Obesity and Hepatitis B Infection and Risk Of Primary CNS Lymphoma

The incidence of primary central nervous system lymphoma (PCNSL) has markedly increased during the past three decades. Advanced HIV disease, as well as congenital and iatrogenic immunodeficiency states are the only established risk factors. While the incidence of PCNSL continues to rise among older patients (>60), the vast majority of newly-diagnosed PCNSL patients are not overtly immune suppressed. The goal of this study is to identify novel risk factors for PCNSL that may explain the continued rise in incidence among non-HIV infected, immunocompetent populations.

A cohort of 72 HIV-negative patients diagnosed with primary and secondary CNS lymphoma who received ambulatory follow-up evaluation at University of California at San Francisco between 2009-2013 were frequency-matched to Bay Area population-based controls by age-group, sex and race with 1:4 case:control ratio. We regarded HBsAg positivity at baseline as evidence of chronic HBV infection, and HBcAb positivity at baseline as prior HBV infection. Body mass index (BMI) was modeled as normal (reference,<25), overweight (25-30) and obese (30+). Multivariable unconditional logistic regression was used to compute odds ratios (OR) as estimates of relative risk. Models were adjusted for matching factors and statistical significance was based on a two-sided p<0.05. Having been born in a country with a high prevalence of HBV was assessed as a potential confounder.

64 patients with PCNSL were identified. Among these, 28 (44%) were male, 69% Caucasian, median age at diagnosis was 61.5 years, 6 (10%) died during the follow-up period, and 7 (11%) had intraocular involvement. HBV infection (chronic or prior) and increased BMI were independently associated with increased risk of PCNSL; HBV infection: OR=14.8 (5.0-44), p<0.0001; BMI: obese vs. normal, OR=2.8 (1.2-6.5), p for trend=0.04. There was no evidence of confounding and no statistical interaction between HBV and BMI (p=0.72). HCV positivity also was assessed but analysis was constrained as only 3 patients were HCV antibody positive (1 also HBV positive). Results from descriptive analyses of intraocular involvement provided some evidence that these PCNSL patients were more likely to have been born in a country with moderate/high HBV prevalence (chi-square p=0.006). Obese PCNSL patients were statistically significantly younger (median age 54) than other patients (overweight median age 64.5, normal 63). Interestingly the Asian patients were younger (median age 57) than Caucasian (median age 62). Hep B patients were older (median age 66) compared with non Hep B (median age 56) but the difference was not statistically significant. All PCNSL patients were treated with high dose methotrexate-based systemic chemotherapy. 37 (58%) were determined to be in complete remission at the end of the follow-up period. For B-cell PCNSL patients who were treated with high-dose methotrexate (n=55), those who were obese had improved progression-free survival (PFS) compared to non-obese (P<0.04; HR 0.3). In contrast, a history of hepatitis B infection was associated with shorter PFS (P=0.04; HR 2.7). Notably, the apparent risk factors of obesity and/or hepatitis B impact approximately half of the 55 non-HIV-infected PCNSL patients in this analysis.

We believe this to be the first report of associations among obesity, hepatitis B infection and PCNSL. These findings may partly explain the increasing incidence of this subtype of NHL. We hypothesize that both hepatitis B infection as well as obesity may each promote inflammatory states that contribute to CNS lymphomagenesis. Further studies are warranted to confirm these findings and to explore underlying mechanisms of pathogenesis. Supported by Leukemia and Lymphoma Society and NIH R01CA139-83-01A1.

Rubenstein:Genentech: Research Funding; Celgene: Research Funding.