled antibodies and color development substrates, and the cost of a single test is high. Its sensitivity is limited by the efficiency of the enzymatic reaction, and the detection limit is usually only at the pg / mL level. It also requires a professional laboratory environment and large-scale microplate reader equipment, making it difficult to achieve real-time bedside detection and dynamic monitoring. Cross-reactions are prone to occur when multiple indicators are tested together, seriously affecting the specificity of the test.

[0008] In summary, the existing ELISA technology for detecting PD-1 has problems such as cumbersome procedures, harsh conditions, and long time consumption, making it difficult to meet the needs of real-time clinical monitoring. Many scientists are researching CNT-FET biosensors in the field of biomarker detection, but there are still problems such as loss of antibody activity, need for improved sensitivity, complex preparation, and high cost. Furthermore, there are no reports on the use of CNT-FET biosensors for detecting PD-1, let alone reports on the detection of both human and mouse PD-1. Summary of the Invention

[0009] The purpose of the present invention is to address the problems and shortcomings of the existing technology and provide a carbon-based biosensor with low-resistance ohmic contact between electrodes and s-SWCNTs, a simple preparation process, guaranteed antibody activity, and the first non-invasive detection of human and mouse PD-1 protein, which outputs the detection results in real time, as well as a preparation method and non-invasive detection method in PD-1.

[0010] The present invention is a carbon-based biosensor, comprising a silicon dioxide substrate, a source electrode, a drain electrode, and semiconductor single-walled carbon nanotubes (s-SWCNTs). A conductive channel is provided in the middle of the upper surface of the silicon dioxide substrate, with a source electrode and a drain electrode on either side. The source electrode and the drain electrode are both Ti / Au stacks, and the conductive channel material is s-SWCNTs. Ohmic contacts are formed between the s-SWCNTs and the source electrode and the drain electrode. The s-SWCNTs, as the conductive channel material, are terminated at the upper surfaces of the source electrode and the drain electrode, and form low-resistance ohmic contacts with the Ti / Au stacks of the source electrode and the drain electrode, respectively, forming a sensing interface of the biosensor as a whole. No oxide dielectric layer is provided on the upper surface of the device, and no covalent coupling modification is required, so the device directly serves as the sensing interface of the biosensor. The s-SWCNTs in the conductive channel on the biosensor sensing interface are either slightly higher than the upper surfaces of the source electrode and the drain electrode, or flush with the upper surfaces of the source electrode and the drain electrode.

[0011] The present invention also provides a method for preparing a carbon-based biosensor, which is characterized by comprising the following steps:

[0012] Step 1: Design a CNT-FET biosensor: comprising a silicon dioxide substrate, a source electrode, a drain electrode, and semiconducting single-walled carbon nanotubes (s-SWCNTs). A conductive channel is provided on the upper surface of the silicon dioxide substrate, with a source electrode and a drain electrode on either side. The source and drain electrodes are both Ti / Au stacks. The s-SWCNTs serve as the conductive channel material and are terminated on the upper surfaces of the source and drain electrodes, forming low-resistance ohmic contacts with the Ti / Au stacks of the source and drain electrodes, respectively, to form a sensing interface of the biosensor. No oxide dielectric layer is provided on the upper surface of the device, requiring no covalent coupling modification, and the device directly serves as the sensing interface of the biosensor. The s-SWCNTs in the conductive channel on the biosensor sensing interface are either slightly higher than the upper surfaces of the source and drain electrodes, or flush with the upper surfaces of the source and drain electrodes.

[0013] Step 2: Clean the silicon dioxide substrate: Use a p-type silicon dioxide substrate with a resistivity of 0-0.0015Ω·cm and a 100nm thermal oxide layer. <100> The silicon wafer was used as the substrate and ultrasonically cleaned with acetone, isopropyl alcohol and ultrapure water in sequence, each cleaning lasting 3 minutes, and then purged with nitrogen.

[0014] Step 3: Photolithographic patterning of the device on the upper surface of the substrate: First, design a device photolithographic pattern on the upper surface of the substrate. The photolithographic pattern is a conductive channel with a diameter of 10.0±0.2μm in the middle, and a device mask pattern with source and drain windows on either side of the conductive channel. The pattern shapes of the source and drain windows are designed to be two symmetrical squares or two symmetrical rectangles, and the source and drain are symmetrical about the center line of the conductive channel. Use a photolithographic process to prepare the designed device mask pattern on the upper surface of the device substrate to complete the patterning of the upper surface of the substrate.

[0015] Step 4: magnetron sputtering to prepare the Ti / Au stack for the source and drain electrodes: on the patterned upper surface of the substrate, a titanium Ti adhesion layer is blanket deposited using a magnetron sputtering process, and a gold Au conductive layer is further sputtered on the upper surface of the Ti layer. Then, a stripping process is used to selectively remove the photoresist and redundant metal covering outside the source and drain electrode windows, retaining the Ti / Au stack in the electrode window area, thereby completing the preparation of the Ti / Au stack electrodes for the source and drain electrodes on the upper surface of the substrate;

[0016] Step 5: Depositing s-SWCNTs to form a low-resistance ohmic contact between the channel and the electrodes: On the upper surface of the device where the source and drain electrodes have been stacked, high-purity s-SWCNTs are uniformly deposited on the entire upper surface of the device, including the Ti / Au electrode surface and the 10.0±0.2μm exposed conductive channel region in the middle, using a solution deposition process. The s-SWCNTs form a continuous, uniform, and large-area randomly arranged network-like thin film on the Ti / Au electrode surface and the conductive channel region in the middle; the s-SWCNTs in the conductive channel form low-resistance ohmic contacts with the source and drain electrodes;

[0017] Step 6: Remove s-SWCNTs outside the conductive channel to complete device fabrication: After the s-SWCNTs in the conductive channel form low-resistance ohmic contacts with the source and drain electrodes, an oxygen plasma etching process is used to selectively retain 10.0±0.2μm conductive channel s-SWCNTs and remove redundant s-SWCNTs outside the conductive channel area. Finally, ultrasonic cleaning is performed with acetone, isopropyl alcohol, and ultrapure water, respectively, for 3 minutes each time. After nitrogen purging, the device fabrication is completed.

[0018] The present invention also provides a non-invasive detection method for PD-1 protein using a carbon-based biosensor, which is characterized by comprising the following steps:

[0019] Step S1, initial electrical performance characterization: connect the unmodified CNT-FET biosensor to a 4200-SCS semiconductor characterization system and fix the drain-source voltage V ds =-1V, gate voltage V gsThe initial transfer characteristics test was performed under the condition of scanning range from -5V to +25V to ensure the good performance of CNT-FET biosensor and record the benchmark stable on-state current I0;

[0020] Step S2, immobilizing PD-1 monoclonal antibody sites: 20 μL of 0.1 mg / mL PD-1 monoclonal antibody was dripped onto the conductive channel surface composed of s-SWCNTs of the CNT-FET biosensor, and incubated at a constant temperature of 4°C for 6 hours to allow the PD-1 monoclonal antibody to be directly immobilized on the s-SWCNTs surface through non-covalent electrostatic interactions, forming PD-1 specific binding sites. The areas where the PD-1 monoclonal antibody was not immobilized are non-specific binding sites, i.e., redundant sites, thereby obtaining a CNT-FET biosensor with immobilized PD-1 monoclonal antibody sites.

[0021] Step S3, blocking nonspecific binding: blocking excess sites, immersing the CNT-FET biosensor with fixed PD-1 monoclonal antibody sites in 1 mg / mL bovine serum albumin (BSA) blocking solution, incubating at room temperature for 1 hour at 25°C, and rinsing with deionized water three times to remove unbound BSA molecules, effectively blocking nonspecific binding sites on the surface of the CNT-FET biosensor, and obtaining a CNT-FET biosensor after BSA blocking treatment;

[0022] Step S4, benchmark electrical characteristics test: Connect the BSA-blocked CNT-FET biosensor to a 4200-SCS semiconductor characteristics analysis system to obtain a transfer characteristic curve after antibody immobilization. The transfer characteristic curve of the CNT-FET biosensor is obtained by varying the gate voltage. The stable on-state current I1 of the source-drain of the CNT-FET biosensor at a specific operating voltage is recorded and determined as the benchmark current value I1 for PD-1 protein detection, thereby completing the benchmark electrical characteristics test.

[0023] Step S5, specific binding of PD-1 protein and antibody: 1 μL of the serum sample to be tested was diluted to a specific concentration and evenly dripped onto the surface of the CNT-FET biosensor that had completed the benchmark electrical test. The sample was placed in a constant temperature environment at 4°C and allowed to react for 1 hour to allow the PD-1 protein in the serum to specifically bind to the PD-1 monoclonal antibody immobilized on the s-SWCNTs surface. The sample was then rinsed three times with 1× PBS buffer to remove unbound serum proteins and other interfering substances. After drying with nitrogen, the specific binding of the PD-1 protein and the PD-1 monoclonal antibody was completed, and electrical properties testing was immediately performed.

