HIV-1 DNA levels in peripheral blood mononuclear cells of patients with HIV-related non-Hodgkin's lymphoma

To the Editor: Human immunodeficiency virus (HIV) patients have an increased risk of non-Hodgkin's lymphoma (NHL), even in the era of antiretroviral therapy (ART).[1] The specific mechanisms of HIVrelated NHL (HIV-NHL) remain unclear. It is currently thought that HIV causes lymphoma primarily through chronic immune activation and an increased risk of infection with oncoviruses. In addition, HIV may have some other mechanisms that directly contribute to the occurrence of lymphoma.[2] Even with sustained ART, HIV-DNA remains in infected cells, creating a persistent reservoir of HIV infection. It is not known whether HIV reservoirs are involved in the development of HIV-NHL. In the present study, we detected HIV-1 DNA levels in peripheral blood mononuclear cells (PBMCs) and plasma inflammatory cytokines levels in HIV-NHL, and then evaluated their effects on HIV-NHL development and prognosis. The study enrolled four groups of patients admitted to the Shanghai Public Health Clinical Center between May 1, 2016, and November 30, 2020. The inclusion criteria were: (1) ART-naïve HIV-NHL patients (TN-NHL): who was diagnosed with HIV-NHL, had not started ART at the time of lymphoma diagnosis, and had no prior treatment for lymphoma. (2) ART-naïve HIV patients (TN-HIV): who was diagnosed with HIV infection, had not started ART, and had no evidence of NHL. (3) ART-experienced HIV-NHL patients (TE-NHL): who was diagnosed with HIV-NHL, had received ART ≥1 year and had undetectable HIV-RNA levels (<40 copies/mL) at the time of lymphoma diagnosis, had no prior treatment for lymphoma, and had no evidence of other tumors or infections. (4) ART-experienced HIV patients (TE-HIV): who was diagnosed with HIV infection, had received ART ≥1 year and had undetectable HIV-RNA levels (<40 copies/mL), and had no evidence of tumors or infections. Patients with incomplete clinical data or with no blood samples were excluded. Ethical approval has been obtained from the Ethics Committee of Shanghai Public Health Clinical Center (No. 2017-S022-04). Written informed consent was obtained from all participants. We measured CD4+ and CD8+ T-cell counts using flow cytometry (BD Biosciences, Franklin Lakes, NJ, USA) and plasma HIV-1 RNA levels using polymerase chain reaction (Cobas Amplicor, Roche, Basel, Switzerland). Genomic DNA was obtained from PBMCs using QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA, USA). HIV-1 DNA in PBMCs was quantified by a fluorescence-based, real-time HIV quantitative detection kit (SUPBIO, Guangzhou, China). The quantification range of this assay was 10–5 × 106 copies/106 PBMCs. A total of 10 kinds of inflammatory cytokines in plasma, including interleukin (IL)-1b, interferon (IFN)-α2, IFN-γ, IL-8, IL-10, IL-12 p70, IL-17a, IL-18, IL-23, and IL-33 were quantitatively detected using LEGENDplexTM (BioLegend, USA) Multi-analytic Flow Assay for Inflammatory Factors (Cat. No. 740809). SPSS 24.0 was used for statistical analysis (IBM Corp., Armonk, NY, USA). The normally distributed continuous variables were expressed as mean ± standard deviation (SD) and compared using the t-test; continuous variables with skewed distributions were shown as median (interquartile range [IQR]) and compared using the Mann–Whitney U test. Categorical variables are expressed in frequency and percentage terms and compared using chi-squared (χ2) tests or Fisher's exact tests. The correlation between HIV-1 DNA levels and plasma inflammatory cytokines was analyzed using Spearman correlation analysis and linear regression. Survival rates were evaluated using the Kaplan–Meier method, and differences were analyzed using a log-rank test. All tests were two-tailed tests, and P <0.05 was considered significant. A total of 68 ART-naïve patients were enrolled in the study, including 39 TN-NHL patients and 29 TN-HIV patients. There were no significant differences in gender (P = 0.733), age (Z = -0.099, P = 0.921), CD4+ counts (Z = -2.096, P = 0.061), CD8+ counts (Z = -1.488, P = 0.224), and HIV-RNA levels (163,500 copies/mL vs. 283,000 copies/mL, z = -1.096, P = 0.264) between the two groups. A total of 48 ART-experienced patients were included in the study, including 21 TE-NHL patients and 27 TE-HIV patients. There were no significant differences in gender (P >0.999), age (Z = -0.499, P = 0.618), and CD4 counts (Z = -1.507, P = 0.132) between the two groups. However, the duration of ART in the TE-HIV group was longer than that in the TE-NHL group (χ2 = 8.451, P = 0.015), and CD8+ counts were higher than that in the TE-NHL group (707 cells/μL vs. 521 cells/μL; Z = -2.109, P = 0.035). The main pathological types of enrolled HIV-NHL were diffuse large B-cell lymphoma (DLBCL) and Burkitt lymphoma (BL). There were 24 DLBCL cases (61.5%) and 13 BL cases (33.3%) in the TN-NHL group and 14 DLBCL cases (66.7%) and 7 BL cases (33.3%) in the TE-NHL group, respectively. The detailed demographic and clinical characteristics of the study participants were summarized in Supplementary Table 1, https://links.lww.com/CM9/B778. The median HIV-1 DNA levels in PBMCs were 791.15 (362.58–1894.98), 1132.65 (555.80–2166.49), 228.18 (97.99–343.24), and 353.49 (188.17–487.99) copies/1 × 106 PBMCs for TN-NHL, TN-HIV, TE-NHL, and TE-HIV groups, respectively. Significant differences in PBMC HIV-1 DNA levels were observed between the four groups of patients (χ2 = 40.541, P <0.001). HIV-1 DNA levels in ART-naïve patients were higher than that in ART-experienced patients (TN-NHL vs. TE-NHL, Z = -4.471, P <0.001; TN-HIV vs. TE-HIV, Z = -4.419, P <0.001). There were no significant differences in PBMCs HIV-1 DNA levels between the two ART-naïve patients with or without lymphoma (Z = -0.961, P = 0.337). However, among ART-experienced patients, HIV-1 DNA levels were lower in TE-NHL patients than that in TE-HIV patients (Z = -1.985, P = 0.047) [Figure 1].Figure 1: HIV-1 DNA levels in PBMCs of HIV-1-infected patients with or without NHL. HIV-1 DNA levels are expressed as Log copies per million PBMCs (Log copies/1 × E + 6 PBMCs). * P <0.05; † P <0.001. ART: Antiretroviral therapy; NHL: Non-Hodgkin's lymphoma; ns: No statistical significance; PBMCs: Peripheral blood mononuclear cells; TE-HIV: ART-experienced HIV-infected patients without lymphoma; TE-NHL: ART-experienced HIV-related non-Hodgkin's lymphoma; TN-HIV: ART-naïve HIV-infected patients without lymphoma; TN-NHL: ART-naïve HIV-related non-Hodgkin's lymphoma.We then compared inflammatory cytokine levels between the TE-NHL and TE-HIV groups and analyzed the correlation between HIV-1 DNA levels and inflammatory cytokine levels. The results showed that plasma IL-18 levels were significantly higher in TE-NHL patients than in TE-HIV patients (1496.14 pg/mL vs. 801.77 pg/mL; Z = -2.380, P = 0.017), and PBMC HIV-1 DNA levels were negatively correlated with the plasma IL-18 levels (r = -0.309, P = 0.032) [Supplementary Figures 1 and 2, https://links.lww.com/CM9/B778]. The survival analysis was performed on 46 HIV-NHL patients who had received anti-lymphoma therapy based on plasma inflammatory cytokine levels before chemotherapy. The results showed that IFN-γ levels were correlated with prognosis; patients with lower IFN-γ levels had better survival prognosis (hazard ratio [HR] = 0.193, 95% confidence interval [CI] 0.055–0.674; P = 0.019) [Supplementary Figure 3, https://links.lww.com/CM9/B778]. The clinical significance of HIV-1 DNA as a biomarker of viral reservoirs has been continuously studied and explored. In this study, we investigated potential risk factors for HIV-NHL development and prognosis in terms of HIV-1 DNA levels and plasma inflammatory cytokine levels. We found that for virologically suppressed patients, HIV-1 DNA levels in the PBMCs of the lymphoma group were significantly lower than those in the non-lymphoma group. This result is contrary to our expectations and requires further research to understand the exact cause and mechanism, which may provide new insight into the pathogenesis of HIV-NHL. HIV-1 DNA levels are affected by various factors, including viral characteristics, host and immune factors, and antiviral therapy.[3] It has been shown that after peripheral HIV-1 RNA is suppressed by ART, integrated HIV-1 DNA levels remain relatively constant with no apparent dependence on treatment length.[4] In this study, we enrolled virologically suppressed participants who had been on ART for at least one year, thus excluding the effect of ART on HIV-1 DNA levels prior to peripheral HIV-1 RNA inhibition. HIV-1 DNA levels may be closely related to CD8+ T cell activation and the inflammatory state in virologically suppressed HIV-infected patients.[5] Among the ART-experienced patients included in our study, CD8+ T cell counts in the lymphoma group were significantly lower than in the non-lymphoma group. As shown in other studies, correlations between total CD8+ T cell counts and HIV-DNA have also been observed.[6] Previous studies have revealed that the occurrence of lymphoma in people with HIV is related to immune activation. This study showed that IL-18 levels were significantly higher in the lymphoma group than in the non-lymphoma group among ART-experience patients, suggesting that elevated IL-18 levels may be a potential biomarker of HIV-NHL occurrence. IL-18 is a proinflammatory and immunomodulatory cytokine and a member of the IL-1 family. Elevated serum IL-18 levels have been found in patients with cutaneous T-cell lymphoma and NK-T-cell lymphoma, and are associated with a poor prognosis in DLBCL.[7] However, the role of IL-18 in the development of NHL in HIV patients requires further study due to the complexity of the human immune system and the impact of HIV and lymphoma on the immune system. The role of inflammation in the pathogenesis and prognosis of lymphoma has been extensively studied. Multiple abnormal immunological markers are independent prognostic factors in HIV-negative DLBCL.[8] In this study, we found that plasma IFN-γ level before chemotherapy is a prognostic factor for HIV-NHL, which needs to be verified with a larger cohort study. There are several shortcomings in this study. The sample size of the cohort is limited. There is a lack of longitudinal observations of the prognostic value of inflammatory cytokines for HIV-NHL. In addition, we analyzed the correlation between inflammatory cytokine levels and prognosis only by univariate analysis, rather than multivariate analysis. In summary, in patients with virologically suppressed HIV, PBMC HIV-1 DNA levels were significantly lower in NHL patients than in those without NHL, and immune activation was involved in NHL development. Plasma IL-18 levels, which are negatively correlated with PBMC HIV-1 DNA levels, may be potential biomarkers for NHL development, and increased IFN-γ level before chemotherapy is a risk factor for poor prognosis of HIV-NHL. Funding This work was supported by grants from the Shanghai Municipal Health Commission (No. shslczdzk01102) and the Shanghai Public Health Clinical Center (No. KY-GW-2023-03). Conflicts of interest None.