Methods and compositions for rapid intraoperative histopathologic diagnosis

Abstract

Histopathologic diagnoses, obtained from biopsies, guide surgery and postoperative care for patients. However, cunent diagnostic techniques can be too slow to provide specific diagnoses during surgery, which may lead to sub-optimal surgical care and potentially necessitate additional surgeries for patients. Systems, methods, kits, and compositions described herein addresses this unmet need for improving patient care by combining applications of novel aptamer-based molecular nanoprobes that quickly label clinically proven biomarkers and new deep learning image analysis to provide physicians and surgeons fast and precise intraoperative histopathologic diagnoses.

Description

September 23, 2025METHODS AND COMPOSITIONS FOR RAPID INTRAOPERATIVE HISTOPATHOLOGIC DIAGNOSISCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of U.S. Provisional Application No. 63 / 724,196 filed November 22, 2024, and U.S. Provisional Application No. 63 / 697,813 filed September 23, 2024, the specifications of which are incorporated herein in their entirety by reference.REFERENCE TO AN ELECTRONIC SEQUENCE LISTING

[0002] The contents of the electronic sequence listing (SKSG 24 03 PCT.xml; Size: 8,192 bytes; and Date of Creation: September 22, 2025) is herein incorporated by reference in its entirety.FIELD OF THE INVENTION

[0003] The present disclosure features methods and compositions for fast and precise intraoperative histopathologic diagnosis facilitated by multivalent aptamer nanoprobe designs and deep learning image analysis.BACKGROUND OF THE INVENTION

[0004] The current challenge in intraoperative tumor management is the need for both rapid and precise histopathologic diagnoses to guide critical surgical decisions. Conventional clinical approaches, such as frozen tissue sections followed by antibody staining, suffer from extended processing times and lack the specificity required for effective real-time interventions. The present disclosure addresses this unmet need by developing an innovative system that combines programmable DNA nanoprobes with an electric field (ELF)-driven staining method and advanced Al-based image analysis to achieve ultra-fast, high-precision tissue diagnostics.

[0005] The present disclosure employs DNA nanostructure scaffolded multivalent aptamer-based probes and multiple fluorophores attached on to the scaffold, designed for swift and efficient binding of tumor-specific biomarkers. The use of ELF technology accelerates the migration and penetration of these nanoprobes into tissues, drastically reducing the staining time compared to traditional methods. To complement this, a generative Al model was introduced that enhances image clarity by denoising ELF-stained images, eliminating the need for blocking and washing steps and further expediting the workflow. This Al framework, equipped with interpretability features such as GradCAM, supports pathologists and surgeons by automating image analysis and facilitating accurate, real-time decision-making.September 23, 2025BRIEF SUMMARY OF THE INVENTION

[0006] It is an objective of some embodiments to provide compositions and methods that reduce the specific diagnosis time to less than 20 minutes, which is critical for effective intraoperative decision-making, as specified in the independent claims. Embodiments of the invention are given in the dependent claims. Embodiments of the present disclosure can be freely combined with each other if they are not mutually exclusive.

[0007] The present disclosure addresses the urgent need for the rapid intraoperative diagnosis of tumor tissues by developing multivalent DNA nanoprobes that quickly and specifically label cancer biomarkers. Electric filed enhanced staining (ELF) is used and coupled with advanced deep learning models for fast image analysis and reporting of results, which will enable surgeons to make more informed, real-time decisions. This integration of ELF staining, molecular nanotechnology, and Al transforms patient outcomes by significantly reducing diagnosis times with excellent accuracy during surgery where every minute counts.

[0008] In some embodiments, the present disclosure describes an application for brain tumor surgeries, where rapid and accurate diagnosis is critical. Specifically, it aims to distinguish between primary CNS lymphoma (PCNSL), which is typically non-operative, and glioblastoma (GBM), which often requires extensive resection. However, the technology described herein may also be extended to other cancers, such as skin cancer / melanoma, breast cancer, gynecological cancer, colorectal cancer, prostate cancer, and others, where rapid and precise diagnosis is essential.

[0009] In some embodiments, methods described herein may be used for the prognosis or diagnosis of cancers including but not limited to skin cancer, breast cancer, colorectal cancer, prostate cancer, renal cancer, brain cancer, metastatic cancer, pancreatic cancer, lung cancer, liver cancer, bladder cancer, bone sarcoma, ovarian cancer, rectal cancer, blood cancer, gastrointestinal cancer, among others. For example, in cancer applications, the technology described herein may be directed towards prognosis and the use of prognostic markers to tailor treatment, employing a precision medicine approach. In some embodiments, this technology is used in conjunction with flow cytometry, utilizing fluorescence-based methods for analysis.

[0010] In non-limiting embodiments, this technology can use either light microscopy such as with chromogens or fluorescent microscopy for better results in melanoma applications. Kappa and lambda light chains are used to help diagnose B cell lymphomas and other lymphoproliferative disorders. This technology can target CD20 and kappa and lambda diagnosis to diagnose several types of melanomas than current methods, potentially all main types of melanomas.September 23, 2025

[0011] In some embodiments, the present disclosure features a multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target protein. In some embodiments, the nanoprobe comprises a) a DNA nanostructure; b) at least two aptamers bound to the DNA nanostructure, and c) one or more reporters bound to the DNA nanostructure. In some embodiments, each of the aptamers is configured to bind to the target protein.

[0012] In some embodiments, the aptamers are comprised of DNA sequences. In other embodiments, the aptamers are comprised of RNA sequences. In some embodiments, the aptamers are linked to the DNA nanostructure. In some embodiments, the reporter is covalently conjugated to the DNA nanostructure.

[0013] In some embodiments, the target protein is a biomarker for cancer (e.g., brain cancer, skin cancer, breast cancer, colorectal cancer, prostate cancer, or the like). In some embodiments, the target protein is glial fibrillary acidic protein (GFAP) or CD20.

[0014] In some embodiments, the reporter is a fluorescence reporter (e.g., fluorescence dye), fluorochrome, or an enzyme reporter. Non-limiting examples of fluorescence dyes t include but are not limited to Alexa Fluor™ 488 fluorescent dye, Alexa Fluor™ 647 fluorescent dye, or Alexa Fluor™ 633 fluorescent dye. In some embodiments, the chromogen is magenta. Additionally, nonlimiting examples of enzyme reporters may include a chromogen such as 3, 3 '-diaminobenzidine tetrahydrochloride (DAB), horseradish peroxidase (HRP), or alkaline phosphatase (AP). Other reporters may be used in accordance with the present invention.

[0015] In some embodiments, the present disclosure may further comprise a method of diagnosing cancer in a subject in need thereof. In some embodiments, the method comprises a) obtaining or having obtained a tissue sample from the subject, b) staining the tissue sample using the nanoprobe as described herein, for a period of time; and c) analyzing the stained tissue sample using an imaging-based deep learning (DL) model; wherein the imaging-based DL model detects a level of staining in the tissue sample. In some embodiments, the subject is diagnosed with cancer if the staining is above a predetermined threshold as determined by the imaging-based DL model.

[0016] In some embodiments, the method further comprises applying an electric field to the tissue sample to accelerate staining.

[0017] In some embodiments, the tissue sample is from a solid tumor. In some embodiments, the solid tumor is within the brain of the subject. In some embodiments, the cancer is GlioblastomaSeptember 23, 2025 multiforme (GBM).

[0018] In some embodiments, a kit is featured comprising a multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target protein, one or more of stains, reagents, or combinations thereof, and instructions comprising combining the nucleotide composition and the one or more of stains, reagents, or combinations thereof, and staining a tissue sample. The nanoprobe may comprise a DNA nanostructure and at least two aptamers bound to the DNA nanostructure, and one or more reporters bound to the DNA nanostructure. Each of the aptamers is configured to bind to the target protein.

[0019] In some embodiments, the technology described herein is used for drug development for oncology including use as a companion diagnostic or biomarker diagnostic. It may aid in patient stratification, addressing heterogeneity, and supporting clinical development of molecularly targeted drugs. It may be used to detect and monitor tumor-targeted drug delivery processes, tumor accumulation of medicines, and accurately assess and predict medicine target-site localization. It may be used with nanomedicines.

[0020] One of the technical features of the system, method, kit, and composition is the integrated solution (e.g., molecular nanotechnology and Al innovations). The technical features of some embodiments advantageously reduce the total histopathologic diagnosis staining process to under 5 minutes. By integrating molecular nanotechnology with Al innovations, the present systems, methods, kits, and compositions can be expanded to several types of tumors, significantly enhancing surgical outcomes and patient care. None of the presently known prior references or works have the unique technical feature of the present disclosure.

[0021] Another technical feature of the system, method, kit, and composition is the use of multivalent nanoprobes comprising multiple aptamers and fluorophores. As illustrated in FIG. 9B, when the particle concentration is held constant, a multivalent nanoprobe containing three fluorophores demonstrates improved performance compared to a mono-aptamer probe with a single fluorophore. In additional experiments using equivalent total fluorophore concentrations (e.g., 0.1 pM for the nanoprobe and 0.3 pM for the mono-aptamer), staining intensity was comparable; however, the multivalent nanoprobe exhibited significantly reduced background and nonspecific signal relative to the mono-aptamer. This demonstrates that the multivalent design enhances signal specificity while maintaining effective fluorescence intensity.

[0022] Any feature or combination of features described herein are included within the scope ofSeptember 23, 2025 the present disclosure provided that the features included in any such combination are not mutually inconsistent as will be apparent from the context, this specification, and the knowledge of one of ordinary skill in the art. Additional advantages and aspects of the present disclosure are apparent in the following detailed description and claims.BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING(S)

[0023] The features and advantages of further embodiments will become apparent from a consideration of the following detailed description presented in connection with the accompanying drawings in which:

[0024] FIG. 1 shows an overview diagram of the staining process. By developing a multivalent aptamer DNA nanoprobe and novel ELF staining method with Al denoising, a tissue section may be stained in under 5 minutes.

[0025] FIG. 2 shows a shortened post-frozen section process by ELF staining and Al denoising. The most time-consuming steps after frozen section are blocking and staining, may be dramatically shortened with the developed methods.

[0026] FIG. 3 shows xenograft Biopsy Staining with PCNSL-Specific Aptamer. Eleven-minute staining protocol with 1 pM PCNSL-specific aptamer identifies 76.7 ± 15.1% of PCNSL cells with a false positive rate of 0.81 ± 1.75% from negative control GBM xenograft biopsies. Note areas with fluorescent artifacts that lack ring-like staining pattern (arrows). Scale bars equal 20p.m.

[0027] FIG. 4 shows standard Histologic Staining of PCNSL Xenograft Biopsies Compared to Aptamer. Hematoxylin and Eosin staining show regions of hypercellular tumor (left, top and bottom). PCNSL biopsies show strong CD20 staining (middle, top), in contrast to GBM without CD20 staining (middle, bottom). High magnification imaging shows ring-like staining pattern of CD20 (middle, top), and similar aptamer staining pattern of RFP-expressing PCNSL cells (top, right). High magnification image of GBM biopsy lacks CD20 staining (bottom, middle), and lack of ring-like aptamer staining in RFP-expressing GBM biopsy (bottom, right). Scale bars equal 20 p.

[0028] FIG. 5 shows a schematic of multivalent aptamers nanoprobe binding to homo-oligomers ofCD20.

[0029] FIGs. 6A and 6B show a schematic of multivalent aptamers DNA nanostructure. FIG. 6A shows a multivalent DNA nanoprobe and single aptamer probe. FIG. 6B shows a multivalent aptamer and multiple fluorophore DNA nanoprobe and mono-aptamer binding with intermediate filament which assembly with GFAP. IF is intermediate filament.

[0030] FIGs. 7A, 7B, and 7C show aptamer candidates against GFAP. FIG. 7A shows predicted secondary structure for the three most promising aptamer candidates by mFold. FIG. 7B showsSeptember 23, 2025SPR based assessment of aptamer against recombinant GFAP. Lines from top to bottom represent 1 / 2X, 1 / 4X, 1 / 8X, 1 / 16X, 1 / 32X, respectively. FIG. 7C shows cell binding assay. Percentage of aptamer binding determined by subtracting the unbound fraction from 100% in Ramos and U251 cell lines.

[0031] FIGs. 8A and 8B show aptamer intracellular staining and specificity. FIG. 8A shows three different aptamers against GFAP were used to stain human U251 cell lines expressing GFAP protein. FIG. 8B shows three cell lines that were stained by fluorescent-labeled Apt3 and imaging by confocal microscopy.

[0032] FIGs. 9A, 9B, and 9C show the evaluation of multivalent aptamer DNA nanoprobe and mono-aptamer. FIG. 9A shows a schematic representation of both multi-apt3 and mono-apt3. FIG. 9B shows U251 cells were stained by multi- Apt3 and mono-Apt3. FIG. 9C shows a comparison of intergrade density of intracellular staining from two probes.

