Original Article Treatment of Chronic Hepatitis D Patients with Pegylated Interferon: a Real-World Experience

Antiviral Therapy 2014; 19:463–468 (doi: 10.3851/IMP2728)

Original article Treatment of chronic hepatitis D patients with pegylated interferon: a real-world experience

Zaigham Abbas1,2*, Mohammad S Memon3, Hammad Mithani2, Wasim Jafri1, Saeed Hamid1

1Department of Medicine, The Aga Khan University Hospital, Karachi, Pakistan 2OMI Hospital, Karachi, Pakistan 3Isra University, Hyderabad, Pakistan

*Corresponding author e-mail: [email protected]

Background: Published experience of treating chronic response in 23 (22.1%) patients. Sustained virological hepatitis D patients with pegylated interferon (PEG‑IFN)‑a response (SVR) at 24 weeks post-treatment was seen in is limited. The aim of this study was to determine the 24 (23.1%) patients each for the virological and bio‑ efficacy of 48 weeks of treatment with PEG-IFN in naive chemical responses and in 13 (12.5%) as combined patients outside the clinical trial setting, in the real world. response. Both ETR and SVR were associated with a neg‑ Methods: Patients with chronic hepatitis D were treated ative HDV RNA at 24 weeks of treatment (P=0.001 and with PEG-IFN. The primary end points were sustained P=0.000, respectively). Detectable HDV RNA at this point clearance of HDV RNA and normal alanine aminotrans‑ had a positive predictive value of 0.95 (range 0.85–0.99) ferase (ALT) at 24 weeks post-treatment. for detectable RNA at 6 months post-treatment. End of Results: The total number of patients treated with PEG- treatment biological response, that is, normal ALT at the IFN was 104; 91 males, mean age ±sd 30.1 ±10.0 years end of treatment was also a predictor of ETR and SVR (range 15–55). Cirrhosis was present in 41 patients. (P=0.004 and P=0.041, respectively). With an intention-to-treat analysis, end of treatment Conclusions: Treatment with PEG-IFN for hepatitis D is virological response (ETR) was achieved in 44 (42.3%), of limited efficacy. Detectable HDV RNA at 24 weeks of normalization of ALT in 38 (35%) and a combined treatment is a predictor for a failed SVR.

Introduction

HDV is a defective RNA agent needing the presence of treatment led to sustained elimination of HDV RNA HBV for its own replication [1]. Persistent HDV repli- in only about 20–25% of patients [1,3]. Its efficacy cation leads to cirrhosis and hepatocellular carcinoma in a real-world scenario is not known. The aim of at annual rates of 4% and 2.8%, respectively [2]. These this study was to determine the efficacy and safety figures underscore the need to treat this infection. Unfor- of PEG-IFN given for 48 weeks for treatment-naive tunately, there is no effective antiviral therapy available patients with chronic hepatitis D in a real-world clini- for the treatment of hepatitis D. The only established cal setting. therapy for chronic hepatitis D is interferon-a, standard or pegylated, which has to be administered for at least Methods 48 weeks [3]. A recent Cochrane review [4] showed that therapy Participants with standard interferons may result in an end of treat- All the patients with chronic hepatitis D who were ment virological and biochemical response in one-third of treated with PEG-IFN-a from 2009 to 2011 were patients but sustained virological response (SVR), judged considered for inclusion in this retrospective analy- by loss of HDV RNA from serum at six months, was sis. The patients came from three centres that were seen in only 17.4% of patients. Histological improve- following the same algorithm to treat hepatitis D. ment was seen in one-quarter of the patients [4]. The liver biopsy and PCR studies were done in a The experience of using pegylated interferon (PEG- single centre. Inclusion criteria were patients aged IFN)-a in hepatitis D is limited to a few studies. The between 18 and 60 years with chronic hepatitis D

AVT-13-OA-3109_Abbas.indd 463 02/10/2014 16:05:53 Z Abbas et al.

