Accepted Manuscript Clinical Trial Watch Clinical Trial Watch: Reports from the EASL International Liver Congress (ILC), Vienna, April 2015 Martina Gambato, Augusto Villanueva, Juan G. Abraldes, Jose Altamirano, Xavier Forns PII: DOI: Reference:
S0168-8278(15)00401-8 http://dx.doi.org/10.1016/j.jhep.2015.06.010 JHEPAT 5714
To appear in:
Journal of Hepatology
Received Date: Accepted Date:
10 June 2015 14 June 2015
Please cite this article as: Gambato, M., Villanueva, A., Abraldes, J.G., Altamirano, J., Forns, X., Clinical Trial Watch: Reports from the EASL International Liver Congress (ILC), Vienna, April 2015, Journal of Hepatology (2015), doi: http://dx.doi.org/10.1016/j.jhep.2015.06.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Clinical Trial Watch: Reports from the EASL International Liver Congress (ILC), Vienna, April 2015 Martina Gambato1, Augusto Villanueva2, Juan G. Abraldes3, Jose Altamirano4,5 and Xavier Forns1.
1
Liver Unit, Hospital Clínic, IDIBAPS and CIBEREHD, Barcelona, Spain; 2Liver Cancer
Program, Divisions of Liver Diseases and Hematology / Medical Oncology, Tisch Cancer Institute, Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA, 3Division of Gastroenterology (Liver Unit), University of Alberta, Edmonton, Alberta, Canada, 4Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain and 5 Vall d’Hebrón Institut de Recerca (VHIR). Barcelona, Spain.
Corresponding author: Xavier Forns. Liver Unit, Hospital Clínic, IDIBAPS and CIBEREHD, Barcelona, Spain. Phone 34 93 227 57 53 email: [email protected]
Electronic word count: 6039 Number of Figures: 1 Number of Tables: 2 List of Abbreviations: SOF: sofosbuvir, DCV: daclatasvir, RBV: ribavirin, CP: Child-Pugh, SVR: sustained virologial response, HVPG: hepatic venous pressure gradient, G: genotype, LDV: ledipasvir, GZR:grazoprevir, EBR:elbasvir, SMV: simeprevir, CKD: chronic kidney disease, RAVs: resistance associated variants, TACE: transarterial chemoembolization, AAV2:adeno-
associated virus 2, CSC: cancer stem cells, NSBBs: non-selective beta blockers, ALD: Alcoholic liver disease, GWA: genome-wide association, ARC: alcohol related cirrhosis, MAIT: Mucosal Associated Invariant T-cells, NAFDL: Non-alcoholic fatty liver disease
Keywords: DAA, complications of cirrhosis, NASH, hepatocellular carcinoma. Conflicts of interest: XF received unrestricted grant from Janssen and acted as advisor for Janssen, Gilead and Abbvie. Financial support: Authors contributions: MG, AV, JGA, JA and XF contributed to this manuscript by selecting and summarizing the information delivered at the ILC held in Vienna in April 2015 in their respective fields of expertise.
Summary
Viral Hepatitis: efficacy against hepatitis C is improving but resistance should not be neglected The ILC brought very exciting news in the hepatitis C field, with relevant data on new treatment combinations and important studies on already approved treatment regimens. Probably, the most expected studies were those dealing with treatment of “difficult to cure populations”, which are summarized in the following paragraphs. Patients with decompensated cirrhosis Patients with advanced liver disease, especially those with decompensated cirrhosis, have a poor prognosis and there is only limited experience with direct acting antivirals (DAAs) in this setting. Several clinical trials and studies in real life cohorts have addressed safety and efficacy issues in these patients.
The ALLY-1 study is a phase 3 trial that assessed the safety and efficacy of sofosbuvir (SOF), daclatasvir (DCV) and ribavirin (RBV) in patients with advanced cirrhosis and in post-liver transplant hepatitis C recurrence [1]. Sixty cirrhotic patients (20% Child-Pugh [CP]-A, 53% CP-B, and 27% CP-C) infected with genotypes 1 (G1) to 4 (75% G1) received 12 weeks of this oral regimen. The majority of patients had a MELD score between 10 and 15. The overall sustained virological response 12 weeks after treatment interruption (SVR12) was 83%, with excellent data in CP-A and B (92% and 94%, respectively), while efficacy dropped in patients with CP-C (56%), though the number of patients in this group was very small
(n=16). Non response was explained by relapse in 9 of the 10 virological failures. No severe adverse events were attributed to study medication. A question that remains open is if extension to 24 weeks of therapy would reduce the rate of relapse in individuals with more advanced disease.
In the SOLAR-2 study [2], G1 or G4 decompensated cirrhotic patients (n=160), both preand post-liver transplant (n=53) were randomized to receive 12 or 24 weeks of ledipasvir (LDV), SOF and RBV. More than 50% of these patients had ascites, hepatic encephalopathy, and around 25% had a MELD score >15. Overall, this regimen resulted in high SVR12 rates (85%-88%) independently of the treatment duration, both in the pre- and post-LT setting. It is important to notice that virological response was associated with improvements in MELD and CP scores from baseline, when assessed 4 weeks after treatment interruption. This was largely explained by a decrease in bilirubin and/or an increase in liver synthetic function (i.e albumin). Indeed, 35% CP-B (n=100) improved to CTP- A, and 48% of CP-C (n=54) improved to CP-B.
Despite this effect of viral clearance on liver function,
improvements in portal pressure may take longer. Indeed, the effect of viral suppression on hepatic venous pressure gradient (HVPG) was characterized in 50 cirrhotic patients (CPA: 18 and CP-B: 32) with portal hypertension ( ~ 85% HVPG > 12 mmHg). Patients were randomized to receive 48 weeks of open-label SOF and RBV; a control group received the same therapy after a 24-weeks observation period [3]. When comparing baseline and endof-treatment HVPG measurements in 37 cirrhotics with available paired assessments, HVPG decreased > 20% in only 24% of individuals. The latter indicates that most patients would remain at risk of clinical decompensation, suggesting that the remodeling of fibrous
septa and hepatic vascular changes in liver cirrhosis require a longer follow-up to improve or, that at some point, these changes are not reversible any more.
