Host–Virus Genetic Interactions Drive Nasopharyngeal Cancer Risk

Explaining this divergence requires understanding how human and viral factors jointly influence disease risk. However, studies have typically analyzed host or pathogen genomes in isolation. A new study shows that specific interactions between human and viral genomes shape nasopharyngeal cancer risk.

Researchers at Columbia University Mailman School of Public Health implemented a large-scale genome-to-genome analysis to interrogate host–virus genetic interactions implicated in cancer. The approach jointly assessed human genetic variation and EBV sequence diversity rather than evaluating each genome separately. This design was used to identify combinations of human leukocyte antigen (HLA) alleles and EBV variants associated with nasopharyngeal carcinoma.

The team analyzed human genome-wide association data alongside EBV whole‑genome sequencing using a stepwise analytical framework that integrates statistical genetics with causal inference methods. The pipeline accounted for population structure in both human and viral genomes, relatedness among samples, and multiple testing. This enabled a systematic search for interaction effects across both genomes that would be missed by single‑genome analyses.

A key interaction emerged between the HLA allele HLA‑A*11:01 and a single‑nucleotide variant (85841G) in the EBV gene EBNA3B. Individuals with susceptible HLA backgrounds who were infected with high‑risk EBV strains carrying 85841G showed substantially elevated risk of nasopharyngeal carcinoma. Functional experiments demonstrated that the viral mutation generates a peptide presented by HLA‑A*11:01, eliciting HLA‑A*11:01‑restricted CD8+ T‑cell responses that kill EBV‑transformed B cells harboring 85841G strains.

The findings are published in Nature on April 15, 2026 under the title “EBV strain interacts with host HLA to drive nasopharyngeal carcinoma risk”. Provided materials emphasize the growing role of statistical genomics and causal inference as multiple genomic systems are integrated to clarify complex disease mechanisms. The study framework highlights how combined host–pathogen genetics can refine understanding of cancer susceptibility.

“Most genetic studies examine the host genome or the pathogen genome separately,” said Zhonghua Liu, ScD, assistant professor of Biostatistics at Columbia Mailman School and co‑senior author. “By analyzing both genomes together using advanced statistical genetics methods and causal inference-inspired interaction frameworks, we can identify and quantify genetic interactions that would otherwise remain hidden.”

Related Links
Columbia University Mailman School of Public Health