microRNA levels in paraffin-embedded indolent B-cell non-Hodgkin lymphoma tissues from patients chronically infected with hepatitis B or C virus

Abstract

Background: Epidemiological evidence links Hepatitis B Virus (HBV) and Hepatitis C Virus (HCV) to B-cell non-Hodgkin lymphoma (B-NHL). These B-NHLs, particularly those associated with HCV, may represent a distinct sub-group with peculiar molecular features, including peculiar expression of microRNAs (miRs).

Methods: Fourteen formalin fixed paraffin embedded (FFPE) tissues from HBV+, HCV+ and HBV-/HCV- indolent B-NHL patients were analyzed for levels of 34 selected miRs by quantitative Real-Time PCR. Reactive lymph nodes (RLNs) from HBV-/HCV- patients were included as non-tumor control. Statistical analysis of output data included Pearson and Spearman correlation and Mann-Whitney test and were carried out by the STATA software.

Results: MiR-92a was decreased exclusively in HBV-/HCV- B-NHLs, while miR-30b was increased in HBV+ and HCV+ samples, though only the HCV+ achieved full statistical significance. Analysis of a small subset of B-NHLs belonging to the same histological subtype (Nodal Marginal Zone Lymphoma) highlighted three miRs associated with HCV infection (miR-223, miR-29a and miR-29b) and confirmed decreased level of miR-92a in HBV-/HCV- samples also when considering this restricted B-NHL group.

Conclusions: Although caution is needed due to the limited number of analyzed samples, overall the results suggest that differences at the miR expression level exist between indolent B-NHLs developed in patients with or without HBV or HCV infection. The identification of three further miRs associated with HCV by analyzing histologically homogeneous samples suggests that variations of miR levels possibly associated with HBV or HCV may be obscured by the tissue-specific variability of miR level associated with the different histological subtypes of B-NHL. Thus, the identification of further miRs will require, in addition to an increased sample size, the comparison of B-NHL tissues with the same histological classification.

Figure 1

Level of miR-92a and miR-30b in HBV+ B-NHL, HCV+ B-NHL, Ctrl B-NHL and RLN groups. Box-plot representation of the level of miR-92a (A) and miR-30b (B) in HBV+ B-NHL, HCV+ B-NHL, Ctrl B-NHL and RLN sample groups. HBV+ B-NHL: B-NHLs from HBV positive patients; HCV+ B-NHL: B-NHLs from HCV positive patients; Ctrl B-NHL: control B-NHLs from patients negative for both HBV and HCV. The level is reported as 2^-Ct values. p values < 0.1 are reported, and those < 0.05 are shown in bold. C and D: summary of fold-change (vs. reference Ctrl B-NHL and RLN, respectively) and p-value (Mann-Withney test). Fold-change was calculated as ratio between median level of each group vs. median level of the reference group.

Figure 2

Level of miR-223, miR-92a, miR-29a and miR-29b in HCV+ NMZL, Ctrl NMZL and RLN groups. A: box-plot representation of the level of miR-223, miR-92a, miR-29a and miR-29b in HCV+ NMZL, Ctrl NMZL and RLN sample groups. HCV+ NMZL: NMZLs from HCV positive patients; Ctrl NMZL: NMZLs from HCV negative patients; RLN: non-neoplastic Reactive Lymph Nodes. The level is reported as 2^-Ct values. p values <0.1 are reported, and those < 0.05 are shown in bold. B and C: summary of fold-change (vs. reference Ctrl NMZLs and RLNs, respectively) and p-value (Mann-Whitney test). Fold-change was calculated as ratio between median level of each group vs. median level of the reference group.