The safety and efficacy of a regimen not containing SOF to treat patients with advanced liver disease was presented at the ILC. The C-SALT trial [4] is a Phase 2/3 study which assessed a 12-week all-oral combination of the NS3 protease inhibitor grazoprevir (GZR) and the NS5A inhibitor elbasvir (EBR). The study included 30 GT 1-infected patients with CP-B (CP-7 70%) cirrhosis and 10 matched non-cirrhotic controls. Overall SVR12 was 90% (27/30), with a good safety profile.
Data on real life experience using SOF-containing regimens were also presented at the ILC. An update of SVR12 data coming from the TARGET cohort [5], which assessed the safety and efficacy of all-oral regimens in patients with a MELD score ≥ 10, confirmed the results presented at the last AASLD [6,7]. SVR12 rates were around 75% in G1 patients treated with SOF and SMV±RBV (with slightly higher efficacy in G1b), whereas SVR12 rates were significantly lower in G3 patients treated with SOF plus RBV (~ 40%). Foster et al. [8] presented an observational study conducted in 467 patients with decompensated cirrhosis infected with genotype 1 or 3 who underwent treatment with SOF plus LDV or DCV (with or without RBV). Around 65% of patients were CP-B with a mean MELD score of 12. Most G1 infected patients received SOF/LDV; most G3-infected patients received SOF/DAC (see below). SVR12 rates in G1 patients were very high, ranging from 81 to 86%. Four weeks after treatment interruption, MELD scores improved > 2 points in 42% of patients but worsened in 11%. Preliminary analysis suggested that in older individuals (age > 65) and those with more advanced disease (albumin < 35 g/dL) treatment did not result in an
improvement in liver function. Nevertheless, trying to identify those patients with decompensated cirrhosis who will benefit (or not) from viral clearance will require more studies and a longer follow-up after treatment interruption. Additonal studies including compensated and decompensated cirrhotic patients presented at the ILC are depicted in Table 1 [9-13].
G3-infected cirrhotic patients Genotype 3 has become the most difficult to treat genotype, especially in treatmentexperienced cirrhotic patients. Previous studies have already shown that in this setting, the combination of SOF and RBV, even when extended to 24 weeks, SOF/DCV (without RBV) or SOF/LDV and RBV for 12 weeks achieve only around 60%-70% of SVR rates, which is clearly suboptimal in the current therapeutic scenario [14-16]. The BOSON study [17] is a multicentric, randomized, open-label clinical trial, assessing the efficacy of three different regimens: PEG-IFN, RBV and SOF for 12 weeks, and SOF plus RBV for 24 or 16 weeks. A total of 592 patients were included: G2 treatment-experienced with cirrhosis (n=48) and G3 treatment naïve or experienced, with or without cirrhosis (n=544). For G3-infected patients the highest SVR12 rates were obtained with PEG-IFN, RBV and SOF (93%), as compared with 84% and 71% with the 24- and 16-week SOF/RBV regimens, respectively. The results favoring a PEG-IFN-based regimen were even clearer in treatment-experienced cirrhotics, with an SVR12 of 86% in the PEG-IFN arm and only 77% and 47% in the 24- and 16-week SOF/RBV arms, respectively.
Regarding IFN-free therapy, SOF/DCV plus RBV for 12 weeks appeared to be a better choice than SOF/LDV plus RBV in G3 patients included in the English Extended Access Program [8]. Indeed, in this cohort which included mainly patients with decompensated cirrhosis, SVR12 rates were 70% in the 114 individuals who underwent SOF/DCV plus RBV compared to 59% in the 61 individuals who underwent SOF/LDV plus RBV. Based on data from the French compassionate use program [18], it seems that extending SOF plus DAC to 24 weeks might increase the efficacy in patients with advanced liver disease. Finally, data from a small phase 2 study showed excellent efficacy in a small group of G3 infected patients (n=40) who underwent treatment with the combination of SOF with GZP/EBV for 12 weeks. SVR12 was achieved in 38 (95%) of the 40 patients (10 of 11 cirrhotics, 91%) [19].
Liver transplant recipients with severe hepatitis C recurrence As stated above, the SOLAR-2 study [2] assessed the safety and efficacy of SOF/LDV plus RBV (12 or 24 weeks) in 168 patients with hepatitis C recurrence after liver transplantation, (including patients with compensated and decompensated cirrhosis). In individuals with F0F3 and compensated cirrhosis SVR12 rates were 95% and 98% with 12 and 24 weeks of treatment, respectively. Interestingly, a small subanalysis of patients with biopsy-proven fibrosing cholestatic hepatitis (n=11) included in this study showed an SVR12 rate of 100% in this very sick population [20]. Overall, there were no major safety concerns regarding the use of this regimen in this population and no changes in immunosupression were necessary. As part of the ALLY-1 study [1] the efficacy of 12 weeks of DCV/SOF +RBV was evaluated in 53 infected patients with HCV recurrence (77% G1): 30% of patients had cirrhosis and 25% bridging fibrosis. SVR12 rate was 94%; 3 individuals presented a
virological relapse. Only 4 patients had to interrupt RBV due to anemia and no modification of the immunosupressive drugs were required due to the favorable drug-drug interaction profile of this combination. Coilly et al. presented data assessing SOF/DCV as part of the French multicentric real life CULPIT cohort [21]. Most of the 130 patients included in the analysis received a 24-week treatment (approximately half with RBV). More than 50% of the patients had bridging fibrosis or cirrhosis, including some with decompensated disease. G1 was predominant. The reported efficacy was very high, ranging from 96 to 100%. In patients receiving RBV anemia requiring RBV dose reductions and erithropoietin use were common. Overall, data presented at the ILC strongly supports that all-oral regimens are as effective in liver transplant recipients as in the immunocompetent population. This is particularly true for patients with less advanced disease, and emphasizes that antiviral therapy should be indicated before individuals reach advanced disease. Importantly, neither SOF/LDV nor SOF/DCV have significant interactions with tacrolimus and cyclosporine, facilitating the use of these regimens in this setting. Real life data indicate that in LT recipients who cannot tolerate RBV, a 24-week regimen with SOF/DCV achieves very high SVR rates. Although the combination of SOF/LDV for 24 weeks (without RBV) appears to be as efficacious as a 12week regimen with RBV in cirrhotic patients [22], data are not yet available in liver transplant recipients.
