Methods behind neoantigen prediction for personalized anticancer vaccines

Next to conventional cancer therapies, immunotherapies such as immune checkpoint inhibitors have broadened the cancer treatment landscape over the past decades. Recent advances in next generation sequencing and bioinformatics technologies have made it possible to identify a patient's own immunogenic neoantigens. These cancer neoantigens serve as important targets for personalized immunotherapy which has the benefit of being more active and effective in targeting cancer cells. This paper is a step-by-step guide discussing the different analyses and challenges encountered during in-silico neoantigen prediction. The protocol describes all the tools and steps required for the identification of immunogenic neoantigens.

Challenges in neoantigen-directed therapeutics

A fundamental prerequisite for the efficacy of cancer immunotherapy is the presence of functional, antigen-specific T cells within the tumor. Neoantigen-directed therapy is a promising strategy that aims at targeting the host's immune response against tumor-specific antigens, thereby eradicating cancer cells. Initial forays have been made in clinical environments utilizing vaccines and adoptive cell therapy; however, many challenges lie ahead. We provide an in-depth overview of the current state of the field with an emphasis on in silico neoantigen discovery and the clinical aspects that need to be addressed to unlock the full potential of this therapy.

A synthetic DNA template for fast manufacturing of versatile single epitope mRNA

A flexible, affordable, and rapid vaccine platform is necessary to unlock the potential of personalized cancer vaccines in order to achieve full clinical efficiency. mRNA cancer vaccine manufacture relies on the rigid sequence design of multiepitope constructs produced by laborious bacterial cloning and time-consuming plasmid preparation. Here, we introduce a synthetic DNA template (SDT) assembly process, which allows cost- and time-efficient manufacturing of single (neo)epitope mRNA. We benchmarked SDT-derived mRNA against mRNA derived from a plasmid DNA template (PDT), showing that monocyte-derived dendritic cells (moDCs) electroporated with SDT-mRNA or PDT-mRNA, encoding HLA-I- or HLA-II-restricted (neo)epitopes, equally activated T cells that were modified to express the cognate T cell receptors. Furthermore, we validated the SDT-mRNA platform for neoepitope immunogenicity screening using the characterized HLA-A2-restricted neoepitope DHX40B and four new candidate HLA-A2-restricted melanoma neoepitopes. Finally, we compared SDT-mRNA with PDT-mRNA for vaccine development purposes. moDCs electroporated with mRNA encoding the HLA-A2-restricted, mutated Melan-A/Mart-1 epitope together with TriMix mRNA-generated high levels of functional Melan-A/Mart-1-specific CD8⁺ T cells. In conclusion, SDT single epitope mRNA can be manufactured in a more flexible, cost-efficient, and time-efficient way compared with PDT-mRNA, allowing prompt neoepitope immunogenicity screening, and might be exploited for the development of personalized cancer vaccines.

Abstract CC01-02: Novel algorithm discovers up to 35% more epitopes translated from non-coding regions in cold tumours

Currently, most neoantigen pipelines often focus on the detection of neoantigens derived from mutations in the coding regions of the genome. However, in some cancer indications, the number of mutations detectable in tumours can be very low (low tumour mutational burden). This limits the number of actionable neoantigens and results in so-called “cold” tumours. In these cases, non-canonical neoantigens resulting from alterations in non-coding regions of the human genome could represent a high potential alternative for treatment. Indeed, recent research has revealed that previously presumed non-coding regions of the human genome, such as long non-coding RNAs (lncRNAs), can contain translatable small open reading frames (smORFs) generating micropeptides. Some of these micropeptides have already been shown to be involved in cancer development, but these small peptides could also represent a high potential source of non-canonical neoantigens for personalised therapy. Here, we present smORFin, a machine learning algorithm specifically trained to identify smORFs in transcripts and to assess their coding potential. While most tools are focused on longer sequences, smORFin is specifically developed to target small ORFs (<303 nucleotides). Furthermore, smORFin also accounts for smORFs with alternative initiation codons, thereby improving its sensitivity for the detection of novel unannotated smORFs. As a result, the smORFin model reaches a precision of 0.98 and an accuracy of 0.95 on its testing dataset. Using this new prediction tool, a library of human smORFs was assembled, the so-called smORFeome. This library of smORFs, and their associated proteins, was evaluated as a reference for spectrum to peptide matching in mass spectrometry data (MS) analysis. Indeed, the evaluation of seven MS samples revealed and validated the presence of smORFeome-related micropeptides and HLA-Iassociated epitopes originating from smORFs. The impact of mutations in allegedly non-coding regions of tumour genomes and its influence on the neoantigen repertoire, was evaluated through integration of smORFin in a neoantigen identification pipeline targeting lncRNA-derived mutated epitopes; lncRNeos. It was observed that these epitopes only represent a minor fraction of the total neoantigen load. Strikingly, when only focusing on tumours with a low neoantigen load, lncRNeos represented up to 27% of the total neoantigen load. This indicates that for tumours with a low TMB, and therefore with a low neoantigen load, lncRNeos allows to significantly expand the neoantigen repertoire. Biological in vivo/in vitro validation remains necessary to assess the existence, presentation, and actionability of lncRNeos.

