WO2025228445A1

WO2025228445A1 - Methods for preparing mhc-peptide complexes

Info

Publication number: WO2025228445A1
Application number: PCT/CN2025/092746
Authority: WO
Inventors: Yujiang Geno SHI; Gang Song; Jiuxing FENG; Yu Li
Original assignee: Fudan University; Immunoracle Inc
Current assignee: Fudan University; Immunoracle Inc
Priority date: 2023-05-03
Filing date: 2025-05-02
Publication date: 2025-11-06
Anticipated expiration: 2026-11-03
Also published as: AU2024266326A1; WO2025228444A1; WO2024227447A1

Abstract

Methods of producing a plurality of folded MHC-peptide complexes and identifying a pool of peptides suitable for cancer diagnosis. The methods of producing a plurality of folded MHC-peptide complexes comprise: a) adding a plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and b) adding an MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes.

Description

METHODS FOR PREPARING MHC-PEPTIDE COMPLEXES

CROSS REFERENCE TO RELATED APPLICATION

This application claims the benefit of International Application No. PCT/CN2024/091080, filed May 3, 2024, the contents of which are incorporated herein by reference in their entirety for all purposes.

FIELD

The application relates to methods for producing MHC-peptide complexes and methods for screening a pool of peptides suitable for e.g., cancer diagnosis.

BACKGROUND

Decades of cancer research and clinical practice have highlighted a critical challenge: patients diagnosed at later stages often miss the optimal window for curative treatment, resulting in poorer outcome. In contrast, survival rates improve significantly when cancer is detected, diagnosed, and treated early. The primary goal of early detection is to identify tumors that are still localized, have not metastasized, and remain responsive to treatment. Achieving this requires detecting extremely small tumors-less than a few millimeters in diameter of the tumor and consisting of fewer than 10⁶ tumor cells-posing substantial challenges for conventional methods such as physical examinations, laboratory tests, standard imaging technologies, and traditional tissue biopsies.

Liquid biopsy has emerged as a promising non-invasive approach, enabling the analysis of circulating tumor DNA (ctDNA) , circulating tumor cells (CTCs) , and other biomarkers from blood samples. However, significant challenges remain. One of the primary limitations is the low concentration of ctDNA in early-stage cancers, coupled with variability in tumor shedding, which can impede accurate detection. Moreover, the risks of false positives and the high costs associated with these tests pose further barriers.

The immune system, the body’s intrinsic defense network, plays a crucial role in recognizing and combating cancer. Tumor-specific neoantigens (TSNAs) -unique peptides derived from tumor-specific mutations-serve as ideal targets for anti-cancer immune responses. Dendritic cells capture and present TSNAs via major histocompatibility (MHC) molecules to CD8⁺ T cells, triggering T cell receptor (TCR) rearrangement and a targeted immune response. Unlike self-antigens, TSNAs are highly immunogenic and evade central tolerance, enhancing the specific activation and expansion of cytotoxic T lymphocytes. This TSNA-driven immune response has been instrumental in advancing precision immunotherapies, such as personalized cancer vaccines and adoptive T cell therapies. Despite these advances, the potential of harnessing the highly specific immune response for early cancer detection remains underexplored.

Current methods for producing MHC-peptide complexes include refolding recombinant MHC molecules with synthetic peptides, peptide exchange using UV-cleavable or conditional ligands, cell-based expression systems with peptide-loaded MHCs in mammalian or insect cells, and in vitro reconstitution using purified MHC and peptides. Refolding is cost-effective but labor-intensive and limited by peptide solubility. Peptide exchange offers flexibility for high-throughput screening but suffers from incomplete exchange and potential MHC instability. Cell-based systems mimic physiological conditions but are complex, costly, and yield variable peptide loading efficiency. In vitro reconstitution allows precise control but struggles with scalability and MHC denaturation. Each method’s drawbacks-such as inefficiency, instability, high costs, or limited peptide diversity-pose challenges for large-scale production.

The disclosures of all publications, patents, patent applications and published patent applications referred to herein are hereby incorporated herein by reference in their entirety.BRIEF SUMMARY

The present application in one aspect provides methods of producing a plurality of folded MHC-peptide complexes, comprising: a) adding a plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and b) adding an MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes.

The present application in another aspect provides methods of identifying a pool of peptides suitable for cancer diagnosis, comprising: a) generating a plurality of distinct peptides associated with one or more oncogenes; b) adding the plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and c) adding a MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes comprising a pool of peptides suitable for cancer diagnosis. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Tables 8A-8B or 4A.

The present application in another aspect provides methods of identifying a pool of immunogenetic peptides (one or more) associated with an antigen (e.g., for cancer vaccine) , comprising: a) generating a plurality of distinct peptides associated with the antigen; b) adding the plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and c) adding a MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes comprising a pool of immunogenic peptides.

In some embodiments according to any of the methods described above, refolding buffer comprises L-Arginine, oxidized glutathione, reduced glutathione, a chelating agent, a protease inhibitor, and a buffering agent, and wherein the refolding buffer has a pH of about 7.8 to about 8.2. In some embodiments according to any of the methods described above, the chelating agent is EDTA, the protease inhibitor is PMSF, and/or the buffering agent is Tris-HCl. In some embodiments according to any of the methods described above, the refolding buffer comprises 80-125 mM Tris-HCl, 300-500 mM L-Arginine, 1.5-2.5 mM EDTA, 0.4-0.6 mM oxidized glutathione, 4.0-6.0 mM reduced glutathione, and 0.15-0.25 mM PMSF, wherein the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10. In some embodiments according to any of the methods described above, the refolding buffer has a pH of 8.0. In some embodiments according to any of the methods described above, the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF.

In some embodiments according to any of the methods described above, the MHC molecule consists of the same MHC molecule. In some embodiments, the MHC molecule comprises two or more distinct MHC molecules.

In some embodiments according to any of the methods described above, the MHC molecule comprises a MHC I molecule comprising an HLA molecule and a β-2M.

In some embodiments according to any of the methods described above, the molar ratio of the HLA molecule and the β-2M is at least 0.8: 1 and less than 1.5: 1. In some embodiments, the molar ratio of the HLA molecule and the β-2M is about 1: 1. In some embodiments, the HLA molecule has a concentration of about 1-3 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 2 nmol/mL in the refolding buffer. In some embodiments, the β-2M has a concentration of about 1-3 nmol/mL in the refolding buffer, optionally wherein the β-2M has a concentration of about 2 nmol/mL in the refolding buffer.

In some embodiments, the HLA molecule is added to the refolding buffer in batches for three times ( “the first batch, the second batch and the third batch in chronological order” ) , and wherein every adjacent two times are separated by at least 12-24 hours. In some embodiments, the first HLA molecule batch has about half of the HLA molecule, and the second and third HLA molecule batch each has about a quarter of the HLA molecule.

In some embodiments, the HLA molecule is selected from the group consisting of a HLA-A molecule, a HLA-B molecule, a HLA-C molecule, a HLA-DP molecule, a HLA-DQ molecule, and a HLA-DR molecule, optionally wherein the HLA molecule is a HLA-A molecule, further optionally wherein the HLA molecule is selected from the Table 10.

In some embodiments, the HLA molecule comprises a spacer at the C-terminus of the HLA molecule. In some embodiments, the spacer has a length of a peptide of about 5-50 amino acids. In some embodiments, the spacer has a length of a peptide of about 20-30 amino acids. In some embodiments, the spacer comprises a peptide linker of at least five amino acids. In some embodiments, the spacer comprises a peptide linker of 10-20 amino acids, optionally wherein the peptide linker is a GS linker. In some embodiments, the peptide linker is a (G4S) n linker, wherein n is any of 2, 3, and 4. In some embodiments, the spacer comprises a peptide tag, optionally wherein the peptide tag is an Avi-Tag, an AP-tag, a biotin acceptor domain (BAD) , a Sortase A (SrtA) tag, a His-tag, a FLAG-tag, or Strep-tag II. In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-20 amino acids, and b) an Avi-tag.

In some embodiments according to any of the methods described above, the incubation is at 4 ℃.

In some embodiments according to any of the methods described above, the HLA molecule and the β-2M are both recombinant proteins produced in E. coli.

In some embodiments according to any of the methods described above, the volume of the refolding buffer is about or at least about 50 mL.

In some embodiments according to any of the methods described above, the method further comprises adding a salt to the refolding buffer. In some embodiments, the salt is NaCl. In some embodiments, the final concentration of NaCl is about 200 mM.

In some embodiments according to any of the methods described above, the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange i) via ultrafiltration or ii) between ultrafiltration and SEC.

In some embodiments according to any of the methods described above, the plurality of distinct peptides comprise at least about 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different peptides.

In some embodiments according to any of the methods described above, the plurality of folded MHC-peptide complexes comprise at least about 40%, 45%, 50%, 55%, or 60%of the plurality of distinct peptides. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 50%to about 60%.

In some embodiments according to any of the methods described above, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different mutations, wherein the plurality of different mutations comprise mutations (e.g., driver mutations) on the one or more oncogenes.

In some embodiments, the plurality of folded MHC-peptide complexes comprise peptides that target at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%or 85%of the plurality of different mutations, optionally wherein the plurality of folded MHC-peptide complexes comprise about 80%to about 85%of the plurality of different mutations.

In some embodiments, the oncogenes comprise one or more oncogenes selected from Tables 8A-8B or 4A.

In some embodiments, the plurality of mutations comprise one or more mutations selected from Tables 8A-8B or 4B.

Provided herein in another aspect are MHC-peptide complex refolding systems comprising a refolding buffer, a plurality of distinct peptides, an HLA molecule and a β-2M, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL.

In some embodiments according to any of the systems described above, refolding buffer comprises L-Arginine, oxidized glutathione, reduced glutathione, a chelating agent, a protease inhibitor, and a buffering agent, and wherein the refolding buffer has a pH of about 7.8 to about 8.2. In some embodiments according to any of the systems described above, the chelating agent is EDTA. In some embodiments according to any of the systems described above, the protease inhibitor is PMSF. In some embodiments according to any of the systems described above, the buffering agent is Tris-HCl. In some embodiments according to any of the systems described above, the pH of the refolding buffer is about 8.0. In some embodiments according to any of the systems described above, the refolding buffer comprises 80-125 mM Tris-HCl, 300-500 mM L-Arginine, 1.5-2.5 mM EDTA, 0.4-0.6 mM oxidized glutathione, 4.0-6.0 mM reduced glutathione, and 0.15-0.25 mM PMSF. In some embodiments according to any of the systems described above, the refolding buffer has a pH of 8.0 and comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF.

In some embodiments according to any of the systems described above, the molar ratio of the HLA molecule and the β-2M is at least 0.8: 1 and less than 1.5: 1. In some embodiments, the molar ratio of the HLA molecule and the β-2M is about 1: 1.

In some embodiments according to any of the systems described above, the HLA molecule has a concentration of about 1-3 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 2 nmol/mL in the refolding buffer. In some embodiments according to any of the systems described above, the β-2M has a concentration of about 1-3 nmol/mL in the refolding buffer. In some embodiments, the β-2M has a concentration of about 2 nmol/mL in the refolding buffer.

In some embodiments according to any of the systems described above, the HLA molecule is selected from the group consisting of a HLA-A molecule, a HLA-B molecule, a HLA-C molecule, a HLA-DP molecule, a HLA-DQ molecule, and a HLA-DR molecule, optionally wherein the HLA molecule is a HLA-A molecule. In some embodiments according to any of the systems described above, the HLA molecule consists of the same HLA molecule. In some embodiments, the HLA molecule comprises two or more distinct HLA molecules.

In some embodiments according to any of the systems described above, the HLA molecule comprises a spacer at the C-terminus of the HLA molecule, optionally wherein the spacer has a length of a peptide of about 5-50 amino acids. In some embodiments, the spacer has a length of a peptide of about 20-30 amino acids. In some embodiments, the spacer comprises a peptide linker of at least five amino acids. In some embodiments, the spacer comprises a peptide linker of 10-20 amino acids, optionally wherein the peptide linker is a GS linker, further optionally wherein the peptide linker is a (G4S) n linker, wherein n is any of 2, 3, and 4. In some embodiments, the spacer comprises a peptide tag, optionally wherein the peptide tag is an Avi-Tag, an AP-tag, a biotin acceptor domain (BAD) , a Sortase A (SrtA) tag, a His-tag, a FLAG-tag, or Strep-tag II. In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-20 amino acids, and b) an Avi-tag.

In some embodiments according to any of the systems described above, the HLA molecule and the β-2M are both recombinant proteins produced in E. coli.

In some embodiments according to any of the systems described above, the plurality of distinct peptides comprise at least about 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different peptides.

In some embodiments according to any of the systems described above, the plurality of distinct peptides comprises neoantigen peptides that target at least two, three, four, five, six, seven, eight, night, ten, twelve, fifteen, or twenty different oncogenes. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Tables 8A-8B. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Table 4A. In some embodiments, the oncogene comprises one or more of KRAS, EGFR, POLE, TP53, SMAD4, BRCA1, CDKN2A, TNN, MUC16, BRAF and ALK. In some embodiments, the oncogene comprises TP53, KRAS, MUC16, SMAD4, and CDKN2A.

In some embodiments according to any of the systems described above, the plurality of folded MHC-peptide complexes comprise one or more peptides that have a dissociation constant (KD) of more than about any of 10 nM, 100 nM, 1μM, 10 μM, 20 μM, 30 μM, 40 μM, 50 μM, 60 μM, 70 μM, 80 μM, 90 μM, 100 μM, 120 μM, 140 μM, 150 μM, 160 μM, 180 μM, or 200 μM between the peptide and MHC molecule. In some embodiments, the plurality of folded MHC-peptide complexes comprise a) one or more peptides that have a KD of about 0.1 μM to about 10 μM, and b) one or more peptides that have a KD of about 10 μM to about 100 μM (e.g., about 30 μM to 100 μM, about 40 μM to 100 μM, about 50 μM to 100 μM, about 60 μM to 100 μM, about 70 μM to 100 μM, about 80 μM to 100 μM) between the MHC and peptide. In some embodiments, the plurality of folded MHC-peptide complexes comprise a) one or more peptides that have a KD of about 0.1 μM to about 10 μM, b) one or more peptides that have a KD of about 10 μM to about 100 μM and c) one or more peptides that have a KD of more than 100 μM between the MHC and peptide.

In some embodiments according to any of the systems described above, the plurality of folded MHC-peptide complexes comprise one or more peptides that have a dissociation constant (KD) of more than about any of 10 nM, 100 nM, 1μM, 10 μM, 20 μM, 30 μM, 40 μM, 50 μM, 60 μM, 70 μM, 80 μM, 90 μM, 100 μM, 120 μM, 140 μM, 150 μM, 160 μM, 180 μM, or 200 μM between MHC-peptide complexes and a T cell receptor (e.g., a T cell) . In some embodiments, the plurality of folded MHC-peptide complexes comprise a) one or more peptides that have a KD of about 0.1 μM to about 10 μM, and b) one or more peptides that have a KD of about 10 μM to about 100 μM (e.g., about 30 μM to 100 μM, about 40 μM to 100 μM, about 50 μM to 100 μM, about 60 μM to 100 μM, about 70 μM to 100 μM, about 80 μM to 100 μM) . In some embodiments, the plurality of folded MHC-peptide complexes comprise a) one or more peptides that have a KD of about 0.1 μM to about 10 μM, b) one or more peptides that have a KD of about 10 μM to about 100 μM and c) one or more peptides that have a KD of more than 100 μM.

In some embodiments, the KD value for MHC-peptide-TCR interactions is measured using surface plasmon resonance (SPR) . In some embodiments, the KD value for MHC-peptide-TCR interactions is measured using a flow cytometry-based assay.

In some embodiments, the method further comprises performing a second SEC on refolded MHC-peptide complexes following associating a binding component (e.g., a biotin) to the MHC molecule in the MHC-peptide complexes. In some embodiments, the binding component is associated with the MHC molecule via the protein tag (e.g., Avi-tag) .

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 exemplifies Kras neoantigens originated from Kras mutations (Kras G12V, Kras G12D, Kras G12R) .

FIG. 2A depicts the characterization of MHC-monomer bound with multiple antigenic peptides on PAGE.

FIG. 2B depicts the characterization of MHC-tetramer bound with multiple antigenic peptides on PAGE.

FIG. 2C depicts the peptide standards in LC-MS/MS detection.

FIG. 2D depicts the detection of antigenic peptides from PAGE.

FIG. 2E depicts the characterization of MHC-tetramer bound with multiple antigenic peptides in LC-MS/MS detection.

FIG. 2F depicts the characterization of MHC-monomer and MHC-tetramer on PAGE.

FIG. 3A a schematic of tNET: a liquid biopsy platform targeting tumor-engaged T cells for cancer diagnosis.

FIG. 3B shows an overview of tNET: a liquid biopsy platform targeting tumor-engaged T cells for cancer diagnosis. Center circle: Tumor-specific neoantigens (TSNAs) arise from somatic mutations in intracellular proteins and are presented as TNSA-MHC-I complexes (TNSA^PeptideMHC-IC) on tumor cells. These complexes are recognized by tumor neoantigen-engaged CD8⁺ T cells (TNETs) , initiating antitumor immunity. The capture of TNETs cells involves the following steps: 1. Analyzing tumor tissue for nonsynonymous somatic mutations and filtering RNA sequencing and immunopeptidomic data for expressed genes; 2. Generation and algorithmic/MS-based filtering of mutant peptide sequences; 3. Synthesis of candidate peptides and purification of MHC-I subunits; 4. Assembly of peptides with MHC-I to construct a TSNA^peptideMHC-IC^biotin library; 5. Conjugation of pMHC-I complexes to streptavidin-coated magnetic beads for isolation of TNETs from PBMCs; 6. Detection of captured T cells via CD8-HRP and ELISA; 7. A positive ELISA signal indicates prior neoantigen-specific immune responses, suggesting its utility in early pancreatic cancer (PDAC) detection. Abbreviation: SA, streptavidin; PBMCs, Peripheral blood mononuclear cells; HRP, horseradish peroxidase; ELISA, enzyme-linked immunosorbent assay.

FIG. 3C shows KRAS mutation-specific CD8⁺ T cells were captured in peripheral blood of PDAC patients with matching KRAS mutations (KRAS^G12V or KRAS^G12D) and HLA-A alleles profiles using neoantigen-specific tetramer libraries. Tetramer⁺CD8⁺ cell population was detected in P3 (KRAS^G12D/HLA-A*24: 02& HLA-A*11: 01) and P4 (KRAS^G12V/HLA-A*02: 01&HLA-A24: 01) , P6 (KRAS^G12V/HLA-A*02: 01&HLA-A*02: 03) , and P7 (KRAS^G12D/HLA-A*02: 01&HLA-A*02: 07) . No tetramer⁺CD8⁺ T cell population was detected in P1 (KRAS^wildtype/HLA-A*11: 01& HLA-A*11: 01) and P2 (KRAS^wildtype/HLA-A*24: 02& HLA-A*02: 01) , and P5 (KRAS^wildtype/HLA-A*24: 02& HLA-A*02: 06) . P1, P2, P3, and P4 are PDAC patients, and P5, P6, and P7 are intrahepatic cholangiocarcinoma (ICC) patients.

FIG. 3D shows the information of PDAC and ICC patient. KRAS mutations, HLA-A types, tumor types, and tumor stages are listed.

FIG. 4A-4C show design, purification, and assembly of MHC-I complexes for Kras neoantigen.

FIG. 4A shows the structural diagrams of the designed amino acid sequences of the heavy and light chains of MHC-I. Four types of human HLA-A (HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02) along with the light chain B2m were designed.

FIG. 4B shows four types of HLA-A and B2m subunits purified from inclusion bodies using the E. coli system and the purity were evaluated by SDS-PAGE gel running.

FIG. 4C shows the tetramer library of Kras mutations (Kras^G12V or Kras^G12D) were prepared. Kras^G12V or Kras^G12D peptides were assembled into MHC-IC^Biotin monomers, which were then bind with SA-PE. A total of twenty different lengths of neoantigens derived from Kras^G12V or Kras^G12D mutations were included.

FIG. 5A-5E shows engineering and profiling Kras neoantigen-expressing Pan02 cells for KNETs detection.

FIG. 5A shows the strategy to express six Kras mutations-derived neoantigens in Pan02 cell line. The DNA sequences corresponding to 12 amino acids flanking the mutation codon of six Kras mutations was cloned into an expression plasmid for neoantigen production in Pan02 cells.

FIG. 5B shows transcriptional levels of six Kras-derived neoantigens were detected by real-time PCR in Pan02-EV and Pan02-Neo cells. Successful transcription of cloned Kras^{G12V/D/R/C/I/A}sequences was confirmed in Pan02 cells. Quantification of mRNA expression levels was carried out using RT-qPCR in both cell lines.

FIG. 5C shows a structural diagram of the amino acid sequence of mouse MHC class I heavy chain (H2-Kb) and light chain (B2m) .

FIG. 5D shows the purity of H2-Kb and B2m subunits purified from E. coli inclusion bodies was assessed by SDS-PAGE.

FIG. 5E shows the dynamic curve of KNETs detected with mixed tetramers in mice that were intravenously challenged with 1×10³ and 1×10⁴ Pan02-EV or Pan02-Neo. All tetramers were individually assembled through the in vitro refolding of neoantigens derived from Kras^{G12V/D/R/C/I/A}mutations, H2K^b, and B2m. Data are presented as mean ± s.d.

FIG. 6A shows that circulating TNETs are specific and sensitive indicators of nascent tumor initiation and progression. Kras^{G12V/D/R/C/I/A}mutations-derived neoantigen engaged CD8⁺ T cells (KNETs) detection was performed using flow cytometry with tetramers. All tetramers were assembled separately and then were mixed together to detect KNETs in peripheral blood after the subcutaneous challenge with either Pan02-empty vector (Pan02-EV cell line) or Pan02 overexpressing six Kras-mutant peptides (Pan02-Neo cell line) . Ten neoantigens were designed for each Kras mutation site.

FIG. 6B shows the experimental timeline for peripheral blood sampling and KNETs detection in subcutaneous tumor models.

FIG. 6C shows on the left: KNETs frequency in blood was measured 4 days post-Pan02-EV or -Neo cells challenge; and on the right: Tumor growth kinetics (day 4 and 16) showing detectable KNETs (～1 mm tumor diameter) at early timepoints. The tumor diameter was less than 2 mm when KNETs were detectable with mixed tetramers library in peripheral blood of mice subjected to the tumor challenge.

FIG. 6D shows KNETs dynamics detected by mixed Kras^{G12V/D/R/C/I/A}tetramers in mice following a subcutaneous challenge with Pan02-EV or Pan02-Neo cells. The data are expressed as mean ± standard deviation (s.d. ) .

FIG. 6E shows the timepoints for peripheral blood sampling and KNETs detection in intravenous tumor model.

FIG. 6F shows KNETs detection in blood at days 1, 4, 8, and 16 post-intravenous injection of 4×10⁵ Pan02-EV or Pan02-Neo cells.

FIG. 6G shows dose-dependent KNETs kinetics after intravenous challenge (4×10⁵ or 2×10⁴ cells) . Data show mean ± s.d.

FIG. 6H shows the KNETs detection following the second challenge with Pan02-EV or Pan02-Neo cells. Tumor cells were rechallenged when the percentage of KNETs decreased to baseline levels (on 76th day after the first tumor challenge) . Data are presented as mean ± s.d.

FIG. 7A-7E shows the streptavidin bead-based MHC-I neoantigen multimerization and TNETs aggregation assay.

FIG. 7A shows a schematic diagram of the aggregation test principle and procedure. Streptavidin (SA) magnetic beads were used to capture biotinylated MHC class I (MHC-I) monomers loaded with different neoantigens.

FIG. 7B shows SEC purification profiles of OVA^peptideMHC-IC and Kras^G12VMHC-IC monomers, including biotinylated monomers. SEC effectively remove misfolded H2-Kb-B2m, H2-Kb, and B2m. Using 20 Kras^G12V neoantigens, MHC-IC monomers (left) and MHC-IC^Biotin monomers (right) were prepared via “one-pot” refolding.

FIG. 7C shows tetramerization efficiency evaluation for OVA^peptideMHC-IC^Biotin and Kras^G12VMHC-IC^Biotin. Streptavidin (SA) was conjugated to Peptide-MHC-IC^Biotin monomers, with tetramer formation verified by native PAGE. An optimal SA concentration ensured complete binding of Peptide-MHC-IC^Biotin monomers and assembly of tetramers.

FIG. 7D shows the successful loading of OVA^peptideMHC-IC^Biotin monomers onto SA beads was confirmed using SDS-PAGE.

FIG. 7E shows the results of an aggregation assay with beads-based OVA^peptideMHC-IC^Biotin and Kras^G12VMHC-IC^Biotin and PBMCs. PBMCs from peptide-immunized mice (OVA, Kras^G12V, or control peptide) were co-cultured with corresponding SA bead-antigen-MHC-IC to isolate positive aggregation.

FIG. 7F shows the statistical analysis for positive aggregation is shown in FIG. 7E. Data are presented as mean ± s.d. Two-tailed Student's t-test was performed.

[Corrected under Rule 26, 09.07.2025]
FIG. 8A-8H show the development of a high-throughput TSNA^peptideMHC-IC^biotin library system for TNETs capture coupled with an integrated ELISA detection platform.

FIG. 8A shows the procedures for the capture of KNETs or model antigens (OVA) -specific CD8⁺ T cells were conducted with ELISA assays. Kras^PeptideMHC-IC^Biotin monomers loaded with twenty distinct Kras^G12V neoantigen peptides (varying lengths) were prepared by using “one-pot” refolding system and were bound to SA beads to form bead-based heterogeneous Kras^PeptideMHC-IC^Biotin multimers.

