Quantitative T-cell repertoire analysis of peripheral blood mononuclear cells from lung cancer patients following long-term cancer peptide vaccination

BERTrand-peptide:TCR binding prediction using Bidirectional Encoder Representations from Transformers augmented with random TCR pairing

MOTIVATION

The advent of T-cell receptor (TCR) sequencing experiments allowed for a significant increase in the amount of peptide:TCR binding data available and a number of machine-learning models appeared in recent years. High-quality prediction models for a fixed epitope sequence are feasible, provided enough known binding TCR sequences are available. However, their performance drops significantly for previously unseen peptides.

RESULTS

We prepare the dataset of known peptide:TCR binders and augment it with negative decoys created using healthy donors' T-cell repertoires. We employ deep learning methods commonly applied in Natural Language Processing to train part a peptide:TCR binding model with a degree of cross-peptide generalization (0.69 AUROC). We demonstrate that BERTrand outperforms the published methods when evaluated on peptide sequences not used during model training.

AVAILABILITY AND IMPLEMENTATION

The datasets and the code for model training are available at https://github.com/SFGLab/bertrand.

Analysis of thymic generation of shared T-cell receptor α repertoire associated with recognition of tumor antigens shows no preference for neoantigens over wild-type antigens

BACKGROUND

The number of mutations in cancer cells is an important predictor of a positive response to cancer immunotherapy. It has been suggested that the neoantigens produced by these mutations are more immunogenic than nonmutated tumor antigens, which are likely to be protected by immunological tolerance. However, the mechanisms of tolerance as regards tumor antigens are incompletely understood.

METHODS

Here, we have analyzed the impact of thymic negative selection on shared T-cell receptor (TCR) repertoire associated with the recognition of either mutated or nonmutated tumor antigens by comparing previously known TCR-antigen-pairs to TCR repertoires of 21 immunologically healthy individuals.

RESULTS

Our results show that TCRα chains associated with either type of tumor antigens are readily generated in the thymus, at a frequency similar to TCRα chains associated with nonself. In the peripheral repertoire, the relative clone size of nonself-associated chains is higher than that of the tumor antigens, but importantly, there is no difference between TCRα chains associated with mutated or nonmutated tumor antigens.

CONCLUSION

This suggests that the tolerance mechanisms protecting nonmutated tumor antigens are non-deletional and therefore potentially reversible. As unmutated antigens are, unlike mutations, shared by a large number of patients, they may offer advantages in designing immunological approaches to cancer treatment.

Distinct Predictive Immunogenomic Profiles of Response to Immune Checkpoint Inhibitors and IL2: A Real-world Evidence Study of Patients with Advanced Renal Cancer

Recombinant human high-dose IL2 (HD-IL2; aldesleukin) was one of the first approved immune-oncology agents based upon clinical activity in renal cell carcinoma (RCC) and metastatic melanoma but use was limited due to severe toxicity. Next-generation IL2 agents designed to improve tolerability are in development, increasing the need for future identification of genomic markers of clinical benefit and/or clinical response. In this retrospective study, we report clinical and tumor molecular profiling from patients with metastatic RCC (mRCC) treated with HD-IL2 and compare findings with patients with RCC treated with anti-PD-1 therapy. Genomic characteristics common and unique to IL2 and/or anti-PD-1 therapy response are presented, with insight into rational combination strategies for these agents. Residual pretreatment formalin-fixed paraffin embedded tumor samples from n = 36 patients with HD-IL2 mRCC underwent RNA-sequencing and corresponding clinical data were collected. A de novo 40-gene nearest centroid IL2 treatment response classifier and individual gene and/or immune marker signature differences were correlated to clinical response and placed into context with a separate dataset of n = 35 patients with anti-PD-1 mRCC. Immune signatures and genes, comprising suppressor and effector cells, were increased in patients with HD-IL2 clinical benefit. The 40-gene response classifier was also highly enriched for immune genes. While several effector immune signatures and genes were common between IL2 and anti-PD-1 treated patients, multiple inflammatory and/or immunosuppressive genes, previously reported to predict poor response to anti-PD-L1 immunotherapy, were only increased in IL2-responsive tumors. These findings suggest that common and distinct immune-related response markers for IL2 and anti-PD-1 therapy may help guide their use, either alone or in combination.