Citation Format: Lena Pfitzer, Lore Van Oudenhove, Cedric Bogaert, Bruno Fant. Novel algorithm discovers up to 35% more epitopes translated from non-coding regions in cold tumours [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2021 Oct 7-10. Philadelphia (PA): AACR; Mol Cancer Ther 2021;20(12 Suppl):Abstract nr CC01-02.

Abstract P006: Novel machine-learning tools improve cost-effective development of personalised immunotherapies: Lowering false positive rates in the search for actionable (personalised and largely shared) immunogenic neoantigens

Personalised immunotherapy approaches rely on the ability of tumour-derived neoantigens to elicit a T-cell immune reaction able to recognise and kill the tumour cells expressing them. Clinical attempts to leverage the power of neoantigens have however yielded mixed results. This can mostly be attributed to the difficulty of finding truly immunogenic peptides from the set of novel peptides generated by mutations in a given cancer patient. In silico approaches can help alleviate this heavy cost by reducing the neoantigen search space, prioritising epitopes based on various parameters such as epitope expression or MHC binding likelihood. Here we present a suite of tools aimed at further assisting clinicians in selecting the most actionable peptides from a set of potential candidates. We developed neoMS, a neural network algorithm able to predict epitope presentation at the cell surface with unparalleled performance. The model achieves up to 0.61 precision at recall 0.4 on its test set, vastly outperforming the current industry standards. In addition, due to his sequence-based comparison method, neoMS exhibits extrapolation capabilities, achieving non-zero predictive power when evaluated on ground truth ligandome data derived from an HLA allele completely absent from the training set. In some cancer indications we showed, moreover, that the neoMS-predicted rate of neoantigen presentation can be used in combination with tumour mutational burden as a high-specificity predictor of response to immune checkpoint inhibitor treatment. Furthermore, the neoIM algorithm is able to discriminate, in an HLA-agnostic fashion, which of the presented peptides will elicit a T-cell immune reaction. This first-in-class algorithm is a random forest classifier specifically trained to classify short peptides of length 9-11 amino acids as immunogenic or non-immunogenic. neoIM vastly outperforms the currently available methods and can predict peptide immunogenicity with high accuracy (AUC=0.88). Interestingly, neoIM confirmed ELISPOT data obtained by Dillon et al. (2017) showing a response in 4 out of 11 breast cancer patients to a vaccine consisting of 9 MHC class-I restricted breast cancer-associated peptides. The 2 antigens that resulted in a CD8+ T-cell specific response were predicted by neoIM as the highest scoring showing its potential in finding the truly immunogenic neoantigens. Taken together, these tools decrease false positive rates significantly as they enable improved identification of immunogenic peptides and the predictions correlate with intensity of immune response and clinical benefits. As such, these tools represent a cost-efficient preliminary step in the search for actionable, immunogenic neoantigens.