FIG. 8B shows the detection of KNETs and OVA-specific CD8⁺ T cells using beads-based heterogeneous Kras^PeptideMHC-IC^Biotin multimers and beads-based OVA^PeptideMHC-IC^Biotin multimers in peripheral blood of mice pre-stimulated with Kras^G12V and OVA peptides. Data show mean ± s.d.

FIG. 8C the detection sensitivities of Luciferase, MRI, c-flow, and tNETs for the early diagnosis of pancreatic tumors on the third day after KPC orthotopic tumor inoculation were performed. SA-PE and CD8-HRP were used as signals for flow cytometry detection and ELISA detection, respectively. KPC tumors were dissected from the mouse pancreas following MRI detection, and luciferase was employed as an indicator to confirm the presence of tumor cells in vivo.

FIG. 8D shows the dynamic profile of TNETs in peripheral blood of a KPC-challenged orthotopic pancreatic tumor model was examined by tNET. KPC cell line harbors Kras^G12D and TP53^R172H mutations. Beads-based Kras^G12D MHC-IC^Biotin and TP53^R172HMHC-IC^Biotin multimers were utilized to capture TNETs in peripheral blood at different timepoint after tumor challenge. Data are presented as mean ± s.d.

FIG. 8E shows the ELISA quantification of peripheral blood TNETs signals (day 8 post-KPC challenge) on the eighth day after KPC cell challenge. Data represent mean ± s.d. Two-tailed t-test was performed.

FIG. 8F shows the pictures of resected KPC tumors on the eighth day after KPC cell challenge from orthotopic pancreatic cancer mouse model when tNET successfully detected Kras^G12D and TP53^R172H neoantigen-specific TNETs. These resected pancreatic tumors from the pancreas of the orthotopic tumor mouse model were analyzed using H&E staining. The tumors before and after resection from pancreas were marked by red arrows.

FIG. 8G shows the dynamic of Kras^G12D neoantigen-specific TNETs in the peripheral blood of patient-derived xenografts (PDX) tumor model in humanized-immunodeficient C-NKG mice. Clinical Kras^G12D PDAC tumors (HLA-A*02: 01) were initially inoculated and passaged in NSG mice to train tumors. Then, these trained PDX tumors were xenografted into humanized severe immunodeficient C-NKG mice (HLA-A*02: 01) to simulate the initiation of PDAC in humans. Kras^G12D neoantigen-specific TNETs were detected using an HLA-A*02: 01 matched beads-based Kras^G12D MHC-IC^Biotin multimers library. Data are presented as mean ± s.d. Two-tailed paired Student's t-test was performed.

FIG. 8H shows the PBMCs were isolated from PDX challenged humanized C-NKG mice, and KRAS^G12D neoantigen-specific TNETs were detected using tNET.

FIG. 9A-9G shows establishment and validation of an orthotopic Kras^G12D pancreatic tumor model for TNETs detection.

[Corrected under Rule 26, 25.07.2025]
FIG. 9A shows Kras mutation profiling in KPC cell line. Western blot analysis identified the Kras^G12D mutation in KPC cells, with Pan02 (Negative control) and ASPC-1 cell lines (Positive control) serving as controls.

FIG. 9B shows in vivo imaging of orthotopic pancreatic tumors. D-Luciferin reporter gene was introduced into KPC cells via lentiviral infection to confirm tumor initiation in orthotopic pancreatic tumor model. This cell line was utilized for in vivo imaging of orthotopic pancreatic tumors.

FIG. 9C shows MRI detection on the third day following the KPC challenge revealed no tumors after the administration of 2×10⁵ KPC cells.

FIG. 9D shows the detection of luciferase activity on the third day after tumor challenge. The mouse were challenging with 2x10⁵ KPC cells, and in vivo bioluminescence imaging for luciferase activity was carried out with IVIS Lumina K device. The images clearly showed the localization of the luciferin-related signals in the pancreatic tumor region. The false color images, where red represented high-intensity signals and blue represented low-intensity signals, visually depicted the spatial distribution of the luciferin-emitting areas within the mice.

FIG. 9E shows that on the third day after tumor cells challenge when TNETs are detectable, the size of the KPC tumor is approximately 1 mm, and its weight is around 1 mg.

FIG. 9F shows the tNET signal threshold was evaluated by constructing orthotopic pancreatic tumors using different numbers of KPC cells. A total of 4×10⁵, 2×10⁵, and 1×10⁴ cells were used respectively. As the number of tumor cells inoculated increased, the appearance of tumor neoantigen-specific CD8⁺ T cells occurred earlier.

FIG. 9G shows that humanized C-NKG mice were constructed and key immune cells, including CD4⁺ T cells, CD8⁺ T cells, plasmacytoid dendritic cells (pDCs) , and conventional dendritic cells (cDCs) , involved in the presentation and recognition of tumor neoantigens were identified using flow cytometry

FIGS. 10A-10D shows generation of the First Human TSNA^PeptideMHC-IC^Biotin Library of tNET Platform for Efficient and Accurate Detection of Early PDAC: tNET V1.0.

FIG. 10A shows neoantigen library construction. Computational prediction of tumor-specific neoantigens for PDAC, ICC and colorectal cancer (CRC) was performed using integrated multi-omics analysis of whole exome sequencing, RNA sequencing, proteomic, and microRNA profiling data.

FIG. 10B shows compositional analysis of PDAC-specific neoantigen library. Big circle on the left shows the distribution of LC-MS/MS-certified neoantigens after one-pot refolding along the mutational genes used for the establishment of the library. Four small circles on right: HLA-A restriction analysis reveals coverage of four prevalent alleles (HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, HLA-A*24: 02) , with similar distribution patterns of mutation-derived neoantigens.

FIG. 10C shows the refolding rates of TNSAs and TSNA^peptideMHC-IC^Biotin prepared by the one-pot refolding system were quantitatively evaluated. The evaluation included peptide-MHC refolding efficiency, mutation coverage rate, and the loading rates of TSNA^peptideMHC-IC^Biotin multimers on SA-beads.

FIG. 10D shows immunogenicity validation of LC-MS/MS-detected and -undetected neoantigens loaded on TSNA^peptideMHC-IC^Biotin monomers were evaluated by ELISpot assay. A panel of 14 neoantigens-10 derived from KRAS and 4 from TP53-were evaluated for their ability to elicit CD8⁺ T-cell responses. Among these, three KRAS neoantigens (No. 1, 2, and 3) and three TP53 neoantigens (No. 4 and 5) were detected from TSNA^peptideMHC-IC^Biotin monomers by LC-MS/MS. The other KRAS (No. 6-12) and TP53 (No. 13 and 14) neoantigens were not detected from TSNA^peptideMHC-IC^Biotin monomers by LC-MS/MS. The immunogenic potential of these peptides was assessed by measuring IFN-γ secretion in an ELISpot assay.

FIG. 11A-11B shows compositional analysis of PDAC-specific neoantigen library.

FIG. 11A shows the library of neoantigens designed based on the mutation types and frequency distribution in the PDAC population is categorized by mutation types. A total of 1, 535 neoantigens were included.

FIG. 11B shows the distribution of PDAC specific library of neoantigens designed based on mutation types and frequency distributed among different types of HLA-A types. Four types of most prevalent HLA-A were included.

FIG. 12A-12F shows high-throughput “one-pot” refolding and multimerization of TSNA^peptideMHC-IC^Biotin complexes for TNETs detection.

FIG. 12A shows that the technical strategy for detecting TNETs involves the use of mixed neoantigen-MHC class I capture beads for clinical samples. In a “one-pot” refolding system, fifty neoantigens are refolded together with MHC-I subunits in a single reaction. After refolding, the MHC class I-neoantigen complexes are purified and biotinylated, and then bound to streptavidin (SA) beads to form capture complexes. This method encompasses four HLA-A types (HLA-A02: 01, HLA-A03: 01, HLA-A11: 01, and HLA-A24: 02) .

FIG. 12B shows a comparison on the refolding products containing varying numbers of neoantigens in the one-pot refolding system. Neoantigens were set in quantities of 20, 50, and 100 types in per reaction (only 20 and 100 neoantigens were shown) .

FIG. 12C shows that an investigation was made on the effect of neoantigen quantity on the formation of TSNA^peptideMHC-IC^Biotin and hetero-tetramers in the “one-pot” refolding system. Three groups, including 20, 50, and 100 neoantigens, were set. The formation of biotin monomers and hetero-tetramers were evaluated using non-reducing, non-denaturing electrophoresis.

FIG. 12D shows the effect of the amount of neoantigens on the formation of biotin-monomers and hetero-tetramers in “one pot” refolding system was investigated. Amounts of 5 mg, 1.5 mg, 0.5 mg, and 10μg of neoantigens were included.

FIG. 12E shows SEC analysis of TSNA^peptideMHC-IC monomers. The main monomers bands were marked.

FIG. 12F shows SEC analysis of TSNA^peptideMHC-IC^Biotin monomers. The main monomers bands were marked.

FIG. 13 shows long-term stability evaluation of biotinylated MHC-I monomer libraries under different storage conditions. The assembled MHC-I monomer library was stored at-80℃, either with or without 10%glycerol. The stability of the biotin-monomers was assessed at 1 week, 2 weeks, 1 month, 3 months, 6 months, and 12 months using PAGE gel electrophoresis. All gels had the same loading order of samples corresponding to different storage durations. The non-reduced marker and reduced marker were loaded in lane #1 and lane #7, respectively. Streptavidin (SA) was added to evaluate the stability of biotinylation on the monomers (lanes #2 and #4) , which is crucial for binding with SA beads. Monomers with (lane #3) or without (lane #5) 10%glycerol, as well as SA (lane #6) , were included as controls.

FIGS. 14A-14I shows that tNET V1.0 demonstrates high-throughput, high sensitivity and specificity in clinical application.

FIG. 14A shows Diagnostic Receiver Operating Characteristic (ROC) curve for tNET detection in PDAC cohort. The ROC curve was generated using 107 cases of mutation-and HLA-A-matched PBMCs samples from PDAC patients.

FIG. 14B shows the detection sensitivity of tNET at 100%specificity (95%CI) in early-and advanced-stage PDAC patients. The cohort was divided into three groups: healthy groups (n=107) , early-stage (I/II, n=87) , and advanced-stage (III/IV, n=20) . Data are presented as mean ± s.d.

FIG. 14C shows the performance of HtNET for the diagnosis of PDAC in a blind study (without prior knowledge of HLA-A*and mutation profiles) was evaluated. A mixed library consisting of HLA-A*02: 01, HLA-A*03: 01, HLA-A*24: 02, and HLA-A*11: 01 specific beads-TSNA^peptideMHC-IC^Biotin hetero-multimers was prepared. A total of 57 PDAC patients and 76 healthy individuals were randomly selected for analysis.

FIG. 14D shows the prognostic value of tNET signals in pancreatic cancer. Kaplan-Meier curves compare progression-free survival (PFS) between patients with positive versus negative tNET detection.

FIG. 14E shows correlation analysis of CA19-9 levels with PFS in PDAC patients. No significant association was observed between serum CA19-9 levels and PFS in the same patient cohort as FIG. 14D.

FIG. 14F shows the false-negative rates of tNET detection and CA19-9 biomarker for PDAC prediction with 100%specificity in healthy individuals and PDAC patients.

FIG. 14G shows the false-negative rate of tNET detection and CA19-9 biomarker for PDAC prediction with 100%specificity in healthy individuals and early-stage PDAC patients.

FIG. 14H shows the diagnostic performance of CA19-9, tNETs, and their combination was evaluated in early-stage PDAC patients. A total of 87 cases of mutation-and HLA-A matched PBMCs samples from early-stage PDAC patients were analyzed.

FIG. 14I shows the diagnostic performance of CA19-9, tNETs, and their combination was evaluated in early-stage PDAC patients. A total of 87 cases of mutation-and HLA-matched PBMCs samples from early-stage PDAC patients were analyzed.

FIG. 15A-15B shows the diagnostic performance of tNET detection for pancreatic cancer across disease stages.

FIG. 15A shows the ROC curve for tNET detection in the diagnosis of PDAC at various stages.

FIG. 15B shows the sensitivity of tNET detection at a specificity of 100%, with 95%CI, in patients with PDAC at various stages. Data are presented as mean ± s.d.

DETAILED DESCRIPTION

The present application in one aspect provides methods of producing a plurality of folded MHC-peptide complexes, comprising: a) adding a plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and b) adding an MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes. The present application in another aspect provides methods of identifying a pool of peptides suitable for cancer diagnosis, comprising: a) generating a plurality of distinct peptides associated with one or more oncogenes; b) adding the plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and c) adding a MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes comprising a pool of peptides suitable for cancer diagnosis. In some embodiments, the MHC molecule comprises an MHC I molecule comprising an HLA molecule and a β-2M. In some embodiments, the molar ratio of the HLA molecule and the β-2M is at least 0.8: 1 and less than 1.5: 1, optionally wherein the molar ratio of the HLA molecule and the β-2M is about 1: 1. In some embodiments, the HLA molecule comprises a spacer at the C-terminus of the HLA molecule, optionally wherein the spacer has a length of a peptide of about 5-50 amino acids, further optionally wherein the spacer has a length of a peptide of about 20-30 amino acids. In some embodiments, the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange i) via ultrafiltration or ii) between ultrafiltration and SEC. In some embodiments, the plurality of distinct peptides comprise at least about 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different peptides. In some embodiments, the plurality of folded MHC-peptide complexes comprise at least about 40%, 45%, 50%, 55%, or 60%of the plurality of distinct peptides, optionally wherein the plurality of folded MHC-peptide complexes comprise about 50%to about 60%. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different mutations, wherein the plurality of different mutations comprise mutations (e.g., driver mutations) on the one or more oncogenes. In some embodiments, the plurality of folded MHC-peptide complexes comprise peptides that target at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%or 85%of the plurality of different mutations, optionally wherein the plurality of folded MHC-peptide complexes comprise about 80%to about 85%of the plurality of different mutations. In some embodiments, the oncogenes comprise one or more oncogenes selected from Table 8A-8B or 4A. In some embodiments, the plurality of mutations comprise one or more mutations selected from Table 8A-8B or 4B.

As demonstrated in e.g., the examples, the methods described herein have been demonstrated to be a scalable and powerful tool to efficiently prepare MHC-peptide complexes in one pot that involve a pool of at least 20, 30, 50, 70, or 100 distinct peptides. In some cases, the folded MHC-peptide complexes have a comprehensive representation of the pool of the distinct peptides, for examples, peptides that target potential mutations associated with cancer. As described in detail in Examples, with the methods described herein, peptides with weaker bindings to the MHC molecule can be preserved. These methods can also be used to identify a pool of neoantigen peptides that are suitable for cancer diagnosis. As shown in the Examples, the MHC-peptides produced with these methods have a pool of neoantigen peptides and can be sensitively (>80%sensitivity) and specifically (about 100%) captured by patients that have related mutations. The methods also effectively select immunogenic peptides.Definitions

In general, terms used in the claims and the specification are intended to be construed as having the plain meaning understood by a person of ordinary skill in the art. Certain terms are defined below to provide additional clarity. In case of conflict between the plain meaning and the provided definitions, the provided definitions are to be used.

As used herein the term "antigen" is a substance that induces an immune response.

As used herein the term "neoantigen" is an antigen that has at least one alteration that makes it distinct from the corresponding wild-type, parental antigen, e.g., via mutation in a tumor cell or post-translational modification specific to a tumor cell. A neoantigen can include a polypeptide sequence. A mutation that results in a neoantigen can include a frameshift or non-frameshift indel, missense or nonsense substitution, splice site alteration, genomic rearrangement or gene fusion, or any genomic or expression alteration giving rise to a neoORF. A mutations can also include a splice variant. Post-translational modifications specific to a tumor cell can include aberrant phosphorylation. Post-translational modifications specific to a tumor cell can also include a proteasome-generated spliced antigen. See Liepe et al., A large fraction of HLA class I ligands are proteasome-generated spliced peptides; Science. 2016 Oct 21; 354 (6310) : 354-358.

As used herein the term "tumor neoantigen" or “cancer neoantigen” is a neoantigen present in a subject's tumor cell or tissue but not in the subject's corresponding normal cell or tissue.

As used herein the term "missense mutation" is a mutation causing a substitution from one amino acid to another.

As used herein the term "nonsense mutation" is a mutation causing a substitution from an amino acid to a stop codon.

As used herein the term "frameshift mutation" is a mutation causing a change in the frame of the protein.

As used herein the term "indel" is an insertion or deletion of one or more nucleic acids.

As used herein, the term percent "identity, " in the context of two or more nucleic acid or polypeptide sequences, refer to two or more sequences or subsequences that have a specified percentage of nucleotides or amino acid residues that are the same, when compared and aligned for maximum correspondence, as measured using one of the sequence comparison algorithms described below (e.g., BLASTP and BLASTN or other algorithms available to persons of skill) or by visual inspection. Depending on the application, the percent "identity" can exist over a region of the sequence being compared, e.g., over a functional domain, or, alternatively, exist over the full length of the two sequences to be compared.

For sequence comparison, typically one sequence acts as a reference sequence to which test sequences are compared. When using a sequence comparison algorithm, test and reference sequences are input into a computer, subsequence coordinates are designated, if necessary, and sequence algorithm program parameters are designated. The sequence comparison algorithm then calculates the percent sequence identity for the test sequence (s) relative to the reference sequence, based on the designated program parameters. Alternatively, sequence similarity or dissimilarity can be established by the combined presence or absence of particular nucleotides, or, for translated sequences, amino acids at selected sequence positions (e.g., sequence motifs) .

As used herein the term "epitope" is the specific portion of an antigen typically bound by an antibody or T-cell receptor.

As used herein the term "immunogenic" is the ability to elicit an immune response, e.g., via T-cells, B cells, or both.

As used herein the term "HLA binding affinity" "MHC binding affinity" means affinity of binding between a specific antigen and a specific MHC allele.

As used herein the term "bait composition" is a composition comprising a molecule (e.g., an antigenic peptide) used to enrich a cell that specifically binds to the bait from a sample. To clarify, “antigenic peptide” described herein is not limited to peptides that are capable of inducing an immune response. It is exchangeable to “peptide” and includes any peptide that can be presented by an MHC molecule.

As used herein the term "variant" is a difference between a subject's nucleic acids and the reference human genome used as a control.

As used herein the term "allele" is a version of a gene or a version of a genetic sequence or a version of a protein.

As used herein the term "HLA type" is the complement of HLA gene alleles.

As used herein the term "exome" is a subset of the genome that codes for proteins. An exome can be the collective exons of a genome.

As used herein the term "dextramers" is a dextran-based peptide-MHC multimers used for antigen-specific immune-cell staining in flow cytometry.

As used herein the term "MHC multimers" is a peptide-MHC complex comprising multiple peptide-MHC monomer units.

As used herein the term "MHC tetramers" is a peptide-MHC complex comprising four peptide-MHC monomer units. As used herein the term "MHC monomers" is a peptide-MHC complex comprising one peptide-MHC monomer units. As used herein the term "MHC dimers" is a peptide-MHC complex comprising two peptide-MHC monomer units. As used herein the term "MHC trimers" is a peptide-MHC complex comprising three peptide-MHC monomer units.

As used herein, "sample" refers to an aliquot of body fluid or a tissue obtained from a subject which contains an immune cell.

The term "mammal" encompasses both humans and non-humans and includes but is not limited to humans, non-human primates, canines, felines, murines, bovines, equines, and porcines.

A “reference” as used herein, refers to any sample, standard, or level that is used for comparison purposes. A reference may be obtained from a healthy and/or non-diseased sample. In some examples, a reference may be obtained from an untreated sample. In some examples, a reference is obtained from a non-diseased or non-treated sample of an individual. In some examples, a reference is obtained from one or more healthy individuals who are not the individual or patient.

The terms “subject, ” “individual, ” and “patient” are used interchangeably herein to refer to a mammal, including, but not limited to, human, bovine, horse, feline, canine, rodent, or primate. In some embodiments, the individual is a human.

It is understood that embodiments of the application described herein include “consisting” and/or “consisting essentially of” embodiments.

Reference to “about” a value or parameter herein includes (and describes) variations that are directed to that value or parameter per se. For example, description referring to “about X” includes description of “X” .

As used herein, reference to “not” a value or parameter generally means and describes “other than” a value or parameter. For example, the method is not used to treat cancer of type X means the method is used to treat cancer of types other than X.

The term “about X-Y” used herein has the same meaning as “about X to about Y. ”

It should be noted that, as used in the specification and t e appended claims, the singular forms "a, " "an, " and "the" include plural referents unless the context clearly dictates otherwise.

Ranges: throughout this disclosure, various aspects of the invention can be presented in a range format. The description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from 10%to 60%should be considered to have specifically disclosed subranges such as from 10%to 30%, from 10%to 40%, from 10%to 50%, from 20%to 40%, from 20%to 60%, from 30%to 60%etc.. This applies regardless of the breadth of the range.

The methods described herein (including methods of analyzing immune cells and methods of detecting antigen-specific immune cells described in following sections) have high specificity and sensitivity and enable it to detect antigen-specific immune cells even when they are rare in the sample (e.g., less than 1%) . Sensitivity is calculated as the %of the true positive/ (true positive +false negative) . Specificity is calculated as the %of the true negative/ (true negative + false positive) . In some embodiments, the method described herein demonstrated a specificity of more than about 50%, 60%, 70%, 80%, 90%, 95%, or 100%. In some embodiments, the method described herein demonstrated a sensitivity of more than about 50%, 55%, 60%, 65%, 70%, 75%or 80%. In some embodiments, the sample comprises PBMC and the immune cells comprise or are T cells or B cells contained in the PBMC. In some embodiments, the immune cells are not subject to an enrichment step for the immune cells in general or immune cells that specifically binds to the antigenic peptide prior to contacting with the bait composition. In some embodiments, the immune cells are not subject to an enrichment step for the immune cells in general or immune cells that specifically binds to the antigenic peptide throughout this methods. In some embodiments, the sample comprise no more than 1x10⁸, 5x10⁷, 2x10⁷, 1x10⁷, 5x10⁶, 2x10⁶, or 1x10⁶ PBMCs. In some embodiments, the sample comprise no more than 1x10⁶, 5x10⁵, 2x10⁵, 1x10⁵, 5x10⁴, 2x10⁴, 1x10⁴, 7x10³, 5x10³, 2x10³, 1x10³, 5x10², 2x10², or 1x10² immune cells (e.g., CD8 T cells, e.g., immune cells that bind to antigenic peptide (s) in the bait composition, e.g., CD4 or CD8 T cells that bind to antigenic peptide (s) in the bait composition) .

Any terms not directly defined herein shall be understood to have the meanings commonly associated with them as understood within the art of the invention. Certain terms are discussed herein to provide additional guidance to the practitioner in describing the compositions, devices, methods and the like of aspects of the invention, and how to make or use them. It will be appreciated that the same thing may be said in more than one way. Consequently, alternative language and synonyms may be used for any one or more of the terms discussed herein. No significance is to be placed upon whether or not a term is elaborated or discussed herein. Some synonyms or substitutable methods, materials and the like are provided. Recital of one or a few synonyms or equivalents does not exclude use of other synonyms or equivalents, unless it is explicitly stated. Use of examples, including examples of terms, is for illustrative purposes only and does not limit the scope and meaning of the aspects of the invention herein.Methods of producing folded MHC-peptide complexes and Methods of a pool of peptides suitable for cancer diagnosis

In some embodiments, there is provided a method of producing a plurality of folded MHC-peptide complexes, comprising: a) adding a plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and b) adding an MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes. In some embodiments, refolding buffer comprises L-Arginine, oxidized glutathione, reduced glutathione, a chelating agent, a protease inhibitor, and a buffering agent, and wherein the refolding buffer has a pH of about 7.8 to about 8.2 (e.g., 8.0) . In some embodiments, the refolding buffer comprises 80-125 mM Tris-HCl, 300-500 mM L-Arginine, 1.5-2.5 mM EDTA, 0.4-0.6 mM oxidized glutathione, 4.0-6.0 mM reduced glutathione, and 0.15-0.25 mM PMSF, wherein the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10. In some embodiments, the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF. In some embodiments, the MHC molecule consists of the same MHC molecule. In some embodiments, the MHC molecule comprises a MHC I molecule comprising an HLA molecule and a β-2M. In some embodiments, the molar ratio of the HLA molecule and the β-2M is at least 0.8: 1 and less than 1.5: 1, optionally wherein the molar ratio of the HLA molecule and the β-2M is about 1: 1. In some embodiments, the HLA molecule is selected from the group consisting of a HLA-A molecule, a HLA-B molecule, a HLA-C molecule, a HLA-DP molecule, a HLA-DQ molecule, and a HLA-DR molecule, optionally wherein the HLA molecule is a HLA-A molecule, further optionally wherein the HLA molecule is selected from the Table 10. In some embodiments, the HLA molecule comprises a spacer at the C-terminus of the HLA molecule, optionally wherein the spacer has a length of a peptide of about 5-50 amino acids, further optionally wherein the spacer has a length of a peptide of about 20-30 amino acids. In some embodiments, the spacer comprises a peptide linker of at least five amino acids. In some embodiments, the spacer comprises a peptide linker of 10-20 amino acids, optionally wherein the peptide linker is a GS linker, further optionally wherein the peptide linker is a (G4S) n linker, wherein n is any of 2, 3, and 4. In some embodiments, the spacer comprises a peptide tag, optionally wherein the peptide tag is an Avi-Tag, an AP-tag, a biotin acceptor domain (BAD) , a Sortase A (SrtA) tag, a His-tag, a FLAG-tag, or Strep-tag II. In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-20 amino acids, and b) an Avi-tag. In some embodiments, the method further comprises adding a salt to the refolding buffer, optionally wherein the salt is NaCl, and further optionally wherein the final concentration of NaCl is about 200 mM. In some embodiments, the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange i) via ultrafiltration or ii) between ultrafiltration and SEC. In some embodiments, the plurality of distinct peptides comprise at least about 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different peptides. In some embodiments, the plurality of folded MHC-peptide complexes comprise at least about 40%, 45%, 50%, 55%, or 60%of the plurality of distinct peptides, optionally wherein the plurality of folded MHC-peptide complexes comprise about 50%to about 60%. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different mutations, wherein the plurality of different mutations comprise mutations (e.g., driven mutations) on the one or more oncogenes. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Tables 22A-22B or 17A. In some embodiments, the plurality of folded MHC-peptide complexes comprise peptides that target at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%or 85%of the plurality of different mutations, optionally wherein the plurality of folded MHC-peptide complexes comprise about 80%to about 85%of the plurality of different mutations. In some embodiments, the oncogenes comprise one or more oncogenes selected from Tables 22A-22B or 17A. In some embodiments, the plurality of mutations comprise one or more mutations selected from Tables 22A-22B or 17B. In some embodiments, the method further comprises performing a second SEC on refolded MHC-peptide complexes following associating a binding component (e.g., a biotin) to the MHC molecule in the MHC-peptide complexes. In some embodiments, the binding component is associated with the MHC molecule via the protein tag (e.g., Avi-tag) .

