Background Small cell lung cancer (SCLC) is the most aggressive subtype of lung cancer. SCLC responds well to chemotherapy, but almost inevitably relapses with limited treatment options available. Despite having a high mutational burden, many SCLC patients fail to benefit from immune checkpoint inhibition. Therefore, new immunotherapeutic approaches, like peptide-based vaccination or cellular therapy, are needed. Since SCLC-associated T cell antigens are still unknown, we aim to analyze naturally presented HLA-I and HLA-II peptides by extensive immunopeptidomics of SCLC tumors.

Methods SCLC tumors were obtained from fresh/frozen cryorecanalizations or biopsies, and patient-derived mouse xenograft models. After immunoprecipitation of peptide-HLA-I and -HLA-II complexes from tumor lysates, presented peptides were identified with liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS). Sequence annotation was performed using a reference proteome containing the reviewed human (and mouse) proteins from UniProt to identify non-mutated sequences and -if available- patient-individual whole exome sequencing data to identify mutated sequences. By comparative analysis with our published data of peptides presented by healthy tissue (containing >2 Mio HLA-I/-II peptides from e.g. lung, liver, brain, PBMCs), tumor-exclusive peptides were identified. Additionally, PBMCs from SCLC patients and healthy donors were sampled.

Results We analyzed n=12 samples of cryorecanalizations or biopsies and n=17 samples of patient-derived xenograft models. We optimized the peptide identification and selection strategy for xenograft models to improve overall peptide yields and to exclude mouse-derived MHC-peptides. We identified more than 77,000 HLA-I binders and 44,000 HLA-II peptides. Approx. 15,000 non-mutated peptides (7,988 for HLA-I and 7,517 for HLA-II) were tumor exclusive and not found to be presented on any healthy tissue. Among these, we found peptides deriving from proteins associated with SCLC, e.g. TP53, RB1, CHGA, ASCL1 or YAP1. So far, no mutated neoantigens were identified. The SCLC cohort covered 11 HLA-A, 20 HLA-B and 15 HLA-C allotypes. Following an in-house strategy, we have now selected frequent HLA-I binders shared between different samples, as well as the most interesting candidates for common HLA-I types in our SCLC cohort that will be tested for T cell recognition with cells from either healthy donors (priming) or SCLC patients (recall).

Conclusions We could extract HLA-bound peptides from different SCLC tumor sources and identified tumor-exclusive peptides that were shared between different samples and/or derived from proteins associated with SCLC. In following immunogenicity experiments we will further characterize SCLC-associated T-cell antigens that could be useful for immunotherapeutic approaches to combat SCLC.

Ethics Approval This study was approved by the ethics committee of the University of Tübingen (project number 526/2021BO2) and the ethics committee of the University of Cologne (project number 19–1164).