| |
Pioglitazone in chronic hepatitis C with Insulin Resistance not responding to pegylated interferon-β and ribavirin
|
| |
| |
from Jules: Perhaps it would be necessary to fully treat insulin first, that is get glucose abnormality or diabetes under much better control for a much longer period of time before starting Peg/RBV. Or perhaps this intervention would only be successful in patients with full blown diabetes, and it would be necessary to first gain full control of diabetes or glucose abnormaity for a long period of time before starting Peg/RBV.
Journal of Hepatology
Articles in Press
Kathrin Overbeck12, Daniel Genne4, Alain Golay3, Francesco Negro12, on behalf of the Swiss Association for the Study of the Liver (SASL)
Received 29 February 2008; received in revised form 26 March 2008; accepted 26 March 2008. published online 22 May 2008.
Uncorrected Proof
ABSTRACT
Background/Aims
Insulin resistance reduces the response to interferon alfa-based therapy of chronic hepatitis C patients. It has been speculated that improvement of insulin sensitivity might increase the chances of responding to treatment of such individuals.
Methods
We started a multicenter clinical trial of retreatment of chronic hepatitis C patients, who had failed to respond to the pegylated interferon alfa/ribavirin combination, with a triple therapy consisting in these same antivirals plus an insulin-sensitizer (pioglitazone) (The INSPIRED-HCV study).
Results
None of the first five patients fulfilling the inclusion criteria and included in the trial had a satisfactory virological response after 12 weeks of retreatment, despite the fact that in at least three of them the insulin resistance score improved. As a result, the study was terminated.
Conclusions
Different schedules are warranted to improve insulin sensitivity prior to attempting retreatment of chronic hepatitis C patients with insulin resistance.
Introduction
Insulin resistance and diabetes are major disease modifiers in chronic hepatitis C, as they increase liver fibrogenesis [1], [2], [3], [4] and reduce the rate of response to antivirals [5], [6], [7]. Regarding the latter, a sustained virological response (SVR) was reported to occur in 23 of 70 (32.8%) of patients with genotype 1 and insulin resistance (measured as homeostasis assessment of insulin resistance, HOMA-IR>2) vs. 26 of 43 (60.5%) of genotype 1 patients without insulin resistance (p=.007; OR, 3.12, 95% CI, 1.42-6.89) [5]. These findings were independently confirmed [6] and recently extended to non-responders with genotypes 2 and 3 [7]. Thus, based on our recent results [7], we [7], [8] and others [5] have suggested that insulin resistance should be corrected in patients with chronic hepatitis C not responding to currently available antiviral treatment, in order to improve response to retreatment. The modalities of this intervention, however, have not been established. In addition, the optimal HOMA-IR score to be attained has not been identified.
We planned a prospective, multicenter study to investigate the efficacy and safety of the insulin-sensitizer pioglitazone (ActosTM, Takeda Pharma AG, Lachen, Switzerland) 15mg QD, added to the pegylated interferon-β2a (PEG-IFN-_-β2a) (PegasysTM, Roche Pharma Schweiz AG, Reinach, Switzerland) 180_g QW/ribavirin (CopegusTM, Roche) 1000-1200mg QD combination therapy in chronic hepatitis C patients who had previously failed to respond (i.e. had detectable serum HCV RNA after 12 weeks of therapy) to a pegylated interferon-β/ribavirin combination without the insulin-sensitizer ("a pilot study of treatment with pegylated interferon-alpha2a, ribavirin and insulin-sensitizer pioglitazone of insulin resistance (with the exception of diabetes) in hepatitis C virus infection", The INSPIRED-HCV study, registered as NCT00433069 at Clinicaltrials.gov). All patients had a baseline HOMA-IR score >2 as additional inclusion criterion, because this was the threshold discriminating responders from non-responders in previous works [5], [7]. Diabetic patients were excluded. The trial was approved by the Ethical Committee of the University Hospitals of Geneva, and all patients consented to participate.
Multiple case reports
Patient #1 was a 46-year-old female who had failed previous treatment of PegasysTM and CopegusTM four years earlier. A liver biopsy performed in 2004 showed a Metavir score of A2 F3, with mild steatosis affecting 20% of hepatocytes. She had a normal body weight with a BMI of 20.8, and was not known for a hypertension or a dyslipidemia. As concomitant medications, she was receiving duloxetine 60mg QD and zopiclone 7.5mg QD. After 12 weeks of treatment, serum HCV RNA level did not change (from 176,000IU/ml to 154,000IU/ml, measured by Cobas TaqMan HCV, Roche Diagnostics, Rotkreuz, Switzerland) and, oddly enough, the HOMA-IR score increased from 2.65 to 11. Thus, therapy was discontinued. Her body weight remained stable during therapy.
