Evidence of lesions from Epstein-Barr virus infection in human breast cancer genomes

Abstract

Epstein-Barr virus (EBV) infects essentially all humans and provides no benefit. EBV can cause nasopharyngeal cancer (NPC), Burkitt’s lymphoma (BL), and perhaps breast cancer. Breast tissues from patients with breast cancer are more likely to be EBV-positive than tissues from healthy controls. However, EBV is not a proven cause of breast cancer because the tissues are not consistently EBV-positive. If EBV causes breast cancer, it would have to do it without an active infection. Other cancers with known viral origins do not require continuing presence of the virus. However, the "hit and run" theory is difficult to test for breast cancer without a proven EBV connection.

Here, I test this theory with multiple independent bioinformatic analyses. First, hundreds of breast cancer genomes contained characteristic methylation scars that indicate a cleared EBV infection. The genomes had further differential hypermethylation near positions where EBV reprograms normal cells into malignancy. Second, genomes from EBV cancers and breast cancers inactivated the same tumor-suppressive mechanisms. Third, deletions were identified on chromosome 3p in EBV cancers that shift cells to oxidative glycolysis, a prominent breast cancer phenotype known as the Warburg effect. Similar 3p deletions were found in breast cancer genomes. Fourth, somatic hypermutation clusters in EBV-cancers marked genome positions in breast cancers near translocations and focal oncogene amplification. EBV deregulation of deaminase and estrogen-induced topoisomerase explain these translocation breakpoints. Fifth, several alternate explanations for these results were ruled out. Finally, only limited segments of EBV DNA matched the human genome, making it possible that a childhood vaccine would end breast cancer.