[0024] Step S6, electrical property testing after specific binding of PD-1 protein to antibody: After site-specific binding of PD-1 protein to PD-1 monoclonal antibody, electrical property testing is immediately performed. The test conditions are the same as those in step S3. The electrical property testing after specific binding of PD-1 protein to antibody is performed using the same semiconductor property analysis system to obtain a transfer characteristic curve and a stable on-state current value I2. This completes the electrical property testing after specific binding of PD-1 protein to antibody.

[0025] Step S7, detection data analysis, providing the detection results of PD-1 protein: After obtaining the detection data of PD-1 protein by the CNT-FET biosensor in real time, the data is analyzed, including the following: comparing the sizes of I1 and I0, as well as I2 and I1, to characterize the antibody fixation efficiency and the concentration of PD-1, respectively; drawing and obtaining standard curves for different PD-1 concentrations, and providing the detection results of PD-1 protein.

[0026] The present invention solves technical problems such as the loss of antibody activity and decreased sensitivity caused by complex chemical modification and covalent coupling of fixed antibodies in CNT-FET biosensors, high contact resistance caused by the "first CNT deposition and then electrode preparation" process, high cost caused by device preparation and antibody fixation, and poor detection performance in complex human serum environments.

[0027] Compared with the prior art, the technical advantages of the present invention are as follows:

[0028] Improving the sensitivity of carbon-based biosensors: The CNT-FET biosensor of this invention achieves low-resistance ohmic contact between the source and drain electrodes and the s-SWCNTs in the conductive channel through a modified fabrication process. The device also lacks an oxide dielectric layer on its top surface, and does not require any additional modification or protective layers. This eliminates the need for covalent coupling and modification during antibody immobilization, minimizing the biological activity of the antibody and allowing it to serve directly as the biosensor's sensing interface. These two factors work together to successfully improve the device's sensitivity.

[0029] Simplified fabrication process: Existing CNT-FETs often use an yttrium oxide floating gate structure. This is complex to manufacture and requires high cost and fabrication process requirements, significantly increasing fabrication costs and limiting clinical applications. This invention eliminates the use of an oxide floating gate structure as a dielectric layer on s-SWCNTs, significantly reducing fabrication costs. The simplified process facilitates large-scale and clinical application.

[0030] The first non-invasive and rapid detection of PD-1 using CNT-FET biosensor: This invention uses CNT-FET biosensor to detect PD-1 for the first time, realizing non-invasive, rapid, and trace detection, effectively avoiding the trauma of tissue biopsy, and shortening the detection time to 30 minutes. Only 1-5μL of serum is required, making non-invasive, immediate, and high-frequency clinical monitoring and small animal experiments a reality, providing a revolutionary solution for the dynamic evaluation of immunotherapy.

[0031] Cross-species detection is achieved, and it has the advantages of low misdiagnosis rate and real-time detection: the present invention uses direct electrostatic adsorption to fix monoclonal antibodies to effectively inhibit the nonspecific adsorption of substances other than PD-1 in serum, thereby significantly reducing the risk of misdiagnosis due to sample factors; the present invention directly causes changes in channel current through the specific binding of PD-1 to the monoclonal antibody on the surface of the CNT-FET biosensor, which can achieve real-time bedside detection and data output.

[0032] It has prospects for clinical transformation and multi-index detection applications: the preparation process of the present invention has low cost and simple procedures, and can achieve second-level response and continuous data output, laying the foundation for the development of handheld portable devices and alleviating the site and cost limitations of clinical promotion; at the same time, the miniaturized multi-channel integrated design of the CNT-FET sensor of the present invention is expected to simultaneously monitor immune checkpoints such as PD-1 / PD-L1, CTLA-4, and LAG-3, which can not only accurately capture the treatment window period and drug resistance mechanism, but also provide an efficient technical platform for personalized drug screening. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 : is a schematic structural diagram of the CNT-FET biosensor of the present invention;

[0034] Figure 2 : is a schematic diagram of the top view of the source electrode and drain electrode of the CNT-FET of the present invention;

[0035] Figure 3 : It is a flow chart of the preparation process of the present invention;

[0036] Figure 4 : It is a schematic diagram of the preparation process of the present invention;

[0037] Figure 5 : is a flow chart of the detection process for PD-1 detection of the present invention;

[0038] Figure 6 : is a schematic diagram of the binding of PD-1 protein and antibody on the sensing surface in the detection method of the present invention;

[0039] Figure 7 : is a transfer characteristic curve diagram of the present invention for detecting mouse serum;

[0040] Figure 8 : is a transfer characteristic curve diagram of the present invention for detecting different concentrations of mouse serum;

[0041] Figure 9 : is a transfer characteristic curve diagram of the present invention for detecting human serum;

[0042] Figure 10 : is a contact angle diagram of the CNT-FET before the antibody is immobilized;

[0043] Figure 11 : is a contact angle diagram of the CNT-FET after antibody immobilization of the present invention;

[0044] Figure 12 : This is a contact angle diagram of CNT-FET after PD-1 of the present invention is combined with PD-1 monoclonal antibody.

[0045] The technical solutions of the present invention are described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention. DETAILED DESCRIPTION

[0046] Example 1

[0047] With the rapid development of biosensor technology and nanomaterials science, carbon nanotube field-effect transistor (CNT-FET) biosensors have shown great application potential in disease diagnosis, drug screening, and biomolecule detection. Existing CNT-FET sensors still face key technical challenges such as complex preparation processes, severe loss of antibody activity, insufficient detection stability, and difficulties in clinical translation. To address these technical bottlenecks, the present invention proposes a carbon-based biosensor and a method for its preparation and non-invasive detection in PD-1, providing a new solution for non-invasive, rapid, and high-precision biomolecule detection.

[0048] The present invention firstly relates to a carbon-based biosensor, also known as a CNT-FET biosensor, hereinafter referred to as a biosensor or sensor. The carbon-based biosensor of the present invention comprises a silicon dioxide substrate, a source electrode, a drain electrode, and semiconducting single-walled carbon nanotubes (s-SWCNTs). A conductive channel is provided on the upper surface of the silicon dioxide substrate, with a source electrode and a drain electrode on either side. The source and drain electrodes are both Ti / Au stacks, and the conductive channel material is s-SWCNTs. In the prior art, ohmic contact also exists between the s-SWCNTs and the source and drain electrodes, but the contact resistance is relatively high. Figure 1 , Figure 1This is a schematic diagram of the structure of the CNT-FET biosensor of the present invention. In the present invention, s-SWCNTs 7 are used as the conductive channel material and are terminated on the upper surface of the source electrode 4 and the drain electrode 5. Figure 2 , Figure 2 It is a schematic diagram of the source electrode and drain electrode of the present invention. The s-SWCNTs 7 form a low-resistance ohmic contact with the Ti / Au stack of the source electrode 4 and the drain electrode 5, respectively, and the whole constitutes the sensing interface of the biosensor. The upper surface of the CNT-FET biosensor device of the present invention does not have an oxide dielectric layer, and no additional modification or any protective layer is added. It is directly the sensing interface of the carbon-based biosensor of the present invention. Different from the prior art that uses yttrium oxide as a floating gate dielectric layer and chemically modifies the yttrium oxide surface to achieve antibody fixation. When the CNT-FET biosensor device of the present invention is used to detect PD-1, no covalent coupling modification is required during the fixation of the PD-1 monoclonal antibody. The s-SWCNTs on the upper surface of the CNT-FET biosensor device of the present invention are in direct contact with the antibody, which improves the detection sensitivity and shortens the detection time. See Figure 2 The s-SWCNTs 7 in the conductive channel 6 on the sensing interface of the carbon-based biosensor of the present invention is either slightly higher than the upper surface jointly formed by the source electrode 4 and the drain electrode 5, or flush with the upper surfaces of the source electrode and the drain electrode. In other words, the conductive channel material is either slightly higher than the upper surface jointly formed by the source electrode 4 and the drain electrode 5, or flush with the upper surfaces of the source electrode and the drain electrode.

[0049] In existing technologies, because the antibody is ultimately fixed to the device's dielectric layer, it lacks direct contact with the s-SWCNTs' conductive channel. This blocks the charge interaction between the sensing interface and the target molecule, severely weakening the inherent high sensitivity of CNT-FETs. The present invention's sensing interface is directly exposed on the s-SWCNT surface, eliminating the need for traditional covalent modification steps such as silanization and carboxyl activation, and without the addition of any additional protective or dielectric layers. Instead, the antibody can be efficiently immobilized at 4°C through electrostatic interaction between the s-SWCNTs and the PD-1 antibody.

