Precision cancer medicine with small tyrosine kinase inhibitors (TKIs) directed toward oncogenic drivers, are important treatment regimens for solid tumours. The epidermal growth factor receptor (EGFR)-TKI osimertinib is a preferred therapy for patients with non-small cell lung cancer (NSCLC) driven by activating mutations in EGFR , unfortunately responses are heterogeneous. This calls for non-invasive methods to predict or monitor treatment response, for example, via biomarker analyses in blood. To reveal such putative biomarkers, we analysed the proteome of extracellular vesicles (EVs) from osimertinib resistant or responsive NSCLC cells in vitro and from EVs isolated from serum samples of NSCLC patients treated with osimertinib in second line within the phase II clinical trial TREM. The protein cargo of the EVs was analysed by mass spectrometry (MS) and proximity extension assay (PEA). Western blotting, ELISA and single vesicle analysis was performed to validate and further confirm the expression of certain proteins. MS profiling of the NSCLC cells and their released EVs revealed a protein signature associated with osimertinib refractoriness. Among them were CSPG4, HSPG2, MCAM, L1CAM, TAGLN, THBS1 and TNC. GO-pathway analysis related several of these proteins to the focal adhesion and proteoglycan in cancer pathways. Some of these proteins, including CSPG4, which when suppressed by transient siRNA transfection in NSCLC cells resulted in reduced cell viability, were expressed also in EVs from serum of the NSCLC patients. Moreover, PEA profiling of the serum-isolated EVs revealed signatures associated with immune cells, best response and/or progression-free survival, including PD-L1, CD73/NT5E, FR-alpha/FOLR1, LAMP3, FASLG1 and ANXA1. In summary, we demonstrate that protein profiling of EVs in relation to osimertinib refractoriness has the potential to identify possible biomarkers that can indicate osimertinib treatment resistance, for example, CSPG4, HSPG2, TAGLN, TNC, THBS1, ANXA1 and CD73/NT5E. Studies in expanded cohorts should be conducted to further validate these putative osimertinib biomarkers.

Extracellular vesicles (EVs) are lipid membrane-bound nanoscale (20 nm – 1000 nm) objects shed by all cells. As EVs play an important role in intercellular signaling, these have emerged as promising biomarkers for many diseases including cancer and neurodegenerative disorders. A major bottleneck in research into EVs and their smooth translation to clinic as disease biomarkers is a lack of access to affordable and user-friendly technologies to quickly isolate EVs from complex biological and clinical samples with acceptable purity and yield. Here, we report two different designs of inertial microfluidic devices (e.g. LoAR-Exo and HiAR-Exo) that can isolate EVs from different kinds of biological samples in a single label-free step using a spiral microchannel design with different aspect ratios. LoAR-Exo has a height-to-width aspect ratio ∼ 1, while HiAR-Exo has a height-to-width aspect ratio of 2.5, optimized by COMSOL simulations. We tested both designs by separating polystyrene particles of size < 1 μm from a heterogeneous mixture. We then benchmarked the performance of the microfluidic chips against ultracentrifugation and a precipitation kit by isolating EVs from the cell culture-conditioned media (CCM) of MDA-MB-231 cells. We also compared the microchip with SEC by using the CCM from H1975/OR cells. The size distribution of EVs isolated by the microfluidic chip was comparable with ultracentrifugation, precipitation and SEC. In summary, both devices isolated EVs from as little as 1 mL of sample volume using a label-free technique in a continuous manner and without any user intervention. Both microfluidic platforms offer a simple, efficient, and scalable alternative to conventional methods for EV isolation. ### Competing Interest Statement DP, SA and AA have filed an Indian patent application (202421016839) and a PCT application (WO 2025/186844 A1) titled Device and method for label-free isolation of extracellular vesicles. The authors have no other conflicts of interest to declare. Wadhwani Research Centre for Bioengineering, IIT Bombay, DO/2022-WRCB002-082, DO/2025-WRCB002-105 Blockchain for Impact-BIOME, DO/2024-INCO002-003 VAIBHAV fellowship, INAE/DST-VF/2024/I/17 Seed Funding for Collaboration and Partnership Projects, IOE cell, IIT Bombay, RD/0523-IOE00I0-053 WRCB-HiMedia Entrepreneur-in-Residence fellowship, DO/2025-HIME004

