LAUSR

Background: Human polyomaviruses have a suspected role in carcinogenesis. In the setting of human immunodeficiency virus (HIV) infection, coinfection with MCV polyomavirus is associated with high risk of Merkel cell carcinoma. A role in lymphomagenesis has also been suggested by studies in the general population, but there are no prior studies in individuals with the acquired immunodeficiency syndrome (AIDS). Methods: We measured antibodies against 11 common polyomaviruses in archived serum and plasma samples from two prospective cohort studies of HIV infection. Patients with incident (n=28) and prevalent (n=38) AIDS-related non-Hodgkin lymphoma (NHL) were matched by age, sex, and CD4 count to 67 HIV-positive AIDS-free controls. Seroreactivity was measured by fluorescent bead-based multiplex serology, quantified as median fluorescence intensity (MFI). Logistic regression models were used to estimate adjusted odds ratios (ORs) and 95% confidence intervals (95% CIs) for NHL risk. Multinomial logistic regression compared incident and prevalent NHL cases to lymphoma-free controls and heterogeneity between case groups was evaluated by Wald tests. Differences in MFI were assessed by Kruskal-Wallis tests. Based on log-transformed MFI values for 15 masked duplicate samples, estimated coefficients of variation (CV) were 90% for all antibodies except LPV (CV 29%, ICC 67%). Results: Cases had significantly lower prevalence of antibodies to WU polyomavirus than lymphoma-free controls (64% vs. 85%), with OR 0.28 (95% CI 0.12-0.68). Cases had non-significantly lower prevalence of antibodies to JC, TSV, MCV and KI polyomaviruses (ORs 0.42-0.71); similar prevalence of antibodies to BK, HPyV6 and HPyV7 polyomaviruses (ORs 0.98-1.18); and non-significantly higher prevalence of antibodies to HPyV10, LPV and HPyV9 polyomaviruses (ORs 1.35-1.77). MFIs for anti-WU antibodies were also lower for cases (median 552.6, interquartile range [IQR] 140.3-1298.3) than controls (median 1096.3, IQR 410.3-2044.3; p-value 0.047). The association of anti-WU antibody with NHL was stronger in samples obtained post-diagnosis (OR 0.18, 95% CI 0.07-0.48) than pre-diagnosis (OR 0.66, 95% CI 0.19-2.31; p-heterogeneity=0.037). Conclusion: Our data do not directly implicate known polyomaviruses as lymphomagenic in the setting of HIV-associated immunodeficiency. Nevertheless, impaired antibody response to WU may be a harbinger of AIDS-related lymphoma. Citation Format: Minkyo Song, Noemi Bender, James J. Goedert, Cheryl A. Winkler, Nicole Brenner, Tim Waterboer, Charles S. Rabkin. Associations of polyomavirus seroreactivity with AIDS-related non-Hodgkin9s lymphoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3243.