Neoantigen polypeptide vaccines induce effective antitumor response in colorectal cancer

Background: The role of neoantigens in cancer immunotherapy is crucial. However, the effectiveness and safety of personalized neoantigen vaccines in colorectal cancer (CRC), especially in Chinese population, has not been well studied. This paper mainly explores the feasibility and effectiveness of personalized neoantigen vaccines in CRC treatment. Methods: Whole-exome sequencing and transcriptome sequencing were used to identify somatic mutations, RNA expression and human leukocyte antigen (HLA) alleles. Neoantigens were predicted, and the immunogenicity of neoantigen candidates was evaluated by ELISPOT in vitro. To verify the immunogenicity in vivo , neoantigen candidates from HLA-A0201 + PW11 were used to immunized female 6-8-week-old HLA-A2.1/K b -transgenic (Tg) mice. Neoantigen-reactive T cells (NRTs) were induced by immunogenic peptides from autologous HLA-A2.1/K b to adoptive transfer transgenic mice, and C57BL/6 nu/nu mice were used for in vivo antitumor response assays. Results: Compared to medium alone (no peptide) or the unrelated peptide VSV-NP 43-69 , the neoantigens TSHZ3-L523P, RARA-R83H, TP53-R248W, EYA2-V333I and NRAS-G12D from Patient 4 (PW4); HAVCR2-F39V, SEC11A-R11L, TASP1-P161L, RAP1GAP-S215R, MOSPD1-V63I and NAV2-D1973N from Patient 10 (PW10); and SMPDL3B-T452M, LRFN3-R118Q and ULK1-S248L from Patient 11 (PW11) induced notable peptide-specific T cell responses. The results indicated that about half of the predicted neoantigens for all 3 patients can stimulate T cell responses and antitumor effects in CRC. In addition, predicted neoantigens from PW11 (HLA-A0201) showed promising antitumor efficacy in HLA-A2.1/K b -Tg mice and tumor-bearing mouse models. Conclusion: With the application of next-generation sequencing (NGS) sequencing of patient specimens, neoantigen prediction and a rapid immunoassay system, an evaluation system utilizing in vitro studies and in vivo transgenic and tumor-bearing mouse models can be used to screen strong immunogenic neoantigens in CRC patients. Accurate identification of neoantigens with strong immunogenicity would promote personalized cancer vaccine development. vaccines. Tumor tissue samples and peripheral blood cells were collected for neoantigens identification. Antitumor effects of personalized neoantigen vaccines for CRC patients were verified by an neoantigen-reactive T cell (NRT)-induced in vitro cytotoxicity assay and in vivo experiments in tumor-bearing mice model. The results indicated that about half of the predicted from fresh frozen tumor samples. The construction of RNA-seq libraries were performed using the TruSeq Stranded Autologous PBMCs from a patient were used to assess the in vitro immunogenicity of candidate neoantigens. We used a simple and rapid culture method to detect and monitor circulating antigen peptide-specific cytotoxic T lymphocyte precursors (CTL-P), which was only slightly modified in previous reports(20, 21). Peripheral blood was obtained from patients that had been undergoing postoperative chemoradiotherapy for more than 1 month. PBMCs were isolated and cultured. The human PBMCs were suspended in complete medium (RPMI 1640 medium containing 10% foetal bovine serum) for collection. PBMC-derived DCs were harvested on the sixth day after cultivation, adjusted for cell concentration using a complete medium for human DCs (RPMI 1640 complete medium supplemented with 500 U/ml rhGM-CSF and 10 ng/ml th-4), Frozen T cells collected in RPMI 1640 medium were suspended in 10% foetal bovine serum and co-culture for 3 stimulations. The cells were collected 7 days after the last stimulation, and positive selection with CD3 immunomagnetic beads (Miltenyi Biotec) was performed. The T cells were selected in strict accordance with the instructions provided by the manufacturer. In vitro gene transcription using the mMESSAGE mMACHINE UItra Kit (Thermo Fisher) kit was performed. Minigene RNA was transferred into PBMCs by a nuclear transfection kit (Amaxa Cell Line Nucleofector Kit V, Lonza). On day 10, the specific response of T cells to each peptide was assessed by ELISPOT assays. Comparing the response induced by the peptide to that induced by medium alone (no peptide) or the unrelated peptide VSV-NP 43 − 69 (STKVALNDLRAYVYQGIKSGNPSILHI), and phytohemagglutinin was used as the positive control. T cell reactivity was assessed by DCs peptide pulses co-cultured with T cells in some cases. Finally, T cell activity was detected by an IFN-y ELISPOT, an ELISA kit and intracellular flow cytometry staining. TD, Gartner JJ, Jia L, et al. Unique Neoantigens Arise from Somatic Mutations in Patients with Gastrointestinal Cancers.