The elastic properties of nanoscale extracellular vesicles (EVs) are believed to influence their cellular interactions, thus having a profound implication in intercellular communication. However, accurate quantification of their elastic modulus is challenging due to their nanoscale dimensions and their fluid-like lipid bilayer. We show that the previous attempts to develop atomic force microscopy-based protocol are flawed as they lack theoretical underpinning as well as ignore important contributions arising from the surface adhesion forces and membrane fluctuations. We develop a protocol comprising a theoretical framework, experimental technique, and statistical approach to accurately quantify the bending and elastic modulus of EVs. The method reveals that membrane fluctuations play a dominant role even for a single EV. The method is then applied to EVs derived from human embryonic kidney cells and their genetically engineered classes altering the tetraspanin expression. The data show a large spread; the area modulus is in the range of 4 to 19 mN/m and the bending modulus is in the range of 15 to 33 k B T , respectively. Surprisingly, data for a single EV, revealed by repeated measurements, also show a spread that is attributed to their compositionally heterogeneous fluid membrane and thermal effects. Our protocol uncovers the influence of membrane protein alterations on the elastic modulus of EVs.

Fluorescence-based single-molecule detection has been widely investigated and applied in biosensing and bioimaging due to its ultrahigh sensitivity and specificity. However, bulky and expensive commercial fluorescence microscopes are usually required. The Stokes shift property of most commonly used fluorophores requires optical sets such as dichroic mirrors and specific filters in the optical pathway before a photodetector to eliminate excitation and scattering lights from the fluorescence signals. The fluorescence signal collected by an objective is further unavoidably attenuated, and the optical resolution is diffraction-limited. Herein, a proof of concept of a lab-on-a-chip compatible molecular sensor is shown by integrating upconversion nanoparticles (UCNPs) and amorphous hydrogen-doped InGaZnO (InGaZnO:H) thin-film phototransistor (IGZO:H TFTs) aiming to alleviate those issues. Upon illumination with a 980 nm infrared light, the phototransistor shows no photocurrent without UCNPs but yields a high photocurrent with UV-visible fluorescent light emitted from the UCNPs. The molecular detection is enabled by further involving the Forster resonance energy transfer (FRET) mechanism, with the UCNPs as donors. The photocurrent falls back to its original low level when biotinylated gold nanoparticles are added to selectively bind and quench the UCNPs via biotin-streptavidin coupling. Each UCNP shows an estimated photocurrent-to-dark current ratio of 103 and each biotinylated gold nanoparticle causes at least 1 order of magnitude decrease of the photocurrent. Our integrated setup presents a promising platform for further development toward an optoelectronic biosensor capable of single-molecule detection.

Detection of analytes using streaming current has previously been explored using both experimental approaches and theoretical analyses of such data. However, further developments are needed for establishing a viable microchip that can be exploited to deliver a sensitive, robust, and scalable biosensor device. In this study, we demonstrated the fabrication of such a device on silicon wafer using a scalable silicon microfabrication technology followed by characterization and optimization of this sensor for detection of small extracellular vesicles (sEVs) with sizes in the range of 30 to 200 nm, as determined by nanoparticle tracking analyses. We showed that the sensitivity of the devices, assessed by a common protein-ligand pair and sEVs, significantly outperforms previous approaches using the same principle. Two versions of the microchips, denoted as enclosed and removable-top microchips, were developed and compared, aiming to discern the importance of high-pressure measurement versus easier and better surface preparation capacity. A custom-built chip manifold allowing easy interfacing with standard microfluidic connections was also constructed. By investigating different electrical, fluidic, morphological, and fluorescence measurements, we show that while the enclosed microchip with its robust glass-silicon bonding can withstand higher pressure and thus generate higher streaming current, the removable-top configuration offers several practical benefits, including easy surface preparation, uniform probe conjugation, and improvement in the limit of detection (LoD). We further compared two common surface functionalization strategies and showed that the developed microchip can achieve both high sensitivity for membrane protein profiling and low LoD for detection of sEV detection. At the optimum working condition, we demonstrated that the microchip could detect sEVs reaching an LoD of 104 sEVs/mL (when captured by membrane-sensing peptide (MSP) probes), which is among the lowest in the so far reported microchip-based methods.