[0050] Example 2

[0051] The overall structure of the CNT-FET biosensor is the same as in Example 1. The conductive channel length of the present invention is set to 10 μm. This 10 μm channel length ensures sufficient sensing area while achieving superior carrier transfer efficiency. The shorter channel length effectively reduces signal attenuation, and the resulting compact molecular recognition space promotes the stability of biomolecular interactions, making it particularly suitable for rapid detection of samples of conventional concentrations.

[0052] The biosensor of the present invention is described in Figure 2In the Ti / Au stack of the source 4 and the Ti / Au stack of the drain 5, Ti is used as an adhesion layer 3 located below the Ti / Au stack with a thickness not exceeding 20 nm; Au is used as a conductive layer 2 located above the Ti / Au stack with a thickness of 50 nm to 100 nm. Figure 1 .

[0053] In this example, the Ti adhesion layer is no thicker than 20nm, forming a strong bond with the SiO2 substrate, ensuring the stability of the electrode structure while preventing interference with biomolecule detection signals caused by excessively thick metal layers. The Au conductive layer is between 50-100nm thick, ensuring excellent conductivity while achieving low contact resistance. This carefully designed metal stack structure not only ensures the device's electrical performance, but also creates an ideal interface environment for biomolecule immobilization and detection.

[0054] In this embodiment, the source electrode 4 and the drain electrode 5 adopt a Ti (5nm) / Au (50nm) stacked structure, and the source electrode 4 and the drain electrode 5 are both composed of a 5nm Ti adhesion layer 3 and a 50nm Au metal layer 2. The 5nm Ti adhesion layer 3 minimizes the interference of the metal layer on the detection of biological molecules while ensuring that the electrode is firmly bonded to the silicon dioxide substrate. The thickness of the 50nm Au metal layer 2 has been optimized to ensure good conductivity while avoiding the problem of increased surface roughness caused by an overly thick metal layer. This design is suitable for detecting low-concentration biomarkers. The thinner Ti layer reduces the risk of nonspecific adsorption, while the moderate Au layer thickness ensures a stable electrical signal output, showing excellent sensitivity and repeatability in applications such as serum PD-1 detection.

[0055] Example 3

[0056] The overall structure of the CNT-FET biosensor is the same as that of Example 1-2. The CNT-FET biosensor of this embodiment adopts a Ti (13nm) / Au (75nm) stacked electrode structure, and the length of the conductive channel is 20μm. The 20μm conductive channel length provides a larger sensing area, enhances the antibody loading capacity, and is conducive to the capture and detection of low-concentration biomarkers; and the extended channel design improves the uniformity of the electric field distribution, making the detection signal more stable and reliable. The 13nm Ti layer provides a stronger interface binding force, which is suitable for scenarios that require repeated detection or long-term monitoring. The 75nm Au layer thickness optimizes the conductive performance of the 75nm Au layer and maintains signal stability when detecting complex samples such as blood. This combination is more suitable for the detection of complex biological samples (such as whole blood or high-viscosity body fluids). The thicker Ti layer can better buffer thermal stress, while the increased Au layer thickness improves resistance to electrochemical corrosion in biological environments.

[0057] Example 4

[0058] The structure of the CNT-FET biosensor is the same as that of Examples 1-3. The CNT-FET biosensor of this embodiment adopts a Ti (5nm) / Au (100nm) laminated electrode structure. The 5nm Ti layer not only ensures the firm bonding between the electrode and the substrate, but also minimizes the influence of the metal layer on the activity of biomolecules. The 100nm thick Au layer has excellent mechanical strength and surface stability, can withstand multiple cleanings and reuses, and is suitable for repeatability detection experiments. Combined with the strong interfacial bonding provided by the 5nm Ti adhesion layer, the reliability of this electrode structure in long-term monitoring applications provides a stable and reliable detection platform for clinical dynamic monitoring of biomarkers. This design is particularly suitable for implantable monitoring devices or continuous detection systems that require long-term stability. The thicker metal laminate can effectively resist the erosion of the biological environment and ensure the long-term stability of the sensor in a complex body fluid environment.

[0059] The biosensor based on carbon nanotube field effect transistor (CNT-FET) of the present invention has a simple structure and low preparation cost, and can detect the expression level of PD-1 in peripheral blood non-invasively and rapidly. Compared with the prior art, the Ti adhesion layer in the present invention does not exceed 20nm, and forms a strong bond with the SiO2 substrate, while ensuring the stability of the electrode structure and avoiding the interference of the excessively thick metal layer on the biological molecule detection signal; the thickness of the Au conductive layer is set between 50-100nm to ensure excellent conductivity while achieving low contact resistance, and low resistance contact is related to the sensitivity of the device. The metal stacked structure of the electrode of the present invention creates an ideal interface environment for the fixation and detection of biological molecules while ensuring the electrical performance of the device. The present invention does not require complex dielectric layers or nanoparticle modifications, and achieves high-sensitivity detection by directly adsorbing anti-PD-1 antibodies.

[0060] Example 5

[0061] The present invention is also a method for preparing a CNT-FET biosensor, which is directed to the preparation of a CNT-FET biosensor. The structure of the CNT-FET biosensor is the same as that of Examples 1-4, see Figure 3 , Figure 3 It is a flowchart of the preparation method of the present invention, which includes the following steps:

[0062] Step 1: Design a CNT-FET biosensor: This includes a silicon dioxide substrate, a source electrode, a drain electrode, and semiconducting single-walled carbon nanotubes (s-SWCNTs). A conductive channel is defined on the top surface of the silicon dioxide substrate, flanked by a source and drain electrode. Both the source and drain electrodes are constructed of a Ti / Au stack, and the conductive channel material is s-SWCNTs. The s-SWCNTs form ohmic contacts with the source and drain electrodes. In the present invention, the s-SWCNTs serve as the conductive channel material, terminated on the top surfaces of the source and drain electrodes, and form low-resistance ohmic contacts with the Ti / Au stacks of the source and drain electrodes, respectively, forming the biosensor's sensing interface. The device's top surface lacks an oxide dielectric layer and no protective layer. Covalent coupling modification is not required during antibody immobilization, and the device directly serves as the biosensor's sensing interface. The conductive channel on the biosensor's sensing interface is either slightly higher than or flush with the top surfaces of the source and drain electrodes.

[0063] See also Figure 4 , Figure 4 It is a schematic diagram of the preparation process of the present invention.

[0064] Step 2: Clean the silicon dioxide substrate: Use a p-type silicon dioxide substrate with a resistivity of 0-0.0015Ω·cm and a 100nm thermal oxide layer. <100> The silicon wafer was used as the substrate, and the silicon dioxide substrate was ultrasonically cleaned with acetone, isopropyl alcohol and ultrapure water in sequence, each cleaning lasting 3 minutes, and then purged with nitrogen.

[0065] Step 3, photolithographic patterning of the device on the upper surface of the substrate: first design the device photolithographic pattern on the upper surface of the substrate. The photolithographic pattern is a conductive channel with a diameter of 10.0±0.2μm in the middle, and a device mask pattern with a source window and a drain window on both sides of the conductive channel. The pattern shapes of the source window and the drain window are designed to be two symmetrical squares or two symmetrical rectangles, and the source and the drain are symmetrical about the center line of the conductive channel; use a photolithographic process to prepare the designed device mask pattern on the upper surface of the device substrate to complete the patterning of the upper surface of the substrate.

[0066] Step 4, magnetron sputtering to prepare the Ti / Au stack of the source and drain: on the patterned upper surface of the substrate, a titanium (Ti) adhesion layer is deposited coveringly by a magnetron sputtering process, and a gold (Au) conductive layer is continued to be sputtered on the upper surface of the Ti layer. At this time, the metal layer covers the entire upper surface of the device, and then the photoresist and the covering redundant metal outside the source and drain electrode windows are selectively removed by a stripping process, retaining the Ti / Au stack in the electrode window area, and a semi-finished device with active and drain electrodes is obtained, completing the preparation of the Ti / Au stacked electrodes of the source and drain on the upper surface of the substrate; at this time, the upper surface of the device is the source and drain electrode surface area composed of the Ti / Au stack, and the conductive channel surface area located between the source and drain without the Ti / Au stack, and the source and drain are not electrically connected.

