Tumor extracellular vesicle-based protease activity assay for disease detection and treatment monitoring

Abstract

A method of assaying for activity of a protease enzyme corresponding to a disease in a subject includes selectively capturing an extracellular vesicle (EV) from a sample from the subject; contacting the EV after the capturing with a probe molecule, the probe molecule including a peptide, a fluorescent moiety on a first end of the peptide and a quenching moiety on a second end of the peptide to provide a Fluorescence Resonance Energy Transfer (FRET) pair separated by the peptide, the peptide containing a cleavage sequence for the protease enzyme; and measuring a fluorescence of at least the fluorescent moiety after the contacting to indicate a presence or absence of the occurrence of cleavage of the peptide by the protease enzyme.

Description

Tumor Extracellular Vesicle-Based Protease Activity Assay for Disease Detection and Treatment MonitoringCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present patent application claims priority benefit to U.S. Provisional Patent Application No. 63 / 546,684, filed on October 31, 2023, the entire content of which is incorporated herein by reference. All references cited anywhere in this specification, including the Background and Detailed Description sections, are incorporated by reference as if each had been individually incorporated.FEDERAL FUNDING

[0002] This invention was made with government support under CA255727, and CA253651 awarded by the National Institutes of Health. The government has certain rights in the invention.BACKGROUND1. Technical Field

[0003] The currently claimed embodiments of the present invention relate to methods and kits for assaying, and more particularly to methods and kits for assaying for activity of a protease enzyme corresponding to a disease in a subject.2. Discussion of Related Art

[0004] Proteases have a crucial role in the progression and spread of cancer. They are involved in signaling pathways1’2and the degradation of extracellular matrixes (ECM).3-4Dysregulated protease function can contribute to abnormal growth and survival of cancercells, as well as angiogenesis signaling in tumors.2’5The proteolytic activity of proteases facilitates the invasion and metastasis of cancer cells by locally and distantly degrading ECM.6’7Given their significant roles in cancer pathology, proteases are considered as important targets for both cancer treatment and detection.Current methods for detecting protease and measuring protease activities

[0005] Direct quantification of proteases. The abundance of proteases is commonly quantified at the transcriptomic level using reverse transcription polymerase chain reaction (RT- PCR)8-9and atthe proteomic level through western blot analysis.8-10Similarly, techniques such as in situ hybridization (isH)11-12and immunohistochemistry (IHC)13 14allow for direct visualization of the protease levels and their distribution within tumor tissue. However, it is important to note that measuring protease abundance may not fully reflect their functional behavior in diseases, as protease activity can be inhibited by endogenous inhibitors.15-16

[0006] Zymography. To address the limitation of direct quantification, various methods have been developed to measure protease activity based on their proteolytic function.17In gel zymography (1GZ) is a conventional approach for evaluating protease activity.18-19In this method, proteins extracted from homogenized specimens are separated using SDS-PAGE in a gelatincontaining gel. The enzymes are then renatured to allow for the degradation of gelatin substrate, and the degraded gel is visualized through staining.20-21In spite of functional assessment, gelatin-based zymography is primarily limited to proteases capable of degrading gelatin, such as matrix metalloproteinase (MMP) 2 and 9, as well as certain cathepsins (CTSs).22, 23Additionally, IGZ is a time- consuming process (>24 h of procedure) and there may be concerns regarding the loss of protease activity during the renaturation step.23-25There thus remains a need for improved methods and kits for assaying for activity of a protease enzyme corresponding to a disease in a subject.SUMMARY

[0007] A method of assaying for activity of a protease enzyme corresponding to a disease in a subject according to an embodiment of the current invention includes selectively capturing an extracellular vesicle (EV) from a sample from the subject; contacting the EV after the capturing with a probe molecule, tire probe molecule including a peptide, a fluorescent moiety on a first end of the peptide and a quenching moiety on a second end of the peptide to provide a Fluorescence ResonanceEnergy Transfer (FRET) pair separated by the peptide, the peptide containing a cleavage sequence for the protease enzyme; and measuring a fluorescence of at least the fluorescent moiety after the contacting to indicate a presence or absence of the occurrence of cleavage of the peptide by the protease enzyme.

[0008] A kit for assaying for activity of a protease enzyme corresponding to a disease in a subject according to an embodiment of the current invention includes at least one of a structure substrate or a plurality of magnetic particles; a first solution comprising a plurality of binding moieties, each binding moiety functionalized to attach to the at least one of the structure substrate or a magnetic particle of the plurality of magnetic particles; a second solution comprising a plurality of binding moieties, each binding moiety functionalized with a selected antibody to attach to an extracellular vesicle (EV) of interest; and a third solution comprising a plurality of probe molecules, each probe molecule comprising a peptide, a fluorescent moiety on a first end of the peptide and a quenching moiety on a second end of the peptide to provide a Fluorescence Resonance Energy Transfer (FRET) pair separated by the peptide, the peptide containing a cleavage sequence for a protease enzyme of interest.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present disclosure, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description and the appended claims with reference to the accompanying drawings, all of which form a part of this specification, wherein like reference numerals designate corresponding parts in the various figures. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the invention.

[0010] FIG. 1 is a conceptual illustration of the working mechanism of the OS EV Protease Activity Assay according to an embodiment of the present invention. Each OS EV protease subpopulation is identified via the colocalization of an OS-specific marker (by capturing) and OS-associated active protease (by quantification), enabling unique advantages of this technology platform: sensitive, specific, quantitative, and accessible.

[0011] FIG. 2 is a schematic illustration of the overall workflow of the OS EV Protease Activity Assay according to an embodiment of the present invention, i) EV Click MagBeads for OS EV enrichment. EV Click MagBeads enable OS EV enrichment through biorthogonal Click Chemistry. Trans-cyclooctene (TCO)-modified antibodies targeting three OS-specific markers (i.e., B7H3, GPNMB, and LRRC15) are used to label the EVs. These TCO-labeled OS EVs are then mixed with the EV Click MagBeads, magnetic beads that have been covalently functionalized with methyltetrazine (mTz). The resulting Click Chemistry reaction between the mTz-grafted EV Click MagBeads and the TCO-grafted OS EVs immobilizes the OS EVs on the bead surface. The OS EV-immobilized magnetic beads are then magnetically isolated, resulting in enrichment of OS EVs. ii) FRET- based peptide cleavage assay and downstream analysis. For membrane-bound protease, MMP-14, the enriched OS EVs on the magnetic beads were directly subjected to the FRET-based peptide cleavage assay. For the inter-EV protease, MMP-2 / 9, the enriched OS EVs were lysed. The lysed solutions are incubated with FRET peptide substrates, which generate a fluorescence readout via the enzymatic reaction of OS-associated active proteases (i.e., MMP-14 and MMP-2 / 9). The resulting OS EV Protease Activity Signatures are compiled to generate OS EV Protease Activity Scores for differentiating metastatic from localized OS and detection of disease progression.

[0012] FIGS. 3A-3B provide: a) A schematic summary of stepwise functional group transformation employed for the preparation of EV Click MagBeads and TCO-labeled EVs. b) The quality and the lifetime of EV Click MagBeads and TCO-Ab are evaluated using fluorescent probes labeled with a click chemistry motif (i.e., TCO-Cy5 and mTz-Cy5) as indicators of their functional counterparts.

[0013] FIGS. 4A-4B show: a) Representative scanning electron microscopy (SEM) images (scale bar, 500 nm) of EV Click MagBeads with immobilized HOS EVs in the presence of TCO-anti-B7H3. b) A representative TEM image (scale bar, 100 nm) of HOS EVs immobilized on a Click MagBead after immunogold labelling of CD63, a representative EV marker.

[0014] FIGS. 5A-5C show: a) Michaelis-Menten curve of MMP-14 based on FRET peptide substrate, b) MMP-14 activity of pure active MMP-14 (rhMMP-14). c) MMP-14 activity OS EVs in artificial plasma sample, captured by anti-GPNMB as OS- specific marker.

[0015] FIGS. 6A-6D show: a) Conceptual illustration of the working mechanism of OS EV MMP-14 Activity Assay, where OS-specific anti-GPNMB was used to achieve click chemistry-mediated capture of OS EVs, followed by FRET -based peptide cleavage assay to quantify MMP-14 activity, b) Box plots representing the fluorescence readouts of OS EV MMP-14 Activity Assay observed for healthy donors (HD, n=16), localized (n = 26) and metastatic (n = 4) OS patients. Significant differences between different groups were evaluated using one-way ANOVA. c) ROC curves for OS EV MMP-14 Activity Assay, including HD versus all OS patients (i.e., localized + metastatic) (AUC = 0.92), HD versus localized OS (AUC = 0.91), and localized versus metastatic (AUC = 0.82). c) Box plots showing that there were significant background signals in plasma sample, thus no significant difference was observed between HD and OS patients.

[0016] FIGS. 7A-7C are for OS EV MMP-2 / 9 Activity Assay based on the click chemistry-mediated EV enrichment targeting 3 OS-specific markers (i.e., B7H3, GPNMB, and LRRC15). a) OS EV MMP-2 / 9 Activity Assay observed for GPNMB(+)-OS EVs. ROC curves for HD versus all OS patients (i.e., localized + metastatic) (AUC = 0.90), HD versus localized OS (AUC = 0.88), and localized versus metastatic (AUC = 0.66). b) OS EV MMP- 2 / 9 Activity Assay observed for B7H3(+)-OS EVs. ROC curves for HD versus all OS patients (i.e., localized + metastatic) (AUC = 0.87), HD versus localized OS (AUC = 0.88), and localized versus metastatic (AUC = 0.61). c) OS EV MMP- 2 / 9 Activity Assay observed for LRRC15(+)-OS EVs. ROC curves for HD versus all OS patients (i.e., localized + metastatic) (AUC = 0.99), HD versus localized OS (AUC = 0.99), and localized versus metastatic (AUC = 0.87).

[0017] FIGS. 8A-8B show clinical significance of OS EV Protease Activity Assay. A plot of MMP-14 activity (RFU) measured from GPNMB+ OS EVs with time of the assay depicting the disease progression of OS patients, a) For an OS patient OST-003, Magneticresonance (MR) images were taken before and after cycle 1-2 ADM / DDP plus cycle 1 MTX chemotherapy treatment. High activity of MMP 14 activity was maintained throughout the treatment, which indicated disease progression, b) For an OS patient OST-005, MR. images were taken before and after cycle 1 ADM / DDP chemotherapy treatment. The level of activity of MMP14 was significantly increased after treatment, which indicated disease progression.

[0018] FIG. 9 shows overall workflow of the OS EV MMP Activity Assay according to an embodiment pf the current invention.

[0019] FIGS. 10A-10E shows an example of characterization and validation of OS EVs with the workflow of FIG. 9.

[0020] FIG. 11 shows clinical validation of OS EV MMP Activity Assay for detecting OS and metastatic status with the workflow of FIG. 9.

[0021] FIG. 12 shows clinical validation of OS EV MMP Activity Assay for detecting OS and metastatic status with the workflow of FIG. 9.

[0022] FIG. 13 shows dynamic performance of OS EV MMP Activity Assay for monitoring OS with the workflow of FIG. 9.

[0023] FIG. 14 shows dynamic performance of OS EV MMP Activity Assay for monitoring OS with the workflow of FIG. 9.

[0024] FIG. 15 provides a schematic illustration of a streamlined workflow developed for the two-step OS EV MMP Activity Assay according to an embodiment of the current invention. Step 1 : Click chemistry-mediated enrichment of three subpopulations of OS EVs using EV Click MagBeads. After labeling OS EVs in 250 pL of plasma by one of the transcyclooctene (TCO)-grafted antibodies targeting the respective OS EV surface marker (i.e., TCO-anti-LRRC15, TCO-anti-GPNMB, or TCO-anti-B7-H3), EV Click MagBeads functionalized with methyltetrazine (mTz) were employed to immobilize the respective subpopulation of OS EVs. Step 2: A FRET peptide probe for assessing the proteolytic activity of MMP14 or MMP2 of each subpopulation of OS EVs. To assess the MMP activities, a subpopulation of OS EVs enriched on EV Click MagBeads was lysed andincubated with a corresponding FRET peptide probe. Among the OS EV MMP Activity profiles assessed across six OS EV marker / MMP combinations, the three top-performing — LRRC15 / MMP2, GPNMB / MMP 14, and B7-H3 / MMP14 OS EVs— were selected to establish OS EV MMP Activity Scores via logistic regression. These Scores were utilized to monitor OS patient’s disease progression and treatment responses.

[0025] FIG. 16 shows an integrated EV biomarker selection framework employed for identifying the three OS EV surface markers and two OS EV-associated MMPs according to an embodiment of the current invention. This framework started from selecting OS-specific markers from two datasets: i) cell line protein (CCLE) dataset and ii) OS tissue RNA dataset, followed by excluding housekeeping and hematopoietic proteins. EV-specific markers were then identified through Vesiclepedia (Number of EV studies > 2). Refinement of OS EV surface markers included: i) selecting the top 20% OS proteins from CCLE, ii) filtering the top 2.5% RNA levels in CCLE and TARGET-OS, and iii) selecting those with low expression in hematopoietic lineage (DMAP < 7). Within these OS EV marker pools, CD276 and MMP14 were added from the caGESP list of sarcomas, resulting in 62 OS EV marker candidates. For the selection of OS EV surface markers, five proteins were identified as cancer surface proteins from the Cancer Surfacesome Atlas (TCSA), and those which have been used for targeted therapies (i.e., mAb, ADC, and CAR-T) were further selected, yielding three OS EV surface markers: LRRC15, GPNMB, and CD276 (B7-H3). For OS EV- associated proteases, seven proteins exhibiting protease activity were identified, and MMP 14 and MMP2 were selected as representatives of membrane-type and soluble MMPs, respectively, with i) the highest plasma level and ii) commercially available FRET peptide probes. DEG: Differentially Expressed Genes; caGESP: Cancer-specific Genes Encoding Surface Proteins, DMAP: The Differentiation Map.

[0026] FIG. 17A-17C show validation of three OS EV surface markers and two OS EV- associated MMPs using OS tissue microarray (TMA) according to an embodiment of the current invention. (A) Representative hematoxylin and eosin (H&E) and immunohistochemistry (1HC) images of the three OS EV surface markers and two OS EV- associated MMPs on OS TMA slides. Scale bar = 100 pm. (B and C) Percentage of OSsamples categorized by IHC staining intensity of strong (3+), moderate (2+), weak (1+), and negative (0) for each of the (B) three OS EV surface markers and (C) two OS EV-associated MMPs were summarized in both pie charts and bar charts. The percentage of IHC staining for each OS EV surface marker and their combinations, and each MMP and their combination were summarized in the bar chart.

[0027] FIGS. 18A-18G provide an example of validation of OS EV MMP Activity Assay using synthetic plasma samples. (A) A schematic illustration of the workflow developed for assessing the performance of OS EV MMP Activity Assay. Synthetic plasma samples were prepared by spiking HOS cell-derived EVs into the EV-depleted healthy donor (HD) plasma. The OS EVs were enriched using EV Click MagBeads in conjunction with TCO-grafted antibodies targeting the three OS EV surface markers (LRRC15, GPNMB, or B7-H3). To obtain the activity readouts of membrane-type MMP14, the enriched OS EVs on EV Click MagBeads were directly incubated with a MMP14 FRET probe. In contrast, for the activity of intravesicular MMP2, the enriched OS EVs were lysed, followed by incubation with a MMP2 FRET probe. (B-D) Dynamic linearity ranges of MMP14 activity observed for the OS EVs enriched by TCO-grafted (B) anti-LRRC15, (C) anti-GPNMB, and (D) anti-B7- H3. (E-G) Dynamic linearity ranges of MMP2 activity observed for the OS EVs enriched by TCO-grafted (E) anti-LRRC15, (F) anti-GPNMB, and (G) anti-B7-H3.

[0028] FIGS. 19A-19H provides evaluation of OS EV MMP Activity Assay and establishment of OS EV MMP Activity Scores using clinical samples. (A) A general workflow for the OS EV MMP Activity Assay using plasma samples, collected from three groups: healthy donors (HD, n=34), patients with localized OS (n=24), and patients with metastatic OS (n=10). (B) Heatmaps summarizing quantitative MMP activity profiles from six OS marker / MMP combinations — LRRC15 / MMP14, GPNMB / MMP14, B7-H3 / MMP14, LRRC15 / MMP2, GPNMB / MMP2, and B7-H3 / MMP2 OS EVs— in plasma samples. The three top-performing combinations (LRRC 15 / MMP2, GPNMB / MMP14, and B7- H3 / MMP14 OS EVs, labeled as *) were used to establish OS EV MMP Activity Scores. (C and D) OS EV MMP Activity profiles of the three top-performing combinations (LRRC 15 / MMP2, GPNMB / MMP14, and B7-H3 / MMP14 OS EVs) show significantlyhigher (P < 0.001) MMP activities in (C) OS patients compared with HDs and (D) metastatic (MET) OS compared to localized (LOC) OS patients. Their corresponding Area Under Receiver Operating Characteristic (AUROC) curves demonstrate robust diagnostic performance for (C) separating OS from HDs and (D) distinguishing metastatic OS from localized OS. (E) The OS EV MMP Activity Score was established using logistic regression of the three top-performing combinations for the enhanced performance to distinguish metastatic OS from localized OS. (F) Score performance, (G) heatmap, and (H) ROC curve of OS EV MMP Activity Score. Student’ s unpaired t-test was used for statistical analysis between two groups.

[0029] FIGS. 20A-20C show performance of OS EV MMP Activity Assay for dynamic monitoring of disease progression and treatment response of OS patients. (A) Workflow for assessing OS EV MMP activity in longitudinally collected plasma samples from OS patients under therapeutic intervention. The resulting OS EV MMP Activity Scores at each time point was plotted, and the Score change was compared with radiographic imaging of each patient to evaluate their clinical correlation. (B) Stable Scores observed post-local control surgery in a localized OS patient (OST-005). (C) Rising Scores post-amputation, indicating rapid recurrence in a metastatic OS patient (OST-003), along with computed tomography (CT) imaging for disease monitoring. The swimmer plots for each patient are shown as individual bars, with blood draw points marked as circles (white for Scores below 0, red for Scores above 0). Clinical events from the start of treatment are depicted as triangles along the bar: blue triangles above represent radiographic imaging, while red triangles below represent surgical events (e.g., local control, amputation). Blue bars denote types and period of chemotherapy, including MAP (Methotrexate / Doxorubicin / Cisplatin; light blue) and ICE (Ifosfamide / Carboplatin / Etoposide; blue) treatments.

[0030] FIG. 21 shows validation of three OS EV surface markers and two OS EV- associated MMPs using OS cell line. Immunofluorescence micrographs of HOS cells, representative OS cell line, showing the colocalization of each of the three OS EV surface markers (LRRC15, GPNMB, and B7-H3; shown in orange) on the plasma membranes,membrane-type MMP14 (in green), and intravesicular MMP2 (in red). Cell nuclei were stained with DAPI (blue). Scale bar = 10 pm.

[0031] FIG. 22A-22E shows characterization of EV Click MagBeads. (A) Stepwise preparation of EV Click MagBeads. (B) Zeta potentials of amine-grafted Dynabeads™ (NH2-Dynabeads), mTz-modified Dynabeads™ (mTz-Dynabeads), and EV Click MagBeads, indicating the functional group transformations on the MagBeads surfaces. (C) Fluorescence micrographs of EV Click MagBeads before and after labeling with TCO-Cy5 (scale bar = 10 pm) and (D) corresponding histograms of mean fluorescence intensity (MFI). (E) Lifetime of the surface mTz group on EV Click MagBeads.

[0032] FIGS. 23A-23D show characterization of OS EVs. (A-B) Size distribution of HOS cell EVs measured by (A) dynamic light scattering (DLS) and (B) nanoparticle tracking analysis (NT A). (C-D) Representative images of HOS EVs obtained by (C) transmission electron microscopy (TEM; scale bar = 200 nm) and (D) scanning electron microscopy (SEM; scale bar = 500 nm).

[0033] FIGS. 24A-24B show validation of EV Click MagBeads with OS EVs. (A) Representative SEM images (scale bar = 500 nm) of EV Click MagBeads with immobilized HOS EVs using TCO-anti-B7-H3. Inset depicts how TCO-labeled EVs are immobilized onto an EV Click MagBead. (B) Representative TEM images (scale bar = 100 nm) of HOS EVs immobilized on EV Click MagBeads followed by immunogold (nanogold) labelling of CD63, a representative EV surface marker.

[0034] FIGS. 25A-25E show optimization of the MMP14 activity assay. (A) Schematic illustration of the MMP14 activity assay using activated recombinant human MMP14 (rhMMP14). (B) Initial rate of hydrolysis of the MMP14 FRET probe by rhMMP14 (20 nM) and (C) the corresponding Michaelis-Menten plot for determining the kinetic parameters (Vmax = 5.55 ARFU / s, Km = 4.54 pM) of MMP14. (D) Dynamic range of the MMP14 activity assay. (E) Verification of the specificity of the MMP14 FRET probe. Activities of MMP14 and MMP2 were assayed with the MMP14 FRET probe, and the signals were normalized against that of MMP14.

[0035] FIGS. 26A-26E show optimization of the MMP2 activity assay. (A) Schematic illustration of the MMP2 activity assay using activated recombinant human MMP2 (rhMMP2). (B) Initial rate of hydrolysis of the MMP2 FRET probe by rhMMP2 (1 .3 nM) and (C) corresponding Michaelis-Menten plots for determining the kinetic parameters of MMP2 (Vmax = 3.93 ARFU / s, Km = 11.5 pM). (D) Dynamic range of the MMP2 activity assay. (E) Verification of the specificity of the MMP2 FRET probe. Activities of MMP2 and MMP14 were assayed with the MMP2 FRET probe, and the signals were normalized against that of MMP2.

[0036] FIGS. 27A-27B show significance of EV enrichment. To evaluate the significance of EV enrichment, MMP activities in whole plasma and enriched OS EVs were compared. (A) MMP 14 and MMP2 activities in whole plasma from healthy donors (HD, n = 6) and OS patients (n = 6) showed no significant difference (P > 0.05). (B) However, MMP activities of OS EVs from the same plasma samples, enriched using EV Click MagBeads in conjunction with TCO-anti-B7-H3, significantly separated (P < 0.001) OS patients from HDs.

[0037] FIG. 28 shows clinical validation of the OS EV MMP Activity Assay for detecting OS. The MMP14 and MMP2 activity readouts were analyzed using the OS EV MMP Activity Assay on plasma samples collected from healthy donors (HD, n = 34) and OS patients at all stages (OS, n = 34). Significantly higher (P < 0.001) MMP14 and MMP2 activities were observed in all three subpopulations of OS EVs (LRRC15+, GPNMB+, and B7-H3+ OS EVs) in plasma samples from OS patients compared to those from heathy donors. Corresponding AUROC also showed remarkable separation (AUROC > 0.85) of OS patients from HDs. Statistical analysis was performed using Student’s unpaired t-test.

[0038] FIG. 29 shows clinical validation of OS EV MMP Activity Assay for distinguishing healthy donors, localized OS patients, and metastatic OS patients. The MMP 14 and MMP2 activity profiles were analyzed using the OS EV MMP Activity Assay on plasma samples collected from healthy donors (HD, n = 34), patients with localized OS (LOC, n = 24), and patients with metastatic OS (MET, n = 10). Significantly higher (P < 0.001) MMP14 activities were observed in all three subpopulations of OS EVs (LRRC15+,GPNMB+, and B7-H3+ OS EVs) in plasma samples from metastatic patients compared to those from localized patients. Distinguishing between metastatic and OS localized patients based on MMP2 activity was only evident in LRRC15+ EVs. Statistical analysis was performed using one-way ANOVA and Tukey's multiple comparison analysis, ns = not significant (P > 0.05), ***p < 0.001, ****p < 0.0001.

[0039] FIGS. 30A-30D show dynamic performance of OS EV MMP Activity Assay for monitoring OS. Plots illustrating OS EV MMP Activity Scores, generated from MMP activity profiles of the three top-performing OS marker / MMP combinations (LRRC15 / MMP2, GPNMB / MMP14, and B7-H3 / MMP2 OS EVs). (A-B) The Scores were observed to drop over the therapeutic intervention in a localized OS patient (A, OST-OOl) and a resolved metastatic patient (B, OST-006). (C-D) Increasing or very high (>5) scores were observed in metastatic OS patients (C, OST-004; D, OST-008), aligning with routine surveillance computed tomography (CT) imaging.

[0040] FIG. 31 provides a swimmer plot showing therapeutic interventions for six OS patients (Localized: OST-1, 5, 6; Metastatic: OST-3, 4, 8). Each bar represents an individual patient, with circles denoting blood draw points: red for OS scores > 0 (higher chance of metastasis) and white for OS scores < 0 (lower chance of metastasis). Clinical events from the start of pre-treatment are indicated by triangles above the bars — blue for radiographic imaging and red for surgery (e.g., local control, amputation). Types and durations of chemotherapy are shown as blue bars: MAP (sky blue), ICE (blue), and Gem / Pax (deep blue). MAP, Methotrexate / Doxorubicin / Cisplatin; ICE, Ifosfamide / Carboplatin / Etoposide; Gem / Pax, Gemcitabine / Paclitaxel; VATS, Video-assisted thoracic surgery.

[0041] FIG. 32 provides Table 1. Clinical characteristics of the healthy donors.

[0042] FIG. 33 provides Table 2. Clinical characteristics of the osteosarcoma patients.

[0043] FIG. 34 provides Table 3. Clinical characteristics of the osteosarcoma patients with disease monitoring.

