Endocrine Abstracts

Introduction: Adrenocortical carcinoma (ACC) is one of the most aggressive endocrine malignancies and confers a poor prognosis in advanced stages. Effective treatments are lacking. The results of immune checkpoint inhibition were disappointing with few responders only deriving clinical benefit. For the development of novel immunotherapies such as tumor vaccines and T cell-based treatments, target identification is essential. Tumor-specific mutant neoantigens that may be recognized by T cells in the context of major histocompatibility complex (MHC) I are promising candidates.

Methods: We performed whole exome-sequencing in 10 ACC samples with matched blood controls. Somatic mutations were identified using an in house bioinformatics pipeline. By coupling POLYSOLVER for HLA typing with netMHCpan, in silico binding affinity of tumor-specific neoantigens to MHC was calculated taking into account both peptide and HLA sequence information. Strong binding (SB) was defined as <0.5% rank, weak binding (WB) as 0.5-1.9% rank and no binding (NB) as >2.0% rank of all peptides to MHC.

Results: Across 10 ACC patients, we identified 1067 unique somatic mutations (median 49.5, ranging from 15-590), affecting 989 different genes. Binding affinity changed for 414 predicted neoantigens from NB to WB, 80 from NB to SB and 82 from WB to SB. The mutant neoantigen load per patient ranged from 10 to 235 (mean 66.2) and was positively correlated to the total number of non-synonymous single nucleotide variants (R² 0.8977, 95% CI 0.79-0.99; P<0.0001).

Discussion: This is the first study that demonstrates successful in silico neoantigen profiling in ACC. Mutant neoantigens were predicted to be present both in ACC with high and low tumor mutational burden. These data will pave the way for in vitro validation and hold potential to develop therapeutic cancer vaccines and T cell-based cancer immunotherapy in ACC.