Abstract
Background/Aim: The antiviral agent ritonavir is a substrate for cytochrome P450 3A4 (CYP3A4); therefore, concomitant use of CYP3A4-metabolising drugs might cause adverse reactions to this drug. We investigated the plasma level of calcium channel blockers (CCBs) as CYP3A4 substrates and peripheral edema as a potential adverse drug reaction possibly caused by the anti-hepatitis C virus (HCV) regimen of ombitasvir/paritaprevir/ritonavir (OPR) and CCBs. Patients and Methods: We enrolled Japanese patients prescribed OPR for HCV infection. Peripheral edema was graded according to the Common Terminology Criteria for Adverse Events ver. 4. Plasma samples were collected on days 0, 7, 14, 28, and 42 after antiviral treatment, at the trough level. Results: Out of 52 patients, 64% experienced grade 1 or grade 2 peripheral edema, but not grade 3. Concomitant use of CCBs significantly increased the emergence of grade 2 edema (62%), compared with patients treated solely with OPR (48%). The use of OPR significantly increased the plasma concentration of amlodipine. Conclusion: Peripheral edema in patients treated with OPR and CCBs, although tolerable, should be closely monitored.
The hepatitis C virus (HCV) is a major cause of chronic liver disease. New oral direct-acting antiviral agents are marketed as interferon-free therapies and are highly effective for treating HCV (1–7). Oral direct-acting antiviral medications for HCV are generally well tolerated and cause few adverse drug reactions (ADRs) (8).
Among the direct-acting antiviral agents, the three-drug combination (hereafter referred to as OPR) contains ombitasvir (a nonstructural protein 5A inhibitor), paritaprevir (a nonstructural protein 3/4A protease inhibitor), and ritonavir. Ritonavir inhibits cytochrome P450 3A4 (CYP3A4) to allow higher peak and trough concentrations of paritaprevir since paritaprevir is mainly metabolized by CYP3A4 (9). However, the inhibition of CYP3A4 by ritonavir can cause drug–drug interactions with other CYP3A4-metabolized drugs such as calcium channel blockers (CCBs) (9, 10).
Approximately 5% of Japanese patients administered OPR experienced peripheral edema. It should be noted that all the patients who experienced peripheral edema were also administered CCBs (1, 11). Furthermore, co-administration of protease inhibitors and CCBs increased the plasma levels of CCBs (9, 10, 12). These findings suggest that CYP3A4 inhibition by ritonavir might elevate the plasma CCB level, potentially worsening peripheral edema since both OPR and CCBs cause edema (6, 11, 13). CCBs are widely used for treating hypertension and are frequently administered in combination with other drugs. Considering that most patients with HCV are elderly (14), we anticipate that elderly patients with HCV who are prescribed OPR may also already be taking CCBs to treat hypertension.
We investigated the occurrence of peripheral edema as an ADR in Japanese patients taking OPR, with or without CCBs. We assessed the timing of symptom onset, the rate and severity of peripheral edema, and measured plasma concentrations of amlodipine, a CCB. Our study objective was to assess the effects of concomitant use of CCBs and OPR on ADRs and the plasma level of CCBs.
Patients and Methods
Study design. Fifty-five elderly Japanese patients with HCV infection who were receiving 12 weeks of OPR treatment were enrolled at Gifu Municipal Hospital between December 2015 and December 2016. Exclusion criteria were: Already experiencing administration of OPR; having similar study symptoms caused by factors other than taking CCBs. All patients provided written informed consent, and the study was approved by the Ethics Committees of Gifu Pharmaceutical University (no. 27-19) and Gifu Municipal Hospital (no. 301).
In order to investigate the association of concomitant drugs with edema, we checked the use of drugs suggested to be associated with edema immediately before OPR was administered. Among the candidate drugs, CCBs were used in many patients and considered to be more influenced by OPR due to competition for CYP3A4. Therefore, patients were divided into two groups, namely those with and those without co-administration of CCBs. The decision to treat with CCBs was made by the physician rather than according to a trial protocol. There were no patients who stopped CCBs administration throughout the observation period.
Safety assessment. Peripheral edema was evaluated by measuring changes in ankle diameter according to the Common Terminology Criteria for Adverse Events (CTCAE version 4.0) (15) in all patients, before starting OPR treatment and at the time of hospital visits. The observation period in which patients were treated with OPR ranged from 84 to 91 days. Additionally, we confirmed concomitant use of CCBs or other medications for all patients. CCB dosing and discontinuation were at the discretion of the treating physician. Clinical data, including age, sex, body weight, and laboratory values, such as levels of albumin, aspartate aminotransferase, alanine aminotransferase, total bilirubin, total protein, creatinine, and HCV RNA log10, were also recorded. Other possible ADRs were checked by review of nursing records.
Sample collection. Blood samples were collected from all patients on days 0, 7, 14, 28 and 42, at trough levels. Collected blood samples were centrifuged; then the resultant plasma was stored at –30°C until analysis. We measured the plasma concentration of amlodipine, since amlodipine was the most frequently prescribed CCB.