Patients with chronic kidney disease (CKD)
There were practically no data on interferon-free regimens in HCV-infected patients with CKD. At the ILC, data from two clinical trials were presented. In the first one, 224 G1 infected patients (naïve or treatment experienced) with CKD stage 4 (eGFR 15-29 ml/min) or 5 (eGFR < 15 ml/min) were randomized to an observational group or to receive GZV/EBV for 12 weeks [23]. Half of the patients were infected with G1a, 75% were on dialysis, most were treatment naïve and less than 10% were cirrhotic. Of the 117 patients who received medication, all but one (99%) achieved SVR12. Adverse events and laboratory abnormalities were similar between the treatment and the observational groups. In a smaller study [24], 20 G1-infected treatment naïve patients with CKD stage 4 or 5 received paritaprevir/ritonavir, ombitasvir and dasabuvir for 12 weeks (no RBV in 7 G1b; with RBV in 13 G1a). Despite the starting dose of RBV was 200 mg/d for G1a patients, 8 of 13 interrupted the drug. At the meeting presentation all 14 patients with end-of-treatment data and 10 who had reached week 4 of follow-up remained HCV-RNA undetectable. Except for anemia in those who received RBV, no major safety issues were recorded. Overall, it seems that in the near future two regimens (one pending regulatory approval) will be available and highly efficacious in patients with end-stage renal disease.
Retreatment of patients who failed previous DAA-based therapy: the relevance of resistance-associated variants (RAVs) Retreatment of patients who failed previous DAA-based regimens is an increasing problem. At the ILC, data using GZR/EBV with RBV and SOF/LDV were presented. Seventy-nine patients with chronic HCV GT1 infection, 43% with cirrhosis, who failed triple therapy with PEG-IFN/RBV combined with boceprevir, telaprevir or simeprevir, were retreated with GZR,
EBV and RBV for 12 weeks [25]. Almost half of the patients had baseline NS3 resistance associated variants (RAVs). All but 3 patients (96%) achieved SVR12. Similar results were obtained by retreating 40 HCV G1 infected Japanese patients who failed previous PI-based triple therapy, with LDV/SOF with or without RBV for 12 weeks [26]. All patients achieved SVR12, including cirrhotic patients (80% of overall population). Thus, it seems that patients who failed a treatment regimen including first generation protease inhibitors have very good prospects for cure. Treatment of patients who failed a previous IFN-free regimen containing an NS5A inhibitor seems to be more difficult. Indeed, 41 G1 infected patients who failed 8 or 12 weeks of LDV/SOF-based therapy, underwent the same treatment regimen for a longer duration (24 weeks)[26]. Forty-six percent of them were cirrhotic. The overall response rate was 71%. In patients who underwent a 12-week course of SOF/LDV the presence of NS5A RAVs at baseline was more frequent (100%) and the efficacy was lower (SVR12 46%) than in patients who were treated for 8 weeks (NS5A RAVs 63%, SVR12 80%). In a phase 2a clinical trial, patients with early stage (F0-F2) liver fibrosis, with previous exposure to a short course (4-8 weeks) of LDV/SOF in combination with GS-9451 and/or GS-9669 were retreated with LDV/SOF for 12 weeks [27]. Most patients (76%) were infected with HCV genotype 1a and had baseline NS5A mutations (64%) with high levels of resistance to LDV. Nevertheless, SVR12 rate was 85% (17/20; ITT) and 94.4% (17/18; per protocol) after extending therapy for 12 weeks. This suggests that relapse occurring after a “long” course (≥12 weeks) of an NS5A-containing regimen may facilitate the selection of NS5A RAVs with a high level of resistance and enough fitness to hamper future retreatment choices.
Understanding the persistence of RAVs is important to decide on the optimal retreatment strategy of patients who failed DAAs-based therapy. Black et al. [28] evaluated the prevalence and types of NS3 and NS5A variants at baseline and their impact on treatment response in patients treated with GZR+EBR with or without RBV, included in the C-WORTHY clinical trial. By population sequencing, 45% of G1a- and 8% of G1b-infected patients presented baseline NS3 variants resistant to 1st generation protease inhibitor (Q80K and S122G were the most frequent). The presence of baseline NS3 RAVs did not impact on treatment response. The prevalence of NS5A RAVs at baseline was lower (12%, the most common at positions 28, 31 and 93). However, the presence of baseline NS5A RAVs was associated with lower SVR24 in GT1a patients (69% and 96% with and without RAVs, respectively). At treatment failure, around 60% of patients had treatment-emergent NS3 and/or NS5A RAVs (the latter were detected at baseline by deep sequencing methods). Similarly, in the C-EDGE studies [29,30] GT1a naïve and treatment experienced patients with baseline NS5A RAVs that cause >5-fold potency reduction to EBV, achieved low SVR rates after therapy with GZP+EBV with or without ribavirin (20-50%) (Table 2). The emergence of RAVs after treatment with paritaprevir, ombitasvir and dasabuvir was also assessed in patients who did not achieve SVR in the Phase 2 and 3 clinical trials including this combination, over a follow up period of 24-48 weeks post-therapy [31]. In GT1a, the proportion of subjects with RAVs in NS3 declined over time, while most subjects with RAVs in NS5A and NS5B had these variants detectable through post-treatment week 48 (NS3 9%, NS5A 96% and NS5B 57% ). Finally, Sarrazin et al. [32] evaluated the prevalence and effect of baseline HCV NS5A RAVs on treatment outcome in 513 GT1 infected patients with compensated cirrhosis, treated
with LDV/SOF ±RBV, by population and next generation sequencing. Of 513 patients, 20 failed to achieve SVR12. Using a 1% deep-sequencing cut-off baseline NS5A RAVs were detected in 18% (94/511) of cirrhotic patients:
13 % in GT 1a and 25% in GT 1b.
