miRNA-26b Downregulation in Peripheral Blood Mononuclear Cells of Patients with Hepatitis C Associated Lymphomas is Restored by Successful Interferon-Free Antiviral Therapy

Abstract

Background

Patients with chronic HCV infection are at increased risk of developing B-cell non-Hodgkin lymphoma (B-NHL). Regression of HCV-associated B-NHL (HCV-NHL) can be achieved through HCV eradication using interferon (IFN). However, only about two-thirds of patients with sustained virological response (SVR) also had a consecutive lymphoma response. miRNA-26b is associated with HCV-NHL response to antiviral therapy. Recent data suggest that IFN-free direct-acting antiviral (DAA) regimens also have anti-lymphoma activity in this patient population.

Methods

We report four patients with HCV-NHL who were treated with different IFN-free DAA regimens as oncological monotherapy in our centre between 2015 and 2016. We analysed the virological and lymphoproliferative disease response. Moreover, we analysed miRNA-26b expression in peripheral blood mononuclear cells at different time points during antiviral therapy for all included patients as well as for a total of 10 controls with (n=5) and without (n=5) chronic HCV infection.

Results

All patients had marginal zone lymphoma subtype and received different DAA regimens for 12–24 weeks. All four patients achieved SVR, but only three patients also had lymphoma response (one complete response, two partial responses). One patient showed progression to a high-grade lymphoma subtype after SVR. miRNA-26b expression was generally decreased in patients with HCV-NHL. Moreover, miRNA-26b expression was restored in those HCV-NHL patients with lymphoma response after 6 months (P=0.009).

Conclusions

We have demonstrated that IFN-free DAA treatment of HCV can improve or even cure NHL. miRNA-26b-levels could be a potentially useful biomarker to predict lymphoma response in HCV-NHL patients.

Introduction

Worldwide, an estimated 71 million people are chronically infected with HCV, a hepatotropic and potentially lymphotropic virus [1–4]. In the last two decades, evidence from epidemiological studies, biological insights, and especially therapeutic approaches provided strong support for an association between HCV and B-cell non-Hodgkin's lymphomas (NHL) [5]. HCV-related B-cell proliferation represents an important model of virus-driven autoimmune/neoplastic disorder leading to mixed cryoglobulinaemia (MC) and/or NHL [6]. However, the most compelling argument for a causal relationship between HCV and B-NHL is made by interventional studies demonstrating that antiviral therapy using interferon (IFN) or pegylated IFN with and without addition of ribavirin (RBV) frequently leads to regression of HCV-associated lymphomas (HCV-NHL) [7,8].

Piluso et al. [9] showed downregulation of the miRNA miR-26b in peripheral blood mononuclear cells (PBMCs) of patients with HCV-NHL compared with controls. Downregulation of the tumour suppressive microRNA-26b (miR-26b) was first connected to HCV-NHL in microdissected primary tissue of HCV-associated splenic marginal zone lymphoma [10]. Consecutively, the same miRNA was found downregulated in PBMCs of patients with MC and HCV-NHL, also demonstrating restoration of miR-26b expression in MC patients who achieved a sustained virological response (SVR) after treatment with IFN and RBV [11]. These findings indicate that miR-26b could be involved in the mechanism of HCV-driven lymphomagenesis and could potentially serve as a valuable biomarker to monitor lymphoma response during antiviral therapy.

Antiviral therapy of HCV underwent a revolution within the last several years. After almost 25 years of IFN-based therapies, highly effective direct-acting antiviral (DAA) drugs are now standard of care for treatment of chronic HCV infection. With these DAAs, viral eradication can now be achieved in approximately 95% of all patients across different genotypes and fibrosis stages [12].

Data on IFN-free treatment of HCV-NHL is still scarce and mainly based on case reports [13–17]. Now the first multicentre retrospective series of HCV-NHL treated with IFN-free antiviral therapy shows promising results but also reports on patients with SVR who did not show favourable oncological response [18–20].

The aim of this study was to investigate the clinical course of HCV-NHL patients from our centre treated with IFN-free DAA regimen. In addition, we evaluated miR-26b expression as a biomarker for lymphoma response in PBMC of HCV-NHL patients undergoing DAA therapies.

