Hepatitis C virus-associated B-cell non-Hodgkin’s lymphomas: what do we know?

Abstract

Epidemiological studies have shown an increased risk of developing B-cell lymphomas in patients with chronic hepatitis C virus (HCV) infection. There is, however, a great geographic variability and it remains unclear whether additional environmental and genetic factors are involved or whether the international discrepancies represent simply a consequence of the variable prevalence of HCV infection in different countries. Other confounding factors may affect the comparability of the different studies, including the method of HCV assessment, the selection of normal controls, the lymphoma classification used and the year of publication. The most convincing evidence for a causal relationship comes from the observation, mainly limited to some indolent subtypes, of B-cell lymphoma regressions after successful HCV eradication with antiviral treatment. Yet, the molecular mechanism of HCV-induced lymphomagenesis are mainly hypothetical. According to most plausible models, lymphoma growth is a consequence of continuous antigenic stimulation induced by the chronic viral infection. This review will summarize the current knowledge on HCV-associated lymphomas and their management.

Introduction

The hepatitis C virus (HCV) is a hepatotropic and lymphotropic virus responsible for acute hepatitis and chronic liver disease. With almost 160 million people estimated to be chronically infected worldwide [Lavanchy, 2011], HCV is a global health problem. However, spread of the chronic infection is very variable among countries. The highest prevalence of infected people (>10%) is reported in Egypt, Central Africa, Mongolia and Bolivia [Zaltron et al. 2012], while Europe accounts for around 15 million people [Hope et al. 2014]. The Asian continent accounts for the largest number of infected persons, China and India together having more HCV infections than the whole of Europe or the Americas [Sievert et al. 2011].

The liver is not the only target organ; the most common extrahepatic manifestation described in HCV-positive individuals is type II mixed cryoglobulinemia (MC), HCV infection being present in nearly all patients with MC. Moreover, several lines of evidence have suggested a causative link between HCV infection and the development of a portion of malignant non-Hodgkin’s lymphomas (NHLs).

This review will summarize the available epidemiological, clinical and biological evidence on the etiologic association between HCV and some NHLs. The possible pathogenetic mechanisms will also be addressed.

Is there a link between MC type II and HCV-associated lymphomas?

Type II MC is a disorder characterized by circulating cryoglobulins [i.e. complexes of polyclonal immunoglobulin G (IgG) and monoclonal IgM rheumatoid factors that become insoluble at reduced temperature] [Agnello et al. 1992]. HCV infection is present in 80–90% of patients with MC, even in the absence of chronic liver disease, whilst the prevalence of MC in subjects with HCV infection is widely variable depending on geographical areas.

Histological lesions of MC are mainly related to vasculitis, caused by the deposit of immune complexes in the wall of small vessels. These immune complexes have been found to be highly enriched in HCV-RNA [Agnello et al. 1992], but the biological relationship between HCV and MC is not fully understood. Some studies highlight the predisposition of HCV to select a restricted B-cell repertoire in response to chronic antigenic stimulation, favoring the preferential use of IGHV1-69, IGKV3-A27 and IGKV3-20 [Gorevic and Frangione, 1991; Allander et al. 2000; Libra et al. 2010]. Other studies show the presence of monoclonal B lymphocytes in the liver, bone marrow and peripheral blood of patients with MC [Viswanatha and Dogan, 2007].

Moreover, MC has been shown to be associated with a certain risk of developing lymphomas [Zignego et al. 2012; Gragnani et al. 2015] and it has been reported that 5–10% of patients with MC can progress to overt NHL [Monti et al. 2005; Gragnani et al. 2015]. Interestingly, a few studies also showed that the attainment of a sustained virologic response (SVR) after anti-HCV treatment (mainly interferon-α ± ribavirin) can also reduce the risk of lymphoma development [Cacoub et al. 2014; Gragnani et al. 2015].

What evidence comes from epidemiological studies linking NHL and HCV?

