INTRODUCTION

Hepatitis C virus (HCV) may cause several extra-hepatic manifestations, including autoimmune diseases, mixed cryoglobulinemia (MC), and lymphoproliferative disorders (LPD).¹

In the case of LPD, the pathogenetic model is based on the continuous exposure of B-cell receptors (BCRs) to viral antigens due to HCV chronic infection determining a chronic antigen stimulation with subsequent antigen-dependent clonal expansion; specific genetic lesions may play a role by favouring an HCV-independent B-cell survival and proliferation.²

The histological subtypes more closely associated with HCV are marginal zone lymphoma (MZL), lymphoplasmacytic lymphoma (LPL), and, less frequently, diffuse large B-cell lymphoma (DLBCL).³ In addition, patients not developing overt LPD, may have clear documentation, in peripheral blood (PB) and/or bone marrow, of clonal B-cell populations in the absence of other signs or symptoms referred to LPD, thus fitting the diagnosis of monoclonal B lymphocytosis (MBL).⁴

There is evidence that therapies combining interferon (IFN) plus ribavirin (RBV) in HCV patients are associated with virological and haematological responses,⁴ through a mechanism mostly relying upon a direct effect of IFN on the proliferating clone rather than an indirect effect on the virus.⁵

In recent years, IFN-free direct antiviral agents (DAAs) have been introduced to treat HCV infection. Despite a sustained virologic response (SVR) in about 100% of cases,¹ only a fraction of patients affected by HCV-related LPD obtain complete or partial haematological responses, with a minority of cases requiring other conventional treatments.^{6, 7} These data have been recently pinpointed in the sole available prospective study evaluating DAAs in HCV-related LPD.⁸

PATIENTS AND METHODS

Twenty-three HCV-infected patients with LPD/MBL were included in the study from May 2015 to May 2023. Six patients, previously reported,^9-11 were included with updated follow-up. Criteria for patient selection were: LPD (including MBL) at first diagnosis or relapse, indolent course, and HCV infection (HCV-RNA positivity). HBV and HIV patients were excluded. All cases met the eligibility criteria for treatment with DAAs.¹² Clinical and biological data were recorded at baseline, every 4 weeks until the end of treatment (12th–24th week), and every 3/6 months after the end of treatment (Table 1; Table S1; Figure S1). Clinical responses have been evaluated by using response evaluation criteria in lymphoma (RECIL).¹³

Methods for laboratory testing, immunophenotyping, immunoglobulin sequencing and molecular evaluation of BCL2 rearrangements are detailed in Supporting Information.

RESULTS

Pre-treatment characterization

Patients' characteristics, HCV genotyping, and basal HCV-RNA levels are reported in Table S1. A MC diagnosis was done in 17/23 cases, 12 with high rheumatoid factor (RF) or low C4 levels; alanine transaminase (ALT) was elevated in 10/21 cases, and clinical manifestations of vasculitis were observed in 20/23 cases; regarding liver condition at pre-treatment, 8 cases had stage 4 fibrosis, 1 with oesophageal varices (Table 1; Table S1). As reported in Table 1 and Table S1, excluding 8 MBL (2 CLL-like, 4 MZL-like, and 2 not specified), the vast majority (11/15) was MZL.

Molecular and/or flow cytometry analyses revealed the presence of clonality in 19/23 cases (83%), as documented by IGHV gene rearrangement (18/19, 10/18 IGHV1, IGHV1-69 in 6/10; 6/18 IGHV3), κ/λ restriction (12κ/3λ) and BCL2 rearrangement (8 MBR and 2 mcr) (Table 1).

A similarity between HCDR3 sequences, as detected in BCR, to HCDR3 of known RFs was detected (cases #1, #6, #8, #9 and #19).^{10, 11} Moreover, cases #4, #5, #7, #11, #13 and #14 had HCDR3 resembling each other suggesting the recognition of yet unidentified (HCV-related) antigen(s) (Figure S2; Table S2).

Analysing the sub-clonal composition inside each pathological clone, we observed a certain degree of intraclonal diversification (Figure 1), the phenomenon in keeping with a continuous antigen-driven B-cell stimulation due to HCV occurring both at the MBL stage and upon overt lymphoma transformation.²

DISCUSSION

Since 2015 when IFN was replaced by DAAs therapy, few studies demonstrated a certain degree of remission in indolent low-grade Non-Hodgkin's lymphoma (NHL) after HCV eradication.^{8, 9, 14-16}

Our results demonstrated that the viral clearance, achieved in 100% of cases with good tolerability, induced haematological responses in 3/23 patients, that is, 1 MBL and 2 MZL. These data confirmed the capacity of DAA to induce a non-negligible rate of lymphoma responses, as previously reported,⁸ although in our cohort two of these patients received anti-CD20 therapies due to the persistence of neuropathy. At variance to previous studies,^{9, 16} herein we investigated the effects of DAA therapies at the molecular level by documenting the persistence of the B-cell clones bearing identical IGHV-IGHD-IGHJ rearrangements and/or the same t(14;18)-MBR/mcr translocation.