Significance

Next-generation IL2 agents, designed for improved tolerability over traditional HD-IL2 (aldesleukin), are in clinical development. Retrospective molecular tumor profiling of patients treated with HD-IL2 or anti-PD-1 therapy provides insights into genomic characteristics of therapy response. This study revealed common and distinct immune-related predictive response markers for IL2 and anti-PD-1 therapy which may play a role in therapy guidance, and rational combination strategies for these agents.

Cancer/Testis Antigens as Biomarker and Target for the Diagnosis, Prognosis, and Therapy of Lung Cancer

Lung cancer is the leading type of malignant tumour among cancer-caused death worldwide, and the 5-year survival rate of lung cancer patients is only 18%. Various oncogenes are abnormally overexpressed in lung cancer, including cancer/testis antigens (CTAs), which are restrictively expressed in the male testis but are hardly expressed in other normal tissues, if at all. CTAs are aberrantly overexpressed in various types of cancer, with more than 60 CTAs abnormally overexpressed in lung cancer. Overexpression of oncogenic CTAs drives the initiation, metastasis and progression of lung cancer, and is closely associated with poor prognosis in cancer patients. Several CTAs, such as XAGE, SPAG9 and AKAP4, have been considered as biomarkers for the diagnosis and prognostic prediction of lung cancer. More interestingly, due to the high immunogenicity and specificity of CTAs in cancer, several CTAs, including CT45, BCAP31 and ACTL8, have been targeted for developing novel therapeutics against cancer. CTA-based vaccines, chimeric antigen receptor-modified T cells (CAR-T) and small molecules have been used in lung cancer treatment in pre-clinical and early clinical trials, with encouraging results being obtained. However, there are still many hurdles to be overcome before these therapeutics can be routinely used in clinical lung cancer therapy. This review summarises the recent rapid progress in oncogenic CTAs, focusing on CTAs as biomarkers for lung cancer diagnosis and prognostic prediction, and as targets for novel anti-cancer drug discovery and lung cancer therapy. We also identify challenges and opportunities in CTA-based cancer diagnosis and treatment. Finally, we provide perspectives on the mechanisms of oncogenic CTAs in lung cancer development, and we also suggest CTAs as a new platform for lung cancer diagnosis, prognostic prediction, and novel anti-cancer drug discovery.

T cell receptor repertoire characteristics both before and following immunotherapy correlate with clinical response in mesothelioma

Background

Malignant pleural mesothelioma (MPM) is a highly lethal malignancy in need for new treatment options. Although immunotherapies have been shown to boost a tumor-specific immune response, not all patients respond and prognostic biomarkers are scarce. In this study, we determined the peripheral blood T cell receptor β (TCRβ) chain repertoire of nine MPM patients before and 5 weeks after the start of dendritic cell (DC)-based immunotherapy.

Materials and methods

We separately profiled PD1⁺ and PD1⁻CD4⁺ and CD8⁺ T cells, as well as Tregs and analyzed 70 000 TCRβ sequences per patient.

Results

Strikingly, limited TCRβ repertoire diversity and high average clone sizes in total CD3⁺ T cells before the start of immunotherapy were associated with a better clinical response. To explore the differences in TCRβ repertoire prior-DC-therapy and post-DC-therapy, for each patient the TCRβ clones present in the total CD3⁺ T cell fractions were classified into five categories, based on therapy-associated frequency changes: expanding, decreasing, stable, newly appearing and disappearing clones. Subsequently, the presence of these five groups of clones was analyzed in the individual sorted T cell fractions. DC-therapy primarily induced TCRβ repertoire changes in the PD1⁺CD4⁺ and PD1⁺CD8⁺ T cell fractions. In particular, in the PD1⁺CD8⁺ T cell subpopulation we found high frequencies of expanding, decreasing and newly appearing clones. Conversion from a PD1⁻ to a PD1⁺ phenotype was significantly more frequent in CD8⁺ T cells than in CD4⁺ T cells. Hereby, the number of expanding PD1⁺CD8⁺ T cell clones—and not expanding PD1⁺CD4⁺ T cell clones following immunotherapy positively correlated with overall survival, progression-free survival and reduction of tumor volume.