Introduction
Colorectal cancer (CRC) is the third most common cause of cancer-related death worldwide (1). At present, the incidence of CRC ranks fifth among systemic malignant tumors in China (2). Surgery is the main treatment strategy for CRC patients, but adjuvant chemotherapy is available only for patients with stage II and most patients with stage III CRC and stage II or III rectal cancer (3,4). Survival has increased over the past 30 years with the introduction of screening programmes and development of new targeted drugs. However, the 5-year relative survival rate of CRC is still only 68%. It is worth noting that approximately one-third of patients undergoing curative surgery for CRC develop recurrent disease because of incomplete tumor resection, and the standard therapies of surgery and adjuvant chemotherapy are often limited by side effects and resistance to chemotherapy (5,6). Therefore, it is highly important to develop other strategies for CRC treatment.
In recent years, immunotherapy for cancer treatment has made great progress (7,8). However, only < 5% of patients with advanced CRC benefited from programmed cell death protein 1 (PD-1) checkpoint blocking immunotherapy and the efficacy in CRC was limited (9). Personalized neoantigen vaccines, as a form of cancer immunotherapy, have enormous potential to induce potent and long-lasting antitumor effects. With the advent and availability of next-generation sequencing (NGS), the identification of patient-specific neoantigens for personalized vaccines has become feasible.
Neoantigen vaccines were shown to be immunogenic and safe with promising clinical effects in several clinical trials of melanoma, glioblastoma and other types of cancer (10,11). In CRC, HLA-C*08:02-restricted tumor-infiltrating lymphocytes (TIL) immunotherapy targeting oncogene mutation KRAS G12D has been shown to promote tumor regression in one patient (12). However, the effectiveness and safety of personalized neoantigen vaccines in CRC, especially in Chinese population, has not been well studied, so we developed an effective personalized vaccination programme for CRC patients to achieve a significant increase in antitumor effects.
In this research, 13 CRC patients were recruited to study the feasibility and effectiveness of neoantigen polypeptide vaccines. Tumor tissue samples and peripheral blood cells were collected for neoantigens identification. Antitumor effects of personalized neoantigen vaccines for CRC patients were verified by an neoantigen-reactive T cell (NRT)-induced in vitro cytotoxicity assay and in vivo experiments in tumor-bearing mice model. The results indicated that about half of the predicted neoantigens for 3 patients can stimulate T cell responses and antitumor effects in CRC. This study shows that neoantigens can be used to create an ideal cancer vaccine for CRC, thus establishing a path for personalized immunotherapy for CRC patients.

Materials And Methods Patient material and cell lines
Patient tumor samples were obtained from biopsy specimens. A portion of each sample was subjected to formalin fixation and paraffin embedding, and the remaining tumor tissue was carefully manually Cell lines were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). Peripheral blood mononuclear cells (PBMCs) were the original material harvested for immunological monitoring or sample manufacturing; these cells were isolated by density gradient centrifugation using Ficoll-Hypaque (Amersham Biosciences) and collected from the buffy coat of peripheral blood samples from healthy donors or CRC patients, thereby producing immature dendritic cells (DCs). H522 cells were cultured under standard conditions described by the ATCC, and the cells were re-certified by short tandem repeat (STR) analysis at ATCC, while a cell bank for mycoplasma was generated.
Next-generation Sequencing (ngs) For

Bioinformatics And Mutation Discovery
All mutations in a single patient were analysed using the Python programming language. DNA libraries required at least 150 × 10 6 paired-end 50 nt reads, while RNA libraries required at least 75 × 10 6 paired-end 50 nt reads. For mutation detection, DNA reads were consistent with the reference genomes hg19. Duplicate exomes from tumor and matched normal samples were analysed for singlenucleotide variants. Identification and screening of sites assumed to be homozygous genotypes in normal samples, thus preserving a high degree of trust in single-nucleotide variation. Further examination of the remaining sites was performed to determine the presumed homozygous or heterozygous mutation event. Screening suspicious sites to exclude potential false positives, and testing the sum of duplicates and duplicates respectively, and merging duplicates. We compared the genomic coordinates of the identified variants with the detailed known gene transcription coordinates in the UCSC, and further determined the relationship between the mutations and genes, transcription, expression values derived from RNA-seq, and the changes of potential amino acid sequences.

Hla Typing
The HLA alleles of each patient were inferred from the WES data using OptiType (13) with the default settings after the reads were filtered by aligning to the HLA region using RazerS version 3.4.0 (14).