[0033] FIG. 10 shows comparative cellular imaging with DNA nanoprobe tagged with diverse fluorophores. U251 cells were stained with DNA nanoprobes carrying a range of fluorophores quantities.

[0034] FIGs. HA and 1 IB show dynamic evaluation of staining pattern: comparative cellular staining and imaging with DNA nanoprobe and antibodies across time points. FIG. 11A shows an assessment via confocal microscopy. FIG. 1 IB shows analysis via flow cytometry.

[0035] FIG. 12 shows Staining pattern and tissue specificity evaluation of DNA nanoprobe on xenograft tumor section.

[0036] FIGs. 13 A, 13B, and 13C show ELF staining for the GFAP aptamer stains xenograft tumors in approximately 1 minute with exceptionally high specificity. FIG. 13 A shows a diagram of the ELF staining device. FIG. 13B shows voltage values and current directions applied during the 1-minute staining process. FIG. 13C shows xenograft tumor and normal brain tissue stained using the 1-minute ELF method with Alexa647-labeled target aptamer (Apt 3) and a scramble aptamer, then imaged by confocal microscopy. Red indicates the aptamer signal, while blue represents nucleus.

[0037] FIG. 14 shows a denoising Al model for blocking free tissue staining. The diagram illustrates training and application of the denoising model to remove non-specific signals in blocking-free tissue, improving image clarity without requiring a blocking step.

[0038] FIG. 15 shows a denoising Al model for blocking free tissue staining. Left: non-block staining, non-specific binding is shown in nuclei. Middle: BlobCUT rendered synthetic blocking staining. Right: block staining.

[0039] FIG. 16 shows a schematic diagram of tumor margin determination.September 23, 2025

[0040] FIG. 17 shows confocal images of xenograft U251 tumor tissue stained with 3HB and 3HB-Apt3. 3HB was labeled with Alex647. Almost no unspecific staining from 3HB only could be seen on the tissue compared with 3HB-Apt3. Scale bar: 100pmDETAILED DESCRIPTION OF THE INVENTION

[0041] Disclosed are various peptides, solvents, solutions, carriers, and / or components to be used to prepare compositions to be used within the methods disclosed herein. Also disclosed are the various steps, elements, amounts, routes of administration, symptoms, and / or treatments that are used or observed when performing the disclosed methods, as well as the methods themselves. These and other materials, steps, and / or elements are disclosed herein, and it is understood that when combinations, subsets, interactions, groups, etc. of these materials are disclosed, that while specific reference of each various individual and collective combination and permutation of these compounds may not be explicitly disclosed, each is specifically contemplated and described herein. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0042] Unless otherwise explained, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which a disclosed invention belongs. The singular terms "a," "an," and "the" include plural referents unless context clearly indicates otherwise. Similarly, the word "or" is intended to include "and" unless the context clearly indicates otherwise. The term "comprising" means that other elements can also be present in addition to the defined elements presented. The use of "comprising" indicates inclusion rather than limitation. Stated another way, the term "comprising" means "including principally, but not necessarily solely". Furthermore, variation of the word "comprising", such as "comprise" and "comprises", have correspondingly the same meanings. In one respect, the technology described herein related to the herein described compositions, methods, and respective component(s) thereof, as essential to the invention, yet open to the inclusion of unspecified elements, essential or not ("comprising").

[0043] Suitable methods and materials for the practice and / or testing of embodiments of the disclosure are described below. Such methods and materials are illustrative only and are not intended to be limiting. Other methods and materials similar or equivalent to those described herein can be used. For example, conventional methods well known in the art to which the disclosure pertains are described in various general and more specific references, including, for example, Sambrook et al., Molecular Cloning: A Laboratory Manual, 2d ed., Cold Spring Harbor LaboratorySeptember 23, 2025Press, 1989; Sambrook et al., Molecular Cloning: A Laboratory Manual, 3d ed., Cold Spring Harbor Press, 2001; Ausubel et al., Current Protocols in Molecular Biology, Greene Publishing Associates, 1992 (and Supplements to 2000); Ausubel et al., Short Protocols in Molecular Biology: A Compendium of Methods from Current Protocols in Molecular Biology, 4th ed., Wiley & Sons, 1999; Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, 1990; and Harlow and Lane, Using Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, 1999, Gene Expression Technology (Methods in Enzymology, Vol. 185, edited by D. Goeddel, 1991. Academic Press, San Diego, Calif.), "Guide to Protein Purification” in Methods in Enzymology (M. P. Deutshcer, ed., (1990) Academic Press, Inc.); PCR Protocols: A Guide to Methods and Applications (Innis, et al. 1990. Academic Press, San Diego, Calif.), Culture of Animal Cells: A Manual of Basic Technique, 2nd Ed. (R. I. Freshney. 1987. Liss, Inc. New York, N.Y.), Gene Transfer and Expression Protocols, pp. 109-128, ed. E. J. Murray, The Humana Press Inc., Clifton, N.J.), and the Ambion 1998 Catalog (Ambion, Austin, Tex.), the disclosures of which are incorporated in their entirety herein by reference.

[0044] All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety for all purposes. In case of conflict, the present specification, including explanations of terms, will control.

[0045] Although methods and materials similar or equivalent to those described herein can be used to practice or test the disclosed technology, suitable methods and materials are described below. The materials, methods, and examples are illustrative only and not intended to be limiting.

[0046] As used herein, the terms “subject” and “patient” are used interchangeably. As used herein, a subject can be a mammal such as a non-primate (e.g., cows, pigs, horses, cats, dogs, rats, etc.) or a primate (e.g., monkey and human). In specific embodiments, the subject is a human. In one embodiment, the subject is a mammal (e.g., a human) having a disease, disorder or condition described herein. In another embodiment, the subject is a mammal (e.g., a human) at risk of developing a disease, disorder, or condition described herein. In certain instances, the term patient refers to a human.

[0047] Referring now to FIGs. 1-17, the present disclosure features methods and compositions for fast and precise intraoperative histopathologic diagnosis facilitated by multivalent aptamer nanoprobe designs and deep learning image analysis. Effective surgical care and adjuvant therapy for tumor patients depend on accurate histopathological diagnosis of biopsies acquired during surgery. However, current histopathological methods lack the simultaneous speed and specificitySeptember 23, 2025 required for accurate diagnosis and real-time surgical decision-making. Here, the present disclosure describes a comprehensive approach combining novel molecular diagnostic probe designs with an electric field (ELF)-driven staining method and new multi-task deep learning (MTL) models to address the critical needs of intraoperative tumor margin determination and preresection diagnostics.

[0048] In some embodiments, the staining process is completed in less than 20 minutes. In other embodiments, the staining process is completed in less than 10 minutes, and in some cases, less than 5 minutes. While completion in under 5 minutes is not essential for the invention to function effectively, the process can achieve sufficient accuracy in approximately 20 minutes, approximately 10 minutes, or as quickly as around 5 minutes or less. Without intending to limit the present invention to any specific mechanism, it is believed that reducing the staining process time to under 20 minutes, or even under 5 minutes allows pathologists to evaluate frozen sections more efficiently.

[0049] In some embodiments, the technology is applied to brain cancer, with additional applications extending to other cancers such as skin cancer / melanoma, breast cancer, gynecological cancer (such as cervical cancer, ovarian cancer, vulvar cancer, vaginal cancer and endometrial / uterine cancer), colorectal cancer, prostate cancer, and more. While brain tumor intraoperative diagnosis has been selected as the initial focus for development and validation, the methods and compositions can, in some embodiments, be adapted for use with other tumor types requiring rapid and accurate diagnoses.

[0050] Brain tumor surgery is highly dependent on tumor type. For example, primary CNS lymphoma (PCNSL) is generally considered non-operative, whereas glioblastoma (GBM) often requires maximal safe resection. Additionally, the extent of GBM resection is critical, as the percentage of tumor residue left behind directly impacts overall survival (OS) and recurrence rates. Thus, correct diagnosis of tumor types before resection and the identification of GBM residue during surgery are crucial but often depend heavily on the expertise of neuropathologists — scarce resources that require years of training. Even with highly skilled pathologists, rapid diagnosis using cytological smears or frozen sections can lead to misdiagnoses. Targeting specific biomarkers can enhance the reliability of rapid diagnostic methods, particularly in environments where expert oversight is limited. Currently, the most widely employed immunohistochemistry (IHC) methods in clinical practice require 30 to 45 minutes to produce a diagnosis, where over half of this total turnaround time is the staining protocol. The present disclosure achieves fast and preciseSeptember 23, 2025 intraoperative histopathological diagnosis and utilizes two critical components: (1) highly efficient molecular probes capable of rapid tissue penetration and high-affinity, specific binding to targets and (2) an electric field-driven staining method combined with an advanced denoising Al model. The electric field accelerates the migration of DNA-base probes, and the Al model denoises images for blocking- and washing-free staining. Together, these innovations reduce the entire staining process to under 5 minutes while providing comparable or improved staining patterns to traditional methods, thereby expediting the pathologists' decision-making.

[0051] The present disclosure features a DNA nanostructure-scaffolded multivalent aptamerbased nanoprobe with multiple fluorophores, combined with ELF staining to enhance the speed, accuracy, and efficiency of clinical intraoperative histopathologic diagnosis. Additionally, an advanced denoising Al model can be used to eliminate the need for blocking and washing, further simplifying and accelerating the process.

[0052] In some embodiments, the present disclosure features a novel DNA nanoprobe for the rapid cell-membrane staining of PCNSL and evaluation of its clinical potential for specific intraoperative diagnosis of PCNSL. Without wishing to limit the present invention to any theory or mechanism it is believed that the multivalent aptamer nanoprobes described herein allow for the rapid cellular membrane molecule staining of PCNSL frozen sections and ex vivo biopsies to gain rapid and specific intraoperative diagnosis of PCNSL similarly accurate to IHC.

[0053] In other embodiments, the present disclosure features a multivalent-aptamer DNA nanoprobe for intracellular target labeling of GBM. Referring to FIG. 8A, 3 aptamers were selected for glial fibrillary acid protein (GFAP), an intracellular tumor marker for GBM but not significantly expressed in non-glial tumor types such as PCNSL. Referring to FIG. 9A-9C, a multivalent aptamer-based fluorescent DNA nanoprobe was developed, as well as a 25-minute turnaround time of staining protocol for specific intracellular staining of GFAP in GBM cell lines.

[0054] In further embodiments, methods described herein shorten the staining time by electric field (ELF) driven staining method and Generative Al denoising model to support pre-resection tumor type diagnosis and intraoperative tumor margin determination. The methods described herein significantly accelerate the staining process and enhance image clarity for intraoperative diagnosis by integrating an innovative ELF -driven staining method with an advanced Al-based denoising tool. Additionally, stain intensity and patterns may be incorporated as domain knowledge within the segmentation and classification framework of the MTL model, facilitating precise pre-resection tumor type diagnosis and intraoperative tumor margin determination. TheSeptember 23, 2025 classification head of the model may be enhanced with interpretability features, including GradCAM, to support and strengthen clinical validation.

[0055] The present disclosure features may multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target molecule. In some embodiments, the nanoprobe comprises a) a DNA nanostructure; b) at least two aptamers bound to the DNA nanostructure, and c) one or more reporters bound to the DNA nanostructure. In some embodiments, each of the aptamers is configured to bind to the target molecule.

[0056] In some embodiments, the target molecule is a target protein. In some embodiments, the target molecule is a target nucleotide (e.g., DNA or RNA). In some embodiments, the target molecule is a microRNA (miRNA), circular RNA (circRNA), small, non-coding RNAs, or the like. In other embodiments, the target molecule is a cell-free DNA (cfDNA). In further embodiments, the target molecule is a target gene.

[0057] In some embodiments, the nanoprobe comprises a) a DNA nanostructure; b) at least two aptamers bound to the DNA nanostructure, and c) one or more reporters bound to the DNA nanostructure. In other embodiments, the nanoprobe comprises a) a DNA nanostructure; b) at least three aptamers bound to the DNA nanostructure, and c) one or more reporters bound to the DNA nanostructure. In further embodiments, the nanoprobe comprises a) a DNA nanostructure; b) at least four aptamers bound to the DNA nanostructure, and c) one or more reporters bound to the DNA nanostructure. In some embodiments, each of the aptamers is configured to bind to the target protein. In some embodiments, the aptamers are identical to one another. In other embodiments, the aptamers are different from one another.