Extracellular vesicles (EVs) are nanoparticles encapsulated with a lipid bilayer, and they constitute an excellent source of biomarkers for multiple diseases. However, the heterogeneity in their molecular compositions constitutes a major challenge for their recognition and profiling, thereby limiting their application as an effective biomarker. A single-EV analysis technique is crucial to both the discovery and the detection of EV subpopulations that carry disease-specific signatures. Herein, a plasmonic nanohole array is designed for capturing single EVs and subsequently performing fluorescence detection of their membrane proteins by exploiting plasmonic amplification of the fluorescence signal. Unlike other reported methods, our design relies on an exclusive detection of single EVs captured inside nanoholes, thus allowing us to study only plasmonic effects and avoid other metal-induced phenomena while leveraging on the proximity of emitters to the plasmonic hotspots. The method is optimized through numerical simulations and verified by a combination of atomic force, scanning electron microscopy, and fluorescence microscopy. Fluorescence enhancement is then estimated by measuring the CD9 expression of small EVs derived from the human embryonic kidney (HEK293) cell line and carefully considering the spatial distribution of emission and excitation intensities. Fluorescence intensities of immunostained EVs show a moderate overall enhancement of intensity and follow the intensity trend predicted by simulation for nanohole arrays with different nanohole periods. Moreover, the number of observed EVs in the best-performing nanohole array increases by more than 12 times compared with EVs immobilized on a reference substrate, uncovering a vast amount of weakly fluorescent EVs that would remain undetected with the regular fluorescent method. Our nanohole array provides a basis for a future platform of single-EV analyses, also promising to capture the signature arising from low-expressing proteins.

Elastic properties of nanoscale extracellular vesicles (EVs) are believed to influence their cellular interactions, thus having a profound implication in intercellular communication. Yet, an accurate quantification of the elasticity of such small lipid vesicles is difficult even with AFM-based nanoindentation experiments as it crucially depends on the reliability of the theoretical interpretation of such measurements. Here we describe a complete method composed of theoretical framework, experimental procedure, and appropriate statistical approach for an accurate determination of bending modulus and effective elastic modulus of EVs. Further, we experimentally demonstrate that the quantification of EVs by the elastic modulus from AFM-based force spectroscopy measurement is marred by the interplay of their compositionally inhomogeneous fluid membrane with the adhesion forces from the substrate and thermal effects - two exquisite phenomena that could thus far only be theoretically predicted. The effects result in a large spreading of elastic modulus even for a single EV. Our unified model is then applied to genetically engineered classes of EVs to understand how the alterations in tetraspanin expression may influence their elastic modulus.

Precision cancer medicine has changed the treatment landscape of non-small cell lung cancer (NSCLC) as illustrated by the introduction of tyrosine kinase inhibitors (TKIs) towards mutated epidermal growth factor receptor (EGFR). However, as responses to EGFR-TKIs are heterogenous among NSCLC patients, there is a need for ways to early monitor changes in treatment response in a non-invasive way e.g., in patient's blood samples. Recently, extracellular vesicles (EVs) have been identified as a source of tumor biomarkers which could improve on non-invasive liquid biopsy-based diagnosis of cancer. However, the heterogeneity in EVs is high. Putative biomarker candidates may be hidden in the differential expression of membrane proteins in a subset of EVs hard to identify using bulk techniques. Using a fluorescence-based approach, we demonstrate that a single-EV technique can detect alterations in EV surface protein profiles. We analyzed EVs isolated from an EGFR-mutant NSCLC cell line, which is refractory to EGFR-TKIs erlotinib and responsive to osimertinib, before and after treatment with these drugs and after cisplatin chemotherapy. We studied expression level of five proteins; two tetraspanins (CD9, CD81), and three markers of interest in lung cancer (EGFR, programmed death-ligand 1 (PD-L1), human epidermal growth factor receptor 2 (HER2)). The data reveal alterations induced by the osimertinib treatment compared to the other two treatments. These include the growth of the PD-L1/HER2-positive EV population, with the largest increase in vesicles exclusively expressing one of the two proteins. The expression level per EV decreased for these markers. On the other hand, both the TKIs had a similar effect on the EGFR-positive EV population.

A microfabricated biosensor based on the streaming current method is presented in this work. The microfabricated sensor consists of a silicon microchannel, which is enclosed with a glass capping to form a closed microchannel. The depth of the microchannel is approximately 10 µm in width and length varying from 50 to 100 µm. The silicon is etched using deep reactive ion etching (DRIE) to form a microchannel. For the capping of the channel, a glass wafer of type Borofloat is used and anodically bonded to the silicon wafer to form a closed microchannel. The microchannel is then functionalized to be specific to certain biomarkers which can be a potential biomarker for cancer, for example. The method used for detection is called the streaming current method. In this method, fluid is flown through the microchannel with high pressure close to six bars. The surface of the silicon is oxidized, which has a zeta potential of approximately 2.7. Depending on the type of fluid the charge concentration varies. By having a pressure in the channel, the charges get distributed as an anode and cathode at the inlet and outlet electrodes of the microfluidic channels. At a fixed potential, a streaming current is observed, which is proportional to the charge accumulated. The difference between the streaming current with and without the biomarker is correlated to the concentration. Hence, a biosensor based on the streaming current method can be realized, which could be used for potential cancer biomarker detection.