In the past, when HCV-related NHL was treated with IFN-alpha-based therapies, higher haematological responses were usually observed.⁵ Here, the haematologic response at the end of DAAs therapy was lower than in the past, allegedly for the lack of the direct effect of IFN-alpha on proliferating clones. Curiously, the only patient affected by overt NHL reaching a complete molecular response in the present series was an MZL (#16) treated with a DAA protocol which included IFN-alpha.

Previous studies, by investigating in HCV patients the B-cell clonality through light-chain κ/λ surface expression before and after DAAs treatment, found that circulating B-cell clones persisted in lymphoma patients for up to 2 years after HCV infection eradication,⁴ a κ/λ ratio normalization being observed only in patients without overt lymphoma. These results suggest that in the setting of indolent HCV-related LPD, the achievement of CR is not a critical factor for a long-term good outcome.^{8, 9}

The molecular characterization of the BCR in HCV-associated lymphoproliferative disorders may indicate the role of the antigen in driving clonal expansion. Stereotyped BCRs have been described in HCV patients, frequently endowed with both RF activity and specificity for the HCV-E2 protein.¹⁷ In keeping with this observation, we provided evidence of HCDR3 similarities with known RF-HCDR3,^{11, 18} as well as of novel similarities among LPD-derived HCDR3, suggesting the recognition of yet unidentified (HCV-related) antigen(s).

Evidence has also been provided that BCR from HCV-related LPD, even at the MBL stage, may exhibit marks of ongoing mutations of IGHV genes, due to the retention, after the neoplastic transformation, of the physiological machinery for somatic hypermutation, possibly due to chronic stimulation by HCV-related antigens.² Future studies aimed at analysing the Ig-light-chain composition and the mutational pattern of lymphomagenesis-related genes could provide novel correlations with different responses to DAAs, and enlighten the contribution of HCV to the lymphoma maintenance.^{10, 11}

In conclusion, our results indicate that patients with HCV-associated LPD receiving DAAs therapy showed excellent SVR, low toxicity, and good cryoglobulinemic vasculitis response but unsatisfactory haematological response with persistence of molecular B-cell clonality. These data support the recommendation that patients with HCV-associated LPD should be treated with DAAs as a first line, with the objective of eradicating the virus and HCV-related non-haematological symptoms, associating, when necessary, LPD-specific therapies to manage the B-cell tumour. Trials combining the use of DAAs and LPD-specific therapies could be of help in this setting.

AUTHOR CONTRIBUTIONS

Cesare Mazzaro designed the study, interpreted data and wrote the manuscript. Filippo Vit, Erika Tissino, Federico Pozzo, Robel Papotti, Tamara Bittolo, Milvia Casato, Laura Gragnani, Antonella Zucchetto, Massimo Degan and Riccardo Bomben performed and interpreted molecular studies, and contributed to data interpretation. Filippo Vit, and Riccardo Bomben generated the bioinformatics pipeline of analysis, and performed statistical analyses. Cesare Mazzaro, Marcella Visentini, Laura Gragnani, Milvia Casato and Anna Linda Zignego collected clinical data and contributed to data interpretation. Riccardo Bomben and Valter Gattei designed the study, interpreted data and wrote the manuscript.

FUNDING INFORMATION

The present study is supported in part by Progetto Ricerca Finalizzata RF-2018-12365790, Italian Ministry of Health, Rome, Italy; Associazione Italiana Ricerca Cancro (AIRC), Investigator Grant IG-21687; FIL project GR2020 bando Giovani Ricercatori; Fondazione Giulia Maramotti e Fondazione GRADE Onlus; Associazione Italiana contro le Leucemie, linfomi e mielomi (AIL), Venezia Section, Pramaggiore/Veneto Orientale Group, Italy; Linfo-check – Bando ricerca – contributo art. 15, comma 2, lett b, LR 17/2014; “5x1000 Intramural Program”, Centro di Riferimento Oncologico, Aviano, Italy.

CONFLICT OF INTEREST STATEMENT

The authors declare no competing financial interests.

ETHICS STATEMENT

This is a retrospective observational study approved by the Ethics Committee of Friuli-Venezia-Giulia Region (approval no. 2018-OS-147-CRO), Azienda Universitaria Policlinico Umberto I in Rome (Prot. 678/18) and University of Florence (Prot. BIO.16.014); informed consensus was obtained from all patients in accordance with the declaration of Helsinki.