[0058] Without wishing to limit the present invention to any theory or mechanism it is believed that the spatial arrangement of the aptamers and reporters on the DNA nanostructure is important design to ensure proper probe function. In certain embodiments, the position of each aptamer is selected based on the structural pattern of the target protein. For example, glial fibrillary acidic protein (GFAP) assembles into microtubules in which each accessible target site is arranged in an ordered manner. Accordingly, the distance between adjacent aptamers on the nanostructure (e.g., a three-helix bundle (3HB) nanostructure) is preferably greater than the distance between neighboring GFAP target sites. If the spacing between aptamers is too small, binding of one aptamer to a GFAP molecule may prevent the remaining aptamers from reaching additional targets, thereby reducing detection efficiency. Similar spatial considerations apply to reporters, particularly when the reporters carry positive or negative charges that could influence nanostructure formation;September 23, 2025 sufficient separation is maintained to allow proper assembly and signal output. In certain embodiments, aptamers and reporters are positioned on opposite sides of the nanostructure to achieve the required distances. However, on alternative nanostructures — such as planar or other geometries — placement on the same side may be used provided that the necessary spacing between aptamers and reporters is maintained to preserve structural integrity and binding performance.

[0059] In some embodiments, for targets with an ordered structure, such as the cytoskeleton, multivalency can be designed to target multiple proteins, whereas for individual proteins, multivalency can be designed to bind multiple sites on a single protein.

[0060] In some embodiments, the nanoprobe comprises homo-aptamers, with three identical aptamers assembled on a DNA nanostructure such as a three-helix bundle (3HB) or a tetrahedral scaffold. In alternative embodiments, the aptamers need not be identical; for example, different aptamers may be designed to bind distinct target sites on a single protein, thereby increasing binding versatility and avidity. In certain embodiments, the manner of aptamer construction varies according to the nanostructure used. For example, a 3HB nanostructure may be designed with three loading sites to accommodate three aptamers, whereas a tetrahedral nanostructure, having four vertices, may provide four potential loading sites.

[0061] In some embodiments, the multivalent DNA nanoprobe is formed by self-assembly of three fluorophore-conjugated DNA strands onto the nanostructure scaffold (FIG. 5). The underlying DNA scaffold can be tuned to control the spatial arrangement of aptamers — including interaptamer distance, orientation, and number — to optimize binding kinetics and imaging performance

[0062] In some embodiments, the aptamers are comprised of DNA sequences. In other embodiments, the aptamers are comprised of RNA sequences. In some embodiments, the aptamers are linked to the DNA nanostructure. In some embodiments, the reporter is covalently conjugated to the DNA nanostructure.

[0063] In some embodiments, the nanoprobe comprises one or more reporters (e.g., two reporters) bound to the DNA nanostructure. In some embodiments, the nanoprobe comprises two or more reporters (e.g., three reporters) bound to the DNA nanostructure. In some embodiments, the nanoprobe comprises three or more reporters (e.g., four reporters) bound to the DNA nanostructure. In some embodiments, the nanoprobe comprises four or more reporters (e.g., five reporters) bound to the DNA nanostructure.September 23, 2025

[0064] In some embodiments, the reporter is a fluorescence reporter (e.g., fluorescence dye), fluorochrome, a luminescent reporter, or an enzyme reporter. Non-limiting examples of fluorescence dyes include but are not limited to Alexa Fluor™ fluorescent dyes (e.g., Alexa Fluor™ 488, Alexa Fluor™ 647, or Alexa Fluor™633), fluorescein isothiocyanate (FITC), tetramethyl rhodamine isothiocyanate (TRITC), Texas Red, cyanine dyes (Cy3, Cy5, Cy7), Atto dyes (Atto 488, Atto 550, etc.). Non-limiting examples of For luminescent reporters include but are not limited to luciferases (e.g., firefly luciferase, Renilla luciferase, NanoLuc®), 0- galactosidase (LacZ / X-gal system). In some embodiments, the chromogen is magenta. Additionally, non-limiting examples of enzyme reporters may include a chromogen such as 3,3'- diaminobenzidine tetrahydrochloride (DAB), horseradish peroxidase (HRP), glucose oxidase, or alkaline phosphatase (AP). In other embodiments, the reporter is quantum dots, upconversion nanoparticles, gold nanoparticles, or other nanomaterials. Other reporters may be used in accordance with the present invention.

[0065] In some embodiments, the target protein is a biomarker for cancer. For example, the biomarkers may comprise pl 6, EGFR, PD-L1, cyclin DI, or Ki-67. In other embodiments, isocitrate dehydrogenase (IDH) mutations may be used as a biomarker for grading glioblastoma (GBM), wherein primary GBM is classified as IDH-wild type, and tumors harboring an IDH mutation are classified as astrocytoma.

[0066] In some embodiments, the number of aptamers that may be bound to the DNA nanostructure is not limited to a fixed value but instead depends on the structural configuration of the nanostructure and the available loading sites. For example, a three-helix bundle (3HB) nanostructure may be designed with three attachment sites to accommodate three aptamers, whereas a tetrahedral nanostructure, which contains four vertices, can be configured to load four aptamers. The binding sequence of the aptamers generally does not require modification provided that it does not interfere with nanostructure formation. Similarly, the number of reporter molecules (e.g., fluorophores) may vary with the nanostructure design; for a 3HB nanostructure, experimental results indicate that up to four fluorophores can be incorporated without disrupting structural integrity, while higher numbers may adversely affect nanostructure stability.

[0067] The technology described herein may be readily generalized and applied to a wide range of cancer types, e.g., brain cancer, breast cancer, gynecological cancer (including cervical cancer, ovarian cancer, vulvar cancer, vaginal cancer and endometrial / uterine cancer) skin cancer, lymphoma, colorectal cancer, renal cancer, prostate cancer, lung cancer, liver cancer, pancreaticSeptember 23, 2025 cancer, bladder cancer, gastrointestinal cancer, or the like.

[0068] In some embodiments, the target protein is a biomarker for brain cancer (e.g., glioblastoma). For example, the target protein may glial fibrillary acidic protein (GFAP), CD20, or primary CNS lymphoma (PCNSL). In some embodiments, the target protein is a biomarker for breast cancer, e.g., biomarkers may include estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), Ki-67, CA 15-3, CA 27-29, and CA 125.

[0069] In other embodiments, the target protein is a biomarker for skin cancer (e.g., melanoma). In some embodiments, the target molecule is a biomarker skin cancer. Non-limiting examples of biomarker for that may be used to detect skin cancer, but are not limited to antigens (MAAs), microRNAs (miRNAs), SIOOB, CRP, circulating tumor cells (CTCs), Human Melanoma Black- 45 (HMB-45), Melan-A, tyrosinase, microphthalmia transcription factor, SI 00, PRAME, lymphovascular invasion (LVI) / endothelial markers (e.g., D2-40 or CD31), Serum lactate dehydrogenase (LDH), S100 0, Melanoma inhibitory activity (MIA), and Vascular endothelial growth factor (VEGF). Other prognostic or predictive markers such as lactate dehydrogenase (LDH) levels may also be used.

[0070] In some embodiments, the target protein is a biomarker for lymphoma. For example, biomarkers for lymphoma may include plasma cell markers (e.g., CD138 and CD38), B-cell markers (e.g., CD20, CD79a, CD3, CD20, CD79a, CD56, Bcl-2, Cyclin D-l, Ki-67, CD30, CD23, CD19, CD22, and Pax-5;), T-cell markers (e.g., CD2, CD3, CD4, CD5, CD7, and CD8), Cytotoxic markers (e.g., TIA-1, granzyme B, and perforin), germinal center markers (e.g., CD 10, Bcl-6, and LMO2), and kappa and lambda light chains. For example, diagnosing lymphoma may include analyzing kappa and lambda light chain (e.g., target proteins) expression in a biological sample from a subject. The method may involve determining the ratio of kappa to lambda light chains, wherein a balanced ratio indicates a normal immune system, while a predominance of either kappa or lambda light chains signifies a monoclonal population, which is characteristic of lymphoma. This method can be implemented using immunoassays, flow cytometry, or sequencing technologies to detect and quantify light chain levels accurately.

[0071] In some embodiments, the target protein is a biomarker for colorectal cancer, e.g., biomarkers may include carcinoembryonic antigen (CEA), CA 19-9 proteins, HER2, or tissue inhibitor of metalloproteinase (TIMP)-l. In other embodiments, target genes may be used as a biomarker for colorectal cancer, including but not limited to methylated genes (e.g., MLH1, VIM, and SEPT9). Colorectal cancer may also be diagnosed using microsatellite instability (MSI)),September 23, 2025Mismatch repair (MMR) status, or RAS / RAF mutation status.

[0072] In some embodiments, the target protein is a biomarker for renal cancer. For example, biomarkers for renal cancer may include but are not limited to cell-free DNA (cfDNA), Carbonic anhydrase IX (CA9), Aquaporin-1 (AQP-1), perilipin-2 (PLIN2), Glycosaminoglycans (GAGs), Heat shock protein 27 (HSP27), MicroRNAs (miRNAs), Circular RNA (circRNA), Epithelial membrane proteins CK and EPCAM, Mesenchymal markers VIM, TWIST, and N-cadherin, Carbonic anhydrase 12 (CA12), CXCL16, ADAM10, B7-H1, Ki-67, Survivin, and P53.

[0073] In some embodiments, the target protein is a biomarker for prostate cancer. For example, biomarkers for prostate cancer may include but are not limited to Appll, Sortilin, Syndecan-1, Prostatic Acid Phosphatase (PAP), Prostate-specific antigen (PSA), HMW-CK, p63, PD-1 / PD-L1, CD276, CD73. Other biomarkers may be used to detect prostate cancer such as Gleason score; Epigenetic changes; Androgen receptor splice variant 7 (AR-V7); Tumor-associated macrophages (TAMs); Cytotoxic CD8 tumor-infiltrating lymphocytes (TILs); Regulatory T cells (Tregs); or Mast cell testing.

[0074] Biomarkers for lung cancer may include markers associated with various histological types, such as adenocarcinoma, adenosquamous carcinoma, squamous cell carcinoma, large cell carcinoma, and large cell neuroendocrine carcinoma. Methods may involve testing for abnormalities in DNA or protein levels within a tumor, as well as cell-free DNA or circulating tumor DNA (cfDNA or ctDNA) analysis. Additional approaches may include serum tumor marker testing, such as carcinoembryonic antigen (CEA), cytokeratin-19 fragment (CYFRA), and squamous cell carcinoma-associated antigen (SCC), as well as the detection of specific microRNAs (miRNAs).

[0075] Liver cancer biomarkers may include blood biomarkers such as proteins, cytokines, enzymes, and gene transcripts, including Alpha-fetoprotein (AFP). Tissue markers may also be utilized, such as glypican-3, heat shock protein 70, glutamine synthetase (GS), arginase- 1, polyclonal carcinoembryonic antigen, CD 10, and bile salt export pump. Additionally, microRNAs (miRNAs), small non-coding RNAs, have been investigated as diagnostic biomarkers for liver cancer, with certain miRNAs shown to accurately predict poor prognosis in hepatocellular carcinoma (HCC).

[0076] Pancreatic cancer biomarkers may include immunohistochemical (IHC) markers such as pentraxin-3, ENO1, REG4, POSTN, CA125, CA242, Galectin-1, Galectin-9, Maspin, pVHL,September 23, 2025MUC1, MUC5AC, THBS2, LTBP2, CPA4, IMP3, CD13, Dkkl, KOC, SIOOP, Mesothelin, and PAM4. Serum markers such as CA19-9, CA125, and CEA may also be utilized, along with proteins including biglycan, pigment epithelium-derived factor, thrombospondin-2, and transforming growth factor 0 induced protein Ig-h3 precursor. Other methods may involve the detection of exosomes, microRNAs (miRNAs), and SMAD4.

[0077] Ovarian cancer biomarkers include CA-125, sialyl-Tn antigen (STn), and tumor suppressor genes such as BRCA1 and BRCA2. For bladder cancer, biomarkers may include proteins like alfa- defensin, apolipoprotein A-l (APOA1), and alfa 1-antitrypsin (A1AT), as well as Bcl-2 and somatic mutations in genes such as FGFR3, PIK3CA, KDM6A, and TP53.

[0078] In gastrointestinal cancers, relevant biomarkers may include carcinoembryonic antigen (CEA), CA19-9, HER2 overexpression, HER2 / neu amplification, MSI-H, and PD-L1+, along with CLDN18.2, HIF-3a, p-PKC a / 0 II, miRNAs, and Epstein-Barr virus (EBV). These biomarkers play a crucial role in diagnosis, prognosis, and treatment decision-making across various cancer types.

[0079] In some embodiments, aforementioned nanoprobes may be paired with an electric field (ELF)-driven staining method in combination with a denoising Al model, reducing staining time to within approximately 5 minutes. Additionally, the methods described herein may be integrated with interpretable deep learning tools to facilitate rapid intraoperative decision-making for both pre-resection tumor type diagnosis and the determination of intraoperative tumor margins.