[0067] Step 5: Deposit s-SWCNTs to form a low-resistance ohmic contact between the channel and the electrode: On the upper surface of the device where the source and drain electrodes have been stacked, high-purity s-SWCNTs are uniformly deposited on the entire upper surface of the device, including the Ti / Au electrode surface and the 10.0±0.2μm exposed conductive channel area in the middle, through a solution deposition process. The s-SWCNTs form a continuous, uniform, and covering large-area randomly distributed network film on the Ti / Au electrode surface and the conductive channel area in the middle; the s-SWCNTs located in the conductive channel build a conductive network channel connecting the source and drain electrodes on the exposed substrate surface. On the upper surface of the electrode, the s-SWCNTs in the conductive channel form a low-resistance ohmic contact with the source and drain electrodes.

[0068] Step 6: Removing s-SWCNTs outside the conductive channel to complete device fabrication: Selectively retaining the channel s-SWCNTs: After the s-SWCNTs in the conductive channel form low-resistance ohmic contacts with the source and drain electrodes, oxygen plasma etching is used to selectively retain 10.0±0.2μm of the conductive channel s-SWCNTs and remove redundant s-SWCNTs outside the conductive channel region. Finally, ultrasonic cleaning is performed using acetone, isopropyl alcohol, and ultrapure water, each for 3 minutes. After nitrogen purging, device fabrication is complete. Compared to the "CNT deposition followed by electrode formation" process, this method avoids contact issues caused by uneven metal coverage during electrode deposition, and the contact resistance between the CNTs and the prefabricated electrodes after etching is lower.

[0069] The CNT-FET biosensor preparation method provided by the present invention proposes innovative solutions to several key problems existing in the prior art, such as complex preparation, high cost, and stringent process requirements. The traditional CNT-FET preparation process usually adopts the process of "depositing CNTs first and then preparing electrodes", which easily leads to poor contact between the metal electrode and the carbon nanotube, resulting in high contact resistance, which will affect the charge transfer efficiency of the device. The present invention has made innovative improvements and adopted the preparation sequence of "preparing electrodes first and then depositing CNTs". By accurately preparing Ti / Au stacked electrodes in advance, and then depositing s-SWCNTs and selectively etching them, it ensures that a low-resistance ohmic contact is formed between the carbon nanotubes and the electrodes, significantly improving the electrical performance and stability of the device, reducing contact resistance, and enhancing charge transfer efficiency. This makes the electrical signal response generated by the sensor when detecting PD-1 protein more sensitive, thereby improving the sensitivity of the CNT-FET biosensor in detecting PD-1 protein.

[0070] In existing technologies, biosensor fabrication often requires complex surface modification and protective layer deposition, which not only increases process complexity but also raises production costs. The present method completely avoids the deposition of additional oxide dielectric and protective layers. By optimizing process parameters, it directly utilizes the exposed s-SWCNT network as the sensing interface. The removal of the dielectric layer brings the sensing interface closer to the conductive channel, facilitating improved detection sensitivity. This simplifies the fabrication process, significantly reducing material costs and preparation time.

[0071] Example 6

[0072] The CNT-FET biosensor and its preparation method are the same as those in Examples 1-5. The photolithographic patterning of the device on the substrate surface in step 3 of the preparation method of the present invention includes the following steps:

[0073] 3.1 Spin-coating photoresist: AZ5214 photoresist is spin-coated on a clean SiO2 substrate at a speed of 5000 rpm / s for 30 minutes to form a 1.25 μm thick photoresist layer. This process effectively eliminates the coffee ring effect, ensures that the steepness of the pattern edges and the deviation of key dimensions after exposure and development are within the appropriate range, fully meeting the requirements for accuracy and consistency in the preparation of CNT-FET sensors, and laying a solid foundation for the subsequent preparation of high-precision source and drain electrodes.

[0074] 3.2 UV alignment exposure: Use a mask to perform UV alignment exposure on the upper surface of the SiO2 substrate after spin coating of photoresist. Use an MA6 photolithography machine and use a mask to UV expose for 22s with a light intensity of 400W / cm 2, ensuring that the pattern accuracy reaches the sub-micron level. In the present invention, the source and drain are first patterned by photolithography, and the designed device mask pattern is prepared on the surface of the device substrate by using a photolithography process, see Figure 2 The source 4 and drain 5 are symmetrically located on either side of the conductive channel 6, mirror-symmetric about the y-axis, or the channel's centerline. This means that if the channel lies on the y-axis, the source and drain are symmetrical about the y-axis, or the channel's centerline. This process offers high resolution and excellent repeatability, with minimal exposure uniformity error, providing precise positioning for subsequent electrode patterning and significantly improving device consistency and yield.

[0075] 3.3 Forming the electrode pattern of source and drain: The SiO2 substrate after UV exposure is placed in a developer for 60 seconds to develop a clear electrode pattern, and then baked at 110℃ for 90 seconds to improve the pattern stability. The following pattern is formed: the source and drain are symmetrically distributed on the surface of the silicon dioxide substrate, forming a 50×50μm 2 The two pads are separated by a rectangular conductive channel 10.0±0.2μm long. Precise control of the development time during this process ensures sharp pattern edges, while a post-baking process improves pattern dimensional stability. The channel length error is less than 2%, ensuring uniformity in subsequent CNT deposition and laying the foundation for the fabrication of high-performance CNT-FET devices.

[0076] Example 7

[0077] The CNT-FET biosensor and its preparation method are the same as those in Examples 1-6, wherein the Ti / Au stacked electrode structure is prepared by magnetron sputtering as described in step 4, including the following steps:

[0078] 4.1 Sputtering of Ti adhesion layer: First, the Ti adhesion layer is sputtered. After the substrate is cleaned, a 5nm Ti layer is sputtered in a vacuum chamber as an adhesion layer. The present invention enhances the bonding strength between the metal electrode and the substrate. The power is 100W and the time is 50s. The precisely controlled sputtering parameters ensure uniform coverage of the Ti layer on the SiO2 substrate, which is beneficial to the subsequent deposition of the Au layer.

[0079] 4.2 Au conductive layer sputtering: Then the Au conductive layer is deposited. After the Ti layer is deposited, a 100 nm thick Au layer is immediately sputtered without breaking the vacuum. The sputtering power is adjusted to 100 W and the time is 8.5 minutes. The precisely controlled sputtering time makes the Au layer have excellent surface flatness.

[0080] 4.3 Selective Stripping and Cleaning: After sputtering, the substrate is immediately immersed in acetone and allowed to stand for 12 hours before being stripped with a purge process. Ultrasonic cleaning is used to remove the photoresist and redundant metal. Finally, the silicon dioxide substrate is ultrasonically cleaned in acetone, isopropyl alcohol, and ultrapure water, respectively, for 3 minutes each time. This allows the deposition of high-quality Ti / Au metal stack electrodes on the substrate, providing an ideal contact interface for subsequent carbon nanotube deposition.

[0081] In Examples 6 and 7, electrodes were fabricated directly on the substrate surface. By preforming a complete electrode structure on the substrate surface, the physical damage to the carbon nanotubes that can occur during post-deposition of electrodes in conventional processes was avoided. This fabrication process not only ensures precise electrode patterning but also provides an ideal contact interface for subsequent s-SWCNT deposition, laying a solid foundation for future biosensing applications.

[0082] Example 8

[0083] The CNT-FET biosensor and its preparation method are the same as those in Examples 1-7, except that the s-SWCNTs are deposited in step 5 to form a low-resistance ohmic contact between the channel and the electrode, including the following steps:

[0084] 5.1 Uniform deposition of s-SWCNTs: The SiO2 substrate with completed source and drain electrodes was immersed in a s-SWCNTs solution with a purity of ≥99.9% and allowed to stand at room temperature for 48 hours to form a uniform thin film of s-SWCNTs with a thickness of approximately 5 nm on the substrate surface. This high-purity s-SWCNTs solution can effectively avoid the interference of metallic CNTs on the electrical performance of the device, significantly improving the device switching ratio and reducing background noise. At the same time, the mild room temperature deposition conditions can maintain the integrity of the CNT structure, avoid defects caused by high temperature or mechanical force, and lay the foundation for the subsequent formation of a high-quality conductive network.

[0085] 5.2 Formation of s-SWCNTs thin film: A continuous, uniform, large-area s-SWCNTs network film is formed on the surface of the source and drain electrodes and in the conductive channel area between the source and drain electrodes, so that the source and drain electrodes form low-resistance ohmic contacts with the s-SWCNTs respectively; the full contact between the s-SWCNTs and the source and drain electrodes forms an ohmic contact, effectively reducing the contact resistance. This uniformly distributed s-SWCNTs network structure can effectively improve the consistency of the device and provide sufficient active sites for subsequent biomolecule modification. It is a key step in obtaining a highly sensitive biosensor.

[0086] Depositing the carbon nanotubes after electrode preparation avoids damage to the carbon nanotube structure caused by the high temperature and stress of the metal sputtering process, ensuring the integrity of the carbon nanotube lattice. This preserves the intrinsic electrical properties of the carbon nanotubes and helps control the contact quality of the carbon nanotube / electrode interface, reducing contact resistance and providing optimal conditions for constructing high-performance biosensing interfaces.