Citation Format: Cedric Bogaert, Lena Pfitzer, Nil Adell Mill, Bruno Fant. Novel machine-learning tools improve cost-effective development of personalised immunotherapies: Lowering false positive rates in the search for actionable (personalised and largely shared) immunogenic neoantigens [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2021 Oct 7-10. Philadelphia (PA): AACR; Mol Cancer Ther 2021;20(12 Suppl):Abstract nr P006.

Novel biomarkers to assess in utero effects of maternal opioid use: First steps toward understanding short- and long-term neurodevelopmental sequelae

Maternal opioid use disorder is common, resulting in significant neonatal morbidity and cost. Currently, it is not possible to predict which opioid-exposed newborns will require pharmacotherapy for neonatal abstinence syndrome. Further, little is known regarding the effects of maternal opioid use disorder on the developing human brain. We hypothesized that novel methodologies utilizing fetal central nervous system-derived extracellular vesicles isolated from maternal blood can address these gaps in knowledge. Plasma from opioid users and controls between 9 and 21 weeks was precipitated and extracellular vesicles were isolated. Mu opioid and cannabinoid receptor levels were quantified. Label-free proteomics studies and unbiased small RNA next generation sequencing was performed in paired fetal brain tissue. Maternal opioid use disorder increased mu opioid receptor protein levels in extracellular vesicles independent of opioid equivalent dose. Moreover, cannabinoid receptor levels in extracellular vesicles were upregulated with opioid exposure indicating cross talk with endocannabinoids. Maternal opioid use disorder was associated with significant changes in extracellular vesicle protein cargo and fetal brain micro RNA expression, especially in male fetuses. Many of the altered cargo molecules and micro RNAs identified are associated with adverse clinical neurodevelopmental outcomes. Our data suggest that assays relying on extracellular vesicles isolated from maternal blood extracellular vesicles may provide information regarding fetal response to opioids in the setting of maternal opioid use disorder. Prospective clinical studies are needed to evaluate the association between extracellular vesicle biomarkers, risk of neonatal abstinence syndrome and neurodevelopmental outcomes.

Selective expansion of myeloid and NK cells in humanized mice yields human-like vaccine responses

Mice engrafted with components of a human immune system have become widely-used models for studying aspects of human immunity and disease. However, a defined methodology to objectively measure and compare the quality of the human immune response in different models is lacking. Here, by taking advantage of the highly immunogenic live-attenuated yellow fever virus vaccine YFV-17D, we provide an in-depth comparison of immune responses in human vaccinees, conventional humanized mice, and second generation humanized mice. We demonstrate that selective expansion of human myeloid and natural killer cells promotes transcriptomic responses akin to those of human vaccinees. These enhanced transcriptomic profiles correlate with the development of an antigen-specific cellular and humoral response to YFV-17D. Altogether, our approach provides a robust scoring of the quality of the human immune response in humanized mice and highlights a rational path towards developing better pre-clinical models for studying the human immune response and disease.

Environmental, genetic and epigenetic contributions to cocaine addiction

Decades of research on cocaine has produced volumes of data that have answered many important questions about the nature of this highly addictive drug. Sadly, none of this information has translated into the development of effective therapies for the treatment of cocaine addiction. This review endeavors to assess the current state of cocaine research in an attempt to identify novel pathways for therapeutic development. For example, risk of cocaine addiction is highly heritable but genome-wide analyses comparing cocaine-dependent individuals to controls have not resulted in promising targets for drug development. Is this because the genetics of addiction is too complex or because the existing research methodologies are inadequate? Likewise, animal studies have revealed dozens of enduring changes in gene expression following prolonged exposure to cocaine, none of which have translated into therapeutics either because the resulting compounds were ineffective or produced intolerable side-effects. Recently, attention has focused on epigenetic modifications resulting from repeated cocaine intake, some of which appear to be heritable through changes in the germline. While epigenetic changes represent new vistas for therapeutic development, selective manipulation of epigenetic marks is currently challenging even in animals such that translational potential is a distant prospect. This review will reveal that despite the enormous progress made in understanding the molecular and physiological bases of cocaine addiction, there is much that remains a mystery. Continued advances in genetics and molecular biology hold potential for revealing multiple pathways toward the development of treatments for the continuing scourge of cocaine addiction.