[0080] The present invention may also feature a method of diagnosing cancer in a subject in need thereof. In some embodiments, the method comprises obtaining or having obtained a tissue sample from the subject, staining the tissue sample using the nanoprobe as described herein for a period of time, and detecting a level of staining in the tissue sample. In some embodiments, the subject is diagnosed with cancer if the staining is above a predetermined threshold. In some embodiments, the stained tissue is analyzed using image analysis or Al-assisted image analysis (e.g., imagine based deep learning model); wherein the image analysis detects the level of staining in the tissue sample; wherein the subject is diagnosed with cancer if the staining is above a predetermined threshold as determined by the image analysis.

[0081] In some embodiments, the present disclosure may further comprise a method of diagnosing cancer in a subject in need thereof. In some embodiments, the method comprises a) obtaining or having obtained a tissue sample from the subject, b) staining the tissue sample using the nanoprobeSeptember 23, 2025 as described herein, for a period of time; and c) analyzing the stained tissue sample using an imaging-based deep learning (DL) model; wherein the imaging-based DL model detects a level of staining in the tissue sample. In some embodiments, the subject is diagnosed with cancer if the staining is above a predetermined threshold as determined by the imaging-based DL model.

[0082] In some embodiments, the method further comprises applying an electric field to the tissue sample to accelerate staining. In some embodiments, staining is performed by an electric field (ELF)-driven reaction (e.g., aptamer-target recognition), and the direction of the current is periodically switched such that unbound probes are moved within the tissue to increase the likelihood of target binding, while specifically bound probes remain localized, thereby enhancing signal intensity and reducing background noise. In some embodiments, the current is switched twice. In other embodiments, the current is switched three times. In some embodiments, the current is switched four times. In some embodiments, the current is switched five or more times. However, the present invention is not limited to any particular number of current-switching events.

[0083] Referring to FIG. 13 A, as a non-limiting example, in some embodiments, the DNA nanoprobe is prepared in a buffer (e.g., PBS) at the desired concentration (e.g., 0.1 pM) and 100 pL is applied to the bottom slide (e.g., an indium tin oxide (ITO) slide). The top slide, containing the tissue section, is then placed over the bottom slide. The ELF device is activated using the programmed code to run the staining protocol. After completion, the tissue is imaged using standard confocal microscopy procedures.

[0084] The number of current-switching cycles and the duration of each cycle may be adjusted based on various factors, including the charge characteristics of the nanoprobe, the binding affinity of the aptamer-target pair, and the thickness of the tissue sample. In certain embodiments, the forward voltage is greater than the reverse voltage because experimental results indicate that a relatively short reverse cycle is sufficient to enhance signal and remove nonspecifically bound probes, whereas prolonged reverse application may undesirably displace specifically bound probes.

[0085] In some embodiments, the samples is stain for a period of twenty minutes or less (e.g., the staining is completed in twenty minutes of less). In some embodiments, the samples is stain for a period of ten minutes or less (e.g., the staining is completed in ten minutes of less). In some embodiments, the samples is stain for a period of five minutes or less (e.g., the staining is completed in five minutes of less).September 23, 2025

[0086] In certain embodiments, the presence of staining above a predetermined threshold is indicative of tumor type or tumor margin status, thereby guiding surgical decisions. For example, in brain cancer, the tumor type may determine whether surgical resection is appropriate, while in breast cancer, detection of tumor cells in a lymph node may indicate the need for lymph node removal. Similarly, determination of residual tumor cells at a surgical margin may indicate the need for additional excision.

[0087] In some embodiments, the tissue sample is from a solid tumor. In some embodiments, the solid tumor is within the brain of the subject. In some embodiments, the cancer is Glioblastoma multiforme (GBM).

[0088] In some embodiments, a kit is featured comprising a multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target protein as described herein, one or more of stains, reagents, or combinations thereof, and instructions comprising combining the nucleotide composition and the one or more of stains, reagents, or combinations thereof, and staining a tissue sample. The nanoprobe may comprise a DNA nanostructure and at least two aptamers bound to the DNA nanostructure, and one or more reporters bound to the DNA nanostructure. Each of the aptamers is configured to bind to the target protein. In some embodiments, the kit further comprise a microfluidic chip for staining. In some embodiments, the microfluidic chip is a disposable cartridge for intraoperative or point-of-care use. In further embodiments, the kit may further comprise instructions for intraoperative staining. Alternatively, or in addition to, in some embodiments, the kit may further comprise instructions for point-of-care staining.

[0089] In some embodiments, the technology described herein is used for drug development for oncology including use as a companion diagnostic or biomarker diagnostic. It may aid in patient stratification, addressing heterogeneity, and supporting clinical development of molecularly targeted drugs. It may be used to detect and monitor tumor-targeted drug delivery processes, tumor accumulation of medicines, and accurately assess and predict medicine target-site localization. It may be used with nanomedicines.

[0090] In some embodiments, the present invention features a system for using artificial intelligence (Al) to generate analyze the staining of tissue samples from a subject. In some embodiments, the system may comprise a processor configured to execute computer-readable instructions, and a memory component communicatively coupled to the processor. The memory component may comprise an Al model comprising one or more neural networks, trained with a training data set comprising imagining-based training data. Training the Al model may compriseSeptember 23, 2025 feeding the training data set as input into the Al model. The Al model may be trained to quantify the amount of staining in a given tissue sample. The memory component may further comprise computer-readable instructions. The computer-readable instructions may comprise inputting imaging-based data into the Al model.

[0091] The Al model may be stored, trained, and / or executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. The Al model may be stored in the form of program code, as described above. The one or more neural networks of the Al model, in some embodiments, may comprise a perceptron neural network, a feed-forward neural network, a multilayer perceptron neural network, a radial basis functional neural network, a recurrent neural network, a long short-term memory neural network, a sequence-to-sequence neural network model, a modular neural network, a graph-based convolutional neural network, an instance-based model, a feature attribution model, or the like.

[0092] In a non-limiting example, the Al model of the presently claimed invention may comprise a perceptron neural network, configured to accept the imaging-based data as input, execute one or more functions on the input, multiply the output of the one or more functions by a plurality of weights, and generating a final output. In another non-limiting example, the Al model of the presently claimed invention may comprise a multilayer perceptron neural network comprising a plurality of layers, each layer configured to execute the process of a single perceptron network, each layer linked to an input and output feed such that the output of one layer is the input of a subsequent layer.

[0093] EXAMPLE

[0094] The following is a non-limiting example of the present disclosure. It is to be understood that said example is not intended to limit the present disclosure in any way. Equivalents or substitutes are within the scope of the present disclosure.

[0095] Fast and precise intraoperative histopathologic diagnosis is essential for optimal surgical management of tumors. Surgery is a key pillar in the treatment of solid tumors. However, the success of surgical intervention often critically depends on precise intraoperative decision-making guided by rapid imaging and diagnosis during the surgery. Intraoperative decision-making plays a pivotal role in a range of tumor surgeries within oncology. For example, in breast, cervical, endometrial, vulvar cancers, and melanoma, sentinel lymph node (SLN) mapping has become central to such decision-making to detect the metastasis of the cancer. SLNSeptember 23, 2025 is typically too small to be detected on CT, MRI, or ultrasound, making its biopsy challenging preoperatively due to the need for surgical exposure to identify the true SLN and follow lymphatic channels. In early-stage cancer, SLN biopsy can safely replace full lymphadenectomy, significantly reducing surgical morbidity without compromising oncologic safety. Intraoperatively, identifying whether SLNs contain neoplastic cells informs whether to proceed with extensive node dissection or adjuvant therapy. Similarly, intraoperative diagnosis during ovarian tumor surgeries helps rapidly distinguish between benign, borderline, and malignant lesions, enabling decisions about staging, debulking, or conservative treatment in real time. Beyond lymph node metastasis detection and tumor malignancy determination, the degree of cytoreduction is a critical factor influencing the success of surgery. Incomplete resections increase the risk of recurrence and represent one of the strongest unmet needs for oncology patients. Some rapid tumor margin determination methods have been developed, such as rapid flow cytometry to detect cancer at the margin after tumor resection in melanoma and brain tumors. However, this approach is time-consuming, highly dependent on differentiating cell cycle phases, and lacks the ability to detect specific biomarkers, limiting their accuracy and clinical application. Based on the clinical needs mentioned above, a specific, rapid, and biomarker-sensitive intraoperative diagnostic methodology is urgently needed to improve surgical precision and patient outcomes.

[0096] Rapid staining methods need to be developed to overcome the limitations of traditional histologic imaging processes. Intraoperative decision-making and the setting of surgical goals are profoundly influenced by tumor pathology. An ideal method for intraoperative histology delivers near real-time diagnostic images with a streamlined workflow. This workflow can be divided into two key components: sample preparation and sample imaging. Traditional histologic methods, including formalin-fixed paraffin-embedded (FFPE) tissue sections and frozen tissue sections, are time-consuming, often requiring 20 minutes to several hours to preserve tissue morphology and prepare sufficiently thin sections. Such delays hinder timely intraoperative decision-making.

[0097] Rapid sample preparation techniques have been developed to address these limitations. These include direct slide contact methods — such as touch imprinting, scraping, or crushing tissues to create smears — for rapid staining with hematoxylin and eosin (H&E), toluidine blue, or suspending dissociated cells in buffer for flow cytometry. While such approaches can reduce preparation time to under 10 minutes, nonspecific interactions between dyes and cellular components or nucleic acids limit both accuracy and diagnostic utility. Targeted staining approaches, such as immunohistochemistry (IHC), traditionally require hours to days due toSeptember 23, 2025 antibody binding and multiple washing steps. Accelerated IHC protocols achieving 20-30 minutes often compromise specificity and sensitivity. Raman scattering (RS) microscopy offers rapid, high- resolution imaging of unstained tissues, and machine learning can automate histopathologic classification. However, RS specificity may be affected by overlapping spectra and tumor heterogeneity, while traditional statistical machine learning requires separate feature engineering and predictive modeling, complicating interpretation.

[0098] The present invention addresses these challenges by providing compositions and methods that combine novel electric field-driven diagnostic probe designs with generative Al-based imaging strategies. This approach decreases diagnosis time while increasing specificity, supporting rapid decision-making by pathologists and surgeons. In certain embodiments, the biopsy tissue staining protocol is completed in under five minutes, reducing the total turnaround time from tissue sampling to reporting results to the surgeon from over 30 minutes to approximately 15 minutes, and providing critical information in near real-time to guide intraoperative neurosurgical decisions. This is accomplished through two key elements (FIG. 2): (1) a DNA probe configured to move faster than molecular diffusion in free solution, allowing rapid binding to target proteins and reducing staining time from approximately 10 minutes to about 1 minute; and (2) computational tools empowered by artificial intelligence to denoise nonspecific staining without blocking the tissue, reducing the staining process from approximately 11 minutes to about 1 minute while supporting the pathologist in making diagnostic decisions. The diagnostic probes are also synthetically accessible and cost-effective, enabling practical use in clinical settings. Brain tumor intraoperative diagnosis was selected as an initial application to develop and validate the technology; however, without wishing to be bound by theory, the compositions and methods described herein are believed to be generalizable to a wide range of tumors requiring fast and precise intraoperative diagnosis.

[0099] Brain tumor intraoperative diagnosis. Currently, treatments for brain tumor patients are guided by histopathologic diagnoses derived from biopsies. Surgical intervention is required to remove the lesion and obtain tissue for histopathological evaluation. Most often, brain tumor biopsies are obtained during craniotomy to access the brain tissue for open surgical resection. With the exception of surgery for recurrent tumors, the specific histopathologic tumor diagnosis is typically unknown prior to the surgery. Biopsies obtained during surgery and intraoperative diagnosis play a crucial role in guiding real-time surgical care for brain tumors patients, such as: dictating aggressive surgical resection, halting surgery, or determining if a patient is eligible for a specific intraoperative intervention, such as placement of treatment catheters or intracavitarySeptember 23, 2025 chemotherapy or brachytherapy. In a minority of cases, stereotactic biopsies (STB) are performed to obtain intraoperative tissue samples in a minimally invasive fashion when open surgery is not recommended. However, intraoperative histopathologic sampling error from STB can be 10%, and worsens with smaller and deeper brain lesions. Significantly, primary central nervous system lymphoma (PCNSL) is considered a non-operative tumor and surgery is often limited to biopsy alone. Glioblastoma (GBM), the most common malignant primary brain tumor, is often surgically treated with maximal safe resection. Failure to specifically differentiate these tumors intraoperatively can lead to early termination of a surgery, resection of a tumor best treated without surgery, or exclusion of beneficial intraoperative techniques. Additionally, the GBM recurrence rate is positively associated with traditional IHC such as antibody-based staining, which is a primary mode of diagnosis for brain tumors by identifying tumors by identifying tumor-specific markers. However, it takes hours to confer an accurate diagnosis, much too long a timeframe for intraoperative feedback. Intraoperative fluorescent tissue staining methods have shown a large variation of sensitivity ranging from 55% to 95% and specificity from 67% to 78% in the detection of brain tumor tissue during surgery. Furthermore, tumor residue after resection significantly influences GBM recurrence rates and overall survival (OS). Clinical studies have shown that grosstotal resection (GTR), which removes 100% of the contrast-enhanced tumor as seen on postoperative imaging, yields a considerably higher OS (1-, 3-, and 5-year OS rates of 91.3%, 38.6%, and 25.3%, respectively) compared to subtotal resection (STR), (1-, 3-, and 5-year OS rates of 68.0%, 17.6%, and 7.2%, respectively). Supratotal resection (SupTR), which involves removing the entire tumor plus additional surrounding brain tissue, has demonstrated a significantly lower recurrence rate — about 17% lower than GTR, reducing from 80.5% to 63.4%.