Conclusion

We conclude that the clinical response to DC-mediated immunotherapy is dependent on both the pre-existing TCRβ repertoire of total CD3⁺ T cells and on therapy-induced changes, in particular expanding PD1⁺CD8⁺ T cell clones. Therefore, TCRβ repertoire profiling in sorted T cell subsets could serve as predictive biomarker for the selection of MPM patients that benefit from immunotherapy.

Trial registration number

NCT02395679.

Progress and challenges of immunotherapy in triple-negative breast cancer

Triple-negative breast cancer (TNBC), a subtype of breast cancer, is defined as lacking estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor-2 (HER2) expression. Compared with other subtypes in breast cancer, TNBC is more likely to recur and metastasize, with a lower survival rate. Due to the absence of definitive targets, there was limited novel therapeutic interventions and chemotherapy remained the primary treatment in the past decades. Following the development of immune checkpoint inhibition (ICI) in solid tumors and validation of the immunogenicity in TNBC, immunotherapy has attracted more and more attentions. On basis of accumulating clinical studies, we reviewed the current progress targeting different immune checkpoints in several-lines treatment for TNBC, including programmed death-1 (PD-1)/programmed death ligand-1 (PD-L1) inhibitors, cytotoxic T-lymphocyte associated antigen-4 (CTLA-4) inhibitor, and other novel immunotherapeutic approaches (e.g., individualized peptide vaccine, cancer-testis antigen (CTA), new antigen vaccine, RNA vaccine and chimeric antigen receptor modified T cells (CAR-T)). In order to improve the survival outcome of TNBC populations, we further discussed potential predictive biomarkers for immunotherapy (e.g., PD-L1 expression, tumor mutational burden (TMB), tumor-infiltrating lymphocytes (TILs), microsatellite instability (MSI)/mismatch repair (MMR) deficiency) and challenges in the future treatment of TNBC.

Stimulation of Oncogene-Specific Tumor-Infiltrating T Cells through Combined Vaccine and αPD-1 Enable Sustained Antitumor Responses against Established HER2 Breast Cancer

PURPOSE

Despite promising advances in breast cancer immunotherapy, augmenting T-cell infiltration has remained a significant challenge. Although neither individual vaccines nor immune checkpoint blockade (ICB) have had broad success as monotherapies, we hypothesized that targeted vaccination against an oncogenic driver in combination with ICB could direct and enable antitumor immunity in advanced cancers.

EXPERIMENTAL DESIGN

Our models of HER2⁺ breast cancer exhibit molecular signatures that are reflective of advanced human HER2⁺ breast cancer, with a small numbers of neoepitopes and elevated immunosuppressive markers. Using these, we vaccinated against the oncogenic HER2Δ16 isoform, a nondriver tumor-associated gene (GFP), and specific neoepitopes. We further tested the effect of vaccination or anti-PD-1, alone and in combination.

RESULTS

We found that only vaccination targeting HER2Δ16, a driver of oncogenicity and HER2-therapeutic resistance, could elicit significant antitumor responses, while vaccines targeting a nondriver tumor-specific antigen or tumor neoepitopes did not. Vaccine-induced HER2-specific CD8⁺ T cells were essential for responses, which were more effective early in tumor development. Long-term tumor control of advanced cancers occurred only when HER2Δ16 vaccination was combined with αPD-1. Single-cell RNA sequencing of tumor-infiltrating T cells revealed that while vaccination expanded CD8 T cells, only the combination of vaccine with αPD-1 induced functional gene expression signatures in those CD8 T cells. Furthermore, we show that expanded clones are HER2-reactive, conclusively demonstrating the efficacy of this vaccination strategy in targeting HER2.