Over the last 20 years, studies have repeatedly reported an epidemiological association between HCV infection and lymphoproliferative disorders. Epidemiological data show no risk or only a slightly increased risk of T-cell NHL and Hodgkin’s lymphoma [De Sanjose et al. 2008; Nieters et al. 2006; Dal Maso and Franceschi, 2006], and very few and inconsistent data exist on the risk of multiple myeloma with respect to HCV seropositivity [Dal Maso and Franceschi, 2006]. The strongest evidence is available for B-cell NHL, including cases without MC [Musto, 2002; Matsuo et al. 2004; Negri et al. 2004; Dal Maso and Franceschi, 2006]. Remarkably, the consistency of the association between HCV and B-cell NHL varies by geographical region and the correlation is particularly evident in countries with a high prevalence of HCV infection. The percentage of NHL attributable to HCV ranges between 0% and 50% and this huge variability can be, at least in part, explained by the epidemiology of the infection, which is sporadic in some areas and endemic in others. Meta-analyses of the published studies [Gisbert et al. 2003; Matsuo et al. 2004; Negri et al. 2004; Nieters et al. 2006; De Sanjose et al. 2008] (Tables 1 and 2) show that at least 10% of NHLs may be due to HCV infection in endemic areas such as Italy or Japan and that the overall relative risk of being infected for patients with B-cell NHL is roughly two to four times higher than in the general population (e.g. in the large European cohort study EPILYMPH, enrolling patients from northern Europe, Spain and Italy, an estimated two fold global risk of lymphoma has been described [Nieters et al. 2006]). On the other side, in most low HCV prevalence areas, such as Scandinavia, the UK or Canada, lower or no evident correlation has been demonstrated, maybe due to numbers of infected cases, while in the USA a modest but significant association is found [Engels et al. 2004; Giordano et al. 2007]. Nevertheless, the possible presence of some confounding factors related to the study design and its year of publication should be taken into account, because they may affect the comparability of the published studies [Gisbert et al. 2003; Engels et al. 2004; Matsuo et al. 2004; Negri et al. 2004; Dal Maso and Franceschi, 2006; Nieters et al. 2006; Giordano et al. 2007; De Sanjose et al. 2008].

Table 1.

Systematic reviews and meta-analyses of studies on the risk of HCV-associated lymphomas.

Study	No. of patients with NHL	No. of studies	Overall prevalence of HCV-associated NHL		Odds ratio* (95% CI)
			Weighted mean (95% CI)	Range across individual studies
Gisbert et al. [2003]	5542	48	13% (12.1–13.9%)	0.0–50%	n.a.
Negri et al. [2004]	4574	60	18.3% (17.1–19.4%)	0.0–37.1%	n.a.
Gisbert et al. [2003]	818	10	16.7% (14.2–19.5%)	0.0–32%	10.8 (7.4–16)
Negri et al. [2004]	1942	14	16.9% (15.3–18.7%)	0.0–37.1%	3.4 (2.9–4.1)
Matsuo et al. [2004]	4049	23	12.8% (11.8–13.9%)	0.5–37.1%	5.7 (4.1–7.9)
Dal Maso and Franceschi [2006]	4678	15	10.8% (9.9–11.7%)	0.5–42.7%	2.5 (2.1–3.0)
De Sanjose et al. [2008]	4784	7	3.6% (3.1–4.2%)	0.0–30.6%	1.8 (1.4–2.3)

^*Estimated in case-control studies.

CI, confidence interval; HCV, hepatitis C virus; NHL, non-Hodgkin’s lymphoma.

Table 2.

Geographic variation of HCV prevalence in patients with NHL^*.

Country	No. of HCV+	No. of NHL	No. of studies	Prevalence	(95% CI)
Italy	623	3487	24	17.9%	(16.6–19.2%)
Spain	59	607	4	9.7%	(7.4–12.4%)
France	28	808	7	3.5%	(2.3–5%)
Germany	5	174	2	2.8%	(0.9–6.6%)
UK	0	201	4	0%	(0–0.2%; one-sided 97.5% CI)
Switzerland	17	180	1	9.4%	(5.6–14.7%)
Eastern Europe	30	110	2	27.3%	(19.2–36.6%)
Europe	762	5567	44	13.7%	(12.7–14.6%)
Japan	85	771	7	11.0%	(8.9–13.5%)
Others$	41	677	8	6.1%	(4.4–8.1%)
Asia	126	1448	15	8.7%	(7.3–10.2%)
USA	111	1871	7	5.9%	(4.9–7.1%)
Canada	2	188	2	1.1%	(0.1–3.8%)
Brazil	8	87	1	9.2%	(4.0–17.3%)
North America	113	2059	9	5.5%	(4.5–6.6%)

^*Data from Gisbert et al. [2003]; Negri et al. [2004]; Matsuo et al. [2004]; Dal Maso and Franceschi [2006].

^$Others = Turkey (five studies), Thailand (one study), Saudi Arabia (one study), South Korea (one study).

CI, confidence interval; HCV, hepatitis C virus; NHL, non-Hodgkin’s lymphoma.

Which NHL histological subtypes are more frequently associated with HCV?

Some early epidemiological studies indicated a possibly higher prevalence of some B-cell NHL subtypes in the HCV-infected population [Musto, 2002; Arcaini et al. 2007; Libra et al. 2010]. Although initially unclear [Dal Maso and Franceschi, 2006], recent studies with larger number of cases show a prevalence of diffuse large B-cell lymphomas (DLBCLs), lymphoplasmacytic lymphomas (LPLs) and marginal zone lymphomas (MZLs) (especially splenic MZL and non-gastric MZL of mucosa-associated lymphoid tissue). The EPILYMPH study described DLBCL as the lymphoma subtype most frequently associated with HCV infection [odds ratio (OR) 2.19] [Nieters et al. 2006]. A very similar risk was later found in the very large International Lymphoma Epidemiology Consortium (Interlymph) case–control study, which provided a subtype-specific analysis, showing that HCV infection is increasingly associated not only with DLBCL (OR 2.24), but also with MZL (OR 2.47) and LPL (OR 2.57) [De Sanjose et al. 2008].