Patient #2 was a 49-year-old male with a liver biopsy performed in 2003 that showed a mild chronic hepatitis, focal interface hepatitis, mild steatosis and a Metavir fibrosis score of F3. Prior to this biopsy, in 2001, he had been enrolled in an international clinical trial [9] where he had received PegasysTM and CopegusTM at the same doses tested in the INSPIRED-HCV study. Immediately before enrolment in the latter, the liver stiffness (measured by FibroScan, Echosens, Paris, France) was 11.1kPa. At baseline, the patient had a serum HCV RNA level of 11,000,000IU/ml (genotype 1b), a HOMA-IR score of 4.64, a marked overweight (98kg, 182cm for a BMI of 29.6 and a waist circumference of 120cm), but no dyslipidemia (serum total cholesterol 5.4mmol/L, triglycerides 0.3mmol/l and HDL cholesterol 2.0mmol/l) nor arterial hypertension (145/85mmHg). After 12 weeks of therapy, the HCV RNA and the HOMA-IR decreased, respectively, to 64,000IU/ml and 2.66. The body weight fell to 88kg and the waist circumference to 111cm. No dose changes of the three drugs had been necessary, as they were all sufficiently well tolerated. Despite the fall in viral load, the patient had his therapy discontinued. In fact, the _2.23 Log reduction in serum HCV RNA was superimposable to that observed on occasion of the first treatment six years before (_2.24 Log), when no pioglitazone had been added.
Patient #3 was a 59-year-old female, with histologically documented cirrhosis and a mild steatosis, functionally compensated, related to a HCV genotype 1b infection. She had no overweight (62kg, waist circumference 83cm), no arterial hypertension and a baseline HOMA-IR score of 5.71. Treatment was very well tolerated, with no dose reductions. After 12 weeks of therapy, the HOMA-IR score decreased slightly to 4.52, while serum HCV RNA went from 410,000 to 61,000IU/ml. All treatment were stopped. Body weight and waist circumference were virtually unchanged (63kg and 82cm at week 12).
Patient #4 was a 50-year-old male with histologically proven cirrhosis, and mildly overweight (BMI 25.6, waist circumference 93cm) and a poorly controlled arterial hypertension discovered one year prior to enrolment. At this time, he was taking lisinopril 20mg QD and hydrochlorothiazide 12.5mg QD but blood pressure was 170/115mmHg. Serum cholesterol and triglycerides were within the normal ranges, but HDL cholesterol was low (0.5mmol/L). The baseline HOMA-IR score was 3.79 and serum HCV RNA 1,100,000IU/ml (genotype 1b). Antiviral treatment was rather well tolerated, with mild fatigue not necessitating drug dose modifications. After 12 weeks, the HOMA-IR was virtually unchanged (3.57) but HCV RNA had fallen to 11,000IU/ml. Since during a previous course of antiviral treatment the viremia had decreased by less than 1 Log, we resorted to continue the triple therapy. After 24 weeks, however, viremia had increased slightly to 42,000IU/ml and all treatment were definitely stopped. Body weight remained unchanged throughout the whole course of therapy.
Patient #5 was a 46-year-old male. A liver biopsy carried out one year before enrolment had shown mild activity and portal fibrosis (Metavir A1 F1), with a mild steatosis affecting 20% of hepatocytes. The patient had received a 12-week course of combination with PegasysTM and CopegusTM resulting in a _0.95 Log change in his serum HCV RNA level one year prior to enrolment. His baseline HCV RNA was 2,700,000IU/ml (genotype 1a), and the HOMA score 2.88. He had normal body weight (60kg, waist circumference 87cm) and normal blood pressure. As concomitant medications, he was receiving olanzapine 15mg QD and escitalopram 20mg QD for a well controlled depression. The study drugs were well tolerated, but after 12 weeks from start the HOMA-IR had increased to 3.72, while the serum HCV RNA had decreased to 300,000IU/ml, i.e. _0.85 Log, even less than on the occasion of the first treatment. Body weight and waist circumference were slightly reduced (58kg and 85cm).
