In Utero Cytomegalovirus Infection Increases Risk of Developing Acute Lymphocytic Leukemia in Pediatric Patients

Children who had in utero cytomegalovirus (CMV) infection are more likely to develop childhood acute lymphocytic leukemia (ALL) than matched controls without infection, suggesting a role of infection in the etiology of ALL, according to the results of a next-generation sequencing (NGS) study published in Blood. The study authors, led by Stephen S. Francis, PhD, from the Department of Epidemiology & Biostatistics at the University of California, San Francisco/Berkeley, also found that congenital CMV infection is more prominent in Hispanic children than in children in any other ethnic group.

Dr. Francis and authors conducted two complementary studies to determine which infections are unique to childhood ALL (compared with acute myeloid leukemia [AML]) and whether these infections are risk factors for ALL.

The first study was a comprehensive NGS-based metagenomics analysis of pretreatment diagnostic bone marrow (BM) data from 127 children with ALL and 38 children with AML who were enrolled in the California Childhood Leukemia Study, which identified prevalence of infections. In the second study, researchers analyzed blood specimens from 268 newborn patients with ALL and 270 newborn patients without cancer (controls) identified through the California Department of Public Health, and then screened the samples for two herpesviruses that were identified in the discovery set at diagnosis (including CMV and Epstein-Barr virus).

The children with ALL tended to be younger than those with AML, but all other demographic characteristics were similar.

NGS-based analyses revealed a manifold higher prevalence of CMV in the BM samples of patients with ALL, compared with patients with AML (odds ratio [OR] = 18; p=0.003), suggesting that CMV was associated with ALL. Screening revealed that children who developed ALL were 3.71 times more likely to be CMV-positive at birth (p=0.0016).

The prevalence of congenital CMV infection was higher in white children, and the authors observed an increased risk for developing ALL in this group, though this association was not statistically significant (OR=2.1; 95% CI 0.69-7.13). Hispanic ethnicity, however, was associated with a statistically significant 5.9-fold increased risk of developing ALL, compared with children of other ethnicities (95% CI 1.9-26; p=0.006).

The mean CMV viral load among positive case samples was 0.214 copies/μL, which was higher than the CMV viral load among control samples (0.071 copies/μL; p=0.003). Age did not affect the risk of ALL diagnosis between CMV-positive (mean age = 4.8 years) and CMV-negative patients (mean age = 5.15 years). See TABLE for the results of the neonatal blood specimen screening.

The majority of congenital CMV infections are caused by vertical transmission of reactivated virus, Dr. Francis and authors noted, because most women of childbearing age have been previously infected with CMV.

“Timing of CMV infection has important implications,” the authors wrote, explaining that the presence of CMV in the child prior to birth may have important immune control implications. “Infection prior to the development of a robust adaptive immune response in the fetus and neonate may affect central tolerance and enable CMV to persist in an infectious course.”

Dr. Francis and authors noted two features of CMV infection that support its role in the oncogenesis of ALL: First, congenital CMV infection can cause chromosomal instability, and, second, CMV has the largest genome of any known human viral pathogen and harbors many immune evasion genes.

The study is limited by its retrospective design and lack of information on ALL subtype and cytogenetic characteristics. In addition, the low input quantity of dried blood specimen DNA used in the second study phase may have resulted in false negatives.

“Our findings lead us to hypothesize that in utero or perinatal CMV infection initiates immune dysregulation during the critical period of fetal immune development, allowing a greater number of and more fulminant infections later in life,” the authors concluded, adding that studies that define in utero CMV infection in terms of ALL subtype, cytogenetics, and overall prevalence are warranted.


Francis SS, Wallace AD, Wendt JA, et al. In utero cytomegalovirus infection and development of childhood acute lymphoblastic leukemia. Blood. 2016 December 15. [Epub ahead of print]

TABLE. Results of Neonatal Blood Spot Screen
CMV CMV+ OR p Value EBV− EBV+ OR p Value
ALL 242 26 3.71

(95% CI 1.71-8.95)

0.0016 257 11 1.01

(95% CI 0.42-2.24)

Controls 262 8 259 11
According to Race
Hispanic ALL 134 3 5.9

(95% CI 1.89-25.96)

0.006 141 8 1.38

(95% CI 0.46-4.35)

Hispanic controls 152 10 149 6
White ALL 71 5 2.1

(95% CI 0.69-7.13)

0.204 79 2 0.53

(95% CI 0.09-2.94)

White controls 82 1 83 5
Asian/PI ALL 25 0 N/A 1 25 1 N/A 1


21 0 21 0
Black ALL 10 0 10 0
Black controls 4 0 4 0
Other ALL 2 0 2 0 1
Other controls 3 0 2 1
CMV = cytomegalovirus; OR = odds ratio; EBV = Epstein-Barr virus; ALL = acute lymphocytic leukemia; PI = Pacific Islander; N/A = not applicable