Importantly, 9% and 17% of them had NS5A RAVs that conferred a >100-fold shift in EC50 to LDV. SVR12 rates were very high even in patients with baseline NS5A RAVS, but an impact on treatment efficacy was evident in G1a treatment-experienced cirrhotics with highly resistant NS5A RAVs (Table 2). In a 3-year registry study NS5A RAVs characterization was performed in patients who failed HCV treatment with LDV in the absence SOF. A high prevalence of NS5A RAVs at virologic failure was observed in these patients and RAVs persisted in >95% of patients through week 48 and in 86% through week 96 [33].
New drugs and shorter treatment duration The “raise” for a very short treatment duration (< 12 weeks) prompted some small phase 2 studies to combine 3 or even 4 DAAs targeting different NS HCV proteins with treatment durations ranging from 4 to 8 weeks [19,34-36]. Overall, data from these studies suggest that 4 weeks of therapy is clearly too short even for easy-to-treat patients, whereas 6-8 weeks might work in selected treatment-naïve individuals with mild fibrosis. Moving to phase 3 studies assessing regimens below 12 weeks in individuals with more advaned disease or who are treatment-experienced require more data that those currently presented at the ILC (Fig 1).
Hepatitis B: can we stop therapy after years of undetectable HBV-DNA?
Berg et al [37] presented the interim results of the FINITE study, a large phase 3 trial where non cirrhotic patients with HBeAg-Negative chronic hepatitis B who had been under virological suppression (with tenofovir) for a long time (> 3,5 years) were randomized to continue tenofovir or interrupt treatment. As expected, all 21 patients who continued on NUC therapy remained suppressed (HBV-DNA < 400 IU/L).
Regarding patients who
interrupted treatment, assessment at week 48 demonstrated that 2 individuals (10%) lost HBsAg, 12 (57%) had HBV-DNA < 2000 IU/L and ALT > 2 xULN, 4 (19%) had HBV-DNA > 2000 IU/L and ALT < 2 xULN and 3 (14%) had to restart tenofovir due to significant increase in ALT and HBV-DNA. This interim analysis suggests that treatment interruption is possible in selected patients who have been under long-term viral suppression with NUC therapy.
Liver Cancer: more negative trials, but better molecular characterization None of the systemic agents tested in phase 3 trials after the approval of sorafenib [38] in 2007 has been able to improve or even parallel the results of this drug in first line. Combination strategies with transarterial chemoembolization (TACE) have also been explored. For instance, the FGFR inhibitor brivanib was tested in combination with TACE but failed to improve the anti-tumoral effects of TACE alone [39]. Park et al presented results of a large phase 3 study that further evaluated the combination of a molecular agent with TACE (i.e., ORIENTAL trial [40]). ORIENTAL is an Asian multicenter phase 3 randomized controlled trial aimed at determining whether TSU-68 (orantinib) was able to increase survival in HCC patients treated with TACE. Preliminary data showed that TSU-68, a VEGFR/PDGFR/FGFR inhibitor, had no impact in overall survival and a modest effect in progression-free survival in a randomized phase 2 trial that enrolled 103 HCC patients [41].
The ORIENTAL trial included 889 patients, who were randomized 1:1 to receive either TSU68 or placebo after the first TACE procedure. The trial was prematurely stopped due to futility, with a median overall survival of 31.1 and 32.2 months (P=0.43) for TSU-68 and placebo, respectively. In terms of safety, patients receiving TSU-68 had more ascites and peripheral edema. Overall, it seems unlikely that TSU-68 may have a role in combination with TACE in patients with intermediate HCC (BCLC-B). There were other studies that addressed the role of extended indications of TACE, exploring its performance in patients beyond stage BCLC-B, the current accepted indication. Hucke et al presented a retrospective study of HCC patients at BCLC stage C, and evaluated the performance of their previously reported STATE score [42] to identify the ideal candidates for TACE among these patients [43]. The study analyzed a total of 80 BCLC-C patients following a trainingvalidation scheme. The STATE score was able to discriminate patient’s survival (5.2 versus 14.3 months, P20% or to less than 12 mmHg) was significantly greater in the combination group (82%) than in the control group (50%), and this was achieved with a lower dose of propranolol and less side effects. Though several questions remain, among them if the effects of rifaximin are limited to patients with ascites, the population in whom bacterial translocation is relevant, this study provides further rationale for the ongoing randomized trials to evaluate if rifaximin can prevent complications of portal hypertension (NCT02190357 and NCT01904409).
Repeated large-volume paracentesis is still the first-line therapy for the management of refractory ascites [54], but has a significant impact in the quality of life of the patients. A recent uncontrolled study has shown that alfapump, a device that drives ascites from the peritoneum into the bladder, might be effective in removing ascites in these patients [55]. Adebayo [56] and colleagues reported the effects of alfapump on systemic hemodynamics, nutritional parameters, markers of systemic inflammation and clinical response in a subset of patients (n=15) from a large multicenter ongoing randomized trial comparing alfapump with paracentesis. One out of 7 patients required paracentesis in the alfapump group, as compared with 8/8 in the paracentesis group. Alfapump patients experience reversible acute kidney injury. There was a trend for increased creatinine and renin activity, and for decreased albumin, without significant differences were observed in systemic hemodynamics or systemic inflammatory markers. There was a clear improvement in nutritional paramenters in the alfapump group. The authors suggested that albumin infusion might be useful in these patients to prevent the mild circulatory dysfunction induced by the continuous removal of ascites. These results will require confirmation in the whole sample of the randomized trial. Two large observational studies presented in the parallel sessions provided relevant information for clinical practice. Bossen et al [57] addressed the impact on mortality of non-selective beta blockers (NSBBs) in patients with ascites. This has been a highly controversial issue since the publication of a prospective cohort study [58] suggesting an increased mortality in patients with refractory ascites treated with NSBBs. The authors analyzed data from 1198 patients included in three randomized controlled studies assessing the effects of satavaptan (an aquaretic drug) in the treatment of ascites. Mortality rates were not different in patients treated with NSBBs (n=562) than in patients
not treated with NSBBs (n=636), even after adjusting for relevant prognostic variables. This was also the case in the subgroup of patients with refractory ascites (n=595), in patients with high or low doses of NSBBs, in patients with low or high MELD, with mean arterial pressure ≥82mmHg or
S0168-8278(15)00401-8 http://dx.doi.org/10.1016/j.jhep.2015.06.010 JHEPAT 5714
To appear in:
Journal of Hepatology
Received Date: Accepted Date:
10 June 2015 14 June 2015
Please cite this article as: Gambato, M., Villanueva, A., Abraldes, J.G., Altamirano, J., Forns, X., Clinical Trial Watch: Reports from the EASL International Liver Congress (ILC), Vienna, April 2015, Journal of Hepatology (2015), doi: http://dx.doi.org/10.1016/j.jhep.2015.06.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Clinical Trial Watch: Reports from the EASL International Liver Congress (ILC), Vienna, April 2015 Martina Gambato1, Augusto Villanueva2, Juan G. Abraldes3, Jose Altamirano4,5 and Xavier Forns1.