What do we know about the biological mechanisms of HCV-related lymphomagenesis?

HCV is an enveloped, positive-stranded RNA virus belonging to the Flaviviridae family [Weng and Levy, 2003]. It does not integrate into the host genome and does not contain an evident oncogene [Marcucci et al. 2012]. Viral replication has been clearly demonstrated in hepatocytes and this proves the direct role of HCV in the development of hepatocellular carcinoma [Weng and Levy, 2003; Bouvard et al. 2009]. It has also been demonstrated that HCV is able to infect lymphocytes; however, its ability to replicate in vivo in the infected lymphocytes is yet to be established [Sung et al. 2003; Machida et al. 2004; Ito et al. 2010; Pawelczyk et al. 2013; Tucci et al. 2013] and, even so, replication in the lymphocytes would not be sufficient to fully demonstrate a pathogenetic role in NHL development.

A causal relationship between HCV and NHL may be based on somewhat dissimilar theoretical mechanisms: the direct oncogenic effect or an indirect mechanism. The latter, in the form of an antigen-driven lymphocyte proliferation, analogous to what is observed in Helicobacter pylori-associated gastric MZL [Zucca et al. 2014], seems to be the most plausible link [Marcucci et al. 2012]. Nevertheless, diverse oncogenic stimuli do not necessarily have to be mutually exclusive; while the oncogenic potential of single events may be low, the accumulation of at least two oncogenic signals can be the prerequisite for lymphoma occurrence [Marcucci and Mele, 2011].

The ability of HCV to stimulate B-cell proliferation finds confirmation in numerous reports [Weng and Levy, 2003; Rosa et al. 2005; Ng et al. 2014]. While some studies confirm that the double binding of CD19/CD21/CD81 complex (via HCV-E2 protein) and the B-cell receptor (BCR, by specific HCV antigens) results in a reduced threshold for B-cell activation and proliferation [Weng and Levy, 2003], others suggest that the engagement of CD81 may be sufficient to elicit the B-cell proliferation even in the absence of a BCR coligation and the BCR involvement in this context remains controversial [Rosa et al. 2005]. Chronic antigenic stimulation may nonetheless play an important role in the development of an initial polyclonal B-cell expansion and predispose people to genetic aberrations. A restricted combination of the IGHV gene repertoire expressed in the monoclonal rheumatoid factor-like IgM component of the type II MC and in monoclonal antibodies from patients with HCV infection has been detected, with high frequency of V regions encoded by IGHV1-69 and IGHVk3-A27 [Gorevic and Frangione, 1991; Allander et al. 2000; Libra et al. 2010]. Also HCV-NHLs have been found to express analogue regions of the IgV genes [Ng et al. 2014]. HCV-associated NHLs may derive from B-cell clones activated during HCV infection (the malignant counterpart of type II MC), which following additional genomic alterations can develop into overt lymphoma [Libra et al. 2008]. Further reports also identified genetic instability, increased aneuploidy and reduced Rb protein expression in HCV-infected cells, potentially favoring a neoplastic transformation [Machida et al. 2009a, 2009b; Ito et al. 2011; Marcucci and Mele, 2011]. Moreover, HCV positive individuals were characterized by high levels of induced cytidine deaminase (AID) in the B-cells. The inappropriate expression of AID might acts as a DNA mutator, thus enhancing genetic susceptibility to mutagenesis [Ito et al. 2011]. It has also been reported that HCV infection is responsible for nitric oxide production, and the oxidative stress associated with the production of nitric oxide and reactive oxygen species can cause mitochondrial and DNA damage [Machida et al. 2006].

Immune system deregulation caused by HCV may also be involved in lymphoma development. Chronic hepatitis C is characterized by impaired T-lymphocyte function, allowing virus to escape immune surveillance and establish a persistent infection. HCV core protein differentially regulates T- and B-cell functions through two negative signaling pathways, programmed death 1 (PD-1) and suppressor of cytokine signaling 1 (SOCS-1) [Yao et al. 2005, 2007; Frazier et al. 2010]. Recently, it has been shown that the PD-1 pathway is involved in regulating T-regulatory cell number and function, suppressing T-cell responses in patients with HCV-associated B-cell NHL. A more profound T-cell suppression and higher PD-1 levels have been detected in HCV-associated lymphomas compared with non-HCV-positive lymphomas, chronic HCV infection and normal controls [Ni et al. 2011]. Moreover, in the setting of HCV-NHL, it has been observed that blocking PD-1 signal leads to T-cell function recovery, supporting a mechanism by which HCV might cause a T-cell immune dysregulation that is associated with NHL development [Yao et al. 2011].