High heterogeneity in the membrane protein expression of small extracellular vesicles (sEVs) means that bulk methods relying on antibody-based capture for expression analysis have a drawback that each type of antibody may capture a different sub-population. An improved approach is to capture a representative sEV population, without any bias, and then perform a multiplexed protein expression analysis on this population. However, such a possibility has been largely limited to fluorescence-based methods. Here, we present a novel electrostatic labelling strategy and a microchip-based all-electric method for membrane protein analysis of sEVs. The method allows us to profile multiple surface proteins on the captured sEVs using alternating charge labels. It also permits the comparison of expression levels in different sEV-subtypes. The proof of concept was tested by capturing sEVs both non-specifically (unbiased) as well as via anti-CD9 capture probes (biased), and then profiling the expression levels of various surface proteins using the charge labelled antibodies. The method is the first of its kind, demonstrating an all-electrical and microchip based method that allows for unbiased analysis of sEV membrane protein expression, comparison of expression levels in different sEV subsets, and fractional estimation of different sEV sub-populations. These results were also validated in parallel using a single-sEV fluorescence technique.

Fluorescence-based optical sensing techniques have continually been explored for single-molecule detection targeting myriad biomedical applications. Improving signal-to-noise ratio remains a prioritized effort to enable unambiguous detection at single-molecule level. Here, we report a systematic simulation-assisted optimization of plasmon-enhanced fluorescence of single quantum dots based on nanohole arrays in ultrathin aluminum films. The simulation is first calibrated by referring to the measured transmittance in nanohole arrays and subsequently used for guiding their design. With an optimized combination of nanohole diameter and depth, the variation of the square of simulated average volumetric electric field enhancement agrees excellently with that of experimental photoluminescence enhancement over a large range of nanohole periods. A maximum 5-fold photoluminescence enhancement is statistically achieved experimentally for the single quantum dots immobilized at the bottom of simulation-optimized nanoholes in comparison to those cast-deposited on bare glass substrate. Hence, boosting photoluminescence with optimized nanohole arrays holds promises for single-fluorophore-based biosensing.

Detection of analyte using streaming current has been previously explored using both experimental and theoretical approaches. However, little has been done to develop a viable microchip which can be exploited to deliver a sensitive, robust, and scalable biosensor device. In this study, we demonstrate the fabrication of such a device on silicon wafer using a scalable silicon microfabrication technology followed by their characterization and optimization for highly sensitive detection of small extracellular vesicles (sEVs). We show that the sensitivity of the devices, estimated using a common protein-ligand pair and sEVs significantly outperforms previous reports using the same principle. Two versions of the microchips, denoted as enclosed and open-top microchip, were developed and compared aiming to discern the importance of high-pressure measurement vs easier and better surface preparation capacity. A custom-built chip-manifold allowing easy interfacing with standard microfluidic connections was also developed. By investigating different electrical, fluidic, morphological, and fluorescence measurements, we show that while the enclosed microchip with its robust glass-silicon bonding can withstand higher pressure and thus generate higher streaming current, the open-top configuration offers several practical benefits including easy surface preparation, uniform probe conjugation, and improvement in the limit of detection (LoD). We further compare two common surface functionalization strategies and show that the proposed microchip can achieve both high sensitivity for membrane protein profiling and low LoD for sEV detection. At the optimum condition, we demonstrate that the microchip can detect sEVs reaching a LoD of 10 4 sEV/mL, which is among the lowest in the reported microchip-based methods.

The MeMC is an open-source software package for monte-carlo simulation of elastic shells. It is designed as a tool to interpret the force-distance data generated by indentation of biological nano-vesicles by atomic force microscopes. The code is written in c++ and python. The code is customizable – new modules can be added in a straightforward manner.