Methods

Study design and patients

The current study is a retrospective analysis, which was approved by the local Ethics Committee of the University Hospital Frankfurt, Germany (Ethik-Kommission des Fachbereichs Medizin der Goethe-Universität). Written informed consent was obtained from all patients. Clinical data were abstracted from electronic records. Patients with HCV-NHL (n=4) and matched individuals with chronic HCV without lymphoma (n=5) were recruited from our outpatient clinic specialized in HCV treatment. Additionally, PBMCs from healthy volunteers (n=5) were acquired from blood donors.

Clinical and virological end points

The primary oncological end point was overall tumour response. Response evaluation was based on Lugano classification criteria for lymphomas [12]. Bone marrow biopsy as well as CT scans were performed before start of antiviral therapy and repeated at the end and 6 months after treatment. SVR was defined as undetectable HCV viral load with a sensitive real-time PCR assay 12 weeks after completion of therapy. HCV viral load was measured using the COBAS AmpliPrep/COBAS TaqMan HCV Test, v2.0 (Roche Diagnostics, Mannheim, Germany). HCV genotypes were determined by the VERSANT HCV Genotype 2.0 assay (Siemens Healthcare Diagnostics, Erlangen, Germany).

miR-26b expression analysis

PBMCs were isolated from fresh anticoagulated blood by gradient precipitation on Lymphoprep (Axis-Shield PoC AS, Oslo, Norway) according to the manufacturer's instructions. After the second wash, the cells were counted and stored at −80°C. RNA extraction was performed using Trizol reagent (Invitrogen, Carlsbad, CA, USA) from 5x106 PBMCs according to the manufacturer's instructions. Reverse transcription was done using the TaqMan MicroRNA RT kit (Applied Biosystems, Foster City, CA, USA) and 100 ng of total RNA. Expression levels of human miR-let-7g and miR-26b were evaluated by real-time PCR using specific TaqMan MicroRNA Assays (Applied Biosystems), according to the manufacturer's instructions.

Relative expression levels of the different miRNAs were evaluated with the delta-Ct method, using miR-let-7d as internal control to normalize miRNA expression levels, as previously described [11].

Statistical analysis

Statistical analyses and illustrations were performed using GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA). Student's two-sample t-test for either paired or unpaired samples was used to compare miRNA expression values. All P-values were two-tailed and were considered statistically significant for levels ≤0.05.

Results

Cases

Case 1

The 49-year-old male was diagnosed with HCV GT3a infection in 2005 following intravenous drug use. Viral load was 541,000 IE/ml. Pretreatment transient elastography (FibroScan, Echosens, Paris, France) showed a liver stiffness of 5.9 KPa, corresponding to F0-F1 fibrosis. In 2015, the patient presented to his primary care physician with abdominal pain in the left upper quadrant. There were no B symptoms present. On physical examination, the treating physician diagnosed splenomegaly, and the blood count showed an increased leukocyte/lymphocyte cell count (60,000/mm³/42,000/mm³). The patient was referred to a specialized centre for haematology/oncology. The high lymphocyte count was confirmed showing a monoclonal B-cell immune phenotype with the morphology of villous lymphocytes. Cryoglobulinaemia was ruled out. CT scan showed splenomegaly (longitudinal diameter of 27 cm) and an enlargement of paraaortic, paracaval and mesenteric as well as supraclavicular and mediastinal lymph nodes. Bone marrow biopsy showed lymphoid nodular infiltration of more than 40%. The final diagnosis was stage IV leukaemic splenic marginal zone lymphoma. As the patient was asymptomatic at presentation at the haematology centre, he was then referred to our centre for antiviral therapy. The patient received a combination of sofosbuvir 400 mg once daily and daclatasvir 60 mg once daily for 12 weeks. HCV RNA was undetectable from week 2 and remained undetectable at 12 and 24 weeks after the end of treatment. At week 4 of treatment, the patient presented with abdominal pain due to splenomegaly and symptom control was achieved with use of non-steroidal anti-inflammatory drugs in combination with transdermal fentanyl. At week 8 of treatment, the pain subsided gradually with no need for any pain medication after week 12. The blood lymphocyte count increased initially from baseline to week 2 (44,000 to 54,000/mm³), but steadily decreased (39,000 at follow-up week 12) to nearly normal values 1 year after antiviral therapy, with the B-cell clone still detectable. Moreover, bone marrow biopsy 3 months after initiation of treatment showed residual but reduced lymphoma infiltration (25%). The CT scan 6 months after antiviral therapy showed decrease of all lymph node diameters by more than 50% and a decrease in spleen size to nearly normal size. The response was classified as partial response.

