Abstract

The risk pattern of adolescent/young adult Hodgkin lymphoma (HL) is consistent with causation by a relatively late infection to a common childhood virus. However a causal virus has not yet been convincingly found. Susceptibility is known to be heritable, and a lower genetically determined interleukin-12 [IL-12] response is a risk factor. A lower IL-12 response can also be produced by diminished gastrointestinal exposure to the environmental microbiome, with resulting persistence of a Th2-skewed cytokine phenotype. We have therefore studied twin pairs discordant for HL to search for early life differences in sources of both viral and non-viral infection. HL-discordant twin pairs volunteered for the International Twin Study in ignorance of specific hypotheses. 70% of the questionnaires sent to individual twins were returned producing 188 informative pairs. The Kappa statistic was used to assess between-twin agreement. Designating the twin with HL as the case and the unaffected twin as the control, odds ratios (ORs) and 95% confidence intervals (CI) were calculated using conditional logistic regression for matched pairs. Tonsillectomy or appendectomy at least 5 years prior to diagnosis were associated respectively with a 2- and 3-fold statistically significant increase in risk. Infection with 3 or more childhood exanthems was associated with a 60% decreased risk (95% confidence intervals 0.2, 0.9). Behaviors likely to produce oral exposure to the environmental microbiome conveyed statistically significant decreases in risk (OR=0.2–0.5). A history of eczema increased risk (OR= 2.8, 95% CI= 1.0, 7.8). Kappa statistics were high (>0.8) for significant findings. Our evidence supports a role for early exposure to various infections in the etiology of adolescent/young adult Hodgkin lymphoma.

Table 1. History of infections, immune-related surgeries and other childhood experience at least 5 years prior to diagnosis and risk of adolescent-young adult Hodgkin Lymphoma diagnosed between 13–50 years of age in disease-discordant twin pairs from the International Twin International Twin Registry (n = 188 total pairs).

Exposures Kappa1 Ratio of Exposure Discordant Pairs2 Odds Ratio3 95% CI4 
1. Kappa calculated from the 93 double respondent twin pairs (pairs in whom both twins returned questionnaires) 
2 Total number of twin pairs in which case-twin was exposed and the unaffected co-twin was unexposed/Total number of twin pairs in which unaffected co-twin was exposed and the case-twin was unexposed 
3 Odds ratio estimated using conditional logistic regression using SAS Version 8.1 
4 Confidence interval estimated using conditional logistic regression 
3 or more childhood exanthems 6/15 0.4 0.2, 0.9 
Infectious mononucleosis 0.72 22/19 1.2 0.6, 2.1 
Cold Sores (Herpes Simplex 1) 0.70 18/10 1.8 0.8, 3.9 
Tonsillectomy 0.86 25/12 2.1 1.1, 4.1 
Appendectomy 0.84 33/11 3.0 1.5, 5.9 
Eczema 0.79 14/5 2.8 1.0, 7.8 
Sucked pacifier/thumb/fingers more as an infant/young child? 0.87 18/34 0.5 0.3, 0.9 
Put more things in mouth as an infant/young child? 0.86 10/30 0.3 0.2, 0.7 
Exposures Kappa1 Ratio of Exposure Discordant Pairs2 Odds Ratio3 95% CI4 
1. Kappa calculated from the 93 double respondent twin pairs (pairs in whom both twins returned questionnaires) 
2 Total number of twin pairs in which case-twin was exposed and the unaffected co-twin was unexposed/Total number of twin pairs in which unaffected co-twin was exposed and the case-twin was unexposed 
3 Odds ratio estimated using conditional logistic regression using SAS Version 8.1 
4 Confidence interval estimated using conditional logistic regression 
3 or more childhood exanthems 6/15 0.4 0.2, 0.9 
Infectious mononucleosis 0.72 22/19 1.2 0.6, 2.1 
Cold Sores (Herpes Simplex 1) 0.70 18/10 1.8 0.8, 3.9 
Tonsillectomy 0.86 25/12 2.1 1.1, 4.1 
Appendectomy 0.84 33/11 3.0 1.5, 5.9 
Eczema 0.79 14/5 2.8 1.0, 7.8 
Sucked pacifier/thumb/fingers more as an infant/young child? 0.87 18/34 0.5 0.3, 0.9 
Put more things in mouth as an infant/young child? 0.86 10/30 0.3 0.2, 0.7 

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author