[0044] FIGS. 35A-35B show A) A molecular hallmark of ADNPC: Amyloid-P plaque accumulation. P-secretase drives the formation of amyloid-P plaques, a molecular hallmark of ADNPC, by initiating an abnormal cleavage of APP on the neuron membrane. NEVs mediate the transport of P-secretase from their parental neurons, while retaining its functional activity. Given these features, NEVs serve as ideal blood-based biomarkers, offering a noninvasive approach to assess P-secretase activity for detecting the early onset of ADNPC. B) NEV P-secretase Activity Assay for differentiating AD from HC. Step 1. Click chemistry- mediated enrichment of two subpopulations of NEVs using Click MagBeads. After labeling NEVs in 0.4 m of either CSF or plasma by one of the two TCO-grafted antibodies targeting NEV markers (i.e., LI CAM or NCAM), Click MagBeads functionalized with methyltetrazine (mTz) were employed to immobilize the respective subpopulations of NEVs. Step 2. Assessment of P-secretase activity of enriched NEVs using FRET probes. To assess P-secretase activity, NEVs enriched on Click MagBeads were lysed and incubated with P- secretase FRET probes. The central premise is that positive readouts are generated only when NEVs exhibit colocalization of one of the two NEV markers (i.e., L1CAM or NCAM) with P-secretase. The resulting P-secretase activity profiles across two subpopulations of NEVs, LlCAM(+)-NEVs and NCAM(+)-NEVs, were employed to differentiate AD patients from HCs.

[0045] FIGS. 36A-36E show preparation and characterization of Click MagBeads. A) Schematic illustration of stepwise preparation of the Click MagBeads. B) Zeta potentials of the amine-grafted Dynabeads™ (NH2-Dynabeads), mTz-modified Dynabeads™ (mTz- Dynabeads), and Click MagBeads. C) Fluorescence micrographs of the Click MagBeads before and after labeling with TCO-Cy5 (scale bar = 10 pm) and D) corresponding histograms of mean fluorescence intensity (MFI). E) Measurement of mTz motif concentration on the Click MagBeads over time after synthesis.

[0046] FIGS. 37A-37D show verification of click chemistry-mediated enrichment of two subpopulations of NEVs and confirmation of colocalization of NEV markers with P-secretase on the enriched NEVs. A) A schematic illustration of the workflow for characterizing interface between NEVs and Click MagBeads. Step 1. Click chemistry-mediatedimmobilization of NEVs onto Click MagBeads from CSF of AD patient. NEVs in 0.4 mL CSF of AD patient were labeled with one of the two TCO-grafted antibodies targeting NEV markers (i.e., LI CAM or NCAM). Click MagBeads were then employed to immobilize the respective subpopulations of NEVs. Step 2. Immunogold labeling to confirm colocalization of one of the two NEV markers with CD63 or P-secretase. To verify the colocalization of one of the two NEV markers with CD63, a well-established EV marker, or P-secretase, enriched NEVs were labeled with one of the two antibodies targeting CD63 or P-secretase, followed by immunogold labeling using AuNP-grafted secondary antibody. B) SEM images of LlCAM(+)-NEVs (top) and NCAM(+)-NEVs (bottom) immobilized onto the Click MagBeads using TCO-anti-LlCAM and TCO-anti-NCAM, respectively. C-D) TEM images of LlCAM(+)-NEVs (top) and NCAM(+)-NEVs (bottom) immobilized onto the Click MagBeads with immunogold labeling of C) CD63 and D) P-secretase.

[0047] FIGS. 38A-38C show proof-of-concept study of the NEV P-secretase Activity Assay with clinical CSF samples in a study cohort. A) A general workflow adopted for the NEV P-secretase Activity Assay using clinical CSF samples, each with a volume of 0.8 mL, from a study cohort of 15 AD patients and 15 HCs. The two subpopulations of NEVs from the CSF samples were enriched onto the Click MagBeads using one of the two TCO-grafted antibodies targeting NEV markers, i.e., L1CAM and NCAM. To assess P-secretase activity, the enriched NEVs were lysed and incubated with a P-secretase FRET probe to generate the P-secretase activity profile of study cohort. B) Heatmap summarizing quantitative P-secretase activity readouts obtained from the two subpopulations of NEVs in the clinical CSF samples. C) Box charts and corresponding ROC curves of two subpopulations of NEVs for differentiating AD patients from HCs.

[0048] FIGS. 39A-39C show validation of the NEV P-secretase Activity Assay with synthetic plasma samples. A) A schematic illustration of the workflow developed for assessing the capability of the NEV P-secretase Activity Assay to detect NEVs in synthetic plasma samples. Synthetic plasma samples were prepared by spiking the CSF of AD patient into EV-depleted HC plasma. B) Time-dependent measurement of fluorescence generated by the hydrolysis of the P-secretase FRET probes by P-secretase of two subpopulations of NEVsenriched with TCO-anti-LlCAM (left) and TCO-anti-NCAM (right) across varying concentrations of spiked NEVs. C) Corresponding calibration curves of P-secretase activity of two subpopulations of 'NEVs across varying concentrations of spiked NEVs.

[0049] FIGS. 40A-40I show retrospective case-control study with clinical plasma samples in a validation cohort. A) A general workflow adopted for the NEV -secretase Activity Assay using clinical plasma samples from a validation cohort of 55 AD patients and 55 HCs. B) Heatmap summarizing quantitative P-secretase activity readouts obtained from the two subpopulations of NEVs in the clinical plasma samples from the validation cohort and their corresponding MMSE scores. C) Box charts and D) corresponding ROC curves of P-secretase activity readouts from two subpopulations of NEVs for differentiating AD patients from HCs. E) NEV P-secretase Activity Score established based on a logistic regression model that synergistically combines the readouts from two subpopulations of NEVs into a single metric. F) Box chart, G) heatmap, and H) ROC curve of NEV P-secretase Activity Score to demonstrate the enhanced differentiation of AD patients from HCs. I) Correlation between NEV P-secretase Activity Scores and cognitive performance of AD patients indicated by MMSE scores.

[0050] FIG. 41 is a diagram showing severity of biomarker abnormality as a function of different stages of AD progression. The existing methods identify amyloid-P accumulation, an early indicator of ADNPC compared to different biomarkers, by detecting amyloid-P peptides and plaques through CSF tests and PET scans, respectively. However, these methods can only identify the disease once the plaques have formed and reached detectable levels. To overcome this limitation, a promising alternative is to evaluate upstream P- secretase activity, potentially enabling the detection of ADNPC at an even earlier stage than current approaches.

[0051] FIGS. 42A-42C show characterization of NEVs in the CSF of AD patient. NEVs isolated from the CSF of AD patient were characterized by A) NTA to measure their size and concentration, and B) TEM and C) SEM (scale bar = 200 nm) to confirm the morphology and integrity of NEVs.

[0052] FIGS. 43A-43E provides an investigation of P-secretase enzymatic kinetics. A) Schematic illustration of the P-secretase activity assay using recombinant human p-secretase. B) Time-dependent measurement of fluorescence generated by the hydrolysis of the P- secretase FRET probes by recombinant human P-secretase and C) corresponding Michaelis- Menten plot for determining the kinetic parameters (Vmax = 2.512 * 10-3 pM / s, KM = 3.80 pM, kcat = 0.0157 s-1, and kcat / KM = 4,130.98 s-1 M-l). D) Dynamic range of the P- secretase activity assay. E) Verification of the specificity of the P-secretase FRET probes.

[0053] FIGS. 44A-44B provide box charts and corresponding ROC curves of the two subpopulations of NEVs for differentiating AD patients from A) non-age-matching and B) age-matching HCs.

[0054] FIG. 45 provides Table 4. Summary of literature survey for identifying markers for enriching NEVs.

[0055] FIG. 46 provides Table 5. Clinical characteristics of HCs and AD patients for proof-of-concept demonstration of the NEV P-secretase Activity Assay using clinical CSF samples.

[0056] FIG. 47 provides Table 6. Clinical characteristics of HCs and AD patients for evaluation of the NEV P-secretase Activity Assay using clinical plasma samples.DETAILED DESCRIPTION

[0057] Some embodiments of the current invention are discussed in detail below. In describing embodiments, specific terminology is employed for the sake of clarity. However, the invention is not intended to be limited to the specific terminology so selected. A person skilled in the relevant art will recognize that other equivalent components can be employed and other methods developed without departing from the broad concepts of the current invention. All references cited anywhere in this specification, including the Background and Detailed Description sections, are incorporated by reference as if each had been individually incorporated.

[0058] Some embodiments of this invention relate to the concept of tumor-derived extracellular vesicle (EV) protease activity assays. Given the crucial role of EV-mediated protease transport activity in the progression of solid tumors, quantifying protease activity using tumor EVs can be utilized to detect disease progression in cancer patients. Using osteosarcoma (OS) as an example, we demonstrated OS EV Protease Activity Assay (FIG. 1), capable of selectively purifying OS-derived EVs, followed by quantification of two proteases in the purified OS EVs. However, the general concepts of the current invention are not limited to this example.

[0059] In this example, we synergistically integrate two very powerful platform technologies: i) EV Click MagBeads for OS EV enrichment via click chemistry-mediated EV enrichment targeting three OS-specific markers (i.e., B7H3, GPNMB, and LRRC15), and ii) FRET-based peptide cleavage assays that quantify the activity of OS-associated active proteases, including MMP-14 and MMP-2 / 9, both of which play a crucial role in modulating OS progression and metastasis. The successful development of this OS EV Protease Activity Assay is rapidly translatable, offering pediatric oncologists a sensitive noninvasive assay for dynamic monitoring of treatment response and detecting disease progression in OS patients.

[0060] Fluorescence resonance energy transfer (FRET)-based peptide cleavage assay is an alternative to conventional zymography for measuring protease activity. In this method, a peptide substrate for the specific protease is labeled with a fluorophore and a quencher pair. The presence of target protease leads to the cleavage of the peptide substrate, followed by releasing fluorophore from the quencher, resulting in a fluorescence signal.26-27This assay allows for real-time measurement of protease activity using multi-well plate readers, enabling scalable analysis within a few hours. Moreover, it can be applied to various proteases, as long as suitable peptide substrates are available.28-31

[0061] Protease activity of tumor-derived EVs as cancer biomarkers. Liquid biopsy provides a repeatable and non-invasive method for analyzing the protease activity in tumor tissue, addressing the limitations of traditional tissue biopsies. Tumor-derived proteases present in body fluids, such as blood or urine, can be analyzed to assess tumor growth and metastasis systematically.20-32’33For instance, elevated levels and activity ofMMPs have been observed in the blood of cancer patients.34-36However, it is challenging to identify the specific source of proteases in the bloodstream, as the level of MMPs can be influenced by various disease conditions.37’38Extracellular vesicles (EVs), heterogeneous particles containing proteins and genetic materials,39-40are released by all types of cells, including tumor cells, for intercellular communication.41’42Tumor-derived EVs can transport proteases involved in tumor invasion and dissemination.43-45Moreover, protein cargos of EV are highly stable over a prolonged time-period.46’47Therefore, tumor-derived EVs48-51provide an ideal source to systemically and repeatedly assess the activity of proteases originating from tumors, offering valuable information for diagnosis, prognosis, and metastasis assessment.3, 4Ideally, a tumor EV-based protease activity assay is composed of two functional components: i) an efficient tumor EV purification system, and ii) a sensitive and specific protease activity assay to quantify the trace amount of protease on the surface of tumor EVs or extracted from the tumor EVs.

[0062] Tumor EV enrichment technologies. Conventional methods like ultracentrifugation,52’53filtration,54’55precipitation,56and size-based microfluidic enrichment57-59can isolate total EVs based on their physical properties, but they cannot differentiate between tumor-derived and non-tumor-derived EVs. To address this, researchers have implemented immunoaffinity -based capture techniques to enrich tumor- derived EVs,52-53’60such as anti-GPCl -coated beads for pancreatic cancer61and anti- EGFRvIII-integrated herringbone microfluidic devices for glioblastoma.62Over the last decade, our joint UCLA team has developed new EV enrichment platforms, i.e., EV Click Chips and EV Click Beads, which can effectively enrich tumor-derived EVs from plasma samples, enabling downstream molecular and functional analysis.63-67The EV Click MagBeads represent our ongoing efforts to further streamline tumor-derived EV enrichment (FIG. 2) using click chemistry -mediated magnetic beads for facile separation of the tumor- derived EVs.

[0063] Osteosarcoma and clinical unmet needs. OS is the most common bone cancer of childhood with high metastatic potential.68Current treatment protocols involve a combination of chemotherapy and surgery depending on the patient’s risk profile.69Most OSpatients are treated with neoadjuvant chemotherapy for about 10 weeks. In some people with osteosarcoma in an arm or leg bone, this can shrink the tumor, which might help make surgery easier. The adjuvant chemotherapy is then given after surgery for up to a year. However, early detection of disease recurrence and progression is crucial for better outcomes, as it allows for timely interventions and reducing the number of unnecessary and ineffective therapies. Therefore, there is a need for dynamic monitoring tools to detect disease progression prior to radiographic evidence of treatment failure or disease recurrence, allowing pediatric oncologists to make more informed and timely alterations to their management.70

[0064] OS-associated active proteases for detecting disease progression. Like other cancers, OS express and activate proteases to promote disease progression. MMP-14, transported by EVs due to its membrane-bound nature, is a representative predictor of poor prognosis in OS.71’72High activity of MMP-2 / 9 is also linked to aggressive characteristics of OS.73In addition to MMP-14 and MMP-2 / 9, ADAMI 7 mutation was observed in a dedifferentiated lesion of OS patient, potentially indicating a link to metastasis.74CTSB is up- regulated and increased in activity invasive OS.75Therefore, analyzing the activity signatures of OS-associated active proteases could provide essential information in detecting disease progression.

[0065] Enrichment of OS EVs based on surface protein markers. Recent studies identified surface protein markers associated with OS, including B7H3 (CD276),76-78GPNMB,79and LRRC15. B7H3 level is increased in the blood ofOS patients,80and antibody-drug conjugates targeting GPNMB have demonstrated antitumor activity in a preclinical study.81Our preliminary studies (FIGS. 5A-5C and 6A-6D) have shown the feasibility of targeting B7H3, GPNMB, and LRRC15 for antibody-mediated enrichment of OS EVs in clinical samples, laying a strong foundation for this approach according to an embodiment of the current invention.

[0066] Measuring protease activities of OS EVs for detecting disease progression. OS EVs are ideal “biomarker reservoirs” of biomol ecular cargos, including active proteases, for downstream molecular analysis in non-invasive OS diagnosis.82Emerging evidence suggests that overexpression of MMP-14 and other proteases carried byOS EVs71’72-76may serve as biomarkers for detecting OS progression.83The preliminary data demonstrated for this OS EV Protease Activity Assay is capable of discerning and quantifying two protease activities across three OS EV subpopulations (defined by OS- specific surface markers, i.e., B7H3, GPNMB, and LRRC15) to give the respective OS EV Protease Activity Signatures. Exploring the OS EV Protease Activity Signatures as biomarkers for detecting OS progression is therefore of great significance as it may augment the current diagnostic modalities available for OS patients (e.g., radiographic and / or PET imaging).

[0067] Further aspects of some embodiments of the current invention are summarized below:

[0068] EV Click MagBeads. A unique feature of EV Click MagBeads is utilizing a click chemistry-mediated EV purification (FIGS. 1 and 2). Click chemistry84is a type of rapid bio-orthogonal organic reaction commonly used for biological labeling and bioconjugation. It is noteworthy that click chemistry was recently recognized with the 2022 the Nobel Prize in Chemistry for its advances in diagnostics, pharmaceutical development, and medicine. Compared to conventional immunoaffinity-based EV capture approaches, click chemistry-mediated EV capture can overcome poor EV capture performance caused by the low number of antigens present on the surface of individual EVs. In our previous studies,66’67we used the inverse-electron-demand Diels- Alder (IEDDA) cycloaddition85between tetrazine (Tz) and TCO motifs86to enable specific and irreversible immobilization of the EVs, with improved capture efficiency and reduced background. Our recent report demonstrates that the early version of EV Click Beads can isolate tumor-derived EVs with an isolation yield up to 80%, significantly higher than conventional immunoaffmity methods.67In an embodiment of the current invention, EV Click MagBeads were prepared and optimized to achieve easy handling and to improve scalability.

[0069] FRET-based peptide cleavage assay. FRET-based peptide cleavage assays provide a faster alternative to conventional zymography. This approach allows for real-time measurement of protease activity using multi-well plate readers, enabling scalable analysis. Our data of FIGS. 5A-7C demonstrate that this approach can be used for measuring metastaticactivity and dynamic monitoring of treatment response as well as disease progression in OS patients.

[0070] Advantages and transformative potential of OS EV Protease Activity Assay can include: (i) a sensitive and specific EV protease activity assay, which integrates the 2022 Nobel Prize-winning click chemistry -mediated EV enrichment and FRET-based peptide cleavage assays, ONLY the colocalization of a pair of an OS-specific marker and a protease will generate a signal, enabling highly sensitive detection of OS metastasis; (ii) generating a quantitative score, allowing for parallel quantification of 12 unique OS EV Protease Activity Signatures to be statistically integrated into a combined score; (iii) accessibility, the readout of the OS EV Protease Activity Assay is compatible with widely available well plate fluorescence readers; (iv) fast turnaround time, generating OS EV Protease Activity Scores within 3 h; (v) cost efficient, <$5 for OS EV enrichment and <$5 for FRET-based peptide cleavage assays; (vi) reproducible and scalable, the streamlined workflow and quantification method will minimize user-to-user variation.

[0071] Blood Biospecimen Bank (BBB) at UCLA. Since 2012, Drs. Tseng / Zhu laboratory has been collecting and annotating blood specimens from pediatric patients diagnosed with sarcoma, including osteosarcoma and Ewing’s sarcoma, through collaboration with the Division of Medical Oncology, the Division of Pediatric Hematology / Oncology and the UCLA Pediatric Bone and Soft Tissue Sarcoma program. To date, nearly 100 samples from 50 pediatric cancer patients at all clinical stages have been banked. Along with the expanded collaboration with the clinical teams led by Drs. Jonas and Federman, the UCLA BBB would provide a unique opportunity for validation of the OS EV Protease Activity Assay for detection of metastatic OS.Experimental Results

[0072] EV Click Beads and EV Click MagBeads. Our recent reports66-67demonstrated the efficient capture of EVs using the EV Click Bead platform, mediated by click chemistry between Tz and TCO. This technology was successfully transferred to newlydesigned EV Click MagBeads to efficiently capture OS EVs from patient plasma (FIGS. 3A-3B)

[0073] mTz-grafted EV Click MagBeads. We prepared EV Click MagBeads using a surface modification method similar to our recent work on silica microbead-based EV Click Beads.67For the purity of isolated OS EVs by minimizing the trapping of background EVs, we introduced surface linkers with elongated polyethylene glycol (PEG) chains (i.e., PEG24), which reduces biofouling (FIG. 3A). Quality control (QC) protocols of EV Click MagBeads and TCO-grafted antibodies will be established by using click chemistry motif- grafted fluorescent probes (e.g., TCO-Cy5 and mTz-Cy5) to assess the presence and lifetime of the click chemistry motifs (FIG. 3B).

[0074] Preparation of TCO- grafted antibodies. The three OS EV-associated antibodies, i.e., anti-B7H3, anti- GPNMB, and anti-LRRC15, were covalently grafted with click chemistry motif, TCO using N- hydroxysuccinimide (NHS) chemistry. To test the successful conjugation of TCO onto each OS EV-associated antibody, we adopted an existing flow cytometry protocol where HOS and 143B cells are used. Using TCO- grafted anti- B7H3 as an example, HOS cell suspension is first prepared, fixed, and incubated with TCO- anti-B7H3 in tubes. After washing away the free antibodies, HOS cells conjugated with TCO-anti-B7H3 are exposed to Tz-Cy5 (the fluorescent reporter) for the click chemistry reaction. The resulting cell suspension is then subjected to flow cytometry analysis to verify the affinities of TCO-grafted antibodies. With the TCO-tagged multi -antibody cocktail in place, we will rationally examine operational parameters associated with the workflow (FIG. 2).

[0075] Characterization of the interfaces between EV Click MagBeads and OS EVs. Scanning electron microscopy (SEM) and Transmission electron microscopy (TEM) were employed to characterize the interfaces between EV Click MagBeads and OS EVs, where freshly harvested HOS cell line-derived OS EVs in PBS were used as a model system. FIG. 4A shows representative SEM images (scale bar, 500 nm) of EV Click MagBeads with immobilized HOS EVs in the presence of TCO-anti-B7H3. The cartoon depicts the click chemistry-medicated immobilization of OS EVs. FIG. 4B shows a representative TEMimage (scale bar, 100 nm) of HOS EVs immobilized on a Click MagBead after immunogold labelling of CD63, a representative EV marker.

[0076] Quantification of protease activity. We evaluated the activity of two OS- associated active proteases, i.e., MMP-14 and MMP-2 / 9. Before assessing protease activity of OS EVs, we conducted an enzyme kinetics study to determine the appropriate substrate concentration. Using MMP-14 as an example (FIG. 5A), we obtained the Kmvalue of MMP-14 (4.53 pM), which is comparable with previous reports,87’88This value was used as optimal substrate concentration. We will further determine Kmand kcat^Km of other proteases and compare them with reported values to evaluate the performance of our FRET- based peptide cleavage assay.

[0077] Quantification of protease activity. Linear relationship between activity and fluorescence readout is important for estimating the enzymatic activity of a given sample. In this regard, we sought to quantify the enzyme activity via the proposed FRET -based peptide cleavage assay using activated MMP-14 (recombinant human MMP-14, rhMMP-14) as a representative OS-associated active protease. Linearity with R^ = 0.991 was obtained for the dynamic range of 1-8 pmole of recombinant human MMP-14 (FIG. 5B). We will assess the titration of other active proteases using a similar approach.

[0078] Quantification of OS EV protease activity. We evaluated the concentration dependence of the OS EV protease activity of artificial plasma samples using GPNMB as an OS-specific marker. MMP-14 activity was measured via the FRET -based peptide cleavage assay, and a linear titration curve (R^ = 0.987) was observed with a dynamic range of 0.25-4 pg of spiked EV protein (FIG. 5C). We will demonstrate titration for other active proteases in OS EVs spiked artificial plasma with varying OS-specific markers.

[0079] OS EV MMP-14 Activity Assay. Based on the conditions optimized for MMP-14 (FIGS. 5A-5C), we explored the diagnostic potential of OS EV MMP-14 Activity Assay (FIG. 6A), where anti-GPNMB was used as the OS EV-specific capture agent to achieve click chemistry -mediated capture of GPNMB(+)-OS EVs. Plasma samples collected from healthy donors and OS patients with metastatic and localized diseases were used. FIG.6B summarizes the fluorescence readouts of OS EVMMP-14 Activity Assay observed for healthy donors (HD), localized and metastatic OS patients. The elevated MMP-14 activities were observed for GPNMB(+)-OS EVs in both localized and metastatic OS patients. ROC curves (FIG. 6C) were generated for OS EV MMP-14 Activity Assay in the following groups: HD versus all OS patients (i.e., localized + metastatic) (AUC = 0.92), HD versus localized OS (AUC = 0.91), and localized versus metastatic (AUC = 0.82). As a control experiment, plasma sample from HD and OS patients were directly subjected to the FRET- based peptide cleavage assay for MMP-14 without OS EV purification. No significant difference was observed (FIG. 6D), thus highlighting the crucial role of OS EV purification in the OS EV MMP-14 Activity Assay.

[0080] OS EV MMP-2 / 9 Activity Assay. For OS EV MMP-2 / 9 Activity Assay, OS EV lysis is needed to release MMP-2 / 9 from OS EV prior to conducting the FRET-based peptide cleavage assay. We were able to carry out click chemistry- mediated EV enrichment targeting three OS-specific markers (i.e., B7H3, GPNMB, and LRRC15), followed by FRET- based peptide cleavage assay after OS EV lysis. FIGS. 7A-7C summarize the results of OS EV MMP-2 / 9 Activity Assay observed for GPNMB(+)-OS EVs, B7H3(+)-OS EVs, and LRRC15(+)-OS EVs. Overall, LRRC15(+)-OS EVs exhibited superior performance in detecting both localized and metastatic OS from HD.

[0081] OS EV MMP-14 Activity Assay for monitoring disease progression. We studied the dynamic changes in OS patients and the correlation to treatment response. This was performed on 2 patients with serial samples taken before the initiation of treatment and up to 9 weeks after the first blood draw. Serial plasma samples were collected from patient OST-003 before treatment, and at week 1, 3, 5, 9 after first blood draw. The corresponding MMP-14 activity on GPNMB(+)-OS EVs for each timepoint were plotted in FIG. 8A. MRI images of the patient showed right femur osteosarcoma with interval increase in soft tissue mass prior to treatment. Following the initiation of neoadjuvant chemotherapy, the patient was found to have a progressive disease after 9 weeks with approximately size 9.1 x 9.6 x 8.3 cm compared to 5.9 x 4.9 x 5.6 cm when measured at a similar level on prior examination before treatment. Notably, there is an increase of MMP-14 activity on GPNMB+ OS EVsobserved at week 5, after switching to Cycle 2 of DOX / DDP, the MMP-14 activity on GPNMB(+)-OS EVs decreased at week 6, but quickly increased again at week 9, which indicated disease progression. These radiographic observations are consistent with the dynamic changes observed in the corresponding MMP-14 activity on GPNMB(+)-OS EVs. Similarly, serial plasma samples were collected from patient OST-005 before treatment, and at week 1, 3, 5, 7 after first blood draw. FIG. 8B shows that MRI images of osteosarcoma at the distal femoral diaphysis before treatment and at week 3 post-treatment delineated aggressive progression from previously 3.5 x 4.5 x 10.4 cm to 5.2 x 5.4 x 8.9 cm, with intramedullary and soft tissue components. These radiographic observations are strongly correlated with the increase of the corresponding MMP-14 activity on GPNMB(+)-OS EVs. Unfortunately, radiographic examinations cannot be performed as frequently as IVD assay.

