Viral lymphomagenesis: beyond the usual suspects.

Abstract

In this issue of the Journal, Fama et al (2018) present the results of a large case-control study evaluating the association between the infection by human pegivirus and the risk of developing lymphoma, as well as lymphoma prognosis. In this study, approximately 2 000 cases of lymphoma and 1 500 controls from the Molecular Epidemiology Resource Cohort Study of the Iowa/Mayo Lymphoma Specialized Program of Research Excellence were included, and the authors showed that infection by human pegivirus is associated with a 2-fold higher risk of developing lymphoma, after adjusting for other relevant risk factors. Specifically, human pegivirus infection was associated with an increased risk of developing follicular lymphoma (3.6-fold), mantle cell lymphoma (3.4-fold), T cell lymphoma (3.4-fold), diffuse large B-cell lymphoma (DLBCL; 2.4-fold) and marginal zone lymphoma (1.9-fold). There was no association with risk of developing chronic lymphocytic leukaemia or Hodgkin lymphoma. Also, there was no demonstrable risk associated with a worse prognosis in any of the lymphoma subtypes evaluated. Based on these results, human pegivirus joins the list of viruses associated with an increased risk of lymphomagenesis, which include human immunodeficiency virus (HIV), human T-lymphotrophic virus 1 (HTLV-1), Epstein–Barr virus (EBV) and human herpesvirus 8 (HHV-8), among others (Castillo et al, 2014). HIV infection has been associated with an increased risk of DLBCL and Hodgkin lymphoma, HTLV-1 is associated with adult T cell leukaemia/lymphoma (ATLL), EBV with DLBCL, Hodgkin lymphoma, primary central nervous system lymphoma and Natural Killer/T-cell lymphoma, and HHV-8 with primary effusion lymphoma. The quality of a case-control study is evaluated by having in mind the selection of cases and controls, their comparability and the assessment of the exposure (Wells et al, 2014). Let's evaluate the quality and methodology of the present case-control study in depth. With regard to patient selection, the cases were defined by a central pathology evaluation, minimizing selection bias. Also, the cases can be considered representative, as they included a large number of clinic-based cases. Similarly, the controls originated from the same clinics as the cases. Finally, the controls did not have a history of lymphoma, providing a true incidence of lymphoma in this group as a comparator. With regard to comparability, the cases and controls were frequency-matched with regards to age, sex and location of residence. Relating to exposure, the assessment of human pegivirus infection was evaluated in both cases and controls following the same method of real-time polymerase chain reaction, with a similar rate of non-response of approximately 35% for both cases and controls. Based on the recommendations of the Newcastle-Ottawa Scale (Wells et al, 2014), the present case-control study meets most of the features of studies of the highest quality. However, no study is perfect and there are limitations in the present study that are worth discussing. The only one weak feature in the study design is the inclusion of clinic-based cases and controls instead of community-based cases and controls. In my opinion, this could have introduced a small risk of selection bias, although not large enough to adversely impact the validity of the study. A second limitation is that in about a third of the cases, the blood samples analysed for human pegivirus infection were obtained after therapy for lymphoma was initiated. The authors do not provide information on the timing of blood sampling in this group of patients, which could have been of interest. Also, the authors do not provide the cumulative incidence of lymphoma in patients with active human pegivirus infection. Previous studies, however, have suggested a latency period of at least 4 years from human pegivirus infection to a statistically significant increased risk of lymphoma development (Chang et al, 2014). Finally, the proportion of seropositive individuals was higher in this study than in previous studies, raising concerns on the selection of patients. However, it is possible that the higher rate of human pegivirus infection in this study was a consequence of performing tests with higher sensitivity. To finalize, I would like to point out the biggest limitation of this study, which, in my opinion, is the clinical applicability of the findings. I believe the results of this study are of clinical relevance and open up a series of avenues for new biological and epidemiological research. I do not believe, however, that clinicians (or patients) should obtain (or request) human pegivirus testing in the light of these results. Additional research is needed to further elucidate the true incidence of human pegivirus infection, and also to better understand the underlying biology of lymphomagenesis in the context of an active human pegivirus infection.