Long-read RNA sequencing reveals full-length HERV transcripts and novel ERVE-4 isoforms with potential immunotherapeutic relevance in clear cell renal cell carcinoma

Introduction Human endogenous retroviruses (HERVs) are increasingly recognized as sources of tumor-associated antigens and potential targets for cellular immunotherapy. However, their repetitive sequences and noncanonical splicing patterns complicate accurate transcript annotation using common short-read RNA sequencing. This study aimed to improve HERV transcriptome annotation using long-read RNA sequencing. Methods We developed a bioinformatics workflow for comprehensive annotation of HERV expression using long-read RNA sequencing, with validation using short-read RNA sequencing. We applied this workflow to ERVE-4, a HERV element previously shown to be selectively expressed in clear cell renal cell carcinoma (ccRCC) and to encode immunogenic peptides currently under clinical investigation as targets for cellular immunotherapy. Results Long-read sequencing captured full-length HERV transcripts, improved mapping across repetitive regions, and enabled reconstruction of HERV isoforms. The workflow identified hundreds of HERV loci in long-read datasets absent from corresponding short-read annotations. Analysis of ERVE-4 revealed previously uncharacterized isoforms, whose expression was validated across multiple ccRCC tumors using RT-dPCR. Discussion Long-read RNA sequencing improves the resolution and completeness of HERV transcriptome annotation. Novel ERVE-4 isoforms substantially expand the known antigenic coding potential of ERVE-4 possibly implying discovery of novel peptide targets for immunotherapy in ccRCC.