1
Liver Unit, Hospital Clínic, IDIBAPS and CIBEREHD, Barcelona, Spain; 2Liver Cancer
Program, Divisions of Liver Diseases and Hematology / Medical Oncology, Tisch Cancer Institute, Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA, 3Division of Gastroenterology (Liver Unit), University of Alberta, Edmonton, Alberta, Canada, 4Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain and 5 Vall d’Hebrón Institut de Recerca (VHIR). Barcelona, Spain.
Corresponding author: Xavier Forns. Liver Unit, Hospital Clínic, IDIBAPS and CIBEREHD, Barcelona, Spain. Phone 34 93 227 57 53 email: [email protected]
Electronic word count: 6039 Number of Figures: 1 Number of Tables: 2 List of Abbreviations: SOF: sofosbuvir, DCV: daclatasvir, RBV: ribavirin, CP: Child-Pugh, SVR: sustained virologial response, HVPG: hepatic venous pressure gradient, G: genotype, LDV: ledipasvir, GZR:grazoprevir, EBR:elbasvir, SMV: simeprevir, CKD: chronic kidney disease, RAVs: resistance associated variants, TACE: transarterial chemoembolization, AAV2:adeno-
associated virus 2, CSC: cancer stem cells, NSBBs: non-selective beta blockers, ALD: Alcoholic liver disease, GWA: genome-wide association, ARC: alcohol related cirrhosis, MAIT: Mucosal Associated Invariant T-cells, NAFDL: Non-alcoholic fatty liver disease
Keywords: DAA, complications of cirrhosis, NASH, hepatocellular carcinoma. Conflicts of interest: XF received unrestricted grant from Janssen and acted as advisor for Janssen, Gilead and Abbvie. Financial support: Authors contributions: MG, AV, JGA, JA and XF contributed to this manuscript by selecting and summarizing the information delivered at the ILC held in Vienna in April 2015 in their respective fields of expertise.
Summary
Viral Hepatitis: efficacy against hepatitis C is improving but resistance should not be neglected The ILC brought very exciting news in the hepatitis C field, with relevant data on new treatment combinations and important studies on already approved treatment regimens. Probably, the most expected studies were those dealing with treatment of “difficult to cure populations”, which are summarized in the following paragraphs. Patients with decompensated cirrhosis Patients with advanced liver disease, especially those with decompensated cirrhosis, have a poor prognosis and there is only limited experience with direct acting antivirals (DAAs) in this setting. Several clinical trials and studies in real life cohorts have addressed safety and efficacy issues in these patients.
The ALLY-1 study is a phase 3 trial that assessed the safety and efficacy of sofosbuvir (SOF), daclatasvir (DCV) and ribavirin (RBV) in patients with advanced cirrhosis and in post-liver transplant hepatitis C recurrence [1]. Sixty cirrhotic patients (20% Child-Pugh [CP]-A, 53% CP-B, and 27% CP-C) infected with genotypes 1 (G1) to 4 (75% G1) received 12 weeks of this oral regimen. The majority of patients had a MELD score between 10 and 15. The overall sustained virological response 12 weeks after treatment interruption (SVR12) was 83%, with excellent data in CP-A and B (92% and 94%, respectively), while efficacy dropped in patients with CP-C (56%), though the number of patients in this group was very small
(n=16). Non response was explained by relapse in 9 of the 10 virological failures. No severe adverse events were attributed to study medication. A question that remains open is if extension to 24 weeks of therapy would reduce the rate of relapse in individuals with more advanced disease.
In the SOLAR-2 study [2], G1 or G4 decompensated cirrhotic patients (n=160), both preand post-liver transplant (n=53) were randomized to receive 12 or 24 weeks of ledipasvir (LDV), SOF and RBV. More than 50% of these patients had ascites, hepatic encephalopathy, and around 25% had a MELD score >15. Overall, this regimen resulted in high SVR12 rates (85%-88%) independently of the treatment duration, both in the pre- and post-LT setting. It is important to notice that virological response was associated with improvements in MELD and CP scores from baseline, when assessed 4 weeks after treatment interruption. This was largely explained by a decrease in bilirubin and/or an increase in liver synthetic function (i.e albumin). Indeed, 35% CP-B (n=100) improved to CTP- A, and 48% of CP-C (n=54) improved to CP-B.
Despite this effect of viral clearance on liver function,
improvements in portal pressure may take longer. Indeed, the effect of viral suppression on hepatic venous pressure gradient (HVPG) was characterized in 50 cirrhotic patients (CPA: 18 and CP-B: 32) with portal hypertension ( ~ 85% HVPG > 12 mmHg). Patients were randomized to receive 48 weeks of open-label SOF and RBV; a control group received the same therapy after a 24-weeks observation period [3]. When comparing baseline and endof-treatment HVPG measurements in 37 cirrhotics with available paired assessments, HVPG decreased > 20% in only 24% of individuals. The latter indicates that most patients would remain at risk of clinical decompensation, suggesting that the remodeling of fibrous
septa and hepatic vascular changes in liver cirrhosis require a longer follow-up to improve or, that at some point, these changes are not reversible any more.