[0087] Example 9

[0088] The CNT-FET biosensor and its preparation method are the same as those in Examples 1-8, except that the s-SWCNTs outside the conductive channel are removed in step 6 to complete the device preparation, including the following steps:

[0089] 6.1 Formation of terminated s-SWCNTs: Using a photolithography process with an overlay accuracy of ±0.25 μm and cross-hair alignment marks with a line width of 1 μm, the s-SWCNTs uniformly distributed on the source and drain surfaces were precisely patterned to define a conductive channel region 10.0 ± 0.2 μm long. The channel region to be retained and the non-channel region to be etched were marked.

[0090] 6.2 Oxygen Plasma Selective Etching: Through an oxygen plasma etching process, the etching parameters (oxygen flow rate of 200 sccm, power of 100 W, and time of 1 min) are precisely controlled to selectively remove s-SWCNTs outside the source, drain, and channel regions. At the same time, a photoresist is used as a mask to protect the s-SWCNTs in the channel region, forming a high-purity s-SWCNT termination structure that remains only in the conductive channel between the source and drain, thereby reducing contact resistance and improving device performance. These s-SWCNTs are mainly distributed in the conductive channel region between the source and drain. The s-SWCNTs in the conductive channel region are connected to the electrode by natural extension.

[0091] 6.3 Cleaning and device completion: Finally, ultrasonic cleaning (power 80-100W) with acetone, isopropyl alcohol and ultrapure water for 3 minutes each was performed to completely remove residual photoresist and etching byproducts, ensuring the cleanliness and structural integrity of the s-SWCNTs in the channel, and finally obtaining a high-cleanliness CNT-FET.

[0092] In the present invention, the ends of s-SWCNTs contact the electrode sidewalls with less stress. This naturally formed interface effectively reduces the contact resistance, avoids the metal coverage defects in traditional processes, and provides a more stable and sensitive signal conduction interface for biosensing.

[0093] Example 10

[0094] The present invention also provides a non-invasive detection method for PD-1 protein using a CNT-FET biosensor, which is a method for detecting PD-1 using a CNT-FET biosensor. The CNT-FET biosensor and its preparation method are the same as those in Examples 1-9, see Figure 5 , Figure 5 This is a flowchart of the detection process for PD-1 detection of the present invention, which includes the following steps:

[0095] Step S1, initial electrical performance characterization: connect the unmodified CNT-FET biosensor to a 4200-SCS semiconductor characterization system and fix the drain-source voltage V ds =-1V, gate voltage V gs The initial transfer characteristic test was performed under the condition of scanning range from -5V to +25V to ensure the good performance of CNT-FET biosensor, and the benchmark stable on-state current I0 was recorded.

[0096] Step S2, immobilizing PD-1 monoclonal antibody sites: 20 μL of 0.1 mg / mL PD-1 monoclonal antibody was dripped onto the conductive channel surface formed by s-SWCNTs in the CNT-FET biosensor. The mixture was incubated at a constant temperature of 4°C for 6 hours. The PD-1 monoclonal antibody was directly immobilized on the s-SWCNT surface through non-covalent electrostatic interactions, forming PD-1-specific binding sites. Unimmobilized areas of the PD-1 monoclonal antibody were nonspecific binding sites, or redundant sites, resulting in a CNT-FET biosensor with immobilized PD-1 monoclonal antibody sites. The immobilization process did not introduce any covalent modifiers or cross-linkers, and the antibody molecules maintained their native conformation and biological activity through electrostatic adsorption. Direct electrostatic adsorption was used to directly immobilize the PD-1 monoclonal antibody on the device surface, eliminating the need for complex modification processes, maximizing antibody activity, and ensuring that the immobilized PD-1 monoclonal antibody could specifically recognize and bind to the PD-1 protein.

[0097] Step S3, blocking nonspecific binding: blocking excess sites, immersing the CNT-FET biosensor with fixed PD-1 monoclonal antibody sites in 1 mg / mL bovine serum albumin (BSA) blocking solution, incubating at room temperature of 25°C for 1 hour, and rinsing with deionized water three times to remove unbound BSA molecules, effectively blocking the nonspecific binding sites on the surface of the CNT-FET biosensor, and obtaining a CNT-FET biosensor after BSA blocking treatment; the BSA blocking treatment covers the remaining active sites on the surface of s-SWCNTs by electrostatic adsorption without affecting the activity of the fixed PD-1 monoclonal antibody, thereby significantly reducing nonspecific adsorption interference in subsequent detection.

[0098] Step S4, baseline electrical characteristics test: connect the CNT-FET biosensor after BSA blocking treatment to the 4200-SCS semiconductor characteristics analysis system to obtain the transfer characteristic curve after antibody fixation, obtain the complete transfer characteristic curve of the CNT-FET biosensor by gate voltage scanning, record and determine the stable on-state current I1 of the source-drain of the CNT-FET biosensor at a specific operating voltage, and use it as the baseline current value I1 for PD-1 protein detection to complete the baseline electrical characteristics test. The determination of the reference current I1 is completed before PD-1 protein binding and is used as a quantitative control for subsequent specific binding detection. During the test, the device is kept in a stable electrical environment to ensure data reliability.

[0099] Step S5, specific binding of PD-1 protein and antibody: 1 μL of the serum sample to be tested was diluted to a specific concentration and evenly added to the surface of the CNT-FET biosensor that had completed the benchmark electrical test. The sample was placed in a constant temperature environment at 4°C and allowed to react for 1 hour to allow the PD-1 protein in the serum to specifically bind to the PD-1 monoclonal antibody immobilized on the surface of the s-SWCNTs. The sample was rinsed three times with 1×PBS buffer to remove unbound serum proteins and other interfering substances. After drying with nitrogen, the specific binding of the PD-1 protein and the PD-1 monoclonal antibody was completed, and the electrical properties test was immediately performed.

[0100] Step S6, electrical property detection after specific binding of PD-1 protein to antibody: After specific binding of PD-1 protein to PD-1 monoclonal antibody site, electrical property detection is immediately performed. The test conditions are the same as step S3. The same semiconductor property analysis system is used to detect the electrical property of PD-1 protein after specific binding to antibody, and the transfer characteristic curve is obtained. The stable on-state current value I2 is obtained. After the electrical property detection of PD-1 protein after specific binding to antibody is completed, the detection data of PD-1 protein by CNT-FET biosensor is obtained in real time. Thanks to the excellent structural characteristics of carbon tubes, the sensor has the ability to detect mouse PD-1 with a concentration of 0.0239fg / mL. Figure 7 , Figure 7 This is a transfer characteristic curve diagram of the present invention for detecting mouse serum. It can be seen that the CNT-FET biosensor of the present invention has excellent detection performance for mouse PD-1, and the entire detection process can be completed within half an hour, which fully meets the needs of clinical real-time monitoring.

[0101] Step S7, detection data analysis, providing the detection results of PD-1 protein: After obtaining the detection data of PD-1 protein by the CNT-FET biosensor in real time, the data is analyzed, including the following: comparing the sizes of I1 and I0, as well as I2 and I1, to characterize the antibody fixation efficiency and the concentration of PD-1, respectively; drawing and obtaining standard curves for different PD-1 concentrations, and providing the detection results of PD-1 protein.

[0102] The results of non-invasive detection of PD-1 protein using CNT-FET are analyzed as follows:

[0103] 1) Establishment of reference current system:

[0104] I0: Steady on-state current of unmodified bare CNT-FET;

[0105] I1: Steady on-state current after PD-1 antibody immobilization;

[0106] I2: Steady on-state current after PD-1 protein binds to PD-1 antibody;

[0107] 2) Analysis of current data changes:

[0108] ΔI1=I1-I0: characterizes the antibody fixation efficiency;

[0109] ΔI2=I2-I1: specificity reflects the amount of PD-1 binding;

[0110] A standard curve of PD-1 concentration was established to obtain the quantitative detection results of PD-1.

[0111] In the present invention, PD-1 is used as a key biomarker for evaluating the efficacy of immunotherapy for non-small cell lung cancer. Existing ELISA and other medical detection technologies for PD-1 have many steps, rely on professional laboratories and experimental personnel, have strict requirements on the test operation process, operating environment, and sample amount, and have problems such as complex operation and long time consumption, which makes it difficult to meet the needs of clinical real-time monitoring. The present invention is an overall technical solution that realizes the rapid and accurate detection of PD-1 levels, which is conducive to timely adjustment of treatment plans. The present invention only requires 5μL of trace serum to complete the detection, which reduces the sampling burden on patients and makes long-term longitudinal studies of small animal models possible. The entire detection process does not require labeling operations, avoids errors caused by fluctuations in labeling efficiency, and the test results have higher reliability and repeatability. The present invention has broad application prospects in the fields of immunotherapy efficacy evaluation and drug resistance mechanism research.

