Ribavirin Steady-State Plasma Level Is a Predictor of Sustained Virological Response in Hepatitis C-Infected Patients Treated With Direct-Acting Antivirals
M. van Tilborg,
F. I. Lieveld,
E. J. Smolders,
K. J. van Erpecum,
C. T. M. M. de Kanter,
R. Maan,
M. van der Valk,
J. E. Arends,
A. S. M. Dofferhoff,
H. Blokzijl,
M. Bijmolen,
J. P. H. Drenth,
R. J. de Knegt,
D. M. Burger,
on behalf of the HepNed Study Group
M van Tilborg et al.
Summary
Background: In the era of highly effective direct-acting antivirals (DAAs) for treatment of patients with chronic hepatitis C virus (HCV) infection, ribavirin (RBV) is still considered beneficial in certain patients.
Background: In the era of highly effective direct-acting antivirals (DAAs) for treatment of patients with chronic hepatitis C virus (HCV) infection, ribavirin (RBV) is still considered beneficial in certain patients.
Aim: To assess the association between RBV steady-state plasma levels and sustained virological response (SVR).
Methods: Consecutive HCV-infected patients treated with DAAs plus RBV from four Dutch academic medical centres were enrolled. RBV steady-state plasma levels were prospectively measured at treatment week 8 using validated assays. Logistic regression analyses were performed to assess the influence of RBV steady-state plasma level on SVR, and RBV therapeutic range was explored using area under the ROC curve analyses.
Results: A total of 183 patients were included, of whom 85% had one or more difficult-to-cure characteristics (ie treatment experienced, HCV genotype 3, cirrhosis). The majority was treated with a sofosbuvir-based regimen and 163 (89%) patients achieved SVR. Median RBV dose was 12.9 (interquartile range 11.2-14.7) mg/kg/d, and median RBV steady-state plasma level was 2.66 (1.95-3.60) mg/L. In multivariable analyses, higher RBV steady-state plasma level (adjusted odds ratio 1.79 [95% CI 1.09-2.93]) was an independent predictor of SVR. With regard to the optimal RBV therapeutic range, 2.28 mg/L was the optimal lower cut-off for achieving SVR and 3.61 mg/L was the upper cut-off for preventing significant anaemia (Haemoglobin < 10 g/dL).
Conclusion: In this cohort of mainly difficult-to-cure patients treated with DAAs plus RBV, higher RBV steady-state plasma level was an independent predictor of SVR.
Conclusion: In this cohort of mainly difficult-to-cure patients treated with DAAs plus RBV, higher RBV steady-state plasma level was an independent predictor of SVR.
DISCUSSION ONLY
Full Text Available Online
In this prospective study, the influence of RBV steady-state plasma level on achieving SVR in HCV-infected patients treated with the combination of DAAs and RBV was determined. The main finding of this study is that a higher RBV steady-state plasma level is an independent predictor of achieving SVR. In addition, the optimal RBV steady-state plasma level therapeutic range that balances SVR with risk of anaemia was found to be 2.28-3.61 mg/L.
Despite the great successes of DAAs, some patients remain difficult-to-cure.[29] Adding RBV to DAAs in these patients can increase SVR rates.[8, 30] Studies in the interferon era established that higher RBV steady-state plasma levels were positively correlated with SVR.[12, 23, 31, 32] We similarly found that higher RBV steady-state plasma levels led to a 1.8-fold increase in SVR in DAA-treated patients. This demonstrates that higher RBV plasma levels are effective to improve response rates, particularly in difficult-to-cure patients with a lower a priori chance of achieving SVR. Possibly, therapeutic drug monitoring can help achieving these high RBV plasma levels.
A recent study including patients treated with sofosbuvir/RBV found no association between RBV steady-state plasma levels and SVR.[13] However, this study had a small sample size of 47 treatment naive, predominantly genotype 1-infected, noncirrhotic patients and retrospectively analysed RBV plasma samples. Moreover, SVR rates were low (55%), probably because sofosbuvir/RBV is a suboptimal treatment for genotype 1,[17, 18] and these results should, therefore, be carefully interpreted. Our study, however, represents a large real-world cohort of mainly difficult-to-cure patients on various DAA combinations with prospectively analysed plasma samples and SVR rates similar to clinical trials and real-world cohorts.[17, 18]
In case the addition of RBV is considered beneficial and is administered, it is important to balance efficacy and side effects. Although side effects of RBV in combination with DAAs are generally considered very mild, almost one in five of our patients developed significant anaemia. This study found an optimal RBV steady-state plasma level therapeutic range of 2.28-3.61 mg/L. Interestingly, this study as well as studies with triple-therapy including a first-generation DAA[32, 33] found somewhat higher therapeutic ranges compared with studies done with PEG-IFN/RBV alone.[23, 34, 35] The most likely explanation for this is that PEG-IFN contributes to anaemia through bone marrow suppression and as such augments the toxicity of RBV at any given dose/concentration. Another explanation for this could be that with the excellent safety profile of DAAs, sicker patients than those in the PEG-IFN/RBV era can now be treated.[36] A significant proportion of our patients had (decompensated) cirrhosis, were liver transplant recipients and/or had renal impairment, which could result in higher RBV levels. Finally, to significantly increase the likelihood of achieving SVR in the context of the already highly effective DAAs, a more pronounced increase in RBV exposure is necessary.