Thus, addition of pioglitazone at start of the antiviral combination resulted in a weak reduction of the HOMA-IR score in three patients, while in the remaining two the level of insulin resistance increased (Table 1). The HCV viral load decreased in all patients, albeit to a varying extent and unrelated to the HOMA-IR score variation. Treatment was discontinued at week 12 in patients 1, 2, 3 and 5 and prolonged in patient 4 until week 24, to no avail. Based on these discouraging results, we took the decision to terminate the study.
Discussion
The rationale of increasing insulin sensitivity in non-responders lies on the premise that insulin resistant state directly or indirectly inhibits the antiviral action of IFN-_-β, or increases the viral fitness making it more resistant to therapy, or both. We [10] and others [11], [12], [13] have reported, based on in vitro experimental models and/or observations in patients, that intracellular factors dysregulated by HCV and responsible for the insulin resistant state often play promiscuous roles, as they are also involved in regulating the IFN-_-β signaling transduction. These include some members of the suppressor of cytokine signaling (SOCS) family [10], [11], [12] and the protein phosphatase 2Ac [13]. Modulating the levels and/or the activity of these factors may not only reverse hepatic insulin resistance but also help in establishing the IFN-_-β-induced antiviral state at the very site of HCV replication. However, specific inhibitors of SOCS family members and of the protein phosphatase 2Ac are either not available for in vivo administration or toxic [14]. Alternatively, one may consider increasing systemic insulin sensitivity by correcting the circulating levels of specific cytokines, such as TNF-_, associated with insulin resistance [15], [16], [17], although some caution has been suggested while administering anti-TNF-_ therapy to chronic hepatitis C patients [18]. Insulin-sensitizers may also inhibit HCV replication by decreasing serum free fatty acids overflow to hepatocytes, since saturated and monounsaturated free fatty acids have been shown to stimulate HCV replication [19].
Thus, although increasing insulin sensitivity may be a rational option in chronic hepatitis C patients not responding to current combination therapy, our approach was clearly inadequate to reach the stated goal. If this working hypothesis is acceptable, we prompt other investigators to evaluate different schedules.
In our study, it remains to be explained the paradoxical increase of the HOMA-IR score observed in two of our patients during therapy, which was among the reasons that prompted us to stop the trial. It is noteworthy to mention that both patients were taking additional drugs for their psychiatric comorbidities: patient #1 was taking duloxetine, a serotonin norepinephrine reuptake inhibitor, and zopiclone, a cyclopyrrolone-derived hypnotic agent, while patient #5 was taking escitalopram and olanzapine, an antipsychotic drug. Although some serotonin reuptake inhibitors may induce hyperglycemia [20], in general they ameliorate insulin sensitivity [21]. Furthermore, the dual-mechanism antidepressant duloxetine has not been reported to alter glucose homeostasis [21]. Zopiclone overdoses have been only exceptionally associated with hyperglycemia [22]. On the other hand, olanzapine is well-known for impairing whole body insulin sensitivity [23]. In addition to the potential effects of these drugs, also the acute administration of IFN-_-β may induce some degree of insulin resistance in both healthy subjects [24] and chronic hepatitis C patients [25]. Thus, although the unwanted action of IFN-_-β on glucose metabolism may have been overcome by pioglitazone in at least three patients, this beneficial effect may have been nullified in the two remaining patients taking drugs for psychiatric indications. The interactions among all these drugs are however speculative, but require further studies, in view of the frequent use of antidepressants in chronic hepatitis C patients.
Thus, further clinical trials aiming at reducing the insulin resistance in chronic hepatitis C via different pharmacological interventions are warranted. It is not clear whether the best approach would be using thiazolidindiones such as pioglitazone, with a preeminent peripheral effect [26], or a biguanide such as metformin, whose mechanism of action seems focused on the hepatic AMP-activated protein kinase [27], [28]. Another major question lies in the modality of the intervention. It is not clear whether one should start the antiviral retreatment immediately (as we did) or only once the HOMA-IR score has decreased to a level predicting a sufficient SVR rate [5]. As an alternative, higher dosages of thiazolidindiones (as compared to what used in the present trial) or metformin may be used, neither can we exclude that insulin sensitizing therapy should be tailored according to the infecting HCV genotype [10]. Finally, interactions of insulin sensitizing agents with other drugs, most notably those taken for psychiatric comorbidities, should be considered. Further clinical trials are clearly needed to address all these important issues.
|
|
| |
| |
|
|
|