We present an approach to improve the detection sensitivity of a streaming current-based biosensor for membrane protein profiling of small extracellular vesicles (sEVs). The experimental approach, supported by theoretical investigation, exploits electrostatic charge contrast between the sensor surface and target analytes to enhance the detection sensitivity. We first demonstrate the feasibility of the approach using different chemical functionalization schemes to modulate the zeta potential of the sensor surface in a range -16.0 to -32.8 mV. Thereafter, we examine the sensitivity of the sensor surface across this range of zeta potential to determine the optimal functionalization scheme. The limit of detection (LOD) varied by 2 orders of magnitude across this range, reaching a value of 4.9 × 106 particles/mL for the best performing surface for CD9. We then used the optimized surface to profile CD9, EGFR, and PD-L1 surface proteins of sEVs derived from non-small cell lung cancer (NSCLC) cell-line H1975, before and after treatment with EGFR tyrosine kinase inhibitors, as well as sEVs derived from pleural effusion fluid of NSCLC adenocarcinoma patients. Our results show the feasibility to monitor CD9, EGFR, and PD-L1 expression on the sEV surface, illustrating a good prospect of the method for clinical application.

An electrical immuno-sandwich assay utilizing an electrokinetic-based streaming current method for signal transduction is proposed. The method records the changes in streaming current, first when a target molecule binds to the capture probes immobilized on the inner surface of a silica micro-capillary, and then when the detection probes interact with the bound target molecules on the surface. The difference in signals in these two steps constitute the response of the assay, which offers better target selectivity and a linear concentration dependent response for a target concentration within the range 0.2-100 nM. The proof of concept is demonstrated by detecting different concentrations of Immunoglobulin G (IgG) in both phosphate buffered saline (PBS) and spiked in E. coli cell lysate. A superior target specificity for the sandwich assay compared to the corresponding direct assay is demonstrated along with a limit of detection of 90 pM in PBS. The prospect of improving the detection sensitivity was theoretically analysed, which indicated that the charge contrast between the target and the detection probe plays a crucial role in determining the signal. This aspect was then experimentally validated by modulating the zeta potential of the detection probe by conjugating negatively charged DNA oligonucleotides. The length of the conjugated DNA was varied from 5 to 30 nucleotides, altering the zeta potential of the detection probe from -9.3 +/- 0.8 mV to -20.1 +/- 0.9 mV. The measurements showed a clear and consistent enhancement of detection signal as a function of DNA lengths. The results presented here conclusively demonstrate the role of electric charge in detection sensitivity as well as the prospect for further improvement. The study therefore is a step forward in developing highly selective and sensitive electrokinetic assays for possible application in clinical investigations.

Liquid biopsies based on extracellular vesicles (EVs) represent a promising tool for treatment monitoring of tumors, including non-small-cell lung cancers (NSCLC). In this study, we report on a multiplexed electrokinetic sensor for surface protein profiling of EVs from clinical samples. The method detects the difference in the streaming current generated by EV binding to the surface of a functionalized microcapillary, thereby estimating the expression level of a marker. Using multiple microchannels functionalized with different antibodies in a parallel fluidic connection, we first demonstrate the capacity for simultaneous detection of multiple surface markers in small EVs (sEVs) from NSCLC cells. To investigate the prospects of liquid biopsies based on EVs, we then apply the method to profile sEVs isolated from the pleural effusion (PE) fluids of five NSCLC patients with different genomic alterations (ALK, KRAS or EGFR) and applied treatments (chemotherapy, EGFR- or ALK-tyrosine kinase inhibitors). The vesicles were targeted against CD9, as well as EGFR and PD-L1, two treatment targets in NSCLC. The electrokinetic signals show detection of these markers on sEVs, highlighting distinct interpatient differences, e.g., increased EGFR levels in sEVs from a patient with EGFR mutation as compared to an ALK-fusion one. The sensors also detect differences in PD-L1 expressions. The analysis of sEVs from a patient prior and post ALK-TKI crizotinib treatment reveals significant increases in the expressions of some markers (EGFR and PD-L1). These results hold promise for the application of the method for tumor treatment monitoring based on sEVs from patient liquid biopsies.