and compensated liver disease, who were hepatitis On-treatment evaluation B surface antigen (HBsAg)-reactive and anti-HDV Patients were evaluated as outpatients for safety, tolerance antibody positive for at least 6 months, had presence and efficacy every 4 weeks during treatment until week 48 of HDV RNA in blood by PCR, had alanine ami- and then at 72 weeks during the follow-up period. At each notransferase (ALT) levels more than the upper limit visit complete blood count and biochemistry was assessed. of normal, haemoglobin ≥10.0 g/dl, white blood cell HDV RNA was checked at screening, 24 weeks, end of counts of ≥2.5×109/l, neutrophils ≥1.5×109/l, plate- treatment (48 weeks) and end of follow-up (24 weeks lets ≥75×109/l and negative tests for HCV and HIV. post-treatment). Serum HBV DNA was tested at screen- Exclusion criteria were the patients who had received ing, week 48 and end of follow-up. antiviral therapy within the previous 6 months, had evidence of other genetic, autoimmune or meta- Adverse events bolic liver diseases, were pregnant or breastfeeding, Serious adverse events were defined as those that were showed evidence of alcohol abuse, uncontrolled psy- fatal, life-threatening, required inpatient hospitalization chiatric disease, severe chronic pulmonary, renal or or discontinuation of treatment. Non-serious adverse cardiac diseases, bleeding disorders, elevated alpha- events and laboratory abnormalities leading to dose fetoprotein >100 ng/ml and active or suspected can- modifications and premature withdrawal from therapy cer elsewhere in the body. Liver biopsy specimens were noted. were available only in 39 patients. Grading of inflammation and the staging of fibrosis was performed Statistical analyses according to the Batts–Ludwig classification [5]. All Intent-to-treat efficacy analysis was conducted on all the patients were managed by the usual standards of those subjects who were started on treatment (n=104). care. Local ethics committees did not find any objec- Patients with missing values were considered as non- tion to publication of the data. responders. Additional per protocol analysis was made on the population of patients who completed 48 weeks Treatment of therapy. Data were expressed as the number of sub- All patients eligible for antiviral therapy were treated jects with percentages for nominal variables or means with PEG-IFN-a2a (Pegasys®; F Hoffmann–La Roche with standard deviation for continuous variables. Treat- Ltd, Basel, Switzerland) 180 mg subcutaneously per ment factors associated with HDV RNA negativity at week or PEG-IFN-a2b (Peg-Intron®; Schering-Plough week 72 (24 weeks post-treatment) were assessed. The Corp., Kenilworth, NJ, USA) 1.5 mg/kg subcutaneous c2 test and Fisher’s exact tests were used to compare weekly in a weight-based dosing regimen. Patients were categorical data. Continuous variables were evaluated educated regarding administration of pegylated inter- using t-test or non-parametric test (Mann-Whitney U feron, expected adverse events, schedule for laboratory test). A two-tailed P-value <0.05 was required for sta- monitoring and clinic appointment. tistical significance.

Outcomes Results Primary end points were undetectable HDV RNA levels (virological response) and normal ALT (bio- A total of 104 patients were treated and followed-up chemical response) at the end of 24 weeks of post- between January 2009 and December 2011. These treatment follow-up. Secondary end points were the included 69 patients who received PEG-IFN-a2a normalization of ALT and HDV RNA negativity at (180 mg once weekly) and 35 patients who received the end of treatment. PEG-IFN-a2b (1.5 mg/kg once weekly). There were 91 male patients, mean age 30.1 ±10.0 years (range Viral nucleic acids testing 15–55). Demographic and baseline clinical characteris- Viral nucleic acids were isolated from patients’ serum tics of the patients are shown in Table 1. samples by High Pure Viral Nucleic Acid Kit, according Hepatitis B e antigen (HBeAg) was reactive in 29 to the manufacturer’s instructions (Roche Diagnostics, (27.9%) cases while HBV DNA was detectable in 37 Indianapolis, IN, USA). HBV DNA PCR qualitative (35.6%); 22 were HBeAg-positive. Liver biopsy was analysis was performed by the Cobas Amplicor HBV available in only in 39 cases. Mild inflammation was Monitor (Roche Diagnostic Systems, Basel, Switzer- seen in 1, moderate in 12 and severe in 26 patients. Stage land) with a lower detection limit of 500 copies/ml. For of the disease was 1–2 in 30 and 3–4 in 9 patients. Cir- HDV RNA, the qualitative test was based on the reverse rhosis of the liver as evident from histology, ultrasound transcription PCR of the target gene. Quantification of or clinical examination, was present in 41 (39.4%) HDV RNA was done by real-time PCR having a lower patients. One patient had diabetes and five had arterial limit of detection of 500 IU/ml. hypertension.

