One limitation of our study may be that the population of cases was highly heterogeneous, particularly

in terms of history of antiretroviral therapy. The cases were diagnosed between 1988 and 2007, i.e. before and during the cART era. We tried to minimize this effect by matching cases and controls for click here time period as well as for CD4 cell counts, in order that cases and controls should be similar in terms of antiretroviral regimens received. Another limitation of this study is that results of EBER staining were not available and therefore some lymphoma cases may have been EBER negative and not EBV-driven NHL. This may have minimized the predictive value of high EBV loads in PBMCs for progression to systemic lymphoma in our study. However, this potential

bias does not invalidate our finding that a high EBV load in PBMCs was associated with an increased risk of developing systemic B lymphoma. Finally, one could argue that EBV load may only be a surrogate marker for immunosuppression rather than an independent marker for the risk of occurrence of B systemic lymphoma. However, a high level of EBV DNA in PBMCs remained significantly associated with a higher risk of subsequent progression to systemic B lymphoma, after adjustment for CD4 cell count at sample date or for CD4 cell count nadir. Immune reconstitution is probably the main explanation for the lower Janus kinase (JAK) incidence of ARL following the widespread use Alectinib of cART. Nevertheless, some treated patients with satisfactory immune recovery (CD4 cell count > 350 cells/μL) still develop systemic B lymphoma [7, 27]. This underlines the need to identify additional risk factors for lymphoma in HIV-infected patients. Gasser et al. demonstrated that a lack of EBV-specific CD4 T-cell immunity was associated with

the occurrence of PBL irrespective of CD4 cell count [28]. Different groups reported that uncontrolled HIV replication during cART, assessed by HIV cumulative viraemia, was predictive of the development of AIDs-related lymphoma independently of the CD4 cell count, but the underlying mechanisms of this association remain unclear [6, 27, 29]. Recently, Bohlius et al. reported that, among patients under cART, those who had experienced decreasing CD4 cell counts despite suppression of HIV-1 replication were at a higher risk of developing Hodgkin lymphoma [30]. Jaffe et al. reported that, in untreated patients, initially low and further decreasing CD4 cell counts within 12 months before the diagnosis were predictive of both NHL and Kaposi sarcoma [31]. High EBV DNA blood loads have been reported in up to 20% of asymptomatic HIV carriers and high viral loads persisted over time in more than 80% of this subset of patients [32].