[0082] References1 . Tan GJ, Peng ZK. Lu JP, Tang FQ. Cathepsins mediate tumor metastasis. World J Biol Chem. 2013:4(4):91-101. doi: 10.4331 / wjbc.v4.i4.91. PubMed PMID: 24340132; PMCID: PMC3856311.2. Olson OC, Joyce JA. Cysteine cathepsin proteases: regulators of cancer progression and therapeutic response. Nat Rev Cancer. 2015; 15( 12): 712-29. doi: 10.1038 / nrc4027. PubMed PMID: 26597527.3. Winkler J, Abisoye-Ogunniyan A, Metcalf KJ, Werb Z. Concepts of extracellular matrix remodelling in tumour progression and metastasis. Nat Commun. 2020; 11 ( 1 ) : 5120. Epub 20201009. doi: 10.1038 / s41467-020-18794-x. PubMed PMID: 33037194; PMCID: PMC7547708.4. Roy R, Yang J, Moses MA. Matrix metalloproteinases as novel biomarkers and potential therapeutic targets in human cancer. J Clin Oncol. 2009;27(31):5287-97. Epub 20090908. doi: 10.1200 / JC0.2009.23.5556. PubMed PMID: 19738110; PMCID: PMC2773480.5. Kessenbrock K, Plaks V, Werb Z. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 2010; 141 ( 1): 52-67. doi: 10.1016 / j .cell.2010.03.015. PubMed PMID: 20371345; PMCID: PMC2862057.6. Overall CM, Kleifeld O. Tumour microenvironment - opinion: validating matrixmetalloproteinases as drug targets and anti-targets for cancer therapy. Nat Rev Cancer. 2006;6(3):227-39. doi: 10.1038 / nrcl821. PubMed PMID: 16498445. Mason SD, Joyce JA. Proteolytic networks in cancer. Trends Cell Biol. 2011;21(4):228- 37. Epub 20110112. doi: 10.1016 / j .tcb.2010. 12.002. PubMed PMID: 21232958; PMCID: PMC3840715. Kunz P. Sahr H, Lehner B, Fischer C, Seebach E, Fellenberg J. Elevated ratio of MMP2 / MMP9 activity is associated with poor response to chemotherapy in osteosarcoma. BMC Cancer. 2016:16:223. Epub 20160315. doi: 10. 1186 / s 12885-016- 2266-5. PubMed PMID: 26979530; PMCID: PMC4793522. Yousef EM, Tahir MR, St-Picrrc Y, Gaboury LA. MMP-9 expression varies according to molecular subtypes of breast cancer. BMC Cancer. 2014;14:609. Epub 20140823. doi: 10.1186 / 1471-2407-14-609. PubMed PMID: 25151367; PMCID: PMC4150970. Kim TD, Song KS, Li G, Choi H, Park HD, Lim K, Hwang BD, Yoon WH. Activity and expression of urokinase -type plasminogen activator and matrix metalloproteinases in human colorectal cancer. BMC Cancer. 2006;6:211. Epub 20060818. doi: 10.1186 / 1471- 2407-6-211. PubMed PMID: 16916471; PMCID: PMC1563482. Wood M, Fudge K, Mohler JL, Frost AR, Garcia F, Wang M, Steams ME. In situ hybridization studies of metalloproteinases 2 and 9 and TIMP-1 and TIMP-2 expression in human prostate cancer. Clin Exp Metastasis. 1997;15(3):246-58. doi: 10. 1023 / a: 1018421431388. PubMed PMID: 9174126. Ohtani H, Motohashi H, Sato H, Seiki M, Nagura H. Dual over-expression pattern of membrane-type metalloproteinase- 1 in cancer and stromal cells in human gastrointestinal carcinoma revealed by in situ hybridization and immunoelectron microscopy. Int J Cancer. 1996;68(5):565-70. doi: 10.1002 / (SICI)1097- 0215(19961127)68:5<565::AID-IJC2>3.0.CO;2-X. PubMed PMID: 8938135. Ingvarsen SZ, Gardsvoll H, van Putten S, Norregaard KS, Krigslund O, Meilstrup JA, Tran C, Jurgensen HJ, Melander MC, Nielsen CH, Kjaer A, Bugge TH, Engelholm LH, Behrendt N. Tumor cell MT1-MMP is dispensable for osteosarcoma tumor growth, bone degradation and lung metastasis. Sci Rep. 2020:10(1): 19138. Epub 20201105. doi: 10.1038 / s41598-020-75995-6. PubMed PMID: 33154487: PMCID: PMC7645741. Trudel D, Fradet Y, Meyer F, Harel F, Tetu B. Significance of MMP-2 expression in prostate cancer: an immunohistochemical study. Cancer Res. 2003;63(23):8511-5. PubMed PMID: 14679018.Stetler-Stevenson WG. The tumor microenvironment: regulation by MMP-independent effects of tissue inhibitor of metalloproteinases-2. Cancer Metastasis Rev. 2008;27(l):57-66. doi: 10.1007 / sl0555-007- 9105-8. PubMed PMID: 18058195; PMCID: PMC2254553. Sloane BF, Moin K, Krepela E, Rozhin J. Cathepsin B and its endogenous inhibitors: the role in tumor malignancy. Cancer Metastasis Rev. 1990;9(4):333-52. doi: 10.1007 / BF00049523. PubMed PMID: 2097084. Vandooren J, Geurts N, Martens E, Van den Steen PE, Opdenakker G. Zymography methods for visualizing hydrolytic enzymes. Nat Methods. 2013; 10(3):211-20. doi: 10.1038 / nmeth.2371. PubMed PMID: 23443633. Heussen C. Dowdle EB. Electrophoretic analysis of plasminogen activators in polyacrylamide gels containing sodium dodecyl sulfate and copolymerized substrates. Anal Biochem. 1980; 102(1): 196-202. doi: 10.1016 / 0003-2697(80)90338-3. PubMed PMID: 7188842. Hibbs MS, Hasty KA, Seyer JM, Kang AH, Mainardi CL. Biochemical and immunological characterization of the secreted forms of human neutrophil gelatinase. J Biol Chem. 1985;260(4):2493-500. PubMed PMID: 2982822. Snoek-van Beurden PA, Von den Hoff JW. Zymographic techniques for tire analysis of matrix metalloproteinases and their inhibitors. Biotechniques. 2005;38(l):73-83. doi: 10.2144 / 05381RV01. PubMed PMID: 15679089. Kleiner DE, Stetler-Stevenson WG. Quantitative zymography: detection of picogram quantities of gelatinases. Anal Biochem. 1994;218(2):325-9. doi: 10.1006 / abio.1994.1186. PubMed PMID: 8074288. Wilder CL, Park KY, Keegan PM, Platt MO. Manipulating substrate and pH in zymography protocols selectively distinguishes cathepsins K, L, S, and V activity in cells and tissues. Arch Biochem Biophys. 201 l;516(l):52-7. Epub 20110929. doi: 10.1016 / j.abb.201 L 09.009. PubMed PMID: 21982919; PMCID: PMC3221864. Hawkes SP. Li H, Taniguchi GT. Zymography and reverse zymography for detecting MMPs and TIMPs. Methods Mol Biol. 2010;622:257-69. doi: 10.1007 / 978-1-60327- 299-5 16. PubMed PMID: 20135288. Manrow RE, Dottin RP. Renaturation and localization of enzymes in polyacrylamide gels: studies with UDPglucose pyrophosphorylase of Dictyostelium. Proc Natl Acad SciU S A. 1980;77(2):730-4. doi: 10.1073 / pnas.77.2.730. PubMed PMID: 6153803; PMCID: PMC348354. Lacks SA, Springhom SS. Renaturation of enzymes after polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. J Biol Chem. 1980;255(15):7467-73. PubMed PMID: 6156169. Cummings RT, Salowe SP, Cunningham BR, Wiltsie J, Park YW, Sonatore LM, Wisniewski D, Douglas CM, Hermes JD, Scolnick EM. A peptide-based fluorescence resonance energy transfer assay for Bacillus anthracis lethal factor protease. Proc Natl Acad Sci U S A. 2002;99(10):6603-6. Epub 20020507. doi: 10.1073 / pnas.062171599. PubMed PMID: 11997440; PMCID: PMC124449. Shi L, De Paoli V, Rosenzweig N, Rosenzweig Z. Synthesis and application of quantum dots FRET- based protease sensors. J Am Chem Soc. 2006;128(32): 10378-9. doi: 10.1021 / ja063509o. PubMed PMID: 16895398. Naing SH, Vukoti KM, Drury JE, Johnson JL, Kalyoncu S, Hill SE, Torres MP, Liebennan RL. Catalytic Properties of Intramembrane Aspartyl Protease Substrate Hydrolysis Evaluated Using a FRET Peptide Cleavage Assay. ACS Chem Biol. 2015;10(9):2166-74. Epub 20150715. doi: 10.1021 / acschembio.5b00305. PubMed PMID: 26118406. Hu HY, Gehrig S, Rcithcr G, Subramanian D, Mall MA, Plcttcnburg O, Schultz C. FRET- based and other fluorescent proteinase probes. Biotechnol J. 2014;9(2):266-81. Epub 20140127. doi: 10.1002 / biot.201300201. PubMed PMID: 24464820. Zauner T, Berger-Hoffmann R, Muller K, Hoffmann R, Zuchner T. Highly adaptable and sensitive protease assay based on fluorescence resonance energy transfer. Anal Chem. 2011;83(19):7356-63. Epub 20110914. doi: 10.1021 / ac201274f. PubMed PMID: 21892820. Guerrero JL, O'Malley MA, Daugherty PS. Intracellular FRET-based Screen for Redesigning the Specificity of Secreted Proteases. ACS Chem Biol. 2016; 11(4):961-70. Epub 20160114. doi: 10. 102 l / acschembio.5b01051. PubMed PMID : 26730612. Kotaska K, Dusek P. Prusa R, Vesely S. Babjuk M. Urine and serum cathepsin B concentrations in the transitional cell carcinoma of the bladder. J Clin Lab Anal. 2012;26(2):61-5. doi: 10.1002 / jcla.21483. PubMed PMID: 22467319; PMCID: PMC6807346.Yeh HC, Lin SM, Chen MF, Pan TL, Wang PW, Yeh CT. Evaluation of serum matrix metalloproteinase (MMP)-9 to MMP-2 ratio as a biomarker in hepatocellular carcinoma. Hepatogastroenterology. 2010;57(97):98-102. PubMed PMID: 20422881. Hadler-Olsen E, Winberg JO, Uhlin-Hansen L. Matrix metalloproteinases in cancer: their value as diagnostic and prognostic markers and therapeutic targets. Tumour Biol. 2013;34(4):2041-51. Epub 20130517. doi: 10.1007 / sl3277-013-0842-8. PubMed PMID: 23681802. Morgia G. Falsaperla M, Malaponte G, Madonia M, Indelicate M, Travali S. Mazzarino MC. Matrix metalloproteinases as diagnostic (MMP-13) and prognostic (MMP-2, MMP- 9) markers of prostate cancer. Urol Res. 2005;33(l):44-50. Epub 20041022. doi: 10.1007 / s00240-004-0440-8. PubMed PMID: 15517230. Wu ZS, Wu Q, Yang JH, Wang HQ, Ding XD, Yang F, Xu XC. Prognostic significance of MMP-9 and TIMP-1 serum and tissue expression in breast cancer. Int J Cancer. 2008;122(9):2050-6. doi: 10.1002 / ijc.23337. PubMed PMID: 18172859. Gidday JM, Gasche YG, Copin JC, Shah AR. Perez RS, Shapiro SD. Chan PH, Park TS. Leukocyte- derived matrix metalloproteinase-9 mediates blood-brain barrier breakdown and is proinflammatory after transient focal cerebral ischemia. Am J Physiol Heart Circ Physiol. 2005;289(2):H558-68. Epub 20050311. doi: 10.1152 / ajpheart.O1275.2004. PubMed PMID: 15764676. C DA-M, Couto AES. Campos LCB, Vasconcelos TF, Michelon-Barbosa J, Corsi CAC, Mestriner F, Petroski-Moraes BC, Garbellini-Diab MJ, Couto DMS, Jordani MC, Ferro D, Sbragia L, Joviliano EE, Evora PR, Carvalho Santana R, Martins-Filho OA, Polonis K, Menegueti MG, Ribeiro MS, Auxiliadora- Martins M, Becari C. MMP-2 and MMP-9 levels in plasma are altered and associated with mortality in COVID- 19 patients. Biomed Phannacother. 2021;142: 112067. Epub 20210820. doi: 10.1016 / j.biopha.2021.112067. PubMed PMID: 34449310; PMCID: PMC8376652. Raposo G, Stoorvogel W. Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol. 2013;200(4):373-83. doi: 10.1083 / jcb.201211138. PubMed PMID: 23420871; PMCID: 3575529. Revenfeld AL, Baek R, Nielsen MH, Stensballe A, Vanning K, Jorgensen M. Diagnostic and prognostic potential of extracellular vesicles in peripheral blood. Clin Ther. 2014;36(6):830-46. doi: 10.1016 / j.clinthera.2014.05.008. PubMed PMID: 24952934.van Niel G, D'Angelo G, Raposo G. Shedding light on the cell biology of extracellular vesicles. Nat Rev Mol Cell Biol. 2018; 19(4):213-28. Epub 2018 / 01 / 18. doi: 10.1038 / nim.2017.125. PubMed PMID: 29339798. Gao Z, Pang B, Li J, Gao N, Fan T, Li Y. Emerging Role of Exosomes in Liquid Biopsy for Monitoring Prostate Cancer Invasion and Metastasis. Front Cell Dev Biol. 2021;9:679527. Epub 2021 / 05 / 22. doi: 10.3389 / fcell.2021.679527. PubMed PMID: 34017837; PMCID: PMC8129505. Mashouri L, Yousefi H. Aref AR, Ahadi AM, Molaei F, Alahari SK. Exosomes: composition, biogenesis, and mechanisms in cancer metastasis and drug resistance. Mol Cancer. 2019; 18(1):75. Epub 20190402. doi: 10. 1186 / s 12943-019-0991-5. PubMed PMID: 30940145; PMCID: PMC6444571. Xu R. Rai A, Chen M, Suwakulsiri W, Greening DW. Simpson RJ. Extracellular vesicles in cancer - implications for future improvements in cancer care. Nat Rev Clin Oncol. 2018;15(10):617-38. doi: 10.1038 / s41571-018-0036-9. PubMed PMID: 29795272. Shimoda M, Khokha R. Proteolytic factors in exosomes. Proteomics. 2013; 13(10- 11): 1624-36. doi: 10.1002 / pmic.201200458. PubMed PMID: 23526769. Erozenci LA, Piersma SR, Pham TV, Bijnsdorp IV, Jimenez CR. Longitudinal stability of urinary extracellular vesicle protein patterns within and between individuals. Sci Rep. 2021;l 1(1): 15629. Epub 20210802. doi: 10.1038 / s41598-021-95082-8. PubMed PMID: 34341426: PMCID: PMC8329217. Jeyaram A, Jay SM. Preservation and Storage Stability of Extracellular Vesicles for Therapeutic Applications. AAPS J. 2017;20( 1): 1. Epub 20171127. doi: 10. 1208 / s 12248- 017-0160-y. PubMed PMID: 29181730; PMCID: PMC6582961. King HW, Michael MZ, Gleadle JM. Hypoxic Enhancement of Exosome Release by Breast Cancer Cells. BMC Cancer. 2012;12:421. doi: 10.1 186 / 1471-2407-12-421. PubMed PMID: 22998595; PMCID: 3488584. Taylor DD, Gercel-Taylor C. The origin, function, and diagnostic potential of RNA within extracellular vesicles present in human biological fluids. Front Genet. 2013;4: 142. doi: 10.3389 / fgene.2013.00142. PubMed PMID: 23908664; PMCID: 3726994. Thietart S, Rautou PE. Extracellular vesicles as biomarkers in liver diseases: Aclinician's point of view. J Hepatol. 2020;73(6):1507-25. Epub 2020 / 07 / 19. doi: 10. 1016 / j .jhep.2020.07.014. PubMed PMID: 32682050. Lee YT. Tran BV, Wang JJ, Liang IY. You S. Zhu Y, Agopian VG, Tseng HR, YangJD. The Role of Extracellular Vesicles in Disease Progression and Detection of Hepatocellular Carcinoma. Cancers (Basel). 2021;13(12). Epub 2021 / 07 / 03. doi:IO.3390 / cancersl3123076. PubMed PMID: 34203086; PMCID: PMC8233859. Tauro BL Greening DW, Mathias RA, Ji H, Mathivanan S, Scott AM, Simpson RJ. Comparison of Ultracentrifugation, Density Gradient Separation, and Immunoaffinity Capture Methods for Isolating Human Colon Cancer Cell Line LIM1863-Derived Exosomcs. Methods. 2012;56(2):293-304. doi: 10.1016 / j.ymcth.2012.01.002. PubMed PMID: 22285593. Kalra H, Adda CG, Liem M, Ang CS, Mechler A, Simpson RJ, Hulett MD, Mathivanan S. Comparative Proteomics Evaluation of Plasma Exosome Isolation Techniques and Assessment of the Stability of Exosomes in Normal Human Blood Plasma. Proteomics. 2013;13(22):3354-64. doi: 10.1002 / pmic.201300282. PubMed PMID: 24115447. Heinemann ML, Ilmer M, Silva LP, Hawke DH. Recio A, Vorontsova MA. Alt E, Vykoukal J. Benchtop isolation and characterization of functional exosomes by sequential fdtration. J Chromatogr A. 2014;1371: 125-35. doi: 10.1016 / j.chroma.2014. 10.026. PubMed PMID: WOS:000345810200014. Heinemann ML. Vykoukal J. Sequential Filtration: A Gentle Method for the Isolation of Functional Extracellular Vesicles. In: Kuo WP, Jia S, editors. Extracellular Vesicles: Methods and Protocols. Totowa: Humana Press Inc; 2017. p. 33-41. Rekker K, Saare M, Roost AM, Kubo AL, Zarovni N, Chiesi A, Salumets A, Peters M. Comparison of serum exosome isolation methods for microRNA profding. Clin Biochem. 2014;47(l-2): 135-8. doi: 10. 1016 / j.clinbiochem.2013.10.020. PubMed PMID: 24183884. Kanwar SS, Dunlay CJ, Simeone DM, Nagrath S. Microfluidic Device (ExoChip) for On-Chip Isolation, Quantification and Characterization of Circulating Exosomes. Lab Chip. 2014;14( 11): 1891-900. doi: 10.1039 / c41c00136b. PubMed PMID: 24722878; PMCID: 4134440. Zhao Z, Yang Y, Zeng Y, He M. A Microfluidic ExoSearch Chip for Multiplexed Exosome Detection towards Blood-Based Ovarian Cancer Diagnosis. Lab Chip.2016;16(3):489-96. doi: 10.1039 / c51c01117e. PubMed PMID: 26645590; PMCID: 4729647. Wunsch BH, Smith JT, Gifford SM. Wang C, Brink M, Bruce RL, Austin RH, Stolovitzky G, Astier Y. Nanoscale lateral displacement arrays for the separation of exosomes and colloids down to 20 nm. Nat Nanotechnol. 2016;l 1(11):936-40. doi: 10.1038 / nnano.2016.134. PubMed PMID: 27479757. Y ang F, Liao X, Tian Y, Li G. Exosome separation using microfluidic systems: size-based, immunoaffmity- based and dynamic methodologies. Biotechnol J. 2017:12(4). doi: 10.1002 / biot.201600699. PubMed PMID: 28166394. Melo SA, Lucckc LB, Kahlcrt C, Fernandez AF, Gammon ST, Kaye J, LcBlcu VS, Mittendorf EA, Weitz I, Rahbari N. Reissfelder C. Pilarsky C, Fraga MF, Piwnica- Wonns D, Kalluri R. Glypican-1 Identifies Cancer Exosomes and Detects Early Pancreatic Cancer. Nature. 2015:523(7559): 177-82. doi: 10. 1038 / nature 14581. PubMed PMID: 26106858; PMCID: 4825698. Reategui E, van der Vos KE, Lai CP, Zeinali M, Atai NA, Aldikacti B, Floyd FP, Jr., A HK, Thapar V, Hochberg FH, Sequist LV, Nahed BV, B SC, Toner M, Balaj L. D TT, Breakefield XO, Stott SL. Engineered nanointerfaces for microfluidic isolation and molecular profiling of tumor-specific extracellular vesicles. Nat Commun. 2018;9(1): 175. doi: 10.1038 / s41467-017-02261-l . PubMed PMID: 29330365; PMCID: 5766611. Dong J, Zhang RY, Sun N, Smalley M, Wu Z, Zhou A, Chou SJ, Jan YJ, Yang P, Bao L, Qi D, Tang X. Tseng P, Hua Y, Xu D. Kao R, Meng M, Zheng X. Liu Y, Vagner T. Chai X, Zhou D, Li M, Chiou SH, Zheng G, Di Vizio D, Agopian VG, Posadas E, Jonas SJ, Ju SP, Weiss PS, Zhao M, Tseng HR, Zhu Y. Bio-Inspired NanoVilli Chips for Enhanced Capture of Tumor-Derived Extracellular Vesicles: Toward Non-Invasive Detection of Gene Alterations in Non-Small Cell Lung Cancer. ACS Appl Mater Interfaces. 2019;! 1(15): 13973-83. Epub 20190402. doi: 10.1021 / acsami.9b01406. PubMed PMID: 30892008; PMCID: PMC6545291. Dong J, Zhang RY, Sun N, Hu J, Smalley MD, Zhou A, Yue H, Rothermich W, Chen M, Chen J, Ye J, Teng PC, Qi D, Toretsky JA, Tomlinson JS, Li M, Weiss PS, Jonas SJ, Federman N, Wu L, Zhao M. Tseng HR. Zhu Y. Coupling Nanostructured Microchips with Covalent Chemistry Enables Purification of Sarcoma-Derived Extracellular Vesicles for Downstream Functional Studies. Adv Funct Mater. 2020;30(49). Epub 20200913. doi: 10.1002 / adfin.202003237. PubMed PMID: 34220409: PMCID:PMC8248519. Wang JJ, Sun N, Lee YT, Kim M. Vagner T, Rohena-Rivera K, Wang Z, Chen Z, Zhang RY, Lee J, Zhang C, Tang H. Widjaja J, Zhang TX, Qi D. Teng PC, Jan YJ, Hou KC, Hamann C, Sandler HM, Daskivich TJ, Luthringer DJ, Bhowmick NA, Pei R, You S, Vizio DD, Tseng HR, Chen JF, Zhu Y, Posadas EM. Prostate cancer extracellular vesicle digital scoring assay - a rapid noninvasive approach for quantification of diseaserelevant rnRNAs. Nano Today. 2023;48. Epub 20230103. doi: 10.1016 / j.nantod.2022.101746. PubMed PMID: 36711067; PMCID: PMC9879227. Sun N, Tran BV, Peng Z, Wang J, Zhang C, Yang P, Zhang TX, Widjaja J, Zhang RY, Xia W, Keir A, She JW, Yu HH, Shyue JJ, Zhu H, Agopian VG, Pei R, Tomlinson JS, Toretsky JA, Jonas SJ, Federman N, Lu S, Tseng HR, Zhu Y. Coupling Lipid Labeling and Click Chemistry Enables Isolation of Extracellular Vesicles for Noninvasive Detection of Oncogenic Gene Alterations. Adv Sci (Weinh). 2022;9(14):e2105853. Epub 20220309. doi: 10.1002 / advs.202105853. PubMed PMID: 35486030; PMCID: PMC9108594. Sun N, Zhang C, Lee YT. Tran BV, Wang J, Kim H. Lee J, Zhang RY, Wang JJ, Hu J, Zhang Z, Alsudaney MS, Hou KC, Tang H. Zhang TX, Liang IY. Zhou Z, Chen M, Yeh AH, Li W, Zhou XJ, Chang HR, Han SB, Sadeghi S, Finn RS, Saab S, Busuttil RW, Noureddin M, Ayoub WS, Kuo A, Sundaram V, Al-Ghaieb B, Palomique J, Kosari K, Kim IK, Todo T, Nissen NN, Tomasi ML, You S, Posadas EM, Wu JX, Wadehra M, Sim MS, Li Y, Wang HL, French SW, Lu SC, Wu L, Pei R, Liang L, Yang JD, Agopian VG, Tseng HR, Zhu Y. HCC EV ECG score: An extracellular vesicle-based protein assay for detection of early-stage hepatocellular carcinoma. Hepatology. 2022. Epub 2022 / 08 / 01. doi: 10. 1002 / hep.32692. PubMed PMID: 35908246. Beird HC, Bielack SS, Flanagan AM, Gill J, Heymann D, Janeway KA, Livingston JA, Roberts RD. Strauss SJ, Gorlick R. Osteosarcoma. Nat Rev Dis Primers. 2022;8(l):77. Epub 20221208. doi: 10.1038 / s41572- 022-00409-y. PubMed PMID: 36481668. Meltzer PS, Helman LJ. New Horizons in the Treatment of Osteosarcoma. N Engl J Med. 2021;385(22):2066-76. doi: 10.1056 / NEJMra2103423. PubMed PMID: 34818481. Gill J, Gorlick R. Advancing therapy for osteosarcoma. Nat Rev Clin Oncol. 202I;I8(10):609-24. Epub 20210615. doi: 10. 1038 / s41571-021-00519-8. PubMed PMID: 34131316. Spencer HLM, Shnyder SD, Loadman PM, Falconer RA. The role of MT1-MMP in theprogression and metastasis of osteosarcoma. Journal of Cancer Metastasis and Treatment. 2022;8:2. doi: 10.20517 / 2394-4722.2021.174. Thuault S. Ghossoub R. David G. Zimmermann P. A Journey on Extracellular Vesicles for Matrix Metalloproteinases: A Mechanistic Perspective. Front Cell Dev Biol. 2022;10:886381. Epub 20220520. doi: 10.3389 / fcell.2022.886381. PubMed PMID: 35669514; PMCID: PMC9163832. Roomi MW. Kalinovsky T, Rath M, Niedzwiecki A. In vitro modulation of MMP-2 and MMP-9 in pediatric human sarcoma cell lines by cytokines, inducers and inhibitors. Int J Oncol. 2014;44( l):27-34. Epub 20131031. doi: 10.3892 / ijo.2013.2159. PubMed PMID: 24190483; PMCID: PMC3867366. Chi X, Ji T, Li J, Xu J, Tang X, Xie L, Meng F, Guo W. Genomic Analysis Revealed Mutational Traits Associated with Clinical Outcomes in Osteosarcoma. Cancer Manag Res. 2021;13:5101-11. Epub 20210628. doi: 10.2147 / CMAR.S317809. PubMed PMID: 34234554; PMCID: PMC8254031. Hamer I, Delaive E, Dieu M, Abdel-Sater F, Mercy L. Jadot M, Amould T. Upregulation of cathepsin B expression and enhanced secretion in mitochondrial DNA- depleted osteosarcoma cells. Biol Cell. 2009;101(l):31-41. doi: 10.1042 / BC20080043. PubMed PMID: 18598236. Wang Y, Tian X, Zhang W, Zhang Z, Lazcano R, Hingorani P, Roth ME, Gill JD, Harrison DJ, Xu Z, Jusu S, Kannan S. Wang J, Lazar AJ, Earley EJ, Erickson SW, Gelb T, Huxley P, Lahdenranta J, Mudd G, Kurmasheva RT, Houghton PJ, Smith MA, Kolb EA, Gorlick R. Comprehensive Surfaceome Profiling to Identify and Validate Novel Cell-Surface Targets in Osteosarcoma. Mol Cancer Ther. 2022;21(6):903-13. doi: 10.1158 / 1535-7163. MCT-21-0836. PubMed PMID: 35312779; PMCID: PMC9167717. Wang L, Zhang Q, Chen W, Shan B, Ding Y, Zhang G, Cao N, Liu L. Zhang Y. B7-H3 is overexpressed in patients suffering osteosarcoma and associated with tumor aggressiveness and metastasis. PLoS One. 2013;8(8):e70689. Epub 20130805. doi:10.1371 / joumal.pone.0070689. PubMed PMID: 23940627; PMCID: PMC3734259. Kendsersky NM, Lindsay J, Kolb EA. Smith MA, Teicher BA, Erickson SW. Earley EJ, Mosse YP, Martinez D. Pogoriler J, Krytska K, Patel K, Groff D, Tsang M, Ghilu S, Wang Y, Seaman S. Feng Y, Croix BS. Gorlick R. Kurmasheva R. Houghton PJ, Maris JM. The B7-H3-Targeting Antibody-Drug Conjugate m276-SL-PBD Is Potently Effective Against Pediatric Cancer Preclinical Solid Tumor Models. Clin Cancer Res.2021;27(10):2938-46. Epub 20210222. doi: 10.1158 / 1078-0432. CCR-20-4221. PubMed PMID: 33619171; PMCID: PMC8127361. Roth M, Barris DM, Piperdi S, Kuo V, Everts S, Geller D, Houghton P, Kolb EA, Hawthorne T, Gill J, Gorlick R. Targeting Glycoprotein NMB With Antibody-Drug Conjugate, Glembatumumab Vedotin, for the Treatment of Osteosarcoma. Pediatr Blood Cancer. 2016;63(l):32-8. Epub 20150825. doi: 10.1002 / pbc.25688. PubMed PMID: 26305408. Wang L, Kang FB, Zhang GC, Wang J, Xie MF, Zhang YZ. Clinical significance of serum soluble B7-H3 in patients with osteosarcoma. Cancer Cell Int. 2018; 18: 115. Epub 20180813. doi: 10.1186 / sl2935-018- 0614-z. PubMed PMID: 30123093; PMCID: PMC6090643. Kolb EA, Gorlick R, Billups CA, Hawthorne T, Kurmasheva RT, Houghton PJ, Smith MA. Initial testing (stage 1) of glembatumumab vedotin (CDX-011) by the pediatric preclinical testing program. Pediatr Blood Cancer. 2014;61( 10): 1816-21. Epub 20140609. doi: 10.1002 / pbc.25099. PubMed PMID: 24912408; PMCID: PMC4280502. Raposo G, Stahl PD. Extracellular vesicles: a new communication paradigm? Nature Reviews Molecular Cell Biology. 2019. doi: 10. 1038 / s41580-019-0158-7. PubMed PMID: 31324871. Gao X, Gao B, Li S. Extracellular vesicles: A new diagnostic biomarker and targeted drug in osteosarcoma. Front Immunol. 2022;13:1002742. Epub 20220923. doi: 10.3389 / fimmu.2022.1002742. PubMed PMID: 36211364; PMCID: PMC9539319. Kolb HC, Finn MG, Sharpless KB. Click Chemistry: Diverse Chemical Function from a Few Good Reactions. Angewandte Chemie International Edition. 2001;40(l l):2004-21. doi: 10.1002 / 1521- 3773(20010601)40: l l<2004::aid-anie2004>3.3.co;2-x. PubMed PMID: 11433435. Blackman ML, Royzen M, Fox JM. Tetrazine ligation: fast bioconjugation based on inverse-electron- demand Diels-Alder reactivity. J Am Chem Soc. 2008; 130(41): 13518- 9. doi: 10.102 l / ja8053805. Karver MR, Weissleder R, Hilderbrand SA. Synthesis and evaluation of a series of 1,2,4,5-tetrazines for bioorthogonal conjugation. Bioconjug Chem. 2011 ;22(11):2263- 70. doi: 10.1021 / bc200295y. Neumann U, Kubota H. Frei K, Ganu V. Leppert D. Characterization of Mca-Lys-Pro-Leu-Gly-Leu-Dpa- Ala-Arg-NH2, a fluorogenic substrate with increased specificity constants for collagenases and tumor necrosis factor converting enzyme. Anal Biochem. 2004;328(2): 166-73. doi: 10.1016 / j.ab.2003.12.035. PubMed PMID: 15113693.88. Decaneto E, Suladze S, Rosin C, Havenith M, Lubitz W, Winter R. Pressure and Temperature Effects on the Activity and Structure of the Catalytic Domain of Human MT1-MMP. Biophys J. 2015;109(l 1):2371-81. doi: 10.1016 / j.bpj.2015.10.023. PubMedPMID: 26636948; PMCID: PMC4675863.

