RNA-sequencing of cutaneous melanoma and non-small cell lung cancer biopsies

We used a proteogenomic approach combining RNA-sequencing and mass spectrometry to study the MHC I-immunopeptidome of cutaneous melanoma and non-small cell lung cancer (NSCLC) samples. The RNA-sequencing of each sample was used to construct sample-specific databases for the mass spectrometry-based identification of MHC I-associated peptides (MAPs). MAPs were then filtered based on their RNA expression in the respective cancer types from The Cancer Genome Atlas (TCGA-SKCM for melanoma MAPs, or TCGA-LUSC and TCGA-LUAD for NSCLC MAPs) vs. benign tissues (from the Genotype-Tissue Expression (GTEx) Project, medullary thymic epithelial cells, purified blood and bone marrow cells, and normal melanocytes for melanoma or bronchial brushing samples for NSCLC). MAPs were classified as mutated tumor-specific antigens (mTSAs, derived from non-synonymous mutations expressed in the sample of origin), or unmutated tumor antigens: aberrantly expressed tumor-specific antigens (aeTSAs, no/low expression in benign tissues and at least two times higher expression in TCGA), tumor-associated antigens (TAAs, significant expression in benign tissues and at least two times higher expression in TCGA), lineage-associated antigens (LSAs, specific expression to cancer and normal tissue of origin, i.e., lung and bronchial brushing samples for NSCLC or skin and melanocytes for melanoma). The tumor antigens described here represent attractive targets for immunotherapy of melanoma and NSCLC. The 12 flash-frozen cutaneous melanoma specimens used in this study were purchased from Tissue Solutions, whereas the 26 flash-frozen NSCLC samples were purchased from Tissue Solution, BioIVT or Reprocell. Between 30 mg and 70 mg were used for RNA-seq. Total RNA extraction was done using TRIzol (Invitrogen) and further purification with the RNeasy Micro Kit (QIAGEN) for all primary tumor samples. The RNA quantification was performed using a QuBIT (Life Technologies), and the RNA quality was assessed using a Bioanalyzer Nano (Agilent). cDNA library preparation was done using the KAPA Hyperprep RNAseq stranded kit (KAPA) with polyA capture. The amount of RNA used for cDNA library preparation, and the number of PCR cycles used for library amplification can be found in Supplementary Table 4. Libraries were quantified by QuBit, and average library length was evaluated with the BioAnalyzer DNA1000. All libraries were diluted to 10 nM and normalized by qPCR using the KAPA library quantification kit (KAPA). Libraries were pooled to equimolar concentration. Sequencing was performed with the Illumina HiSeq2000 or the Illumina Nextseq500 using PE100 or PE75 sequencing.