If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address requests for reprints to: Peter Ferenci, MD, Prof, Universitätsklinik für Innere Medizin III, Medical University of Vienna, AKH, Waehringer Guertel 18-20, A 1090 Wien, Austria. fax: (43) 1 0 40400 4735
Oral Silibinin (SIL) is widely used for treatment of hepatitis C, but its efficacy is unclear. Substantially higher doses can be administered intravenously (IV).
Pedigreed nonresponders to full-dose pegylated (Peg)-interferon/ribavirin (PegIFN/RBV) were studied. First, 16 patients received 10 mg/kg/day SIL IV (Legalon Sil; Madaus, Köln, Germany) for 7 days. In a subsequent dose-finding study, 20 patients received 5, 10, 15, or 20 mg/kg/day SIL for 14 days. In both protocols, PegIFNα-2a/RBV were started on day 8. Viral load was determined daily.
Unexpectedly, in the first study, HCV-RNA declined on IV SIL by 1.32 ± 0.55 log (mean ± SD), P < .001 but increased again in spite of PegIFN/RBV after the infusion period. The viral load decrease was dose dependent (log drop after 7 days SIL: 0.55 ± 0.5 [5 mg/kg, n = 3], 1.41 ± 0.59 [10 mg/kg, n = 19], 2.11 ± 1.34 [15 mg/kg, n = 5], and 3.02 ± 1.01 [20 mg/kg, n = 9]; P < .001), decreased further after 7 days combined SIL/PegIFN/RBV (1.63 ± 0.78 [5 mg/kg, n = 3], 4.16 ± 1.28 [10 mg/kg, n = 3], 3.69 ± 1.29 [15 mg/kg, n = 5], and 4.85 ± 0.89 [20 mg/kg, n = 9]; P < .001), and became undetectable in 7 patients on 15 or 20 mg/kg SIL, at week 12. Beside mild gastrointestinal symptoms, IV SIL monotherapy was well tolerated.
IV SIL is well tolerated and shows a substantial antiviral effect against HCV in nonresponders.
; the remaining components are silydianin, silycristin, isosilybinA, isosilybinB, isosilycristin, and taxifolin. Silybum marianum has a history as a medical plant for almost 2 millennia. The mode of action of silymarin is unknown. Silibinin has strong antioxidative
which makes it a potentially useful drug for treatment of chronic liver diseases. The largest randomized, controlled trial performed in the 1970s indicated that long-term treatment with silymarin may decrease mortality in patients with cirrhosis.
Part of this uncertainty is due to the lack of data on its pharmacokinetics and optimal dosing regimens. Silymarin is poorly water soluble, and, thus, the oral preparation has a limited bioavailability. In humans, a water-soluble form of silibinin (silibinin hemisuccinate) is used for treatment of death cup (Amanita phalloides) intoxication as a specific antidote of amanitin.
oxidative stress may contribute to fibrosis and carcinogenesis in chronic hepatitis C virus (HCV) infection. Because the balance of the oxidative and reductive potentials within the cell (cellular redox state) has profound consequences on signal transduction pathways
the antioxidative properties of silibinin in chronic hepatitis C may improve the response to interferon in nonresponders to pegylated (Peg)-IFN/ribavirin (RBV) (PegIFN/RBV). To increase its concentration, silibinin was administered intravenously (IV). Unexpectedly, we observed that silibinin had potent antiviral properties against the HCV. Consequently, we explored in detail the antiviral potency of silibinin.
Patients and Methods
Patients with previous nonresponse to full dose of PegIFN/RBV combination therapy were selected for these studies (Table 1). Nonresponse was defined by the lack of a >2-log drop of viral load after 12 weeks of therapy and/or by not achieving an end-of-treatment response. Patients were required to have had a liver biopsy within 2 years prior to inclusion into this study. Standard inclusion/exclusion criteria for PegIFN/RBV therapy were applied.
During a screening phase within 35 days prior to the first dose of study drug, eligibility of patients according to inclusion/exclusion criteria was established. All patients had at least 1 quantitative HCV RNA test within the 6 months before the screening phase.