[0100] Rapid intraoperative staining as described herein can assist surgeons in more accurately removing tumor residues with tissue surrounding, thereby reducing the damage associated with SupTR by providing targeted staining guidance. Consequently, a refined, rapid intraoperative histopathology approach for generating specific brain tumor diagnoses is crucial for improving surgical decision-making, expediting patient care, decreasing surgical time, and minimizing the need for additional brain surgeries. There is a significant unmet need to develop methods for diagnosis of brain tumors with specificity that are fast and precise enough to guide intraoperative therapeutic decision making. Development of such a method could have an impact on improving the standard of care for brain tumor patients by hastening diagnosis and the selection of proper treatment.

[0101] Multivalent Aptamer-Based DNA Nanoprobe: Aptamers offer several advantagesSeptember 23, 2025 compared to antibodies: (1) they are 10-20 times smaller than antibodies and can be chemically modified in a defined and precise manner; (2) they are easily stored and can be reversibly heat- denatured, with high batch-to-batch reproducibility; and (3) they exhibit superior and faster tissue penetration. For example, a fluorescent aptamer probe targeting the extracellular protein CD20 has demonstrated IHC-like diagnostic capability for primary central nervous system lymphoma (PCNSL) using xenograft and human tissue sections in approximately 11 minutes, including staining and imaging time after protocol optimization. Additionally, incorporation of a nanostructure increases valency, enhancing binding affinity and further reducing the time required for staining.

[0102] Glial fibrillary acidic protein (GFAP), as an intermediate filament-III protein found in astrocytes, is overexpressed in GBM, and has been used as clinical biomarker for diagnosing GBM. Aptamers against GFAP with nanomolar affinities were selected, including an aptamer sequence capable of distinguishing U251 (a GFAP-positive GBM cell line) from Ramos (a GFAP-negative lymphoma cell line).

[0103] CD20 is a B cell-restricted tetra-spanning protein organized in the plasma membrane as multimeric molecular complexes. Both the homo-oligomeric feature of CD20 and the filamentous structure of GFAP provide repetitive binding domains suitable for engineering multivalent aptamers with improved binding avidity and kinetics. An aptamer-based molecular probe for detecting CD20-positive B cells has been previously developed; this probe binds to the heavy mu chain of membrane-bound immunoglobulin in CD20-positive Burkitt’s lymphoma cells and was shown to specifically differentiate B-cell lymphoma from astrocytoma cells in vitro and ex vivo using a rapid staining protocol. In contrast to the aforementioned probe, the present invention features a modified version of a CD20 aptamer that is elongated to include a complementary sequence enabling incorporation into a DNA nanostructure scaffold. This structural modification facilitates the formation of a multivalent, DNA nanostructure-scaffolded aptamer nanoprobe comprising multiple fluorophores, thereby enhancing binding avidity, kinetics, and staining speed beyond what was achieved with the previously reported monovalent aptamer probe.

[0104] In some embodiments, the target protein does not need to contain repetitive binding domains to be useful with the multivalent aptamer-based probe described herein. For targets with an ordered structure, such as the cytoskeleton, multivalency can be designed to target multiple proteins, whereas for individual proteins, multivalency can be designed to bind multiple sites on a single protein.September 23, 2025

[0105] In certain embodiments, a DNA nanostructure-scaffolded multivalent aptamer-based nanoprobe comprising multiple fluorophores attached to the scaffold is provided to improve the speed, accuracy, and efficiency of tissue staining during clinical intraoperative histopathologic diagnosis. In such embodiments, the binding region of the aptamer is carefully designed to provide sufficient affinity for the target while avoiding disruption of nanostructure formation. Additionally, fluorophores are attached to the scaffold rather than directly to the aptamer strand, thereby preserving aptamer-target recognition and functional performance. Without wishing to limit the present invention to any theories or mechanisms, it is believed that the multivalent aptamer-based nanoprobe described herein may shorten the staining time, due to 1) increased avidity; 2) reduced dissociation rate; 3) increased local concentration; 4) flexibility to adopt optimal orientation and conformations for binding. In some embodiments, the nanoprobe provides rapid intracellular GFAP staining, enhancing the visualization of glial brain tumors and histopathological analyses, and enabling evaluation of intraoperative tissue samples within approximately 25 minutes (FIG. 1, left). In some embodiments, the compositions and methods described herein provide an intracellular staining technique that enhances the visualization of glial brain tumors and histopathological analyses.

[0106] Referring to FIG. 3, the modified aptamer-based approach enables rapid and specific diagnosis of primary central nervous system lymphoma (PCNSL) from xenograft biopsies within approximately 11 minutes — a turnaround significantly faster than traditional frozen-section diagnosis and conventional aptamer staining protocols, which typically require 30 minutes to 1 hour. In this embodiment, the aptamer sequence was elongated with a complementary region to facilitate incorporation into the nanostructure scaffold, improving structural stability and enabling multivalent binding. As shown in FIG. 3, the optimized probe labeled a substantially higher percentage of RFP-labeled lymphoma cells (positive control) compared to non-RFP-labeled cells (negative control), and aptamer staining strongly correlated with RFP expression. These results demonstrate that the modified PCNSL aptamer probe provides highly specific diagnostic information in a fraction of the time required for conventional immunohistochemistry (IHC).

[0107] Referring to FIG. 4, clinical staining protocols for ex vivo and frozen section diagnostics were refined to evaluate the performance of a monovalent aptamer probe in comparison with conventional immunohistochemistry (IHC). As shown in FIG. 4, hematoxylin and eosin (H&E) staining identifies regions of hypercellular tumor, while CD20 IHC demonstrates strong ring-like staining in primary central nervous system lymphoma (PCNSL) biopsies but not in glioblastoma (GBM) samples. The aptamer probe produces a similar ring-like staining pattern in RFP-expressingSeptember 23, 2025PCNSL cells and shows an absence of staining in RFP -expressing GBM biopsies, thereby confirming specificity. These results illustrate that the aptamer probe can replicate the diagnostic morphology of standard CD20 IHC while enabling rapid, fluorescence-based detection of PCNSL.

[0108] Multivalent Aptamer-Based DNA Nanoprobe for CD20 cell membrane detection: In certain embodiments, a DNA nanostructure scaffolded multivalent aptamer-based nanoprobe comprising multiple fluorophores is optimized for CD20 cell-membrane detection to improve the speed, sensitivity, and specificity of imaging, thereby enabling faster and more accurate histopathological diagnosis of primary central nervous system lymphoma (PCNSL). In such embodiments, the DNA nanoprobe includes a multivalent PCNSL-specific aptamer and multiple fluorophores arranged on a DNA nanostructure scaffold to facilitate rapid and high-contrast staining.

[0109] In some embodiments, the optimization process focuses on the binding sequence of the aptamer, as the length of this sequence determine the distance between each aptamer and the flexibility of this sequence (e.g., that connects the aptamer to the nanostructure. These parameters influence aptamer recognition and overall binding performance. In further embodiments, the number of fluorophores incorporated into the nanoprobe is adjusted, as an excessive number of fluorophores may interfere with nanostructure formation or reduce aptamer target recognition.

[0110] Characterization of DNA nanoprobe sensitivity and specificity in vitro. The stability of the multivalent fluorescent aptamer DNA nanoprobe will be optimized by assessing performance under simulated physiological conditions, including variations in temperature, pH, ionic strength, and enzymatic activity. Nanoprobes containing multiple Alexa Fluor™ 633 fluorophores will be incubated for one hour, and fluorescence intensity, degradation, and aggregation will be evaluated using agarose gel electrophoresis and fluorescence monitoring. Sensitivity will be determined by comparing the detection limits of the multivalent nanoprobe to those of a monovalent PCNSL- specific fluorescent aptamer and an antibody probe. Ramos, Raji, Jeko, and Daudi cell lines will be stained with serial probe concentrations (1 nM-1 pM), and fluorescence intensity will be quantified by microscopy and flow cytometry to establish the minimum effective concentration. Specificity will be confirmed by staining CD20-positive (Ramos, Raji) and CD20-negative (U251, A549) cell lines and comparing fluorescence signals. Competitive inhibition assays will further verify target-specific binding by pre-treating CD20-positive cells with anti-CD20 antibodies and assessing any reduction in nanoprobe staining.

[0111] Optimization of multivalent aptamer nanoprobe for rapid clinical staining. To enableSeptember 23, 2025 clinical application, the concentration of the multivalent fluorophore-DNA nanoprobe will be optimized using mouse xenograft tissues derived from Ramos cells. Serial concentrations will be applied to tissue sections to identify the lowest concentration that achieves strong, specific staining. Time-course studies will determine the minimum incubation period (e.g., 5, 10, 15, and 20 minutes) required to maintain staining intensity and specificity. Staining stability and reproducibility will be assessed by repeating the optimized protocol on replicate tissue sections. For clinical evaluation, a pilot study will enroll up to ten IHC-confirmed PCNSL cases; randomized, de-identified aptamer images including positive and negative controls will be generated and independently reviewed by a clinical pathologist to confirm diagnostic performance.

[0112] Specificity enhancement in tissue sections. The DNA nanoprobe will be tested on tissue sections with and without CD20 expression as positive and negative controls to confirm target specificity. Cross-reactivity and compatibility with different tissue preparation methods will be evaluated by applying the staining protocol to both frozen and formalin-fixed tissues.

[0113] Without wishing to limit the present invention to any theory or mechanism it is believed that the multivalent aptamer and multiple fluorescent DNA nanoprobes will have a higher sensitivity compared to the monovalent aptamer and antibody probe due to the multiple binding sites and fluorophores, leading to an amplified signal. Additionally, the nanoprobes described herein provide minimal non-specific staining, ensuring clear contrast between the target cells and the background. Minimum effective staining time and optimal blocking condition will be established to provide sensitive and specific staining

[0114] Multivalent Aptamer-Based DNA Nanoprobe for GFAP detection: Aptamers against GBM specific intracellular biomarker GFAP with nanomolar affinities were identified and selected, including an aptamer sequence capable of distinguishing U251 (a GFAP -positive GBM cell line) from Ramos (a GFAP-negative lymphoma cell line). In certain embodiments, a DNA nanostructure-scaffolded multivalent aptamer-based nanoprobe comprising multiple fluorophores (FIG. 6) for specific intracellular staining of the filamentous GFAP in glioblastoma cell line U251 is provided. The fixation and staining time combined for the U251 cells is currently about 25 minutes, which is faster than the traditional intracellular staining protocol for GFAP. The optimized staining method described herein will allow the pathologist to rapidly and precisely identify GFAP- positive cells within the complex cellular environment of the brain, such as astrocytes and glioma cells which may then be morphologically differentiated.

[0115] Candidate aptamers of GFAP. To differentiate more effectively between PCNSL andSeptember 23, 2025GBM, aptamer selection was focused on targeting the GFAP protein. Multiple GFAP aptamer sequences were identified that exhibit nanomolar range affinities. Binding affinity of the aptamers were quantitatively assessed using surface plasmon resonance (SPR). As illustrated in FIG. 7, the results of SPR indicated that the equilibrium dissociation constant (KD) is 59.2 nM for Aptl, 385 nM for Apt2, 58.1 nM for Apt3. These three aptamers demonstrated efficient staining in U251 glioblastoma cell line and effectively differentiation from the Ramos cells (lymphoma) (FIG. 7).