[0083] The following provides some further examples according to an embodiment of the current invention.

[0084] Overall workflow of the OS EV MMP Activity Assay (FIG. 9). Step 1. Click chemistry-mediated enrichment of three subpopulations of OS EVs using EV Click MagBeads. OS EVs are enriched using EV Click MagBeads through biorthogonal Click Chemistry. 7ra / / .s-cyclooctene (TCO)-modified antibodies for three OS-specific markers label the EVs, which are then mixed with the EV Click MagBeads functionalized with methyltetrazine (mTz). Click reaction between the mTz-grafted EV Click MagBeads and TCO-grafted OS EVs immobilizes OS EVs on the bead surface. Magnetic isolation yields enriched OS EVs. Step 2. FRET-based peptide cleavage assay for detecting MMP activity of each subpopulation of OS EVs. Enriched OS EVs immobilized on EV Click MagBeads are directly incubated with surface MMP14 substrate or incubated with intra-vesicular MMP2 / 9 substrate after lysis. Proteolytic activity of MMPs generate a fluorescence readout, which yields OS EV MMP Activity Signatures correlating with progression of OS.

[0085] Characterization and validation of OS EVs (FIGS. 10A-10E). A) Size distribution of HOS and 143B EVs measured using dynamic light scattering (DLS). B) Representative transmission electron microscopy (TEM) images of HOS EVs and 143B EVs (scale bar = 200 nm). C) Scanning electron microscopy (SEM) images of HOS EVs and 143B EVs (scale bar = 500 nm). D) Representative SEM images (scale bar = 500 nm) of EV Click MagBeads with immobilized HOS EVs (left) and 143B EVs (right) using anti-B7-H3. Inset depicts how TCO-labeled EVs are immobilized onto an EV Click MagBead. F)Representative TEM images (scale bar = 100 nm) of HOS EVs (top) and 143B EVs (bottom) immobilized on EV Click MagBeads followed by immunogold labelling of CD63, a representative EV marker.

[0086] Clinical validation of OS EV MMP Activity Assay for detecting OS and metastatic status (FIG. 11). Heatmaps of MMP14 and MMP2 / 9 activity readouts analyzed by FRET-based MMP activity assay of plasma samples from healthy donors (HD, n = 6), patients with localized OS (LOC, n = 24), and patients with metastatic OS (MET, n = 4).

[0087] Clinical validation of OS EV MMP Activity Assay for detecting OS and metastatic status (FIG. 12). Significantly higher (P > 0.05) MMP14 activity from LRRC15+EVs, GPNMB+EVs, and B7-H3+EVs, and MMP2 / 9 activity from LRRC15+EVs, GPNMB+EVs, and B7-H3+EVs were observed in plasma from patients with metastatic OS compared to those from heathy donors.

[0088] Dynamic performance of OS EV MMP Activity Assay for monitoring OS (FIG. 13). Fluorescence readouts from OS EV MMP Activity Assay were obtained from plasma samples of two OS patients over the course of treatment and compared with conventional images. For an OS patient OST-003, magnetic resonance (MR) and CT images were taken before and after treatment. The MMP activities of OS EVs were maintained highly throughout the treatment, indicating possibility of disease progression (CTx, chemotherapy; RTx, radiotherapy; S, surgery).

[0089] Dynamic performance of OS EV MMP Activity Assay for monitoring OS (FIG. 14). Fluorescence readouts from OS EV MMP Activity Assay were obtained from plasma samples of two OS patients over the course of treatment and compared with conventional images. For an OS patient OST-005, MR image was taken before and X-ray images were taken after treatment. The MMP activities of OS EVs were found to be reduced and stabilized after the treatment, indicating possibility of disease stabilization (CTx, chemotherapy; RTx, radiotherapy; S, surgery).FURTHER EXAMPLE 1 - Noninvasive assessment of protease activity in osteosarcoma via click chemistry-mediated enrichment of extracellular vesiclesIntroduction

[0090] Osteosarcoma (OS) is the most common bone cancer of childhood, characterized by aggressive primary lesions with a high propensity for metastasis (7). While most OS patients are believed to have subclinical metastasis at diagnosis (2), 15-20% present with determinable metastasis initially (3). Current standard-of-care treatment protocols for OS typically involve a combination of neoadjuvant chemotherapy, surgery, and adjuvant chemotherapy, depending on histological grade and the risk of subclinical metastasis (4). Tire presence of metastatic lesions significantly reduces overall survival rates in OS patients, from -70% for localized OS to 35% or lower for metastatic OS (5). Consequently, early detection of OS metastasis and frequent monitoring of metastatic progression during treatment could markedly improve patient outcomes (2). Radiographic imaging, including CT, MRI, and PET scans, remains the gold standard for assessing disease progression and treatment responses of OS patients. However, implementation of frequent imaging assessment in OS patients faces several major challenges, including limited resolution, as well as resource restrictions associated with the frequency, costs, and access in most clinical settings. This highlights an unmet need for complementary, novel diagnostic tools capable of noninvasive and frequent monitoring of OS patients to assess the disease progression and treatment responses of OS patients.

[0091] Proteases serve as a cornerstone for tumor invasion and metastasis (6. 7) by remodeling the tumor microenvironment (5, 9). Therefore, developing quantitative and reproducible methods to assess protease activity in tumors offers a promising avenue for novel diagnostic approaches (70). One of the most common methods for assessing protease activity employs fluorescence resonance energy transfer (FRET) peptide probes. However, current approaches for assessing protease activities within tumor tissues involve complicated procedures, including the invasive acquisition of the tumor tissues, followed by the labor-intensive purification of proteases (11-13). This is particularly challenging for osteosarcoma and other hard-tissue sarcomas, as the decalcification requirement for bone tissue biopsies severely reduce the protease activity. Alternatively, nanoparticle -based protease probes, known as activity-based biomarkers, have been developed by Bhatia and coworkers (14). After systemic administration, the end-product of the probes — cleaved by the target proteases — was collected from urine samples for mass spectrometry analysis to assess the activity of the target proteases. Although promising, it faces clinicaltranslational limitations (75). In light of these technical challenges, “liquid biopsy” approaches, which capture proteases secreted by parental tumors into blood, could offer a more feasible and minimally invasive means of assessing protease activities (76, 77).

[0092] Extracellular vesicles (EVs) represent a heterogeneous group of phospholipid bilayer- enclosed nanoparticles (18, 19) secreted by all cell types, including tumor cells and those within the tumor microenvironment (20, 21). Notably, tumor-derived EVs circulate in the bloodstream at relatively early stages of disease and remain detectable (108— IO10per m of plasma) throughout all stages (22. 23). The half-life of tumor EVs ranges from 1 to 5.5 hours, with 90% undergoing hepatic clearance (24-26). The membranes of tumor-derived EVs not only display surface markers (27) of the parental tumor but also protect fragile biomolecular cargos (e.g., proteins, RNA, and DNA), thus reflecting the biomolecular profile of tire parental tumor. Importantly, tumor-derived EVs play crucial roles in transporting proteases to facilitate tumor invasion and metastasis (28-30). Given these distinguishing features, tumor-derived EVs serve as an ideal surrogate for tumor tissue, offering noninvasive access to tumor-associated proteases (8. 9). In particular, a family of matrix metalloproteinases (MMP), including MMP14 and MMP2, are transported by OS EVs (31-33) to promote OS invasion and metastasis (34). MMP14 is a potent predictor of poor prognosis in OS (32, 33), while elevated MMP2 activity is correlated with aggressive OS characteristics (35). Therefore, assessing MMP activity in OS-derived EVs could facilitate noninvasive and timely monitoring of disease progression and treatment response for OS patients, complementing radiographic imaging data for more comprehensive disease management.

[0093] To harness the potential of OS EVs for the noninvasive assessment of OS-associated MMP activities, it is essential to develop a system that effectively enriches OS EVs with intact MMP activities. Our team has pioneered the development of click chemistry-mediated tumor EV enrichment technologies, namely Click Chips (36-39) and Click Beads (40, 41). These approaches offer the following enhancements: i) reduced antibody consumption (by 10-100 times) and labeling time via labeling of tumor-derived EVs with click motif-grafted antibodies in small volumes, and ii) a highly biorthogonal click reaction (42. 43) between / ra -cyclooctciic (TCO) and tetrazine (Tz) motifs, effectively circumventing biological interference. Our recent studies have shown that click chemistry -mediated enrichment of tumor-derived EVs can be seamlessly coupled with various downstream molecular analyses, such as mRNA profiling (36-40) and protein detection (47). By integrating FRET peptide probes as downstream readouts (44, 45), we may noninvasively assess MMP activities in OS.

[0094] Here, we introduce an OS EV MMP Activity Assay, a noninvasive platform to assess MMP activities in OS through a two-step process (FIG. 15): i) click chemistry-mediated enrichment of three subpopulations of OS EVs using EV Click MagBeads in tire presence of one of the three TCO-grafted antibodies targeting the respective OS EV surface markers (LRRC15, GPNMB. and B7- H3), and ii) FRET peptide probes for the assessment of membrane-type MMP 14 or intravesicular MMP2 activities in the enriched OS EVs. Using only 1.5 mL of plasma (250 pL per each combination), the OS EV MMP Activity Assay can assess MMP activities across six OS EV marker / MMP combinations, resulting in a unique MMP Activity profile for each OS patient. To build the six marker / MMP combinations, a biomarker discovery framework was utilized to identify markers that were validated through immunofluorescence (IF) of OS cell lines and immunohistochemistry (IHC) of an OS tissue microarray (TMA). Following validation, the assay was conducted on plasma samples from 34 healthy donors (HD), 24 localized OS patients, and 10 metastatic OS patients, generating a distinct OS EV MMP activity profile for each subject. Based on the receiver operating characteristic (ROC) analysis of each combination for separating OS from HD, three top-perfonning combinations — LRRC15 / MMP2. GPNMB / MMP14. and B7-H3 / MMP14 OS EVs — were identified. To maximize their diagnostic performance in distinguishing metastatic OS from localized OS, an OS EV MMP Activity Score was developed by integrating these combinations via logistic regression. This Score exhibited excellent diagnostic performance, achieving an area under receiver operating characteristic (AUROC) curve of 0.97. Applying this OS EV MMP Activity Score to longitudinally monitor six OS patients through treatment, including chemotherapy and surgery, revealed strong correlations between the OS EV MMP Activity Score and patient disease progression and treatment response, as validated by radiographic imaging such as CT. PET. and MRI.ResultsIdentification of OS EV surface markers and OS EV-associated MMPs

[0095] We developed an integrated EV biomarker selection framework (FIG. 16) to identity the three OS EV surface markers (i.e., LRRC15, GPNMB, and B7-H3) and the two OS EV-associated MMPs (i.e.. membrane-type MMP14 and intravesicular MMP2). Our approach entailed compiling databases of OS cell line proteins (CCLE) and OS tissue RNA (46), followed by excluding nonspecific markers, such as housekeeping and hematopoietic proteins, to select OS-specific markers. We then identified EV-spccific markers by assessing the number of studies observed in cancer EV studies. Further refinement steps were taken to identify OS EV-specific markers, including thefiltering of highly expressed markers in OS and excluding tire markers related to the hematopoietic lineage, yielding 62 OS EV marker candidates. These candidates were subsequently categorized into two groups: OS EV surface markers and OS EV-associated proteases. For OS EV surface markers, we selected three candidates— LRRC 15 (47-49), GPNMB (50-52), and CD276 (B7-H3) (53, 54)— by filtering through the Cancer Surfacesome Atlas (55), and considering their relevance as molecular targets for OS therapies, such as monoclonal antibodies (mAb), antibody-drug conjugates (ADC), and chimeric antigen receptor T-cell (CAR-T) therapies. For OS EV-associated proteases, two representative MMPs — membrane-type MMP14 and intravesicular MMP2 — were selected due to their abundance and the commercial availability of compatible FRET peptide probes.Validation of OS EV surface markers and OS EV-associated MMPs

[0096] Tire essential component of the OS EV MMP Activity Assay (FIG. 15) is the colocalization of a pair of OS EV surface marker (i.e., LRRC15. GPNMB, or B7-H3) and OS- associated MMP (i.e., MMP 14 or MMP2), which we identified with an integrated EV biomarker selection framework. We validated their expression and colocalization using IF staining in the HOS cell line, a representative OS cell line, as well as IHC staining in 110 OS TMA samples. The representative IF images (FIG. 21) showed strong expression of all three surface markers and both proteases, with the surface markers predominantly localized on the cell membranes. In addition, colocalization of each OS EV surface marker with membrane-type MMP 14 and intravascular MMP2 was evident.

[0097] Next, expression and colocalization of three OS EV surface markers and two OS EV- associated MMPs were analyzed by hematoxylin and eosin (H&E) and IHC staining of the OS TMA samples (FIG. 17A). FIGS. 17B and 17C summarize the quantification of the IHC results for the three OS EV surface markers and two OS EV-associated MMPs, respectively. Regarding LRRC 15, among 80 evaluable OS samples (excluding those cases without tumor tissues captured on the TMA slides or tissue detached from tire glass slides during IHC staining), 3 (3.8%), 14 (17.5%), and 50 (62.5%) samples showed strong (3+). moderate (2+), and weak (1+) staining, respectively. GPNMB exhibited the highest positivity among the three OS EV surface markers, with 30 (29.7%), 47 (46.5%), and 23 (22.8%) out of 101 OS samples showing strong (3+), moderate (2+), and weak (1+) staining, respectively. For B7-H3, 6 (6.5%), 39 (41.9%), and 45 (48.4%) out of 93 OS samples showed strong (3+), moderate (2+), and weak (1+) staining, respectively. In addition, most OS samples showed positive staining for the two OS EV-associated MMPs. Specifically, 31 (31.6%), 49(50.0%), and 16 (16.3%) OS samples exhibited strong (3+), moderate (2+), and weak (1+) staining for MMP14. Additionally, for MMP2, 18 (18.9%), 38 (40.0%), and 39 (41.1%) OS samples were stained as strong (3+), moderate (2+), and weak (1+). Overall, all OS samples (75 out of 75) stained positive for at least two OS EV surface markers, with 82.7% showing triple positivity and 17.3% showing dual positivity (FIG. 17B, bottom). These data underscore tire complementary role of the three OS EV surface markers in achieving highly efficient enrichment of heterogeneous OS EVs using EV Click MagBeads. Moreover, all the OS samples (87 out of 87) stained positive for at least one of the OS EV-associated MMPs, with 96.6% showing dual positivity (FIG. 17C, bottom), validating the suitability of adopting both membrane-type MMP14 and intravesicular MMP2 as downstream readouts for the OS EV MMP Activity Assay. Overall, the IF and IHC results i) confinned the suitability of the identified OS EV surface markers and MMPs for the OS EV MMP Activity Assay, and ii) suggest that the combined use of these OS surface markers and MMPs could effectively account for the heterogeneous OS EVs, reflecting the diversity of OS tissues.Preparation and characterization of EV Click MagBeads

[0098] Expanding upon our previous click chcmistry-mcdiatcd EV enrichment technology, EV Click Beads (40. 41), here we prepared the EV Click MagBeads (FIGS. 22A-22E). This new EV enrichment tool offers key improvements: i) transition from silica beads to magnetic beads for quicker and efficient EV enrichment, and ii) adoption of a methyltetrazine (mTz) click motif for enhanced stability7, extending the storage life from two weeks to over three months. The synthetic process of EV Click MagBeads (FIG. 22A) started with the modification of amine-grafted Dynabcads™ (2.7 pm) using mcthyltctrazinc-PEG4-V-hydroxysuccinimidc (mTz-PEG4-NHS) ester via NHS ester chemistry to obtain mTz-Dynabeads. Subsequently, any residual amine groups were covalently capped with methoxypolyethylene glycol succinate-N-hydroxysuccinimide (mPEG4-NHS) ester, conferring anti-fouling properties to the resultant EV Click MagBeads. The alteration in surface chemistry of the magnetic beads were monitored by observing shifts in their zeta potentials (FIG. 22B). Starting from a significant positive charge (+10 mV) of the amine-grafted Dynabeads™, the stepwise addition of mTz-PEG4-NHS and mPEG4-NHS gradually neutralized tire zeta potential, indicating successful attachment of mTz-PEG and mPEG groups. Hie incorporation of mTz motifs on EV Click MagBeads was confirmed via a click reaction with TCO-labeled Cy5 (TCO-Cy5). Labeling the with TCO-Cy5 resulted in pronounced Cy5 fluorescence, observable through fluorescence microscopy (FIG. 22C). Image-based cytometry (FIG. 22D) further confirmed the presence of mTz groups and their capability to undergo click reaction with TCO groups. The TCO-Cy5 labeling method was enabled quantification of the mTz capacity on the EV Click MagBeads for monitoring their stability (FIG. 22E). The initial mTz level was measured at 0.464 ± 0.024 nmol / mg beads, with approximately 60% retention after six months. Based on these stability studies, all experiments were conducted within three months of bead synthesis to ensure that over 80% of the mTz remained active. Notably, the quantity of residual mTz on the EV Click MagBeads (exceeding 80% retention, 9 pmol mTz per batch) proved to be more than sufficient for capturing TCO-grafted antibodies (100 ng antibody, equating to 2 pmol TCO per batch, see below).Testing EV Click MagBeads with OS cell-derived EVs

[0099] To evaluate the performance of EV Click MagBeads, we harvested OS EVs from the conditioned media of HOS cell lines via ultracentrifugation (detailed in below). Harvested OS EVs were characterized by dynamic light scattering (DLS, FIG. 23A). nanoparticle tracking analysis (NTA. FIG. 23B). transmission electron microscopy (TEM, FIG. 23C). and scanning electron microscopy (SEM, FIG. 23D), in accordance with the guidelines outlined by the International Society for Extracellular Vesicles (MISEV 2023) (56).