AVT-13-OA-3109_Abbas.indd 464 02/10/2014 16:05:53 Pegylated interferon for hepatitis D

With an intention-to-treat analysis, clearance of HDV End of treatment virological response and SVR at RNA at the end of treatment was achieved in 44 patients week 24 post-treatment was associated with negative (42.3%) and normalization of ALT in 38 (35%). Both HDV RNA at 24 weeks of treatment (P-values 0.001 virological and biochemical responses were seen in 23 and 0.000), risk ratio 2.29 (range 1.29–4.33) and 7.75 (22.1%) patients. The efficacy of PEG-IFN was sus- (range 1.98–47.16), odds ratio 3.92 (range 1.55–10.11) tained in 24 (23.1%) both in terms of virological and and 11.56 (range 2.38–76.39), respectively (Figure 1). biochemical responses and in 13 (12.5%) as a combined If the HDV RNA was detectable at 24 weeks, the posi- response at 24 weeks follow-up. This data included two tive predictive value (PPV) for a detectable RNA at end patients who had detectable HDV RNA at the end of of treatment was 0.76 (range 0.54–0.85) and the nega- treatment but negative PCR 6 months post-treatment. tive predictive value (NPV) 0.56 (range 0.47–0.63). This PPV further increased to 0.95 (range 0.85–0.99) for a detectable PCR at 6 months post-treatment and Table 1. Baseline characteristics of the study patients NPV decreased to 0.36 (range 0.29–0.39). End of treat- Characteristics Value ment biological response, that it, normal ALT at the end of treatment was also a predictor of end of treatment Mean age, years ±sd 30.1 ±10.0 virological response and SVR (P=0.004 and 0.041), Median age, years (range) 30.0 (15–55) risk ratio 1.90 (range 1.17–2.96) and 2.05 (range Male gender, n (%) 91 (87.5) 0.95–4.45), odds ratio 3.29 (range 1.32–8.25) and 2.60 Mean body mass index, kg/m2 ±sd 23.5 ±5.02 (range 0.93–7.32), respectively (Table 2). Palpable liver, n (%) 28 (26.9) Among 80 patients who could not achieve SVR, Palpable spleen, n (%) 30 (28.8) Cirrhosis (clinical/biopsy/ultrasound), n (%) 41 (39.4) 58 were non-responders, which included 7 break- Mean haemoglobin, g/dl ±sd 13.5 ±1.9 throughs, that is, negative PCR at 24 weeks and Mean TLC×106/l ±sd 6.5 ±2.6 detectable HDV RNA at 48 weeks. Three patients had Mean number of platelets ×109/l ±sd 186 ±72 undetectable HDV RNA at 24 weeks but were lost Mean bilirubin, mg/dl ±sd 0.94 ±0.78 to follow-up between 24 and 48 weeks of treatment Mean ALT, IU/l ±sd 114 ±100 and were considered non-responders in the intention- Mean GGT, IU/l ±sd 92 ±98 to-treat analysis. The rest of the non-responders were Mean alkaline phosphatase, IU/l ±sd 152 ±88 ‘null’ responders including three out of four patients HBeAg-positive patients, n (%) 29 (27.9) who decompensated during treatment. Twenty-two HBV DNA detected, n (%) 37 (35.6) patients had end of treatment virological response but relapsed. This figure included two patients who were ALT, alanine aminotransferase; GGT, gamma glutamyl transferase; HBeAg, hepatitis B e antigen; TLC, total leukocyte count. lost to follow-up after completion of treatment.

