Ovarian carcinosarcoma (OCS) are rare, aggressive cancers with poor prognosis due to limited effective treatments1. OCS tumorigenesis is not driven by somatic mutation, but via reprogramming of gene expression2. Therefore, the tumour mutation burden (TMB) is often low, and they are relatively non-responsive to single agent immunotherapy3.
We have established a paired OCS patient-derived xenograft (PDX) model, SFRC01177 from baseline chemonaïve and post-chemotherapy specimens in Nod-SCID-Gamma (NSG) major histocompatibility complex (MHC)null mice. These mice are deficient in MHC class I/II expression to reduce acute graft-versus-host disease following injection of human peripheral mononuclear cells4. The tumour is refractory to cisplatin and to 1 million human leukocyte antigens (HLA)-match CD8 T-cell injection consistent with expected tumour phenotype.
Whole genome/exome sequencing was performed on the baseline and recurrent tumours. Despite a low TMB of < 4 mutations/Mb, a total of 14,419 variants of mutant peptides ranging from 8 to 16 amino acid length were identified: 13,170 in the baseline and 10,998 in the recurrent tumour with 9,750 shared. Immunoprecipitation of the tumour HLA-complexes with their bound peptides followed by tandem mass spectrometry have identified over 2,300 HLA-bound peptides: 2,215 in the baseline and only 317 in the recurrent tumour with 239 shared. The analysis of the HLA-peptide motifs suggested that down-regulation of tumour cell MHC class I/II may account for this reduction. By combining the genomic and the immunopeptidomic data, a mutant HLA-bound neoepitope was identified. Neuropeptide W E100Q homologous to COSV61585942 (known immunogenic peptide; reported over 5 times in colon cancer) was associated with high areas under the curve in both the baseline and recurrent tumours. Ex vivo pulse neuropeptide W E100Q with HLA-matched dendritic and CD8 T-cells demonstrated delayed in OCS tumour in vivo growth and increased CD8 T-cells intra-tumour infiltration.
In summary, we can identify potential immunogenic mutant neoepitope using an integrated multi-omics approach in a TMB-low tumour. High affinity T-cell receptor discovery against this peptide is currently underway.