[0100] After click-chemistry mediated immobilization of OS EVs onto EV Click MagBeads (FIG. 15), the interfaces between EV Click MagBeads and the immobilized OS EVs were analyzed. The inset of FIG. 24A illustrates the immobilization of TCO-anti-B7-H3 -labeled OS EVs on EV Click MagBeads through click chemistr -mediated EV enrichment. SEM imaging revealed HOS EVs on the surface of EV Click MagBeads, showing similar size and morphology to intact EVs (FIGS. 23A-23D), thus confirming the successful immobilization of OS EVs (FIG. 24A). The identity of the enriched EVs were further validated by immunogold TEM imaging (FIG. 24B). Following capture of HOS EVs, immunogold labeling of CD63, a representative EV surface marker, were detected on the surface of EV Click MagBeads, confirming that the immobilized particles are OS EVs. Overall, these studies validate the surface functionalization of EV Click MagBeads with mTz groups and their ability to immobilize OS EVs.Optimization of two MMP activity assays through enzyme kinetics

[0101] Following the OS EV enrichment process, we focused on optimizing the assessment of MMP proteolytic activities using two FRET peptide probes (detailed in below). We performed enzyme kinetics studies to determine the optimal substrate concentrations for MMPs. A kinetic study was conducted on activated forms of recombinant human MMP14 (rhMMP14, FIG. 25A) and MMP2(rhMMP2, FIG. 26A) by measuring the initial hydrolysis rate of the respective MMP FRET probe with a fixed amount of the MMP (FIGS. 25B and 26B). The resulting data were fitted to Michaelis- Menten equation to determine tire kinetic parameters of each MMP, yielding Kmvalues of 4.54 pM for MMP14 (FIG. 25C) and 11.5 pM for MMP2 (FIG. 26C), consistent with previous reports. (57, 58). We chose a concentration of 4.5 pM for FRET probes of both MMPs for the activity assays. Subsequently, we assessed the dynamic range and linearity of the MMP activity assay. The results revealed that these assays showed good linearities (R2> 0.95) within dynamic ranges of 0-80 nM (MMP14, FIG. 25D) and 0-20 nM (MMP2, FIG. 26D), respectively, sufficient for covering typical concentrations of MMPs in the clinical samples (59, 60). Given the structural similarity of MMPs, we evaluated the specificity of the MMP FRET probes against their target and non-target MMPs. The fluorescence readouts from non-target MMPs. including the MMP 14 FRET probe against MMP2 (FIG. 25E) or vice versa (FIG. 26E), were less than 10% of the target readout, confirming the specificity of the MMP FRET probes utilized in this study.Validation of OS EV MMP Activity Assay using synthetic plasma samples

[0102] Wc then validated the linearity of the two-step OS EV MMP Activity Assay (FIG. 15) using synthetic plasma samples, prepared by spiking HOS EVs into EV-depleted HD plasma. Concentration of HOS EVs in the stock solution was determined by measuring the number of EVs with NTA (detailed below). OS EVs in 250 pL of synthetic plasma were labeled with TCO-grafted antibodies targeting three validated OS EV surface markers (LRRC15, GPNMB, or B7-H3), followed by enrichment using EV Click MagBeads. To obtain the readouts of membrane-type MMP 14, the enriched OS EVs on EV Click MagBeads were directly incubated with a MMP14 FRET probe. In contrast, for the activity of intravesicular MMP2, the enriched OS EVs underwent lysis to release MMP2 from OS EVs. followed by incubation with a MMP2 FRET probe. The combination of three OS EV surface markers and two FRET probes for OS EV-associated MMP enabled the assessment of MMP activities across six distinct OS marker / MMP combinations of OS EVs (FIG. 18B-18G). A robust linear correlation (R2> 0.90) was observed between the concentration of spiked HOS EVs and the detected MMP activity across all six combinations, within the range of 0.5-4 x 109EVs / mL, which adequately covers typical concentrations of tumor-derived EVs observed from the patients’ plasma (22, 23).Diagnostic performance of OS EV MMP Activity Assay for detecting OS

[0103] Before performing OS EV MMP Activity Assays on clinical samples, we conducted a pilot study to examine the essential role of OS EV enrichment. Specifically, we compared MMP activities before and after tire enrichment of OS EVs from plasma samples to separate OS patients (n=6) from HDs (n=6). As shown in FIG. 27A-27B, high MMP 14 and MMP2 activities were detected in both HD and OS plasma samples prior to EV enrichment, while B7-H3-enriched OS EVs enabled separating OS patients from HDs. These findings indicate that, without appropriate OS EV enrichment, the MMP activities could not demonstrate effective performance.

[0104] Next, we examined the feasibility of OS EV MMP Activity Assay (FIG. 19A) by measuring MMP activities of the six OS marker / MMP combinations using 1.5 mL of plasma samples (six combinations with 250 pL for each). These plasma samples were obtained from three groups: healthy donors (HD, n=34), patients with localized OS (n=24), and patients with metastatic OS (n=10). The clinical characteristics of the 34 patients with either localized or metastatic OS. alongside the 34 healthy donors are summarized in Tables 1 and 2 (FIGS. 32, 33). A pediatric oncologist, who was blinded to the assay results, performed the clinical annotation of all plasma samples. The quantitative OS EV MMP Activity profile obtained through the six OS marker / MMP combinations were organized and presented in a heatmap, as shown in FIG. 19B.

[0105] In our analysis of six OS EV marker / MMP combinations, we identified three topperforming OS EVs (LRRC15 / MMP2, GPNMB / MMP14, and B7-H3 / MMP14) that demonstrated significantly higher MMP activities (P < 0.001) in OS patients compared to HDs (FIG. 19C and FIG. 28). The diagnostic performance of these combinations was further assessed using AUROCs, showing 0.99 for LRRC15 / MMP2, 0.95 for GPNMB / MMP14, and 0.94 for B7-H3 / MMP14 OS EVs, respectively. These three top-performing combinations also demonstrated the highest effectiveness (FIG. 20D and FIG. 29) to distinguish metastatic OS from localized OS, with AUROCs of 0.89 for LRRC15 / MMP2, 0.87 for GPNMB / MMP14, and 0.86 for B7-H3 / MMP14 OS EVs, respectively.

[0106] By adopting a logistic regression to synergistically integrate the top-performing combinations, we could establish the OS EV MMP Activity Score (FIG. 19E). This score is specifically designed to enhance the performance to distinguish metastatic OS from localized OS, addressing the needs for a noninvasive diagnostic tool to monitor diseases progression and treatment responses. The OS EV MMP Activity Score significantly distinguished (P < 0.0001) metastatic OS from localized OS patients (FIG. 19F, 19G). Hie diagnostic performance, resulting in AUROC = 0.97 (95% CI = 0.911-1.00) with sensitivity of 95.83% and specificity of 90% (FIG. 19H), wasoutperforming compared to the individual OS marker / MMP combinations, validating the use of the logistic regression model.Dynamic performance of OS EV MMP Activity Assay for monitoring OS.

[0107] To address the gap in developing noninvasive diagnostics that complement radiographic imaging for monitoring disease progression and treatment responses in OS patients, we utilized the OS EV MMP Activity Score to monitor (FIG. 20A) six OS patients (detailed in Table 3 (FIG. 34)) longitudinally over their therapeutic intervention, which included chemotherapy and surgery. We focused on two cases as examples: OST-005 with stable disease, and OST-003 with progressive disease. For OST-005, plasma samples were collected at 16, 37, 60, 95, 121, 134, 154 days after initiating chemotherapy. Surgery was performed on Day 15 to excise a substantial OS lesion near the femur. As shown in FIG. 21B, tire OS EV MMP Activity Scores dropped significantly as the result of the surgery. The low scores maintained, indicating effective disease management by the ongoing chemotherapy (from Day 1 to 160). CT imaging was instrumental in visualizing four key stages of the disease: the untreated tumor (Day -4), a bone fracture (Day 40), and the post-surgery states (Days 35 and 137). These images were in strong concordance with the trends observed in the OS EV MMP Activity Scores. For patient OST-003, serial plasma samples were collected on Day 11, 102, 106, 158. 223. and 230, subsequent to the initiation of their first chemotherapy cycle. Tire corresponding OS EV MMP Activity Scores for these time points are summarized in FIG. 21 C. Following amputation to remove a large OS lesion, the patient was found to have a partial response. Over the course of treatment, lung metastasis was detected on Day 158. Tire lung metastasis continued to grow, and a new metastatic site was identified on the spine as shown in CT images on Day 218. These radiographic imaging results accurately mirrored the progression of the disease, which was also reflected in the escalating OS EV MMP Activity Scores, indicating the spread of the disease. For other OS patients, we summarized the longitudinal monitoring of their OS EV MMP Activity Scores over their therapeutic interventions in FIGS. 30A-30D.

[0108] (Summary) The OS EV MMP Activity Assay addresses a critical gap in OS management by providing a noninvasive alternative for tracking disease progression and treatment response. Although radiographic imaging remains the gold standard for evaluating OS progression, practical limitations, such as cost, restricted access, and logistical barriers to frequent use, highlight the need for complementary diagnostic methods. Furthermore, current treatments involving long cycles of chemotherapy and surgery, along with frequent radiographic imaging, imposes a physicaland psychological burden for pediatric patients. The OS EV MMP Activity Assay harnesses the tumor-promoting activities of MMPs, pivotal in OS tumor invasion and metastasis, to offer a promising solution. By enabling the frequent, noninvasive detection and quantification of MMP activities directly from tumor-derived EVs in blood, our assay provides a real-time snapshot of the tumor microenvironment.

[0109] (Overall flow) To develop the OS EV MMP Activity Assay (FIG. 15), we first conducted marker selection through a rigorous four-layer EV biomarker selection framework (FIG. 16). which involved: i) identifying and selecting OS-specific markers, ii) selecting EV-specific markers, iii) refining these markers to focus on OS EV-specific targets, and iv) finalizing the combination of OS EV-specific markers and MMPs. This systematic approach identified three optimal OS EV surface markers (LRRC15, GPNMB, and B7-H3) along with two OS EV-associated MMPs (MMP2 and MMP14). Subsequent validation using IF on OS cell lines and IHC on an OS TMA confirmed the colocalization of these markers and MMPs in OS tissues (FIGS. 17A-17C).

[0110] To meet the standardized guidelines for EV characterization outlined in MISEV2023 (56). we performed comprehensive analyses of OS EVs using various techniques, including NTA, TEM, and SEM (FIGS. 23A-23D). We observed the intact EVs immobilized on the surface of EV Click MagBeads, confinning the click chemistry-mediated enrichment of OS EVs using the identified OS EV surface marker.

[0111] After testing the performance of EV Click MagBeads, we optimized the perfonnance of the OS EV MMP Activity Assay using synthetic plasma samples. The Michaelis-Menten equation was employed to model the initial hydrolysis rate of the FRET peptide probes, allowing us to optimize kinetic parameters of MMPs (e.g., Vmax and Km). This kinetic optimization establishes conditions that enhance detection accuracy and ensure robust sensitivity across a wide dynamic range. Importantly, the assay demonstrated strong linearity in both MMP (FIGS. 24A-24B and FIGS. 25A- 25E) and OS EV (FIGS. 18A-18G) detection ranges in clinical samples.

[0112] (Establishing OS score) Following assay optimization using synthetic plasma samples, we conducted a retrospective case-control study, applying the OS EV MMP Activity Assay to clinical plasma samples from 34 HDs, 24 localized OS patients, and 10 metastatic OS patients. The Assay demonstrated excellent diagnostic perfonnance (AUROC > 0.95 for MMP 14 activity and AUROC > 0.85 for MMP2 activity) in distinguishing OS patients from HDs (FIG. 18B, FIG. 28 and FIG. 29). Among the six tested marker / MMP combinations, the top three performers —LRRC15 / MMP2, GPNMB / MMP14, and B7-H3 / MMP14 OS EVs —were selected for effectively distinguishing (AUROC > 0.85) metastatic from localized OS patients (FIG. 19D). By integrating these three combinations via logistic regression, we established the OS EV MMP Activity Score, achieving an exceptional AUROC of 0.97 for distinguishing metastatic from localized OS patients (FIG. 19H). This high diagnostic performance demonstrates the Assay’s potential in clinical applications, offering a noninvasive, quantitative, and reproducible approach to monitoring OS progression and stratifying disease stage.

[0113] (Patient monitoring) Once the OS EV MMP Activity Score was established, we utilized it to longitudinally monitor six OS patients throughout their treatment, comparing results with standard radiographic imaging to validate its clinical utility. This Score specifically addressed a critical disease window in distinguishing metastatic from localized OS patients — a challenging yet crucial aspect of monitoring disease progression and treatment response. Blood samples were collected at significant clinical milestones, including chemotherapy and surgery, throughout the standard-of-care treatment for OS patients to correlate the Assay’s output with treatment events. In the case of a patient with stable disease (FIG. 20B), the OS EV MMP Activity Score dropped significantly following surgery and remained low, indicating effective disease control achieved through chemotherapy. This result aligned with CT imaging, confirming the patient’s stable disease status. Conversely, in the patient with progressive disease (FIG. 20C), the Score initially decreased post-amputation, which removed a large OS lesion, but subsequently increased, correlating with the appearance of lung metastasis and a new lesion on the spine, as confirmed by imaging. The rising Scores thus accurately reflected disease progression as confirmed by radiographic imaging. Similar trends were observed in four other patients (FIGS. 30A-30D), where declining Scores correlated with stable disease, while rising Scores indicated progression. These findings reinforce the potential of the OS EV MMP Activity Assay to reliably track patients through this crucial disease window, providing frequent, noninvasive monitoring to support clinical management for OS patients.

[0114] (Specificity / sensitivity) The OS EV MMP Activity Assay operates by detecting the colocalization of a pair of OS EV surface markers (for enrichment) and OS EV-associated MMPs (for quantification), thereby enabling high specificity and sensitivity. The core concept of this technology is that OS EVs exhibiting colocalization of a pair of OS EV surface marker (i.e.. LRRC15. GPNMB, or B7-H3) and OS-associated MMP (z.e., MMP14 or MMP2) will exclusively generate positive readouts, thereby enhancing specificity. Additionally, the high sensitivity of the OS EV MMP Activity Assay is significantly enhanced through click chemistry-mediated EV enrichmenttechnology, which allows for greater capture efficiency and minimizes the risk of biological interference, surpassing traditional EV isolation methods 61-64). Sensitivity is further improved by signal amplification through the enzymatic cleavage of FRET peptide probes, which specifically target MMP activity. Additionally, the OS EV MMP Activity Assay requires only 1.5 mL of plasma (~3 mL of whole blood), offering a minimally invasive alternative that can be performed frequently without additional risk to the pediatric patient.

[0115] (Therapeutic implication) This accessibility is especially valuable given the severe toxicity of current chemotherapeutic regimes and the urgent need for alternative, novel treatments, such as ADCs and CAR-T therapies, in refractory OS cases. Leveraging the Assay’s EV biomarker selection framework (FIG. 16), we purposefully selected LRRC15, GPNMB, and B7-H3 as not only for diagnostic potential but also for their therapeutic implications. LRRC15, for instance, is overexpressed in OS (65) and represents a promising candidate for OS-specific ADCs (47-49); similarly, GPNMB. the target of glembatumumab vedotin {50, 51), and B7-H3 {53. 54), make promising targets for both ADC and CAR-T therapies and are already in various stages of clinical evaluation. We envision the OS EV MMP Activity Assay as a companion diagnostic tool to support the development of these OS immunotherapies, guiding targeted therapeutic interventions to improve outcomes in OS management.

[0116] (Broad applicability) Moreover, given the crucial role of proteases in tumor invasion and metastasis, substantial efforts have focused on developing quantitative and reproducible technologies to assess tumor protease activities. The OS EV MMP Activity Assay stands as a pioneering example, offering a noninvasive approach to monitor tumor-associated protease activity with high sensitivity and specificity. While we developed this assay specifically for OS in the current embodiment, the technology’s foundation — a platfonn for assessing MMP activity in enriched EVs — can extend beyond OS to address other cancer types. Since most cancers often overexpress proteases to facilitate invasion and metastasis, detecting the protease activity of cancer-derived EVs could serve as a reliable indicator of the metastatic potential of parental solid tumors. Furthermore, given the role of MMPs in inflammatory diseases and secretases in neurodegenerative conditions like Alzheimer’s, the assay’s adaptable framework — modifying targets to enrich disease-specific EVs to assess the activity of relevant proteases — positions it as a transformative tool for a broad spectrum of diseases. As a result, it promises versatile, minimally invasive monitoring across diverse clinical contexts.

[0117] (Limitations) As with any novel technology, the OS EV MMP Activity Assay has limitations. One concern is the potential variability in EV marker expression across different patient populations, which could impact the sensitivity and specificity of the assay. To address this, our biomarker selection framework prioritized surface markers with consistent expression across a wide range of OS cases, validated through extensive bioinformatics analysis and experiments. Additionally, the relatively small cohort in this proof-of-concept and retrospective case-control study limit the statistical power to fully demonstrate the assay’s diagnostic capabilities in clinical settings. Although the OS cases — regarded as rare cancer — and tire number of available samples are limited, larger-scale studies will be necessary to validate these findings and further refine the assay for clinical application.

[0118] (Summary) Overall, the OS EV MMP Activity Assay represents a significant advancement in noninvasive, quantitative, and reproducible assessment of MMP activities within OS EVs. This study also highlights a next generation of our click chemistry-mediated EV enrichment technology, integrating FRET peptide probes for functional analyses of tumor-derived EVs beyond proteins and mRNA quantification (36-41). While clinical translation may present challenges, the potential impact on patient outcomes is immense, not only redefining the approach to liquid biopsy but also paving the way for improved management of rare and pediatric cancers.Materials and MethodsBiomarker discovery framework for OS EV surface markers and OS EV-associated MMPs

[0119] Two datasets were employed for the selection of overall OS-related markers: i) Cancer Cell Line Encyclopedia (CCLE) (66) protein dataset, 11,642 proteins detected in OS samples, and ii) OS tissue RNA data (46), 472 upregulated differentially expressed genes (DEGs) were identified. After excluding housekeeping and hematopoietic proteins (67), tire markers identified in more than two cancer studies from Vesiclepedia were selected as EV-specific markers. With filtration in the selection of high-expression proteins using expression level (top 20%, > 0.42) from CCLE protein data, both cell line protein- and tissue RNA-based OS EV markers underwent additional filtration based on RNA data. In specific, cell line protein markers were filtered based on their bone cancer expression levels from CCLE RNA data (top 2.5%, > 8.19) (68), while tissue RNA- based OS EV markers were filtered based on their expression from TARGET-OS data (top 2.5%, > 73.47) (69). Next, the markers were further refined based on their expression levels in 38 distinct human hematopoietic lineages from The Differentiation Map (DMAP) data (70). Markers below themedian expression level of 7 in DMAP data were chosen to remove background signals from the blood cell types. Consequently, 32 OS EV markers were identified from the cell line protein dataset, and 31 OS EV markers were identified from the tissue RNA dataset. Furthermore, CD276 (B7-H3) and MMP14 were added based on cancer-specific Gene Encoding Surface Protein (caGESP) (55) list of sarcomas including osteosarcoma as a subtype, resulting in the pooled 62 OS EV marker candidates. Among the marker candidates, five surface markers (z.e., MMP14, MYO 10, GPNMB, LRRC15, CD276) identified from the caGESP in at least one cancer type of The Cancer Surfacesome Atlas (TCSA) (55) were selected, and the final three markers — LRRC15, GPNMB, and CD276 (B7- H3) — which have been used for OS-targeted therapies (e.g., mAb, ADCs, or CAR-T) were selected as OS EV surface markers in this study. On the other hand, seven OS EV-associated proteases (i.e., MMP14, MMP2, MMP9. MMP13, UCHL1, CTSK. and CPE) were identified based on their protease activity, and the final two proteases — MMP 14 and MMP2 — from top ranking membrane-type and intravesicular MMP, respectively, were selected as OS EV-associated MMPs in this study.Patient enrollment

[0120] For monitoring study, a total of 6 patients with osteosarcoma were enrolled between February 2016 to August 2023 at University of California, Los Angeles (UCLA) Santa Monica Medical Center (Table 3). Additionally, 34 healthy donors were enrolled between November 2022 to November 2023 at UCLA Blood and Platelet Center (Table 1). All participants provided written informed consent for this study according to the institutional review board (IRB) protocols #14- 000197, #10-000236-AM-00021, #20-001197 at UCLA.Clinical Blood Sample Processing

[0121] All blood samples from osteosarcoma patients and healthy donors at UCLA were collected with written infonned consent according to the institutional review board protocols (IRB #19-000857). Each blood sample was collected in a BD Vacutainer plastic tube (BD, Cat. #366643) with EDTA. Plasma samples were isolated within 4 h of collection. Briefly, blood samples were centrifuged at 500 x g for 10 min, supernatant was taken, followed by centrifugation at 4,600 x g for 10 min to remove any cells and debris. The obtained plasma samples were then aliquoted and stored at -80 °C before use. For the evaluation of clinical samples, 250 pL of plasma per OS marker / MMP combinations was analyzed by OS EV MMP Activity Assay under the conditions optimized through the synthetic plasma samples.

[0122] 28 plasma samples from osteosarcoma patients (Table 2) were purchased fromProteoGenex (Inglewood, CA). All osteosarcoma cases for establishing OS EV MMP Activity Score were treatment-naive at the time of blood collection.Statistical Analysis

[0123] The results are reported as mean ± SD. The statistic differences between groups were assessed using one-way ANOVA with Tukey's post-hoc test for multiple group comparison or an unpaired Student’s t-test for two-group comparisons. Significance levels are indicated as follows: ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05, and ns (not significant) > 0.05. Statistical analy sis was performed using GraphPad Prism. The optimal cutoff points for ROC analysis were determined to maximize sensitivity and specificity.