[0112] Example 11

[0113] The non-invasive detection method of PD-1 protein using CNT-FET biosensor is the same as that in Example 10, except that the PD-1 monoclonal antibody is fixed in step S2, comprising the following steps:

[0114] S2-1, Dilution of PD-1 monoclonal antibody: PD-1 monoclonal antibody was diluted to 0.1 mg / mL with 1× PBS buffer to form a homogeneous antibody solution. This concentration was systematically optimized to ensure sufficient antibody coverage of the carbon nanotube surface while avoiding steric hindrance caused by excess antibody, laying the foundation for subsequent high-sensitivity detection.

[0115] S2-2, non-covalent fixation of PD-1 monoclonal antibody on s-SWCNTs: The diluted PD-1 monoclonal antibody solution was directly added to the conductive channel surface of the s-SWCNTs of the CNT-FET without adding covalent modifiers or chemical cross-linkers. The mixture was incubated at 4°C and intermittently rinsed with 1× PBS buffer every 10 minutes for 2 hours to reduce nonspecific adsorption. The mixture was then allowed to stand for another 4 hours to allow the PD-1 monoclonal antibody to be stably adsorbed on the s-SWCNTs surface through electrostatic interactions, forming a high density of specific binding sites. The innovative intermittent rinsing combined with segmented incubation method not only ensured that the antibody was uniformly and stably adsorbed on the carbon tube surface through electrostatic interactions, but also effectively reduced nonspecific binding, allowing the antibody to maintain its optimal bioactive conformation.

[0116] S2-3, Removal of non-binding sites: After incubation, the CNT-FET surface is dynamically rinsed with 1×PBS buffer and ultrapure water to remove unbound PD-1 monoclonal antibodies and potential nonspecific adsorbed molecules again, and finally blown dry for use. The sensing interface is further purified by dynamic rinsing, and unbound antibody molecules and other interfering substances are removed. The antibody-functionalized surface formed has high uniformity and specificity, providing ideal conditions for the subsequent specific recognition of PD-1 protein. In the present invention, the entire antibody fixation process completely avoids the influence of traditional covalent modification on the electrical properties of carbon tubes, while retaining the biological activity of the antibody to the greatest extent, ensuring the detection sensitivity and specificity of the sensor.

[0117] Example 12

[0118] The non-invasive detection method of PD-1 protein using CNT-FET biosensor is the same as that in Example 10-11. The specific detection of PD-1 protein in step S5 is divided into two cases, including the following steps:

[0119] S5-1, detection of mouse serum PD-1: dilute the mouse serum sample containing 0.478 ng / mL PD-1 to 0.478 fg / mL, pipette 10 μL and drop it onto the antibody immobilization area of ​​the CNT biosensor. Incubate at 4°C for 1 hour to complete antigen-antibody specific binding. Rinse with a large amount of 1x PBS solution to remove non-bound components and purge with nitrogen. Figure 7 The red curve in the figure is the transfer curve when detecting mouse PD-1 protein. Compared with the yellow curve, the current is higher, which verifies that the sensor of the present invention has the ability to detect mouse PD-1 with a concentration of 0.478fg / ml. Mouse serum samples with a concentration of 0.478ng / mL PD-1 were gradiently diluted to a concentration range of fg / mL-pg / mL for testing. Figure 8 , Figure 8 This is a transfer characteristic curve for detecting different concentrations of mouse serum. The CNT-FET biosensor has good detection capabilities for various concentrations of mouse PD-1. The current increases with increasing PD-1 concentration. The CNT-FET biosensor also has good detection performance at 0.0239 fg / mL of mouse PD-1. The successful detection of low concentrations demonstrates the sensor's good sensitivity. The requirement for a small sample volume of 10 μL is particularly suitable for long-term monitoring in small animal models.

[0120] S5-2, detection of human serum PD-1: Dilute a human serum sample containing 0.171 ng / mL PD-1 to 1.171 pg / mL. Pipette 10 μL and drop it onto the antibody-immobilized area of ​​the CNT biosensor. Incubate at 4°C for 1 hour to complete antigen-antibody binding. Rinse with a large amount of 1x PBS solution to remove unbound components and purge thoroughly with nitrogen. Figure 9 , Figure 9 This is a transfer characteristic curve diagram for detecting human serum in the present invention. Figure 9 The red curve is the transfer curve when detecting human PD-1 protein. Compared with the yellow curve, the current decreases. The sensor of the present invention has good detection ability for human serum with a concentration of 1.171pg / mL and is also suitable for detecting PD-1 in human serum. From the perspective of scientific experiments, the universality of the present invention for PD-1 detection is verified. The stable detection of PD-1 in 1.171pg / mL human serum shows that the sensor of the present invention has excellent resistance to matrix interference.

[0121] The present invention uses a CNT-FET biosensor to detect human and mouse PD-1 proteins. It is applicable to the specific binding detection of both human and mouse PD-1 antibodies and their corresponding antigens, achieving compatibility testing for both human and mouse samples. This invention possesses excellent species versatility and can support the full process from basic research to clinical translation, providing a unified detection platform for immunotherapy research.

[0122] Example 13

[0123] The non-invasive detection method of PD-1 protein using CNT-FET biosensor is the same as that in Examples 10-12, see Figure 6 , Figure 6 Schematic diagram of the binding of PD-1 protein and antibody on the sensing surface in the detection method of the present invention; the present invention first detects PD-1 in mouse serum using CNT-FET biosensor, see Figure 7 In a PBS environment with a pH of 7-7.4, since the isoelectric point (pI) of the mouse PD-1 monoclonal antibody is between 8 and 9, which is higher than the ambient pH, the positive charge of the antibody repels the holes in the p-type CNT-FET, causing the current to decrease. Figure 7 The yellow curve in the middle shows that the antibody is successfully adsorbed on the sensor surface; the change in contact angle also proves this fact, see Figure 10 , Figure 10 This is a contact angle diagram of the CNT-FET of the present invention before the antibody is fixed; before the antibody is attached, the contact angle of the CNT-FET biosensor is about 77°. After the highly hydrophilic antibody is attached to the device, the contact angle decreases to about 56°. Figure 11 , Figure 11 This is the contact angle diagram of the CNT-FET after the antibody is fixed in the present invention; after the PD-1 protein with weak hydrophilicity binds to the PD-1 monoclonal antibody, the contact angle rises slightly to about 62°, see Figure 12 , Figure 12 This is a contact angle diagram of the CNT-FET after PD-1 and PD-1 monoclonal antibody binding. A test was conducted on mouse serum containing 0.478 fg / mL of PD-1 protein. The current increased after specific adsorption of PD-1 protein. This is because the pI of mouse PD-1 is between 5.8 and 0.5, which is lower than the ambient pH, making PD-1 negatively charged. This attracts holes in the p-type CNT-FET, increasing the number of holes and the current. Figure 7 The red curve in the middle. At the same time, the CNT-FET biosensor can detect different concentrations of PD-1 and exhibits excellent PD-1 concentration response characteristics. Figure 8 The detection current signal shows a good growth relationship with the increase of PD-1 concentration. The CNT-FET biosensor prepared by the present invention realizes the rapid and non-invasive detection of PD-1 in mouse serum, and can dynamically monitor the slight changes in PD-1 expression during NSCLC immunotherapy, providing a new real-time monitoring method for clinical efficacy evaluation.

[0124] The present invention uses the same CNT-FET biosensor to detect human serum with a PD-1 protein concentration of 1.171 pg / mL. When human PD-1 monoclonal antibody (pI of about 8) is attached to the CNT-FET, the current decreases. Figure 9 The yellow curve shows that after the PD-1 protein in human serum is specifically bound, the current continues to decrease because the PI of human PD-1 protein is about 8.05, which is higher than the PBS environment with pH = 7-7.4. Figure 9 The red curve shows that the present invention has a good detection ability for PD-1 in human serum.