Patients first received daily 10 mg/kg silibinin (Legalon Sil; Madaus, Köln, Germany) infused over 4 hours for 7 consecutive days. On day 1, blood was drawn for determination of oxidative stress parameters at baseline, every 30 minutes during the infusion, and 2 hours after the end of the infusion. On day 8, treatment was changed to 140 mg silymarin (Legalon; Madaus) 3 times daily per os in combination with 180 μg/wk PegIFNα-2a (PEGASYS; Roche, Basel, Switzerland) and 1–1.2 g/day RBV (COPEGUS; Roche).
After obtaining the results for the first protocol, treatment with silibinin was extended for 2 weeks, and different doses of silibinin were administered. Patients first received daily 5, 10, 15, or 20 mg/kg silibinin infused over 4 hours for 14 consecutive days. On day 8, treatment with 180 μg/wk PegIFNα-2a and 1–1.2 g/day RBV was started. After day 14, patients received 280 mg silymarin (Legalon; Madaus) 3 times daily per os (orally). During the 14-day infusion period, blood was obtained daily for determination of viral load.
In both protocols in case of intolerance to Peg-IFNα-2a or ribavirin, standard dose adjustment guidelines were used. Antiviral combination therapy was given for a total of 24 weeks (with the option to stop treatment in patients without a >2-log drop at week 12); virologic responders at week 24 were offered to continue treatment for a further 48 weeks. After end of the infusion period, patients were tested after weeks 2 and 4 and then monthly until the end of therapy at week 24. The protocol was approved by the Ethics Committee of the Medical University of Vienna. The details of the study were explained to the patients, and all signed an informed consent.
Serum HCV RNA level was determined by the TaqMan PCR assay (Cobas Ampliprep/Cobas TaqMan HCV Test; limit of detection, 15 IU/mL; Roche Diagnostics, Pleasanton, CA). In a preliminary experiment, it was shown that silibinin (even at the concentration in the undiluted solution) did not inhibit the assay.
Reactive oxidative metabolites in blood were measured by the d-ROMs test (Reactive Oxygen Metabolites derived compounds; Diacron, Grosseto, Italy),
and the amounts of antioxidants by the BAP test (Biological Antioxidant Potential; Diacron) using the portable, free radicals determination system (FRAS 4; SEAC, Calenzano, Italy) before, every 30 minutes during (on day 1), and 2 hours after the silibinin infusions. The d-ROM test measures reactive oxygen metabolites (primarily hydroperoxides) released from plasma proteins by an acidic buffer, which in presence of iron generate alkoxyl and peroxyl radicals, according to the Fenton's reaction. Such radicals, in turn, are able to oxidize an alkyl-substituted aromatic amine (N,N-dietylparaphenylendiamine), thus producing a pink-colored derivative that is photometrically quantified at 505 nm. Results for reactive oxidative metabolites are expressed as Caratelli units (Ucarr; normal: 250–300, 1 Ucarr = 0.08 mg hydrogen peroxide/dL). The BAP test measures the decoloration intensity of a ferric chloride solution mixed with a thiocyanate derivative by the added plasma sample photometrically at 505 nm, which is proportional to the ability to reduce ferric ions by the amounts of antioxidants in plasma (normal: >2200 μmol/L). The description of the assays by the manufacturer does not specify which substances are actually measured.
Originally, the primary outcome variable was the virologic response defined as the percentage of patients being polymerase chain reaction (PCR) negative at end of treatment (week 24). Secondary efficacy variables were virologic response rates at week 12, safety and tolerability of treatment with Peg-IFN/RBV/silymarin; quality of life at baseline, week 24, week 48, week 72 (SF-36, Fatigue Severity Scale); and the oxidative status after silibinin infusions. Because of the unexpected strong virologic response after 7 days of silibinin infusions, the recruitment was halted, and the study was redesigned based on virologic response parameters using longer infusion periods and higher doses of silibinin. For the original study, the sample size was estimated based on Gehan's 2-stage design. According to previous studies, a response rate of >10% seems to warrant further investigation of the treatment regime. Twenty-nine patients had to be recruited in the first stage accordingly (error probability, β = 5%).
Sixteen pedigreed nonresponders (Table 1) were included. All patients had received full-dose treatment with PegIFN (12 PegIFN-α-2a, 2 PegIFN-α-2b) and RBV (1000–1200 mg/day) for at least 12 weeks. Parameters of oxidative stress measured did not change during silibinin infusions (Figure 1).