The safety and efficacy of a regimen not containing SOF to treat patients with advanced liver disease was presented at the ILC. The C-SALT trial [4] is a Phase 2/3 study which assessed a 12-week all-oral combination of the NS3 protease inhibitor grazoprevir (GZR) and the NS5A inhibitor elbasvir (EBR). The study included 30 GT 1-infected patients with CP-B (CP-7 70%) cirrhosis and 10 matched non-cirrhotic controls. Overall SVR12 was 90% (27/30), with a good safety profile.
Data on real life experience using SOF-containing regimens were also presented at the ILC. An update of SVR12 data coming from the TARGET cohort [5], which assessed the safety and efficacy of all-oral regimens in patients with a MELD score ≥ 10, confirmed the results presented at the last AASLD [6,7]. SVR12 rates were around 75% in G1 patients treated with SOF and SMV±RBV (with slightly higher efficacy in G1b), whereas SVR12 rates were significantly lower in G3 patients treated with SOF plus RBV (~ 40%). Foster et al. [8] presented an observational study conducted in 467 patients with decompensated cirrhosis infected with genotype 1 or 3 who underwent treatment with SOF plus LDV or DCV (with or without RBV). Around 65% of patients were CP-B with a mean MELD score of 12. Most G1 infected patients received SOF/LDV; most G3-infected patients received SOF/DAC (see below). SVR12 rates in G1 patients were very high, ranging from 81 to 86%. Four weeks after treatment interruption, MELD scores improved > 2 points in 42% of patients but worsened in 11%. Preliminary analysis suggested that in older individuals (age > 65) and those with more advanced disease (albumin < 35 g/dL) treatment did not result in an
improvement in liver function. Nevertheless, trying to identify those patients with decompensated cirrhosis who will benefit (or not) from viral clearance will require more studies and a longer follow-up after treatment interruption. Additonal studies including compensated and decompensated cirrhotic patients presented at the ILC are depicted in Table 1 [9-13].
G3-infected cirrhotic patients Genotype 3 has become the most difficult to treat genotype, especially in treatmentexperienced cirrhotic patients. Previous studies have already shown that in this setting, the combination of SOF and RBV, even when extended to 24 weeks, SOF/DCV (without RBV) or SOF/LDV and RBV for 12 weeks achieve only around 60%-70% of SVR rates, which is clearly suboptimal in the current therapeutic scenario [14-16]. The BOSON study [17] is a multicentric, randomized, open-label clinical trial, assessing the efficacy of three different regimens: PEG-IFN, RBV and SOF for 12 weeks, and SOF plus RBV for 24 or 16 weeks. A total of 592 patients were included: G2 treatment-experienced with cirrhosis (n=48) and G3 treatment naïve or experienced, with or without cirrhosis (n=544). For G3-infected patients the highest SVR12 rates were obtained with PEG-IFN, RBV and SOF (93%), as compared with 84% and 71% with the 24- and 16-week SOF/RBV regimens, respectively. The results favoring a PEG-IFN-based regimen were even clearer in treatment-experienced cirrhotics, with an SVR12 of 86% in the PEG-IFN arm and only 77% and 47% in the 24- and 16-week SOF/RBV arms, respectively.
Regarding IFN-free therapy, SOF/DCV plus RBV for 12 weeks appeared to be a better choice than SOF/LDV plus RBV in G3 patients included in the English Extended Access Program [8]. Indeed, in this cohort which included mainly patients with decompensated cirrhosis, SVR12 rates were 70% in the 114 individuals who underwent SOF/DCV plus RBV compared to 59% in the 61 individuals who underwent SOF/LDV plus RBV. Based on data from the French compassionate use program [18], it seems that extending SOF plus DAC to 24 weeks might increase the efficacy in patients with advanced liver disease. Finally, data from a small phase 2 study showed excellent efficacy in a small group of G3 infected patients (n=40) who underwent treatment with the combination of SOF with GZP/EBV for 12 weeks. SVR12 was achieved in 38 (95%) of the 40 patients (10 of 11 cirrhotics, 91%) [19].
Liver transplant recipients with severe hepatitis C recurrence As stated above, the SOLAR-2 study [2] assessed the safety and efficacy of SOF/LDV plus RBV (12 or 24 weeks) in 168 patients with hepatitis C recurrence after liver transplantation, (including patients with compensated and decompensated cirrhosis). In individuals with F0F3 and compensated cirrhosis SVR12 rates were 95% and 98% with 12 and 24 weeks of treatment, respectively. Interestingly, a small subanalysis of patients with biopsy-proven fibrosing cholestatic hepatitis (n=11) included in this study showed an SVR12 rate of 100% in this very sick population [20]. Overall, there were no major safety concerns regarding the use of this regimen in this population and no changes in immunosupression were necessary. As part of the ALLY-1 study [1] the efficacy of 12 weeks of DCV/SOF +RBV was evaluated in 53 infected patients with HCV recurrence (77% G1): 30% of patients had cirrhosis and 25% bridging fibrosis. SVR12 rate was 94%; 3 individuals presented a
virological relapse. Only 4 patients had to interrupt RBV due to anemia and no modification of the immunosupressive drugs were required due to the favorable drug-drug interaction profile of this combination. Coilly et al. presented data assessing SOF/DCV as part of the French multicentric real life CULPIT cohort [21]. Most of the 130 patients included in the analysis received a 24-week treatment (approximately half with RBV). More than 50% of the patients had bridging fibrosis or cirrhosis, including some with decompensated disease. G1 was predominant. The reported efficacy was very high, ranging from 96 to 100%. In patients receiving RBV anemia requiring RBV dose reductions and erithropoietin use were common. Overall, data presented at the ILC strongly supports that all-oral regimens are as effective in liver transplant recipients as in the immunocompetent population. This is particularly true for patients with less advanced disease, and emphasizes that antiviral therapy should be indicated before individuals reach advanced disease. Importantly, neither SOF/LDV nor SOF/DCV have significant interactions with tacrolimus and cyclosporine, facilitating the use of these regimens in this setting. Real life data indicate that in LT recipients who cannot tolerate RBV, a 24-week regimen with SOF/DCV achieves very high SVR rates. Although the combination of SOF/LDV for 24 weeks (without RBV) appears to be as efficacious as a 12week regimen with RBV in cirrhotic patients [22], data are not yet available in liver transplant recipients.