CONCLUSIONS

Combining oncogenic driver targeted vaccines with selective ICB offers a rational paradigm for precision immunotherapy, which we are clinically evaluating in a phase II trial (NCT03632941).

Immunization With the CSF-470 Vaccine Plus BCG and rhGM-CSF Induced in a Cutaneous Melanoma Patient a TCRβ Repertoire Found at Vaccination Site and Tumor Infiltrating Lymphocytes That Persisted in Blood

The CSF-470 cellular vaccine plus BCG and rhGM-CSF increased distant metastases-free survival in cutaneous melanoma patients stages IIB-IIC-III relative to medium dose IFN-α2b (CASVAC-0401 study). Patient-045 developed a mature vaccination site (VAC-SITE) and a regional cutaneous metastasis (C-MTS), which were excised during the protocol, remaining disease-free 36 months from vaccination start. CDR3-TCRβ repertoire sequencing in PBMC and tissue samples, along with skin-DTH score and IFN-γ ELISPOT assay, were performed to analyze the T-cell immune response dynamics throughout the immunization protocol. Histopathological analysis of the VAC-SITE revealed a highly-inflamed granulomatous structure encircled by CD11c+ nested-clusters, brisk CD8+ and scarce FOXP3+, lymphocytes with numerous Langhans multinucleated-giant-cells and macrophages. A large tumor-regression area fulfilled the C-MTS with brisk lymphocyte infiltration, mainly composed of CD8+PD1+ T-cells, CD20+ B-cells, and scarce FOXP3+ cells. Increasing DTH score and IFN-γ ELISPOT assay signal against the CSF-470 vaccine-lysate was evidenced throughout immunization. TCRβ repertoire analysis revealed for the first time the presence of common clonotypes between a VAC-SITE and a C-MTS; most of them persisted in blood by the end of the immunization protocol. In vitro boost with vaccine-lysate revealed the expansion of persistent clones that infiltrated the VAC-SITE and/or the C-MTS; other persistent clones expanded in the patient's blood as well. We propose that expansion of such persistent clonotypes might derive from two different although complementary mechanisms: the proliferation of specific clones as well as the expansion of redundant clones, which increased the number of nucleotide rearrangements per clonotype, suggesting a functional antigenic selection. In this patient, immunization with the CSF-470 vaccine plus BCG and rhGM-CSF induced a T-cell repertoire at the VAC-SITE that was able to infiltrate an emerging C-MTS, which resulted in the expansion of a T-cell repertoire that persisted in blood by the end of the 2-year treatment.

Application of immune repertoire sequencing in cancer immunotherapy

With the prominent breakthrough in the field of tumor immunology, diverse cancer immunotherapies have attracted great attention in the last decade. The immune checkpoint inhibitors, adoptive cell therapies, and therapeutic cancer vaccines have already achieved impressive clinical success. However, the fact that only a small subset of patients with specific tumor types can benefit from these treatments limits the application of cancer immunotherapy. To seek out the molecular mechanisms behind this challenge and to select cancer precision medicine for different individuals, researchers apply the immune repertoire sequencing (IRS) to evaluate genetic responses of each patient to current immunotherapies. This review summarizes the technical advances and recent applications of IRS in cancer immunotherapy, indicates the limitations of this technique, and predicts future perspectives both in basic studies and clinical trials.

Immunological therapy: A novel thriving area for triple-negative breast cancer treatment

Triple-negative breast cancer (TNBC) refers to cancers that are low in expression of the estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). TNBC tends to behave more aggressively than other types of breast cancer. Unlike other breast cancer subtypes (ie, ER-positive, HER2-positive subtypes), there are no approved targeted treatments available, other than the administration of chemotherapy. Immunotherapy is a new kind of treatment approach for TNBC when compared with the surgical treatment, chemotherapy, endocrine therapy, and molecular targeting therapy. The present article reviews the research progresses of immunotherapy for TNBC in recent years. The full text structure covers molecular classification of TNBC, active immunotherapy of TNBC, passive immunotherapy of TNBC, oncolytic immunotherapy and the prospect of immunotherapy for TNBC.