Serum HCV RNA declined in all patients on IV silibinin-monotherapy (Figure 2) (baseline: 6.59 ± 0.53, day 8: 5.26 ± 0.81 log IU/mL [mean ± SD], P < .001) with a mean log decline of 1.32 ± 0.55 within 1 week. In parallel, alanine aminotransferase decreased from 162 ± 133 to 118 ± 107 U/L (P = .004). In all patients, HCV RNA remained detectable at initiation of PegIFN/RBV therapy. Three patients declined PegIFN/RBV combination therapy. In 11 of the remaining 13 patients, HCV RNA increased again after the end of the silibinin infusions in spite of initiation of PegIFN/RBV therapy. At week 12, all patients were still HCV RNA positive, but 5 patients had a >2-log drop and continued treatment (Figure 3). None of them became HCV RNA negative at week 24; 1 patient had a 5.5-log drop and continues treatment by own choice.
Twenty pedigreed nonresponders (Table 1) were included. All patients received full-dose treatment with PegIFN (18 PegIFN-α-2a, 4 PegIFN-α-2b; 2 patients received 2 treatment courses) and RBV (1000–1200 mg/day) for at least 12 weeks.
Figure 4 shows the viral kinetics in these patients. Viral load declined continuously. After 7 days of silibinin monotherapy the 5-mg/kg dose was marginally effective (n = 3; log drop, 0.55 ± 0.5), whereas the 10 mg/kg (n = 19 [including the patients in protocol 1], log drop 1.41 ± 0.59), 15 mg/kg (n = 5; log drop, 2.11 ± 1.15), and 20 mg/day doses (n = 9; log drop, 3.02 ± 1.01) led to a highly significant decrease in viral load (P < .001).
After 1 week of combined silibinin and PegIFN/RBV therapy, viral load decreased further (log drop, 5 mg/kg: 1.63 ± 0.78; 10 mg/kg: 4.16 ± 1.28; 15 mg/kg: 3.69 ± 1.29; 20 mg/kg: 4.8 ± 0.89; all P < .0001 vs baseline) (Figure 5). Two of the 5 patients in the 15 mg/kg group and 4 of the 9 patients in the 20 mg/kg group had HCV RNA <15 IU at day 15. HCV RNA was <15 IU/mL in 8 and 7 patients at week 4 (week 5 of the study protocol) and week 12 (week 13 of the study protocol) after start of PegIFN/RBV, respectively. All patients are still on antiviral combination therapy (Figure 6).
Silibinin was generally well tolerated. Five patients complained of mild gastrointestinal symptoms (abdominal pain, 5; diarrhea, 2; nausea, 1), 2 of headache, and 1 of arthralgia. All of these were rated to be mild by the patients and subsided after the end of the infusions; changes in the dosing were not required. All patients in the 15- and 20-mg/kg groups noted a sensation of heat when the infusion was started, which subsided within 30 minutes without treatment. No serious adverse events occurred. On monotherapy, no changes of hemoglobin, leucocytes, platelets, and creatinine were observed. The typical adverse effects of antiviral combination therapy were observed (including 1 patient suffering from increasing dyspnea because of Haemophilus influenzae-induced pneumonia, requiring termination PegIFN/RBV therapy after 8 weeks).
The unexpected finding of this study was the demonstration that IV silibinin is a potent antiviral agent in patients with chronic hepatitis C not responding to standard antiviral combination therapy. Intravenous silibinin was well tolerated, and no serious adverse effects were observed. The most commonly reported adverse effect was a transient sensation of heat. The antiviral effect was dose dependent but was not maintained after the end of the infusion period by the oral administration of silymarin. Thus, the original protocol (study 1) failed to improve the outcome of nonresponders by augmenting the response to interferon, which was the primary hypothesis of the study. Therefore, the planned trial to retreat nonresponders in a larger prospective trial was halted, and a phase 2 dose-finding study (study 2) was initiated. Using the 15- and 20-mg/kg/day dose, HCV RNA was <15 IU/mL in 6 of 14 patients after the silibin infusion period of 14 days; these patients are still on treatment.
The present study confirms the described in vitro antiviral effects of silibinin in patients with chronic hepatitis. To the best of our knowledge, this is the first documentation of the antiviral properties of silibinin in humans. A rigorously standardized silymarin preparation (MK-001) had direct antiviral effects against HCV infection in the HCV replicon system.