Patients with chronic kidney disease (CKD)
There were practically no data on interferon-free regimens in HCV-infected patients with CKD. At the ILC, data from two clinical trials were presented. In the first one, 224 G1 infected patients (naïve or treatment experienced) with CKD stage 4 (eGFR 15-29 ml/min) or 5 (eGFR < 15 ml/min) were randomized to an observational group or to receive GZV/EBV for 12 weeks [23]. Half of the patients were infected with G1a, 75% were on dialysis, most were treatment naïve and less than 10% were cirrhotic. Of the 117 patients who received medication, all but one (99%) achieved SVR12. Adverse events and laboratory abnormalities were similar between the treatment and the observational groups. In a smaller study [24], 20 G1-infected treatment naïve patients with CKD stage 4 or 5 received paritaprevir/ritonavir, ombitasvir and dasabuvir for 12 weeks (no RBV in 7 G1b; with RBV in 13 G1a). Despite the starting dose of RBV was 200 mg/d for G1a patients, 8 of 13 interrupted the drug. At the meeting presentation all 14 patients with end-of-treatment data and 10 who had reached week 4 of follow-up remained HCV-RNA undetectable. Except for anemia in those who received RBV, no major safety issues were recorded. Overall, it seems that in the near future two regimens (one pending regulatory approval) will be available and highly efficacious in patients with end-stage renal disease.
Retreatment of patients who failed previous DAA-based therapy: the relevance of resistance-associated variants (RAVs) Retreatment of patients who failed previous DAA-based regimens is an increasing problem. At the ILC, data using GZR/EBV with RBV and SOF/LDV were presented. Seventy-nine patients with chronic HCV GT1 infection, 43% with cirrhosis, who failed triple therapy with PEG-IFN/RBV combined with boceprevir, telaprevir or simeprevir, were retreated with GZR,
EBV and RBV for 12 weeks [25]. Almost half of the patients had baseline NS3 resistance associated variants (RAVs). All but 3 patients (96%) achieved SVR12. Similar results were obtained by retreating 40 HCV G1 infected Japanese patients who failed previous PI-based triple therapy, with LDV/SOF with or without RBV for 12 weeks [26]. All patients achieved SVR12, including cirrhotic patients (80% of overall population). Thus, it seems that patients who failed a treatment regimen including first generation protease inhibitors have very good prospects for cure. Treatment of patients who failed a previous IFN-free regimen containing an NS5A inhibitor seems to be more difficult. Indeed, 41 G1 infected patients who failed 8 or 12 weeks of LDV/SOF-based therapy, underwent the same treatment regimen for a longer duration (24 weeks)[26]. Forty-six percent of them were cirrhotic. The overall response rate was 71%. In patients who underwent a 12-week course of SOF/LDV the presence of NS5A RAVs at baseline was more frequent (100%) and the efficacy was lower (SVR12 46%) than in patients who were treated for 8 weeks (NS5A RAVs 63%, SVR12 80%). In a phase 2a clinical trial, patients with early stage (F0-F2) liver fibrosis, with previous exposure to a short course (4-8 weeks) of LDV/SOF in combination with GS-9451 and/or GS-9669 were retreated with LDV/SOF for 12 weeks [27]. Most patients (76%) were infected with HCV genotype 1a and had baseline NS5A mutations (64%) with high levels of resistance to LDV. Nevertheless, SVR12 rate was 85% (17/20; ITT) and 94.4% (17/18; per protocol) after extending therapy for 12 weeks. This suggests that relapse occurring after a “long” course (≥12 weeks) of an NS5A-containing regimen may facilitate the selection of NS5A RAVs with a high level of resistance and enough fitness to hamper future retreatment choices.
Understanding the persistence of RAVs is important to decide on the optimal retreatment strategy of patients who failed DAAs-based therapy. Black et al. [28] evaluated the prevalence and types of NS3 and NS5A variants at baseline and their impact on treatment response in patients treated with GZR+EBR with or without RBV, included in the C-WORTHY clinical trial. By population sequencing, 45% of G1a- and 8% of G1b-infected patients presented baseline NS3 variants resistant to 1st generation protease inhibitor (Q80K and S122G were the most frequent). The presence of baseline NS3 RAVs did not impact on treatment response. The prevalence of NS5A RAVs at baseline was lower (12%, the most common at positions 28, 31 and 93). However, the presence of baseline NS5A RAVs was associated with lower SVR24 in GT1a patients (69% and 96% with and without RAVs, respectively). At treatment failure, around 60% of patients had treatment-emergent NS3 and/or NS5A RAVs (the latter were detected at baseline by deep sequencing methods). Similarly, in the C-EDGE studies [29,30] GT1a naïve and treatment experienced patients with baseline NS5A RAVs that cause >5-fold potency reduction to EBV, achieved low SVR rates after therapy with GZP+EBV with or without ribavirin (20-50%) (Table 2). The emergence of RAVs after treatment with paritaprevir, ombitasvir and dasabuvir was also assessed in patients who did not achieve SVR in the Phase 2 and 3 clinical trials including this combination, over a follow up period of 24-48 weeks post-therapy [31]. In GT1a, the proportion of subjects with RAVs in NS3 declined over time, while most subjects with RAVs in NS5A and NS5B had these variants detectable through post-treatment week 48 (NS3 9%, NS5A 96% and NS5B 57% ). Finally, Sarrazin et al. [32] evaluated the prevalence and effect of baseline HCV NS5A RAVs on treatment outcome in 513 GT1 infected patients with compensated cirrhosis, treated
with LDV/SOF ±RBV, by population and next generation sequencing. Of 513 patients, 20 failed to achieve SVR12. Using a 1% deep-sequencing cut-off baseline NS5A RAVs were detected in 18% (94/511) of cirrhotic patients:
13 % in GT 1a and 25% in GT 1b.