In some embodiments, there is provided a method of producing a plurality of folded MHC-peptide complexes, comprising sequentially: a) adding a plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; b) adding an MHC molecule into the refolding buffer and incubating for about one to five days (about 1-3 days) ; thereby producing a plurality of folded MHC-peptide complexes, c) adding a salt to the refolding buffer, optionally wherein the salt is NaCl, and further optionally wherein the final concentration of NaCl is about 200 mM, d) concentrating folded MHC-peptide complexes via ultrafiltration in a ratio of no less than 1: 100, 1: 80, 1: 70, 1: 60, or 1: 50, e) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , f) performing a second SEC on refolded MHC-peptide complexes following associating a binding component (e.g., a biotin) to the MHC molecule in the MHC-peptide complexes, wherein the method does not comprise a buffer exchange between ultrafiltration and SEC.

In some embodiments, there is provided a method of identifying a pool of peptides suitable for cancer diagnosis, comprising: a) generating a plurality of distinct peptides associated with one or more oncogenes; b) adding the plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and c) adding a MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes comprising a pool of peptides suitable for cancer diagnosis. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Tables 22A-22B or 17A. In some embodiments, refolding buffer comprises L-Arginine, oxidized glutathione, reduced glutathione, a chelating agent, a protease inhibitor, and a buffering agent, and wherein the refolding buffer has a pH of about 7.8 to about 8.2 (e.g., 8.0) . In some embodiments, the refolding buffer comprises 80-125 mM Tris-HCl, 300-500 mM L-Arginine, 1.5-2.5 mM EDTA, 0.4-0.6 mM oxidized glutathione, 4.0-6.0 mM reduced glutathione, and 0.15-0.25 mM PMSF, wherein the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10. In some embodiments, the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF. In some embodiments, the MHC molecule consists of the same MHC molecule. In some embodiments, the MHC molecule comprises a MHC I molecule comprising an HLA molecule and a β-2M. In some embodiments, the molar ratio of the HLA molecule and the β-2M is at least 0.8: 1 and less than 1.5: 1, optionally wherein the molar ratio of the HLA molecule and the β-2M is about 1: 1. In some embodiments, the HLA molecule is selected from the group consisting of a HLA-A molecule, a HLA-B molecule, a HLA-C molecule, a HLA-DP molecule, a HLA-DQ molecule, and a HLA-DR molecule, optionally wherein the HLA molecule is a HLA-A molecule, further optionally wherein the HLA molecule is selected from the Table 10. In some embodiments, the HLA molecule comprises a spacer at the C-terminus of the HLA molecule, optionally wherein the spacer has a length of a peptide of about 5-50 amino acids, further optionally wherein the spacer has a length of a peptide of about 20-30 amino acids. In some embodiments, the spacer comprises a peptide linker of at least five amino acids. In some embodiments, the spacer comprises a peptide linker of 10-20 amino acids, optionally wherein the peptide linker is a GS linker, further optionally wherein the peptide linker is a (G4S) n linker, wherein n is any of 2, 3, and 4. In some embodiments, the spacer comprises a peptide tag, optionally wherein the peptide tag is an Avi-Tag, an AP-tag, a biotin acceptor domain (BAD) , a Sortase A (SrtA) tag, a His-tag, a FLAG-tag, or Strep-tag II. In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-20 amino acids, and b) an Avi-tag. In some embodiments, the method further comprises adding a salt to the refolding buffer, optionally wherein the salt is NaCl, and further optionally wherein the final concentration of NaCl is about 200 mM. In some embodiments, the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange i) via ultrafiltration or ii) between ultrafiltration and SEC. In some embodiments, the plurality of distinct peptides comprise at least about 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different peptides. In some embodiments, the plurality of folded MHC-peptide complexes comprise at least about 40%, 45%, 50%, 55%, or 60%of the plurality of distinct peptides, optionally wherein the plurality of folded MHC-peptide complexes comprise about 50%to about 60%. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different mutations, wherein the plurality of different mutations comprise mutations (e.g., driven mutations) on the one or more oncogenes. In some embodiments, the plurality of folded MHC-peptide complexes comprise peptides that target at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%or 85%of the plurality of different mutations, optionally wherein the plurality of folded MHC-peptide complexes comprise about 80%to about 85%of the plurality of different mutations. In some embodiments, the oncogenes comprise one or more oncogenes selected from Tables 22A-22B or 17A. In some embodiments, the plurality of mutations comprise one or more mutations selected from Tables 22A-22B or 17B. In some embodiments, the method further comprises performing a second SEC on refolded MHC-peptide complexes following associating a binding component (e.g., a biotin) to the MHC molecule in the MHC-peptide complexes. In some embodiments, the binding component is associated with the MHC molecule via the protein tag (e.g., Avi-tag) .

In some embodiments, there is provided a method of identifying a pool of peptides suitable for cancer diagnosis, comprising sequentially: a) adding a plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; b) adding an MHC molecule into the refolding buffer and incubating for about one to five days (about 1-3 days) ; thereby producing a plurality of folded MHC-peptide complexes, c) adding a salt to the refolding buffer, optionally wherein the salt is NaCl, and further optionally wherein the final concentration of NaCl is about 200 mM, d) concentrating folded MHC-peptide complexes via ultrafiltration in a ratio of no less than 1: 100, 1: 80, 1: 70, 1: 60, or 1: 50, e) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , f) performing a second SEC on refolded MHC-peptide complexes following associating a binding component (e.g., a biotin) to the MHC molecule in the MHC-peptide complexes, wherein the method does not comprise a buffer exchange between ultrafiltration and SEC.

In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.01 mg/mL to about 0.1 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.01 mg/mL to about 0.03 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.03 mg/mL to about 0.05 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.05 mg/mL to about 0.07 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.07 mg/mL to about 0.9 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.09 mg/mL to about 1.0 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.01 mg/mL, 0.02 mg/mL, 0.025 mg/mL, 0.03 mg/mL, 0.04 mg/mL, 0.05 mg/mL, 0.06 mg/mL, 0.07 mg/mL, 0.075 mg/mL, 0.08 mg/mL, 0.09 mg/mL, 0.1 mg/mL.

In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Tables 8A-8B. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Table 4A. In some embodiments, the oncogene comprises one or more of KRAS, EGFR, POLE, TP53, SMAD4, BRCA1, CDKN2A, TNN, MUC16, BRAF and ALK. In some embodiments, the oncogene comprises TP53, KRAS, MUC16, SMAD4, and CDKN2A.Refolding Buffer

In some embodiments, the refolding buffer comprises L-Arginine, oxidized glutathione, reduced glutathione, a chelating agent, a protease inhibitor, and a buffering agent, and wherein the refolding buffer has a pH of about 7.8 to about 8.2.

In some embodiments, the refolding buffer has a pH of about 7.8 to about 8.2. In some embodiments, the refolding buffer has a pH of about 7.8 to 8.0. In some embodiments, the refolding buffer has a pH of about 8.0 to 8.2. In some embodiments, the refolding buffer has a pH of 8.0.

In some embodiments, the chelating agent is EDTA, the protease inhibitor is PMSF, and/or the buffering agent is Tris-HCl. In some embodiments, the chelating agent is EDTA. In some embodiments, the protease inhibitor is PMSF. In some embodiments, the buffering agent is Tris-HCL. In some embodiments, the refolding buffer comprises 80-125 mM Tris-HCl, 300-500 mM L-Arginine, 1.5-2.5 mM EDTA, 0.4-0.6 mM oxidized glutathione, 4.0-6.0 mM reduced glutathione, and 0.15-0.25 mM PMSF, wherein the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10.

In some embodiments, the refolding buffer comprises 80-125 mM Tris-HCl. In some embodiments, the refolding buffer comprises about 80-90 mM Tris-HCl. In some embodiments, the refolding buffer comprises about 90-100 mM Tris-HCl. In some embodiments, the refolding buffer comprises about 100-110 mM Tris-HCl. In some embodiments, the refolding buffer comprises about 110-120 mM Tris-HCl. In some embodiments, the refolding buffer comprises about 120-125 mM Tris-HCl. In some embodiments, the refolding buffer comprises about 80 mM, 90 mM, 100 mM, 110 mM, 120 mM, or 125 mM Tris-HCl. In some embodiments, the refolding buffer comprises about 100 mM Tris-HCl.

In some embodiments, the refolding buffer comprises about 300-500 mM L-Arginine. In some embodiments, the refolding buffer comprises about 300-350 mM L-Arginine. In some embodiments, the refolding buffer comprises about 350-400 mM L-Arginine. In some embodiments, the refolding buffer comprises about 400-450 mM L-Arginine. In some embodiments, the refolding buffer comprises about 450-500 mM L-Arginine. In some embodiments, the refolding buffer comprises about 300mM, 350 mM, 400 mM, 450 mM, or 500 mM L-Arginine. In some embodiments, the refolding buffer comprises about 400 mM L-Arginine.

In some embodiments, the refolding buffer comprises about 1.5-2.5 mM EDTA. In some embodiments, the refolding buffer comprises about 1.5-1.75 mM EDTA. In some embodiments, the refolding buffer comprises about 1.75-2.0 mM EDTA. In some embodiments, the refolding buffer comprises about 2.0-2.25 mM EDTA. In some embodiments, the refolding buffer comprises about 2.25-2.50 mM EDTA. In some embodiments, the refolding buffer comprises about 1.5 mM, 1.75 mM, 2.0 mM, or 2.25 mM, 2.50 mM EDTA In some embodiments, the refolding buffer comprises about 2 mM EDTA.

In some embodiments, the refolding buffer comprises about 0.4-0.6 mM oxidized glutathione. In some embodiments, the refolding buffer comprises about 0.4-0.5 mM oxidized glutathione. In some embodiments, the refolding buffer comprises about 0.5-0.6 mM oxidized glutathione. In some embodiments, the refolding buffer comprises about 0.4 mM, 0.5 mM, or 0.6 mM oxidized glutathione. In some embodiments, the refolding buffer comprises about 0.5 mM oxidized glutathione. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10.

In some embodiments, the refolding buffer comprises about 4.0-6.0 mM reduced glutathione. In some embodiments, the refolding buffer comprises about 4.0-4.5 mM reduced glutathione. In some embodiments, the refolding buffer comprises about 4.5-5.0 mM reduced glutathione. In some embodiments, the refolding buffer comprises about 5.0-5.5 mM reduced glutathione. In some embodiments, the refolding buffer comprises about 5.5-6.0 mM reduced glutathione. In some embodiments, the refolding buffer comprises about 4.0 mM, 4.5 mM, 5.0 mM, 5.5 mM, or 6.0 mM reduced glutathione. In some embodiments, the refolding buffer comprises about 5.0 mM reduced glutathione. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10.

In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 6. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 6 to 1: 7. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 7 to 1: 8. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 8 to 1: 9. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 9 to 1: 10. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5, 1: 6, 1: 7, 1: 8, 1: 9, or 1: 10. In some embodiments, the molar ratio of oxidized glutathione and reduced glutathione is about 1: 10.

In some embodiments, the refolding buffer comprises about 0.15-0.25 mM PMSF. In some embodiments, the refolding buffer comprises about 0.15-0.20 mM PMSF. In some embodiments, the refolding buffer comprises about 0.20-0.25 mM PMSF. In some embodiments, the refolding buffer comprises about 0.15 mM, 0.2 mM, or 0.25 mM PMSF. In some embodiments, the refolding buffer comprises about 0.2 mM PMSF.

In some embodiments, the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF. In some embodiments, the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF, and the refolding buffer has a pH of 8.0.MHC molecule

In some embodiments, the MHC molecule consists of the same MHC molecule. In some embodiments, the MHC molecule comprises two or more distinct MHC molecules. In some embodiments, the MHC molecule comprises two, three, four, five, six, seven, eight, nine, or ten or more distinct MHC molecules.

In some embodiments, the MHC molecule comprises an MHC I molecule comprising an HLA molecule and a β-2M. In some embodiments, the molar ratio of the HLA molecule and the β-2M is at least 0.8: 1 and less than 1.5: 1. In some embodiments, the molar ratio of the HLA molecule and the β-2M is about 1: 1.

In some embodiments, the MHC molecule comprises an MHC I molecule comprising an HLA molecule. Human leukocyte antigens (HLA) are genes in major histocompatibility complexes (MHC) that help code for proteins that differentiate between self and non-self. HLA genes are highly polymorphic, meaning they have many different versions, and each person inherits two copies of each HLA gene, one from each parent. These two copies determine an individual’s HLA type.

In some embodiments, the HLA molecule has a concentration of about 1-3 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 1-1.5 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 1.5-2.0 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 2.0-2.5 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 2.5-3.0 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 1 nmol/mL, 1.5 nmol/mL, 2 nmol/mL, 2.5 nmol/mL, or 3 nmol/mL in the refolding buffer. In some embodiments, the HLA molecule has a concentration of about 2 nmol/mL in the refolding buffer.

In some embodiments, the HLA molecule is selected from the group consisting of a HLA-Amolecule, a HLA-B molecule, a HLA-C molecule, a HLA-DP molecule, a HLA-DQ molecule, and a HLA-DR molecule. In some embodiment, the HLA molecule is a HLA-A molecule. In some embodiments, the HLA molecule is selected from the Table 10. In some embodiments, the HLA molecule allele is selected from the group consisting of HLA-A*11: 01, HLA-A*02: 01, HLA-A*24: 02, HLA-A*01: 01, HLA-A*33: 03, HLA-A*26: 01, HLA-A*32: 01, HLA-A*02: 05, HLA-A*02: 06, HLA-A*03: 02, HLA-A*02: 11, HLA-A*68: 01, HLA-A*30: 01, HLA-A*29: 01, and HLA-A*29: 02. In some embodiments, the HLA molecule allele is selected from the group consisting of HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02.

In some embodiments, the MHC molecule comprises a MHC I molecule comprising a β-2M. The beta-2-microglobulin (β-2M) protein is a C1-type Ig superfamily (IgSF) protein that associates with most MHC-I proteins and is essential for their proper protein structure and function. In some embodiments, the β-2M molecule has a concentration of about 1-3 nmol/mL in the refolding buffer. In some embodiments, the β-2M molecule has a concentration of about 1-1.5 nmol/mL in the refolding buffer. In some embodiments, the β-2M molecule has a concentration of about 1.5-2.0 nmol/mL in the refolding buffer. In some embodiments, the β-2M molecule has a concentration of about 2.0-2.5 nmol/mL in the refolding buffer. In some embodiments, the β-2M molecule has a concentration of about 2.5-3.0 nmol/mL in the refolding buffer. In some embodiments, the β-2M molecule has a concentration of about 1 nmol/mL, 1.5 nmol/mL, 2 nmol/mL, 2.5 nmol/mL, or 3 nmol/mL in the refolding buffer. In some embodiments, the β-2M molecule has a concentration of about 2 nmol/mL in the refolding buffer.

In some embodiments, the HLA molecule is added to the refolding buffer in batches for three times ( “the first batch, the second batch and the third batch in chronological order” ) . In some embodiments, the first HLA molecule batch has about half of the HLA molecule, and the second and third HLA molecule batch each has about a quarter of the HLA molecule. In some embodiments, the first HLA molecule batch comprises about 1 nmol of the HLA molecule per mL refolding buffer, the second HLA molecule batch comprises about 0.5 nmol of the HLA molecule per mL refolding buffer, and the third HLA molecule batch comprises about 0.5 nmol of the HLA molecule per mL refolding buffer. In some embodiments, every adjacent two times are separated by at least 12-24 hours. In some embodiments, every adjacent two times are separated by at least 12, 16, 20, or 24 hours. In some embodiments, a first HLA molecule batch comprising about 1 nmol of the HLA molecule per mL refolding buffer is added to the refolding buffer, followed by a second HLA molecule batch comprising about 0.5 nmol of the HLA molecule per mL refolding buffer, followed by a third HLA molecule batch comprising about 0.5 nmol of the HLA molecule per mL refolding buffer, wherein the second batch is added to the refolding buffer 12-24 hours after the first batch, and the third batch is added to the refolding buffer 12-24 hours after the second batch.

In some embodiments, the HLA molecule comprises a spacer at the C-terminus of the HLA molecule. In some embodiments, the spacer has a length of a peptide of about 5-50 amino acids. In some embodiments, the spacer comprises a peptide linker of at least five amino acids. In some embodiments, the spacer comprises a peptide linker of at least 10 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 15 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 20 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 25 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 30 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 35 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 40 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 45 amino acids. In some embodiments, the spacer comprises a peptide linker of at least 50 amino acids. In some embodiments, the spacer comprises a peptide linker of 5-10 amino acids. In some embodiments, the spacer comprises a peptide linker of 10-20 amino acids. In some embodiments, the spacer has a length of a peptide of about 20-30 amino acids. In some embodiments, the spacer has a length of a peptide of about 30-40 amino acids. In some embodiments, the spacer has a length of a peptide of about 40-50 amino acids. In some embodiments, the spacer has a length of a peptide of about 10-30 amino acids. In some embodiments, the spacer has a length of a peptide of about 20-40 amino acids.

In some embodiments the peptide linker is a GS linker. In some embodiments, the peptide linker is a (G4S) n linker, wherein n is any of 2, 3, and 4.

In some embodiments, the spacer comprises a peptide tag. In some embodiments, the peptide tag is an Avi-Tag, an AP-tag, a biotin acceptor domain (BAD) , a Sortase A (SrtA) tag, a His-tag, a FLAG-tag, or Strep-tag II. In some embodiments, the spacer comprises a polyhistidine tag. In some embodiments, the polyhistidine tag is his*10. In some embodiments, the polyhistidine tag is at the C-terminal to the peptide linker. In some embodiments, the peptide tag is an Avi-Tag.

In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-50 amino acids, and b) an Avi-tag. In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-30 amino acids, and b) an Avi-tag. In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-20 amino acids, and b) an Avi-tag. In some embodiments, the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 20-30 amino acids, and b) an Avi-tag.

MHC molecules are also described in PCT/US2022/078953, the disclosures of which are hereby incorporated herein by reference in their entirety.Linkers[0001] In some embodiment, a linker (such as peptide linker) comprises flexible residues (such as glycine and serine) so that the adjacent domains are free to move relative to each other. For example, a glycine-serine doublet can be a suitable peptide linker. In some embodiments, the linker is a non-peptide linker. In some embodiments, the linker is a peptide linker. In some embodiments, the linker is a non-cleavable linker. In some embodiments, the linker is a cleavable linker.[0001] Other linker considerations include the effect on physical or pharmacokinetic properties of the resulting compound, such as solubility, lipophilicity, hydrophilicity, hydrophobicity, stability (more or less stable as well as planned degradation) , rigidity, flexibility, immunogenicity, modulation of antibody binding, the ability to be incorporated into a micelle or liposome, and the like.Peptide linkers[0002] The peptide linker may have a naturally occurring sequence, or a non-naturally occurring sequence. For example, a sequence derived from the hinge region of heavy chain only antibodies may be used as the linker. See, for example, WO1996/34103.[0003] The peptide linker can be of any suitable length. In some embodiments, the peptide linker is at least about any of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, or 50 amino acids long. In some embodiments, the peptide linker is no more than about any of 50, 40, 35, 30, 25, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5 or fewer amino acids long. In some embodiments, the length of the peptide linker is any of about 1 amino acid to about 10 amino acids, about 1 amino acid to about 20 amino acids, about 1 amino acid to about 30 amino acids, about 5 amino acids to about 15 amino acids, about 10 amino acids to about 25 amino acids, about 5 amino acids to about 30 amino acids, about 10 amino acids to about 30 amino acids long, about 10 amino acids to about 50 amino acids.[0004] An essential technical feature of such peptide linker is that said peptide linker does not comprise any polymerization activity. The characteristics of a peptide linker, which comprise the absence of the promotion of secondary structures, are known in the art and described, e.g., in Dall’Acqua et al. (Biochem. (1998) 37, 9266-9273) , Cheadle et al. (Mol Immunol (1992) 29, 21-30) and Raag and Whitlow (FASEB (1995) 9 (1) , 73-80) . A particularly preferred amino acid in context of the “peptide linker” is Gly. Furthermore, peptide linkers that also do not promote any secondary structures are preferred. The linkage of the domains to each other can be provided by, e.g., genetic engineering. Methods for preparing fused and operatively linked bispecific single chain constructs and expressing them in mammalian cells or bacteria are well-known in the art (e.g. WO 99/54440, Ausubel, Current Protocols in Molecular Biology, Green Publishing Associates and Wiley Interscience, N.Y. 1989 and 1994 or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 2001) .[0005] The peptide linker can be a stable linker, which is not cleavable by proteases, especially by Matrix metalloproteinases (MMPs) .[0006] The linker can also be a flexible linker (such as a GS linker) . Exemplary flexible linkers include glycine polymers (G) _n, glycine-serine polymers (including, for example, (GS) _n (SEQ ID NO: 108) , (GSGGS) _n (SEQ ID NO: 109) , (GGGGS) _n (SEQ ID NO: 110) , and (GGGS) _n (SEQ ID NO: 111) , where n is an integer of at least one) , glycine-alanine polymers, alanine-serine polymers, and other flexible linkers known in the art. Glycine and glycine-serine polymers are relatively unstructured, and therefore may be able to serve as a neutral tether between components. Glycine accesses significantly more phi-psi space than even alanine, and is much less restricted than residues with longer side chains (See Scheraga, Rev. Computational Chem. 11 173-142 (1992)) .[0007] Furthermore, exemplary linkers also include the amino acid sequence of such as (GGGGS) n (SEQ ID NO: 112) , wherein n is an integer between 1 and 8. In some embodiments, the peptide linker comprises the amino acid sequence of (GSTSGSGKPGSGEGS) n (SEQ ID NO: 113) , wherein n is an integer between 1 and 3.Non-peptide linkers[0008] Coupling of two moieties may be accomplished by any chemical reaction that will bind the two molecules so long as both components retain their respective activities. This linkage can include many chemical mechanisms, for instance covalent binding, affinity binding, intercalation, coordinate binding and complexation. In some embodiments, the binding is covalent binding. Covalent binding can be achieved either by direct condensation of existing side chains or by the incorporation of external bridging molecules. Many bivalent or polyvalent linking agents may be useful in coupling protein molecules in this context. For example, representative coupling agents can include organic compounds such as thioesters, carbodiimides, succinimide esters, diisocyanates, glutaraldehyde, diazobenzenes and hexamethylene diamines. This listing is not intended to be exhaustive of the various classes of coupling agents known in the art but, rather, is exemplary of the more common coupling agents (See Killen and Lindstrom, Jour. Immun. 133: 1335-2549 (1984) ; Jansen et al., Immunological Reviews 62: 185-216 (1982) ; and Vitetta et al., Science 238: 1098 (1987)) .[0009] Linkers that can be applied in the present application are described in the literature (see, for example, Ramakrishnan, S. et al., Cancer Res. 44: 201-208 (1984) describing use of MBS (M-maleimidobenzoyl-N-hydroxysuccinimide ester) . In some embodiments, non-peptide linkers used herein include: (i) EDC (1-ethyl-3- (3-dimethylamino-propyl) carbodiimide hydrochloride; (ii) SMPT (4-succinimidyloxycarbonyl-alpha-methyl-alpha- (2-pridyl-dithio) -toluene (Pierce Chem. Co., Cat. (21558G) ; (iii) SPDP (succinimidyl-6 [3- (2-pyridyldithio) propionamido] hexanoate (Pierce Chem. Co., Cat #21651G) ; (iv) Sulfo-LC-SPDP (sulfosuccinimidyl 6 [3- (2-pyridyldithio) -propianamide] hexanoate (Pierce Chem. Co. Cat. #2165-G) ; and (v) sulfo-NHS (N-hydroxysulfo-succinimide: Pierce Chem. Co., Cat. #24510) conjugated to EDC.[0010] The linkers described above contain components that have different attributes, thus may lead to bispecific antibodies with differing physio-chemical properties. For example, sulfo-NHS esters of alkyl carboxylates are more stable than sulfo-NHS esters of aromatic carboxylates. NHS-ester containing linkers are less soluble than sulfo-NHS esters. Further, the linker SMPT contains a sterically hindered disulfide bond, and can form antibody fusion protein with increased stability. Disulfide linkages, are in general, less stable than other linkages because the disulfide linkage is cleaved in vitro, resulting in less antibody fusion protein available. Sulfo-NHS, in particular, can enhance the stability of carbodimide couplings. Carbodimide couplings (such as EDC) when used in conjunction with sulfo-NHS, forms esters that are more resistant to hydrolysis than the carbodimide coupling reaction alone.Incubation

In some embodiments, the incubation is at about 3-5 ℃. In some embodiments, the incubation is at about 4 ℃. In some embodiments, the incubation is stirred between about 200 rpm to 700 rpm. In some embodiments, the incubation is stirred between about 700 rpm prior to addition of the HLA molecule and β-2M/peptide mixture. In some embodiments, the incubation is stirred between about 200 rpm after addition of the HLA molecule and β-2M/peptide mixture. In some embodiments, the volume of the refolding buffer is about or at least about 50 mL. In some embodiments, the volume of the refolding buffer is about 50 mL.