Two other commercial preparations of silymarin also displayed antiviral activity, indicating that antiviral action is not unique to MK-001. In addition, MK-001 displayed anti-inflammatory actions via inhibition of nuclear factor-κB-induced transcription in human liver cell cultures and inhibition of inflammatory cytokine induction in human peripheral blood mononuclear cells. Other studies have also shown antiproliferative and antinuclear factor-κB actions at relatively high doses of silymarin.
The reasons for these discrepancies are primarily due to the dose and route of administration of the drug. In contrast to modern drug development, which requires a precise knowledge of the pharmacologic properties of a compound before it can be tested in humans, silymarin is a prescription drug in Europe and a food additive in the United States without data on pharmacokinetics and optimal dosing.
Because of the complexicity of absorption, metabolism, and disposition nature of various flavonoids, it is still largely unclear which form, ie, the parent flavonoid or its metabolite(s) contributes to the overall effects in the body. Although flavonoids are rapidly absorbed after oral ingestion, their plasma concentrations are very low, whereas the phase II metabolites such as glucuronides, sulfates, and methylated conjugates seem to be predominant in blood circulation.
Extensive first-pass metabolism in the intestine and the liver are responsible for low oral bioavailabilities of flavonoids. Accordingly, pharmacokinetics of silymarin and its components may be different when given orally or parenterally. Oral administration may not approach the levels in plasma/liver that were tested in vitro. Similar amounts of silymarin given orally had no effect on HCV load,
reflecting differences in bioavailability and metabolism of silibinin resulting in far lower plasma levels. After oral dosing, silymarin flavonolignans are quickly glucuronidated and rapidly eliminated with short half-lives.
Silymarin conjugates were the primary components present in human plasma. The area-under-the-curve 0→∞ values of the conjugated silymarin flavonolignans were 4- to 30-fold higher than those of their free fractions. The individual silymarin flavonolignans exhibited quite different plasma profiles for both the free and conjugated fractions. In patients with liver disease, silymarin pharmacokinetics are altered,
with cirrhotic patients having highest silymarin exposures. Currently, no data are available for IV administration of silibinin. The more favorable antiviral response in study 2 may reflect that it included fewer patients with cirrhosis than study 1. The mechanism of the antiviral effect of silibinin is unknown. The effect may be either a direct inhibition of viral replication as suggested by the replicon data
all of which can modulate antiviral responses independently of the IFN→Jak-Stat→interferon-stimulated gene paradigm. The observation in this study that PegIFN was not able to amplify the antiviral effect of silibinin supports this concept. Naringenin, an other flavonoid (extracted from grapefruit) reduces HCV secretion in infected cells by 80%,
as well as the transcription of 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase and acyl-coenzyme A:cholesterol acyltransferase 2 in infected cells. Similar effects of silibinin should be explored.
The clinical use of silibinin/silymarin for treatment of chronic hepatitis C will depend on future studies addressing the pharmacokinetics, the drug interaction profiles, and the optimal dose and dosing of this compound. The daily IV administration of silibinin limits it use in clinical practice, and other dosing schedules need to be explored. Based on the currently available data, only patients with a viral load <100 IU/mL at week 2 became HCV RNA negative on continued PegIFN/RBV therapy. The planned phase III studies should probably include longer administration of silibinin infusions in those who do not respond within 2 weeks. Nevertheless, our observation suggests that silibinin is well tolerated even in higher doses and may be a very useful drug for treatment of IFN/RBV nonresponders or in future combinations with HCV protease or polymerase inhibitors.
The authors thank Kerstin Zinober and Reza Pourbiabany for technical support.
Molecular structure and stereochemistry of silybin A, silybin B, isosilybin A, and isosilybin B, isolated from Silybum marianum (milk thistle).
The authors disclose the following: Conflicts of interest: P. Ferenci: Roche, Basel, CH; member of the Global Advisory Board and of the speakers bureau and receives financial support for clinical studies. All other authors have no conflicts of interest.
Supported by an unrestricted research grant by Roche Austria, and silibinin (Legalon Sil) and silimaryin (Legalon) were provided by Rottapharm-Madaus, Köln, Germany. None of the companies was involved in the design, conduct, and evaluation of this protocol.