[0116] Table 1 provides a non-limiting example of a 3HB nanostructure comprising three aptamers attached thereto. The sequences of the six DNA strands forming the 3HB and the aptamer are shown, with the complementary regions that mediate aptamer binding to the 3HB indicated in bold. Each aptamer is linked to the 3HB through the same complementary binding arrangement.Table 1:

[0117] Validation of aptamer-fluorophore probe for enhanced intracellular staining with high specificity. To determine the specificity of the three aptamers, the aptamers are labeled withSeptember 23, 2025 a red fluorophore, Alexa Fluor™ 647, and are used as probes to detect the localization and distribution of the specific aptamers by fluorescence microscopy. As shown in FIG. 8A, the U251 cells were fixed and stained by three aptamer-probe (Aptl, Apt2, Apt3) individually, the red fluorescence of Apt3 showed intense and fibrous staining, which is consistent with GFAP, given its role as an intermediate filament protein. Aptl and Apt2 were more concentrated in the nucleus with less cytoplasmic staining. Since GFAP is not a nuclear protein, this pattern indicated that Aptl and Apt2 were unlikely GFAP-specific. The fibrous nature of the staining of Apt3 suggests that aptamer3 has a higher specificity for GFAP, binding to its filamentous structures throughout the cellular cytoplasm. To further evaluate the specificity of Apt3-probe, three cell lines including U251, SKBR3 (a breast cancer cell), and HepG2 (hepatoblastoma cells), were stained with Apt3- probe, which is conjugated to the Alexa Fluor™ 647. The staining pattern of U251 cells shows filamentous structure, compared with HepG2 and SKBR3 staining, indicating the good binding specificity of Apt3-probe to GFAP (FIG. 8B). There is less fluorescence signal in the remaining cell lines, indicating minimal aptamer binding.

[0118] Development of multivalent aptamer and multiple fluorophore DNA nanoprobe for enhanced intracellular rapid staining. To improve both the binding avidity and fluorescent intensity, a compact DNA nanostructure was designed and engineered as depicted in FIG. 9A. This DNA nanostructure is capable of displaying three aptamers and accommodating multiple fluorescents on a single molecular-DNA nanoprobe, enabling rapid and specific intracellular staining. As shown in FIG. 9B, it indicated that an assembled DNA nanoprobe shows a much stronger fluorescence signal and filament shape compared to those with the mono-aptamer probe. The enhanced signal demonstrates that the multivalent DNA nanoprobe may have a higher binding affinity and is able to bind to multiple GFAP targets simultaneously, resulting in a more intense signal. To quantify analysis of the binding efficiency using integrate density by ImageJ software, the multivalent DNA nanoprobe is significantly higher than the mono-aptamer, which quantitatively supports the qualitative observations from the fluorescence images, indicating that the multivalent design enhances the sensitivity and density of binding compared to the monovalent aptamer. This approach can be useful for increasing the detection capabilities of cellular imaging and potentially improve the efficacy of diagnostics in biomedical applications.

[0119] Optimization of multiple fluorophores for improving signal intensity. Utilizing multiple fluorophores excited by a single light source can yield a stronger and more specific signal. To augment the intensity of the DNA nanoprobe for improving rapid staining cells, different quantities of Alex647 fluorophores were assessed, ranging from one to four onto a singular DNASeptember 23, 2025 nanoprobe. U251 cells were stained with these varied DNA nanoprobes for 10 minutes and obtained images with confocal. The results revealed that increasing of Alexa Fluor™647 fluorophores on the DNA nanoprobe substantially increased the fluorescence signal, thus enhancing the visibility and detectability of the DNA nanoprobe when bound to the GFAP targets (FIG. 10).

[0120] Building on the fluorophore optimization, multivalent nanoprobes displaying homologous GFAP-specific aptamers were designed and self-assembled with multiple fluorescent molecules using Tiamat software. DNA nanostructures (DX tile, 3HB, tetrahedron TH) were modeled with oxDNA to guide construction. Binding avidity was quantified across five probe concentrations (1 pM, 100 nM, 10 nM, 1 nM, 100 pM) using gel-shift assays, filter binding assays, and surface plasmon resonance (SPR) to determine the apparent dissociation constant (KD). The optimal Alexa Fluor™ 633-to-DNA ratio was identified by preparing probes with varying fluorophore densities (1 : 1 to 1 :5) and measuring fluorescence intensity to avoid quenching. Spatial distribution of Alexa Fluor™633 on the DNA nanostructure was further evaluated by fluorescence spectroscopy and Forster resonance energy transfer (FRET) to minimize quenching and maximize signal. Nanoprobe stability was assessed by incubating Alexa Fluor™ 633-labeled constructs under physiological conditions and monitoring fluorescence, degradation, and aggregation by agarose gel electrophoresis. The optimized nanoprobe was finally tested for in-vitro efficacy by staining GFAP-positive U251 and U373 cells and GFAP-negative Ramos and Miacapa If cells, followed by confocal imaging to confirm selective fluorescence labeling.

[0121] Validate the specificity and efficacy of DNA nanoprobe. The specificity and efficacy of the DNA nanoprobe were validated through a series of control and comparative assays. Negative control validation was performed using tissue sections prepared from GFAP-negative rodent xenograft tumors, which were stained with the DNA nanoprobe and imaged by fluorescence microscopy. Absent or significantly reduced fluorescence compared to GFAP-positive tissue indicated high specificity. Nanoprobe staining was further compared with traditional antibody staining by dividing tissue samples into two groups and quantifying fluorescence using standard stereology techniques. A competitive inhibition assay was also conducted by pre-incubating tissue sections with a GFAP-specific antibody prior to nanoprobe staining, followed by fluorescence quantification. Staining was evaluated based on intensity, uniformity, and background signal, as well as pattern characteristics including the filamentous distribution of GFAP, cytoplasmic and process localization, and overall structural clarity. Side-by-side imaging enabled direct comparison of staining pattern, intensity, and clarity between the DNA nanoprobe and antibody methods.September 23, 2025

[0122] Comparative analysis between DNA nanoprobe vs traditional antibody for intracellular staining. To assess the efficiency of intracellular staining, a time- dependent staining assay with side-by-side comparison of cellular imaging over time was implemented using both the DNA nanoprobe and antibody staining approaches under the same condition. As shown in FIG. 11, the top row of images shows the intracellular stained with multivalent aptamers and multiple fluorophore DNA nanoprobe, the bottom as antibody stained, as indicated by the schematic on the left, with blue as binding domain, red as fluorescent Alexa Fluor™ 647, green as FAM, respectively. Following the incubation of the DNA nanoprobe and antibody, cells were stained at intervals of 5 minutes, 30 minutes, 1 hour, and 6 hours. As shown FIG. 11A-11B, images were captured by confocal microscopy. Over time, there was a significant increase in red fluorescence intensity of the DNA nanoprobe. There is a significant fluorescence visible at 5-30 minutes for the multivalent aptamer and multiple fluorophore DNA nanoprobe. In contrast, the antibody staining seems to achieve a slower binding, with a mediate signal only at 6 hours. Our DNA nanoprobe significantly increased the signal of staining and specificity over time, achieving saturation more rapidly, compared with antibody staining (FIG 11 A-l IB).

[0123] Staining pattern and tissue specificity evaluation of DNA nanoprobe on xenograft tumor section. To evaluate the tissue staining ability of the DNA nanoprobe, xenograft tumors were grown with U251 cell line and prepared frozen sections with a thickness of 5 pm from both the xenograft tumor and normal brain tissue. Both types of sections were processed using a standard staining protocol, which includes fixation, permeabilization, blocking, and staining, followed by incubation with 1 p.M DNA nanoprobe for 10 minutes. Additionally, the xenograft tumor section was stained overnight with an Alexa Fluor ™488-labeled antibody for comparison. As shown in FIG. 12, the top row illustrates the xenograft tissue stained with the labeled antibody. The second and third rows display the DNA nanoprobe staining for the xenograft tumor and normal brain, respectively. These results demonstrate that DNA nanoprobe produced a staining pattern consistent with the antibody, showing high specificity and no non-specific binding. Only the tumor tissue exhibited significant staining, while the normal brain tissue showed minimal signal.

[0124] Optimize rapid DNA nanoprobe staining protocol for detecting intracellular GFAP in the fresh tissue sections. The concentration of the multivalent aptamer and fluorescent DNA nanoprobe was optimized for effective GFAP staining using xenograft brain tumor tissue. Tissue sections were processed using a standard staining protocol — including fixation, permeabilization, blocking, and DNA nanoprobe incubation — with serial nanoprobe concentrations ranging from 10 nM to 1 pM. Staining intensity and specificity were evaluated by confocal microscopy andSeptember 23, 2025 quantified using ImageJ / FIJI software. Nanoprobe penetration efficiency was assessed by preparing tissue sections of varying thickness (5 pm, 10 pm, and 20 pm) and staining with the optimized concentration, followed by analysis of penetration depth to determine the section thickness that provides adequate probe diffusion without loss of resolution. Staining duration was optimized by testing incubation times of 3, 5, 10, and 20 minutes, with fluorescence quality and specificity evaluated at each time point to identify the minimum effective staining period. Fixation and blocking conditions were further refined by comparing different fixation methods (acetone, paraformaldehyde, methanol) and blocking buffers (BSA, casein, fish gelatin) to preserve tissue morphology and enhance DNA nanoprobe binding.

[0125] Thus, in some embodiments, the present invention features a DNA nanoprobe, equipped with multivalent aptamer and multiple fluorophores, capable of detecting intracellular GFAP staining in tissue samples. Without wishing to limit the present invention to any theory of mechanism it is believed that the DNA nanoprobe described herein will have a high specificity and efficacy, and applicability in both in vitro and ex vivo contexts.

[0126] Electric Fiel-Driven (ELF) Staining Technology to electrophoretically accelerate the nanoprobe movement in the tissue to further shorten the staining time. Due to aptamers’ intrinsic negative charge, they can be driven by an electric field through free liquid solutions, as in electric systematic evolution of ligands by exponential enrichment (SELEX), or through gel-like tissues, as seen in electric FISH. Several studies have confirmed that DNA or RNA can electrophoretically pass-through fixed tissues without disturbing the tissue structure, under low or high voltages.

[0127] The present invention features a novel staining method utilizing an electric field to drive DNA aptamers through tissue, increasing the probability of aptamers binding to target proteins and further accelerating the staining process from minutes to seconds.. After being assembled by DNA nanostructures, the fast fusion speed of the aptamer will retain since the DNA scaffold will not alter the flexibility or 3D conformation of the binding region, allowing it to reach the binding site precisely and rapidly. Furthermore, increased molecular weight will not significantly influence the mobility of nanoprobes in the electric field due to the certain charge-to-mass ratio for different DNA molecules. On the other hand, the assembled nanoprobe as described herein, with increased overall charge, will be under a larger electric force comparing to single aptamer, further helping remove it from unspecific binding to DNA binding proteins such as cyclic GMP-AMP synthase (cGAS) and interferon-inducible protein 16 (IFI16). More importantly, by switching the current direction, only specifically bound probes will remain in the tissue and the unbound probes willSeptember 23, 2025 have more chance to bind during the back-and-forth movement, thereby enhancing the signal and reducing the noise. Thus, in certain embodiments, combining the electric field with the multivalent aptamer-based nanoprobe provides a rapid and precise staining method, enabling histopathologic diagnosis in approximately 14 minutes from obtaining a frozen section sample. (FIG. 1 and 2).

[0128] Development and optimization of the ELF staining device were carried out to enable rapid and precise tissue staining. Voltage accuracy was calibrated using a high-precision voltmeter to ensure consistent output from 1 V to 10 V, followed by electric field uniformity testing with an array of voltage sensors to confirm even distribution across the tissue holder. Adjustments to electrode design were made as needed to achieve uniform nanoprobe delivery. Staining buffer conditions were then optimized by evaluating how varying ionic concentrations affected current decay within the small buffer volume (~100 pL). Higher ion concentrations produced faster decay due to increased conductivity, whereas lower concentrations slowed but destabilized current flow. Buffers were systematically tested to identify conditions that maintained stable current and efficient staining. Finally, the forward (up) and reverse (down) electric field application periods were refined using a programmable current controller. Intervals such as 30 s forward / 10 s reverse and 10 s forward / 2 s reverse were compared to enhance nanoprobe transport and binding, with multiple passes through the tissue shown to increase staining intensity. Together, these optimizations yielded a well-calibrated, efficient ELF device capable of rapid tissue staining.

[0129] The optimized ELF staining system was then applied to clinical samples using the multivalent DNA nanoprobe described herein. Owing to its improved binding pattern, the multivalent aptamer exhibited stronger target affinity and remained bound under the electric field, providing markedly higher specificity than a monovalent aptamer. Voltage settings and forward / reverse field intervals were further fine-tuned during clinical testing to maximize staining uniformity and signal quality.