Case 2

A 56-year-old male, who was diagnosed with HCV genotype-1b infection in 2013 following intravenous drug use in the 1990s. Viral load was 943,000 IE/ml. Pretreatment transient elastography showed a liver stiffness of 15.1 KPa, corresponding to F4 fibrosis score and liver morphology in the abdominal ultrasound was consistent with compensated cirrhosis. At the time of HCV diagnosis, the patient was also diagnosed with a low-grade B-NHL. The patient suffered from occasional fever episodes and night sweats. CT scan showed splenomegaly (longitudinal diameter of 20.7 cm) and paraaortic, mesenteric as well as mediastinal lymph node enlargement. Bone marrow biopsy showed monoclonal lymphoid infiltration of more than 50% with positivity for CD79, CD3, CD20 and CD34. Peripheral blood count showed mild anaemia but otherwise normal results. The final diagnosis was stage IVB low-grade marginal zone lymphoma. The patient was followed-up with staging CT scans every 3 to 6 months and showed a stable course. In June 2015, after interdisciplinary discussion, the patient received antiviral therapy with ombitasvir/paritaprevir (12.5 mg/75 mg two tablets once daily) and dasabuvir (250 mg twice daily) for 12 weeks. HCV RNA was undetectable from week 2 and remained undetectable at 12 and 24 weeks after the end of treatment. The patient was symptom free 8 weeks after initiation of antiviral therapy. The staging CT scan 6 months after the start of treatment showed a minimal decrease in spleen size (21 to 18 cm) but several mediastinal lymph nodes were noted which had grown in size. Biopsies of the mediastinal lymph nodes were taken using endobronchial ultrasound. Histology showed a high grade B-NHL with high proliferation (Ki67 index 80%) consistent with diffuse large B-cell lymphoma (DLBCL). Therefore, response to antiviral treatment was classified as progressive disease with transformation into a high-grade lymphoma. Consecutively the patient received immuno-chemotherapy (R-CHOP) with partial response after three cycles and is currently planned to receive three additional cycles.

Case 3

The 78-year-old female was diagnosed with splenic marginal zone lymphoma in 2015 from an ultrasound guided breast biopsy, after the patient noted a lump in the right breast. Subsequent work-up revealed chronic infection with HCV genotype-2, while no typical risk factors for contracting HCV infection were noted. Viral load was 1,090,000 IE/ml. Transient elastography was not available but conventional ultrasound did not show signs of cirrhosis. The patient reported weight loss of about 4 kilograms but no fevers or night sweats. Staging CT scans revealed that the breast and mediastinal lymph nodes were the only localizations of the lymphoma. There was no bone marrow involvement. The peripheral blood count showed normal results. The final diagnosis was stage I VA low-grade splenic marginal zone lymphoma. The patient was then referred to our centre for HCV treatment. In line with guideline recommendations at the time, the patient received sofosbuvir 400 mg once daily plus weight-based RBV for 12 weeks. HCV RNA was undetectable from week 4 and remained undetectable at 12 and 24 weeks after the end of treatment. The staging CT scan 6 months after the start of therapy showed complete response with minimal scarring tissue at the breast biopsy site, which did not show residual lymphoma after re-biopsy.

Case 4

The 60-year-old male patient was diagnosed with chronic HCV genotype-1b infection following diagnosis of splenic marginal zone lymphoma in 2011. He was initially treated with splenectomy because of symptomatic splenomegaly. After splenectomy, the lymphoma persisted in the bone marrow with a tumour cell infiltration of 40–50%. There were no other disease manifestations and the patient had no B symptoms. The final diagnosis was stage IV splenic marginal zone lymphoma. The patient was then referred to our centre for HCV treatment in 2014. Pretreatment elastography showed a liver stiffness of 4.8 KPa, corresponding to F0-F1 fibrosis. The patient received a combination of sofosbuvir 400 mg once daily and daclatasvir 60 mg once daily plus weight-based RBV for 24 weeks as a participant of a daclatasvir compassionate use programme [21]. HCV RNA was undetectable from week 2 and remained undetectable at 12 and 24 weeks after the end of treatment. There was no evidence of lymphoma manifestation in follow-up CT scans and follow-up bone marrow biopsy 6 months after the start of HCV therapy showed a decrease in lymphoma infiltration to 10%. The response was classified as partial response.

miR-26b expression in PBMCs

For all patients, miR-26b expression was analysed from PBMCs at the start of antiviral therapy, at the end of treatment and 6 months thereafter (clinical patient data are summarized in Table 1).