[0125] In summary, the present invention provides a carbon-based biosensor and its preparation and noninvasive detection method for PD-1. This invention, for the first time, utilizes a CNT-FET biosensor for rapid, noninvasive detection of PD-1. The present invention addresses the technical challenges of reducing ohmic contact between electrodes and s-SWCNTs, directly immobilizing antibodies on the surface of bare s-SWCNTs, and enabling rapid, noninvasive detection of human and mouse PD-1 using a CNT-FET biosensor. The carbon-based biosensor s-SWCNTs of the present invention serve as the conductive channel material, terminated on the upper surfaces of the source and drain electrodes, and forming low-resistance ohmic contacts with the Ti / Au stacks of the source and drain electrodes, respectively. The entire biosensor constitutes the sensing interface of the biosensor. The device's upper surface lacks an oxide dielectric layer, no additional protective layer, and no covalent coupling modification, effectively serving as the biosensor's sensing interface. The carbon-based biosensor preparation method includes the CNT-FET biosensor preparation method described in the present invention, which includes the following steps: designing the CNT-FET biosensor; cleaning the silicon dioxide substrate; photolithographically patterning the device on the substrate surface; magnetron sputtering to prepare the Ti / Au stack of the source and drain electrodes; depositing s-SWCNTs to form a low-resistance ohmic contact between the channel and the electrodes; and removing the s-SWCNTs outside the conductive channel to complete the device preparation. The non-invasive detection of PD-1 protein in the CNT-FET sensor of the present invention includes initial electrical performance characterization; immobilization of PD-1 monoclonal antibody sites; blocking nonspecific binding; baseline electrical property testing; PD-1 protein specific detection; electrical property testing after PD-1 protein binding; and quantitative analysis of PD-1 protein to obtain quantitative detection results of PD-1.

[0126] For the first time, CNT-FETs were used to noninvasively and rapidly detect human and mouse PD-1 in real time, leveraging the sensitivity of CNTs to changes in surface charge to enable PD-1 protein detection. Using a "pre-electrode preparation followed by CNT deposition" fabrication process, a Ti / Au electrode was pre-formed before s-SWCNT deposition, creating an ohmic contact with lower resistance. An antibody immobilization method based on electrostatic adsorption was developed, avoiding the effects of traditional covalent modification on antibody activity.

[0127] Based on the excellent electrical properties of carbon nanotubes, the present invention can achieve rapid detection of PD-1 protein, and the response time is greatly shortened compared with traditional methods, meeting the needs of clinical real-time monitoring. Only a trace amount of sample (<5μL) is required to complete the detection, which greatly reduces the sampling burden on patients and realizes non-invasive detection in the true sense. In terms of the detection process, the present invention has outstanding convenience. The entire detection process does not require complicated sample pretreatment, eliminates the labeling and amplification steps in traditional methods, and the operation process is significantly simplified. The present invention supports continuous dynamic monitoring, which can reflect the changes in PD-1 protein levels in real time, and provide timely and accurate reference for clinical treatment. From the perspective of clinical application, the detection process of the present invention is completely non-invasive, and the preparation process is simple and low-cost. It only requires a small amount of peripheral blood to complete, avoiding the pain brought to patients by traditional tissue biopsy. At the same time, the fast detection speed makes it very suitable for application in clinical scenarios such as outpatient clinics, greatly improving the accessibility and convenience of detection.

[0128] This invention has extremely broad application prospects and will have a profound impact in multiple key areas. In clinical diagnosis and treatment, it can provide real-time, dynamic efficacy monitoring solutions for tumor immunotherapy, enabling precise adjustment of treatment plans. In primary care, it can be developed into a portable, point-of-care testing device, making advanced testing technologies accessible to a wider population. This technology provides a new solution for PD-1 detection, has significant application value and development potential in the field of precision medicine, and has significant clinical value and social benefits.

Claims

1. A carbon-based biosensor comprising a silicon dioxide substrate, a source electrode, a drain electrode, and semiconducting single-walled carbon nanotubes (s-SWCNTs). The silicon dioxide substrate has a conductive channel in the middle and a source electrode and a drain electrode on either side. The source and drain electrodes are both Ti / Au stacks. The s-SWCNTs serve as a conductive channel material, terminated on the upper surfaces of the source and drain electrodes, and form low-resistance ohmic contacts with the Ti / Au stacks of the source and drain electrodes, respectively, forming the biosensor's sensing interface. The device's upper surface does not have an oxide dielectric layer and does not require covalent coupling modification, serving directly as the biosensor's sensing interface. The s-SWCNTs in the conductive channel on the biosensor's sensing interface are either slightly higher than or flush with the upper surfaces of the source and drain electrodes.

2. A carbon-based biosensor according to claim 1, characterized in that: The length of the conductive channel is 10 μm; the Ti / Au stack of the source and drain, Ti as an adhesion layer is located below the Ti / Au stack, with a thickness not exceeding 20 nm; Au as a conductive layer is located above the Ti / Au stack, with a thickness of 50 nm to 100 nm.

3. A method for preparing a CNT-FET biosensor, characterized in that: The following steps are included: Step 1: Design a CNT-FET biosensor: comprising a silicon dioxide substrate, a source electrode, a drain electrode, and semiconducting single-walled carbon nanotubes (s-SWCNTs). A conductive channel is provided on the upper surface of the silicon dioxide substrate, with a source electrode and a drain electrode on either side. The source and drain electrodes are both Ti / Au stacks. The s-SWCNTs serve as the conductive channel material and are terminated on the upper surfaces of the source and drain electrodes, forming low-resistance ohmic contacts with the Ti / Au stacks of the source and drain electrodes, respectively, to form a sensing interface of the biosensor. No oxide dielectric layer is provided on the upper surface of the device, requiring no covalent coupling modification, and the device directly serves as the sensing interface of the biosensor. The conductive channel on the biosensor sensing interface is either slightly higher than the upper surfaces of the source and drain electrodes or flush with the upper surfaces of the source and drain electrodes. Step 2: Clean the silicon dioxide substrate: Use a p-type silicon dioxide substrate with a resistivity of 0-0.0015Ω·cm and a 100nm thermal oxide layer. <100> The silicon wafer was used as the substrate and ultrasonically cleaned with acetone, isopropyl alcohol and ultrapure water in sequence, each cleaning lasting 3 minutes, and then purged with nitrogen. Step 3: Photolithographic patterning of the device on the upper surface of the substrate: First, design a device photolithographic pattern on the upper surface of the substrate. The photolithographic pattern is a conductive channel with a diameter of 10.0±0.2μm in the middle, and a device mask pattern with source and drain windows on either side of the conductive channel. The pattern shapes of the source and drain windows are designed to be two symmetrical squares or two symmetrical rectangles, and the source and drain are symmetrical about the center line of the conductive channel. Use a photolithographic process to prepare the designed device mask pattern on the upper surface of the device substrate to complete the patterning of the upper surface of the substrate. Step 4: magnetron sputtering to prepare the Ti / Au stack of the source and drain electrodes: on the patterned upper surface of the substrate, a titanium (Ti) adhesion layer is blanket deposited using a magnetron sputtering process, and a gold (Au) conductive layer is continuously sputtered on the upper surface of the Ti layer. Then, a stripping process is used to selectively remove the photoresist and the redundant metal covering outside the source and drain electrode windows, retaining the Ti / Au stack in the electrode window area, thereby completing the preparation of the Ti / Au stack electrodes of the source and drain electrodes on the upper surface of the substrate; Step 5: Depositing s-SWCNTs to form a low-resistance ohmic contact between the channel and the electrodes: On the upper surface of the device where the source and drain electrodes have been stacked, high-purity s-SWCNTs are uniformly deposited on the entire upper surface of the device, including the Ti / Au electrode surface and the 10.0±0.2μm exposed conductive channel region in the middle, using a solution deposition process. The s-SWCNTs form a continuous, uniform, and large-area randomly arranged network-like thin film on the Ti / Au electrode surface and the conductive channel region in the middle; the s-SWCNTs in the conductive channel form low-resistance ohmic contacts with the source and drain electrodes; Step 6: Remove s-SWCNTs outside the conductive channel to complete device fabrication: After the s-SWCNTs in the conductive channel form low-resistance ohmic contacts with the source and drain electrodes, an oxygen plasma etching process is used to selectively retain 10.0±0.2μm conductive channel s-SWCNTs and remove redundant s-SWCNTs outside the conductive channel area. Finally, ultrasonic cleaning is performed with acetone, isopropyl alcohol, and ultrapure water, respectively, for 3 minutes each time. After nitrogen purging, the device fabrication is completed.

4. The method for preparing a CNT-FET biosensor according to claim 3, wherein: The photolithographic patterning of the device on the substrate surface described in step 3 includes the following steps: 3.1 Spin coating photoresist: Spin coating photoresist AZ5214 on a clean SiO2 substrate at a spin speed of 5000 rpm / s for 30 s to form a 1.25 μm thick photoresist layer; 3.2 UV alignment exposure: Use a mask to perform UV alignment exposure on the upper surface of the SiO2 substrate after spin coating of photoresist. Use an MA6 photolithography machine and use a mask to UV expose for 22s with a light intensity of 400W / cm 2 ; 3.3 Forming the electrode pattern of source and drain: The SiO2 substrate after UV exposure is placed in a developer for 60 seconds to develop a clear electrode pattern, and then baked at 110℃ for 90 seconds to improve the pattern stability. The following pattern is formed: the source and drain are symmetrically distributed on the surface of the silicon dioxide substrate, forming a 50×50μm 2 There is a rectangular conductive channel with a length of 10.0±0.2μm between the two pads.