[0130] ELF staining for the GFAP aptamer stains xenograft tumors in approximately 1 minute with exceptionally high specificity. Tissue sections were placed on ITO-coated slides, with wires for the positive and negative poles connected to the slides using copper clips, as ITO is a conductive and transparent material. A PLA platform was 3D-printed to embed two slides, maintaining a distance of 0.1 cm between them (FIG. 13A). A constant voltage of 4 V was applied, generating an electric field intensity of 40 V ein'1. Using an Atmega328 chip, an automatic current controller was built that can change current directions at preset intervals, programmed using C++. The current performance over the 80s staining period is shown in FIG. 13B, including 60s forwardSeptember 23, 2025 voltage and 20s reverse voltage. By switching the current direction, only specifically bound probes remain in the tissue and the unbound probes will have more chance to bind during the back-and- forth movement. Both xenograft tumor and normal brain tissues were stained with a GFAP aptamer and a scramble aptamer was used as a negative control with a standard staining protocol (fix- permeabilize-block-stain). As shown in FIG. 13C, ELF staining demonstrated very high specificity, with a strong signal observed only in the xenograft tumor stained with the target aptamer, displaying a consistent staining pattern. In contrast, the scramble aptamer did not stain any tissue.

[0131] Generative Al Models for Advanced Imaging processing and quantification. The electric field-driven (ELF) staining technology as described above reduces the staining workflow from approximately 25 minutes to 14 minutes, including 10 minutes for blocking and 1 minute for staining after fixation and permeabilization (FIG. 2). While further reduction of blocking time could accelerate intraoperative histopathologic diagnosis, blocking remains important to reduce background noise, improve signal-to-noise ratio, and prevent nonspecific binding to positively charged proteins. To address these limitations, the present invention employs deep learning to enhance imaging and quantification.

[0132] In certain embodiments, image processing is treated as an Image-to-Image (121) translation problem, where generative models such as Generative Adversarial Networks (GANs) or diffusion models are trained to translate images obtained without blocking into images corresponding to standard staining. Once trained and validated, the model performs inference in seconds, effectively eliminating the 10-minute blocking step. For example, BlobCUT, an unpaired 121 translation model combining GANs with contrastive learning, learns noise distributions from unblocked images (e.g., noisy images) and performs image translation to generate denoised outputs (e.g., clean or less noisy images). Used with the multivalent aptamer-based nanoprobes described herein, BlobCUT reduces background noise, segments nuclei to remove nonspecific signals, and enables tissue staining within five minutes or less.

[0133] In some embodiments, BlobCUT may serve two purposes (1) reducing imaging background noises and (2) segmenting out and remove non-specific bindings from nuclei. Comparative analyses using standard histopathological segmentation methods (e.g., Unet++, DAN-NucNet, SCA-Net, NucleiSegNet) indicate that BlobCUT achieves comparable performance when adapted from MRI to histopathology datasets (Table 2). Model performance can be further enhanced by expanding training datasets and incorporating nuclei-specific constraints, such asSeptember 23, 2025 convexity and intensity distributions.

[0134] Table 2: Denoising Al model for nuclei segmentation.

[0135] To further shorten the process, the tissue may be stained without the blocking step (FIG. 14, left), resulting in both target and non-specific nucleus staining (left) due to the positive charges in the nucleus. These two types of images may serve as inputs to the BlobCUT, which translate the non-specific nuclei staining (e.g., unblocked staining) into synthetic blocked images and generates probability masks for nuclei. Thresholding these masks yields final nuclei masks that remove nonspecific staining, as illustrated in FIG. 15.

[0136] Given the input (left), a synthetic blocked strain is rendered (middle). As seen, blue boxes show examples of non-specific bindings from nuclei in the block free staining vs. synthetic block staining. Since BlobCUT is unpaired image-to-image translation, there is no block staining as the ground truth for the block free staining example (left). Comparing middle and right, consistent pattern that blocked staining has no nuclei binding is observed, and synthetic image has the same pattern.

[0137] In certain embodiments, a 2D version of BlobCUT is implemented. Synthetic blobs are generated using elliptical Gaussian functions to create training inputs with corresponding masks and shape constraints. The CUT model is then trained with convexity and distribution constraints to translate blocking-free staining into synthetic blocked images. Resulting nuclei masks remove nonspecific signals from tumor and normal tissue, with normal tissues serving as negative controls for validation.

[0138] Deep Learning for Intraoperative Tumor Margin Determination and Pre-resection Diagnosis. Following BlobCUT denoising, a deep learning classifier can determine tumor margins and perform tumor subtyping. Margin detection identifies residual tumor requiring removal, while subtyping distinguishes lymphoma (non-operative) from GBM (operative). A Feature Transfer-September 23, 2025Enabled Multi-Task Learning (FT-MTL) approach leverages geometric features, such as filamentous GFAP shapes quantified as “blobness,” to assist tumor classification. Training datasets include 200 samples, comprising both blocking and blocking-free (N = 50 each) stained GBM and normal tissues. Validation and testing use only blocking-free data with nuclei images as ground truth, each dataset containing 20 samples (10 GBM, 10 normal). Performance is evaluated using Intersection over Union (loU) and Dice coefficients for margin delineation, and sensitivity and specificity for classification accuracy.

[0139] Statistical evaluation of tumor margin determination. Following the successful validation of the staining protocol, the tumor margin detection capabilities of the present invention will be evaluated using clinical samples collected by neurosurgeons. Tumor margins are defined as the boundaries where cancer cells infiltrate into normal tissue. Tissue samples from the apparent intraoperative tumor margin will be stained and assessed to determine the presence of remaining tumor cells (Figure 16). During surgery, approximately 1 cm3of tissue will be collected and sent for immediate frozen section analysis. Samples will be obtained from both the surface of the solid tumor and the resection cavity. Malignant tumors will serve as positive controls, and benign lesions as negative controls. Each sample will be stained using both standard IHC and the rapid DNA nanoparticle-based staining method. IHC will serve as the gold standard. A pathologist will review both results to assess whether infiltrative tumor cells are present in the margins. Concordant results (both positive or both negative) will be recorded as true positives or true negatives. If IHC is positive but the method described herein is negative, the case will be recorded as a false negative; if IHC is negative but the method described herein is positive, it will be recorded as a false positive. No clinical decisions will be based on the experimental staining results, and surgical procedures will not be altered by study participation. All patient data will be de-identified and securely stored. After enrolling 50 cases, statistical analysis will be performed to compare the diagnostic accuracy of the nanoprobe staining with IHC and determine if additional patient enrollment is needed.

[0140] Without wishing to limit the present invention to any theory or mechanism it is believed that ELF staining will be quick enough to stain the clinical tissue and denoising Al model will further expedite the time-to-diagnosis. The DNA nanoprobe described herein can effectively stain frozen tissue sections and, through the development of a deep learning model, bolstered by a robust staining protocol, image capture, and detailed quantitative analysis, will facilitate a comprehensive assessment of its clinical viability. This deep learning model is expected to increase diagnostic precision and objectivity, uncover subtle diagnostic patterns beyond the scope of human detection thus potentially reduce the staining time. Without wishing to limit the present invention to anySeptember 23, 2025 theory or mechanism, it is believed that for margin and residual tumor determination, a deep learning model may report a predicted probability of tumor residue based on staining intensity and pattern within approximately 5 minutes.EMBODIMENTS

[0141] The following embodiments are intended to be illustrative only and not to be limiting in any way.

[0142] Embodiment 1: A multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target protein, the nanoprobe comprising: a) a DNA nanostructure; b) at least two aptamers bound to the DNA nanostructure, wherein each of the aptamers is configured to bind to the target protein; and c) one or more reporters bound to the DNA nanostructure.

[0143] Embodiment 2: The nanoprobe of embodiment 1, wherein the DNA nanostructure is a three-helix bundle (3HB) or a tetrahedral scaffold. Embodiment 3: The nanoprobe of embodiment 1 or embodiment 2, wherein nanoprobe comprises at least three aptamers bound to the DNA nanostructure. Embodiment 4: The nanoprobe of embodiment 1 or embodiment 2, wherein nanoprobe comprises at least four aptamers bound to the DNA nanostructure. Embodiment 5: The nanoprobe of any one of embodiments 1-4, wherein the at least two aptamers bound to the DNA nanostructure are identical. Embodiment 6: The nanoprobe of any one of embodiments 1-4, wherein the at least two aptamers bound to the DNA nanostructure are different from one another.

[0144] Embodiment 7: The nanoprobe of any one of embodiments 1-6, wherein the at least two aptamers are configured to bind multiple proteins. Embodiment 8: The nanoprobe of any one of embodiments 1-6, wherein the at least two aptamers are configured to bind multiple sites on a single protein. Embodiment 9: The nanoprobe of any one of embodiments 1-8, wherein nanoprobe comprises two or more reporters bound to the DNA nanostructure. Embodiment 10: The nanoprobe of any one of embodiments 1 -8, wherein nanoprobe comprises three or more reporters bound to the DNA nanostructure. Embodiment 11: The nanoprobe of any one of embodiments 1- 10, wherein the aptamers and the reporters are positioned on opposite sides of the nanostructure.

[0145] Embodiment 12: The nanoprobe of any one of embodiments 1-11, wherein the reporter is a fluorescent reporter, a luminescent reporter, chromogens, quantum dots, upconversion nanoparticles, gold nanoparticles, or other nanomaterials. Embodiment 13: The nanoprobe of embodiment 12, wherein the fluorescent reporter is a fluorescent dye. Embodiment 14: The nanoprobe of embodiment 12 or embodiment 13, wherein the reporter is Alexa Fluor™ 488September 23, 2025 fluorescent dye, Alexa Fluor™ 647 fluorescent dye, or Alexa Fluor™ 633 fluorescent dye. Embodiment 15: The nanoprobe of embodiment 14, wherein the reporter is a chromogen such as 3,3'-diaminobenzidine tetrahydrochloride (DAB), horseradish peroxidase (HRP), glucose oxidase, or alkaline phosphatase (AP).

[0146] Embodiment 16: The nanoprobe of any one of embodiments 1-15, wherein the target protein is a biomarker for cancer. Embodiment 17: The nanoprobe of any one of embodiments 1- 16, wherein the cancer is brain cancer, skin cancer, breast cancer, gynecological cancer (including cervical cancer, ovarian cancer, vulvar cancer, vaginal cancer and endometrial / uterine cancer), colorectal cancer, renal cancer, prostate cancer, or the like. Embodiment 18: The nanoprobe of any one of embodiments 1-16, wherein the cancer is brain cancer, and the target protein is selected from a group consisting of glial fibrillary acidic protein (GFAP), CD20, and primary CNS lymphoma (PCNSL). Embodiment 19: The nanoprobe of any one of embodiments 1-16, wherein the cancer is breast cancer and the target protein is selected from a group consisting of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), Ki- 67, CA 15-3, CA 27-29, and CA 125. Embodiment 20: The nanoprobe of any one of embodiments 1-16, wherein the cancer is colorectal cancer and the target protein is selected from a group consisting of carcinoembryonic antigen (CEA), CA 19-9 proteins, human epidermal growth factor receptor 2 (HER2), or tissue inhibitor of metalloproteinase (TIMP)-l

[0147] Embodiment 21: A method of diagnosing cancer in a subject in need thereof; the method comprising: a) obtaining or having obtained a tissue sample from the subject; b) staining the tissue sample using the nanoprobe according to any one of embodiments 1-20, for a period of time; c) detecting a level of staining in the tissue sample; wherein the subject is diagnosed with cancer if the staining is above a predetermined threshold. Embodiment 22: A method of diagnosing cancer in a subject in need thereof; the method comprising: a) obtaining or having obtained a tissue sample from the subject; b) staining the tissue sample using a multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target protein, for a period of time, the nanoprobe comprising: i) a DNA nanostructure; ii) at least two aptamers bound to the DNA nanostructure, wherein each of the aptamers is configured to bind to the target protein; and iii) one or more reporters bound to the DNA nanostructure; c) detecting a level of staining in the tissue sample; wherein the subject is diagnosed with cancer if the staining is above a predetermined threshold.

[0148] Embodiment 23: The method of embodiment 21 or embodiment 22, wherein the DNA nanostructure is a three-helix bundle (3HB) or a tetrahedral scaffold. Embodiment 24: The methodSeptember 23, 2025 of any one of embodiments 21-23, wherein nanoprobe comprises at least three aptamers bound to the DNA nanostructure. Embodiment 25: The method of any one of embodiments 21-23, wherein nanoprobe comprises at least four aptamers bound to the DNA nanostructure. Embodiment 26: The method of any one of embodiments 21-25, wherein the at least two aptamers bound to the DNA nanostructure are identical. Embodiment 27: The method of any one of embodiments 21- 25, wherein the at least two aptamers bound to the DNA nanostructure are identical to one another. Embodiment 28: The method of any one of embodiments 21-25, wherein the at least two aptamers bound to the DNA nanostructure are different from one another.

[0149] Embodiment 29: The method of any one of embodiments 21-28, wherein the at least two aptamers are configured to bind multiple proteins. Embodiment 30: The method of any one of embodiments 21-28, wherein the at least two aptamers are configured to bind multiple sites on a single protein. Embodiment 31: The method of any one of embodiments 21-30, wherein nanoprobe comprises two or more reporters bound to the DNA nanostructure. Embodiment 32: The method of any one of embodiments 21-30, wherein nanoprobe comprises three or more reporters bound to the DNA nanostructure. Embodiment 33: The method of any one of embodiments 21-32, wherein the aptamers and the reporters are positioned on opposite sides of the nanostructure.