Paternal cocaine taking elicits epigenetic remodeling and memory deficits in male progeny

This corrects the article DOI: 10.1038/mp.2017.8.

Paternal cocaine taking elicits epigenetic remodeling and memory deficits in male progeny

Paternal environmental perturbations including exposure to drugs of abuse can produce profound effects on the physiology and behavior of offspring via epigenetic modifications. Here we show that adult drug-naive male offspring of cocaine-exposed sires have memory formation deficits and associated reductions in NMDA receptor-mediated hippocampal synaptic plasticity. Reduced levels of the endogenous NMDA receptor co-agonist d-serine were accompanied by increased expression of the d-serine degrading enzyme d-amino acid oxidase (Dao1) in the hippocampus of cocaine-sired male progeny. Increased Dao1 transcription was associated with enrichment of permissive epigenetic marks on histone proteins in the hippocampus of male cocaine-sired progeny, some of which were enhanced near the Dao1 locus. Finally, hippocampal administration of d-serine reversed both the memory formation and synaptic plasticity deficits. Collectively, these results demonstrate that paternal cocaine exposure produces epigenetic remodeling in the hippocampus leading to NMDA receptor-dependent memory formation and synaptic plasticity impairments only in male progeny, which has significant implications for the male descendants of chronic cocaine users.

Comprehensive interactome of Otx2 in the adult mouse neural retina

The Otx2 homeodomain transcription factor exerts multiple functions in specific developmental contexts, probably through the regulation of different sets of genes. Protein partners of Otx2 have been shown to modulate its activity. Therefore, the Otx2 interactome may play a key role in selecting a precise target-gene repertoire, hence determining its function in a specific tissue. To address the nature of Otx2 interactome, we generated a new recombinant Otx2(CTAP-tag) mouse line, designed for protein complexes purification. We validated this mouse line by establishing the Otx2 interactome in the adult neural retina. In this tissue, Otx2 is thought to have overlapping function with its paralog Crx. Our analysis revealed that, in contrary to Crx, Otx2 did not develop interactions with proteins that are known to regulate phototransduction genes but showed specific partnership with factors associated with retinal development. The relationship between Otx2 and Crx in the neural retina should therefore be considered as complementarity rather than redundancy. Furthermore, study of the Otx2 interactome revealed strong associations with RNA processing and translation machineries, suggesting unexpected roles for Otx2 in the regulation of selected target genes all along the transcription/translation pathway. The Otx2(CTAP-tag) line, therefore, appears suitable for a systematic approach to Otx2 protein-protein interactions. genesis 53:685-694, 2015. © 2015 Wiley Periodicals, Inc.

Otx2 ChIP-seq reveals unique and redundant functions in the mature mouse retina

During mouse retinal development and into adulthood, the transcription factor Otx2 is expressed in pigment epithelium, photoreceptors and bipolar cells. In the mature retina, Otx2 ablation causes photoreceptor degeneration through a non-cell-autonomous mechanism involving Otx2 function in the supporting RPE. Surprisingly, photoreceptor survival does not require Otx2 expression in the neural retina, where the related Crx homeobox gene, a major regulator of photoreceptor development, is also expressed. To get a deeper view of mouse Otx2 activities in the neural retina, we performed chromatin-immunoprecipitation followed by massively parallel sequencing (ChIP-seq) on Otx2. Using two independent ChIP-seq assays, we identified consistent sets of Otx2-bound cis-regulatory elements. Comparison with our previous RPE-specific Otx2 ChIP-seq data shows that Otx2 occupies different functional domains of the genome in RPE cells and in neural retina cells and regulates mostly different sets of genes. To assess the potential redundancy of Otx2 and Crx, we compared our data with Crx ChIP-seq data. While Crx genome occupancy markedly differs from Otx2 genome occupancy in the RPE, it largely overlaps that of Otx2 in the neural retina. Thus, in accordance with its essential role in the RPE and its non-essential role in the neural retina, Otx2 regulates different gene sets in the RPE and the neural retina, and shares an important part of its repertoire with Crx in the neural retina. Overall, this study provides a better understanding of gene-regulatory networks controlling photoreceptor homeostasis and disease.