(P 98) A MULTIEPITOPE MRNA VACCINE FOR KAPOSI SARCOMA: AN IMMUNOINFORMATIC APPROACH

Objective: Kaposi sarcoma (KS) is a vascular tumor driven by the Kaposi sarcoma herpesvirus (KSHV), also known as human herpesvirus 8 (HHV8). KS manifests in four clinical forms—epidemic, endemic, iatrogenic, and classical—with KSHV essential for tumor development in all cases. The virus may remain latent, evading humoral immunity. A preventive and therapeutic vaccine should be able to leverage cellular immunity to destroy endothelial and cancer cells with latent infection. We utilized an immunoinformatic approach to design a multi-epitope mRNA vaccine to generate cellular immunity across a broad range of human leukocyte antigens (HLAs).

Methods: Since cellular immunity hinges on the presentation of epitopes in an HLA-specific manner, type 1 HLAs that cover the greatest burden of disease were identified. The frequency of type 1 HLAs in each world region was extracted from the Allele Frequency Net Database , and KS burden data from the World Health Organization. A pooled KS RNA sequencing (RNAseq) database was sourced from the National Center for Biotechnology Information to identify highly expressed viral proteins. Proteins with high expression and previously described functions on tumorigenesis, immune evasion, latency, or infectivity were selected. HLA Athena software was used to identify epitopes among the viral proteins with high antigenicity and cleavability. A final selection of epitopes was made to optimize coverage across the majority of HLAs.

Results: In our analysis of KS prevalence across 11 global regions, the highest prevalence was in Sub-Saharan Africa (66%), followed by Europe (10%) and South and Central America (10%). By weighting the HLA prevalence in each region by their disease prevalence, we found that HLA A*02:01 covers 13.2% of the disease burden, followed by HLA*C06:02 (11.9%) and HLA*30:1 (7.3%). A final list of 20 HLAs was selected for the immunoinformatic approach. Pooled RNAseq data showed that the most expressed latent viral proteins are ORF73(LANA), and ORF74(vGPCR), and the most common lytic are ORF4 (gB) and ORF22(gH). A list of 20 highly expressed proteins was selected, prioritizing functional proteins. After using HLA Athena for type 1 antigenicity and cleavability prediction, 20 epitopes optimized for the highest antigenicity and coverage were selected. The final list of peptides can elicit a high T-cell response in at least 67% of the population affected by the disease.

Conclusion: We designed an equitable multiepitope candidate vaccine predicted to induce effective T-cell responses in a majority of those affected by KSHV worldwide. This vaccine strategy is currently undergoing experimental validation to assess the stimulation capability of T-cells and their subsequent ability to recognize matched tumor samples. Experimental data will be presented at the meeting.