In some embodiments, the HLA molecule and the β-2M are both recombinant proteins produced in E. coli.

In some embodiments, the method further comprises adding a salt to the refolding buffer. In some embodiments, the salt is NaCl. In some embodiments, the final concentration of NaCl is at least about 200 mM. In some embodiments, the final concentration of NaCl is about 170-230 mM NaCl. In some embodiments, the final concentration of NaCl is about 180-220 mM NaCl. In some embodiments, the final concentration of NaCl is about 190-210 mM NaCl. In some embodiments, the final concentration of NaCl is about 200 mM.

In some embodiments, the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange via ultrafiltration. In some embodiments, the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange between ultrafiltration and SEC. In some embodiments, the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange i) via ultrafiltration or ii) between ultrafiltration and SEC. In some embodiments, the method does not comprise a ultrafiltration with more than about a 1: 100 concentration, e.g., 1: 200-1: 1000, 1: 200-1: 500, 1: 200-1: 300, 1: 300-1: 400, 1: 400-1: 500, 1: 200, 1: 300, 1: 400, 1: 500 concentration. As discussed in the examples, it was found that a repeated or strong ultrafiltration such as those that are more than 1: 100 can cause the disappearance of MHC-peptide that have a weaker binding affinity. In some embodiments, the method does not comprise reducing the original volume of the refolding reaction volume to less than 1/100 (e.g., 1/200-1/1000, 1/200-1/500, 1/200-1/300, 1/300-1/400, 1/400-1/500, 1/200, 1/300, 1/400, 1/500) via ultrafiltration concentration or other methods. In some embodiments, the method comprises a gentle ultrafiltration (e.g., wherein the final refolding reaction volume is reduced by no more than 1/50 (e.g., 1/10-1/50, 1/20-1/50, 1/30-1/50, 1/10, 1/20, 1/30, 1/40, 1/50 of the original volume) followed by size exclusion chromatography.

In some embodiments, the plurality of distinct peptides comprise at least about 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different peptides. In some embodiments, the plurality of distinct peptides comprise about 10-100 different peptides. In some embodiments, the plurality of distinct peptides comprise about 10-50 different peptides. In some embodiments, the plurality of distinct peptides comprise about 50-100 different peptides. In some embodiments, the plurality of distinct peptides comprise about 5-10 different peptides. In some embodiments, the plurality of distinct peptides comprise about 10-20 different peptides. In some embodiments, the plurality of distinct peptides comprise about 20-30 different peptides. In some embodiments, the plurality of distinct peptides comprise about 30-40 different peptides. In some embodiments, the plurality of distinct peptides comprise about 40-50 different peptides. In some embodiments, the plurality of distinct peptides comprise about 50-60 different peptides. In some embodiments, the plurality of distinct peptides comprise about 60-70 different peptides. In some embodiments, the plurality of distinct peptides comprise about 70-80 different peptides. In some embodiments, the plurality of distinct peptides comprise about 80-90 different peptides. In some embodiments, the plurality of distinct peptides comprise about 90-100 different peptides.

In some embodiments, the plurality of folded MHC-peptide complexes comprise at least about 40%, 45%, 50%, 55%, or 60%of the plurality of distinct peptides. In some embodiments, the plurality of folded MHC-peptide complexes comprise at least about 40%of the plurality of distinct peptides. In some embodiments, the plurality of folded MHC-peptide complexes comprise at least about 50%of the plurality of distinct peptides. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 40%to about 60%. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 40%to about 45%. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 45%to about 50%. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 50%to about 55%. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 55%to about 60%. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 50%to about 60%.

In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different mutations. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 2-10 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 10-20 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 20-30 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 30-40 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 40-50 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 50-60 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 60-70 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 70-80 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 80-90 different mutation. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 90-100 different mutation.

In some embodiments, the plurality of different mutations comprise mutations (e.g., driver mutations) on the one or more oncogenes. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Table 8A-8B. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Table 4A. In some embodiments, the oncogene comprises one or more of KRAS, EGFR, POLE, TP53, SMAD4, BRCA1, CDKN2A, TNN, MUC16, BRAF and ALK. In some embodiments, the oncogene comprises TP53, KRAS, MUC16, SMAD4, and CDKN2A. In some embodiments, the plurality of different mutations comprise a mutation listed in Tables 8A-8B or 17B.

In some embodiments, the plurality of folded MHC-peptide complexes comprise peptides that target at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%or 85%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise peptides that target at least about 50%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise peptides that target at least about 80%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 50%to about 85%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 50%to about 55%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 55%to about 60%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 60%to about 65%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 65%to about 70%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 70%to about 75%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 75%to about 80%of the plurality of different mutations. In some embodiments, the plurality of folded MHC-peptide complexes comprise about 80%to about 85%of the plurality of different mutations.

In some embodiments, the plurality of mutations comprise one or more mutations selected from Tables 8A-8B. In some embodiments, the plurality of mutations comprise one or more mutations selected from Table 8A-8B or 4B.Refolding system

In some aspects, provided herein is an MHC-peptide complex refolding system comprising a refolding buffer, a plurality of distinct peptides, an HLA molecule and a β-2M, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.01 mg/mL to about 0.1 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.01 mg/mL to about 0.03 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.03 mg/mL to about 0.05 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.05 mg/mL to about 0.07 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.07 mg/mL to about 0.9 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.09 mg/mL to about 1.0 mg/mL. In some embodiments, the concentration of the plurality of distinct peptides in the refolding buffer is about 0.01 mg/mL, 0.02 mg/mL, 0.025 mg/mL, 0.03 mg/mL, 0.04 mg/mL, 0.05 mg/mL, 0.06 mg/mL, 0.07 mg/mL, 0.075 mg/mL, 0.08 mg/mL, 0.09 mg/mL, 0.1 mg/mL.

In some embodiments, the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF. In some embodiments, the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF, and the refolding buffer has a pH of 8.0.

In some embodiments, the HLA molecule is selected from the group consisting of a HLA-Amolecule, a HLA-B molecule, a HLA-C molecule, a HLA-DP molecule, a HLA-DQ molecule, and a HLA-DR molecule. In some embodiment, the HLA molecule is an HLA-A molecule. In some embodiments, the HLA molecule is an HLA molecule listed on Table 10. In some embodiments, the HLA molecule comprises one, two, three, four or more HLA molecules selected from the group consisting of HLA-A*11: 01, HLA-A*02: 01, HLA-A*24: 02, HLA-A*01: 01, HLA-A*33: 03, HLA-A*26: 01, HLA-A*32: 01, HLA-A*02: 05, HLA-A*02: 06, HLA-A*03: 02, HLA-A*02: 11, HLA-A*68: 01, HLA-A*30: 01, HLA-A*29: 01, and HLA-A*29: 02. In some embodiments, the HLA molecule is selected from the group consisting of HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02. In some embodiments, the HLA molecule comprises HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02 molecules.

In some embodiments, the HLA molecule is added to the refolding buffer in batches for three times ( “the first batch, the second batch and the third batch in chronological order” ) . In some embodiments, the first HLA molecule batch has about half of the HLA molecule, and the second and third HLA molecule batch each has about a quarter of the HLA molecule. In some embodiments, the first HLA molecule batch comprises about 1 nmol of the HLA molecule per mL refolding buffer, the second HLA molecule batch comprises about 0.5 nmol of the HLA molecule per mL refolding buffer, and the third HLA molecule batch comprises about 0.5 nmol of the HLA molecule per mL refolding buffer. In some embodiments, every adjacent two times are separated by at least 12-24 hours. In some embodiments, every adjacent two times are separated by at least 12, 16, 20, or 24 hours.

In some embodiments the peptide linker is a GS linker. In some embodiments, the peptide linker is a (G4S) n linker, wherein n is any of 2, 3, and 4. In some embodiments, the spacer comprises a G4S linker and a 10*his-tag.

In some embodiments, the spacer comprises a peptide tag. In some embodiments, the peptide tag is an Avi-Tag, an AP-tag, a biotin acceptor domain (BAD) , a Sortase A (SrtA) tag, a His-tag, a FLAG-tag, or Strep-tag II. In some embodiments, the spacer comprises a polyhistidine tag. In some embodiments, the polyhistidine tag is his-tag*10. In some embodiments, the polyhistidine tag is at the C-terminal to the peptide linker. In some embodiments, the peptide tag is an Avi-Tag.

In some embodiments, the plurality of different mutations comprise mutations (e.g., driver mutations) on the one or more oncogenes. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Table 8A-8B. In some embodiments, the one or more oncogenes comprise the one or more oncogene listed in Table 4A. In some embodiments, the oncogene comprises one or more of KRAS, EGFR, POLE, TP53, SMAD4, BRCA1, CDKN2A, TNN, MUC16, BRAF and ALK. In some embodiments, the oncogene comprises TP53, KRAS, MUC16, SMAD4, and CDKN2A. In some embodiments, the plurality of different mutations comprise a mutation listed in Tables 8A-8B or 4B.

In some embodiments, the plurality of mutations comprise one or more mutations selected from Tables 8A-8B. In some embodiments, the plurality of mutations comprise one or more mutations selected from Table 8A-8B or 4B.

In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target at least two, three, four, five, six, seven, eight, night, ten, twelve, fifteen, or twenty different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target two to 20 different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target two to 10 different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target two to five different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target five to ten different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target at least two different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target at least five different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target at least ten different oncogenes. In some embodiments, the plurality of distinct peptides comprises neoantigen peptides that target at least fifteen different oncogenes.MHC molecules

MHC I and II molecules present protein fragments to CD8+ and CD4+ T cells, respectively. These molecules are essential for cell-mediated immunity and therefore appeared at the inception of the adaptive immune system. For their construction they used two Ig-domains topped by two parallel alpha helixes resting on a platform of beta-pleated sheets. This capital structure generated a peptide-binding groove between the alpha helixes, which is ‘evolutionarily speaking’ likely borrowed from earlier chaperone structures. Yet MHC I and MHC II molecules are unique in the proteome because of their extreme polymorphism (>10,000 different alleles of MHC I molecules have been identified thus far) . This has interesting consequences. Polymorphic residues on the top alpha helixes interact with the TCR and are the basis for the specificity of TCRs for both an antigen peptide plus a particular allelic form of an MHC molecule (a phenomenon called MHC restriction) . Polymorphic residues in the MHC peptide binding groove change the nature and location of so-called pockets. These variable pockets are filled by complementary variable amino acid side chains of peptides (so-called anchor residues) , with the effect that different fragments from a defined antigen are presented by different polymorphic MHC molecules. Yet, next to the anchor residues, most other amino acids in a peptide fill a free space and can be (almost) any of the 20 amino acids. By having pockets with specificity for only a few side chains and allowing the remaining 6–10 amino acids to vary between all possibilities, each kind of MHC molecules can present a very large repertoire of peptides. Moreover, by having 3 to 6 different MHC I as well as 3 to 12 different MHC II molecules (the exact number depending on how many different MHC alleles were inherited from one’s parents and how the MHC II subunits paired) , cells can present a large fraction of the universe of peptides, although not all sequences. See e.g., Trends Immunol. 2016 Nov; 37 (11) : 724–737.

In theory then, MHC I molecules can present a peptidome of around 6 × 20 ^(6–7) different peptides, and MHC II ca display up to 12 × 20 ⁽¹⁰⁾ peptides. In actuality, such a large array of peptides cannot all be presented because there are only around 200,000 MHC I and 20,000 MHC II molecules on cells such as B and T cells per cell. Moreover, since some peptides are present in high number (from highly expressed proteins) , the real number of different peptides presented by one cell is likely less than 10,000. Importantly, when a pathogen alters a critical anchor residue in one of its antigenic epitopes, it may prevent presentation of this antigen in one individual but not in another person with different MHC molecules that will simply select different peptides from the same pathogen. Therefore, MHC polymorphism is good for the survival of the population and not necessarily the individual. See e.g., Trends Immunol. 2016 Nov; 37 (11) : 724–737.

The MHC molecules described herein can be any MHC molecules. See e.g., Cell Mol Immunol. 2015 Mar; 12 (2) : 139-53.

In some embodiments, the MHC molecule is a MHC class I molecule.

In some embodiments, the MHC class I molecule is selected from the group consisting of HLA-A, HLA-B, HLA-C, and HLA-D. In some embodiments, the MHC class I molecule is selected from the group consisting of HLA-A, HLA-B, and HLA-C. In some embodiments, MHC class I molecule comprises a HLA-A molecule. In some embodiments, the HLA-A molecule comprises a mutation that reduces its binding to CD8, further optionally where the HLA-A molecule comprises a HLA A2 heavy chain with a A245V mutation. It was found that this mutation does not result in loss of reactivity towards some of the specific T cells, but in fact with result in a decreased the background staining.

In preferred embodiments, the MHC class I molecule is selected from the group consisting HLA-A*24: 02, HLA-A*11: 01, HLA-A*02: 01, and HLA-A*03: 01. In preferred embodiments, the MHC class I molecule comprises multiple kinds of MHC class I molecules comprising HLA-A* 24: 02, HLA-A*11: 01, HLA-A*02: 01, and HLA-A*03: 01. In another preferred embodiments, the MHC class I molecule comprises multiple kinds of MHC class I molecules comprising HLA-A*69: 01, HLA-A*31: 01, HLA-A*29: 01HLA-A*33: 02, HLA-A*02: 06, HLA-A*02: 07, HLA-A*30: 01, HLA-A*01: 01, HLA-A*02: 03, HLA-A*33: 03.

In some embodiments, the HLA molecule has a His-tag fused to the C-terminus. In some embodiments, the HLA molecule has a polypeptide (e.g., a GS linker, e.g., a G4S linker) that has a length of about 5-15, 7-13, 8-12, 9-11 or 10 amino acids fused to the C-terminus.

In some embodiments, the antigenic peptide complexed with MHC I molecule is about 8 to about 10 amino acids long. In some embodiments, the antigenic peptide is at least 8 (e.g., 8, 9, or 10) amino acids long.

In some embodiments, the MHC molecule is a recombinant MHC I molecule.

In some embodiments, the MHC molecule is a MHC class II molecule.

In some embodiments, the MHC class II molecule is selected from the group consisting of HLA-DR, HLA-DQ, and HLA-DF. In some embodiments, the MHC class II molecule is selected from the group consisting of HLA-DQ and HLA-DR. In some embodiments, the antigenic peptide that complexed with a MHC class II molecule is about 10 to about 20 amino acids long. In some embodiments, the antigenic peptide is at least 10 (e.g., 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20) amino acids long.

In some embodiments, the MHC molecule is a recombinant MHC II molecule.

In some embodiments, the MHC molecule comprise both an MHC I molecule and an MHC II molecule.

In preferred embodiments, the MHC molecule matches with at least one HLA type of the individual from where the sample is obtained. For example, a MHC I molecule comprising HLA-A*24: 02 or HLA-A*11: 01 is used when the individual has both HLA-A*24: 02 and HLA-A*11: 01. Patient-specific NGS data from WGS, WES, or RNA-seq can be used to predict HLA types with computational tools such as Optiptype and Polysolver (polymorphic loci resolver) . See e.g., Szolet et al., Bioinformatics 30, 3310–3316, e.g., Shukla et al., Nat. Biotechnol. 33, 1152–1158. Reads can be selected from the NGS data that potentially derived from the HLA region and then they can be fully aligned to a full-length genomic library of all known HLA alleles. See e.g., Nucleic Acids Res. 41, D1222–D1227.

In some embodiments, the MHC molecule is coupled with a chaperon molecule prior to being complexed to the antigenic peptide. See e.g., Overall et al., Nat Commun. 2020 Apr 20; 11 (1) : 1909.

In some embodiments, the display moiety comprises an HLA-A molecule, an HLA-B molecule and/or an HLA-C molecule. In some embodiments, the display moiety comprises an HLA-A molecule. In some embodiments, the HLA-A molecule comprises a mutation that reduces its binding to CD8. In some embodiments, the HLA-A molecule comprises an HLA A2 heavy chain with a A245V mutation.

In some embodiments, the display moiety comprises an MHC class II molecule. In some embodiments, the MHC class II molecule is selected from the group consisting of an HLA-DO molecule, an HLA-DM molecule, an HLA-DP molecule, an HLA-DQ molecule and an HLA-DR molecule. In some embodiments, the MHC class II molecule is selected from the group consisting of an HLA-DP molecule, an HLA-DQ molecule and an HLA-DR molecule. In some embodiments, the display moiety comprises an MHC class II molecule selected from the group consisting of an HLA-DQ molecule and an HLA-DR molecule.

In some embodiments, the display moiety described herein comprises at least two different MHC molecules.

In some embodiments, the at least two different MHC molecules comprises two different MHC class I molecules. In some embodiments, the two different MHC class I molecules comprise an HLA-A molecule and at least one of an HLA-B molecule and an HLA-C molecule. In some embodiments, the two different MHC class I molecules comprise an HLA-A molecule and an HLA-B molecule. In some embodiments, the two different MHC class I molecules comprise an HLA-A molecule and an HLA-C molecule. In some embodiments, the two different MHC class I molecules comprise an HLA-B molecule and an HLA-C molecule. In some embodiments, the two different MHC class I molecules comprise a) an HLA-A molecule, b) an HLA-B molecule and c) an HLA-C molecule.

In some embodiments, the MHC multimer comprises at least two MHC class I molecules that are at least two different HLA-A molecules. In some embodiments, the at least two different HLA-A molecules are selected from the group consisting of a) HLA-A*24: 02, HLA-A*11: 01, HLA-A*02: 01, and HLA-A*03: 01, or b) HLA-A*69: 01, HLA-A*31: 01, HLA-A*29: 01HLA-A*33: 02, HLA-A*02: 06, HLA-A*02: 07, HLA-A*30: 01, HLA-A*01: 01, HLA-A*02: 03, HLA-A*33: 03.

In some embodiments, the at least two different MHC molecules comprises two different MHC class II molecules. In some embodiments, the two different MHC class II molecules comprise at least two MHC class molecules selected from the group consisting of an HLA-DP molecule, an HLA-DQ molecule and an HLA-DR molecule. In some embodiments, the two different MHC class II molecules comprise an HLA-DP molecule and an HLA-DQ molecule. In some embodiments, the two different MHC class II molecules comprise an HLA-DP molecule and an HLA-DR molecule. In some embodiments, the two different MHC class II molecules comprise an HLA-DR molecule and an HLA-DQ molecule. In some embodiments, the two different MHC class II molecules comprise an HLA-DP molecule, an HLA-DQ molecule, and an HLA-DR molecule.

In some embodiments, the at least two different MHC molecules comprises both an MHC class I molecule and an MHC class II molecule. In some embodiments the at least two different MHC molecules comprise a) an HLA-A molecule and b) an MHC class II molecule selected from the group consisting of an HLA-DP molecule, an HLA-DQ molecule and an HLA-DR molecule. In some embodiments the at least two different MHC molecules comprise a) an HLA-A molecule and b) an HLA-DP molecule. In some embodiments the at least two different MHC molecules comprise a) an HLA-A molecule and b) an HLA-DQ molecule. In some embodiments the at least two different MHC molecules comprise a) an HLA-A molecule and b) an HLA-DR molecule.

In some embodiments, the display moiety comprises two or more (e.g., at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 1000, 10,000, or 100,000) antigenic peptides. In some embodiments, the display moiety comprises four antigenic peptides. In some embodiments, the two or more antigenic peptides in the display moiety are the same or similar. In some embodiments, the two or more antigenic peptides in the display moiety are distinct. In some embodiments, at least one of the two or more antigenic peptides is a truncal antigenic peptide.

In some embodiments, the display moiety comprises an MHC/peptide monomer, dimer, trimer, and/or tetramer. In some embodiments, MHC/peptide complexes are assembled into monomers, dimers, trimers, and/or tetramers, comprising 1, 2, 3, or 4 MHC/peptide complexes bound to a display moiety. In some embodiments, the MHC/peptide complex further comprises a detectable label. The detectable label is a fluorophore, such as phycoerythrin (PE) , allophycocyanin (APC) or any fluorophore known in the art. In some embodiments, the MHC/peptide complex does not comprise a detectable label.

In some embodiments, the MHC/peptide complex is assembled into a multimer (such as, dimer, trimer, tetramer, pentamer, hexamer, or high order multimer) . In some embodiments, the multimer can comprise at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 MHC/peptide complexes. In some embodiments, a high throughput peptide-MHC (pMHC) tetramer library is constructed. See e.g., Overall et al., Nat Commun. 2020 Apr 20; 11 (1) : 1909.

In some embodiments, the display moiety further comprises a barcode (e.g., a DNA barcode) . In some embodiments, each of the one or more display moieties comprises a unique barcode (e.g., a unique DNA barcode) . See e.g., Overall et al., Nat Commun. 2020 Apr 20; 11 (1) : 1909.

In some embodiments, the display moiety comprises a particle. In some embodiments, the particle is selected from the group consisting of a surface, a nanoparticle, a bead, and a polymer. In some embodiments, the particle is a magnetic nanoparticle, e.g. for isolation using a magnet. See e.g., Peng et al., Cell Rep. 2019 Sep 3; 28 (10) : 2728-2738. e7. In some embodiments, the magnetic particle comprises magnetic iron oxide. In some embodiments, the particle is a polystyrene nanoparticle, e.g., for isolation by gravity. In some embodiments, the particle is an agarose bead. In some embodiments, the particle is a sepharose bead. In some embodiments, the particle is a dextran particle. In some embodiments, the particle is a biotinylated dextran or a streptavidin-coated dextran. The particle in the display moiety is not the same as the solid support or substrate used to immobilize the immune cells as described herein.

In some embodiments, the particle is detectable. In some embodiments, the particle is fluorescent. In some embodiments, the particle is attached directly or indirectly to a fluorophore. In some embodiments, the particle is modified with an attachment moiety for attaching additional molecules.

In some embodiments, the antigenic peptide or MHC is directly attached to the particle. In some embodiments, the antigenic peptide or MHC is attached to the particle via a binding pair comprising a first binding component (e.g., biotin) attached to the antigenic peptide and a second binding component (streptavidin) bound to the particle. In some embodiments, the binding components are any suitable moieties known in the art (such as, thiol, maleimide, cyclodextrin, amine, adamantine, carboxy, azide, and alkyne) .

In some embodiments, multiple display moieties (e.g., MHC/peptide complexes) are attached to a single particle.

In some embodiments, the display moiety comprises a cell (e.g., an antigen presenting cell, e.g., a dendritic cell, e.g., a macrophage) . In some embodiments, the cell comprises a polynucleotide encoding the antigenic peptide (e.g., a truncal antigenic peptide) . In some embodiments, the polynucleotide encodes a plurality of antigenic peptides. In some embodiments, the plurality of antigenic peptides are displayed on the surface of the cell (e.g., antigen presenting cells) . The cells used for display antigenic peptides are different from the immune cells in the sample described herein. In some embodiments, the plurality of antigenic peptides are displayed on the surface of the cell in complex with an MHC molecule.

In some embodiments, the cell is obtained from the individual. In some embodiments, the cell has at least one (or two) same HLA type as that of the individual. For example, if the individual has HLA-A*24: 02, the cell in the display moiety also has HLA-A*24: 02.

In some embodiments, the display moiety further comprises a detectable label. In some embodiments, the detectable label is a fluorophore. In some embodiments, the display moiety is itself fluorescent or is attached to a fluorophore directly or indirectly. In some embodiments, the fluorophore is a phycoerythrin (PE) , allophycocyanin (APC) or any fluorophore known in the art. Antigenic peptides (i.e., peptides)

Peptides described herein can be derived from exogenous antigens (e.g., an antigen from a pathogen) or endogenous antigens (e.g., an antigen associated with a cancer cell, e.g., a neoantigen, e.g., an autoantigen) .

The size of a antigenic peptide can comprise, but is not limited to, about 5, about 6, about 7, about 8, about 9, about 10, about 11, about 12, about 13, about 14, about 15, about 16, about 17, about 18, about 19, about 20, about 21, about 22, about 23, about 24, about 25, about 26, about 27, about 28, about 29, about 30, about 31, about 32, about 33, about 34, about 35, about 36, about 37, about 38, about 39, about 40, about 41, about 42, about 43, about 44, about 45, about 46, about 47, about 48, about 49, about 50, about 60, about 70, about 80, about 90, about 100, about 110, about 120 or greater amino molecule residues, and any range derivable therein. In preferred embodiments, the antigenic peptide molecules are equal to or less than 50 amino acids.

In some embodiments, the antigenic peptide complexed with a MHC Class I molecule has 15 residues or less in length and usually consist of between about 8 and about 11 residues, particularly 9 or 10 residues. In some embodiments, the antigenic peptides complexed with a MHC Class II molecule has 6-30 residues.

If desirable, a longer peptide can be designed in several ways. For example, when presentation likelihoods of peptides on HLA alleles are predicted or known, a longer peptide could consist of either: (1) individual presented peptides with an extensions of 2-5 amino acids toward the N-and C-terminus of each corresponding gene product; (2) a concatenation of some or all of the presented peptides with extended sequences for each.

In some embodiments, the one or more antigenic peptides are about 8-50 amino acids in length. In some embodiments, it is about 8-10 amino acids in length. In some embodiments, it is greater than 10 amino acids in length, greater than 15 amino acids in length, greater than 20 amino acids in length, or greater than 30 amino acids in length. In some embodiments, it is about 24-40 amino acids in length.

In some embodiments, the one or more antigenic peptides used in the bait compositions described herein are further optimized based upon one or more selection criteria.