Table 2. Factors associated with SVR at 6 months post-treatment Factor SVR-positive (n=24) SVR-negative (n=80) P-value

Mean age, years ±sd 33.7 ±9.2 30.0 ±10.2 0.111 Male gender, n 22 69 0.728 Mean body mass index, kg/m2 ±sd 23.8 ±3.8 23.4 ±5.6 0.759 Cirrhosis, n 7 34 0.341 Mean number of platelets ×109/l ±sd 191 ±89 184 ±66 0.719 Mean baseline ALT, IU/l ±sd 127 ±135 110 ±87 0.458 Mean baseline GGT, IU/l ±sd 89 ±80 94 ±105 0.901 HBeAg reactive, n 5 24 0.380 HBV DNA detected, n 9 28 0.813 Severe inflammation (biopsy in 39 patients), n 7 19 1.000 Advanced fibrosis, (biopsy in 39 patients), n 2 7 1.000 Mean HDV RNA level ×106 IU/ml (30 patients) ±sd 11.6 ±15.9 7.4 ±14.4 0.464 Pegylated interferon-a2a, n 15 54 0.649 Pegylated interferon-a2b, n 9 26 – Negative HDV RNA at 24 weeks, n 33 26 0.001 Mean ALT at 24 weeks, IU/l ±sd 63 ±40 75 ±43 0.283 Normal ALT at 48 weeks, n 13 25 0.041 Negative HDV RNA at 48 weeks, n 22 22 0.000

ALT, alanine aminotransferase; GGT, gamma glutamyl transferase; HBeAg, hepatitis B e antigen; SVR, sustained virological response.

Antiviral Therapy 19.5 465

AVT-13-OA-3109_Abbas.indd 465 02/10/2014 16:05:53 Z Abbas et al.

Figure 1. Week 24 HDV RNA (week 24 PCR) and normalization of ALT at the end of treatment (week 48 ALT) predicting ETR and SVR

Week 24 PCR and ETR Week 24 PCR and SVR

Week 24 PCR-positive Week 24 PCR-negative Week 24 PCR-positive Week 24 PCR-negative

P=0.001 P=0.000 41 39 33 34

ETR achieved ETR not achieved SVR achieved SVR not achieved

Week 48 ALT and ETR Week 48 ALT and SVR

Week 48 ALT elevated Week 48 ALT normal Week 48 ALT elevated Week 48 ALT normal

P=0.004 45 P=0.041 55

23 21 25 15 11 13

ETR achieved ETR not achieved SVR achieved SVR not achieved

ALT, alanine aminotransferase; ETR, end of treatment virological response; SVR, sustained virological response.

Considering patients who completed 48 weeks of the study according to the protocol and the follow-up therapy (n=98) after excluding dropouts (n=3) and (n=96) was 25% (24/96), biochemical response was therapy discontinuation due to decompensation (n=3), found in 25% (24/96) and a combined virological and the end of treatment virological response was seen in biochemical response in 13/96 (13.5%). Mean ALT ±sd 45% (44/98) and biochemical response in 39% (38/98). in patients with SVR was 51.6 ±60.3 and in patients Mean ALT ±sd was 49.2 ±24.8 in responders and 83.0 without SVR 92.8 ±75.8 (P=0.028). ±60.5 in non-responders (P=0.000). The proportion The side effects reported by these patients were usu- of sustained virological responders who completed ally of PEG-IFN and included nausea, weakness, fever,