[0124] References for FURTHER EXAMPLE 11. H. C. Beird et al.. Osteosarcoma. Nat Rev Dis Primers 8, 77 (2022).2. G. Sheng, Y. Gao, Y. Yang, H. Wu, Osteosarcoma and Metastasis. Front Oncol 11, 780264 (2021).3. S. C. Kaste, C. B. Pratt, A. M. Cain, D. J. Jones-Wallace, B. N. Rao, Metastases detected at the time of diagnosis of primary pediatric extremity osteosarcoma at diagnosis: imaging features. Cancer 86, 1602-1608 (1999).4. P. S. Meltzer, L. J. Helman, New Horizons in the Treatment of Osteosarcoma. N Engl J Med 385, 2066-2076 (2021).5. S. Cole. D. M. Gianferante, B. Zhu. L. Mirabello. Osteosarcoma: A Surveillance. Epidemiology, and End Results program-based analysis from 1975 to 2017. Cancer 128, 2107-21 18 (2022).6. C. M. Overall, O. Kleifeld, Tumour microenvironment - opinion: validating matrix metalloproteinases as drug targets and anti-targets for cancer therapy. Nat Rev Cancer 6, 227- 239 (2006).7. S. D. Mason, J. A. Joyce, Proteolytic networks in cancer. Trends Cell Biol 21, 228- 237 (2011).8. J. Winkler, A. Abisoye-Ogunniyan, K. J. Metcalf, Z. Werb, Concepts of extracellular matrix remodelling in tumour progression and metastasis. Nat Commun 11, 5120 (2020).9. R. Roy. J. Yang, M. A. Moses, Matrix metalloproteinases as novel biomarkers and potential therapeutic targets in human cancer. J Clin Oncol 27, 5287-5297 (2009).10. A. Alaseem et al. , Matrix Metalloproteinases: A challenging paradigm of cancer management. Semin Cancer Biol 56, 100-115 (2019).11. S. P. Hawkes, H. Li, G. T. Taniguchi, Zymography and reverse zymography for detecting MMPs and TIMPs. Methods Mol Biol 622, 257-269 (2010).12. R. E. Manrow, R. P. Dottin, Renaturation and localization of enzymes in polyacrylamide gels: studies with UDPglucose pyrophosphorylase of Dictyostelium. Proc Natl Acad Sci USA 77, 730-734 (1980).13. S. A. Lacks, S. S. Springhom, Renaturation of enzymes after polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. J Biol Chem 255, 7467-7473 (1980).14. J. S. Dudani, A. D. Warren. S. N. Bhatia, Harnessing Protease Activity to Improve Cancer Care. Annual Review of Cancer Biology 2, 353-376 (2018).15. G. A. Kwong et al., Synthetic biomarkers: a twenty-first century path to early cancer detection. Nat Rev Cancer 21. 655-668 (2021).16. K. Kotaska, P. Dusek, R. Prusa, S. Vesely, M. Babjuk, Urine and serum cathepsin B concentrations in the transitional cell carcinoma of the bladder. J Clin Lab Anal 26, 61 -65 (2012).17. H. C. Yeh et cd.. Evaluation of serum matrix metalloproteinase (MMP)-9 to MMP-2 ratio as a biomarker in hepatocellular carcinoma. Hepatogastroenterology 57, 98-102 (2010).18. G. Raposo, W. Stoorvogel, Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol 200, 373-383 (2013).19. A. L. Revenfeld et al.. Diagnostic and prognostic potential of extracellular vesicles in peripheral blood. Clin Ther , 830-846 (2014).20. G. van Niel, G. D'Angelo, G. Raposo. Shedding light on the cell biology of extracellular vesicles. Nat Rev Mol Cell Biol 19, 213-228 (2018).21. Z . Gao et al., Emerging Role of Exosomes in Liquid Biopsy for Monitoring Prostate Cancer Invasion and Metastasis. Front Cell Dev Biol 9, 679527 (2021).22. K. B. Johnsen, J. M. Gudbergsson, T. L. Andresen, J. B. Simonsen, What is the blood concentration of extracellular vesicles? Implications for the use of extracellular vesicles as blood-borne biomarkers of cancer. Biochim Biophys Acta Rev Cancer 1871, 109-116 (2019).23. Y. Tian et al., Quality and efficiency assessment of six extracellular vesicle isolation methods by nano-flow cytometry. J Extracell Vesicles 9, 1697028 (2020).24. S. Ferguson, R. Weissleder, Modeling EV Kinetics for Use in Early Cancer Detection. Adv Biosyst 4, el 900305 (2020).25. M. Kang, V. Jordan, C. Blenkiron, L. W. Chamley, Biodistribution of extracellular vesicles following administration into animals: A systematic review. J Extracell Vesicles 10, 612085 (2021).26. M. Auber, P. Svenningsen, An estimate of extracellular vesicle secretion rates of human blood cells. Journal of Extracellular Biology 1, e46 (2022).27. K. Jerabkova-Roda, A. Dupas, N. Osmani, V. Hyenne, J. G. Goetz, Circulating extracellular vesicles and tumor cells: sticky partners in metastasis. Trends Cancer 8, 799- 805 (2022).28. L. Mashouri et al., Exosomes: composition, biogenesis, and mechanisms in cancer metastasis and drug resistance. Mol Cancer 18, 75 (2019).29. R. Xu et al. , Extracellular vesicles in cancer - implications for future improvements in cancer care. Nat Rev Clin Oncol 15, 617-638 (2018).30. M. Shimoda, R. Khokha, Proteolytic factors in exosomes. Proteomics 13, 1624-1636 (2013).31. Y. Wang et al.. Comprehensive Surfaceome Profiling to Identify and Validate Novel Cell-Surface Targets in Osteosarcoma. Mol Cancer Ther 21, 903-913 (2022).32. S. Thuault, R. Ghossoub, G. David, P. Zimmermann, A Journey on Extracellular Vesicles for Matrix Metalloproteinases: A Mechanistic Perspective. Front Cell Dev Biol 10, 886381 (2022).33. H. L. M. Spencer, S. D. Shnyder, P. M. Loadman, R. A. Falconer, Tire role of MT1- MMP in tire progression and metastasis of osteosarcoma. Journal of Cancer Metastasis and Treatment 8, 2 (2022).34. X. Gao, B. Gao, S. Li, Extracellular vesicles: A new diagnostic biomarker and targeted drug in osteosarcoma. Front Immunol 13, 1002742 (2022).35. M. W. Roomi, T. Kalinovsky, M. Rath, A. Niedzwiecki, In vitro modulation of MMP-2 and MMP-9 in pediatric human sarcoma cell lines by cytokines, inducers and inhibitors. IntJ Oncol 44, 27-34 (2014).36. N. Sun et al. , Purification of HCC-specific extracellular vesicles on nanosubstrates for early HCC detection by digital scoring. Nat Commun 11, 4489 (2020).37. J. Dong et al.. Coupling Nanostructured Microchips with Covalent Chemistry Enables Purification of Sarcoma-Derived Extracellular Vesicles for Downstream Functional Studies. Adv Funct Mater 30, 2003237 (2020).38. J. Dong et al.. Bio-Inspired NanoVilli Chips for Enhanced Capture of Tumor- Derived Extracellular Vesicles: Toward Non-Invasive Detection of Gene Alterations in NonSmall Cell Lung Cancer. ACS Appl Mater Interfaces 11, 13973-13983 (2019).39. J. J. Wang et al., Prostate cancer extracellular vesicle digital scoring assay - a rapid noninvasive approach for quantification of disease-relevant mRNAs. Nano Today 48, 101746 (2023).40. N. Sun et al. , Coupling Lipid Labeling and Click Chemistry' Enables Isolation of Extracellular Vesicles for Noninvasive Detection of Oncogenic Gene Alterations. Adv Set (Weinh) 9, e2105853 (2022).41. N. Sun et al. , HCC EV ECG score: An extracellular vesicle-based protein assay for detection of early-stage hepatocellular carcinoma. Hepatology, 774-788 (2022).42. M. L. Blackman, M. Royzen, J. M. Fox, Tetrazine ligation: fast bioconjugation based on inverse-electron-demand Diels-Alder reactivity. J Am Chem Soc 130, 13518-13 19 (2008).43. M. R. Karver. R. Weissleder, S. A. Hilderbrand, Synthesis and evaluation of a series of 1,2,4.5-tetrazines for bioorthogonal conjugation. Bioconjug Chem 22, 2263-2270 (2011).44. R. T. Cummings et al., A peptide-based fluorescence resonance energy transfer assay for Bacillus anthracis lethal factor protease . Proc Natl Acad Set USA 99, 6603-6606 (2002) .45. L. Shi, V. De Paoli, N. Rosenzweig, Z. Rosenzweig, Synthesis and application of quantum dots FRET-based protease sensors. J Am Chem Soc 128, 10378-10379 (2006).46. Y. Xiong. S. Wu, Q. Du, A. Wang, Z. Wang, Integrated analysis of gene expression and genomic aberration data in osteosarcoma (OS). Cancer Gene Ther 22, 524-529 (2015).47. P. Hingorani et al., ABBV-085, Antibody-Drug Conjugate Targeting LRRC15, Is Effective in Osteosarcoma: A Report by tire Pediatric Preclinical Testing Consortium. Mol Cancer Ther 20, 535-540 (2021).48. K. K. Slemmons, S. Mukheijee, P. Meltzer, J. W. Purcell, L. J. Helman, LRRC15 antibody-drug conjugates show promise as osteosarcoma therapeutics in preclinical studies. Pediatr Blood Cancer 68, e28771 (2021).49. G. D. Demetri et al., First-in-Human Phase I Study of ABBV-085, an Antibody-Drug Conjugate Targeting LRRC15, in Sarcomas and Other Advanced Solid Tumors. Clin Cancer Res 27, 3556-3566 (2021).50. E. A. Kolb etal., Initial testing (stage 1) of glembatumumab vedotin (CDX-011) by the pediatric preclinical testing program. Pediatr Blood Cancer 61, 1816-1821 (2014).51. L. M. Kopp et al. , Phase II trial of the glycoprotein non-metastatic B-targeted antibody-drug conjugate, glembatumumab vedotin (CDX-011), in recurrent osteosarcoma AOST1521: A report from the Children's Oncology Group. Eur J Cancer 121, 177-183 (2019).52. M. Roth et al., Targeting Glycoprotein NMB With Antibody-Drug Conjugate, Glembatumumab Vedotin, for the Treatment of Osteosarcoma. Pediatr Blood Cancer 63, 32- 38 (2016).53. N. M. Kendsersky et al., The B7-H3 -Targeting Antibody-Drug Conjugate m276-SL- PBD Is Potently Effective Against Pediatric Cancer Preclinical Solid Tumor Models. Clin Cancer Res 27, 2938-2946 (2021).54. L. J. Talbot et al., Redirecting B7-H3.CAR T Cells to Chemokines Expressed in Osteosarcoma Enhances Homing and Antitumor Activity in Preclinical Models. Clin Cancer Res 30, 4434-4449 (2024).55. Z. Hu et al.. The Cancer Surfaceome Atlas integrates genomic, functional and drug response data to identify actionable targets. Nat Cancer 2, 1406-1422 (2021).56. J. A. Welsh et al.. Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. J Extracell Vesicles 13, el2404 (2024).57. U. Neumann, H. Kubota, K. Frei, V. Ganu, D. Leppert, Characterization of Mca-Lys- Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH2, a Anorogenic substrate with increased specificity constants for collagenases and tumor necrosis factor converting enzyme. Anal Biochem 328, 166-173 (2004).58. E. Decaneto et al., Pressure and Temperature Effects on the Activity and Structure of the Catalytic Domain of Human MT1-MMP. Biophys . / 109. 2371-2381 (2015).59. A. A. Alrehaili et l., Clinical significance of plasma MMP-2 and MMP-9 levels as biomarkers for tumor expression in breast cancer patients in Egypt. Mol Biol Rep 47, 1153- 1160 (2020).60. P. Laudanski et al., Increased serum level of membrane type 1-matrix metalloproteinase (MT1-MMP / MMP-14) in patients with breast cancer. Folia Histochem Cytobiol 48, 101-103 (2010).61. J. P. Oliveira et al. , Impact of conjugation strategies for targeting of antibodies in gold nanoparticles for ultrasensitive detection of 17beta-estradiol. Set Rep 9, 13859 (2019).62. Y. Jung, J. Y. Jeong, B. H. Chung, Recent advances in immobilization methods of antibodies on solid supports. Analyst 133, 697-701 (2008).63. J. H. T. Luong, S. K. Vashist, Chemistry of Biotin-Streptavidin and the Growing Concern of an Emerging Biotin Interference in Clinical Immunoassays. ACS Omega 5, 10-18 (2020).64. J. H. T. Luong, K. B. Male, J. D. Glennon, Biotin interference in immunoassays based on biotin-strept( avidin) chemistry: An emerging threat. Biotechnol Adv 37, 634-641 (2019).65. J. Cui et al.. Expression and clinical implications of leucine-rich repeat containing 15 (LRRC15) in osteosarcoma. J Orthop Res 38, 2362-2372 (2020).66. D. P. Nusinow et al., Quantitative Proteomics of the Cancer Cell Line Encyclopedia. Cell 180, 387-402 e316 (2020).67. E. Eisenberg, E. Y. Levanon, Human housekeeping genes, revisited. Trends Genet 29, 569-574 (2013).68. J. Barretina et al. , The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature 483, 603-607 (2012).69. N. C. Institute.70. N. Novershtem et al. , Densely interconnected transcriptional circuits control cell states in human hematopoiesis. Cell 144, 296-309 (2011).Supplementary Materials and MethodsInstrumentation

[0125] Dynamic light scattering (DLS) and zeta potential were measured on the Zetasizer Nano instrument (Malvern Instruments Ltd., UK). Nanoparticle tracking analysis (NTA) was measured by ZetaView PMX-120 (Particle-Metrix, Germany). Fluorescence readouts were measured by tire microplate reader (Infinite M200 PRO, Tecan, Switzerland). The concentrations of EV protein were measured on the spectrophotometer (Nanodrop® 2000, Thermo Fisher Scientific).Determination of size distribution and concentration of OS EVs

[0126] Hie size distribution and concentration of OS EVs (z.e., HOS EVs) were determined using DLS and NTA. For DLS, harvested HOS EVs were diluted 10-fold into 100 pL of 0.22 pm filtered PBS. For NTA, samples were diluted into 1 mL of 0.22 pm filtered PBS at dilution rates ranging from 100 to 10,000-fold. Each sample was analyzed in three replicates.SEM imaging

[0127] Scanning electron microscopic (SEM) images were obtained using a Supra 40VP SEM (Zeiss, Germany) at an accelerating voltage of 20 kV. OS EV samples and OS EV-immobilized Click MagBeads were cast onto a silicon wafer and air-dried at room temperature overnight. The dried samples were sputter-coated with a gold target (Ted Pella, USA).TEM imaging

[0128] Transmittance electron microscopic (TEM) images were acquired using a Tecnai 12 Quick Cryo-EM (FEI, USA). Prior to sample preparation, a 400-mesh carbon-coated copper grid (Ted Pella, USA) was glow discharged using the PELCO easiGlow™ Glow Discharge Cleaning System (Ted Pella, USA) to facilitate hydrophilization, and the grid was used within 1 h. OS EV samples and OS EV-immobilized Click MagBeads were applied to the glow discharged grid and incubated for 2 min at room temperature. Excess samples were removed by blotting with filter paper, and the grid was washed three times by 5% uranyl acetate. Subsequently, the samples were stained with 5% uranyl acetate for 1 minute. For unstained samples, distilled water was utilized instead of uranyl acetate solution following the same procedure.OS Cell Line Culture

[0129] HOS cell line was purchased from American Type Culture Collection (ATCC) and culture in Eagle’s Minimum Essential Medium (EMEM, ATCC) with 10% fetal bovine serum (FBS) and 100 U / mL penicillin-streptomycin (Thermo Fisher Scientific) in an incubator with 5% CO2.Immunofluorescence staining of OS EV surface markers and OS EV-associated MMPs

[0130] HOS cells on glass coverslips were stained with the following immunocytochemistry' (ICC) protocol. Briefly, cells were fixed with 4% paraformaldehyde (PFA) for 30 min and subsequently permeabilized with 0.1% Triton X-100 for 10 min at room temperature. Cells were then incubated with 2% donkey serum (Jackson ImmunoResearch) for 30 min and incubated overnight at 4 °C in 200 p.L of PBS containing 2% donkey serum with the following primary antibodies: anti-human GPNMB, anti-human LRRC15, anti-human B7-H3, anti-human MMP2. or anti-human MMP14. After washing with PBS, these cells were incubated with DAPI solution (Invitrogen, 1 : 1000 v / v) and the secondary antibodies anti-goat IgG, anti-mouse IgG, and anti-rabbit IgG (Invitrogen, Alexa Fluor, 1:500 v / v). Thereafter, these cells were imaged using a 40x objective lens on a Nikon Eclipse 90i fluorescence microscope.Immunohistochemical staining of OS EV surface markers and OS EV-associated MMPs

[0131] Pathological analysis, including Hematoxylin and Eosin (H&E) staining, evaluation, as well as immunohistochemistry (IHC) staining and scoring on the tissue microarray (TMA) obtained from the 110 OS patients enrolled at Zhongshan Hospital, Fudan University, Shanghai, were performed by experienced pathologists. The original OS tumor samples were fixed in 10% neutral formalin for 24-48 hours, embedded in paraffin, and then formalin-fixed paraffin-embedded (FFPE) blocks were generated following tire routine pathological tissue procedure. The H&E and IHC staining was performed following Clinical Laboratory Improvement Amendments (CLIA)-compliant protocols. 4-pm-thick tissue sections from TMA blocks were serially cut and mounted on poly-L- lysine-coated glass slides. Standard IHC staining on these 4-pm-thick tissue sections was carried out using the Ventana Benchmark ULTRA Slide Stainer, following protocols optimized for each antibody, as determined by conventional IHC. The IHC analyses for LRRC15, GPNMB, and CD276 (B7-H3), as well as MMP14 and MMP2 were individually conducted under optimal conditions for each antibody. Two experienced pathologists independently conducted the evaluation and scoring of these five protein expressions on the OS tissues in the TMA.Synthesis of TCO-graftcd antibodies (TCO-Ab)

[0132] Antibody (BSA-free / azide-free grade, 1 mg / mL, 20 pL) and TCO-PEG4-NHS (1 mM, 2.7 giL. Click Chemistry Tools) were mixed in net volume 100 pL of PBS (pH 8.5) and incubated for 60 min at RT with shaking. Unconjugated TCO-PEG4-NHS was removed by Zeba™ 40 kDa column as per manufacturer's instruction. The obtained amount of the Ab-TCO was determined by NanoDrop® with A280. The degree of labeling (DOL) of TCO was calculated by methyltetrazine-Cy5 (mTz-Cy5) labeling. In brief, TCO-grafted antibody was reacted with excess amount of mTz-Cy5 for 60 min at RT and purified with Zeba™ 40 kDa column. DOL of Cy5 was determined using the protein Cy5 labeling mode of Nanodrop®, yielding an average of ~3 TCO per antibody across the entire batch, regardless of antibody type.Synthesis of methyltetrazine / PEG-grafted Dynabeads™ (EV Click MagBeads)

[0133] Amine-grafted Dynabeads™ M-270 (200 pL) were washed twice with dry DMF. The beads were then resuspended in dry DMF (50 pL) containing triethylamine (90 pmol) and reacted with mTz-PEG4-NHS (0.9 pmol) for 60 min. Afterward, mPEG4-NHS solution (18 pmol, dry DMF) was added, and the mixture was reacted for additional 60 min. Residual NHS was quenched by adding 120 pL of Tris buffer (pH 8.4) and reacting for 10 min. Tire beads were washed with 3* ddFLO and 2zEtOH, then dried at room temperature. The dry beads were stored at 4 °C with protection from light and reconstituted in TE-Tween 0.05% before use.Characterization of EV Click MagBeads

[0134] The zeta potential was measured to confirm at each step of surface modification of EV Click MagBeads. The concentration of the beads was maintained at 0.1 mg / mL in 10% PBS, and each sample was replicated over three runs. The amount of mTz on EV Click MagBeads was determined by back titration of TCO-Cy5 after incubation with the beads. A standard curve of TCO- Cy5 was generated in the range of 0-5 pM in PBS using fluorescence at Xex=620 nm / em=680 nm. 0.5 mg of EV Click MagBeads was dispersed in 200 pL of 5 pM TCO-Cy5 solution and allowed to react for 1 h at room temperature. The Cy5-labeled EV Click MagBeads were magnetically separated, and the Cy5 fluorescence of the supernatant was measured to determine the reacted amount of TCO-Cy5. Tire Cy5 -labeled EV Click MagBeads were visualized using fluorescence microscopy (Nikon 90i, Nikon) at far red channel.Collection of OS EVs from cell culture medium

[0135] HOS cells were cultured in 18 Nunc™ EasYFlask™ Flasks (175 cm2, Thermo Fisher Scientific) until 80% confluency. Before OS EV collection, the culture medium was replaced to serum-free cultured medium (13 mL per flask) to starve the cells for 24 h. The conditioned medium was collected and centrifugated at 300 - . . 4 °C for 10 min followed by second centrifugation step at 2800zg. 4 °C for 10 min to remove cell debris. The medium was gently transferred to Ultra-Clear Tubes (38 mL per tube, Beckman Coulter, Inc., USA) and then ultracentrifuged at 100,000 * g, 4 °C for 90 min. The OS EV pellet was resuspended in 400 pL of PBS and regarded as OS EV stock solution. The number of OS EVs in the stock solution was determined by NTA as described in Instrumentation section.Characterization of OS EVs

[0136] Tire size distribution of OS EVs was determined using DLS and NTA as aforementioned. Each sample was replicated in three runs. For electron microscopy imaging of OS EVs, harvested OS EV samples were fixed in 4% PFA for 30 min. Tire SEM and TEM samples were prepared as described in Instrumentation section and observed using the Supra 40VP SEM and Tecnai 12 Quick Cryo-EM, respectively.Preparation of synthetic plasma samples

[0137] Synthetic plasma samples mimicking OS patient plasma was prepared to optimize and to validate the capacity of EV Click MagBeads for enriching OS EVs. HOS cell-derived OS EVs were spiked into EV-depleted healthy donors’ plasma with a desired EV number and aliquoted into 250 pL each. For the EV titration study, a synthetic plasma sample with 2.5 x 108EV / mL was prepared and subjected to three times of 2-fold serial dilutions (a total of 4 samples).OS EV enrichment from plasma samples

[0138] OS EVs in both synthetic and clinical plasma samples were enriched using EV Click MagBeads based on the optimized condition from our previous study. In brief. 100 ng of TCO- grafted antibodies (TCO-anti-LRRC15, TCO-anti-GPNMB, TCO-anti-B7-H3) were mixed with 250 pL of plasma sample for 45 min at room temperature. The plasma samples containing TCO-labeled EVs were incubated with 25 pg of EV Click MagBeads for 45 min followed by a magnetic separation for 30 s. OS EV-enriched Click MagBeads were washed three times by washing buffer (50 mM Tris,10 mM CaCU, 1 pM Zn(NO3)2, 0.05% Tween-20, pH 7.5) and further analyzed by FRET-based MMP activity assay.Characterization of OS EV-immobilized EV Click MagBeads

[0139] OS EV-immobilized EV Click MagBeads were analyzed by SEM and immunogold TEM. HOS cell-derived OS EVs were enriched on EV Click MagBead with anti-TCO-B7-H3, using the procedure as described in OS EV Enrichment from Plasma Samples section. The OS EV- immobilized EV Click MagBead samples were washed with PBS and fixed in 4% paraformaldehyde (PFA) solution for 30 min at room temperature. For SEM samples, tire samples were washed with ddH2O and prepared as described in Instrumentation section. For immunogold staining, the fixed EV- bound beads were incubated with anti-CD63 (Abeam, mouse, 1:50 dilution) in 1% BSA for 1 h at room temperature and washed twice with PBS. Subsequently, the samples were incubated with antimouse IgG nanogold (Jackson ImmunoResearch. 12 nm, 1:20 dilution) in 1% BSA for 30 min. After washing twice with PBS, the gold-labeled samples were dropped onto glow discharged TEM grids, and the sample was prepared without staining as described in Instrumentation section.Optimization of the MMP activity assays

[0140] The progress curve and dynamic range of MMPs were monitored using fluorescence microplate reader (MMP 14 FRET probe at Xex=328 nm / em=400 nm, MMP2 FRET probe at Xex=280 nm / Z,m=360 nm). Recombinant human MMPs (rhMMPs, Abeam) and FRET peptide probes (Sigma- Aldrich) were prepared in activation buffer (50 mM Tris, 10 mM CaCty 1 pM Zn(NO3)2, pH 7.5). For the MMP 14 progress curve, final concentration of 0-7.5 pM MMP 14 probe and 20 nM rhMMP14 were prepared; For MMP2 kinetic curves, final concentration of 0-60 pM MMP2 probe with 1.3 nM rhMMP2 were prepared. Tire kinetic curve was started by adding the substrate solution to MMPs and allowed to react at 37°C. A Michaelis-Menten plot at 30 min w as calculated to detennine the kinetic parameters (Vmaxand Km). For dynamic range studies, final concentration of MMP 14 and MMP2 probes was fixed at 4.5 pM, and the probe solution was added in serial dilutions of MMPs solution. For specificity study, 20 nM of activated MMP14 or MMP2 w as incubated with 4.5 pM of MMP14 or MMP2 probes for 1 h at 37°C. Fluorescence readouts were normalized against that of target MMP.Validation of OS EV MMP Activity Assay

[0141] OS EVs within the synthetic plasma were enriched using EV Click MagBeads with the optimized condition. After enrichment, the MMPs were activated with 4-aminophenylmercuric acetate (APMA. Sigma- Aldrich). For MMP 14 activation, samples were incubated with 1 mM APMA for 2 h at 37 °C. For MMP2 activation, samples were lysed by 0.1% Triton X-100, followed by activation with 1 mM APMA for 1 h at 37 °C. The corresponding MMP FRET probe was added in the activated samples, incubated for 1 h at 37 °C, and the fluorescence readouts were measured using microplate reader.MMP activity assay of whole plasma

[0142] 2 pL of whole plasma was used for MMP activity assay. For MMP14 activation, samples were incubated with 1 mM APMA for 2 h at 37 °C. For MMP2 activation, samples were lysed by 0.1% Triton X-100, followed by activation with 1 mM APMA for 1 h at 37 °C. The corresponding MMP FRET probe was added and incubated for 1 h at 37 °C. The fluorescence readout of the probe was measured as described.FURTHER EXAMPLE 2 - Pre-clinical detection of Alzheimer’s disease by noninvasive assessment of p-secretase activity in neuronal extracellular vesicles

[0143] Alzheimer's disease (AD), the most prevalent type of dementia, is characterized by a biological process that begins with the development of AD neuropathologic change (ADNPC) while individuals remain asymptomatic. A key molecular hallmark of ADNPC is the accumulation of amyloid-[3 plaques. P-secretase plays a critical role in the upstream of pathological cleavage of amyloid precursor protein (APP), producing amyloid-P peptides that are prone to misfolding, ultimately contributing to plaque formation. Neuronal extracellular vesicles (NEVs) in the blood transport p-secretase and preserve its activity, allowing for noninvasive profiling of p-secretase activity for detecting early onset of ADNPC. In this study, we introduce a novel approach for noninvasive assessment of P-secretase activity in AD patients using an NEV P-secretase Activity Assay. This assay identifies NEVs exhibiting colocalization of NEV markers with AD-associated P- secretase, generating a P-secretase activity’ profile for each patient. The NEV P-secretase Activity Assay represents a significant advancement in leveraging the diagnostic potential of NEVs, offering a noninvasive, quantitative method for reliably assessing p-secretase activity to detect tire early onset of ADNPC

[0144] Alzheimer's disease (AD) is the most prevalent type of dementia in older adults, accounting for 60-70% of dementia cases worldwide.1As the global population ages, the prevalence of AD is projected to rise significantly, from approximately 32 million people in 20232to an estimated 106.8 million by 2050.3This surge will considerably impact society and healthcare systems, positioning AD as one of the most pressing public health challenges of the 21st century.4AD is characterized by a biological process that begins with AD neuropathologic changes (ADNPCs) while individuals are still asymptomatic.5If left undetected, the increasing neuropathologic burden eventually leads to the onset and progression of clinical symptoms. The time interval between ADNPCs and the first sign of memory impairment can exceed 20 years.6Currently, the only FDA- approved methods for the definitive AD diagnosis are positron emission tomography (PET) scans7and molecular tests of cerebrospinal fluid (CSF).8Given the importance of detecting ADNPCs before symptoms emerge, there is an increasing need for developing noninvasive surveillance methods that can enable early intervention and more effective symptom management.9

[0145] Tire accumulation of amyloid-0 plaques10 12is a molecular hallmark of AD, occurring earlier than other ADNPCs, such as Tau-mediated neuronal injury,13and playing a critical role in disrupting neuronal function and contributing to neurodegeneration in AD. In response, immunotherapies using monoclonal antibodies14 16have been recently developed to target and clear brain amyloid in early-stage AD patients. While these treatments have effectively reduced amyloid burden, the corresponding improvements in cognition and function have been modest, and the disease has continued to progress even after amyloid clearance.17Therefore, despite these therapeutic advances, earlier detection of ADNPC before significant amyloid- accumulation has become crucial for timely intervention and early disease treatment.

[0146] Under normal physiological conditions, a-secretase predominantly cleaves the amyloid-0 domain within the amyloid precursor protein (APP), producing short, soluble peptide fragments. Conversely, in the pathological pathway, 0-secretase (e.g., beta-site APP cleaving enzyme 1, BACE1) competes with a-secretase for APP cleavage, producing amyloid-0 peptides (e.g., amyloid-0 42).18These peptides are prone to misfolding, ultimately leading to tire formation of amyloid-0 plaques (FIG. 35A).10Current diagnostic methods, such as CSF tests19and PET scans,20can only detect amyloid burden after amyloid-0 peptides have been generated and plaques have accumulated. Given that amyloid-0 accumulation is triggered by 0-secretase, detecting elevated 0- secretase activity could offer a promising alternative approach for identifying the onset of ADNPC at even earlier stages (FIG. 41).

[0147] Previous studies have demonstrated significantly elevated P-secretase activity in AD patients' brain tissues.21-23However, assessing p-secretase activity’ in brain tissue is limited to postmortem analysis due to the highly invasive nature of tissue collection. CSF has also been explored as an alternative for assessing p-secretase activity.8,24,25However, the collection of CSF via lumbar puncture remains invasive and carries risks, making it impractical for routine surveillance that requires repeated sampling.20To overcome this challenge, researchers have investigated the assessment of P-secretase activity in plasma and serum samples from AD patients.27-29While elevated enzyme activity has been detected, the ability to distinguish AD patients from healthy controls (HCs) remains limited, likely because p-secretase in blood can originate from multiple organs. Developing a method to selectively assess P-secretase activity from neuronal sources could significantly improve its diagnostic performance, particularly for detecting early onset of ADNPC.

[0148] As a key component of liquid biopsies, extracellular vesicles (EVs) are promising candidates as noninvasive biomarkers for disease diagnosis and surveillance.30,31These heterogeneous, nano-scaled particles, enclosed by phospholipid bilayers, are released by various cell types, including neurons.32EVs protect fragile biomolecular cargos, such as proteins, RNA, and DNA, thereby reflecting the biomolecular profile of the parental cells or tissues throughout disease progression.33Neuronal EVs (NEVs) are particularly suited for noninvasive harvesting of p-secretase of neuronal origin due to three key advantages: i) NEVs mediate the transport of biomolecular cargos, including P-secretase. from neurons across the blood-brain barrier into the bloodstream.34,35ii) NEV membranes protect fragile biomolecules, thus preserving the integrity and activity of P-secretase :3band iii) NEVs mirror the surface markers of their parental neurons, allowing the use of preestablished neuron surface markers for NEV isolation.3738These features permit the assessment of P-secretase activity within NEVs from blood samples, providing an opportunity for detecting the earlier onset of ADNPC in a noninvasive manner.