Finally, different signaling pathways have been demonstrated to potentially mediate HCV-induced lymphoma growth. However, as the majority of the data derives from in vitro or murine models [Feldmann et al. 2006; Sene et al. 2007; Ito et al. 2011; Marcucci and Mele, 2011], further research is required. Gene expression profiling of lymphoma B cells from HCV-transgenic mice has demonstrated that both canonical and an alternative nuclear factor κB pathway as well as the downregulation of some microRNAs may lead to lymphomagenesis [Kasama et al. 2014]. A reduced expression of miR-26b has been found in patients with HCV-positive splenic marginal zone lymphoma (SMZL) [Peveling-Oberhag et al. 2012]. Moreover, a decreased expression of miR-138-5p and an increased expression of miR-511-5p, miR-147a and miR-147b have been associated with bad prognosis in HCV-associated DLBCL [Augello et al. 2014]. Also, deregulation of the NOTCH signaling pathway in a subset of HCV-associated DLBCLs has been detected in a recent paper. In this study a comparison was made between HCV-positive and HCV-negative DLBCLs. Mutations of NOTCH2 were found in 20% and of NOTCH1 in 4% of the cases of HCV-positive DLBCL, while only 1/64 HCV-negative cases was found to harbor a NOTCH2 mutation [Arcaini et al. 2015]. Interestingly, the presence of these mutations was associated with a shorter overall survival (OS) and with the presence in the diagnostic biopsy of a small cell component that suggests a possible histologic transformation of the DLBCL [Arcaini et al. 2015]. It is already known that NOTCH is a pathway involved in the pathogenesis of SMZL: this finding seems to correlate with the observation that a high proportion of DLBCLs in HCV-positive individuals can arise from a preceding MZL [Besson et al. 2006; Michot et al. 2015].

Which is the clinical evidence suggesting a pathogenetic role of HCV in the growth of NHL?

The potential association between HCV and NHL has been examined in many case–control studies, as reported above. The results show a variable strength in this association, which can be due either to methodological differences among studies or to demographic heterogeneity of the populations studied. In part this variability could also reflect a possible specific association between HCV and only certain NHL subtypes. The biological plausibility of this relationship is supported by several reports. Still, the response of NHL to a treatment targeting the infection provides the strongest evidence of a potential causative role of HCV in lymphomagenesis. This was initially observed in nine patients affected by splenic lymphoma who underwent interferon α and ribavirin as front-line treatment [Hermine et al. 2002]. All patients who obtained a HCV clearance (HCV-RNA undetectable) obtained a lymphoma response. On the contrary, in the same study, none of the six additional patients without HCV infection who underwent antiviral therapy achieved lymphoma regression. This validates the argument against a direct antitumor effect of the interferon itself [Hermine et al. 2002]. This clinical observation of lymphoma regression in HCV-positive splenic lymphomas after interferon α treatment is similar to that observed in gastric marginal zone lymphoma of mucosa-associated lymphoid tissue (MALT) after anti-Helicobacter pylori treatment [Zucca et al. 2013, 2014].

Later reports confirmed that antiviral treatment with interferon α with or without ribavirin can be efficacious for treatment of HCV-associated indolent lymphomas, mainly but not only of MZL type [Kelaidi et al. 2004; Saadoun et al. 2005; Vallisa et al. 2005; Mazzaro et al. 2009; Pellicelli et al. 2011; Arcaini et al. 2014; Michot et al. 2015] (Table 3).

Table 3.

Lymphoma responses following anti-HCV treatment in HCV-associated indolent NHLs.

Reference	No. of patients	Lymphoma subtype	MC	Antiviral treatment				Hematological response
				IFN	IFNαRBV	PegIFN/RBV	IFN/DAAs	ORR	CR
Hermine et al. [2002]	9	SLVL (9)	6	7	2	0		9 (100%)	8 (89%)
Kelaidi et al. [2004]	8	SLVL (3)	8	2	1	5		5 (62%)	5 (62%)
		SMZL (1)
		EMZL (2)
		Disseminated MZL (2)
Saadoun et al. [2005]	18	SLVL (18)	18	8	10	0		18 (100%)	14 (78%)
Vallisa et al. [2005]	13	LPL (4)	5	0	0	13		9 (69%)	7 (54%)
		SMZL (4)
		NMZL (2)
		EMZL (2)
		FL (1)
Mazzaro et al. [2009]	18	LPL (16)	13	0	8	10		13 (72%)	9 (50%)
		SLVL (1)
		FL (1)
Pellicelli et al. [2011]	9*	SMZL (3)	5	0	0	9		7 (78%)	5 (55%)
		EMZL (3)
		FL (2)
		NMZL (1)
Arcaini et al. [2014]	100$	SMZL (23)	35	10‡	23	63		77 (77%)	54 (54%)
		NMZL (12)
		EMZL (25)
		LPL (7)
		FL (5)
		CLL/SLL (1)
		Others (27)
Michot et al. [2015]	14	MZL (14)	n.a.		0	8	6	11 (78%)	8 (57%)

^*Only nine patients receiving antiviral treatment as first-line therapy are reported.