AVT-13-OA-3109_Abbas.indd 466 02/10/2014 16:05:54 Pegylated interferon for hepatitis D

bloating, decreased appetite, body aches, headaches pegylated interferon even without virological response and weight loss. These side effects did not require a (Figure 1). The ultimate goal of treatment should be the dose reduction. Significant thrombocytopenia was seen eradication of HBsAg which is difficult to achieve as in 6 patients, which required interruption of therapy for none of our patients became HBsAg-negative. 2 weeks in two cases, and dose reduction in the rest for We followed up our patients only for 24 weeks post- 3–4 weeks. One patient received interleukin-11 as well, treatment and found a sustained response of 23%. From to increase the platelet count. One patient developed our experience of treating hepatitis C with interferon, significant neutropenia and was given filgrastim. Four the possible factors that may influence the outcome of patients decompensated; three during the treatment treatment in the case of hepatitis D could also be geno- after 24 weeks of therapy and one during the follow-up type and viral load, rapid virological response, stage of period. All were male and were aged 32, 40, 40 and 45 fibrosis, duration of disease, previous treatment, dura- years. They were having clinical cirrhosis at the time of tion of therapy, compliance and IL28B polymorphism. initiation of treatment but fulfilled the inclusion crite- The genotype of HBV and baseline HBsAg level might ria. The patient who deompensated during the follow- also be possible predictors. Apparently, hepatitis D up period had SVR, while the rest of the three were behaves in a different way from hepatitis C. We tried non-responders. to find out factors leading to SVR in our patients. The predominant genotype of hepatitis D is 1 [14] and of Discussion HBV is D [15] in our province (Sindh) so genotypes could not confound our results. Role of HBsAg titre in So far, interferon-a, standard or pegylated, is the only predicting response has not been established [9]. Pres- treatment option for hepatitis D. In a systematic meta- ence or absence of cirrhosis, degree of inflammation and analysis, standard interferon-a led to the end of treat- stage of fibrosis did not influence the outcome in our ment virological response in 32.6% of the patients study. Kabaçam et al. [16] also found that response rate compared with 7.8% in the untreated controls [4]. SVR to pegylated interferon is the same in patients with or was achieved in 17.4% of patients on interferon com- without advanced liver disease. Studies have shown that pared with 5.2% of controls. PEG-IFN is considered measuring HDV RNA at 24 weeks of therapy and docu- superior to conventional interferon. However, a head- menting 2–3 log reduction may be of help in predicting to-head comparison of PEG-IFN with standard inter- response [6–8,17]. As we did not have the baseline viral feron has not been carried out. The pilot studies using load available in all patients we relied on qualitative PCR, PEG-IFN-a2b (1.5 mg per kilogram of body weight which was carried out in all cases at 24 weeks. Negative per week) for periods of 48 to 72 weeks showed an HDV RNA was a strong predictor of SVR in our study SVR 6 months after the end of treatment in 17 to 43% (risk ratio 7.75 [range 1.98–47.16]). If the HDV RNA of the treated patients [6–8]. The results of HIDIT-1 was detectable at 24 weeks, PPV for a detectable RNA trial showed that treatment with PEG-IFN-a2a for 48 at 6 months post-treatment was 0.95 (range 0.85–0.99) weeks, with or without adefovir, resulted in sustained and NPV was 0.36 (range 0.29–0.39). This means that HDV RNA clearance in about one-quarter of patients patients who still harbour HDV RNA in their blood at with HDV infection [9]. All these trials were done in a 24 weeks of treatment are less likely to have SVR, and controlled setting. We have gathered here the response at the same time negative PCR at 24 weeks does not in the real-world scenario while treating these patients guarantee an SVR. in clinics. Our experience is not much different from A strength of the study is that it included a reasonable the published clinical trials. number of patients and the sample size is much larger than The incidence of HDV infection in the West is many previous studies. This study tells us the response rate decreasing but the infection is still very much preva- and compliance in the real-world scenario, which was not lent in Pakistan [10]. Though the prevalence of HBsAg worse than the controlled settings. The shortcomings of positivity is 2.5%, there are certain districts with much this study were that all the baseline parameter tests in con- higher prevalence. This ‘hepatitis belt’ also has the high- trolled trials could not be carried out in all patients, these est prevalence of HDV super-infection and represents included quantitative PCR, genotype and pre-treatment a serious public health problem [11,12]; therefore, we biopsies. The patients were followed up for only 24 weeks were able to get enough patients to offer treatment and after treatment and long-term follow-up were not avail- compile this data. able to predict sustained response later on. It has been documented previously that despite low In conclusion, treatment with pegylated interferon rates of eradication of the virus, standard interferon for hepatitis D is of limited efficacy. It is difficult to may favourably affect the natural history of the disease expect who would respond to the treatment. A detect- in the long run [13]. We witnessed normalization of able HDV RNA at 24 weeks of treatment is a good pre- ALT in a significant number of patients treated with dictor for a failed SVR.