In some embodiments, the antigenic peptides are further selected based on their likelihood to be processed and/or presented on the cell surface HLA molecules. In some embodiments, in silico prediction algorithms (such as any of the algorithms described herein) is used as the basis for the selection. In some embodiments, immunopeptidomics analysis is used as the basis for the selection.

Computational algorithms such as NetMHC (See e.g., Andreatta et al., Bioinformatics 32, 511–517, 2016) , NetMHCpan (See e.g., Rammensee et al., Immunogenetics 50, 213–219, 1999) , and MHCflurry (O’Donnell et al., Cell Syst. 7, 129–132. e124, 2018) trained on large in vitro experimental datasets can be used to prioritize candidate neoantigens that bind to the predicted HLA types with high affinity. For example, Neopepsee and pVAC-Seq are representative analysis pipelines for tumor somatic mutations (Hundal et al., Genome Med. 8, 11, 2016; Kim et al., Ann. Oncol. 29, 1030–1036, 2018) . Recently, a new prediction model-EDGE based on tumor HLA peptide mass spectrometry (MS) datasets has increased the positive predictive value up to nine-fold. (Bulik-Sullivan et al., Nat. Biotechnol. 18: 4313) .

In some embodiments, the one or more antigenic peptides are selected based upon its binding affinity to a) an MHC molecule and/or b) a cognate TCR molecule.

In some embodiments, the antigenic peptide has a binding affinity that is less than 5000 nM (IC50) to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity of about 500 nM to 5000 nM (IC50) to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity that is less than 500 nM (IC50) to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity of about 250 nM to 500 nM IC50 to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity that is less than 250 nM (IC50) to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity that is less than 100 nM (IC50) to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity of about 50 nM to 500 nM IC50 to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity that is less than 50 nM (IC50) to an MHC molecule. In some embodiments, the antigenic peptide has a binding affinity of about 1 nM to 50 nM IC50 to an MHC molecule.

In some embodiments, the antigenic peptide has a binding affinity that is less than 5000 nM (IC50) to a cognate TCR molecule. In some embodiments, the antigenic peptide has a binding affinity of about 500 nM to 5000 nM (IC50) to a cognate TCR molecule. In some embodiments, the antigenic peptide has a binding affinity of about 50 nM to 500 nM IC50 to a cognate TCR molecule. In some embodiments, the antigenic peptide has a binding affinity of about 1 nM to 50 nM IC50 to a cognate TCR molecule.

In some embodiments, the antigenic peptide is selected based upon its hydrophobic status. In some embodiments, the antigenic peptide is hydrophobic. In some embodiments, the antigenic peptide has a high content of aromatic residues. In some embodiments, the antigenic peptide has at least about 10%, 20%, 30%, or 40%aromatic residues.

In some embodiments, the antigenic peptide has a binding affinity of about 1nM to about 5000 nM (e.g., about 1nM to about 50 nM, about 50 nM to about 500 nM, about 500 nM to about 5000 nM) to an MHC molecule, a binding affinity of about 1nM to about 5000 nM (e.g., about 1nM to about 50 nM, about 50 nM to about 500 nM, about 500 nM to about 5000 nM) to a cognate TCR molecule, a mutation relative to a wildtype peptide, optionally at the third amino acid position counting from the N-terminus, is hydrophobic, and has high content of aromatic residues.

In some embodiments, the antigenic peptide has low immunogenicity. Immunogenicity of the antigenic peptide can be predicted by algorithm developed for this purpose. See e.g., Riley et al., Front Immunol. 2019 Aug 28; 10: 2047; e.g., Schmidt et al., Cell Rep Med. 2021 Feb 6; 2 (2) : 100194.

In some embodiments, the antigenic peptide may be flanked by universal sequences or portions thereof. In some embodiments, the universal sequences or portions allow for rapid, high throughput methods for replacing or inserting the antigenic peptide encoding nucleotide in the polynucleotide MHC template.

In some embodiments, the antigenic peptides further comprise a unique defined barcode sequence operably associated with the identity of each distinct polypeptide. In some embodiments, the unique defined barcodes provide an antigen-specific sequence for identification during the analysis of the immune cell. See e.g., Peng et al., Cell Rep. 2019 Sep 3; 28 (10) : 2728-2738. e7.

In some embodiments, the peptide is associated with a cancer or tumor antigen. In some embodiments, the peptide is associated with a neoantigen.

In some embodiments, the peptide is associated with an antigen of a pathogen. In some embodiments, the pathogen is a virus, bacteria of fungus.

In some embodiments, the peptide is associated with an autoantigen.Neoantigen peptides

Neoantigens can be formed via various mechanisms. Non-synonymous somatic mutations, which can alter amino acid coding sequences, are the main cause of neo-epitopes. Except for somatic non-synonymous protein-altering mutations, tumor neoantigens can be generated from alternative splicing variations. Multiple computational methods and databases have been developed to identify alternative splicing events from RNA-seq data, such as SplAdder and CancerSplicingQTL2. See e.g., Kahles et al., Bioinformatics 32 1840–1847; Tian et al., Nucleic Acids Res. 47 D909–D916. A computational strategy was developed to identify neoepitopes generated from intron retention events in tumor transcriptomes and confirmed that these neoepitopes were processed and presented on MHC-I. See Smart et al. Nat. Biotechnol. 36 1056–1058.EXAMPLES

The examples below are intended to be purely exemplary of the invention and should therefore not be considered to limit the invention in any way. The following examples and detailed description are offered by way of illustration and not by way of limitation.Example 1. Preparation of the MHC-peptide library complexes.

Materials: The following neoantigen libraries were synthesized by Nanjing Peptide Biotech Ltd. The Kras 20-peptide neoantigen library included Kras G12V (KV1-KV20) and Kras G12R (KR1) . The Kras 50-peptide neoantigen library includes Kras G12V (KV1-KV20) , Kras G12R (KR1-KR20) , and Kras G12D (KD1-KV11) . See FIG. 1. The purity of both neoantigen libraries were above 95%. MHC heavy chain (HLA-A*02: 01) containing an His-tag and MHC light chain (hβ2m) were expressed and purified in Escherichia coli.

Formation of MHC-peptide monomer: In this experiment, the MHC-peptide monomer was formed following the preparation method described in Example 1, with the modification that HLA-A*02: 01 and human β2-microglobulin (hβ2m) were used as the MHC heavy and light chains, respectively. As shown in the non-reducing SDS-PAGE results in FIG. 2A, either a Kras 20-peptide library or a Kras 50-peptide library was mixed with the HLA-A*02: 01 and the hβ2m to form MHC-peptide monomer. Lane #2 showed the His-HLA-A*02: 01 in complex with the Kras 20-peptide library. Lane #3 showed the His-HLA-A*02: 01 in complex with the Kras 20-peptide library. Multiple bands were observed in the approximately 60-70 kDa range, confirming the presence of various peptides, which was attributed to the variation in molecular weight of the plurality of distinct peptides in the library.

Formation of MHC-peptide tetramer: In this experiment, the MHC-peptide monomer was formed following the preparation method described in Example 1, with the modification that HLA-A*02: 01 and human β2-microglobulin (hβ2m) were used as the MHC heavy and light chains, respectively. The biotinylation of the carried out following the suggested protocol in this reference (Altman, J.D., and Davis, M.M. 2016. MHC-peptide tetramers to visualize antigen-specific T cells. Curr. Protoc. Immunol. 115: 17.3.1-17.3.44) . Subsequently, the MHC-peptide tetramer was formed by mixing in streptavidin. As shown in the non-reducing SDS-PAGE results in FIG. 2B, higher molecular weight bands above 70 kDa were observed in the presence of added streptavidin, as shown in Lane #2 and #4. In contrast, bands were only observed in the approximately 60-70 kDa range in the absence of added streptavidin. The MHC-peptide tetramers were formed regardless of either a Kras 20-peptide library or a Kras 50-peptide library was used.

Optimization of MHC to peptide ratio in the formation of MHC-peptide monomer and tetramer: A QC method was developed to identify bound peptides in the MHC-peptide complexes. To establish the detection limit for this QC method, a series dilution of the Kras 50-peptide library was prepared as shown in FIG. 2C, ranging in 100 μg, 10 μg, 1 μg, 0.1 μg, and 0.01 μg. The serially diluted peptides were then desalted and freeze-dried to remove interfering salts.

FIG. 2D showed the formation of the MHC-peptide monomer in the presence of a reducing amount of the Kras 50-peptide library of 50 peptides (specifically 5 mg, 1.5 mg, 500 ug, and 10 ug) in a 50 mL refolding system, corresponding to 100 mg/L, 30 mg/L, 10 mg/L, 0.2 mg/L conc. of peptides) . As demonstrated in the SDS-PAGE, 10 μg of neoantigen was not sufficient to form the MHC-peptide monomer as indicated by the dominant presence of the lower band (e.g. HLA-A*02: 01 in complex with human β2-microglobulin) , while 500 μg was adequate as indicated by the dominant presence of the higher band (e.g. HLA-A*02: 01 in complex with human β2-microglobulin and Kras 50-peptide library) . Additionally, increasing the amount of neoantigens (such as to 5 mg or 1.5 mg) did not further improve the yield of the MHC-peptide monomer in the 50 mL refolding system as indicated by the unchanging ratio between the higher band and the lower band when the peptide was increased from 500 μg to 1.5 mg and 5.0 mg.

FIG. 2E showed a table for the detection rates of neoantigens loaded on the MHC-peptide monomer. In the control, where 1 μg of the Kras 50-peptide library was directly loaded onto the LC-MS/MS for analysis, 41 out of the 51 peptides in the library were identified, thus validating the LC-MS/MS method. In test samples 2, 3, and 4, where 5 mg, 1.5 mg, and 500 μg of the Kras 50-peptide library were allowed to form MHC-peptide complexes and then subjected to LC-MS/MS analysis, 44, 44, and 43 out of the 51 peptides in the Kras 50-peptide library were identified, respectively. As shown in the table, 500 ug is enough to form ≥85%MHC-p monomer in 50mL refolding system.

After refolding under the described in Example 1 and SEC that purifies MHC-p complexes and removes excessive biotin, the detection rate of neoantigens that were loaded on MHC-p complexes was evaluated by LC-MS/MS. More than 70%of the neoantigen peptides were detected.

FIG. 2F further showed the formation of the MHC-peptide monomer and the MHC-peptide tetramer in the presence of streptavidin. Lane #5 showed the presence of the MHC-peptide monomer in the absence of streptavidin. Increasing the amount of MHC-peptide monomer to streptavidin ratio in the order of Lanes 4, 3, and 2 promoted the formation of MHC-peptide tetramers, as indicated by the presence of bands of progressively higher molecular weights. The presence of the MHC-peptide monomer in Lane 2 indicates an excess amount of MHC-peptide monomer in the binding mixture, and all four binding sites on the streptavidin have already been saturated with MHC-peptide monomer, thus preventing the formation of more. In contrast, without biotin attached to the MHC-peptide monomer, the MHC-peptide tetramer could not be formed with streptavidin, as shown in Lane #9. Lanes 6, 7, 8 were various controls, as biotinylated MHC-peptide monomer, HLA-A*02: 01 in complex with human β2-microglobulin MHC complex, and non-biotinylated MHC-peptide monomer respectively.

Collectively, FIGs. 2A-2F demonstrated that the MHC-peptide complexes, including both the MHC-peptide monomer and the MHC-peptide tetramer, can be formed from mixtures of a broader range peptides in one pot, therefore allowing scaling the assembly of heterogenous MHC-peptide molecules efficiently and effectively.Example 2. tNET, a novel liquid biopsy platform that isolates and analyzes tumor engaged CD8⁺ T cells for early cancer detection

This Example describes a novel liquid biopsy platform ( “tNET” ) that isolates and analyzes tumor engaged CD8⁺ T cells for early cancer detection. Tumor neoantigens, recognized as “non-self” by specific T cell receptors, play a key role in antitumor immunity. tNET integrates synthetic tumor-specific neoantigens (TSNAs) with a high-throughput system to enrich TSNA-engaged CD8⁺ T cells (TNETs) from blood samples, providing direct evidence of malignancies. Validated in vitro and through xenograft models, tNET demonstrated significant enrichment of TNETs across cancer stages. In 107 pancreatic ductal adenocarcinoma (PDAC) patients and 87 healthy controls, tNET achieved 83%sensitivity and 100%specificity. When combined with CA19-9 analysis, early detection sensitivity increased to ≥94.33%with 98.72%specificity. Unlike conventional liquid biopsies, tNET captures live tumor-engaged immune cells, offering a highly sensitive, cost-effective, and accessible approach to early cancer screening. Thus, tNET warrants a new paradigm shift in early cancer screening and diagnostics, paving a new way for timely interventions and future improvements of cancer patient outcomes.

Tumor neoantigens trigger antitumor immunity by activating tumor-specific CD8⁺ T cells, which the tNET platform isolates and analyzes through synthetic neoantigen-based enrichment for early cancer detection. This approach directly captures live tumor-engaged immune cells, offering superior sensitivity and specificity compared to conventional liquid biopsies, particularly when combined with existing biomarkers. By enabling noninvasive, cost-effective screening, tNET represents a transformative paradigm in cancer diagnostics with broad clinical potential.

This example presents a first-in-class methodology and platform for early cancer detection, grounded in the principles of cancer immunosurveillance. By harnessing the immune system’s inherent capabilities, this approach aims to overcome the limitations of current liquid biopsy techniques. The immune-based strategy exploits the natural recognition and amplification of cancer signals by the immune system, enabling the detection of early-stage tumors with greater sensitivity and specificity. The methodology and platform are summarized in FIG. 3A.Results:Overview of tNET: A Liquid Biopsy Platform Targeting Tumor-Engaged T Cells (TNETs) for Cancer Diagnosis

TSNAs presented by MHC-I complex on tumor cells attract circulating TNETs. These TNETs recognize and bind to the TSNA–MHC-I complex via their TCRs triggering cytotoxic activity against the tumor cells. The highly specific interaction between TCRs and the TSNA–MHC-I complex directs TNETs to detect and attack tumor cells (FIG. 3B, center circle) .

Capturing and analyzing these live TNETs could serve as bona fide witnesses of tumor cells, and provide compelling evidence of cancer presence. By monitoring the immune response to TSNAs-specifically through isolating and identifying TNETs-cancers may be detected at the earliest stages, even before they manifest clinically or release detectable levels of tumor-derived biomarkers.

tNET was developed as a novel liquid biopsy platform designed to capture and analyze TNETs as reliable indicators of cancer presence (FIG. 3A) . The platform comprises four key components:1. A synthetic TSNA^peptide library presented by engineered MHC-I complex (TSNA^PeptideMHC-IC) to selectively capture TNETs. Given the diversity of TSNAs and HLA subtypes, constructing a comprehensive and stably assembled biotinylated TSNA^PeptideMHC-IC (TSNA^PeptideMHC-IC^Biotin) library was essential for high sensitivity and specificity of the tNET platform.2. A quick and efficient program to isolate and preserve peripheral blood mononuclear cells (PBMCs) , which are critically important to ensure high quality source of live TNETs interacting with a library of TSNA^PeptideMHC-IC^Biotin.3. A high-throughput TSNA^PeptideMHC-IC^Biotin loading system to capture diverse populations of TNETs. 4. A fast and cost-effective detection method to assess TNETs positivity and enrichment, indicating the presence of TSNA-bearing tumors.

To dissect and demonstrate the critical role of each component of the tNET platform, the following study first investigated if the synthesized TSNA in complex with matched recombinant MHC-I can specifically capture the TSNA-engaged peripheral T cells. The TSNA^PeptideMHC-IC^Biotin tetramers strategy was updated to Kras mutations since Kras mutations are the most frequently mutated type with high immunogenicity in many types of deadly cancer, particularly in pancreatic ductal adenocarcinoma (PDAC) and intrahepatic cholangiocarcinoma (ICC) . As exemplified in FIGS. 4A-4B, Avi-tagged recombined H2K^b and β2-microglobulin (B2M) subunits of MHC-I complex (MHC-IC) were purified (FIG. 4A and 4B) and refolded with each of KRAS^G12D or KRAS^G12V mutation-derived TSNA^peptide in vitro to produce Kras^G12D/VMHC-IC capturing monomer. Each of Kras ^G12D/V MHC-IC tetramers were prepared following the saturated binding of monomers with PE-conjugated streptavidin (FIG. 4C, the first step) . The tetramers were separately incubated with the PBMCs isolated from stage I/II PDAC patients with matched HLA types and matched KRAS mutations, or the WT controls. The flow cytometry analysis showed KRAS^G12V TNETs and KRAS^G12D TNETs were specifically isolated and significantly enriched compared to KRAS wild-type patients (FIG. 3B and 3C, and FIG. 4C) .

For proof-of-concept, a set of HLA subtype and mutant TSNA-matched blood samples from patients with ICC, colorectal cancer (CRC) , PDAC, and Liver metastasis of pancreatic ductal adenocarcinoma (LM-PDAC) were selected, and an experiment was carried out to examine the generality, feasibility and viability of the tNET platform (experimental details will be described later) . The results demonstrate significant enrichment of TNETs across all cancer types and stages, indicating the presence of cancer, whereas the patient samples with no matched TSNA mutation, or HLA subtype show negative results (Table 1) . These findings suggest that the tNET platform offers a simple, quick, and sensitive approach to track the presence of various cancers, holding significant potential for early cancer detection.Table 1: Detailed information on multiple biomarkers for detecting cancer. Circulating TNETs are specific and sensitive indicators of nascent tumor initiation and progression

TNETs in peripheral blood exhibit dynamic changes at various time points during tumor initiation and progression post-surgery (15) . It was investigated whether TNETs are sensitive and specific indicators of nascent tumor initiation, and the presence of TNETs can be plotted during the time course of tumor development. To address these questions, a xenografted tumor model with Kras mutations was established since Kras mutations are validated TSNAs with high immunogenicity. To do so, a new PDAC cell line, Pan02-Neo, was created, which overexpresses six tandem Kras mutation-derived neoantigens (Kras^{G12V/D/R/C/I/A}) linked by P2A/T2A elements (FIG. 5A) . These Kras mutation-derived neoantigens were successfully expressed in Pan02-Neo cells (FIG. 5B) .

The TSNA^peptideMHC-IC tetramers designed based on gene mutations identified through whole-exome sequencing can isolate 0.4%to 0.002%of live TNETs from the peripheral blood of patients with melanoma (16) . To confirm that mice inoculated with Pan02-Neo tumor cells indeed produce Kras-specific neoantigens and consequently activate the TNETs’ response, the Kras^{G12V/D/R/C/I/A}MHC-IC tetramers strategy was employed for examination. Avi-tagged recombined H2K^b and B2M subunits of MHC-I were purified and refolded with one of each Kras ^{G12V/D/R/C/I/A}mutation-derived neoantigen in vitro to produce Kras^{G12V/D/R/C/I/A}MHC-IC monomers, and each of Kras ^{G12V/D/R/C/I/A}MHC-I tetramers were prepared following the saturated binding of monomers with PE-conjugated streptavidin (FIG. 5C and 5D) . Overall, a Kras^{G12V/D/R/C/I/A}MHC-IC tetramer library comprising sixty neoantigen peptides derived from Kras^{G12V/D/R/C/I/A}mutations was prepared as a pool of baits to capture the populations of circulating Kras ^{G12V/D/R/C/I/A}-engaged CD8⁺ T cells (KNETs) (FIG. 6A) .

Next, KNETs activation by tumor-produced Kras^{G12V/D/R/C/I/A}neoantigens was investigated and used to monitor the dynamic changes of the KNETs. Kras^{G12V/D/R/C/I/A}MHC-IC tetramer library was incubated with PBMCs isolated at different time points from the C57BL/6J mice which were subcutaneously xenografted with Pan02-Neo tumor cells, then processed and examined by flow cytometry analysis. A distinct KNETs subpopulation was detectable by day 4, when tumors were ～1 mm in diameter (FIG. 6B and 6C) . In a time course study, KNETs subpopulation peak early and gradually declined by day 36 post-tumor challenge, nevertheless, they are continuously detectable by day 69 post-tumor challenge (FIG. 6B and 6D) . Likewise, intravenous injection of Pan02-Neo cells to C57BL/6J mice at equal or lower doses of the tumor cells than that of subcutaneous model showed similar results (FIG. 6E-6G) . Interestingly, it was observed that fewer injected cells correlated with a lower KNETs population, faster decline in a time course, showing nearly undetectable level of KNET at day 77 (FIG. 6G, green line, and FIG. 5E) . These results suggest that the tumor cells are more quickly and efficiently detected and eliminated by the activated KNETs when tumor cells are directly injected into the circulating blood environment. Significantly, the second-round challenge with much fewer Pan02-Neo tumor cells into the C57BL/6J mice at the time point in which KNET was nearly undetectable from the first round, triggered a quick and robust KNETs increase (FIG. 6H) . Taken together, these results suggest that circulating TNETs not only sensitively detect tumor initiation and progression, but also are capable of sensitively and specifically monitoring the new rising of the minimal residual disease (MRD) after second round of tumor cell challenge.Establishment of a high-capacity and high-throughput system for loading a library of TSNA^PeptideMHC-IC^Biotin and Capturing TNETs

Although binding affinity of TSNA^PeptideMHC-IC tetramers to T-cell receptors and their enhanced TNETs capturing ability compared to TSNA^PeptideMHC-IC monomers is a proven advantage of the tetramer strategy, there are limitations of this approach. A major drawback is that each of TSNA^PeptideMHC-IC tetramers has to be individually made for different mutations. Thus, it is labor-intensive and impedes for generating a large library with high-capacity loading of TSNA^PeptideMHC-IC and high-throughput capturing of TNETs. Thus, increasing the binding sites of each TSNA^PeptideMHC-IC monomer on the same spherical surface of a carrier vehicle, such as a streptavidin-conjugated magnetic bead, could improve the low affinity and stabile interaction among a wide range of TSNAs to MHC-I molecule, and TCR of TNETs.

To test this hypothesis, streptavidin-conjugated magnetic beads were used to conjugate TSNA^PeptideMHC-IC^Biotin monomers, producing TSNA^PeptideMHC-IC^Biotin-beads hetero-multimers with significantly more TCR binding sites (FIG. 7A) . To assess the capability and affinity of ^PeptideMHC-IC^Biotin hetero-multimer beads binding to the antigen-engaged CD8⁺ T cells in vitro, OVA^peptideMHC-IC^Biotin and Kras^G12VMHC-IC^Biotin monomers were assembled and purified by size-exclusion chromatography (SEC) (FIG. 7B) . The SEC purified ^PeptideMHC-IC^Biotin monomers were then loaded onto streptavidin-conjugated magnetic beads to form ^PeptideMHC-IC^Biotin hetero-multimer on beads (FIG. 7C and 7D) . The beads were then incubated with PBMCs isolated from mice pre-stimulated with OVA or Kras^G12V antigens. Microscopy analysis revealed significantly more NETs aggregates bound to OVA or Kras^G12V beads compared to controls (FIG. 7E and 7F) , indicating the success of ^PeptideMHC-IC^Biotin hetero-multimer beads as an effective tool for capturing circulating TNETs.Development of an ELISA reporting Assay that seamlessly couples to the TNETs Capture system and efficiently detecting of TNETs

To quantitatively measure the efficacy, sensitivity, and specificity of the above beads-based TNETs capturing system, eloped a high-throughput ELISA method that seamlessly couples the capture/isolation/enrichment step to final TNETs detection step was also developed. With this development, tNET integrates all following experimental procedures, –PMBC preparing/beads priming, Beads-based TNETs-capture and purification, and an anti-CD8 antibody coupled ELISA assay for detecting TNETs in one 96-well plate, providing a quantitative and straightforward readout of circulating TNETs (FIG. 8A) . First, the performance of this tNET system was investigated in animal models stimulated by antigens-Ovalbumin (OVA) or a sub-library of Kras^G12V neoantigens. OVA^peptidesMHC-IC^Biotin monomers and a sub-library of 20 types of Kras^G12VMHC-IC^Biotin monomers were assembled on streptavidin-conjugated magnetic beads followed by anti-CD8 antibody-based ELISA assay. It was found that the ^OVA PBMCs: ^OVA beads and Kras^G12V PBMCs: Kras^G12V beads groups exhibited a significantly higher optical density (OD) readout at 450 nm compared to the ^OVAPBMC: ^Ctr beads, Kras^G12VPBMC: ^Ctr beads, and ^Ctr PBMCs: ^Ctr beads groups (FIG. 8B) .

To investigate the sensitivity and specificity of this tNET platform for capturing and detecting TNETs activated by endogenous genetic mutations ex vivo, an orthotopic pancreatic tumor model using the KPC-Luciferase cell line derived from primary murine PDAC (Kras^LSL-G12D/+/Trp53^LSL- ^R172H/+/Pdx-1 Cre) on C57BL/6 strains was established. The Kras^G12D mutation and luciferase reporter activity were confirmed in the KPC-Luciferase cell line (FIGS. 9A and 9B) . The xenografted mouse PDAC tumors are detected and confirmed by two imaging methods, IVIS Lumina K and MRI, and further verified by anatomical dissection (FIG. 8C, the first three panels, and FIGS. 9C-9E) .

To perform tNET test, PBMCs and corresponding beads-conjugated hetero-multimeric libraries of Kras^G12DMHC-IC^Biotin and Trp53^R172HMHC-IC^Biotin were prepared as described (FIG. 8A, model diagram, and Table 2) . Compared with traditional methods that rely on repeated ultrafiltration for buffer exchange and concentration (for example with at least a 1: 200-1: 1000 concentration) , weakly bound polypeptides have great difficulty in binding to the MHC complex, which poses challenges for the translation of this technology. Such methods were not able to recover the weakly bound MHC-peptide complexes (e.g., with the peptides in the KRAS gene) . The present purification strategy utilizes a gentle ultrafiltration (with about 1: 50 concentration) followed by size exclusion chromatography (SEC) without a buffer exchange to achieve excellent separation results. After the renaturation and concentration processes, the SEC purification can effectively remove polymers and HLA-B2M complexes associated with unbound polypeptides. Subsequently, after the BriA-catalyzed reaction (i.e., biotinylation) , a second SEC purification stage selectively removes excessive free biotin. This approach enables the efficient retention of weakly bound polypeptides derived from gene mutations, such as those in the KRAS gene while provides an effective purification. The present method provides a more streamlined and effective solution, ensuring the preservation of these weakly bound peptide complexes.