Neoantigen Candidates Identification
Non-synonymous somatic mutations, including SNVs, insertion, deletion and frame shift, were used to predict neoantigens by MuPeXI (15), which is a pipeline for neoantigen identification. First, 8-11amino-acid long mutant peptides were generated for MHC I-restricted neoantigen prediction, and 12-15-amino-acid long mutant peptides were generated for MHC II-restricted neoantigen prediction. Then HLA binding affinity for each peptide was calculated by NetMHCpan4.0(16) and NetMHCIIpan3.2 (17) for MHC I and MHC II molecules respectively. Expression level of mutant genes was derived from RNAseq data in transcripts per million (TPM). The number of mismatches between the mutant and normal peptides was considered as the similarity to self-peptides. Mutant allele frequency was detected by the variant caller MuTect2 (18). Each peptide was given a priority score based on HLA binding affinity, expression level, similarity to self-peptides and mutant allele frequency. Peptides with priority score larger than 0 were selected as neoantigen candidates.

Synthesis Of Long Peptidesand Final Vaccine Preparation
The peptides with purity greater than 95% were analysed by mass spectrometry and were synthesized by GL Biochem (Shanghai, China) using fluorenylmethoxycarbonyl chemistry and purified by reversed-phase high-performance liquid chromatography. The lyophilized peptides were dissolved in dimethyl sulfoxide and diluted to a concentration of 10 mM in phosphate buffer brine (pH 7.4), and continued to be stored as isocyanates at − 80 °C.

Minigenes (tmgs)
The construction process for tandem minigenes (TMGs) has been described previously (19). In each TMG construction, 12 small genes were included. A TMG plasmid was constructed by linearization with the restriction enzyme NsiI, and in vitro RNA transcription was performed by using each of the encoded linearized plasmids as a template with the mMESSAGE mMACHINE T7 Transcription Kit (Thermo Fisher Scientific); 20 mg of RNA was purified for cell transfection. The remaining PBMC cells after cryopreservation of removal of DC/T cells in this section were thawed and cultured, and minigene RNA was introduced into the PBMCs using a nuclear transfection kit (Amaxa Cell Line Nucleofector Kit V, Lonza). After 24 hours of transfection, these cells were used as stimulatory cells, and the above procedure was performed strictly in accordance with the manufacturer's instructions.
Induction of neoantigen-reactive T cells by coculture with CRC patient-derived peripheral blood lymphocytes (PBLs) in vitro Autologous PBMCs from a patient were used to assess the in vitro immunogenicity of candidate neoantigens. We used a simple and rapid culture method to detect and monitor circulating antigen peptide-specific cytotoxic T lymphocyte precursors (CTL-P), which was only slightly modified in previous reports (20,21). Peripheral blood was obtained from patients that had been undergoing postoperative chemoradiotherapy for more than 1 month. PBMCs were isolated and cultured. The human PBMCs were suspended in complete medium (RPMI 1640 medium containing 10% foetal bovine serum) for collection. PBMC-derived DCs were harvested on the sixth day after cultivation, adjusted for cell concentration using a complete medium for human DCs (RPMI 1640 complete medium supplemented with 500 U/ml rhGM-CSF and 10 ng/ml th-4), Frozen T cells collected in RPMI 1640 medium were suspended in 10% foetal bovine serum and co-culture for 3 stimulations. The cells were collected 7 days after the last stimulation, and positive selection with CD3 immunomagnetic beads (Miltenyi Biotec) was performed. The T cells were selected in strict accordance with the instructions provided by the manufacturer. In vitro gene transcription using the mMESSAGE mMACHINE UItra Kit (Thermo Fisher) kit was performed. Minigene RNA was transferred into PBMCs by a nuclear transfection kit (Amaxa Cell Line Nucleofector Kit V, Lonza). On day 10, the specific response of T cells to each peptide was assessed by ELISPOT assays. Comparing the response induced by the peptide to that induced by medium alone (no peptide) or the unrelated peptide VSV-NP 43 − 69 (STKVALNDLRAYVYQGIKSGNPSILHI), and phytohemagglutinin was used as the positive control. T cell reactivity was assessed by DCs peptide pulses co-cultured with T cells in some cases.
Finally, T cell activity was detected by an IFN-y ELISPOT, an ELISA kit and intracellular flow cytometry staining.
Generation Of Neoantigen-specific T Cells In Hla-a2.1/k-transgenic Mice All experimental procedures and animal protocols were approved by the AAALAC-accredited Animal Studies Committee of Wenzhou Medical University (AEC numbers:WYDW2019-0924) and were in compliance with all relevant ethical regulations. All procedures are designed in accordance with the ethical standards of the national research council to minimize the suffering and number of animals. All animals used in this study were purchased from the Jackson Laboratory (Bar Harbor, ME, USA).
Environmental conditions were a temperature of 20 ± 2 °C, the humidity of 55% ±10%, the lighting of 350 lux (at bench level) and a 12/12 h light/dark cycle and All the mice were fed the standard rodent diet with free access to water. The health of the animals was monitored twice daily during the feeding period. No adverse events were observed. All parts of the report follow the ARRIVE Guidelines for animal studies (22).
For the in vivo efficacy evaluation studies, we used eight 6-8-week-old female HLA-A2.1/K b -Tg mice (mean body weight: 20 ± 2 g), which were raised under standard conditions and cared for in accordance with the animal care facility guidelines (23). Tumor neoantigen peptides from patients with HLA-A0201 were selected and synthesized to vaccinate HLA-A2.1/K b transgenic mice subcutaneously at 0,7 days. 50 mg poly (I: C) was mixed with 100 mg polypeptide each time. Seven days after the last immunization, the spleen of mice was extracted by aseptic operation, and the residual blood was rinsed with sterile saline, soaked in RPMI1640 medium, and the single-cell suspension was obtained by gently grinding the needle core of aseptic syringe on 400 mesh steel net, and filtered through steel net to remove large tissue blocks. After centrifuging 1000 × g of the filtered single-cell suspension for 5 minutes, the supernatant of the culture medium was discarded and suspended in the low permeable NH 4 Cl solution (0.15 M NH 4 Cl, 1 M KHCO 3 , 0.1 mM Na 2 EDTA, pH 7.2) for 3 minutes to destroy the red blood cells, and then centrifuged and washed twice with RPMI1640 culture medium. The spleen cells of the mice without red blood cells were suspended in RPMI1640 medium containing 10% fetal bovine serum, counted and placed in 6-well plates. Mouse spleen cells were cultured in vitro and stimulated again with corresponding peptide (20 µM). After 7 days, the effector cells were collected and analysed by ELISPOT, intracellular staining and specific killing activity in tumor-bearing mice.