^$Only 100 patients treated with antiviral treatment frontline are reported.

^‡Four additional patients received PegIFNα alone.

CLL/SLL, chronic lymphocytic leukemia/small cell lymphocytic lymphoma; CR, complete response; DAA, new direct antiviral agent; EMZL, extranodal marginal zone lymphoma of the mucosa-associated lymphoid tissue; FL, follicular lymphoma; HCV, hepatitis C virus; IFN, interferon α; LPL, lymphoplasmacytic lymphoma; MC, mixed cryglobulinemia; MZL, marginal zone lymphoma; n.a., not applicable; NHL, non-Hodgkin’s lymphoma; NMZL, nodal marginal zone lymphoma; PegIFN, pegylated interferon α; ORR, overall response rate (complete + partial responses); RBV, ribavirin; SLVL, splenic marginal zone lymphoma with villous lymphocytes; SMZL, splenic marginal zone lymphoma.

The role of antiviral therapy seems also to be protective against the risk of lymphoma, indirectly supporting the hypothesis of a causal relationship between HCV and lymphomas. In this regard, interesting data come from a large Japanese retrospective study which compared 501 patients with chronic HCV, who had never received interferon α therapy, and 2708 patients with HCV infection, who received interferon α therapy. Patients with persistent infection had a significantly higher incidence of lymphoma than patients attaining a SVR, thus indicating that HCV eradication can reduce the risk of developing lymphomas [Kawamura et al. 2007].

Are there peculiar clinical features of HCV-associated lymphomas?

The clinical presentation of HCV-associated NHL is largely related to the histologic type of the lymphoma itself. Some distinctive characteristics can however be highlighted, which may be, at least in part, attributable to the presence of the HCV infection itself. First, patients who are HCV-positive more frequently present high transaminase levels, monoclonal gammopathies, autoimmune phenomena, rheumatoid factor and asymptomatic cryoglobulinemia [Musto, 2002; Michot et al. 2015]. Second, at presentation, the involvement of the spleen or extranodal sites, which are also common target organs of HCV (i.e. the liver and the salivary gland) is quite recurrent [De Vita et al. 1997].

The clinicopathological features of HCV-associated DLBCL has been addressed by different reports over the years [Tomita et al. 2003; Besson et al. 2006; Visco et al. 2006; Park et al. 2008; Merli et al. 2014; Michot et al. 2015]: HCV-positive DLBCLs commonly present in older patients, with advanced stage (mainly due to extranodal localizations), elevated LDH (serum lactate dehydrogenase, which, at least in part, is due to the concomitant hepatitis) and unfavorable International Prognostic Index (Table 4). Further studies also suggest that HCV-associated DLBCL arises more frequently from a preceding low-grade MZL in comparison to de novo DLBCL [Besson et al. 2006; Michot et al. 2015].

Table 4.

Clinical features at presentation of HCV-associated diffuse large B-cell lymphoma.

Clinical features	Studies
Reference	Tomita et al. [2003]	Besson et al. [2006]	Viscoet al. [2006]	Park et al. [2008]	Ennishi et al. [2010]	Merli et al. [2014]	Michot et al. [2015]	Chen et al. [2015]
No. of patients	25	26	156	32	131	535	45	29
Age < 60 years	28%	81%	34%	41%	n.r.	27%	n.r.	n.r.
Female sex	28%	35%	53%	50%	40%	51%	22%	34%
Performance status 0–1	52%	73%	81%	81%	88%	78%	80%	76%
Stage III–IV	80%	73%	53%	34%	49%	68%	71%	62%
Int.–high/high IPI	68%	64%	44%	22%	40%	54%	44%	52%
Extranodal sites >1	24%	46%	67% (⩾1)	53%	27%	35%	73%	41%
Spleen involvement	4%	46%	34%	18%	18%	35%	27%	7%
Bone marrow involvement	28%	n.r.	16%	0	9%	21%	n.r.	10%
Liver involvement	8%	15%	11%	n.r.	9%	15%	22%	10%
High LDH levels	76%	77%	62%	60%	62%	55%	57%	76%
B symptoms	44%	n.r.	37%	22%	n.r.	31%	n.r.	n.r.
High GOT/GPT levels	28%	44%	n.r.	n.r.	7% with grade >1	46%	n.r.	n.r.
Transformed from low grade	n.r.*	32%	8%	n.r.	4%	n.r.	36%	n.r.

^*The working formulation was used to define histology by Tomita, who reports 20 large cell lymphomas and other aggressive histological types (including four cases of T-cell NHL).

GOT, glutamic oxaloacetic transaminase; GPT, glutamic pyruvic transaminase; HCV, hepatitis C virus; int., intermediate; IPI, international prognostic index; LDH, serum lactate dehydrogenase; n.r., not reported.