[0150] Embodiment 34: The method of any one of embodiments 21-33, wherein the reporter is a fluorescent reporter, a luminescent reporter, chromogens, quantum dots, upconversion nanoparticles, gold nanoparticles, or other nanomaterials. Embodiment 35: The method of embodiment 34, wherein the fluorescent reporter is a fluorescent dye. Embodiment 36: The method of embodiment 34, wherein the reporter is a chromogen such as 3, 3 '-diaminobenzidine tetrahydrochloride (DAB), horseradish peroxidase (HRP), glucose oxidase, or alkaline phosphatase (AP).

[0151] Embodiment 37: The method of any one of embodiments 21-36, wherein the target protein is a biomarker for cancer. Embodiment 38: The method of any one of embodiments 21- 37, wherein the cancer is brain cancer, skin cancer, breast cancer, gynecological cancer (including cervical cancer, ovarian cancer, vulvar cancer, vaginal cancer and endometrial / uterine cancer), colorectal cancer, renal cancer, prostate cancer, or the like.

[0152] Embodiment 39: The method of any one of embodiments 21-38, wherein the tissue samples is an ex vivo sample. Embodiment 40: The method of any one of embodiment 21-39, wherein the tissue sample is from a solid tumor. Embodiment 41 : The method of embodiment 40,September 23, 2025 wherein the solid tumor is within a brain of the subject. Embodiment 42: The method of any one of embodiments 21-41, wherein the cancer is Glioblastoma multiforme (GBM).

[0153] Embodiment 43: The method of any one of embodiments 21-42, wherein staining is performed by an electric field (ELF)-driven reaction (e.g., aptamer-target recognition), and wherein the direction of a current is periodically switched such that unbound nanoprobes are moved within the tissue to increase the likelihood of target binding, while specifically bound nanoprobes remain localized, thereby enhancing (e.g., increasing) signal intensity and reducing background noise, e.g., compared to a sample in which ELF was not used. Embodiment 44: The method of embodiment 43, wherein the ELF-driven reaction enhances (e.g., increase) signal intensity and reduces background, e.g., compared to a sample in which ELF was not used.

[0154] Embodiment 45: The method of any one of embodiments 21-44, wherein the period of time is twenty minutes or less (e.g., the staining is completed in 20 minutes or less). Embodiment 46: The method of any one of embodiments 21-44, wherein the period of time is ten minutes or less (e.g., the staining is completed in 10 minutes or less). Embodiment 47: The method of any one of embodiments 21-44, wherein the period of time is 5 minutes or less (e.g., the staining is completed in 5 minutes or less). Embodiment 48: The method of any one of embodiments 21-44, wherein staining the tissue samples comprises using an electric field (ELF) driven method

[0155] Embodiment 49: The method of any one of embodiments 21-48, wherein the stained tissue is analyzed using image analysis or Al-assisted image analysis; wherein the image analysis detects the level of staining in the tissue sample; wherein the subject is diagnosed with cancer if the staining is above a predetermined threshold as determined by the image analysis. Embodiment 50: The method of embodiment 49, wherein the Al-assisted analysis comprises an imaging-based deep learning (DL) model

[0156] Embodiment 51: A kit comprising: a) the multivalent aptamer-based DNA nanoprobe according to any one of embodiment 1-20 for rapid intracellular staining of a target protein; b) one or more of stains, reagents, or combinations thereof; and c) instructions comprising combining the nucleotide composition and the one or more of stains, reagents, or combinations thereof; and staining a tissue sample. Embodiment 52: A kit comprising: a) a multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target protein, the nanoprobe comprising: i) a DNA nanostructure; ii) at least two aptamers bound to the DNA nanostructure, wherein each of the aptamers is configured to bind to the target protein; and iii) one or more reporters bound to the DNA nanostructure; b) one or more of stains, reagents, or combinations thereof; and c) instructionsSeptember 23, 2025 comprising combining the nucleotide composition and the one or more of stains, reagents, or combinations thereof; and staining a tissue sample

[0157] Embodiment 53: The kit of embodiment 51 or embodiments 52, wherein the DNA nanostructure is a three-helix bundle (3HB) or a tetrahedral scaffold. Embodiment 54: The kit of any one of embodiments 51-53, wherein nanoprobe comprises at least three aptamers bound to the DNA nanostructure. Embodiment 55: The kit of any one of embodiments 51-53, wherein nanoprobe comprises at least four aptamers bound to the DNA nanostructure. Embodiment 56: The kit of any one of embodiments 51-55, wherein the at least two aptamers bound to the DNA nanostructure are identical. Embodiment 57: The kit of any one of embodiments 51-55, wherein the at least two aptamers bound to the DNA nanostructure are identical to one another. Embodiment 58: The kit of any one of embodiments 51-55, wherein the at least two aptamers bound to the DNA nanostructure are different from one another.

[0158] Embodiment 59: The kit of any one of embodiments 51-58, wherein the at least two aptamers are configured to bind multiple proteins. Embodiment 60: The kit of any one of embodiments 51-58, wherein the at least two aptamers are configured to bind multiple sites on a single protein. Embodiment 61: The kit of any one of embodiments 51-60, wherein nanoprobe comprises two or more reporters bound to the DNA nanostructure. Embodiment 62: The kit of any one of embodiments 51 -60, wherein nanoprobe comprises three or more reporters bound to the DNA nanostructure. Embodiment 63: The kit of any one of embodiments 51-62, wherein the aptamers and the reporters are positioned on opposite sides of the nanostructure.

[0159] Embodiment 64: The kit of any one of embodiments 51-63, wherein the reporter is a fluorescent reporter, a luminescent reporter, chromogens, quantum dots, upconversion nanoparticles, gold nanoparticles, or other nanomaterials. Embodiment 65: The kit of embodiment 64, wherein the fluorescent reporter is a fluorescent dye. Embodiment 66: The kit of embodiment 64, wherein the reporter is a chromogen such as 3,3'-diaminobenzidine tetrahydrochloride (DAB), horseradish peroxidase (HRP), glucose oxidase, or alkaline phosphatase (AP).

[0160] Embodiment 67: The kit of any one of embodiments 51-66, further comprising a microfluidic chip for staining. Embodiment 68: The kit of embodiment 67, wherein the microfluidic chip is a disposable cartridge for intraoperative or point-of-care use. Embodiment 69: The kit of any one of embodiments 51-68, further comprising instructions for intraoperative staining. Embodiment 70: The kit of any one of embodiments 51-59, further comprising instructions for point-of-care staining.September 23, 2025

[0161] As used herein, the term “about” refers to plus or minus 10% of the referenced number.

[0162] Although there has been shown and described the preferred embodiment of the present disclosure, it will be readily apparent to those skilled in the art that modifications may be made thereto which do not exceed the scope of the appended claims. Therefore, the scope of the invention is only to be limited by the following claims. In some embodiments, the figures presented in this patent application are drawn to scale, including the angles, ratios of dimensions, etc. In some embodiments, the figures are representative only and the claims are not limited by the dimensions of the figures. In some embodiments, descriptions of the inventions described herein using the phrase “comprising” includes embodiments that could be described as “consisting essentially of’ or “consisting of’, and as such the written description requirement for claiming one or more embodiments of the present disclosure using the phrase “consisting essentially of’ or “consisting of’ is met.

Claims

September 23, 2025WHAT IS CLAIMED IS:

1. A multivalent aptamer-based DNA nanoprobe for rapid intracellular staining of a target protein, the nanoprobe comprising: a. a DNA nanostructure; b. at least two aptamers bound to the DNA nanostructure, wherein each of the aptamers is configured to bind to the target protein; and c. one or more reporters bound to the DNA nanostructure.

2. The nanoprobe of claim 1, wherein the DNA nanostructure is a three-helix bundle (3HB) or a tetrahedral scaffold.

3. The nanoprobe of claim 1, wherein nanoprobe comprises at least three aptamers bound to the DNA nanostructure.

4. The nanoprobe of claim 1, wherein nanoprobe comprises at least four aptamers bound to the DNA nanostructure.

5. The nanoprobe of claim 1, wherein the at least two aptamers bound to the DNA nanostructure are identical.

6. The nanoprobe of claim 1, wherein the at least two aptamers bound to the DNA nanostructure are different from one another.

7. The nanoprobe of claim 1 , wherein the at least two aptamers are configured to bind multiple proteins.

8. The nanoprobe of claim 1 , wherein the at least two aptamers are configured to bind multiple sites on a single protein.

9. The nanoprobe of claim 1, wherein nanoprobe comprises two or more reporters bound to the DNA nanostructure.

10. The nanoprobe of claim 1, wherein nanoprobe comprises three or more reporters bound to the DNA nanostructure.

11. The nanoprobe of claim 1, wherein the aptamers and the reporters are positioned on opposite sides of the nanostructure.

12. The nanoprobe of claim 1, wherein the reporter is a fluorescent reporter, a luminescent reporter, chromogens, quantum dots, upconversion nanoparticles, gold nanoparticles, or other nanomaterials.

13. The nanoprobe of claim 12, wherein the fluorescent reporter is a fluorescent dye.September 23, 202514. The nanoprobe of claim 12, wherein the reporter is a chromogen such as 3,3'- diaminobenzidine tetrahydrochloride (DAB), horseradish peroxidase (HRP), glucose oxidase, or alkaline phosphatase (AP).

15. The nanoprobe of claim 1, wherein the target protein is a biomarker for cancer.

16. The nanoprobe of claim 15, wherein the cancer is brain cancer, skin cancer, breast cancer, gynecological cancer, colorectal cancer, renal cancer, prostate cancer, or the like.

17. The nanoprobe of claim 15, wherein the cancer is brain cancer, and the target protein is selected from a group consisting of glial fibrillary acidic protein (GFAP), CD20, and primary CNS lymphoma (PCNSL).

18. The nanoprobe of claim 15, wherein the cancer is breast cancer and the target protein is selected from a group consisting of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), Ki-67, CA 15-3, CA 27-29, and CA 125.

19. The nanoprobe of claim 15, wherein the cancer is colorectal cancer and the target protein is selected from a group consisting of carcinoembryonic antigen (CEA), CA 19-9 proteins, human epidermal growth factor receptor 2 (HER2), or tissue inhibitor of metalloproteinase (TIMP)-l.

20. A method of diagnosing cancer in a subject in need thereof; the method comprising: a. obtaining or having obtained a tissue sample from the subject; b. staining the tissue sample using the nanoprobe according to claim 1, for a period of time; c. detecting a level of staining in the tissue sample; wherein the subject is diagnosed with cancer if the staining is above a predetermined threshold.

21. The method of claim 20, wherein the tissue samples is an ex vivo sample.

22. The method of claim 20, wherein the tissue sample is from a solid tumor.

23. The method of claim 22, wherein the solid tumor is within a brain of the subject.

24. The method of claim 20, wherein the cancer is Glioblastoma multiforme (GBM).

25. The method of claim 20, wherein staining is performed by an electric field (ELF)-driven reaction.

26. The method of claim 25, wherein the ELF-driven reaction enhances signal intensity and reduces background.

27. The method of claim 25, wherein the direction of a current is periodically switched such that unbound nanoprobes are moved within the tissue to increase the likelihood of targetSeptember 23, 2025 binding, while specifically bound nanoprobes remain localized, thereby enhancing signal intensity and reducing background noise.

28. The method of claim 20, wherein the period of time is twenty minutes or less.

29. The method of claim 20, wherein the period of time is ten minutes or less.

30. The method of claim 20, wherein the period of time is 5 minutes or less.

31. The method of claim 20, wherein the stained tissue is analyzed using image analysis or AI- assisted image analysis; wherein the image analysis detects the level of staining in the tissue sample; wherein the subject is diagnosed with cancer if the staining is above a predetermined threshold as determined by the image analysis.

32. The method of claim 31 , wherein the Al-assisted analysis comprises an imaging -based deep learning (DL) model.

33. A kit comprising: a. the multivalent aptamer-based DNA nanoprobe according to claim 1 for rapid intracellular staining of a target protein b. one or more of stains, reagents, or combinations thereof; and c. instructions comprising combining the nucleotide composition and the one or more of stains, reagents, or combinations thereof; and staining a tissue sample.

34. The kit of claim 33, further comprising a microfluidic chip for staining.

35. The kit of claim 34, wherein the micro fluidic chip is a disposable cartridge for intraoperative or point-of-care use.

36. The kit of claim 33, further comprising instructions for intraoperative staining.

37. The kit of claim 33, further comprising instructions for point-of-care staining

Images