5. The method for preparing a CNT-FET biosensor according to claim 3, wherein: The deposition of s-SWCNTs in step 5 to form a low-resistance ohmic contact between the channel and the electrode includes the following steps: 5.1 Uniform deposition of s-SWCNTs: Immerse the SiO2 substrate with source and drain electrodes prepared in a s-SWCNTs solution with a purity of ≥99.9% and allow it to stand at room temperature for 48 hours until a uniform s-SWCNT film with a thickness of approximately 5 nm is formed on the substrate surface. 5.2 Formation of s-SWCNTs thin film: A continuous, uniform, large-area randomly arranged s-SWCNTs network film is formed on the surface of the source and drain electrodes and in the conductive channel area between the source and drain electrodes, so that the source and drain electrodes form low-resistance ohmic contacts with the s-SWCNTs respectively.

6. The method for preparing a CNT-FET biosensor according to claim 3, wherein: The s-SWCNTs outside the conductive channel are removed in step 6 to complete the device preparation, which includes the following steps: 6.1 Formation of terminated s-SWCNTs: Using a photolithography process with an overlay accuracy of ±0.25 μm and crosshair alignment marks with a line width of 1 μm, the s-SWCNTs were precisely patterned on the surface of the source and drain electrodes, defining a conductive channel region 10.0 ± 0.2 μm long. The channel region to be retained and the non-channel region to be etched were marked. 6.2 Oxygen Plasma Selective Etching: Through an oxygen plasma etching process, precisely controlling the etching parameters (oxygen flow rate of 200 sccm, power of 100 W, and time of 1 minute), the s-SWCNTs outside the source, drain, and channel regions are selectively removed. Simultaneously, a photoresist is used as a mask to protect the s-SWCNTs in the channel region, forming a high-purity s-SWCNT termination structure that remains only within the conductive channel between the source and drain electrodes, thereby reducing contact resistance and improving device performance. 6.3 Cleaning and device completion: Finally, ultrasonic cleaning (power 80-100W) with acetone, isopropyl alcohol and ultrapure water for 3 minutes each was performed to completely remove residual photoresist and etching byproducts, ensuring the cleanliness and structural integrity of the s-SWCNTs in the channel, and finally obtaining a high-cleanliness CNT-FET.

7. A non-invasive detection method for PD-1 protein using a CNT-FET biosensor, characterized in that: The following steps are included: Step S1, initial electrical performance characterization: connect the unmodified CNT-FET biosensor to a 4200-SCS semiconductor characterization system and fix the drain-source voltage V ds =-1V, gate voltage V gs The initial transfer characteristics test was performed under the condition of scanning range of -5V to +25V to ensure the good performance of CNT-FET biosensor and record the benchmark stable on-state current I0; Step S2, immobilizing PD-1 monoclonal antibody sites: 20 μL of 0.1 mg / mL PD-1 monoclonal antibody was dripped onto the conductive channel surface composed of s-SWCNTs of the CNT-FET biosensor, and incubated at a constant temperature of 4°C for 6 hours to allow the PD-1 monoclonal antibody to be directly immobilized on the s-SWCNTs surface through non-covalent electrostatic interactions, forming PD-1 specific binding sites. The areas where the PD-1 monoclonal antibody was not immobilized are non-specific binding sites, i.e., redundant sites, thereby obtaining a CNT-FET biosensor with immobilized PD-1 monoclonal antibody sites. Step S3, blocking nonspecific binding: blocking excess sites, immersing the CNT-FET biosensor with fixed PD-1 monoclonal antibody sites in 1 mg / mL bovine serum albumin (BSA) blocking solution, incubating at room temperature for 1 hour at 25°C, and rinsing with deionized water three times to remove unbound BSA molecules, effectively blocking nonspecific binding sites on the surface of the CNT-FET biosensor, and obtaining a CNT-FET biosensor after BSA blocking treatment; Step S4, benchmark electrical characteristics test: Connect the BSA-blocked CNT-FET biosensor to a 4200-SCS semiconductor characteristics analysis system to obtain a transfer characteristic curve after antibody immobilization. The transfer characteristic curve of the CNT-FET biosensor is obtained by varying the gate voltage. The stable on-state current I1 of the source-drain of the CNT-FET biosensor at a specific operating voltage is recorded and determined as the benchmark current value I1 for PD-1 protein detection, thereby completing the benchmark electrical characteristics test. Step S5, specific binding of PD-1 protein and antibody: 1 μL of the serum sample to be tested was diluted to a specific concentration and evenly dripped onto the surface of the CNT-FET biosensor that had completed the benchmark electrical test. The sample was placed in a constant temperature environment at 4°C and allowed to react for 1 hour to allow the PD-1 protein in the serum to specifically bind to the PD-1 monoclonal antibody immobilized on the s-SWCNTs surface. The sample was then rinsed three times with 1× PBS buffer to remove unbound serum proteins and other interfering substances. After drying with nitrogen, the specific binding of the PD-1 protein and the PD-1 monoclonal antibody was completed, and electrical properties testing was immediately performed. Step S6, electrical property testing after specific binding of PD-1 protein to antibody: After site-specific binding of PD-1 protein to PD-1 monoclonal antibody, electrical property testing is immediately performed. The test conditions are the same as those in step S3. The electrical property testing after specific binding of PD-1 protein to antibody is performed using the same semiconductor property analysis system to obtain a transfer characteristic curve and a stable on-state current value I2. This completes the electrical property testing after specific binding of PD-1 protein to antibody. Step S7, analyzing the detection data and providing the detection results of the PD-1 protein: After obtaining the detection data of the PD-1 protein by the CNT-FET biosensor in real time, the data is analyzed, including the following: comparing the sizes of I1 and I0, and I2 and I1, respectively, to characterize the antibody fixation efficiency and the PD-1 concentration; Draw and obtain the standard curve of different PD-1 concentrations, and give the detection results of PD-1 protein.

8. The non-invasive detection method for PD-1 protein using a CNT-FET biosensor according to claim 6, characterized in that: The fixed PD-1 monoclonal antibody site described in step S2 comprises the following steps: S2-1, dilution of PD-1 monoclonal antibody: dilute PD-1 monoclonal antibody to 0.1 mg / mL with 1× PBS buffer to form a homogeneous antibody solution; S2-2, non-covalent immobilization of PD-1 monoclonal antibody on s-SWCNTs: The diluted PD-1 monoclonal antibody solution was directly added dropwise to the conductive channel surface of the s-SWCNTs of the CNT-FET without adding covalent modifiers or chemical cross-linkers. The solution was incubated at 4°C, with intermittent rinsing with 1× PBS buffer every 10 minutes for 2 hours to reduce nonspecific adsorption. The solution was then allowed to stand for another 4 hours to allow the PD-1 monoclonal antibody to be stably adsorbed on the s-SWCNT surface through electrostatic interactions, forming a high density of specific binding sites. S2-3, Removal of non-binding sites: After incubation, the CNT-FET surface was dynamically rinsed with 1× PBS buffer and ultrapure water to remove unbound PD-1 monoclonal antibodies and potential nonspecific adsorbed molecules again, and finally blown dry for later use.

9. The non-invasive detection method for PD-1 protein using a CNT-FET biosensor according to claim 6, characterized in that: The PD-1 protein specific detection described in step S5 is divided into two cases, including the following steps: S5-1, detection of mouse serum PD-1: A mouse serum sample containing 0.478 ng / mL PD-1 was diluted to 0.478 fg / mL. 10 μL was pipetted and added dropwise to the antibody-immobilized area of ​​the CNT biosensor. After incubation at 4°C for 1 hour to complete antigen-antibody specific binding, the sample was rinsed with a large amount of 1x PBS solution to remove non-bound components and purged with nitrogen gas. S5-2, detection of human serum PD-1: A human serum sample containing 0.171 ng / mL PD-1 was diluted to 1.171 pg / mL. 10 μL was pipetted and added dropwise to the antibody-immobilized area of ​​the CNT biosensor. After incubation at 4°C for 1 hour to complete antigen-antibody specific binding, the sample was rinsed with a large amount of 1x PBS solution to remove non-bound components and purged with nitrogen gas.

10. The non-invasive detection method for PD-1 protein using a CNT-FET biosensor according to claims 8-10, characterized in that: The method is universal across species. Its CNT-FET device preparation process, antibody fixation method and protein detection process are applicable to both human and mouse PD-1 proteins, realizing compatible detection of human and mouse samples.

Images