Concerning indolent lymphomas, two peculiar clinical presentations have been outlined:

1. Primary SMZL with type II MC affects mainly female patients (70% of cases), most of them having symptomatic cryoglobulinemia (vasculitis, arthralgia, peripheral neuropathy) and circulating villous lymphocytes [Saadoun et al. 2005].

2. Subcutaneous ‘lipoma-like’ extranodal marginal zone B-cell lymphoma of MALT type is described in older women presenting single or multiple soft and mobile subcutaneous nodules, without involvement of the skin and the cutaneous adnexa; the clinical course is usually indolent. Analogous to what is described in splenic lymphomas, objective lymphoma regressions have been reported after HCV eradication [Paulli et al. 2010].

How to manage NHL in patients with HCV infection

According to the guidelines of the European Association for the Study of the Liver, antiviral therapy in HCV-related chronic hepatitis should aim at preventing hepatic and extrahepatic complications [European Association for the Study of the Liver, 2014]. In patients affected by lymphoma, HCV eradication may be indicated because it may either induce a lymphoma regression (in some subtypes) or prevent a relapse, removing the antigenic stimulation. With the development of new drugs able to induce earlier and better virologic responses, HCV treatment is now rapidly changing, particularly in the case of HCV genotypes less responsive to traditional therapy with pegylated interferon α and ribavirin [Liang and Ghany, 2013, 2014; Pawlotsky, 2014; Ferenci, 2015]. Most of these patients can nowadays be rapidly cured using sofosbuvir combined with simeprevir, daclatasvir or ledipasvir, or the combination of paritaprevir with ritonavir, ombitasvir and with or without dasabuvir [Ferenci, 2015]. Addition of ribavirin may further abbreviate the duration of treatment. Yet, the efficacy and safety of these drugs is not studied in patients with lymphoma. Moreover, their extremely high cost limits their broad application.

Currently, the management of both the HCV infection and the NHL depends on the lymphoma aggressiveness and remains related to the lymphoma histological type.

HCV-associated indolent NHL

The use of antiviral therapy to treat HCV-associated lymphomas is based upon the already mentioned observation that splenic MZL regresses following HCV eradication with pegylated interferon α and ribavirin [Hermine et al. 2002]. In addition to MZL, for which the achievement of a virologic response can lead to a lymphoma response in up to 75% of the cases, hematological responses to antiviral therapy have also been shown in other indolent NHLs, such as follicular lymphoma and LPL. For these histologic subtypes, most data came initially from studies with a very small number of patients (Table 3). A recent, larger, retrospective survey of 134 patients, confirmed that antiviral therapy is effective in MZL as well as in other indolent entities and that HCV-RNA clearance is needed to attain lymphoma response [Arcaini et al. 2014]. In HCV-positive patients with MZL who do not need immediate antilymphoma therapy, up-to-date evidence suggests that antiviral treatment with pegylated interferon α and ribavirin should be considered as first-line treatment [Dreyling et al. 2013; Arcaini et al. 2014]. The same approach may also be applied to the asymptomatic patient and those with HCV infection with LPLs and follicular lymphomas, but the evidence from the published studies is less strong and further investigation is needed.

For patients who are intolerant or resistant to interferon, new antiviral agents are now available for the treatment of HCV infection. These agents have a better tolerability and a higher efficacy than the interferon-based regimens. A very recent case report [Rossotti et al. 2015] describes the activity of 16-week interferon-free treatment combining the NS3/NS4 inhibitor faldaprevir (120 mg daily) with the non-nucleoside NS5B inhibitor deleobuvir (600 mg twice a day) and with ribavirin (1000 mg daily) in a 42-year-old male patient with SMZL with circulating villous lymphocytes and concomitant HCV genotype 1 infection. Following a very fast virologic response (HCV-RNA undetectable from week 4) without relevant side effects, a reduction of the lymphocyte counts was observed and the spleen size returned to normal. At a follow-up of 1 year the patient maintained a SVR, showing only a persistence of clonal circulating lymphocytes by flow-cytometry analysis without splenomegaly and other signs or symptoms of lymphoma [Rossotti et al. 2015]. Although these results are promising, further research is required for HCV genotypes resistant to conventional treatment.