Table 1

Clinical data of HCV-NHL patients

	Patient 1	Patient 2	Patient 3	Patient 4
Gender	Male	Male	Female	Male
Age at start of HCV therapy	49	5G	78	GO
Lymphoma subtype	SMZL	MZL	MZL	SMZL
Ann Arbor Stage	IVA	IVB	IVA	IVA
Baseline HCV RNA, log₁₀ IU/ml	5.7	6.0	7.0	7.0
HCV genotype	3a	1b	2	1b
Cirrhosis	No	Yes	No	No
Previous chemotherapy	No	No	No	No
Previous antiviral therapy	No	No	No	No
Antiviral therapy	S0F/DCV	3D/RBV	SOF/RBV	SOF/DCV/RBV
Treatment duration, weeks	12	12	12	24
Major adverse events associated with antiviral therapy	No	No	No	No
SVR12/24	Yes	Yes	Yes	Yes
NHL response	PR	PD	CR	PR
Duration of response, months	11	0	15	8

CR, complete response; DCV, daclatasvir; MZL, marginal zone lymphoma; NHL, non-Hodgkin lymphoma; PD, progressive disease; PR, partial response; RBV, ribavirin; SMZL, splenic marginal zone lymphoma; SOF, sofosbuvir; SVR, sustained virological response; 3D, ombitasvir/paritaprevir + dasabuvir.

To evaluate baseline miR-26b expression in HCV-NHL cases, values of matched healthy volunteers and patients with chronic HCV infection without NHL were added as controls (demographic data are shown in Additional file 1). There was no difference in miR-26b expression between healthy volunteers and patients with chronic HCV infection without NHL. HCV-NHL patients showed a trend to lower miR-26b expression compared with controls (Figure 1; P=0.06). Moreover, when the one patient whose lymphoma did not respond to HCV therapy was removed from the analysis, the baseline miR-26b level was significantly lower in the HCV-NHL group compared with the HCV-negative and HCV-positive controls, respectively (P=0.001 and P=0.01; Figure 2). When comparing miR-26b expression between HCV-NHL at baseline and end-of-treatment or 6 months after treatment, respectively, a restoration of miR-26b levels could be observed in all patients but one (patient 2 with no lymphoma response to antiviral therapy). Mean miR-26b expression in this comparison was not statistically significant. Again, however, when case 2 was removed from the analysis, we observed a significant increase in miR-26b following successful HCV eradication over time. When analysing the individual HCV-NHL cases, we observed an increase in miR-26b expression after HCV treatment for all patients who also showed oncological response, reaching levels of the control population. Case 2, whose lymphoma progressed and transformed into DLBCL had a higher baseline level of miR-26b and did not show an increase in expression after HCV was cured.

Figure 1

Relative expression of miR-26b for all patients and controls

Figure 2

Relative expression of miR-26b with data from patient 2 (with no lymphoma response to antiviral therapy) removed

Discussion

We present a case series of four patients with HCV-NHL who received IFN-free DAA therapy with the intention to treat the underlying malignant disease. We found evidence that the expression of the tumour-suppressive miRNA, miR-26b, correlates inversely with successful lymphoma treatment.

So far, few case reports have been published reporting on IFN-free DAA treatment of HCV-NHL patients [14–16]. In these publications successful HCV eradication in patients with marginal zone lymphoma or follicular lymphoma led to a complete and lasting oncological response. Furthermore, Carrier et al. [13] reported on five patients with HCV-NHL treated with DAAs, out of which three received exclusive anti-HCV treatment. All treated patients in these publications reached complete virological as well as oncological response. Recently, multicentre retrospective series of HCV-NHL treated with IFN-free antiviral therapy showed promising results but also reported on patients with SVR who did not show favourable oncological response [18–20].