Being a key player in intercellular communications, nanoscale extracellular vesicles (EVs) offer unique opportunities for both diagnostics and therapeutics. However, their cellular origin and functional identity remain elusive due to the high heterogeneity in their molecular and physical features. Here, for the first time, multiple EV parameters involving membrane protein composition, size and mechanical properties on single small EVs (sEVs) are simultaneously studied by combined fluorescence and atomic force microscopy. Furthermore, their correlation and heterogeneity in different cellular sources are investigated. The study, performed on sEVs derived from Human Embryonic Kidney 293, Cord Blood Mesenchymal Stromal and Human Acute Monocytic Leukemia cell lines, identifies both common and cell line-specific sEV subpopulations bearing distinct distributions of the common tetraspanins (CD9, CD63 and CD81) and biophysical properties. Although the tetraspanin abundances of individual sEVs are independent of their sizes, the expression levels of CD9 and CD63 are strongly correlated. A sEV population co-expressing all the three tetraspanins in relatively high abundance, however, having on average diameters <100 nm and relatively low Young moduli, is also found in all cell lines. Such a multiparametric approach is expected to provide new insights regarding EV biology and functions, potentially deciphering unsolved questions in this field.

Nanosized extracellular vesicles (EVs) have been found to play a key role in intercellular communication, offering opportunities for both disease diagnostics and therapeutics. However, lying below the diffraction limit and also being highly heterogeneous in their size, morphology and abundance, these vesicles pose significant challenges for physical characterization. Here, we present a direct visual approach for their accurate morphological and size-based profiling by using scanning electron microscopy (SEM). To achieve that, we methodically examined various process steps and developed a protocol to improve the throughput, conformity and image quality while preserving the shape of EVs. The study was performed with small EVs (sEVs) isolated from a non-small-cell lung cancer (NSCLC) cell line as well as from human serum, and the results were compared with those obtained from nanoparticle tracking analysis (NTA). While the comparison of the sEV size distributions showed good agreement between the two methods for large sEVs (diameter > 70 nm), the microscopy based approach showed a better capacity for analyses of smaller vesicles, with higher sEV counts compared to NTA. In addition, we demonstrated the possibility of identifying non-EV particles based on size and morphological features. The study also showed process steps that can generate artifacts bearing resemblance with sEVs. The results therefore present a simple way to use a widely available microscopy tool for accurate and high throughput physical characterization of EVs.

Electrokinetic principles such as streaming current and streaming potential are extensively used for surface characterization. Recently, they have also been used in biosensors, resulting in enhanced sensitivity and simpler device architecture. Theoretical models regarding streaming current/potential studies of particle-covered surfaces have identified features such as the particle size, shape and surface charge to influence the electrokinetic signals and consequently, the sensitivity and effective operational regime of the biosensor. By using a set of well-characterized proteins with varying size and net surface charge, this article experimentally verifies the theoretical predictions about their influence on the sensor signal. Increasing protein size was shown to enhance the signal when their net surface charge was either opposite to that of the sensor surface, or close to zero, in agreement with the theoretical predictions. However, the effect gradually saturated as the protein size exceeded the coulomb screening length of the electrolyte. In contrast, the proteins containing the same type of charge as the surface showed little or no difference, except that the signal inverted. The magnitude of the surface charge was also shown to influence the signal. The sensitivity of the technique for protein detection varied over two orders of magnitude, depending upon the size and surface charge. Furthermore, the capacity of the electrokinetic method for direct electrical detection of various proteins, including those carrying little or no net electric charges, is demonstrated.

Small extracellular vesicles (sEVs) generated from the endolysosomal system, often referred to as exosomes, have attracted interest as a suitable biomarker for cancer diagnostics, as they carry valuable biological information and reflect their cells of origin. Herein, we propose a simple and inexpensive electrical method for label-free detection and profiling of sEVs in the size range of exosomes. The detection method is based on the electrokinetic principle, where the change in the streaming current is monitored as the surface markers of the sEVs interact with the affinity reagents immobilized on the inner surface of a silica microcapillary. As a proof-of-concept, we detected sEVs derived from the non-small-cell lung cancer (NSCLC) cell line H1975 for a set of representative surface markers, such as epidermal growth factor receptor (EGFR), CD9, and CD63. The detection sensitivity was estimated to be ∼175000 sEVs, which represents a sensor surface coverage of only 0.04%. We further validated the ability of the sensor to measure the expression level of a membrane protein by using sEVs displaying artificially altered expressions of EGFR and CD63, which were derived from NSCLC and human embryonic kidney (HEK) 293T cells, respectively. The analysis revealed that the changes in EGFR and CD63 expressions in sEVs can be detected with a sensitivity in the order of 10% and 3%, respectively, of their parental cell expressions. The method can be easily parallelized and combined with existing microfluidic-based EV isolation technologies, allowing for rapid detection and monitoring of sEVs for cancer diagnosis.