HCV-associated diffuse large B-cell lymphomas

DLBCL is the most frequent aggressive lymphoma in patients with HCV. The management of HCV-associated DLBCLs remains based on immunochemoterapy with antracycline containing regimens in combination with rituximab as in their HCV-negative counterpart, and there are no specific recommendations for this subset of patients [Pellicelli et al. 2011; Borchardt and Torres, 2014]. It is unclear whether patients with HCV-positive DLBCL still have a significantly worse outcome than those with HCV-negative disease [Ennishi et al. 2010; Visco and Finotto, 2014]. Before the introduction of rituximab in lymphoma treatment, a study reported a 2-year OS of 56% in a group of patients with HCV-positive DLBCL treated with anthracycline-containing intensive chemotherapy regimens compared with a 2-year OS of 80% in HCV-negative cases [Besson et al. 2006]. A later report, however, showed a 3-year OS of 71% in patients with HCV-positive disease treated with CHOP chemotherapy (cyclophosphamide, vincristine, prednisone, doxorubicin) plus rituximab (R-CHOP) [Merli et al. 2014]. This result was confirmed by the 3-year OS of 73% observed in a very recent prospective study including 45 patients with HCV infection and DLBCL, all treated frontline with rituximab plus chemotherapy [Michot et al. 2015]. This outcome is similar to that for HCV-negative cases of DLBCL treated with R-CHOP (Table 5).

Table 5.

Outcome of HCV-associated diffuse large B-cell lymphoma according to the type of treatment.

Reference	Study design	No. of patients	Treatment	Outcomes
Tomita et al. [2003]*	retrospective	25*	CHOP/CHOP-like	5-year OS 46%	-
Besson et al. [2006]	retrospective	26	intensive chemotherapy	3-year OS 56%	3-year EFS 53%
Visco et al. [2006]	retrospective	156	R-CHOP(35)/CHOP or CHOP-like	3-year OS 72%	3-year PFS 51%
Park et al. [2008]	retrospective	32	R-CHOP(11)/CHOP	5-year OS 59%	55% (5 year)
Ennishi et al. [2010]	retrospective	131	R-CHOP	3-year OS 75%	3-year PFS 69%
Merli et al. [2014]	retrospective	535	R-CHOP(255)/CHOP	3-year OS 71% (71% R-CHOP)	3-year PFS 55% (58% R-CHOP)
Michot et al. [2015]§	prospective	27$	R-CHOP/R-ACVBP	2-year OS 81%	2-year PFS 80%
Chen et al. [2015]	retrospective		R-CHOP/R-CHOP-like	3-year OS 55%	3-year PFS 52%

^*The working formulation was used to define histology by Tomita, who reports 20 large cell lymphomas and other aggressive histological types (including four cases of T-cell non-Hodgkin’s lymphoma).

^$The analysis of survival was restricted only to the patients prospectively enrolled in the study by Michot.

CHOP, cyclophosphamide, vincristine, prednisone, doxorubicin; EFS, event-free survival; n.r., not reported; OS, overall survival; PFS, progression-free survival; R-ACVBP, doxorubicin, cyclophosphamide, vindesine, bleomycin, prednisone plus rituximab; R-CHOP, cyclophosphamide, vincristine, prednisone, doxorubicin plus rituximab; RFS, relapse-free survival.

Some reports show that patients with HCV receiving rituximab have significantly more hepatic toxicity, with grade 3–4 liver complications ranging between 10% and 28% [Arcaini et al. 2010; Salah-Eldin et al. 2014; Visco and Finotto, 2014]. Liver cirrhosis and impaired liver function are, indeed, associated with early toxicity. In fact, more frequently these patients require treatment delay compared with HCV-negative cases [Mahale et al. 2012]. Nonetheless, differently from hepatiitis B virus (HBV), the immunosuppression induced by rituximab does not appear to trigger a HCV reactivation. Importantly, the HCV-RNA pretreatment level did not seem to correlate with liver damage after immunochemotherapy. The viral load is usually observed to fall when transaminases rise, supporting the hypothesis that the hepatic damage may be not directly due to an increased viral replication, but it can be related to an immune-mediated reaction against hepatocytes [Ennishi et al. 2010].

In a recent large Italian series by Fondazione Italiana Linfomi (FIL, Italian Lymphoma Foundation), neither the HCV infection per se nor the development of hepatic toxicity seem to affect the survival and HCV-positive cases have a benefit from rituximab similarly to HCV-negative cases [Arcaini et al. 2010; Merli et al. 2014; Salah-Eldin et al. 2014; Visco and Finotto, 2014; Chen et al. 2015]. The FIL study also produced a specific prognostic score for this subset of patients. The study retrospectively enrolled 535 consecutive patients with HCV-associated DLBCL, treated with anthracycline-based regimens (with rituximab in 255 cases). This trial reported a severe hepatotoxicity rate of 14%, which did not increase in the patients receiving rituximab. At multivariate analysis, performance status according to ECOG (Eastern Cooperative Oncology Group) scale of 2 or over, serum albumin below 3.5 g/dl and HCV-RNA viral load over 1000 KIU/ml displayed an adverse prognostic significance. The combination of these three factors in a new ‘HCV prognostic score’ allowed one to distinguish between three risk groups with significantly different outcomes (low = zero; intermediate = one; high risk = at least two factors). The estimated OS at 3 years was 92% for the low-risk group, 77% for the intermediate-risk group and 39% for the high-risk group. The 3-year progression-free survival rates were 81%, 61% and 19% in low-, intermediate- and high-risk groups, respectively. This score proved to be better than the International Prognostic Index in the subcategory of HCV-positive DLBCL and also maintained its prognostic value in the subgroups of patients treated with rituximab [Merli et al. 2014].