Our recent meta-analysis on HCV-NHL treated with IFN-based regimens showed that overall lymphoma response rate (either complete or partial response) was 73% (95% CI, 67, 78%) [22]. Across the different included studies there was a strong association between SVR and lymphoma response (overall response, 83%; 95% CI, 76, 88%) compared with a failure in achieving SVR (overall response, 53%; 95% CI, 39, 67%; P=0.0002). Thus, almost one-fifth of patients with IFN-induced SVR did not achieve lymphoma response. Although HCV cure rates now exceed 90% across different patient populations, DAA therapy may not be able to improve the oncological outcome in all HCV-NHL patients. One possible explanation may be that some patients suffer from lymphomas that are coincident and primarily independent of HCV or, being primarily driven by HCV, have become independent of the viral stimulus over time. Patients with low-grade HCV-NHL and lymphoma progression or even transformation into high-malignant lymphoma types can be treated with immune-chemotherapy in succession, as was done in patient number 4 in our study. However, delaying oncological treatment could worsen the prognosis of the respective patient and concomitant antiviral therapy and chemotherapy has been proposed [23]. However, to spare patients from potentially toxic chemotherapy, a biomarker that can predict oncological outcomes after HCV eradication at an early time point would be desirable. In the past, miR-26b, measured in lymphoma tissue as well as patient blood, has been shown to be in a close relationship with HCV-NHL [9,10,24]. Moreover, a restoration of this tumour-suppressive miRNA to normal levels could be observed in patients successfully cured of mixed cryoglobulinaemia by anti-HCV therapy [11]. In the current study, we observed an inverse correlation of HCV viral load as well as lymphoma burden with miR-26b expression in PBMCs of HCV-NHL patients. More importantly, the one patient who did not show lymphoma response following HCV cure did not have a significantly decreased miR-26b level before therapy and did also not show any increase in miR-26b expression after achieving SVR.

In summary, this case series demonstrates that successful HCV eradication following treatment of HCV with novel DAAs can lead to lymphoma regression or even cure. However, we also show that one lymphoma patient showed progressive disease despite achieving SVR. miR-26b-levels could be a potential biomarker to predict lymphoma response in HCV-NHL patients. Given the relatively low prevalence of HCV-NHL, international collaborative studies are required to address open questions with regard to the potential use of miR-26b and other biomarkers for tumour outcome prediction. Larger, and preferably prospective studies could help to evaluate whether the lack of miR-26b downregulation before the start of antiviral therapy expresses an independence of the lymphoma from the viral stimulus that could hint to a lack of oncological efficacy of antiviral therapy. Moreover, it should be further evaluated whether a lack in restoration or increase in miR-26b expression during and after antiviral therapy might predict worse oncological outcomes.

Footnotes

Acknowledgements

This work was supported by the Deutsche Forschungsgemeinschaft (PE2152/2-1).

SZ: consultancy for AbbVie, Gilead, Janssen, Merck/MSD. TMW: consultancies for AbbVie, Bristol-Myers Squibb, Gilead. JV: lectures/consultancies for AbbVie, Gilead, Merck/MSD. The other authors have no conflict of interest to declare in context of this manuscript.

Additional file 1: A table showing clinical data of matched control patients and healthy control subjects can be found at

References

WHO. Hepatitis C Fact sheet No164. (Updated October 2017. Accessed 1 December 2018.) Available from www.who.int/mediacentre/factsheets/fs164/en/

Durand

, Di Liberto

, Colman

Occult infection of peripheral B cells by hepatitis C variants which have low translational efficiency in cultured hepatocytes. Gut 2010; 59: 934–942

Sung

V.M.

, Shimodaira

, Doughty

A.L.

Establishment of B-cell lymphoma cell lines persistently infected with hepatitis C virus in vivo and in vitro: the apoptotic effects of virus infection. J Virol 2003; 77: 2134–2146

Polaris Observatory HCV Collaborators. Global prevalence and genotype distribution of hepatitis C virus infection in 2015: a modelling study. Lancet Gastroenterol Hepatol 2017; 2: 161–176

Peveling-Oberhag

, Arcaini

, Hansmann

M.L.

