Cytomegalovirus as an etiologic and clinico-pathogenic factor in childhood acute lymphoblastic leukemia

ABSTRACT Childhood acute lymphoblastic leukemia (ALL) is most common cancer in children (age 0- 14 years). Despite improvements in treatment, survivors face a lifelong battle with negative health effects of treatment, making prevention a priority. The lack of modifiable etiological factors has been a major barrier to prevention, but we recently made a groundbreaking (albeit preliminary) discovery that neonatal cytomegalovirus (nCMV) infection is a strong risk factor for childhood ALL (estimated odds ratio = 3.7) in a small study with 268 cases and 270 controls. In the same study, we also found CMV sequences within pre-treatment diagnostic tumor tissue. In addition, we conducted a population-based study linking medical records and cancer registry data in Sweden, which revealed a relative risk over 10 for early CMV infection and the child’s hematologic malignancy diagnosis. We propose here to definitively assess the epidemiology of CMV and ALL including CMV’s impact on the ALL tumor genome. We hypothesize that nCMV will contribute to ALL incidence and create specific mutational signatures present in ALL tumor genomes known to be associated with viral and infectious etiologies. In our first aim, we will better define the role of nCMV infection as an ALL risk factor in a study with over 3800 cases and 4897 controls, accounting for other risk factors such as birthweight, birth order, mode of delivery, polygenic risk score for ALL and parental ages neonatal immune phenotype, as well as the role of maternal CMV infection status during pregnancy. We will also assess whether CMV contributes to the higher risk of ALL in Latinos compared to other groups. As a second aim, we will test 1,020 children for CMV involvement at birth and within their matched ALL tumor genomes collected at diagnosis. We will investigate the presence of the APOBEC and Recombinase Activating Gene (RAG), virally-associated mutational signatures, within tumors that were nCMV positive compared to those from children negative for the virus. This research will leverage exceptional resources, including archived newborn blood specimens and population- based childhood ALL cases and controls, and maternal blood specimens collected during mid-pregnancy. In addition, the proposed study benefits from a high fraction of Latinos in the study population who carry a disproportionate burden of childhood ALL, include advanced laboratory techniques to identify neonatal and maternal CMV infection that have been refined in our laboratory, and will evaluate directly a mutation signature putatively caused by the virus. CMV is a highly adept immune manipulator. Identification of CMV as an etiologic agent in childhood ALL could create an unprecedented target for leukemia prevention, either by vaccination against CMV infection or manipulation of the host response to CMV infection after birth. In either scenario, this proposal will better characterize the role of CMV in the genesis of childhood ALL and have profound impact from a prevention perspective. As CMV is also responsible for congenital hearing loss and a host of other neurological outcomes, our research will impact childhood health apart from leukemia.