Still, the question remains if and how RBV therapeutic drug monitoring should be implemented in the treatment of patients without difficult-to-cure characteristics. With current DAA regimens lasting only 12 weeks,[17, 18] when reaching RBV steady-state plasma level, only 4 weeks of treatment are left to adjust RBV dosage. Thus, for most patients, RBV exposure could perhaps best be monitored based on toxicity instead of RBV steady-state plasma levels. However, retreatment of patients who fail therapy is considerably more expensive than therapeutic drug monitoring, and reaching adequate RBV plasma levels is not a certainty when relying on toxicity alone. Comorbidities, hepatic and/or renal dysfunction and ITPA gene polymorphism[37, 38] can affect RBV pharmacokinetics and thus RBV steady-state plasma levels without evident toxicity.[39] For example, if a patient has a stable haemoglobin level, it is uncertain whether RBV steady-state plasma level will be in or below the therapeutic range. Assessment of RBV plasma level at week 8 of treatment (RBV steady-state plasma level) may be too late for intervention. An option for therapeutic drug monitoring and intervention at an earlier stage is measuring RBV plasma levels at treatment week 2. Van Vlerken et al reported that a week 2 RBV level of ≥1.29 mg/L predicted adequate RBV steady-state plasma levels at week 8 in patients treated with PEG-IFN/RBV.[23] Future studies are needed to determine if this also applies to patients treated with DAAs.
Especially in patients with impaired renal function, therapeutic drug monitoring is important. This study found that despite the Food and Drug Administration and European Medicines Agency recommending RBV dose reduction in patients with eGFR <50 mL/min,[40, 41] they still received usual weight-based start dosages. As a result, these patients required more dose adjustments, resulting in lower average RBV doses, but still higher RBV steady-state plasma levels and high anaemia rates. This study indicates that patients with a renal function <90 mL/min (especially those ≤50 mL/min) should be monitored closely during treatment, so that RBV can be timely adjusted when necessary.
To the best of our knowledge, this is the largest prospective study analysing the association between RBV steady-state plasma level and SVR in the current DAA era. Nevertheless, some limitations are present. First, all four participating centres used different assays to measure plasma RBV. However, we expect the influence of the different assays to be very limited, since all assays are validated according to international standards and have undergone strict evaluation and comparison in the Dutch quality control programme. In addition, logistic regression analyses did not show a centre effect. Second, due to the real-life nature of the study, RBV dose adjustment and selection of DAA regimen were at the discretion of the treating physician, thus leading to variation. Nonetheless, DAA regimens were selected according to international and national guidelines. In hindsight, while previously recommended by guidelines,[42] sofosbuvir/RBV is now considered a suboptimal treatment for genotype 3 patients with cirrhosis, which has led to a high proportion of relapsers within this population. Therefore, DAA regimen and HCV genotype were excluded in our logistic regression model. However, even in sensitivity analyses where DAA regimen, HCV genotype (as a combined variable or as separate variables) or both variables were added to the model, RBV steady-state plasma level still remained an independent predictor of SVR. For dose adjustments and other factors that could have influenced RBV exposure, we tried to account by adjusting for average RBV dose during the entire treatment, presence of cirrhosis and creatinine levels in our logistic regression models. Lastly, influence of ITPA gene polymorphism was not assessed.
In conclusion, a higher RBV steady-state plasma level is an independent predictor of SVR in patients treated with DAAs.
In this prospective study, the influence of RBV steady-state plasma level on achieving SVR in HCV-infected patients treated with the combination of DAAs and RBV was determined. The main finding of this study is that a higher RBV steady-state plasma level is an independent predictor of achieving SVR. In addition, the optimal RBV steady-state plasma level therapeutic range that balances SVR with risk of anaemia was found to be 2.28-3.61 mg/L.
Despite the great successes of DAAs, some patients remain difficult-to-cure.[29] Adding RBV to DAAs in these patients can increase SVR rates.[8, 30] Studies in the interferon era established that higher RBV steady-state plasma levels were positively correlated with SVR.[12, 23, 31, 32] We similarly found that higher RBV steady-state plasma levels led to a 1.8-fold increase in SVR in DAA-treated patients. This demonstrates that higher RBV plasma levels are effective to improve response rates, particularly in difficult-to-cure patients with a lower a priori chance of achieving SVR. Possibly, therapeutic drug monitoring can help achieving these high RBV plasma levels.