Importantly, 9% and 17% of them had NS5A RAVs that conferred a >100-fold shift in EC50 to LDV. SVR12 rates were very high even in patients with baseline NS5A RAVS, but an impact on treatment efficacy was evident in G1a treatment-experienced cirrhotics with highly resistant NS5A RAVs (Table 2). In a 3-year registry study NS5A RAVs characterization was performed in patients who failed HCV treatment with LDV in the absence SOF. A high prevalence of NS5A RAVs at virologic failure was observed in these patients and RAVs persisted in >95% of patients through week 48 and in 86% through week 96 [33].
New drugs and shorter treatment duration The “raise” for a very short treatment duration (< 12 weeks) prompted some small phase 2 studies to combine 3 or even 4 DAAs targeting different NS HCV proteins with treatment durations ranging from 4 to 8 weeks [19,34-36]. Overall, data from these studies suggest that 4 weeks of therapy is clearly too short even for easy-to-treat patients, whereas 6-8 weeks might work in selected treatment-naïve individuals with mild fibrosis. Moving to phase 3 studies assessing regimens below 12 weeks in individuals with more advaned disease or who are treatment-experienced require more data that those currently presented at the ILC (Fig 1).
Hepatitis B: can we stop therapy after years of undetectable HBV-DNA?
Berg et al [37] presented the interim results of the FINITE study, a large phase 3 trial where non cirrhotic patients with HBeAg-Negative chronic hepatitis B who had been under virological suppression (with tenofovir) for a long time (> 3,5 years) were randomized to continue tenofovir or interrupt treatment. As expected, all 21 patients who continued on NUC therapy remained suppressed (HBV-DNA < 400 IU/L).
Regarding patients who
interrupted treatment, assessment at week 48 demonstrated that 2 individuals (10%) lost HBsAg, 12 (57%) had HBV-DNA < 2000 IU/L and ALT > 2 xULN, 4 (19%) had HBV-DNA > 2000 IU/L and ALT < 2 xULN and 3 (14%) had to restart tenofovir due to significant increase in ALT and HBV-DNA. This interim analysis suggests that treatment interruption is possible in selected patients who have been under long-term viral suppression with NUC therapy.
Liver Cancer: more negative trials, but better molecular characterization None of the systemic agents tested in phase 3 trials after the approval of sorafenib [38] in 2007 has been able to improve or even parallel the results of this drug in first line. Combination strategies with transarterial chemoembolization (TACE) have also been explored. For instance, the FGFR inhibitor brivanib was tested in combination with TACE but failed to improve the anti-tumoral effects of TACE alone [39]. Park et al presented results of a large phase 3 study that further evaluated the combination of a molecular agent with TACE (i.e., ORIENTAL trial [40]). ORIENTAL is an Asian multicenter phase 3 randomized controlled trial aimed at determining whether TSU-68 (orantinib) was able to increase survival in HCC patients treated with TACE. Preliminary data showed that TSU-68, a VEGFR/PDGFR/FGFR inhibitor, had no impact in overall survival and a modest effect in progression-free survival in a randomized phase 2 trial that enrolled 103 HCC patients [41].
The ORIENTAL trial included 889 patients, who were randomized 1:1 to receive either TSU68 or placebo after the first TACE procedure. The trial was prematurely stopped due to futility, with a median overall survival of 31.1 and 32.2 months (P=0.43) for TSU-68 and placebo, respectively. In terms of safety, patients receiving TSU-68 had more ascites and peripheral edema. Overall, it seems unlikely that TSU-68 may have a role in combination with TACE in patients with intermediate HCC (BCLC-B). There were other studies that addressed the role of extended indications of TACE, exploring its performance in patients beyond stage BCLC-B, the current accepted indication. Hucke et al presented a retrospective study of HCC patients at BCLC stage C, and evaluated the performance of their previously reported STATE score [42] to identify the ideal candidates for TACE among these patients [43]. The study analyzed a total of 80 BCLC-C patients following a trainingvalidation scheme. The STATE score was able to discriminate patient’s survival (5.2 versus 14.3 months, P20% or to less than 12 mmHg) was significantly greater in the combination group (82%) than in the control group (50%), and this was achieved with a lower dose of propranolol and less side effects. Though several questions remain, among them if the effects of rifaximin are limited to patients with ascites, the population in whom bacterial translocation is relevant, this study provides further rationale for the ongoing randomized trials to evaluate if rifaximin can prevent complications of portal hypertension (NCT02190357 and NCT01904409).
Repeated large-volume paracentesis is still the first-line therapy for the management of refractory ascites [54], but has a significant impact in the quality of life of the patients. A recent uncontrolled study has shown that alfapump, a device that drives ascites from the peritoneum into the bladder, might be effective in removing ascites in these patients [55]. Adebayo [56] and colleagues reported the effects of alfapump on systemic hemodynamics, nutritional parameters, markers of systemic inflammation and clinical response in a subset of patients (n=15) from a large multicenter ongoing randomized trial comparing alfapump with paracentesis. One out of 7 patients required paracentesis in the alfapump group, as compared with 8/8 in the paracentesis group. Alfapump patients experience reversible acute kidney injury. There was a trend for increased creatinine and renin activity, and for decreased albumin, without significant differences were observed in systemic hemodynamics or systemic inflammatory markers. There was a clear improvement in nutritional paramenters in the alfapump group. The authors suggested that albumin infusion might be useful in these patients to prevent the mild circulatory dysfunction induced by the continuous removal of ascites. These results will require confirmation in the whole sample of the randomized trial. Two large observational studies presented in the parallel sessions provided relevant information for clinical practice. Bossen et al [57] addressed the impact on mortality of non-selective beta blockers (NSBBs) in patients with ascites. This has been a highly controversial issue since the publication of a prospective cohort study [58] suggesting an increased mortality in patients with refractory ascites treated with NSBBs. The authors analyzed data from 1198 patients included in three randomized controlled studies assessing the effects of satavaptan (an aquaretic drug) in the treatment of ascites. Mortality rates were not different in patients treated with NSBBs (n=562) than in patients
not treated with NSBBs (n=636), even after adjusting for relevant prognostic variables. This was also the case in the subgroup of patients with refractory ascites (n=595), in patients with high or low doses of NSBBs, in patients with low or high MELD, with mean arterial pressure ≥82mmHg or