Antiviral Therapy 19.5 467

AVT-13-OA-3109_Abbas.indd 467 02/10/2014 16:05:55 Z Abbas et al.

9. Wedemeyer H, Yurdaydìn C, Dalekos GN, et al. Disclosure statement Peginterferon plus adefovir versus either drug alone for hepatitis delta. N Engl J Med 2011; 364:322–331. This study was not funded by any agency. The authors 10. Abbas Z, Jafri W, Raza S, Hepatitis D. Scenario in the Asia- declare no competing interests. Pacific region. World J Gastroenterol 2010; 16:554–562. 11. Seetlani NK, Abbas Z, Raza S, Yakoob J, Jafris W. Prevalence of hepatitis D in HBsAg positive patients visiting References liver clinics. J Pak Med Assoc 2009; 59:434–437. 12. Mumtaz K, Hamid SS, Adil S, et al. Epidemiology and 1. Pascarella S, Negro F. Hepatitis D virus: an update. Liver clinical pattern of hepatitis delta virus infection in Pakistan. Int 2011; 31:7–21. J Gastroenterol Hepatol 2005; 20:1503–1507. 2. Romeo R, Del Ninno E, Rumi M, et al. A 28-year study 13. Farci P, Mandas A, Coiana A, et al. Treatment of chronic of the course of hepatitis delta infection: a risk factor for hepatitis D with interferon alfa-2a. N Engl J Med 1994; cirrhosis and hepatocellular carcinoma. Gastroenterology 330:88–94. 2009; 136:1629–1638. 14. Moatter T, Abbas Z, Shabir S, Jafri W. Clinical presentation 3. Wedemeyer H. Hepatitis D revival. Liver Int 2011; 31 Suppl and genotype of hepatitis delta in Karachi. World J 1:140–144. Gastroenterol 2007; 13:2604–2607. 4. Abbas Z, Khan MA, Salih M, Jafri W. Interferon alpha for 15. Abbas Z, Muzaffar R, Siddiqui A, Naqvi SA, Rizvi SA. chronic hepatitis D. Cochrane Database Syst Rev 2011; Genetic variability in the precore and core promoter regions 12:CD006002. of hepatitis B virus strains in Karachi. BMC Gastroenterol 5. Batts KP, Ludwig J. Chronic hepatitis. An update on 2006; 6:20. terminology and reporting. Am J Surg Pathol 1995; 16. Kabaçam G, Dalekos GN, Çakaloğlu Y, et al. Pegylated 19:1409–1417. interferon-based treatment in patients with advanced liver 6. Erhardt A, Gerlich W, Starke C, et al. Treatment of chronic disease due to chronic delta hepatitis. Turk J Gastroenterol hepatitis delta with pegylated interferon-alpha2b. Liver Int 2012; 23:560–568. 2006; 26:805–810. 17. Yurdaydin C, Bozkaya H, Onder FO, et al. Treatment of 7. Castelnau C, Le Gal F, Ripault MP, et al. Efficacy of chronic delta hepatitis with lamivudine vs lamivudine + peginterferon alpha-2b in chronic hepatitis delta: relevance interferon vs interferon. J Viral Hepat 2008; 15:314–321. of quantitative RT-PCR for follow-up. Hepatology 2006; 44:728–735. 8. Niro GA, Ciancio A, Gaeta GB, et al. Pegylated interferon alpha-2b as mono-therapy or in combination with ribavirin in chronic hepatitis delta. Hepatology 2006; 44:713–720. Accepted 19 December 2013; published online 14 January 2014

AVT-13-OA-3109_Abbas.indd 468 02/10/2014 16:05:55