TNETs were captured by prepared beads and followed by tENTs detection. The tNET results showed a strong TNETs signal both by flow cytometry and ELISA (FIG. 8C, the fourth and fifth panel, respectively) analysis even when the tumor size was merely ～ 1 mm in diameter (FIG. 8C, the third panel and FIG. 9E) , which was undetectable by MRI (FIG. 10C, the second panel and FIG. 9C) , whereas could only be detected by the enhanced IVIS Lumina K imaging (FIG. 8C, the first panel and FIG. 9D) . To determine the minimal tumor cell number for TNETs detection and the dynamic level of the circulating TNETs in peripheral blood, mice were orthotopically challenged with gradient numbers of KPCs cells and analyzed in a time course. The results demonstrated that in the orthotopic pancreatic cancer model, higher inoculated cell numbers correlated with earlier detection of TNETs (tumor-associated neutrophil extracellular traps) , with significant signals detectable as early as day 7 at a seeding density of 1×10⁴ cells (FIG. 9F) . The Kras^G12D&Trp53^R172H specific TNETs were detectable by day 4 post-challenge, significantly elevated compared to controls by day 8, and continuously increasing and maintained at high levels up to day 24 (FIG. 8D and 8E) . The presence of the tumors was examined by anatomical resection and confirmed by H&E staining (FIG. 8F) .Table 2: The neoantigen sequences derived from the KrasG12D and Trp53R172H mutations.10 neoantigen sequences are derived from the Kras^G12D mutation, and 4 neoantigen sequences are derived from the Trp53^R172H mutation. These sequences were obtained through the prediction of binding affinity and bioinformatics prediction of antigenicity, and were synthesized and used for assembling the library.

To further verify the feasibility and viability of tNET, humanized immune system of mice (HuHSC-C-NKG-proF) were generated by transplanting HLA-A*02: 01 human hematopoietic stem cells into immunodeficient C-NKG mice (NOD/Shi-Prkdc^scidIl2rg^em1) . Over 50%of CD45⁺ cells were successfully reconstituted, including 30.52%T cells, 59.50%B cells, and 9.49%monocytes, plasmacytoid dendritic cells (pDCs) , and natural killer (NK) (FIG. 9G) . A patient-derived xenograft was then established in the humanized HuHSC-NKG-proF mice using a patient derived KRAS^G12D tumor samples (HLA-A*02: 01) . Human KRAS^G12D TNETs were detected by tNET as early as four days post-xenograft (FIG. 8G) , and the functionality of KRAS^G12D TNETs was verified in vitro by ELISpot (FIG. 8H) .

Based on the positive results of the above murine model, tNET's diagnostic performance using clinical samples was assessed. Using a similar approach as described above, PBMCs samples obtained from 10 different patients across all tumor stages (50%of stage I and 50%of stage II/III/IV) were analyzed using corresponding magnetic beads conjugated with matched mutations and HLA subtypes. The tNET enables to detect TNETs specifically activated by both common (KRAS, TP53) and rare (PDGFRA p. D1071N) PDAC mutations (Table 3) .Table 3: Detailed information on multiple biomarkers for PDAC

Taken together, a novel blood-based liquid biopsy platform, tNET, was developed. The above arrays of animal and human sample experiments demonstrate that tNET is a feasible, reliable, and high-throughput approach by which deadly malignancy such as PDAC can be sensitively and accurately detected at its earliest stage, paving a new way for its wide clinical application in early cancer detection.Generation of the First Human TSNA^PeptideMHC-IC^Biotin Library of tNET Platform for Efficient and Accurate Detection of Early PDAC: tNET 1.0

Using HLA-matched ^peptideMHC-IC^Biotin hetero-multimer libraries covering a specific Kras mutation of murine or human patients, it was shown that detection of TNETs in peripheral blood is likely the best testimony of the presence of cancer in body (FIGs. 6 and 8) . Nonetheless, the highly individualized, hundreds or even thousands of mutation profiles of a specific type (s) of cancer, together with extensive HLA polymorphism of various cancer patients pose significant challenges for TNETs detection in broad clinical practice. A sensitive and high-throughput TNETs detection platform empowered by a comprehensive TSNA^PeptideMHC-IC^Biotin hetero-multimer library, which covers all possible HLA subtypes and major neoantigens of a specific type (s) cancer, would theoretically overcome this key obstacle. However, it raises a technical challenge as to synthesizing a large-scale TSNA^PeptideMHC-IC^Biotin monomer library for a wide range of the diverse neoantigens since it would be extremely labor intensive and not practically cost-effective.

To solve the above issue, a high-throughput refolding system was developed for making a TSNA^PeptideMHC-IC^Biotin hetero-multimer library for capturing TNETs. Firstly, based on immunogenicity, prevalence and NetMHCpan 4.0 analyses, a home-made AI-based TSNA Discovery Algorithm Program for neoantigen data mining and discovery was built (FIG. 11A) . Through a rigorous multi-step filtering process, a total of 3, 266 potential TSNAs derived from CRC, ICC, and PDAC were predicted and prioritized. As exemplified by establishing the PDAC Library (grouped by major HLA subtypes) , 1, 535 stable binding TSNA candidates were selected, including 344 HLA-A02: 01, 526 HLA-A03: 01, 356 HLA-A11: 01, and 309 HLA-A24: 02. These TSNA peptides are synthesized for in vitro refolding (FIGs. 10A and 11B) . These neoantigens encompassed 89 mutated genes and 385 mutation types, evenly distributed across HLA-A types (Tables 4A-4B) .Table 4A: The information of mutation genes and mutation sites of the neoantigen library of PDAC –Information of mutation genes. Table 4B: The information of mutation genes and mutation sites of the neoantigen library of PDAC -Information of mutation gene sites Based on the bioinformatics prediction and the library screening method mentioned in the method, 1535 neoantigens were designed and synthesized. Table 4A: The information of mutation genes and the number of neoantigens corresponding to these mutation genes. Table 4B: The information of mutation sites and the number of neoantigens corresponding to these mutation sites.

Next, a new TSNA^PeptideMHC-IC^Biotin renature/refolding system was invented, which overcomes limitations of traditional single-neoantigen refolding technique (FIG. 12A) . The new system enables concurrent assembly of multiple TSNA^PeptideMHC-IC^Biotin monomers in one reaction vessel (hereafter named as “one-pot” ) , greatly increasing library construction throughput and efficacy (FIGs. 12B-12F) (Experimental procedures and details are described in Supplemental Information and Methods) . To determine the efficacy, viability, and quality of “one-pot” system for library construction, the TSNA peptides coverage of the concurrent assembly of TSNA^PeptideMHC-IC^Biotin monomers was quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS) . The LC-MS/MS analysis revealed that 936 out of 1, 535 TSNA peptides (61%coverage) are easily detected across all four types of HLA in the library of heterogeneous TSNA^PeptideMHC-IC^Biotin monomers made by the “one-pot” system (FIG. 12B) .

Drawing upon established reference materials, our initial attempts to generate distinct MHC-complex bands proved unsuccessful. Specifically, under experimental conditions where the molar ratio of HLA to β2m was fixed at 3: 2 and equivalent quantities of HLA were sequentially introduced across three reaction stages, the resultant electrophoresis profiles failed to exhibit clear, discernible bands. In response, a series of strategic modifications were implemented: adjusting the HLA: β2m ratio to 1: 1 and halving the HLA dosage in the second and third reaction cycles relative to the initial input. These adjustments led to a significant improvement, enabling the consistent visualization of well-defined MHC-complex bands, thereby facilitating more accurate subsequent operations.

Overall, the library covers about 85.2%of the total input mutations selected from PDAC (FIG. 10C, upper panel) . Furthermore, the prepared heterogeneous TSNA^PeptideMHC-IC^Biotin monomer library remained stable for at least one year at -80℃ (FIG. 13) .

Given the high mutation rates of the KRAS (93.0%) and TP53 (63.0%) genes in PDAC (17, 18) , the coverage of KRAS mutation-and Trp53 mutation-derived heterogeneous TSNA^PeptideMHC-IC^Biotin monomers was measured using LC-MS/MS analysis. A total of 95% (19/20) of KRAS mutation types and 88.4% (107/121) of Trp53 mutations had at least one TSNA candidate present in the biotinylated monomer library (Tables 5A-5C and 6A-6C) . To assess whether TSNAs presented in TSNA^PeptideMHC-IC^Biotin monomers are indeed functional and exhibited relatively high immunogenicity, an ELISpot assay was conducted. Ten KRAS-derived TSNAs and four Trp53-derived TNSAs were selected from the library (Table 7) . The ELISpot assay results showed that the LC-MS/MS-detected TSNAs exhibit stronger immunogenicity compared to the LC-MS/MS-undetected TSNAs. This suggests that the “one-pot” refolding system not only overcomes the limitation of traditional one-by-one TSNA^PeptideMHC-IC refolding strategy, but also demonstrates its advantage in effectively selecting immunogenic TSNAs predicted by bioinformatics (FIG. 10D) .Table 5A-5C: The information about the design of KRAS mutation sites and the library detection in the PDAC neoantigen libraryTable 5A: The information of the sites derived from KRAS mutations and the corresponding numbers of neoantigens. Table 5B: The information of KRAS mutation sites detected by LC-MS/MS and the corresponding numbers of neoantigens. Table 5C: The information of KRAS mutation sites undetectable by LC-MS/MS and the corresponding numbers of neoantigens. Based on the bioinformatics prediction and the library screening method mentioned in the methodology, 342 neoantigens derived from KRAS were designed and synthesized, among which 183 neoantigens could be detected by LC-MS after the subsequent library construction. Table 5A: The information of KRAS gene mutation sites and the number of neoantigens corresponding to these mutation sites. Table 5B: The information of KRAS gene mutation sites detected by LC-MS/MS and the number of neoantigens corresponding to these mutation sites. Table 5C: The information of KRAS gene mutation sites not detected by LC-MS/MS and the number of neoantigens corresponding to these mutation sites.Table 6A-6C: The information about the design of TP53 mutation sites and the library detection in the PDAC neoantigen library.Table 6A: The information of the sites derived from TP53 mutations and the corresponding numbers of neoantigens. Table 6B: The information of TP53 mutation sites detected by LC-MS/MS and the corresponding numbers of neoantigens. Table 6C: The information of TP53 mutation sites undetectable by LC-MS/MS and the corresponding numbers of neoantigens. Based on the bioinformatics prediction and the library screening method mentioned in the method, 402 neoantigens derived from TP53 were designed and synthesized, among which 263 neoantigens could be detected by LC-MS after the subsequent library construction. Table 6A: The information of TP53 gene mutation sites and the number of neoantigens corresponding to these mutation sites. Table 6B:The information of TP53 gene mutation sites detected by LC-MS/MS and the number of neoantigens corresponding to these mutation sites. Table 6C: The information of TP53 gene mutation sites not detected by LC-MS/MS and the number of neoantigens corresponding to these mutation sites.Table 7: Information on peptides detected and undetected by “One-pot” LC-MS/MS.10 neoantigen sequences are derived from the Kras^G12D mutation, and 4 neoantigen sequences are derived from the Trp53^R172H mutation. These sequences were obtained through the prediction of binding affinity and bioinformatics prediction of antigenicity, and were synthesized and used for assembling the library and further screening the affinity and antigenicity of neoantigens.

Finally, heterogeneous TSNA^PeptideMHC-IC^Biotin monomer library was incubated with streptavidin-conjugated magnetic beads. A total of 94.9%of the heterogeneous TSNA^PeptideMHC-IC^Biotin monomers were successfully loaded onto the beads, resulting in approximately 6.51 x 10¹³ MHC-I monomers evenly distributed across 1 mg of beads (FIG. 10C, lower panel, Table 8A-8B, and Method section of Loading quantity of monomer on beads) . In summary, this study developed a high-throughput MHC-IC monomer refolding approach and produced PDAC-specific beads-based heterogeneous TSNA^PeptideMHC-IC^Biotin multimers for capturing TNETs.Table 8A-8B: The information of the neoantigen library in the integrity detection.Table 8A: The information of neoantigens detected by LC-MS/MS from the TSNA^PeptideMHC-IC^Biotin Monomer. Table 8B: The information of neoantigens detected by LC-MS/MS from the TSNA^PeptideMHC-IC^Biotin Monomer-beads. To assess the efficiency of monomer loading onto magnetic beads, a library with 59 neoantigens was used for testing by loading onto SA beads. After LC -MS/MS detection, 56 neoantigens were detectable on the magnetic beads post -loading and washing. A. The information of neoantigens detected by LC-MS/MS from the TSNA^PeptideMHC-IC^Biotin Monomer. B. The information of neoantigens detected by LC-MS/MS from the TSNA^PeptideMHC-IC^Biotin Monomer-beads. tNET1.0 demonstrates High-throughput, High sensitivity and Specificity in Clinical Application

To evaluate the performance of the PDAC-specific library for detection, 107 PDAC patients and 78 healthy individuals were selected with matched HLA-A types and genetic mutations from a total of 210 PDAC patients and 185 healthy individuals, respectively. Circulating TNETs were detected by tNET with the PDAC-specific library. The area under the Receiver Operating Characteristic (ROC) curve analysis indicated that the area under the curve (AUC) for detection TNETs responsiveness was 0.863 (95%CI: 0.804-0.909) , with a sensitivity of 83.18% (95%CI: 74.7%-89.7%) and a specificity of 100% (95%CI: 95.4%-100.0%) (p < 0.001) (FIG. 14A, left panel) . Using PBMCs isolated from as few as 1ml blood, tNET exhibited superior detecting and predictive capability for stage I and II PDAC patients (FIG. 14A, Middle Panel) . In contrast to current liquid biopsy methods for cancer detection on market, it shows even better outcome compared to stage III and IV patients (FIG. 14A, right panels) . The tNET assay demonstrated robust diagnostic performance across different stages of PDAC, achieving a sensitivity of 83.91%for stage I & II patients (n=87) and 80%for stage III & IV patients (n=20) at 100%specificity (FIG. 14B) . Notably, when stratified by individual stages, tNET exhibited the best predictive power for stage II patients, with an AUC of 0.907 (p<0.001) , sensitivity of 89.47%, and specificity of 100% (n=38) (FIG. 15A and 15B) . Taken together, these findings highlight the exceptional capability of tNET for detecting early-stage PDAC (stage I & II) at which the disease is mostly treatable.

To estimate the possible application of the tNET in early-stage cancer screening using PBMCs derived from direct blood draw, without knowing HLA subtype and the mutation type (not tumor-informed) , prospective detection of PDAC with a mixed library of heterogeneous TSNA^PeptideMHC-IC^Biotin multimer beads (hereafter called Hybrit tNET or HtNET) was conducted. The HtNET involves the ratio-controlled mixture of the individually pre-prepared TSNA^PeptideMHC-IC^Biotin multimer beads to create a comprehensive HtNET library that covers the majority of PDAC TSNA^peptides in complex with four predominant HLA subtypes (HLA-A02: 01, HLA-A03: 01, HLA-A24: 02, and HLA-A11: 01) . This integrated hybrid tNET system synergistically combines multiple HLA-specific TSNA multimers within a single analytical platform, thereby achieving two critical objectives: (1) extensive population coverage through multi-HLA repertoire integration, and (2) preservation of high-fidelity and high capacity of neoantigen recognition capabilities essential for early cancer detection and screening.

A total of 70 PDAC patients and 95 healthy individuals were enrolled, among whom 57 PDAC patients and 76 healthy individuals that had matched HLA types and mutation profiles were determined post-HtNET test. The area under the ROC curve for the HtNET detection responsiveness was 0.879 (95%CI: 0.811-0.929) , with a sensitivity of 75.44% (95%CI: 62.2%-85.9%) and a specificity of 100% (95%CI: 95.3%-100.0%) (p < 0.001) (FIG. 15C) . These results indicate that tNET-based cancer detection platform not only shows great sensitivity and specificity for retrospective analysis of cancer presence in tumor informed patient samples, but also demonstrates excellent performance on detecting the cancer presence of tumor-uninformed and HLA-unknown biological samples. Thus, tNET technology shows a great potential adapting to detect early cancer in population screening. To further assess the performance of HtNET in other cancers, 35 tumor patients were enrolled, including those with ICC (n=10) , CRC (n=10) , LM-CRC (n=10) , and LM-PDAC (n=5) . Among them, the patients with matched HLA-A and mutation types exhibited strong NETs signals in PBMCs isolated from as few as 1ml peripheral blood (Table 1) . HtNET demonstrated superior performance compared to traditional biomarkers, such as CA19-9, CEA, CA242, CA125, CA50, and CA72-4, in the diagnosis of these cancers (Table 1) . Notably, the signal from TNETs detected by tNET was associated with poorer progression-free survival (PFS) in PDAC samples, whereas no statistically significant difference in PFS was observed in the same cohorts when analyzed using the CA19-9 diagnostic threshold (FIG. 15D and 15E) . Furthermore, the diagnostic power of tNET for PDAC with that of ctDNA based cancer detection was compared. The findings indicated that tNET demonstrated superior predictive ability for both early-stage and advanced-stage PDAC patients, requiring a smaller volume of peripheral blood samples (Table 9) .Table 9: Compare the sensitivity performance of ctDNA and tNET

To further enhance the detection sensitivity of tNET in PDAC, CA19-9, a traditional biomarker used in PDAC diagnosis, was included in tNET analytical and diagnostic algorithms. tNET alone exhibited a much lower false negative rate in both overall cohort and in patients with stage I and II disease (FIG. 15F and 15G) . The area under the ROC curve for the tNET/CA19-9 combined diagnostic approach was 0.971 (95%CI: 0.936-0.990) , with a sensitivity of 95.33% (95%CI, 89.4%-98.5%) and a specificity of 98.72% (95%CI: 93.1%-100.0%) (p < 0.001) . The sensitivity of the combined approach is significantly better than that of tNET (AUC = 0.863, sensitivity = 83.18%) or CA19-9 alone (AUC = 0.885, sensitivity = 77.57%) . In terms of specificity, tNET still gives the best performance (specificity=100%) , while tNET/CA19-9 is slightly compromised (specificity=98.72%) (FIG. 15H) . Importantly, the combined diagnostic method demonstrated strong predictive power for stage I and II PDAC (FIG. 15I) . These findings suggest that tNET, paired with other low-cost traditional liquid biopsy biomarker such as CA19-9 can significantly increase its sensitivity without compromising much of its specificity.

In summary, the tNET platform, with its all-embracing library of heterogeneous TSNA^PeptideMHC-IC^Biotin multimers for capturing TNETs, not only support sensitive and specific multi-cancer early detection, but also holds great promise for transforming cancer diagnosis and monitoring cancer progression. In cases where traditional cancer biomarkers such as CA19-9 levels are inconclusive or where current liquid biopsy approaches such as ctDNA detection are limited due to low tumor-derived DNA content in early-stage cancers, tNET likely offer a first-in-class early cancer detection platform with more reliable outcome and superior sensitivity and specificity.Discussion

The emergence of liquid biopsy has noticeably advanced the field of clinical oncology, offering a new building lot for early detection and diagnosis of cancer, as well as continuous monitoring the disease (19) . Emerging tests such as Galleri, CancerSEEK, and PanSeer focus on cfDNA, ctDNA, methylation and somatic mutations profiles, often combined with biomarkers such as cancer metabolites and protein markers (20-24) . However, these tests rely on blood circulating tumor-derived components, therefore, retaining engrained limitations due to low abundance in early-stage tumors.

This Example reports a new blood based liquid biopsy platform, tNET. Unlike the other liquid biopsy approaches targeting tumor-derived components, tNET leverages TNETs that detect and eliminate tumor cells based on precise and specific recognition mechanism of the TCR towards TSNA^PeptideMHC-IC on tumor cell surface, enabling identification of cancerous cells at their earliest stages with natuarally heightened specificity. On the other hand, the immune system amplifies tumor initiation signals, as each tumor cell interacts with multiple immune cells. TNETs proliferate rapidly in response to stimuli, increasing their presence in peripheral blood. This high-fidelity amplification mechanism allows the tNET platform to detect early-stage tumors with exceptional sensitivity, even before tumor-derived substances are abundant in circulation. Together, these two innate antitumor immunity mechanisms ensure the key criteria for early cancer detection via liquid biopsy: high sensitivity and high specificity.

This study pioneers two disruptive innovations underpinning the tNET platform: (1) a functional TSNA discovery algorithm enabling comprehensive synthetic TSNA^PeptideMHC-IC^Biotin bait libraries, and (2) a high-throughput loading system for efficient TNETs capture. To make the library, the extensive polymorphism of HLA, mutations, and TCR repertoires necessitates a high-throughput TSNA^PeptideMHC-IC library preparation (25) . The traditional "one-by-one" single neoantigen/MHC-I monomer refolding approach is inefficient and economically prohibitive, limiting their scalability. While recent advances, such as high-throughput TCR discovery pipelines (26) and recombinant MHC-I technologies (27) , have improved neoantigen screening, they remain constrained by low yields and complex workflows. In contrast, the "one-pot" TSNA^PeptideMHC-IC system efficiently assembles hundreds of functional MHC-I monomers in a single reaction, overcoming library construction barriers. These innovative technologies are not only conceptually novel and well-grounded, but also disruptive to the current outlook of liquid biopsy for cancer detection.

The performance of tNET has been systematically examined and validated in four different murine models, and ultimately evaluated using more than 200 clinical samples derived from corresponding cancer patients, along with >150 healthy controls. The results demonstrate that the tNET is an operatable, cost-effective, minimally invasive platform, providing a straightforward readout for the early cancer detection of TSNA-bearing tumors with excellent sensitivity and specificity. Moreover, the tNET platform does not require sophisticate instruments, nor expensive reagents. As a proof-of-principle, the tNET1.0 for PDAC, has successfully demonstrate the feasibility, viability and generality of tNET in early cancer detection in clinical setting. Importantly, tNET1.0 also demonstrates robust cross-tumor-type detection capabilities when patient mutation profiles and HLA types align with the detection library. tNET is capable of detecting early-stage (asymptomatic) cancers at an ultra-low cost and with high sensitivity and specificity as long as the TSNAs library can be continuedly expanded to comprise all existing repertoire of TSNAs of any types of cancer.

Intriguingly, the tNET platform demonstrates stage-dependent sensitivity in PDAC detection, with peak performance in stage II patients, followed by stage I and reduced efficacy in advanced stages (III/IV) . This pattern likely reflects dynamic tumor-immune interactions, where stage II tumors exhibit maximal immunogenicity during active expansion, while early-stage tumors may be too immunologically "cold" and advanced tumors show T-cell exhaustion (28) and MHC-I downregulation (29-31) . The limited availability of stage III/IV surgical samples may also contribute to the observed sensitivity differences. These findings highlight how tumor progression dynamics influence tNET performance, confirming this approach's biological validity while offering critical guidance for clinical implementation-especially for early-stage detection where current methods fail.

The tNET platform's minimal blood requirement enables ultrasensitive and population-scale screening. While ctDNA and cfDNA tests typically necessitate 10-20 mL of peripheral blood to achieve reliable sensitivity and specificity (32, 33) , tNET achieves superior performance with only 1 mL, enabling the detection of TNETs and identification of early-stage tumors in patients. With the inclusion of a comprehensive tumor-specific neoantigen repertoire in the bait library, tNET will offer ultra-high sensitivity and specificity for early cancer screening with much less blood volume.