Another unsolved question concerns the role of antiviral therapy in the treatment of HCV-positive DLBCL. In contrast to HBV, for which prophylactic anti-hepatitis B virus therapy during chemotherapy is the standard of care, concurrent administration of immunochemotherapy with interferon-based anti-HCV treatments significantly increases hematologic and liver toxicity [Arcaini et al. 2010]. An Italian pilot study, published only as a meeting abstract, reported an excessive hematological toxicity with the concomitant administration of R-CHOP chemotherapy plus antiviral treatment with pegylated interferon and ribavirin, while a sequential treatment with R-CHOP (six to eight cycles) followed by a 3-month period of maintenance therapy with pegylated interferon plus ribavirin was well tolerated and resulted in a high rate of complete virus clearance in 12 treated patients [Musto et al. 2005]. Other reports show a possible benefit from the addition of anti-HCV therapy after chemotherapy in responding patients. Remarkably, in the study by Michot and colleagues, more than one third of patients with HCV-positive DLBCL received antiviral therapy (mainly pegylated interferon and ribavirin) during the follow up and a trend associating antiviral treatments and longer OS was found [Michot et al. 2015]. In a previous study, La Mura suggested that the hematological response may be reinforced by the addition of antivirals after chemotherapy and may result in prolonged disease-free survival [La Mura et al. 2008]. Nevertheless, the role of the antiviral therapy in patients in remission after cytotoxic chemotherapy has yet to be more thoroughly studied, particularly in the light of the changing paradigm of HCV treatment.

Are the other hepatitis viruses linked with lymphoma?

Individuals infected with HCV are often coinfected by GB virus C (GBV-C, formerly known as hepatitis G virus) and HBV. These viruses were investigated to elucidate their possible association with lymphomas. GBV-C is a lymphotropic, single-stranded RNA virus of the Flaviviridae family, strictly related to HCV. GBV-C replicates in B and T lymphocytes. Some studies, mainly retrospective, suggest an association between GBV-C infection and lymphomas, but do not provide conclusive evidence of a causative role [Krajden et al. 2010; Nicolosi Guidicelli et al. 2012]. There are only limited data about the involvement of HBV in lymphoma development [Marcucci et al. 2012]. Although a meta-analysis of 11 case–control studies in NHL indicates a high prevalence of HBV carrier state among patients with lymphoma (OR for HBV infection 2.6) [Nath et al. 2010], a causal involvement of HBV remains unproven [Bouvard et al. 2009; Marcucci and Mele, 2011; Marcucci et al. 2012].

Conclusion

The combination of pegylated interferon and ribavirin has been until present the only available treatment for patients with HCV infection. Although traditional therapy (pegylated interferon and ribavirin) remains a therapeutic option, the innovative direct-acting antiviral agents are producing faster and improved responses with fewer adverse effects. The combination of new anti-HCV agents and the duration of treatment are based on HCV genotype, patient treatment status and patient risk factors. These drugs allow to obtain a SVR for patients who are unable to tolerate or who are resistant to pegylated interferon [Liang and Ghany, 2013, 2014; Pawlotsky, 2014; Ferenci, 2015]. These innovative treatments are becoming part of the therapeutic armamentarium of HCV infection. In some reports, the use of new drugs in some extrahepatic manifestations of HCV has already been demonstrated [De Nicola et al. 2014; Saadoun et al. 2015] and few data are emerging on NHL treatment [Rossotti et al. 2015], for which their short-term duration may be particularly advantageous. However, whether the safety profile of novel regimens will allow the administration of antiviral treatment concomitantly with chemotherapy is still unknown. At present, it seems advisable to choose treatment modalities in collaboration with an expert hepatologist/gastroenterologist. In our opinion, an appropriate anti-HCV treatment should be offered to all patients affected by indolent B-NHL not fulfilling the criteria for immediate start of chemoimmunotherapy. Patients with other indolent lymphomas and those affected by DLBCL should undergo chemoimmunotherapy according to current guidelines for lymphoma. In HCV-associated DLBCL, rituximab-containing chemotherapy is not associated with a higher frequency of severe hepatic toxicity and R-CHOP should be considered the standard treatment. Baseline elevated GPT is predictive of severe hepatic toxicity; close monitoring of liver function is highly recommended in these cases. HCV-RNA load is prognostic at baseline and increases with R-CHOP, but it is not predictive of severe hepatic toxicity; it should be evaluated at baseline, but close monitoring of HCV-RNA is not recommended.

For patients infected with HCV, the potential for a cure seems realistic and the use of interferon-free antiviral therapy after obtaining complete response may now be considered standard clinical practice in HCV-associated NHL. Nevertheless, new investigations are needed to properly address the appropriate role and timing of these drugs in lymphoma treatment.