[0149] To harness NEVs’ potential for assessing p-secretase activity in the brain, it is crucial to develop a system that enriches NEVs while preserving enzyme activity. Conventional methods relying on immunoaffinity-based approaches targeting NEV markers often require antibody precoating through techniques such as amide coupling39or biotin-streptavidin interaction.40These methods can present challenges in capture efficiency and potential biological interference. To address these challenges, our team pioneered two innovative click chemistry-mediated EV enrichment technologies, namely Click Chips41 42and Click Beads.4344These technologies offer two key advantages: i) click motif-grafted antibodies used to label EVs significantly reduce antibodyconsumption (by 10 - 100 times) and labeling time, and ii) the use of an inverse electron demand Diels-Alder reaction between / ram-cycloocteiic (TCO) and tetrazine (Tz) motifs ensures a bioorthogonal click reaction,45circumventing the biological interferences commonly encountered with biotin-streptavidin interaction. Additionally, our past studies have demonstrated that click chemistry -mediated EV enrichment can be seamlessly coupled with various downstream molecular analyses, such as mRNA profiling46,47and protein detection44

[0150] Here, we introduce an NEV p-secretase Activity Assay for the noninvasive detection of early onset of ADNPC. As illustrated in FIG. 35B. this assay is comprised of two main steps: i) click chemistry-mediated enrichment of two subpopulations ofNEVs using Click MagBeads with one of the two TCO-grafted antibodies targeting NEV markers, z.e., LI cell adhesion molecule (LI CAM) and neural cell adhesion molecule (NCAM), and ii) assessment of P-secretase activity of the enriched NEVs using p-secretase FRET probes, providing a p-secretase activity profile for each AD patient. The core concept of this technology is that NEVs exhibiting colocalization of one of the two NEV markers (z.e., L1CAM or NCAM) with p-secretase can exclusively generate positive readouts. To validate this idea, NEVs were characterized through scanning electron microscopy (SEM) and transmission electron microscopy (TEM) to confinn the colocalization of one of the two NEV markers with p-secretase.

[0151] As a proof-of-concept study, we evaluated the diagnostic performance of the NEV P- secretase Activity Assay by analyzing CSF samples from a study cohort of 15 AD patients and 15 HCs, validating the assay’s ability to distinguish AD patients from HCs. To pave the way for noninvasive detection of ADNPC using plasma samples, we first demonstrated the reliability of our tecbiology in enriching NEVs and quantifying p-secretase activity in synthetic plasma samples. Subsequently, a retrospective case-control study was conducted using clinical plasma samples from a validation cohort of 55 AD patients and 55 HCs to validate the assay’s performance in distinguishing AD patients from HCs. A logistic regression model, termed the NEV P-secretase Activity Score, was formulated by combining p-secretase activity readouts from the two subpopulations of NEVs. This model exhibited excellent diagnostic accuracy in differentiating AD patients from HCs, achieving an area under receiver operating characteristic curve (AUROC) of 0.986. Importantly, the established scoring system showed a strong correlation with the AD patients’ cognitive performance indicated by the Mini-Mental State Examination (MMSE) score.ResultsIdentification of NEV markers

[0152] Since the NEV P-secretase Activity Assay is designed to specifically detect NEVs which exhibit colocalization of NEV markers with -secretase (FIG. 35B), we first conducted a comprehensive literature survey to identify appropriate NEV markers for NEV enrichment. For NEV marker selection, we analyzed a recent systematic review48that focused on the enrichment of NEVs from the blood of neurodegenerative disease patients. This review evaluated publications over the past decade, with a primary emphasis on NEV enrichment from blood via immunoprecipitation and identified L1CAM as the most commonly used marker for isolating NEVs. Despite ongoing debates regarding reliability of LI CAM as a marker for NEV isolation,49’0its widespread application as a primary marker in the NEV research field justified incorporating this maker to our assay. Additionally, we sought to identify complementary' markers to enhance assay performance. As outlined in Table 4 (FIG. 45), our literature analyses highlighted NCAM as another widely used marker that could complement LI CAM for enriching NEVs. Based on these findings, we incorporated both LI CAM and NCAM to improve the robustness and perfonuance of the NEV P- secretase Activity Assay. In parallel, previous studies have demonstrated that NEVs can transport key enzymes involved in the APP cleavage pathway, including P-secretase.’1’4Leveraging these insights, we designed the NEV P-secretase Activity Assay to assess P-secretase activity from two subpopulations of NEVs: LlCAM(+)-NEVs and NCAM(+)-NEVs.Characterization of NEVs in the CSF of AD patient

[0153] As a model target to demonstrate our assay, NEVs were isolated from the CSF of AD patient via polyethylene glycol (PEG)-based precipitation using ExoQuick™ (detailed in Methods below).5’ The characterization of the isolated NEVs followed the guidelines established by the International Society for Extracellular Vesicles (MISEV2023).59Nanoparticle tracking analysis (NTA) determined the size and concentration of NEVs as 171.5 ± 59 nm and 3.19 x 109NEVs / mL, respectively (FIG. 42A). Hie NEVs' morphology and structural integrity were further validated by transmission electron microscopy (TEM, FIG. 42B) and scanning electron microscopy (SEM, FIG. 42C).Preparation and characterization of Click MagBeads

[0154] Building on our previous development of Click Beads,43,44a click chemistry- mediated EV enrichment technology, we have developed the Click MagBeads platform, introducingtwo major enhancements: i) the transition from silica beads to magnetic beads, significantly improving the speed and efficiency of EV enrichment, and ii) the replacement of tetrazine (Tz) with methyltetrazine (mTz), enhancing stability of surface click motifs and extending their storage life from two weeks to over three months. Tire process for synthesis of Click MagBeads is outlined in FIG. 36A. We modified amine-grafted Dynabeads™ (2.7 pm diameter, ThermoFisher) by conjugating with methyltetrazine-PEG4-A-hydroxysuccinimide (mTz-PEG4-NHS) ester via NHS ester chemistry to produce mTz-Dynabeads™. Any residual amine groups were subsequently capped with methoxypolyethylene glycol succinate-V-hydroxysuccinimide (mPEG4-NHS) ester to confer anti-fouling properties, yielding the final Click MagBeads.

[0155] We monitored the changes in the surface chemistry of the magnetic beads by measuring zeta potential shifts (FIG. 36B). The amine-grafted Dynabeads™ initially exhibited a strong positive charge, which gradually decreased following the sequential addition of mTz-PEG4- NHS and mPEG4-NHS. confirming successful attachment of both functional groups. Tire presence of mTz motifs on the Click MagBeads was further validated by a click reaction with TCO-labeled Cy5 (TCO-Cy5), resulting in noticeable Cy5 fluorescence under fluorescence microscopy (FIG. 36C). A corresponding histogram of mean fluorescence intensity (MFI) also confirmed the successful interaction between mTz and TCO motifs (FIG. 36D).

[0156] To measure the density of mTz motifs on the Click MagBeads and assess their stability over time, we employed the TCO-Cy5 labeling method (FIG. 36E, see details in Methods). Tire initial mTz motif concentration was 0.464 ± 0.024 nmol / mg, with approximately 60% retention after six months. Based on this result, all experiments were conducted within two months of bead synthesis, ensuring that over 80% of the mTz motifs remained active. Notably, the quantity of retained mTz motifs on the Click MagBeads (over 80% retention, 9 pmol per batch) was sufficient to capture all TCO-grafted antibodies used in this study (100 ng antibody, equivalent to 2 pmol TCO motifs per batch).Verification of click chemistry-mediated NEV enrichment

[0157] To verify the click chemistry-mediated enrichment of two subpopulations of NEVs, we implemented the workflow illustrated in STEP1 of FIG. 37A. Specifically, 0.4 mL of CSF, collected from AD patient, was incubated with 100 ng of one of the two TCO-grafted antibodies targeting NEV markers, z.e., TCO-anti-LlCAM or TCO-anti-NCAM. The labeled NEVs were subsequently enriched onto the Click MagBeads through the click chemistry interaction between mTzand TCO motifs, as shown in the insets of FIG. 37B. SEM images in FIG. 37B provided direct visualization of the tw o subpopulations of NEVs enriched onto the Click MagBeads, with their size and morphology aligning with those of intact NEVs (FIG. 42A-42B).

[0158] To further confirm the identity of the enriched particles, immunogold staining was performed, as outlined in STEP2 of FIG. 37A. NEVs enriched onto the Click MagBeads were labeled with an antibody against CD63, a well-established EV marker, followed by tagging with a gold nanoparticle (AuNP)-grafted secondary antibody, as depicted in the insets of FIG. 37C. TEM images in FIG. 37C validated the presence of AuNPs on both NEV subpopulations enriched onto the Click MagBeads, confirming the successfill application of our Click MagBeads platform for click chemistry -mediated NEV enrichment.Confirmation of NEV marker / p-secretase colocalization

[0159] With the designed working mechanism of the NEV P-secretase Activity Assay (FIG. 35B), positive readouts are generated only when NEVs exhibit colocalization of one of the two NEV markers (z.e., LI CAM or NCAM) with p-secretase. To confirm this colocalization, we performed the immunogold staining by labeling the enriched NEVs with anti- -secretase antibody, followed by the application of AuNP -grafted secondary antibody, as outlined in STEP2 of FIG. 37A. This approach allowed us to label the AuNPs onto the NEV-associated P-secretase, as illustrated in the insets of FIG. 37D. Tire TEM images in FIG. 37D confirmed the presence of P-secretase on both NEV subpopulations, demonstrating the colocalization of NEV markers w ith p-secretase on NEVs enriched onto the Click MagBeads. This finding highlights the potential for assessing p-secretase activity from the enriched NEVs. Furthermore, these results establish a solid foundation for the targeted assessment of P-secretase activity from specific NEV subpopulations in complex biological fluids, such as blood.Investigation of P-secretase enzymatic kinetics

[0160] After confirming the colocalization of NEV markers with p-secretase on the enriched NEVs, we proceeded to optimize the assessment of p-secretase activity using recombinant human p- secretase and P-secretase FRET probe (FIG. 43A, see details in Methods). An enzyme kinetics study was performed to establish an optimal substrate concentration to operate the NEV P-secretase Activity Assay. Specifically, we measured the hydrolysis rate of P-secretase FRET probes at various concentrations with a fixed amount of P-secretase (FIG. 43B). The resulting data were fitted to theMichaelis-Menten equation to determine the kinetic parameters of P-secretase, yielding a Michaelis constant (KM) of 3.8 pM (FIG. 43C), consistent with previously reported values?7’9Based on this result, we selected the FRET probe concentration of 12 pM, approximately three times the KM, ensuring at least 75% of the maximum reaction rate (Vmax = 2. 12 x 103pM / s) during the assay. We then assessed the dynamic range of -secretase activity, which demonstrated good linearity within the range of 0 - 160 nM P-secretase (FIG. 43D). Furthermore, we evaluated the specificity of the P- secretase FRET probe against non-target proteases, revealing that the activities from non-target proteases were less than 6% compared to P-secretase (FIG. 43E). These results confirmed the high specificity of the p-secretase FRET probe utilized in this study.Proof-of-concept study with clinical CSF sample

[0161] Before testing plasma samples, we performed a proof-of-concept demonstration of the NEV P-secretase Activity Assay using clinical CSF samples, following the workflow shown in FIG. 38A. NEVs were enriched from 0.8 mL of clinical CSF samples (0.4 mL for each NEV subpopulation), collected from a study cohort of 15 AD patients and 15 HCs, onto tire Click MagBcads (see procedures described in Verification of click chemistry-mediated NEV enrichment section and details in Methods). The clinical characteristics of these participants are summarized in Table 5 (FIG. 46). Subsequently, the enriched NEVs underwent lysis to release p-secretase, followed by incubation with P-secretase FRET probes to measure enzyme activity. By assessing P- secretase activity in two subpopulations of NEVs, we generated a P-secretase activity profile for each patient. Tire quantitative P-secretase activity data were organized into a heatmap (FIG. 38B), showing significantly higher p-sccrctasc activity in AD patients compared to HCs, with p-valucs less than 0.0001 for both LlCAM(+)-NEVs and NCAM(+)-NEVs (FIG. 38C). The diagnostic perfonnance was further evaluated using AUROC values, yielding 0.956 and 0.969 for LlCAM(+)-NEVs and NCAM(+)-NEVs, respectively. This study successfully demonstrated the capability of the NEV P- secretase Activity Assay to reliably differentiate AD patients from HCs by assessing P-secretase activity of NEVs.NEV P-secretase Activity Assay with synthetic plasma sample

[0162] Following the successful proof-of-concept study of the NEV p-secretase Activity- Assay using clinical CSF samples, we further validated its reliability for detecting NEVs in plasma, a more complex biological matrix than CSF, by conducting experiments with synthetic plasma samples. As shown in FIG. 39A, we prepared synthetic plasma samples by spiking the CSF of AD patient intoEV-depleted HC plasma. The concentration of NEVs in the CSF of AD patient was determined by NTA (see details in Methods). The prepared synthetic plasma samples were then divided into duplicates, each containing 0.4 mL, to assess -secretase activity of each NEV subpopulation by following the same procedure employed in the CSF sample assay. The results of time -dependent measurement presented in FIG. 39B proved gradual fluorescence signal enhancement in accordance with the increase of the concentration of spiked NEVs, from 0 to 109NEVs / mL, for both NEV subpopulations. Additionally, we established the calibration curves of P-secretase activity of two subpopulations of NEVs across varying concentrations of spiked NEVs (FIG. 39C). The results demonstrated a strong linear correlation (R2> 0.99) of p-secretase activity in both NEV subpopulations across a wide concentration range of spiked NEVs, with a dynamic range of 5 x 107- 109NEVs / mL. This outcome validated the assay's ability to quantitatively detect NEVs even in plasma, ensuring its applicability for quantify ing NEV-associated P-secretase in plasma and holding promise for broader clinical applications in noninvasive ADNPC diagnostics.Retrospective case-control study with clinical plasma sample

[0163] Building on the successful validation of the NEV p-sccrctasc Activity Assay with synthetic plasma samples, we proceeded with a retrospective case-control study with clinical plasma samples, following the workflow illustrated in FIG. 40A. We assessed p-secretase activity of tire two subpopulations of NEVs in the clinical plasma samples collected from the validation cohort of 55 AD patients and 55 HCs, as detailed in Table 6 (FIG. 47). The P-secretase activity data, presented in a heatmap (FIG. 40B), confirmed that the NEV P-secretase Activity Assay can effectively distinguish AD patients from HCs by analyzing clinical plasma samples, with p-valucs less than 0.0001 for both NEV subpopulations (FIG. 40C). Similar to the CSF sample study, the assay’s diagnostic performance was further assessed using AUROC values, which were 0.968 for LlCAM(+)-NEVs and 0.954 for NCAM(+)-NEVs (FIG. 40D). Notably, the diagnostic accuracy in differentiating AD patients from age-matching HCs was comparable to that for non-age-matching HCs (FIG. 44A-44B). Additionally, we established a logistic regression model that synergistically combines the readouts from two subpopulations of NEVs into a single metric, termed the NEV p-secretase Activity Score (FIG. 40E). The NEV p-secretase Activity Score demonstrated significant differentiation (P < 0.0001) of AD patients from HCs (FIG. 40F and 40G) and exhibited enhanced discrimination power, achieving an AUROC of 0.986, with high sensitivity (96%) and specificity (93%) at the optimal cutoff value of -0.99 (FIG. 40H). We also confirmed that the NEV P-secretase Activity Scores from the majority of AD patients were negatively correlated with their Mini-Mental State Examination(MMSE) scores (FIG. 401). These results underscore the capability of this assay to detect AD using clinical plasma samples, enabling a noninvasive assessment of the ADNPC.Discussion

[0164] Hie ideal approach for managing AD is to detect the early onset of ADNPC, enabling timely intervention, early treatment, and slowing of disease progression. Current FDA-approved diagnostic modalities, such as PET scans20and CSF tests.19primarily detect the accumulation of amyloid-P plaques, which are the products of the pathological APP cleavage pathway and key early indicators of ADNPC. PET scans, for instance, utilize PET probes developed from amyloid-binding dyes that insert into the P-sheet structure of amyloid-P plaques.60As a more cost-effective and accessible alternative, CSF samples are analyzed using immunoassays to measure the ratio of amyloid-P 42 to amyloid-P 40.19,61While these existing diagnostic modalities are valuable, they typically identify ADNPC after significant amyloid-P accumulation has occurred. Since amyloid-P accumulation is triggered by the P-secretase,18detecting elevated P-secretase activity could offer an alternative approach for detecting even earlier onset of ADNPC (FIG. 41). To fulfill this purpose, we developed tire NEV p-sccrctasc Activity Assay, a noninvasive blood test that assesses the activity of P-secretase, offering the potential for earlier detection of ADNPC compared to existing methods and therefore paving the way for more timely intervention in AD progression.

[0165] To develop the NEV P-secretase Activity Assay, we first conducted a literature review to identify the most reliable NEV markers for the specific enrichment of NEVs. Among the markers considered. LI CAM48,50and N CAM62, 63emerged as the most well-established NEV markers over the last decade. In parallel, we identified NEVs can transport key enzymes involved in the APP cleavage pathway, including P-secretase.’1’4Upon selecting these NEV marker / protease combinations, we designed the proof-of-concept study of NEV P-secretase Activity Assay by utilizing CSF samples for the following reasons: i) CSF is in direct contact with the brain, as it circulates through the brain's ventricles, resulting in a high concentration of NEVs, ii) CSF is isolated from other body fluids, minimizing the presence of EVs originating from non-neuronal sources, and iii) CSF contains significantly lower levels of soluble proteins compared to other biofluids. These characteristics make CSF an ideal model biofluid for this study, offering a relatively pure environment with a predominance of NEVs.

[0166] To meet the standardized guidelines for EV characterization outlined in MISEV2023,56we performed comprehensive analyses of CSF-derived NEVs using varioustechniques, including NTA, TEM, and SEM. These analyses confirmed the NEVs’ size (171.5 ± 59 nm), concentration (3.19 x 109NEVs / mL), and morphology. Additionally, TEM imaging with immunogold staining verified the colocalization of NEV markers with p-secretase on the enriched NEVs. Following successful characterization, we evaluated the assay’s ability to differentiate AD patients from HCs by testing CSF samples from a study cohort of 15 AD patients and 15 HCs, achieving AUROCs of 0.956 and 0.969 for LlCAM(+)-NEVs and NCAM(+)-NEVs, respectively. It is worth noting that outliers from AD patient group observed in the study were likely due to the use of postmortem CSF samples with inconsistent postmortem intervals, which could have affected the integrity of NEVs in the samples. Despite this, the proof-of-concept study using clinical CSF samples successfully demonstrated the assay’s ability to differentiate AD patients from HCs by assessing p- secretase activity of NEVs.

[0167] Our successful proof-of-concept study enabled us to transition from CSF to synthetic plasma samples for further validation of the assay’s ability to quantify p-secretase activity of NEVs in plasma, a more accessible biofluid compared to CSF. This quantitative capability in plasma condition underpins the assay’s potential for detecting early onset of ADNPC in a noninvasive manner. By spiking varying concentrations of NEVs from the CSF of AD patient into EV-depleted HC plasma, we established a strong linear correlation for quantifying p-secretase activity across a broad NEV concentration range of 5 x 107- 109NEVs / mL. This validated the assay’s capability for quantitative assessment of NEV-associated p-secretase even in plasma sample.

[0168] Following this validation, we proceeded with a retrospective case-control study by applying the assay to the clinical plasma samples from a validation cohort of 55 AD patients and 55 HCs. Tire assay demonstrated excellent diagnostic performance, with AUROCs of 0.968 and 0.954 for LlCAM(+)-NEVs and NCAM(+)-NEVs. respectively, in distinguishing AD patients from HCs. When we independently evaluated the diagnostic accuracy for age-matching and non-age -matching HCs, a slight reduction in discrimination power was observed in the age-matching group. This was attributed to slightly elevated P-secretase activity in age-matching HCs compared to non-agematching group, suggesting that P-secretase activity may increase in older individuals at higher risk, even before the onset of ADNPC. To further enhance diagnostic accuracy, we employed a logistic regression model to combine readouts from both NEV subpopulations, generating the NEV P- secretase Activity Score. This score achieved exceptional accuracy, with an AUROC of 0.986, demonstrating the assay’s high sensitivity (96%) and specificity (93%). Importantly, the NEV P-secretase Activity Score showed a strong correlation with AD patients’ MMSE scores, a standardized measure of cognitive performance.

[0169] Given its diagnostic performance in the retrospective case-control study, the NEV [3- secretase Activity Assay could have strong potential as a complementary tool to PET scans and CSF tests for earlier detection of ADNPC followed by guidance of timely treatment interventions for AD. This potential stems from several distinct advantages: i) its high specificity, achieved through a double-layered workflow that requires colocalization of NEV markers (z.e., LI CAM and NCAM) with AD-associated protease (z.e., [3-secretase) on NEVs to produce positive readout: ii) its high sensitivity, which is ensured by click chemistry-mediated enrichment of NEVs, followed by signal amplification from enzymatic cleavage of P-secretase FRET probes; iii) its targeting of upstream [3- secretase activity rather than downstream amyloid-|3 products, facilitating earlier detection of ADNPC; iv) its potential for noninvasive diagnosis of AD by assessing p-secretase activity in NEVs derived from blood samples; and v) its readout compatibility with widely available fluorescence plate readers, ensuring easy use in various laboratory settings.

[0170] Despite the promising perfonnance of the NEV p-secretase Activity Assay, several limitations must be acknowledged. First, the proof-of-concept and retrospective case-control studies using clinical CSF and plasma samples from relatively small cohorts limit the statistical power to folly demonstrate the assay's diagnostic power in a clinical setting. Larger-scale studies are necessary to address this limitation. Second, further analytical validation is required by correlating the assay’s readouts with PET scans and CSF tests, which are two contemporary' gold standards for detecting AD, to comprehensively confirm the assay’s diagnostic performance. Third, longitudinal studies are required to validate the assay’s capability for monitoring disease progression and treatment effects. Fourth, it is crucial to assess the assay’s accuracy in distinguishing AD from other non-AD dementias, such as frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS), to ensure the development of a highly specific screening method for AD. Lastly, since AD is a multifactorial neurodegenerative disorder characterized by the interplay of various pathological mechanisms, incorporating other enzymes involved in AD pathogenesis should be considered to provide a more comprehensive insight for early ADNPC detection. For instance, increasing evidence suggests that elevated levels of matrix metalloproteinases (MMPs), including MMP2, MMP9. and MMP14. in the brains of AD patients contribute to pathophysiological processes such as neuroinflammation.64’65Expanding the capacity of this assay to include these factors could lead to a more thorough evaluation of AD and offer better insights for personalized care and treatment strategies.MethodsInstrumentation

[0171] Nanoparticle tracking analysis (NTA) was conducted by ViewSizer 3000 (HORIBA Scientific, Japan). Zeta potential was measured on the Zetasizer Nano instrument (Malvern Instruments Ltd., UK). Fluorescence readouts were measured by the microplate reader (CLARIOStar, BMG Labtech, Germany).Clinical CSF and plasma samples

[0172] 15 CSF samples from HCs were purchased from Medix Biochemica (St. Louis, MO,US) and stored at -80 °C without further processing. Additionally, 15 postmortem CSF samples were collected by following protocols of the Department of Pathology and Laboratory Medicine at UCLA. Specifically, the postmortem CSF was obtained during autopsy by inserting a needle attached to a 20 mL syringe into the lateral ventricles: 1 mL aliquots were then stored in 1.5 mL cryotubes at -80 °C until use. The clinical characteristics of these participants are provided in Table 5. Furthermore, 110 plasma samples (55 from HCs and 55 from AD patients) were purchased from ProteoGenex (Inglewood, CA, US) and stored at -80 °C without further processing. The clinical characteristics of these participants are summarized in Table 6.Collection of NEVs from the CSF of AD patient

[0173] NEVs were isolated from the CSF of AD patient using ExoQuick™ (System Biosciences, US) following the manufacturer’s instruction. Specifically. CSF was centrifuged at 3,000 g for 15 min at 4 °C to remove cellular debris. Then, 0.25 mL of CSF supernatant was mixed with 63 pL of ExoQuick™ solution, followed by incubation at 4 °C for 30 min. The mixture was centrifuged at 1,500 g for 30 min at 4 °C, and the supernatant was carefully aspirated not to disturb the precipitated NEVs in pellet. Tire pellet was resuspended in 0.25 mL of 0.22 pm filtered PBS, which was immediately used to characterize NEVs.Characterization of NEVs’ size distribution and concentration

[0174] Tire size distribution and concentration of NEVs were determined by NTA. NEV samples were diluted into 1 mL of 0.22 pm filtered PBS at dilution rates ranging from 100 to 10,000- fold. Each sample was analyzed in three replicates.SEM imaging

[0175] Scanning electron microscopy (SEM) images were obtained using a Supra 40VP SEM (Zeiss, Germany) at an accelerating voltage of 20 kV. NEV samples and NEV-immobilized Click MagBeads were cast onto a silicon wafer and air-dried at room temperature overnight. The dried samples were sputter-coated with a gold target (Ted Pella, US).TEM imaging

[0176] Transmission electron microscopy (TEM) images were acquired using a Tecnai 12 Quick Cryo-EM (FEE US). Prior to sample preparation, a 400-mesh carbon-coated copper grid (Ted Pella, US) was glow discharged using tire PELCO easiGlow™ Glow Discharge Cleaning System (Ted Pella, US) to facilitate hydrophilization, and the grid was used within 60 min. NEV samples and NEV-immobilized Click MagBeads were applied to the glow discharged grid and incubated for 2 min at room temperature. Excess samples were removed by blotting with filter paper, and the grid was washed three times by 5% uranyl acetate. Subsequently, the samples were stained with 5% uranyl acetate for 1 min.Synthesis of Click MagBeads