Elispot Assay
An ELISPOT assay (R&D Systems, Minneapolis, MN) was performed using a commercial kit. T cells sorted by CD3 immunomagnetic beads were suspended in RPMI 1640 medium supplemented with 10% foetal bovine serum and used as reacting cells, and the cell concentration was adjusted to 5 × 10 6 /ml, and the cell suspension was directly transferred to an ELISPOT assay plate (100 µl/well) coated with an anti-IFN-y antibody, PBMC (mutant type (Mut) minigenes), PBMC (wild-type (WT) minigenes) 15 Gy irradiation was used as a stimulating cell and was added to the detection well (100 ul/well). IFN-y-secreting T cell colony was detected by the method described in the IFN-y ELISPOT assay kit instructions. The frequency of nocturnal peptide (80) activation in cytokine-secreting cells was determined by an IFN-γ ELISPOT kit (Dakewei). In this research, multiple culture protocols were used to analyse T cell-mediated immune responses. Peptide-stimulated PBMCs were added to two wells for approximately 18 hours or DCs pulsed with a peptide were added and cocultured with the T cells. The plate was washed first, and then a diluted test antibody (1:100 dilution) was added and incubated at 37 °C for 1 hour. After washing the plate, streptavidin-HRP diluted to the above mentioned year-on-year dilution was added and incubated at the same temperature for 1 hour, and a 3-amino-9-ethylcarbazole (AEC) solution was prepared according to the manufacturers specifications, and added per well. The plate was allowed to incubate at room temperature for approximately [15][16][17][18][19][20] minutes in darkness, after which time development was stopped by the addition of deionized water.
The plates were scanned with an ELISPOT CTL Reader (Cellular Technology Inc) and analysed by ElisPot software (AID), and spots two-times larger than those in the peptide-free (culture medium) control were considered to be T cell reactive positive.

Intracellular Detection Of Ifn-γ
The detection method used to evaluate intracellular IFN-γ production was performed as described above. Briefly, in RPMI 1640 medium supplemented with 10% foetal bovine serum, frozen T cells collected previously were suspended and adjusted to a cell concentration of 4 × 10 6 /ml, and 0.5 ml was added to each of the above collected peptide-sensitized autologous DCs, and co-cultured at 37 °C, 5% carbon dioxide (T:DC ratio of 10:1), and the same stimulation was provided three times per week. Recombinant IL-2 (rhIL-2; 20 U/ml) was added every 3 days during the culture, the medium was

Statistical analysis
The two-tailed Student's t-test was used to determine statistical significance of the differences between means. Tumor sizes were compared between groups using the Mann-Whitney U-test.
Survival analyses were plotted using Kaplan-Meier curves. Values of P < 0.05 were considered as statistically significant. All statistical tests were performed using SPSS manager software (v21; IBM Inc.)