An exciting possible clinical implication of tNET is directly linking cancer early detection to future therapy of the detected disease. The study has demonstrated the efficacy of a "one-pot" refolding system for screening neoantigens with high binding affinity to MHC-I molecules and strong immunogenicity for TCR recognition. Previous studies have shown that generalized peptide vaccines targeting a narrow spectrum of gene mutations are largely ineffective (34) . Thus, the high immunogenic neoantigens identified through the high-throughput "one-pot" refolding system from vast TSNA repertoires could serve as a selected pool of promising candidates for TSNA-based vaccines to prevent cancer progression. Furthermore, tNET by its natural design has preserved its capability to pinpoint the specific TNETs interacting TSNAs, either by sequencing the TCR of the captured TNETs, or further purifying from the bait library by screening a refined, targeted and barcoded candidate TSNA sub-library. The identified and personalized TSNAs can then be used to develop TSNA-based personalized vaccines, or to generate TSNA-directed personalized CAR-T immunotherapy. This unique merit of tNET raises an exciting and valuable possibility: while a patient has detected or diagnosed cancer at its early stage, at the same time, these patients already have the potential solution for preventing or intervening the disease. This is extremely significant as the anxiety caused by the early cancer diagnosis is the major challenge and debate facing to current methods for cancer early detection and diagnosis.References and notes1. R.L. Siegel, K.D. Miller, H.E. Fuchs, A. Jemal, Cancer statistics, 2022. CA Cancer J Clin 72, 7-33 (2022) . doi. org/10.3322/caac. 21708. PMID: 350202042. K. Rangarajan et al., Ultra-high resolution, multi-scale, context-aware approach for detection of small cancers on mammography. Sci Rep 12, 11622 (2022) . doi. org/10.1038/s41598-022-15259-7. PMID: 358039853. C. Alix-Panabieres, K. Pantel, Liquid Biopsy: From Discovery to Clinical Application. Cancer Discov 11, 858-873 (2021) . doi. org/10.1158/2159-8290. CD-20-1311. PMID: 338111214. A.D. Singhi, E.J. Koay, S.T. Chari, A. Maitra, Early Detection of Pancreatic Cancer: Opportunities and Challenges. Gastroenterology 156, 2024-2040 (2019) . doi. org/10.1053/j. gastro. 2019.01.259. PMID: 307216645. Y.M.D. Lo, D.S.C. Han, P. Jiang, R.W.K. Chiu, Epigenetics, fragmentomics, and topology of cell-free DNA in liquid biopsies. Science 372, (2021) . doi. org/10.1126/science. aaw3616. PMID: 338330976. E.K.J. van der Poort, N.T. van Ravesteyn, J.J. van den Broek, H.J. de Koning, The Early Detection of Breast Cancer Using Liquid Biopsies: Model Estimates of the Benefits, Harms, and Costs. Cancers (Basel) 14, (2022) . doi. org/10.3390/cancers14122951. PMID: 357406157.N. Xie et al., Neoantigens: promising targets for cancer therapy. Signal Transduct Target Ther 8, 9 (2023) . doi. org/10.1038/s41392-022-01270-x. PMID: 366044318. M. Yarchoan, B.A. Johnson, 3rd, E.R. Lutz, D.A. Laheru, E.M. Jaffee, Targeting neoantigens to augment antitumour immunity. Nat Rev Cancer 17, 209-222 (2017) . doi. org/10.1038/nrc. 2016.154. PMID: 282338029. J. He et al., Defined tumor antigen-specific T cells potentiate personalized TCR-T cell therapy and prediction of immunotherapy response. Cell Res 32, 530-542 (2022) . doi. org/10.1038/s41422-022-00627-9. PMID: 3516542210. C.A. Klebanoff, S.S. Chandran, B.M. Baker, S.A. Quezada, A. Ribas, T cell receptor therapeutics: immunological targeting of the intracellular cancer proteome. Nat Rev Drug Discov 22, 996-1017 (2023) . doi. org/10.1038/s41573-023-00809-z. PMID: 3789143511. C. Puig-Saus et al., Neoantigen-targeted CD8 (+) T cell responses with PD-1 blockade therapy. Nature 615, 697-704 (2023) . doi. org/10.1038/s41586-023-05787-1. PMID: 3689023012. G.M. Lynn et al., Peptide-TLR-7/8a conjugate vaccines chemically programmed for nanoparticle self-assembly enhance CD8 T-cell immunity to tumor antigens. Nat Biotechnol 38, 320-332 (2020) . doi. org/10.1038/s41587-019-0390-x. PMID: 3193272813. Z. Jing et al., A Potent Micron Neoantigen Tumor Vaccine GP-Neoantigen Induces Robust Antitumor Activity in Multiple Tumor Models. Adv Sci (Weinh) 9, e2201496 (2022) . doi. org/10.1002/advs. 202201496. PMID: 3571277014. G. Cafri et al., mRNA vaccine-induced neoantigen-specific T cell immunity in patients with gastrointestinal cancer. J Clin Invest 130, 5976-5988 (2020) . doi. org/10.1172/JCI134915. PMID: 3301692415. S.D. Martin et al., A library-based screening method identifies neoantigen-reactive T cells in peripheral blood prior to relapse of ovarian cancer. Oncoimmunology 7, e1371895 (2017) . doi. org/10.1080/2162402X. 2017.1371895. PMID: 2929652216. C.J. Cohen et al., Isolation of neoantigen-specific T cells from tumor and peripheral lymphocytes. J Clin Invest 125, 3981-3991 (2015) . doi. org/10.1172/JCI82416. PMID: 2638967317. D. Wei, L. Wang, X. Zuo, A. Maitra, R.S. Bresalier, A Small Molecule with Big Impact: MRTX1133 Targets the KRASG12D Mutation in Pancreatic Cancer. Clin Cancer Res 30, 655-662 (2024) . doi. org/10.1158/1078-0432. CCR-23-2098. PMID: 3783100718. M.K. Kim et al., Prognostic value of the TP53 mutation in patients with pancreatic ductal adenocarcinoma receiving FOLFIRINOX. Ther Adv Med Oncol 16, 17588359241290482 (2024) . doi. org/10.1177/17588359241290482. PMID: 3944973219. S. Connal et al., Liquid biopsies: the future of cancer early detection. J Transl Med 21, 118 (2023) . doi. org/10.1186/s12967-023-03960-8. PMID: 3677450420. J.D. Cohen et al., Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science 359, 926-930 (2018) . doi. org/10.1126/science. aar3247. PMID: 2934836521. E.A. Klein et al., Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set. Ann Oncol 32, 1167-1177 (2021) . doi. org/10.1016/j. annonc. 2021.05.806. PMID: 3417668122. S. Cristiano et al., Genome-wide cell-free DNA fragmentation in patients with cancer. Nature 570, 385-389 (2019) . doi. org/10.1038/s41586-019-1272-6. PMID: 3114284023. R.H. Xu et al., Circulating tumour DNA methylation markers for diagnosis and prognosis of hepatocellular carcinoma. Nat Mater 16, 1155-1161 (2017) . doi. org/10.1038/nmat4997. PMID: 2903535624. R. Ben-Ami et al., Protein biomarkers and alternatively methylated cell-free DNA detect early stage pancreatic cancer. Gut 73, 639-648 (2024) . doi. org/10.1136/gutjnl-2023-331074. PMID: 3812399825. A. Sverchkova, S. Burkholz, R. Rubsamen, R. Stratford, T. Clancy, Integrative HLA typing of tumor and adjacent normal tissue can reveal insights into the tumor immune response. BMC Med Genomics 17, 37 (2024) . doi. org/10.1186/s12920-024-01808-8. PMID: 3828102126. Z. Moravec et al., Discovery of tumor-reactive T cell receptors by massively parallel library synthesis and screening. Nat Biotechnol, (2024) . doi. org/10.1038/s41587-024-02210-6. PMID: 3865379827. M. Meyer et al., MediMer: a versatile do-it-yourself peptide-receptive MHC class I multimer platform for tumor neoantigen-specific T cell detection. Front Immunol 14, 1294565 (2023) . doi. org/10.3389/fimmu. 2023.1294565. PMID: 3823935228. D. Saka et al., Mechanisms of T-Cell Exhaustion in Pancreatic Cancer. Cancers (Basel) 12, (2020) . doi. org/10.3390/cancers12082274. PMID: 3282381429. R. Rosenthal et al., Neoantigen-directed immune escape in lung cancer evolution. Nature 567, 479-485 (2019) . doi. org/10.1038/s41586-019-1032-7. PMID: 3089475230. V.P. Balachandran et al., Identification of unique neoantigen qualities in long-term survivors of pancreatic cancer. Nature 551, 512-516 (2017) . doi. org/10.1038/nature24462. PMID: 2913214631. K. Yamamoto et al., Autophagy promotes immune evasion of pancreatic cancer by degrading MHC-I. Nature 581, 100-105 (2020) . doi. org/10.1038/s41586-020-2229-5. PMID: 3237695132. J.C.M. Wan et al., Liquid biopsies come of age: towards implementation of circulating tumour DNA. Nat Rev Cancer 17, 223-238 (2017) . doi. org/10.1038/nrc. 2017.7. PMID: 2823380333. J.D. Merker et al., Circulating Tumor DNA Analysis in Patients With Cancer: American Society of Clinical Oncology and College of American Pathologists Joint Review. J Clin Oncol 36, 1631-1641 (2018) . doi. org/10.1200/JCO. 2017.76.8671. PMID: 2950484734. A.H. Pearlman et al., Targeting public neoantigens for cancer immunotherapy. Nat Cancer 2, 487-497 (2021) . doi. org/10.1038/s43018-021-00210-y. PMID: 3467637435. M.C. Liu et al., Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA. Ann Oncol 31, 745-759 (2020) . doi. org/10.1016/j. annonc. 2020.02.011. PMID: 3350676636. P. Bailey et al., Genomic analyses identify molecular subtypes of pancreatic cancer. Nature 531, 47-52 (2016) . doi. org/10.1038/nature16965. PMID: 2690957637. A.V. Biankin et al., Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes. Nature 491, 399-405 (2012) . doi. org/10.1038/nature11547. PMID: 2310386938. L. Cao et al., Proteogenomic characterization of pancreatic ductal adenocarcinoma. Cell 184, 5031-5052 e5026 (2021) . doi. org/10.1016/j. cell. 2021.08.023. PMID: 3453446539. Y. Cao et al., Whole exome sequencing of insulinoma reveals recurrent T372R mutations in YY1. Nat Commun 4, 2810 (2013) . doi. org/10.1038/ncomms3810. PMID: 2432677340. K.A. Hoadley et al., Cell-of-Origin Patterns Dominate the Molecular Classification of 10,000 Tumors from 33 Types of Cancer. Cell 173, 291-304 e296 (2018) . doi. org/10.1016/j. cell. 2018.03.022. PMID: 2962504841. Y. Jiao et al., DAXX/ATRX, MEN1, and mTOR pathway genes are frequently altered in pancreatic neuroendocrine tumors. Science 331, 1199-1203 (2011) . doi. org/10.1126/science. 1200609. PMID: 2125231542. A. Scarpa et al., Whole-genome landscape of pancreatic neuroendocrine tumours. Nature 543, 65-71 (2017) . doi. org/10.1038/nature21063. PMID: 2819931443. J. Wu et al., Whole-exome sequencing of neoplastic cysts of the pancreas reveals recurrent mutations in components of ubiquitin-dependent pathways. Proc Natl Acad Sci U S A 108, 21188-21193 (2011) . doi. org/10.1073/pnas. 1118046108. PMID: 2215898844. Y. Jiao et al., Whole-exome sequencing of pancreatic neoplasms with acinar differentiation. J Pathol 232, 428-435 (2014) . doi. org/10.1002/path. 4310. PMID: 2429329345. E. Cerami et al., The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2, 401-404 (2012) . doi. org/10.1158/2159-8290. CD-12-0095. PMID: 2258887746. X. Zhang et al., Characterization of the genomic landscape in large-scale Chinese patients with pancreatic cancer. EBioMedicine 77, 103897 (2022) . doi. org/10.1016/j. ebiom. 2022.103897. PMID: 3523169947. M. Dal Molin et al., Very Long-term Survival Following Resection for Pancreatic Cancer Is Not Explained by Commonly Mutated Genes: Results of Whole-Exome Sequencing Analysis. Clin Cancer Res 21, 1944-1950 (2015) . doi. org/10.1158/1078-0432. CCR-14-2600. PMID: 2562321448. J. Wu et al., TSNAdb: A Database for Tumor-specific Neoantigens from Immunogenomics Data Analysis. Genomics Proteomics Bioinformatics 16, 276-282 (2018) . doi. org/10.1016/j. gpb. 2018.06.003. PMID: 3022304249. K. Wang, M. Li, H. Hakonarson, ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res 38, e164 (2010) . doi. org/10.1093/nar/gkq603. PMID: 2060168550. V. Jurtz et al., NetMHCpan-4.0: Improved Peptide-MHC Class I Interaction Predictions Integrating Eluted Ligand and Peptide Binding Affinity Data. J Immunol 199, 3360-3368Materials and MethodsClinical study design and participants

The clinical study in this Example describes a retrospective clinical study involving a total of 245 cancer patients, which includes 210 patients with PDAC, 10 patients with CRC, 10 with ICC, 10 with liver metastasis of colorectal cancer (LM-CRC) , and 5 with liver metastasis of pancreatic ductal adenocarcinoma (LM-PDAC) . Additionally, 185 healthy volunteers were enrolled from Zhongshan Hospital, affiliated with Fudan University. All participants were divided into three pre-specified sub-studies: (1) discovery and validation, (2) double-blind detection, and (3) superiority of tNET detection. The first sub-study aimed to demonstrate the feasibility of tNET detection for identifying PDAC (n=107) , ICC (n=10) , CRC (n=10) , LM-CRC (n=10) , and healthy volunteers (n=78) who had matched HLA-A types and mutations with the MHC-I sub-library. The second sub-study involved blind tNET detection using the entire MHC-I library, which included the HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02 sub-libraries, in PDAC patients (n=70) and healthy volunteers (n=95) blinded to their HLA-A type and gene mutation profiles. This was conducted to assess the potential of tNET detection for early diagnosis across various cancers. Finally, 57 PDAC patients and 76 healthy volunteers were matched for HLA and gene mutations, and this group was analyzed to investigate the efficiency of tNET detection in the double-blind assessment of PDAC. The third sub-study aimed to compare the sensitivity of tNET detection with that of circulating tumor DNA (ctDNA) detection in diagnosing PDAC. Fresh blood samples from 15 PDAC patients were used for both tNETs and ctDNA detection simultaneously. 1 mL of fresh whole blood was utilized for tNETs detection with the complete MHC-I library, while 5 mL of fresh whole blood were used for ctDNA detection. The detection sensitivity was analyzed based on the differing proportions of PDAC diagnoses obtained through the tNETs and ctDNA strategies in this subgroup. The ethical approval numbers for this study are B2025-024R and B2024-247R.Tumor samples and PBMCs collection, processing, and preparation

De-identified clinical samples, including fresh blood, tumor tissues, and paratumor tissues, were collected from cancer patients, while fresh blood was obtained from healthy volunteers. Specifically, 15 mL of fresh peripheral blood was collected from cancer patients, and 3 mL from healthy volunteers, using anticoagulant tubes and maintained at 4 ℃ during transportation. One milliliter of the fresh blood sample was immediately frozen at -20℃ for human leukocyte antigen (HLA) typing analysis using DNA sequencing of white blood cells. The remaining blood was processed to isolate PBMCs within one hour. The isolated PBMCs were utilized for tNET (1 mL of peripheral blood yields approximately 1×10⁶ PBMCs) , ELISpot, or flow cytometry analysis. Tumor and para-tumor tissues from cancer patients were collected and divided into two portions. One portion was rapidly snap-frozen in liquid nitrogen, while the other was fixed in 4%formaldehyde solution for whole exome sequencing (WES) analysis.Peripheral Blood Mononuclear Cells (PBMCs) isolation

PBMCs were isolated from whole blood using density gradient centrifugation with Lymphocyte Separation Solution (YEASEN, 40503ES60) as previously described with minor modifications. Briefly, the separation solution was carefully underlaid beneath the blood sample in 50 mL PBMCs separation tubes (STEMCELL, SepMate^TM) using a syringe to minimize bubble formation, followed by gentle layering of 1: 1 PBS (pH=7.4) -diluted whole blood (supplemented with 2%FBS and 1%penicillin-streptomycin) . After centrifugation at 1, 200 × g for 10 min at 4℃, the PBMCs-containing supernatant was collected, diluted to 40 mL with PBS (pH=7.4) (2%FBS, 1%P/S) , and pelleted by centrifugation (400×g, 6 min, 4℃) . The cells were washed once with 40 mL PBS (pH=7.4) (2%FBS, 1%P/S) , aliquoted, and centrifuged again under identical conditions. Isolated PBMCs were either cryopreserved in CryoStor^TM CS10 medium (1.5×10⁶ cells/mL) using controlled-rate freezing for liquid nitrogen storage, or immediately resuspended in ImmunoSpot^TM CTL-Test Medium for tNET detection or ELISpot analysis. For murine PBMCs isolation, the manufacturer's protocol for the PBMCs isolation kit (TBD, LDS1090) was followed precisely.Cell lines and mouse models

Pan02 and Aspc1 cell lines were purchased from ATCC. A recombinant DNA sequence encoding six Kras mutation-derived 25 amino acid oligopeptides, linked by ribosomal skipping sequences (T2A and P2A) , was cloned into a CMV expression plasmid to construct the Pan02-neo cell line, which stably overexpresses six Kras mutations (Kras^G12V, Kras^G12D, Kras^G12R, Kras^G12C, Kras^G12I, and Kras^G12A) derived neoantigens. The KPC cell line was derived from primary murine pancreatic ductal adenocarcinoma (PDAC) (Kras^LSL-G12D/+; Trp^{53LSL-R172H/+}; Pdx-1 Cre) in C57BL/6 strains. Both Pan02 and KPC cell lines were cultured in DMEM supplemented with 10%fetal bovine serum (FBS) and 1%penicillin/streptomycin in a humidified incubator at 37℃ with 5%CO2 (Thermo, 6N98) , and were sub-cultured as needed using trypsin-EDTA. Before constructing the tumor model, cells were collected and counted using a cell counter (Life Technologies, Countess II FL) following Trypan Blue staining. For the subcutaneous tumor model, 4 × 10⁵ Pan02-con or Pan02-Neo cells were resuspended in 100 μL of PBS (pH=7.4) and injected into the flank of C57BL/6J mice. The orthotopic pancreatic mouse model was established as previously reported (PMID: 29903803) . Briefly, a 1 cm incision was made in the right flank of the mouse (in a caudal to rostral direction, approximately 5–7 anterior to the spine) , and the pancreatic tissue was gently pulled extracted blunt forceps under sterile conditions to find locate head of the mouse Then, 10³, 10⁴, 2×10⁵, and 4×10⁵ cells were resuspended in 25 μL of and implanted 2mm-deep into head of C57BL/6J mice. Antibiotics (Benzylpenicillin Sodium, 500 IU/g body weight) were administered to prevent postoperative infection. Peripheral blood samples were collected from the retroorbital vein of mice challenged with varying numbers of tumor cells and for different durations using sterile capillaries. PBMCs was isolated as previously described. A total of 4×10⁵ PBMCs were seeded in a 96-well plate for the analysis of dynamic changes in tumor neoantigen-specific CD8⁺ T cells. All animal procedures were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals and were approved by the Animal Ethics Committee of Fudan University.huHSC-C-NKG-ProF construction and PDX establishment

The huHSC-NKG-ProF mice represent an advanced iteration of the huHSC-NKG series, featuring of severe immune deficiency. This mouse model generates a diverse array of human immune cells, including T cells, B cells, natural killer (NK) cells, dendritic cells, and monocytes. In summary, three-day-old female C-NKG mice were irradiated with 1.0 Gy at a rate of 1.23 Gy/min using a RAD SOURCE RS2000 to eliminate bone marrow cells. Subsequently, 1×10⁶ umbilical cord blood-derived hematopoietic stem cells (HLA-A*02: 01) were transplanted via the superficial temporal vein. After 12 weeks, human immune cell reconstitution was assessed using flow cytometry. Following the development of immune cells, patient-derived tumor tissue was prepared. Briefly, KRAS^G12D mutant PDAC tissues were obtained from clinical surgeries and immediately placed in a tissue preservation solution. The tumor samples were then transported to a specific pathogen-free (SPF) facility via a dedicated biosafety transfer port. Tumor tissues were washed three times in 5 mL of precooled PBS (pH=7.4) containing 2%penicillin/streptomycin, and necrotic and normal tissues were removed. The tumor tissue in PBS (pH=7.4) was subsequently transferred to Dulbecco's Modified Eagle Medium (DMEM) containing 2%penicillin/streptomycin, and five tissue fragments (150 to 300 mm³) were reserved. The remaining tissue was cut into smaller pieces measuring 20 to 30 mm³, which were then implanted into the armpit and hind leg of mice using a puncture needle. The growth of the tumor was monitored, and PBMCs were isolated for further analysis.Luciferase detection in orthotopic pancreatic mouse model

The KPC cells were stably transfected with a lentiviral vector carrying puromycin resistance (puro) and firefly luciferase (Luc) (KPC-Luc-puro) (HANBIO, Shanghai) . The KPC cells were cultured in DMEM, supplemented with 10%fetal bovine serum (FBS) (Gibco) and 1%penicillin/streptomycin (P/S) , and maintained at 37℃ in a humidified atmosphere with 5%CO2. Mice bearing orthotopically implanted KPC-Luc-puro tumors were randomly allocated into different groups, and tumor growth was monitored by using IVIS imaging system (PerkinElmer) . D-luciferin potassium salt (150 mg/kg of body weight) (Yeasen Biotechnology, Shanghai) was intraperitoneally injected into mice, and the bioluminescence recorded.RNA extraction and RT-qPCR

Total RNA was extracted using the RNeasy Mini Kit (Qiagen) . RNA concentration was measured with a Nanodrop 2000 (Thermo) , and reverse transcription was performed using the PrimeScript RT Reagent Kit with gDNA Eraser (Takara, RR047Q) . Gene expression levels were quantified using SYBR (Roche) intensity in a 384-well plate on the ABI Prism 7500 (Applied Biosystems) . The expression levels of target genes were normalized to GAPDH using the ΔC (t) method. Data are presented as mean± s.d. from three independent experiments. All primers sequences were as follows: Neo-F: 5′-AACCCACTGCTTACTGGCTTA-3′; Neo-R: 5′-TCATGGTGGCGAATTCGGTG-3′; Actin-F: 5′-CTAGACACCATGTGCGACGA-3′; Actin-R: 5′-ATAGATGGGCACGTTGTGGG-3′.Western blots

Cells were lysed using RIPA buffer containing 10 mM PMSF as a protease inhibitor for 30 minutes on ice. The protein concentration was quantified using the Bradford assay, and a loading sample of 1 μg/μL was prepared with BlueJuice^TM loading buffer (Thermo Fisher, Catalog #10816015) and boiled for 5 minutes in a metal bath. A total of 10 μg of protein, which was prepared using the PAGE Gel Fast Preparation Kit (Yamei, PG112) , was subjected to SDS-PAGE electrophoresis. The proteins were then transferred to PVDF membranes (Thermo Fisher, Catalog #88518) . After blocking with 5%skim milk, the protein bands were incubated overnight at 4℃ with primary antibodies: Anti-KRAS (G12D Mutant) [HL10] (Genetex, Catalog #GTX635362) and β-actin (Abcam, Catalog #ab7817) . Following incubation with secondary antibodies, the bands were visualized using an ECL kit (Smart-Lifesciences, Catalog #H31500-1 & H31500-1) under a Gel Imaging System (Tanon, 2500BR) .H&E staining

Normal pancreatic tissue and pancreatic tumor tissue were resected and fixed in 4%paraformaldehyde overnight, then embedded in standard paraffin wax and sectioned into 5 μm sections. Following standard procedures, these sections were subjected to H&E staining.MRI in vivo imaging

For in vivo imaging, high‐resolution MRI acquisitions were performed on a 9.4T small -animal MRI system (uMR 9.4T, United Imaging Life Science Instrument, Wuhan, China) in combination with a 3-channel receive array coil. High resolution T2 weighted images were acquired with following parameters: TR=1800ms, TE=34.8ms, field of view (FOV) = 37 × 21 mm, acquisition matrix = 240 × 140, 18 slices with a thickness of 0.5 mm, 4 averages, echo train length=9. During the scans the animals were monitored to minimize motion effects in the imaging process via image gating. This was accomplished by using the MR‐compatible small animal monitoring and gating system (SA Instruments Inc., Stony Brook, NY) . The system was setup to monitor the respiration gating to remove motion artifacts by triggering during the MR acquisition.Aggregation of TNETs by Beads-based TSNA^PeptideMHC-IC^Biotin multimers

Targeted model antigens-Ovalbumin (OVA) or a mixture of Kras^G12V-and Kras^G12D-derived neoantigens were refolded with HLA and β-2M in vitro to create TSNA^PeptideMHC-IC^Biotin. The TSNA^PeptideMHC-IC^Biotin monomers were incubated with SA-magnetic beads at room temperature for 1 to 2 hours to produce beads loaded with TSNA^PeptideMHC-IC^Biotin monomers. After thorough washing, 15 μL of the solution containing the beads loaded with the TSNA^PeptideMHC-IC^Biotin multimers (approximately 3.9 pmol of monomers) was incubated with PBMCs isolated from mice that had been pre-stimulated with the corresponding model antigens or the mixed Kras^G12V-and Kras^G12D-derived neoantigens for 4 hours in a 37℃ incubator with 5%CO2. The formation of aggregates was imaged under a microscope using established protocols.Expression and Purification of HLA and B2M Proteins

Recombinant HLA and β-2M were produced in Escherichia coli using the pET-28b vector. The bacterial cells were lysed with a lysis buffer (ACE, BR0005-02) and centrifuged at 12,000 ×g for 10 minutes, discarding the supernatant and retaining the inclusion body. The inclusion body was washed with buffer W1 (50 mM Tris-HCl, pH 8.0, 100 mM NaCl, 1%Triton X-100, 12.5 mM EDTA, 1%deoxycholic acid, 10 mM DTT) by sonication, followed by centrifugation at 12,000×g for 10 minutes; this washing step was repeated three to four times. Next, the inclusion body was washed with buffer W2 (50 mM Tris-HCl, pH 8.0, 1 mM EDTA, 0.5%Triton X-100, 1 mM DTT, 100 mM NaCl) and buffer W3 (50 mM Tris-HCl, pH 8.0, 1 mM EDTA, 1 mM DTT, 100 mM NaCl) by sonication, with each wash repeated three times. Purified proteins were stored in 6 M guanidine-HCl containing 10 mM β-mercaptoethanol at -80 ℃. Protein concentration was measured using the Bradford assay, and protein refolding and characterization were confirmed via SDS-PAGE.Refolding of TSNA^PeptideMHC-IC monomers

1.5 mg of neoantigen peptides (neoantigens were synthesized by Nanjing Peptide Biotech Ltd., with peptide purity exceeding 95%) was dissolved in DMSO at a concentration of approximately 10-20 mg/mL and gradually added to 50 mL of Refolding Buffer (100 mM Tris-HCl, pH 8.0, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, 0.2 mM PMSF) at 4 ℃ while stirring at 700 rpm. Subsequently, 100 nmol of recombinant β-2M was added dropwise to the peptide solution, followed by the addition of 50 nmol of recombinant HLA to the β-2M/peptide mixture. The refolding reaction was allowed to proceed for 24 hours at 4 ℃ with a reduced stirring speed of 200 rpm. On the second and third days, an additional 25 nmol of recombinant HLA was added dropwise to the overnight HLA/β-2M/peptide solution, and the refolding continued for another 24 hours under the same conditions. Once renaturation was complete, 50 mL of HLA/β-2M/peptide refolding solution was mixed with 2 mL of 5 M NaCl to achieve a final concentration of 200 mM. The mixture was then centrifuged at 3000 rpm for 15 minutes at 4 ℃. The resulting supernatant, which contained the refolded TSNA^PeptideMHC-IC monomers, was concentrated to approximately 1-2 mL using a 30-kDa ultrafiltration tube (Millipore) . Finally, proteins were separated from aggregates and unbound peptide complexes using SEC on an AKTA system, and the buffer was exchanged for PBS (pH=7.4) .