[0177] Amine-grafted Dynabeads™ M-270 (200 pL, Thermo Fisher Scientific, US) were washed twice with dry dimethyl sulfoxide (DMSO). The beads were then resuspended in dry DMSO (50 pL) containing triethylamine (90 pmol) and reacted with mTz-PEG4-NHS (0.9 pmol) for 60 min. Then, mPEG4-NHS (18 pmol in dry DMSO) was added, and the mixture was incubated for additional 60 min. Residual NHS was quenched by adding 120 pL of Tris buffer (pH 8.4) and reacting for 10 min. The beads were washed three times with distilled water (DW) and twice with ethanol (EtOH), then dried at room temperature. The dry beads were stored at 4 °C with protection from light and reconstituted in PBST (0.05%) before use.Characterization of Click MagBeads

[0178] The zeta potential was measured at each step of surface modification of Click MagBeads. The concentration of the beads was maintained at 0.1 mg / mL in PBS, and each sample was replicated over three runs. The amount of mTz on Click MagBeads was determined by back titration of TCO-Cy5 after incubation with the beads. A standard curve of TCO-Cy5 was generated in the range of 0 - 5 pM in PBS using fluorescence at Xex= 620 nm and Xem = 680 nm. 0.5 mg of ClickMagBeads was dispersed in 200 pL of 5 pM TCO-Cy5 solution and allowed to react for 60 min at room temperature. Tire Cy5 -labeled Click MagBeads were magnetically separated, and the Cy5 fluorescence of the supernatant was measured to determine the reacted amount of TCO-Cy5. Hie Cy5 -labeled Click MagBeads were visualized using fluorescence microscopy (Nikon 90i, Nikon, Japan) at far red channel.Synthesis of TCO-grafted antibodies

[0179] Anti-LICAM (R&D Systems, US) or anti-NCAM (Abeam, UK) antibody (1 mg / mU, 20 pU) and TCO-PEG4-NHS (1 mM, 2.7 pL, Click Chemistry Tools) were mixed in net volume 100 pL of PBS (pH 8.5) and incubated for 60 min at room temperature with shaking. Unconjugated TCO- PEG4-NHS was removed by Zeba™ 40 kDa column (Thermo Fisher Scientific, US) as per manufacturer’s instruction. The obtained amount of TCO-grafted antibodies was determined by spectrophotometer (NanoDrop® 2000. Thermo Fisher Scientific, US) with absorbance at wavelength of 280 nm. The degree of labeling (DOL) of TCO was calculated by methyltetrazine-Cy5 (mTz-Cy5) labeling method. In details, TCO-grafted antibodies were reacted with excess amount of mTz-Cy5 for 60 min at room temperature and purified with Zeba™ 40 kDa column. DOL of Cy5 was determined using the protein Cy5 labeling mode of NanoDrop®, yielding an average of ~3 TCO per antibody across the entire batch, regardless of antibody type.Click chemistry-mediated NEV enrichment from CSF and plasma

[0180] NEVs in both CSF and plasma were enriched using Click MagBeads based on the optimized condition from our previous study. In brief, 100 ng of one of the two TCO-grafted antibodies (i.e., TCO-anti-LlCAM or TCO-anti-NCAM) in 0.4 mL of PBS were mixed with 0.4 mL of CSF or plasma for 45 min at room temperature with shaking. The CSF or plasma containing TCO- labeled NEVs was incubated with 100 pg of Click MagBeads, which were preblocked in 100 pL of 10% BSA, for 45 min at room temperature with shaking, followed by a magnetic separation for 30 s. The NEV -enriched Click MagBeads were washed three times by PBST (0.05%) before analyses.Characterization of NEV-enriched Click MagBeads

[0181] NEV-enriched Click MagBeads were analyzed by SEM and immunogold TEM. NEVs in the CSF of AD patient were enriched onto Click MagBeads with TCO-anti-LlCAM and TCO-anti-NCAM using the protocol described in Click chemistry-mediated NEV enrichment fromCSF and plasma section. The NEV-enriched Click MagBeads were fixed in 4% paraformaldehyde (PFA) solution for 30 min at room temperature. For SEM specimens, the samples were washed with ddH2O and prepared as described in SEM imaging section. For immunogold staining, the fixed NEV-enriched Click MagBeads were incubated with either anti-CD63 (MyBioSource, goat, 1:50 dilution) or anti-P-secretase (R&D Systems, goat. 1:50 dilution) in 1% BSA for 60 min at room temperature and washed twice with PBS. Subsequently, the sample was incubated with anti -goat IgG nanogold (Jackson ImmunoResearch, 18 nm, 1:20 dilution) in 1% BSA for 30 min. The nanogold- labeled sample was dropped onto glow discharged TEM grids, and the specimen was prepared as described in TEM imaging section without staining.Investigation of -secretase enzymatic kinetics

[0182] Hie progress curve of -secretase was monitored by using microplate reader. 0.3 pL of P-secretase FRET probes (Sigma- Aldrich, US) at varying concentrations in DMSO were added to 24.7 pL recombinant human P-secretase (Sigma-Aldrich, US) solution (20 mM sodium acetate at pH 4.5, 50 mM NaCl, 0.05% Triton X-100, and 160 nM recombinant human P-secretase in final concentration), followed by enzyme reaction at 37 °C for 60 min with fluorescence signal measurement (Lx = 350 nm and Lm = 490 nm) at every 5 min. Hie hydrolysis rates of P-secretase FRET probes at various concentrations obtained from the progress curve were fitted to the Michaelis- Menten equation to determine the kinetic parameters of P-secretase (Vmax, KM, kcat, and kcat / KM). To obtain the dynamic range, the final concentration of P-secretase FRET probes was fixed at 12 pM, approximately three times the KM, and the probe solution was added in serial dilutions of human recombinant P-sccrctasc solution. For FRET probe specificity test, 160 nM of enzymes (P-sccrctasc, a-secretase, MMP14, and MMP2) were incubated with 12 pM of p-secretase FRET probes at 37 °C.Preparation of synthetic plasma

[0183] Synthetic plasma samples mimicking the plasma of AD patients were prepared to validate the assay’s reliability for detecting NEVs in plasma condition. We first prepared EV-depleted HC plasma through ultracentrifugation. In brief, HC plasma was transferred to six Ultra-Clear Tubes (38 mL per tube, Beckman Coulter, Inc., US) and then ultracentrifuged at 100,000 g for 90 min at 4 °C. Tire top two-thirds (approximately 26 mL per tube) of plasma was carefully collected as EV- depleted HC plasma. Tire synthetic plasma samples were then prepared by spiking the CSF of AD patient into EV-depleted HC plasma, adjusting the final NEV concentration to the desired level. The mixture was aliquoted into duplicates of 0.4 mL each for further analysis.NEV P-secretase Activity Assay using CSF and plasma

[0184] NEVs in both CSF and plasma were enriched using Click MagBeads following the protocol described in Click chemistry-mediated NEV enrichment from CSF and plasma section. 10 pL of lysis buffer (26 mM sodium acetate at pH 4.5, 125 mM NaCl, and 0.1% Triton X-100) was added to lyse the NEVs enriched onto Click MagBeads, followed by incubation at room temperature for 30 min with shaking. Tire Click MagBeads were separated by magnetic separation for 30 s and the supernatant was transferred to 384-well plate (Coming, US). 15 pL of p-secretase FRET probe solution (16 mM sodium acetate at pH 4.5 and 20 pM p-secretase FRET probes) was added and timedependent fluorescence readouts were measured using microplate reader with identical measurement setting described in Investigation of P-secretase enzymatic kinetics section. The P-secretase activity was determined by calculating the hydrolysis rate (RFU / s) from the linear range within the timedependent fluorescence readout curve.Statistical analysis

[0185] Hie results were reported as mean ± standard deviation. The statistic differences among groups were calculated by one-way ANOVA. Statistical analyses were carried out using OriginLab and MedCalc. The optimal cutoff was calculated to maximize the sensitivity and specificity for ROC analysis.

[0186] References for FURTHER EXAMPLE 21. Vogt, A.-C. S., Jennings, G. T., Mohsen, M. O., Vogel, M. 8t Bachmann, M. F. Alzheimer's disease: a brief history of immunotherapies targeting amyloid p. / nt. J. Mol. Sci. 24, 3895 (2023).2. Gustavsson, A. et al. Global estimates on the number of persons across the Alzheimer's disease continuum. A / zheimers Dement. 19, 658-670 (2023).3. Vejandla, B., Savani, S., Appalaneni, R., Veeravalli, R. S. & Gude, S. S. Alzheimer's disease: The past, present, and future of a globally progressive disease. Cureus 16 (2024).4. 2024 Alzheimer's disease facts and figures. Atzheimers Dement. 20, 3708-3821 (2024).5. Jack Jr, C. R. et al. Revised criteria for diagnosis and staging of Alzheimer's disease: Alzheimer's Association Workgroup. A / zheimers Dement. 20, 5143-5169 (2024).6. Betthauser, T. J. et al. Multi-method investigation of factors influencing amyloid onset and impairment in three cohorts. Brain 145, 4065-4079 (2022).Rabinovici, G. D. et ai. Updated appropriate use criteria for amyloid and Tau PET. Alzheimers Dement. 19, e078912 (2023). Shaw, L. M. et al. Appropriate use criteria for lumbar puncture and cerebrospinal fluid testing in the diagnosis of Alzheimer's disease. Alzheimers Dement. 14, 1505-1521 (2018). Hansson, O., Blennow, K., Zetterberg, H. & Dage, J. Blood biomarkers for Alzheimer's disease in clinical practice and trials. Nat. Aging 3, 506-519 (2023). Blennow, K., de Leon, M. J. 8t Zetterberg, H. Alzheimer's disease. Lancet 368, 387-403 (2006). Pospich, S. & Raunser, S. The molecular basis of Alzheimer's plaques. Science 358, 45-46 (2017). Kampmann, M. Molecular and cellular mechanisms of selective vulnerability in neurodegenerative diseases. Nat. Rev. Neurosci. 25, 351 -371 (2024). Jack, C. R. et ai. Tracking pathophysiological processes in Alzheimer's disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol 12, 207-216 (2013). Sevigny, J. et ai. The antibody aducanumab reduces Af> plaques in Alzheimer's disease. Nature 537, 50-56 (2016). Van Dyck, C. H. et ai. Lecanemab in early Alzheimer's disease. AZ Engi. J. Med. 388, 9-21 (2023). Mintun, M. A. et ai. Donanemab in early Alzheimer's disease. AZ Engi. J. Med. 384, 1691 -1704 (2021). Shi, M., Chu, F, Zhu, F. & Zhu, J. Impact of anti-amyloid-|3 monoclonal antibodies on the pathology and clinical profile of Alzheimer's disease: a focus on aducanumab and lecanemab. Front. Aging Neurosci. 14, 870517 (2022). O’brien, R. J. & Wong, P. C. Amyloid precursor protein processing and Alzheimer's disease. Annu. Rev. Neurosci. 34, 185-204 (201 1). Schindler, S. E. et ai. Cerebrospinal fluid biomarkers measured by Elecsys assays compared to amyloid imaging. Alzheimers Dement. 14, 1460-1469 (2018). Chetelat, G. et ai. Amyloid-PET and18F-FDG-PET in the diagnostic investigation of Alzheimer's disease and other dementias. Lancet Neu roi. 19, 951 -962 (2020). Tyler, S. J., Dawbarn, D„ Wilcock, G. K. 8t Allen, S. J. a-and -secretase: profound changes in Alzheimer's disease. Biochem. Biophys. Res. Commun. 299, 373-376 (2002). Fukumoto, H., Cheung, B. S., Hyman, B. T. 8t Irizarry, M. C. -Secretase protein and activity are increased in the neocortex in Alzheimer disease. Arch. Neuroi. 59, 1381 -1389 (2002). Ahmed, R. R. et ai. BACE1 and BACE2 enzymatic activities in Alzheimer's disease. J. Neurochem. 112, 1045-1053 (2010). Simonsen, A. H. et ai. Recommendations for CSF AD biomarkers in the diagnostic evaluation of dementia. Alzheimers Dement. 13, 274-284 (2017).Bouwman, F. H. et al Clinical application of CSF biomarkers for Alzheimer's disease: From rationale to ratios. A / zheimers Dement. 14, e12314 (2022). Lee, J. C., Kim, S. J., Hong, S. & Kim, Y. Diagnosis of Alzheimer’s disease utilizing amyloid and tau as fluid biomarkers. Exp. Mol. Med. 51, 1 -10 (2019). Shen, Y. et a / . Increased plasma beta-secretase 1 may predict conversion to Alzheimer's disease dementia in individuals with mild cognitive impairment. Biol. Psychiatry 83, 447-455 (2018). Cervellati, C. et a / Serum beta-secretase 1 (BACE1) activity as candidate biomarker for late- onset Alzheimer's disease. Geroscience 42, 159-167 (2020). Zuliani, G. et at Serum beta-secretase 1 (BACE1) activity increases in patients with mild cognitive impairment. J. Neurochem. 159, 629-637 (2021). Liang, Y, Lehrich, B. M., Zheng, S. 8t Lu, M. Emerging methods in biomarker identification for extracellular vesicle-based liquid biopsy. J. Extracell. Vesicles 10, e12090 (2021). Min, L. et al. Advanced nanotechnologies for extracellular vesicle-based liquid biopsy. Adv. Sci. 8, 2102789 (2021). Blanchette, C. R. 8t Rodal, A. A. Mechanisms for biogenesis and release of neuronal extracellular vesicles. Curr. Opin. Neurobiol. 63, 104-1 10 (2020). Chen, Y, Zhao, Y., Yin, Y., Jia, X. & Mao, L. Mechanism of cargo sorting into small extracellular vesicles. Bioengineered 12, 8186-8201 (2021). Ramos-Zaldivar, H. M. et al. Extracellular vesicles through the blood-brain barrier: a review. Fluids Barriers CNS V , 60 (2022). Abdelsalam, M., Ahmed, M., Osaid, Z., Hamoudi, R. & Harati, R. Insights into exosome transport through the blood-brain barrier and the potential therapeutical applications in brain diseases. Pharmaceuticals (Basel) 16, 571 (2023). Lee, J.-H. et at. Alzheimer's disease protease-containing plasma extracellular vesicles transfer to the hippocampus via the choroid plexus. eBioMedicine ll , 103903 (2022). Mustapic, M. et al. Plasma extracellular vesicles enriched for neuronal origin: a potential window into brain pathologic processes. Front. Neurosci. 11, 278 (2017). Rocha, D. A., Santos, L. E„ Da Fonseca, P. B. & De Felice, F. G. Prospects and challenges in using neuronal extracellular vesicles in biomarker research. Alzheimers Dement. 20, 6632-6638 (2024). You, Y. et at. ATP1 A3 as a target for isolating neuron-specific extracellular vesicles from human brain and biofluids. Sci. Adv. 9, eadi3647 (2023). Delgado-Peraza, F. et al. Neuron-derived extracellular vesicles in blood reveal effects of exercise in Alzheimer's disease. Alzheimers Res. Then 15, 156 (2023).Sun, N. et al. Purification of HCC-specific extracellular vesicles on nanosubstrates for early HCC detection by digital scoring. Nat. Common. 11, 4489 (2020). Dong, J. et al. Coupling nanostructured microchips with covalent chemistry enables purification of sarcoma-derived extracellular vesicles for downstream functional studies. Adv. Funct. Mater. 30, 2003237 (2020). Sun, N. et al. Coupling lipid labeling and click chemistry enables isolation of extracellular vesicles for noninvasive detection of oncogenic gene alterations. Adv. Sci. 9, 2105853 (2022). Sun, N. et al. HCC EV ECG score: an extracellular vesicle-based protein assay for detection of early-stage hepatocellular carcinoma. Hepatology!! , 774-788 (2023). Blackman, M. L, Royzen, M. & Fox, J. M. Tetrazine ligation: fast bioconjugation based on inverse-electron-demand Diels-Alder reactivity. J. Am. Chem. Soo. 130, 13518-13519 (2008). Dong, J. et al. Bio-inspired nanovilli chips for enhanced capture of tumor-derived extracellular vesicles: toward non-invasive detection of gene alterations in non-small cell lung cancer. ACS-13983 (2019). Wang, J. J. et al. Prostate cancer extracellular vesicle digital scoring assay - a rapid noninvasive approach for quantification of disease-relevant mRNAs. Nano Today A , 101746 (2023). Badhwar, A. et a! Assessment of brain-derived extracellular vesicle enrichment for blood biomarker analysis in age-related neurodegenerative diseases: An international overview. A / zheimers Dement. 20, 441 1 -4422 (2024). Norman, M. et al. L1 CAM is not associated with extracellular vesicles in human cerebrospinal fluid or plasma. Nat. Methods 18, 631 -634 (2021). Nogueras-Ortiz, C. J. et al. Single-extracellular vesicle (EV) analyses validate the use of L1 Cell Adhesion Molecule (L1 CAM) as a reliable biomarker of neuron-derived EVs. J. Extracell. Vesicles 13, e12459 (2024). Sharpies, R. A. et at Inhibition of y-secretase causes increased secretion of amyloid precursor protein C-terminal fragments in association with exosomes. FASEB J. 22, 1469-1478 (2008). Soares Martins, T. et at Diagnostic and therapeutic potential of exosomes in Alzheimer's disease. J. Neurochem. 156, 162-181 (2021). Perez-Gonzalez, R„ Gauthier, S. A., Kumar, A. & Levy, E. The exosome secretory pathway transports amyloid precursor protein carboxyl-terminal fragments from the cell into the brain extracellular space. J. Biol. Chem. 287, 43108-431 15 (2012). Natale, F., Fusco, S. & Grassi, C. Dual role of brain-derived extracellular vesicles in dementia- related neurodegenerative disorders: cargo of disease spreading signals and diagnostic- therapeutic molecules. Trans! Neurodegener. 1 1, 50 (2022). Ter-Ovanesyan, D. et al. Framework for rapid comparison of extracellular vesicle isolation methods. eLife 10, e70725 (2021).56. Welsh, J. A. et al. Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. J. Extracell. Vesicles 13, e12404 (2024).57. Griininger-Leitch, F., Schlatter, D., Kung, E., Nelbbck, P. & Dobeli, H. Substrate and inhibitor profile of BACE (|3-secretase) and comparison with other mammalian aspartic proteases. J. Biot. Chem. 277, 4687-4693 (2002).58. Turner, R. T. et al. Subsite specificity of memapsin 2 (p-secretase): implications for inhibitor design. Biochemistry 40, 10001 -10006 (2001).59. Ermolieff, J., Loy, J. A., Koelsch, G. & Tang, J. Proteolytic activation of recombinant pro- memapsin 2 (pro-|3-secretase) studied with new fluorogenic substrates. Biochemistry 39, 12450-12456 (2000).60. Li, Y, Liu, T., Zeng, Q. & Cui, M. Current status of PET tracers for the early diagnosis of Alzheimer's disease. Trends Ana / Chem. 172, 117546 (2024).61. Amft, M. et a / . The cerebrospinal fluid biomarker ratio Af>42 / 40 identifies amyloid positron emission tomography positivity better than A[342 alone in a heterogeneous memory clinic cohort. Alzheimers Res. Then 14, 60 (2022).62. Fiandaca, M. S. et al. Identification of preclinical Alzheimer's disease by a profile of pathogenic proteins in neurally derived blood exosomes: a case-control study. Alzheimers Dement. 11, 600-607. e601 (2015).63. Li, Y. et al. Amyloid-^ protein and MicroRNA-384 in NCAM-Labeled exosomes from peripheral blood are potential diagnostic markers for Alzheimer's disease. CNS Neurosci. Then 28, 1093- 1 107 (2022).64. Zipfel, P., Rochais, C., Baranger, K., Rivera, S. & Dallemagne, P. Matrix metalloproteinases as new targets in Alzheimer's disease: Opportunities and challenges. J. Med. Chem. 63, 10705- 10725 (2020).65. Radosinska, D. 8t Radosinska, J. The link between matrix metalloproteinases and Alzheimer's disease pathophysiology. Mol. Neurobiol. 1 -15 (2024).

[0187] References for Table 4 (FIG. 45)1. Fiandaca, M. S.; Kapogiannis, D.; Mapstone, M.; Boxer, A.; Eitan, E.; Schwartz, J. B.; Abner, E. L.; Petersen, R. C.; Federoff, H. J.; Miller, B. L, Identification of preclinical Alzheimer's disease by a profile of pathogenic proteins in neurally derived blood exosomes: a case-control study. Alzheimer's & Dementia 2015, 77 (6), 600-607. e1 .2. Vandendriessche, C.; Kapogiannis, D.; Vandenbroucke, R. E., Biomarker and therapeutic potential of peripheral extracellular vesicles in Alzheimer's disease. Advanced drug deliveryreviews 2022, 190, 1 14486.3. Pei, J.; Palanisamy, C. P.; Jayaraman, S.; Natarajan, P. M.; Umapathy, V. R.; Roy, J. R.; Thalamati, D.; Ahalliya, R. M.; Kanniappan, G. V; Mironescu, M., Proteomics Profiling of Extracellular Vesicle for Identification of Potential Biomarkers in Alzheimer's Disease: A Comprehensive Review. Ageing Research Reviews 2024, 102359.4. Sathe, G.; Na, C. H.; Renuse, S.; Madugundu, A. K.; Albert, M.; Moghekar, A.; Pandey, A., Quantitative proteomic profiling of cerebrospinal fluid to identify candidate biomarkers for Alzheimer's disease. PROTEOMICS-Clinical Applications 2019, 13 (4), 1800105.5. Croese, T.; Furlan, R., Extracellular vesicles in neurodegenerative diseases. Molecular aspects of medicine 2018, 60, 52-61.6. Shi, M.; Sheng, L; Stewart, T.; Zabetian, C. P.; Zhang, J., New windows into the brain: Central nervous system-derived extracellular vesicles in blood. Progress in neurobioiogy GV), 175, 96-106.7. Boyer, E.; Deltenre, L; Dourte, M.; Colmant, L; Paftre, E.; Sleegers, K.; Suelves, N.; Hanseeuw, B.; Kienlen-Campard, R, Comparison of plasma soluble and extracellular vesicles-associated biomarkers in Alzheimer's disease patients and cognitively normal individuals. Alzheimers' research & therapy 2024, 75 (1), 141.8. Jia, L; Qiu, Q.; Zhang, H.; Chu, L; Du, Y.; Zhang, J.; Zhou, C.; Liang, F.; Shi, S.; Wang, S., Concordance between the assessment of A|342, T-tau, and P-T181 -tau in peripheral blood neuronal-derived exosomes and cerebrospinal fluid. Alzheimers' & Dementia 2019, 75 (8), 1071 - 1080.9. Li, Y; Meng, S.; Di, W.; Xia, M.; Dong, L; Zhao, Y.; Ling, S.; He, J.; Xue, X.; Chen, X., Amyloid- 13 protein and MicroRNA-384 in NCAM-Labeled exosomes from peripheral blood are potential diagnostic markers for Alzheimer's disease. CNS Neuroscience & Therapeutics 2022, 28 (7), 1093- 1 107.10. Li, Y; Xia, M.; Meng, S.; Wu, D.; Ling, S.; Chen, X.; Liu, C., MicroRNA-29c-3p in dual-labeled exosome is a potential diagnostic marker of subjective cognitive decline. Neurobiology of Disease 2022, 171, 105800.

[0188] The embodiments illustrated and discussed in this specification are intended only to teach those skilled in the art how to make and use the invention. In describing embodiments of the invention, specific terminology is employed for the sake of clarity. However, the invention is not intended to be limited to the specific terminology so selected. The above-described embodiments of the invention may be modified or varied, without departing from the invention, as appreciated by those skilled in the art in light of the above teachings. It is therefore to be understood that, within the scope of the claims and their equivalents, the invention may be practiced otherwise than as specifically described.

Claims

We claim:

1. A method of assaying for activity of a protease enzyme corresponding to a disease in a subject, comprising: selectively capturing an extracellular vesicle (EV) from a sample from said subject; contacting said EV after said capturing with a probe molecule, said probe molecule comprising a peptide, a fluorescent moiety on a first end of said peptide and a quenching moiety on a second end of said peptide to provide a Fluorescence Resonance Energy Transfer (FRET) pair separated by said peptide, said peptide containing a cleavage sequence for said protease enzyme; and measuring a fluorescence of at least said fluorescent moiety after said contacting to indicate a presence or absence of the occurrence of cleavage of said peptide by said protease enzyme.

2. Tire method according to claim 1, further comprising lysing said EV prior to said contacting.

3. Tire method according to claim 1 or 2, wherein said selectively capturing said EV from said sample comprises attaching said EV onto a magnetic particle.

4. The method according to claim 1 or 2, wherein said selectively capturing said EV from said sample comprises attaching said EV onto a structured substrate.

5. The method according to any one of claims 1-4, wherein said selectively capturing said EV from said sample comprises exposing said EV to at least a portion of a binding molecule that is functionalized with a selected antibody.

6. A kit for assaying for activity of a protease enzyme corresponding to a disease in a subject, comprising: at least one of a structure substrate or a plurality of magnetic particles;a first solution comprising a plurality of binding moieties, each binding moiety functionalized to attach to said at least one of said structure substrate or a magnetic particle of said plurality of magnetic particles; a second solution comprising a plurality of binding moieties, each binding moiety functionalized with a selected antibody to attach to an extracellular vesicle (EV) of interest; and a third solution comprising a plurality of probe molecules, each probe molecule comprising a peptide, a fluorescent moiety on a first end of said peptide and a quenching moiety on a second end of said peptide to provide a Fluorescence Resonance Energy Transfer (FRET) pair separated by said peptide, said peptide containing a cleavage sequence for a protease enzyme of interest.

Images