Somatic mutation identification and neoantigen prediction for CRC patients
We recruited 13 patients with CRC (Table 1) Table 2). Among all predicted neoantigens, a median of 9 neoantigens (range 5-18) are strong binders with %rank less than 0.5, and a median of 46 neoantigens (range 22-72) are weak binders with %rank larger than 0.5 (Fig. 1B

Neoantigen-specific T-cell responses can be induced in vitro in PBLs from CRC patients
Adoptive T cell therapy has successfully been applied to treat many human solid tumors. There is growing evidence that NRTs can achieve tumor regression in patients receiving adoptive cell therapy (26,27). In addition, to evaluate the ability of NRTs to respond to Mut and WT peptides, we  (Fig. 4A). In a cytotoxicity assay, the NRTs against SEC11A-R11L and ULK1-S248L NRTs significantly killed T2 cells loaded with the corresponding mutant peptide and SW480-minigene expressing the corresponding antigen. No significant killing was detected for T2 cells that were not loaded with the peptide or loaded with the irrelevant peptide VSV-NP 43 − 69 or SW480 cells that did not express the mutant peptide (Fig. 4C, 4D).
NRTs did not exhibit significant cytotoxicity in response to HAVCR2-F39V (Fig. 4B). No adverse reactions were observed in mice throughout the experiment.
Adoptive NRT immunotherapy of C57BL/6 nu/nu mice bearing human CRC tumors To determine whether the SEC11A-R11L and ULK1-S248L peptides can be used as a potent vaccine to limit tumor growth in vivo. We established HLA-A2.1/K b -Tg mouse-derived NRTs, which were adoptively transferred into SW480-minigene human CRC-bearing C57BL/6 nu/nu mice. Thus, we developed an adoptive transfer therapy model in nude mice. As shown in Fig. 5A, the mice subjected to adoptive transfer of SEC11A-R11L and ULK1-S248L-stimulated NRTs exhibited significantly delayed tumor growth, whereas CTLs induced by irrelevant peptide could not prevent tumor growth. In addition, in the SEC11A-R11L and ULK1-S248L vaccination groups, 7/10 (70%) of the ULK1-S248Lvaccinated mice and 6/10 (60%) of the SEC11A-R11-vaccinated mice showed significant long-term survival after tumor inoculation (over 80 days) (Fig. 5B). All the mice of control group died on days 20 to 40. Therefore, these results indicate that immunization with the SEC11A-R11L and ULK1-S248L peptides induces an effective antitumor response in vivo. In this study, there were too few mice to determine whether these represented different tissue cell types that respond to neoantigen polypeptide. No adverse reactions were found in the experimental and control mice.

Discussion
Cancer immunotherapy is rapidly evolving and has been effectively translated into the clinical.
Personalized immunotherapy has demonstrated the ability to enhance NRT cell reactivity, which will further enhance the antitumor effect of cancer immunotherapy on malignant tumors (28) G12D that could stimulate CD8 + T-cell response (12). The method of identifying neoantigens was laborious and time-consuming. Thus, in our study, we used computational tools to identify neoantigens from hundreds of mutations. However, one of the most crucial problems of in silico neoantigen identification is the low immunogenicity of predicted neoantigen candidates. State-of-art tools that regard binding affinity between peptide and MHC molecules as the main consideration, such as NetMHCpan and NetMHCIIpan, can only achieve a low validation rate about 20-30% (29). Therefore, we used MuPeXI which is a computational tool that integrates binding affinity, RNA expression level, mutant allele frequency and similarity to self-peptidesas factors to determine the probability of neoantigen candidates to activate immune responses (15). According to our results, the validation rate increased to about 50% by using MuPeXI to predict neoantigens.
Although we and previous studies have demonstrated the immunogenicity and antitumor effects of the neoantigens (10,11,30,31), the neoantigen-based vaccine is less effective against solid tumors,

Availability of data and materials
The dataset(s) supporting the findings of this study are included within the article.

Ethics approval and consent to participate
In this study, all investigation and experiments have obtained patients' consent and been approved by the Medical Ethics Committee of the Second Affiliated Hospital, Wenzhou Medical University

Consent for publication
Written informed consent for publication was obtained from all participants.