A recent study including patients treated with sofosbuvir/RBV found no association between RBV steady-state plasma levels and SVR.[13] However, this study had a small sample size of 47 treatment naive, predominantly genotype 1-infected, noncirrhotic patients and retrospectively analysed RBV plasma samples. Moreover, SVR rates were low (55%), probably because sofosbuvir/RBV is a suboptimal treatment for genotype 1,[17, 18] and these results should, therefore, be carefully interpreted. Our study, however, represents a large real-world cohort of mainly difficult-to-cure patients on various DAA combinations with prospectively analysed plasma samples and SVR rates similar to clinical trials and real-world cohorts.[17, 18]
In case the addition of RBV is considered beneficial and is administered, it is important to balance efficacy and side effects. Although side effects of RBV in combination with DAAs are generally considered very mild, almost one in five of our patients developed significant anaemia. This study found an optimal RBV steady-state plasma level therapeutic range of 2.28-3.61 mg/L. Interestingly, this study as well as studies with triple-therapy including a first-generation DAA[32, 33] found somewhat higher therapeutic ranges compared with studies done with PEG-IFN/RBV alone.[23, 34, 35] The most likely explanation for this is that PEG-IFN contributes to anaemia through bone marrow suppression and as such augments the toxicity of RBV at any given dose/concentration. Another explanation for this could be that with the excellent safety profile of DAAs, sicker patients than those in the PEG-IFN/RBV era can now be treated.[36] A significant proportion of our patients had (decompensated) cirrhosis, were liver transplant recipients and/or had renal impairment, which could result in higher RBV levels. Finally, to significantly increase the likelihood of achieving SVR in the context of the already highly effective DAAs, a more pronounced increase in RBV exposure is necessary.
Still, the question remains if and how RBV therapeutic drug monitoring should be implemented in the treatment of patients without difficult-to-cure characteristics. With current DAA regimens lasting only 12 weeks,[17, 18] when reaching RBV steady-state plasma level, only 4 weeks of treatment are left to adjust RBV dosage. Thus, for most patients, RBV exposure could perhaps best be monitored based on toxicity instead of RBV steady-state plasma levels. However, retreatment of patients who fail therapy is considerably more expensive than therapeutic drug monitoring, and reaching adequate RBV plasma levels is not a certainty when relying on toxicity alone. Comorbidities, hepatic and/or renal dysfunction and ITPA gene polymorphism[37, 38] can affect RBV pharmacokinetics and thus RBV steady-state plasma levels without evident toxicity.[39] For example, if a patient has a stable haemoglobin level, it is uncertain whether RBV steady-state plasma level will be in or below the therapeutic range. Assessment of RBV plasma level at week 8 of treatment (RBV steady-state plasma level) may be too late for intervention. An option for therapeutic drug monitoring and intervention at an earlier stage is measuring RBV plasma levels at treatment week 2. Van Vlerken et al reported that a week 2 RBV level of ≥1.29 mg/L predicted adequate RBV steady-state plasma levels at week 8 in patients treated with PEG-IFN/RBV.[23] Future studies are needed to determine if this also applies to patients treated with DAAs.
Especially in patients with impaired renal function, therapeutic drug monitoring is important. This study found that despite the Food and Drug Administration and European Medicines Agency recommending RBV dose reduction in patients with eGFR <50 mL/min,[40, 41] they still received usual weight-based start dosages. As a result, these patients required more dose adjustments, resulting in lower average RBV doses, but still higher RBV steady-state plasma levels and high anaemia rates. This study indicates that patients with a renal function <90 mL/min (especially those ≤50 mL/min) should be monitored closely during treatment, so that RBV can be timely adjusted when necessary.
To the best of our knowledge, this is the largest prospective study analysing the association between RBV steady-state plasma level and SVR in the current DAA era. Nevertheless, some limitations are present. First, all four participating centres used different assays to measure plasma RBV. However, we expect the influence of the different assays to be very limited, since all assays are validated according to international standards and have undergone strict evaluation and comparison in the Dutch quality control programme. In addition, logistic regression analyses did not show a centre effect. Second, due to the real-life nature of the study, RBV dose adjustment and selection of DAA regimen were at the discretion of the treating physician, thus leading to variation. Nonetheless, DAA regimens were selected according to international and national guidelines. In hindsight, while previously recommended by guidelines,[42] sofosbuvir/RBV is now considered a suboptimal treatment for genotype 3 patients with cirrhosis, which has led to a high proportion of relapsers within this population. Therefore, DAA regimen and HCV genotype were excluded in our logistic regression model. However, even in sensitivity analyses where DAA regimen, HCV genotype (as a combined variable or as separate variables) or both variables were added to the model, RBV steady-state plasma level still remained an independent predictor of SVR. For dose adjustments and other factors that could have influenced RBV exposure, we tried to account by adjusting for average RBV dose during the entire treatment, presence of cirrhosis and creatinine levels in our logistic regression models. Lastly, influence of ITPA gene polymorphism was not assessed.
In conclusion, a higher RBV steady-state plasma level is an independent predictor of SVR in patients treated with DAAs.
Full Text
No comments:
Post a Comment