, Zeuzem

Hepatitis C-associated B-cell non-Hodgkin lymphomas. Epidemiology, molecular signature and clinical management. J Hepatol 2013; 59: 169–177

Ferri

, Sebastiani

, Giuggioli

Hepatitis C virus syndrome: a constellation of organ- and non-organ specific autoimmune disorders, B-cell non-Hodgkin's lymphoma, and cancer. World J Hepatol 2015; 7: 327–343

Arcaini

, Merli

, Volpetti

, Rattotti

, Gotti

, Zaja

Indolent B-cell lymphomas associated with HCV infection: clinical and virological features and role of antiviral therapy. Clin Dev Immunol 2012; 2012: 638185

Hermine

, Lefrere

, Bronowicki

J.P.

Regression of splenic lymphoma with villous lymphocytes after treatment of hepatitis C virus infection. N Engl J Med 2002; 347: 89–94

Piluso

, Gragnani

, Fognani

Deregulation of microRNA expression in peripheral blood mononuclear cells from patients with HCV-related malignancies. Hepatol Int 2015; 9: 586–593

10.

Peveling-Oberhag

, Crisman

, Schmidt

Dysregulation of global microRNA expression in splenic marginal zone lymphoma and influence of chronic hepatitis C virus infection. Leukemia 2012; 26: 1654–1662

11.

Fognani

, Giannini

, Piluso

Role of microRNA profile modifications in hepatitis C virus-related mixed cryoglobulinemia. PLoS One 2013; 8: e62965

12.

Pawlotsky

J.M.

, Feld

J.J.

, Zeuzem

, Hoofnagle

J.H.

From non-A, non-B hepatitis to hepatitis C virus cure. J Hepatol 2015; 62: S87–S99

13.

Carrier

, Jaccard

, Jacques

HCV-associated B-cell non-Hodgkin lymphomas and new direct antiviral agents. Liver Int 2015; 35: 2222–2227

14.

Rossotti

, Travi

, Pazzi

, Baiguera

, Morra

, Puoti

Rapid clearance of HCV-related splenic marginal zone lymphoma under an interferon-free, NS3/NS4A inhibitor-based treatment. A case report. J Hepatol 2015; 62: 234–237

15.

Sultanik

, Klotz

, Brault

, Pol

, Mallet

Regression of an HCV-associated disseminated marginal zone lymphoma under IFN-free antiviral treatment. Blood 2015; 125: 2446–2447

16.

Hattori

, Ikeda

, Nakano

Curative effects for B-cell lymphoma accomplished by direct-acting antiviral agents of hepatitis C. Open Forum Infect Dis 2017; 4: ofx057

17.

Galati

, Rampa

, Vespasiani-Gentilucci

Hepatitis C and double-hit B cell lymphoma successfully treated by antiviral therapy. World J Hepatol 2016; 8: 1244–1250

18.

Arcaini

, Besson

, Frigeni

Interferon-free antiviral treatment in B-cell lymphoproliferative disorders associated with hepatitis C virus infection. Blood 2016; 128: 2527–2532

19.

Alric

, Besson

, Lapidus

Antiviral treatment of HCV-infected patients with B-cell non-Hodgkin lymphoma: ANRS HC-13 Lympho-C Study. PLoS One 2016; 11: e0162965

20.

Michot

J.M.

, Canioni

, Driss

Antiviral therapy is associated with a better survival in patients with hepatitis C virus and B-cell non-Hodgkin lymphomas, ANRS HC-13 lympho-C study. Am J Hematol 2015; 90: 197–203

21.

Welzel

T.M.

, Petersen

, Herzer

Daclatasvir plus sofosbuvir, with or without ribavirin, achieved high sustained virological response rates in patients with HCV infection and advanced liver disease in a real-world cohort. Gut 2016; 65: 1861–1870

22.

Peveling-Oberhag

, Arcaini

, Bankov

, Zeuzem

, Herrmann

The anti-lymphoma activity of antiviral therapy in HCV-associated B-cell non-Hodgkin lymphomas: a meta-analysis. J Viral Hepat 2016; 23: 536–544

23.

Persico

, Aglitti

, Caruso

Efficacy and safety of new direct antiviral agents in hepatitis C virus-infected patients with diffuse large B-cell non-Hodgkin's lymphoma. Hepatology 2018; 67: 48–55

24.

Gragnani

, Fognani

, Piluso

, Zignego

A.L.

Hepatitis C-associated B-cell non-Hodgkin lymphomas: the emerging role of miRNA-26b. J Hepatol 2013; 59: 1362–1363