Based on the description of the HLA structure in the literature, after renaturing MHC-I, a satisfactory yield was not achieved, and the biotinylation catalytic effect was also suboptimal. During the experiment, the MHC-I bands were faint, and SA could not bind effectively to the biotinylated MHC-I. These observations collectively indicate that the biotinylation modification did not yield the desired results. In response to this, the HLA expression pattern was optimized by introducing a spacer (e.g., a GS linker (e.g., G4S) and a 10*His-tag) . This modification aims to spatially separate HLA from the avi-tag, endowing the avi-tag with greater flexibility and thereby enhancing its susceptibility to catalysis by the BriA.Biotinylation TSNA^PeptideMHC-IC monomers

Biotinylation of the TSNA^PeptideMHC-IC monomers were conducted using the BirA enzyme biotinylation kit from Avidity (Byeotime, P0630M) . The proteins were purified from aggregates and excess biotin through SEC (Cytiva, Superdex 75 Increase 10/300 GL) utilizing an AKTA system (Cytiva, Pure 24) , and the buffer was subsequently exchanged for PBS (pH=7.4) .Non-denaturing and non-reducing gel electrophoresis

The loading samples were prepared using a non-denaturing and non-reducing protein loading buffer (5×) (Beyotime, P0016N) . During the electrophoresis process, a non-denaturing PAGE buffer (Beyotime, P0014G) was employed. The electrophoresis was conducted on ice with the Protein Electrophoresis System (BIO-RAD) at a constant voltage of 180 V for 60 to 90 minutes, alongside the High Molecular Weight Native Electrophoresis Protein Marker (Coolaber, DM2004 -20T) . Notably, the gel utilized in this experiment was obtained from the Polyacrylamide Gel Rapid Preparation Kit (Yamei, PG112) . Upon completion of the electrophoresis, the gel was stained with Coomassie Brilliant Blue solution (composed of 50%methanol, 10%acetic acid, and 0.25% (w/v) Coomassie Brilliant Blue R250) for 1 to 2 hours, followed by destaining with a solution of 20%methanol and 10%acetic acid until the background was clear.Flow cytometry analysis

Fresh peripheral blood was collected into a 1.5-mL microcentrifuge tube containing 50 μL of 0.5 mM EDTA. The red blood cells were removed using 1 mL of lysing buffer (BD, Cat. 555899) according to the manufacturer's instructions. The cell pellet was obtained by centrifugation at 350 ×g for 5 minutes and was then resuspended in 40 μL of staining buffer (PBS + 1%FBS) . The cells were pre-incubated with 2 μL of Fc Block for 10 minutes at 4℃, followed by the addition of 5 μL of specific fluorophore-conjugated antibodies, and incubated for 30 minutes on ice in the dark. After staining, the samples were washed twice with pre-chilled PBS (pH=7.4) and resuspended in 300 μL of pre-cooled PBS (pH=7.4) for quantification using a flow cytometer instrument (LSR Fortessa, BD Biosciences) . The flow cytometry files were analyzed using FlowJo software (V10, TreeStar) .Preparation of beads loaded with TSNA^PeptideMHC-IC^Biotin monomers

Streptavidin-coated M280-SA beads (Thermo Fisher, #11205D) were resuspended by vigorous vortexing to ensure uniformity. A total of 100 μL of M280-SA beads were washed three times in 1 mL of PBS (pH=7.4) and then resuspended in 1 mL of PBS (pH=7.4) . Next, 10 μg of TSNA^PeptideMHC-I monomer was added to the bead solution and incubated for 1 hour at room temperature, followed by an overnight incubation at 4℃ on a rocker. After washing the beads four times with PBS (pH=7.4) , the MHC-peptide-loaded magnetic beads were prepared for subsequent experiments.Loading quantity of monomer on beads

he theoretical loading capacity of 1 mL of magnetic beads at a concentration of 1 mg/mL is 3 to 5 μg of monomer. Consequently, the addition of 60 μL of magnetic beads in a single tNET reaction corresponds to approximately 3.9 to 6.5 pmol of MHC-I monomers.

The formula for calculation is as follows:

N represents the amount of protein in picomoles (pmol) . V denotes the volume of magnetic beads in microliters (μl) . C indicates the loading capacity in micrograms per milligram of magnetic beads (μg/mg) . M refers to the molecular weight of the protein in kilodaltons (kDa) .LC-MS/MS analysis

The complex was treated with a 0.1 M glycine solution (pH 2.5) and then it was boiled for 10 minutes. Subsequently, in accordance with the requirements of the protocol, the resulting peptides were desalted using C18 Zip-Tips (Millipore, ZTC18S096) and dried in a vacuum centrifuge. LC-MS/MS analysis of enriched peptides was performed on a nano-HPLC chromatography system connected to a hybrid trapped ion mobility spectrometry quadrupole time-of-flight mass spectrometer (TIMS-TOF Pro, Bruker Daltonics) via a CaptiveSpray nano-electrospray ion source. A total of 200 ng peptides dissolved in solvent A (0.1%formic acid) was loaded onto the analytical column (75 μm i.d. × 25 cm) and separated with a 60 min gradient (2 -22%solvent B (ACN with 0.1%formic acid) for 45 min, 22 -37%B for 5 min, 37 -80%B for 5 min, and then 80%B for 5 min) . The flow rate was maintained at 300 nL/min. For MS analysis, the accumulation and ramp time were set to 100 ms each. Survey full-scan MS spectra (m/z 100–1700) were obtained in positive electrospray mode. The ion mobility was scanned from 0.75 to 1.37 Vs/cm². The overall acquisition cycle of 1.16 s comprised one full TIMS-MS scan and 10 parallel accumulation-serial frag-mentation (PASEF) MS/MS scans. During PASEF MS/MS scanning, the collision energy was ramped linearly as a function of the mobility from 59 eV at 1/K0 = 1.6 Vs/cm² to 20 eV at 1/K0 = 0.6 Vs/cm². All the mass spectral data generated by Nano-HPLC-MS/MS were online searched using PEAKS ONLINE (X Build, version 1.4.2020-10-02_113407) against corresponding peptide library database. The precursor mass error tolerance was set to 20 ppm, the fragment mass error tolerance was set to 0.05 Da, the enzyme mode was set to no digestion, and the length of the identified peptides was set to 6-45 amino acids. The variable modifications were set to methionine oxidation (+15.995 Da) and acetylation (Protein N-term) (+42.011 Da) . The false discovery rate (FDR) for peptides and protein groups were set to less than 0.01.Identification of Tumor-Specific Somatic Mutations and Derived TSNAs

To identify tumor-specific somatic coding mutations and derive potential neoantigens, two complementary strategies to establish a neoantigen library in this study were employed. Approach 1: High-Frequency DNA Marker, genomic data from 1, 082 pancreatic cancer samples obtained from the cBioPortal database ^[36-44] was analyzed. Within this dataset, non-synonymous mutations with a frequency greater than 2 in pancreatic cancer patients ^[45] was selected for further evaluation. Approach 2:High Peptide Affinity Marker By conducting a thorough review of the literature ^[46, 47] and leveraging insights from TSNAdb ^[48] , five genes-KRAS, TP53, CDKN2A, MUC16, and SMAD4-were identified as harboring mutations associated with pancreatic tumorigenesis. These genes frequently exhibit somatic mutations associated with disease progression and are likely to generate high-affinity peptides suitable for neoantigen discovery. By combining these two approaches, a comprehensive pool of candidate mutations for neoantigen prediction was constructed. To convert the mutations into peptide sequences, ANNOVAR ^[49] , a versatile tool for annotating genetic variants and generating all possible peptide sequences from somatic mutations, was utilized. The resulting peptides were then evaluated using NetMHCpan (v4.0) ^[50] , an algorithm widely recognized for its accuracy in predicting TSNA^PeptideMHC-I binding affinity for specific neoantigens. 8-to13-mer peptides derived from mutation sites were analyzed and assessed for their binding affinity to common HLA class I alleles (HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02) . For the mutation candidates identified in approach 1, up to five peptides per mutation were selected based on their binding strength with these HLA-A types for further testing. For the candidate neoantigen generated by approach 2, three peptides with the strongest binding affinity (lowest IC50 values) for each mutation site were prioritized.

Table 10 shows proportions of HLA-A alleles in white and Asian populations. In white populations, the combined proportion of HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02 represent 48.82%of the population, and in Asian populations the combined proportion of HLA-A*02: 01, HLA-A*03: 01, HLA-A*11: 01, and HLA-A*24: 02 represents 46.74%of the population.Table 10: HLA-A allele frequency tNET detection of the specific CD8⁺ T cells by beads-based TSNA^PeptideMHC-IC^biotin hetero-multimers with ELISA

To detect neoantigen-reactive CD8 T cells from PBMCs, a magnetic enzyme-linked immunosorbent assay (ELISA) was developed based on tNET detection. For fresh PBMCs, an appropriate quantity of cells was resuspended in CTL-Test^TM medium and seeded into a 96-well plate. For frozen samples, PBMCs were thawed at 37℃ and resuspended in 5 mL of CTL-Test^TM medium (CTL, Catalog #CTLT-010) in a 15 mL tube. The cells were then centrifuged at 350 ×g for 5 minutes at 4℃. The supernatant was discarded, and the PBMCs pellet was diluted to 1 x 10⁶ viable cells per 200 μL of CTL-Test^TM medium. The cells were cultured in a CO2 incubator at 37℃ with 5%CO2 in a humidified atmosphere. The isolated PBMCs should be used within three days after thawing, and the percentage of the viable cells must exceed 60%Dasatinib (Med Chem Express, Catalog #BMS-354825) was dissolved in CTL-TEST^TM medium to achieve a final working concentration of 50 nM. Subsequently, 1 μL of Human (BD, Catalog #564765) or mouse Fc-block (BD, Catalog #553141) was added, mixed gently, and incubated at 37℃ for 30 minutes. Peripheral blood mononuclear cells (PBMCs) were then collected by centrifugation at 360 ×g and 4℃ for 6 minutes in a 1.5 mL tube. The cell pellet was gently resuspended in 150 μL of pre-cold PBS (pH=7.4) containing 1%BSA (Sigma, Catalog #WXBD9888V) and transferred to each test tube, with approximately 1 x 10⁶ viable cells per tube. Next, 0.5 μL of CD8-HRP antibody (Catalog #NBP2-25195H, Novusbio) was added to the cells and mixed gently. This was followed by the addition of either 15 μL of single HLA beads or 70 μL of whole library beads for HtNETs testing, and the mixture was incubated at room temperature in the dark on a shaker set to 20-30%power for one hour. After incubation, the beads-PBMCs mixture was transferred to a new 8 mL tube, resuspended in 1 mL of pre-cold wash buffer (PBS (pH=7.4) +1%FBS) , and placed on a DynaMag^TM-5 (Invitrogen, Catalog #12303D) for 35 seconds to separate the beads bound to neoantigen-specific CD8⁺ T cells. The wash buffer was then removed, and the bead-cell mixture was immediately washed with 4 mL of pre-chilled wash buffer five times (35 seconds each time on a magnetic stand) . After the fifth wash, the mixture was transferred to a new tube, and the wash buffer was removed while the tubes were on the magnetic stand. Next, 100 μL of TMB substrate solution (Thermo Scientific, Catalog #34022) was added to resuspend the bead-CD8⁺ T cell mixture, which was subsequently transferred into a 96-well plate and incubated at room temperature in the dark for 20 to 30 minutes, depending on the color change. The plate was then placed on a magnet for 1 minute, and the supernatant was collected into adjacent new wells. Following this, 100 μL of Stop solution (Thermo Scientific, Catalog #PIN600) was added to the supernatant and incubated for 10 minutes. The optical density (OD) of the supernatant was analyzed at 450 nm using a microplate reader (BioTek) .DNA extraction

Genomic DNA was extracted from tumor tissues (formalin-fixed, paraffin-embedded [FFPE] ) with a pathological tumor cellularity of 20%or greater using the HiPure FFPE DNA Kit (D3126, Magen) . A peripheral blood sample was obtained and stored in Cell-Free DNA BCT tubes (Streck Inc., Omaha) . The sample was centrifuged at 1600 ×g and 4℃ for 10 minutes. The sediment was utilized for matched genomic DNA extraction using the TIANGEN TIANamp Genomic DNA Kit (DP304, TIANGEN) , while the supernatant was transferred to a new tube and centrifuged at 16,000 ×g and 4℃ for 10 minutes to isolate circulating cell-free DNA (cfDNA) . The supernatant was then transferred to a new tube. cfDNA was recovered from plasma using the Extiquick^TM cfDNA DNA Kit (EX002, WISGN BIO) . DNA quantification was performed with the Qubit^TM 3.0 (Thermo Fisher) using the Equalbit 1 × dsDNA HS Assay Kit (EQ121, Vazyme) .Library construction and sequencing

The whole-exome sequencing (WES) library was prepared using the VAHTS Universal Plus DNA Library Prep Kit for Illumina V2 (ND627, Vazyme) according to the manufacturer's instructions. Briefly, 200 ng of formalin-fixed paraffin-embedded (FFPE) DNA samples and 100 ng of matched white blood cell (WBC) genomic DNA (gDNA) were fragmented using an enzyme. The fragmented DNA samples were end-repaired and dA-tailed, followed by ligation with Universal Adapters. After post-ligation cleanup, the ligated products were amplified with index primers. Whole-exome and HLA-related fragments were captured separately using the Hi-Exon 35 Probe (P10016, Homgene) and the HLA-Small Probe (P10011, Homgene) , respectively. The final libraries were quantified using the Qubit^TM 3.0 (Thermo Fisher, Carlsbad) with the Equalbit 1 × dsDNA HS Assay Kit (EQ121, Vazyme) . Library size distribution was analyzed with the Qseq100 (Bioptic) according to the manufacturer’s recommendations. All libraries were sequenced on the DNBSEQ-T7 sequencer (MGI) using a 2×150 bp paired-end sequencing mode, achieving a mean target coverage of approximately 500×for tumor samples and approximately 150× for matched normal samples. For HLA samples, an average of 1 Gb of data was obtained for each sample.WES data bioinformatics analysis

Adapters were initially removed using Trimmomatic (v0.39-1) . Clean reads were subsequently mapped to the human reference genome (UCSC, hg19) using the BWA-MEM (v0.7.17-r1194-dirty) algorithm to generate a SAM file. This SAM file was then converted to BAM format using Samtools (v1.3.0) . PCR and optical duplicates were identified with Picard (v2.20.2-0) , and suspicious regions around indels were realigned using GATK (v4.1.3.0) to produce an analysis-ready BAM file. Quality control (QC) information was obtained with BAMdst (v1.0.9) . Somatic variants were called using VarDict (v1.8.2) , retaining only variants with a minimum allele frequency of ≥ 5%. According to the ExAC, 1000 Genomes, and gnomAD databases, variants with a population frequency greater than 1%in tumors were classified as single-nucleotide polymorphisms (SNPs) and excluded from further analysis. All remaining variants were annotated using ANNOVAR (v2016-02-01) .HLA typing

For HLA typing, adapters were removed using Trimmomatic (v0.39-1) . HLA-HD (v1.7.0) was employed for the allele assignment of HLA genes at both four-digit and six-digit resolution levels, encompassing both classical and non-classical HLA genes.Personalized panel design

For each patient, patient-specific somatic single nucleotide polymorphisms (SNPs) were identified through the analysis of primary tumor and matched normal whole exome sequencing (WES) samples. The SNPs were classified using in-house developed criteria based on variant allele frequency (VAF) and their impact on protein function. A maximum of 20 highly classified SNPs with a VAF of 5%or greater were selected for panel design. SNPs located in homologous regions were excluded from consideration. A custom primer design software, named versaTile, was utilized for the panel design.cfDNA library preparation and sequencing

The cfDNA library was prepared using a two-step PCR approach. In the first round of PCR amplification, specific primers containing tag sequences were employed to amplify designated target regions. Dimers were subsequently removed through magnetic bead purification. For the second round of PCR amplification, primers incorporating unique index sequences were utilized. The libraries underwent purification and quantification using Qubit^TM 3.0 (Thermo Fisher, Carlsbad, CA, USA) with the Equalbit 1 × dsDNA HS Assay Kit (EQ121; Vazyme, Nanjing, Jiangsu; China) . For each sample, one cfDNA library was prepared alongside three control libraries derived from matched white blood cells and one no-template control (NTC) library. All libraries were sequenced on the NOVAseq 6000 platform (Illumina) utilizing a paired-end sequencing mode of 2 × 150 bp, achieving an average target coverage of approximately 100,000x.Site-level variant calling and sample-level MRD callingAdapters were initially removed using Trimmomatic (v0.39-1) . Subsequently, clean reads were aligned to the human reference genome (UCSC, hg19) utilizing the BWA-MEM (v0.7.17-r1194-dirty) algorithm to generate a SAM file. This SAM file was then converted into BAM format using Samtools (v1.3.0) . Suspicious regions surrounding indels were realigned with GATK (v4.1.3.0) to produce an analysis-ready BAM file. Quality control information was obtained through BAMdst (v1.0.9) . Somatic variants were identified using Vardict (v1.8.2) . For each personalized tumor-specific variant, a site was classified as positive if the variant was exclusively detected in the cfDNA library with a variant allele frequency (VAF) of ≥ 0.02%, while remaining undetected in any of the three control libraries or the no-template control library; otherwise, it was considered negative. cfDNA samples exhibiting two or more positive single nucleotide variants (SNVs) were categorized as ctDNA-positive, and ctDNA concentration was reported as mean tumor molecules per mL of plasma.Statistical analysis

Sample size (n) represents the number of patients, samples, animals, or cells. Multiple statistical methods were used. Survival curves are analyzed by log -rank (Mantel–Cox) test. Unpaired two -tailed Mann–Whitney test compares two groups; for paired data, two -tailed student’s t -test or two -tailed Wilcoxon test were used. Kruskal–Wallis test is for multiple groups, ch-squared test for categorical variables, and Spearman correlation for related parameters. Error bars (as in figure legends) show SD. All analyses are done with GraphPad Prism 8.0 (USA) . For combined diagnosis data, Logistic regression is used. Medcalc (Belgium) plots ROC curve, calculates AUC, finds optimal cut -off, and determines diagnostic sensitivity and specificity. P < 0.05 is statistically significant. ****: P < 0.0001; ***: P < 0.001; **: P < 0.01; *: P < 0.05; ns: P > 0.05.SEQUENCE TABLE

Claims

A method of producing a plurality of folded MHC-peptide complexes, comprising:

a) adding a plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and

b) adding an MHC molecule into the refolding buffer incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes.
A method of identifying a pool of peptides suitable for cancer diagnosis, comprising:

a) generating a plurality of distinct peptides associated with one or more oncogenes;

b) adding the plurality of distinct peptides into a refolding buffer, wherein the concentration of the plurality of distinct peptides in the refolding buffer is at least about 0.01 mg/mL; and

c) adding a MHC molecule into the refolding buffer and incubating for at least one day; thereby producing a plurality of folded MHC-peptide complexes comprising a pool of peptides suitable for cancer diagnosis.
The method of claim 2, wherein the one or more oncogenes comprise the one or more oncogene listed in Tables 8A-8B or 4A.
The method of any one of claims 1-3, wherein refolding buffer comprises L-Arginine, oxidized glutathione, reduced glutathione, a chelating agent, a protease inhibitor, and a buffering agent, and wherein the refolding buffer has a pH of about 7.8 to about 8.2.
The method of claim 4, wherein the chelating agent is EDTA, the protease inhibitor is PMSF, and/or the buffering agent is Tris-HCl.
The method of any one of claims 1-5, wherein the refolding buffer comprises 80-125 mM Tris-HCl, 300-500 mM L-Arginine, 1.5-2.5 mM EDTA, 0.4-0.6 mM oxidized glutathione, 4.0-6.0 mM reduced glutathione, and 0.15-0.25 mM PMSF, wherein the molar ratio of oxidized glutathione and reduced glutathione is about 1: 5 to 1: 10.
The method of any one of claims 1-6, wherein the refolding buffer has a pH of 8.0.
The method of any one of claims 1-7, wherein the refolding buffer comprises about 100 mM Tris-HCl, 400 mM L-Arginine, 2 mM EDTA, 0.5 mM oxidized glutathione, 5.0 mM reduced glutathione, and 0.2 mM PMSF.
The method of any one of claims 1-8, wherein the MHC molecule consists of the same MHC molecule.
The method of any one of claims 1-8, wherein the MHC molecule comprises two or more distinct MHC molecules.
The method of any one of claims 1-10, wherein the MHC molecule comprises a MHC I molecule comprising an HLA molecule and a β-2M.
The method of claim 11, wherein the molar ratio of the HLA molecule and the β-2M is at least 0.8: 1 and less than 1.5: 1, optionally wherein the molar ratio of the HLA molecule and the β-2M is about 1: 1.
The method of claim 11 or claim 12, wherein the HLA molecule has a concentration of about 1-3 nmol/mL in the refolding buffer, optionally wherein the HLA molecule has a concentration of about 2 nmol/mL in the refolding buffer.
The method of any one of claims 11-13, wherein the β-2M has a concentration of about 1-3 nmol/mL in the refolding buffer, optionally wherein the β-2M has a concentration of about 2 nmol/mL in the refolding buffer.
The method of any one of claims 11-14, wherein the HLA molecule is added to the refolding buffer in batches for three times ( “the first batch, the second batch and the third batch in chronological order” ) , and wherein every adjacent two times are separated by at least 12-24 hours.
The method of claim 15, wherein the first HLA molecule batch has about half of the HLA molecule, and the second and third HLA molecule batch each has about a quarter of the HLA molecule.
The method of any one of claims 11-16, wherein the HLA molecule is selected from the group consisting of a HLA-Amolecule, a HLA-B molecule, a HLA-C molecule, a HLA-DP molecule, a HLA-DQ molecule, and a HLA-DR molecule, optionally wherein the HLA molecule is a HLA-Amolecule, further optionally wherein the HLA molecule is selected from the Table 10.
The method of any one of claims 11-17, wherein the HLA molecule comprises a spacer at the C-terminus of the HLA molecule, optionally wherein the spacer has a length of a peptide of about 5-50 amino acids, further optionally wherein the spacer has a length of a peptide of about 20-30 amino acids.
The method of claim 18, wherein the spacer comprises a peptide linker of at least five amino acids.
The method of claim 18 or claim 19, wherein the spacer comprises a peptide linker of 10-20 amino acids, optionally wherein the peptide linker is a GS linker, further optionally wherein the peptide linker is a (G4S) n linker, wherein n is any of 2, 3, and 4.
The method of any one of claims 18-20, wherein the spacer comprises a peptide tag, optionally wherein the peptide tag is an Avi-Tag, an AP-tag, a biotin acceptor domain (BAD) , a Sortase A (SrtA) tag, a His-tag, a FLAG-tag, or Strep-tag II.
The method of any one of claims 18-21, wherein the spacer comprises, from N-terminus to C-terminus, a) a peptide linker of about 10-20 amino acids, and b) an Avi-tag.
The method of any one of claims 1-22, wherein the incubation is at 4 ℃.
The method of any one of claims 11-23, wherein the HLA molecule and the β-2M are both recombinant proteins produced in E. coli.
The method of any one of claims 1-24, wherein the volume of the refolding buffer is about or at least about 50 mL.
The method of any one of claims 1-25, wherein the method further comprises adding a salt to the refolding buffer, optionally wherein the salt is NaCl, and further optionally wherein the final concentration of NaCl is about 200 mM.
The method of any one of claims 1-26, wherein the method further comprises a) concentrating folded MHC-peptide complexes via ultrafiltration and b) purifying folded MHC-peptide complexes via size-exclusion chromatography (SEC) , wherein the method does not comprise a buffer exchange i) via ultrafiltration or ii) between ultrafiltration and SEC.
The method of any one of claims 1-27, wherein the plurality of distinct peptides comprise at least about 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different peptides.
The method of claim 28, wherein the plurality of folded MHC-peptide complexes comprise at least about 40%, 45%, 50%, 55%, or 60%of the plurality of distinct peptides, optionally wherein the plurality of folded MHC-peptide complexes comprise about 50%to about 60%.
The method of any one of claims 1-29, wherein the plurality of distinct peptides comprises neoantigen peptides that are associated with a plurality of different mutations, at least 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 different mutations, wherein the plurality of different mutations comprise mutations (e.g., driver mutations) on the one or more oncogenes.
The method of claim 30, wherein the plurality of folded MHC-peptide complexes comprise peptides that target at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%or 85%of the plurality of different mutations, optionally wherein the plurality of folded MHC-peptide complexes comprise about 80%to about 85%of the plurality of different mutations.
The method of claim 30 or 31, wherein the oncogenes comprise one or more oncogenes selected from Tables 8A-8B or 4A.
The method of any one of claims 30-32, wherein the plurality of mutations comprise one or more mutations selected from Tables 8A-8B or 4B.
The method of any one of claims 1-33, wherein the plurality of folded MHC-peptide complexes comprise one or more peptides that have a KD of more than about any of 10 nM, 100 nM, 1μM, 10 μM, 20 μM, 30 μM, 40 μM, 50 μM, 60 μM, 70 μM, 80 μM, 90 μM, or 100 μM.