Expansion and optimal Exploitation of individual neoepitope Repertoire

Immunotherapy is fundamentally changing the treatment of cancer patients.  Personalized vaccines eliciting immune responses against individual cancer mutations (see Figure) have moved into the spotlight (Sahin et al. 2017).  We have pioneered the field and moved ´cancer mutanome vaccines´ from a mere vision into a disruptive medical concept compatible with current standards of drug development and health care practice.  Solving key scientific and technological challenges and building on extensive preclinical studies, we showed in a first-in-human trial potent tumor-directed immune responses in every single vaccinated patient, and clinical activity of a novel mRNA-based mutanome vaccine. Given that mutations are a hallmark of cancer, mRNA mutanome vaccines are universal drugs the efficacy of which are unaffected by the cancer type.  

The aim of this research program is to ignite the next wave of advancement by addressing key constraints challenging a full clinical realization of such vaccines.  Our objective is to exploit the full spectrum of mutated gene products for immunotherapy, such as single-nucleotide variations, indels, and fusions. These include mutations within exons and introns from annotated genes as well as from genome regions that are not related to known expressed gene products.

We have made progress to improve neoepitope detection tools. (1) The detection of structural variants (SVs) as a genomic basis for cancer specific changes of the transcriptome was shown to be specific and sensitive (Sethi et al. 2020). A multiomics characterization approach of the 4T1 cell line revealed many shared molecular features with human triple negative breast cancer (Schrörs et al. 2020) and their in-vitro immunogenicity. (2) Combining the TRON-Easyfuse® pipeline with our machine learning model, we are able to detect fusion genes with high specificity and sensitivity. Testing of antigen candidates revealed CD8+- or CD4+-T cell responses to one third of the candidates clearly showing their immunotherapeutic relevance (Weber et al. accepted). (3) We have also developed a tumor-specific splice neoepitope identification pipeline. Using publicly available tools, ~1 neo splice epitope is identified, but our advanced pipeline identifies many more candidates. (4) Furthermore, individually transcribed tumor antigens were evaluated as an additional neoantigen class. On the patient-level, a rather low contribution of this neoantigen-class was found, however transcribed tumor-antigens were more common in kidney renal papillary cell carcinoma, prostate adenocarcinoma and sarcoma.  To optimize detection, existing neoepitope selection algorithms were evaluated in published datasets.  It was revealed that no published predictor is predictive throughout different cohorts. Two manuscripts based on these results have been published (Lang et al. 2021 and 2022). The prediction of further neoantigen features is ongoing.

We are also in the process of developing a novel neoantigen selection tool based on 3D structural modeling to differentiate MHC class I-peptide in complex with wild-type vs. mutated peptides. To accurately capture intratumoral heterogeneity and tumor evolution we are aiming to represent the mutational processes via machine learning and AI models. Moreover, we are developing a deep learning based prediction system for the accurate detection of somatic mutations in sequencing data, in order to complement the existing detection algorithms.

1.    Sahin U, Derhovanessian E, Miller M, Kloke B-P, Simon P, Löwer M, Bukur V, Tadmor AD, Luxemburger U, Schrörs B, Omokoko T, Vormehr M, Albrecht C, Paruzynski A, Kuhn AN, Buck J, Heesch S, Schreeb KH, Müller F, Ortseifer I, Vogler I, Godehardt E, Attig S, Rae R, Breitkreuz A, Tolliver C, Suchan M, Martic G, Hohberger A, Sorn P, Diekmann J, Ciesla J, Waksmann O, Kemmer-Brück A, Witt M, Zillgen M, Rothermel A, Kasemann B, Langer D, Bolte S, Diken M, Kreiter S, Nemecek R, Gebhardt C, Grabbe S, Höller C, Utikal J, Huber C, Loquai C, Türeci Ö. Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer. 2017. Nature. 547(7662):222-226.
2.    Riccha Sethi, Julia Becker, Jos de Graaf, Martin Löwer, Martin Suchan, Ugur Sahin, David Weber. (2020) Integrative analysis of structural variations using short-reads and linked-reads yields highly specific and sensitive predictions. PLoS Comput Biol. 16(11):e1008397. doi: 10.1371/journal.pcbi.1008397
3.    Barbara Schrörs, Sebastian Boegel, Christian Albrecht, Thomas Bukur, Valesca Bukur, Christoph Holtsträter, Christoph Ritzel, Katja Manninen, Arbel D. Tadmor, Mathias Vormehr, Ugur Sahin, and Martin Löwer (2020) Multi-Omics Characterization of the 4T1 Murine Mammary Gland Tumor Model. Front Oncol.;10:1195 doi: 10.3389/fonc.2020.01195
4.    David Weber, Jonas Ibn-Salem, Patrick Sorn, Martin Suchan, Christoph Holtsträter, Urs Lahrmann, Isabel Vogler, Franziska Lang, Barbara Schrörs, Martin Löwer, Ugur Sahin. Accurate detection of tumor-specific fusion genes reveals strongly immunogenic personal neo-antigens. Manuscript accepted for publication.
5.    Franziska Lang, Pablo Riesgo Ferreiro, Martin Löwer, Ugur Sahin, Barbara Schrörs (2021). NeoFox: annotating neoantigen candidates with neoantigen features. Bioinformatics. btab344. doi: 10.1093/bioinformatics/btab344
6.    Franziska Lang, Barbara Schrörs, Martin Löwer, Özlem Türeci, Ugur Sahin (2021). Identification of neoantigens for individualised cancer immunotherapy. Nature Reviews Drug Discovery; 2022 Feb 1. doi: 10.1038/s41573-021-00387-y