Determination of amlodipine. The amlodipine level was determined using liquid chromatography–tandem mass spectrometry (LC-MS/MS) as previously reported (16) with slight modifications. Briefly, the prepared samples were separated using a Cadenza HS-C18 column (50×2 mm, 3 μm; Imtakt Corporation, Kyoto, Japan) with a mobile phase composed of 80% acetonitrile containing 0.1% formic acid. The transitions (m/z: precursor ion → product ion) for multiple reaction monitoring were 409.00 → 238.05 for amlodipine, and 267.00 → 72.20 for desipramine as the internal standard. Amlodipine was extracted from 200 μl of plasma samples using solid-phase extraction systems according to the manufacturer’s protocols (Oasis HLB columns; Waters Corporation, Milford, MA, USA). The residue after desiccation under a stream of nitrogen was reconstituted in 100 μl of mobile phase and vortexed for 15 s. Subsequently, diluents were passed through Millex®-LG filters (0.2 μm, 4 mm; Merck, Darmstadt, Germany), and 10 μl filtered samples were injected into the LC-MS/MS system.
The retention times of amlodipine and the internal standard were both 1.8 min, and the total run time was 3 min. The calibration curves were highly linear (r2 >0.99), from 0.5 to 15.0 ng/ml. Subsequent full validation according to the Food and Drug Administration guidance for biological method validation (17) of recovery, intra- and inter-day precision and accuracy, matrix effects, and freeze/thaw stability demonstrated that the data were highly reproducible (Supplementary Table I and Supplementary Table II).
Statistical analysis. Comparisons of continuous data between groups were performed using Student’s t-test after confirming the normal distribution of the data. We compared the proportion of emerging cases of peripheral edema and of concomitant drug use in Japanese patients, both with and without co-administration of CCBs, using Fisher’s exact probability test. We compared the severity (defined as CTCAE grade >2) and onset of peripheral edema between the two groups using Cox proportional hazards regression models. Statistical significance was determined at p<0.05. All statistical analyses were performed using SPSS software (IBM Corp., Armonk, NY, USA).
Plasma amlodipine concentrations and their percentage elevation are presented as the mean±standard deviation. Comparisons between day 0 and day 7 or later were performed using Student’s t-test, and p<0.05 was considered statistically significant.
Results
Patient characteristics. Fifty-five patients were enrolled. However, in the study cohort, we utilized 52 patients consisting of 25 male and 27 female patients, with an average age of 70 years (range=45-84 years) since three patients were excluded because they discontinued OPR treatment due to i) an enlarged aortic aneurysm, ii) incremental increases in blood pressure with deteriorating liver function, and iii) non-compliance. Table I shows the patient characteristics in patients who received both OPR and CCBs (n=21) and those who received OPR but not CCBs (n=31). Among CCBs, amlodipine was the most prescribed (n=14); diltiazem (n=2) and nifedipine (n=2) were the second most prescribed. Benidipine, cilnidipine, and nilvadipine were also prescribed (n=1 each).
Patient characteristics.
In the CCB group, there were no significant differences in laboratory values, such as the level of albumin, between day 0 (before concomitant use of CCBs) and day 14 (after concomitant use of CCBs) or later (data not shown). Only one patient underwent CCB dose reduction (amlodipine: 5 to 2.5 mg) due to concerns over dizziness. Only one patient discontinued CCBs (amlodipine and benidipine) secondary to their physician’s order. This patient exhibited mild grade 2, but not grade 3, edema. All other patients continued CCB medication without dose adjustment, even though grade 1 or 2 edema occurred. Moreover, no patients experienced significant fluctuations in blood pressure or other serious ADRs during the study period. Regarding comorbidities, candidates that may cause peripheral edema were compared between the two groups. As shown in Table II, no significant differences were observed in candidates except for hypertension. Moreover, although liver fibrosis is one factor that can cause peripheral edema (18), there were no significant differences in fibrosis stage between the CCB and non-CCB groups (Table I).
Frequency of comorbidities that may cause peripheral edema (17) in patients treated with and without calcium channel blockers (CCBs) whilst under treatment for hepatitis C virus infection with ombitasvir/paritaprevir/ritonavir.
It is well-known that some drugs, such as antihypertensive drugs, nonsteroidal anti-inflammatory drugs and corticosteroids used in this study possibly cause peripheral edema (18). The number of patients who used nonsteroidal anti-inflammatory drugs (n=5) or corticosteroids (n=1) were few and negligible. However, the number of patients using other antihypertensive drugs in the CCB group (n=17) was significantly higher than that in the non-CCB group (n=5) (p<0.01, Fisher’s exact probability test). In detail, we found the use of antihypertensive drugs (angiotensin-converting-enzyme inhibitors/angiotensin receptor blockers) to be significantly more frequent in the CCB group (13 vs. 4, p<0.01); as was the use of α-/β-blockers: 6 vs. 0, p<0.01; and diuretics: 6 vs. 2, p<0.01, respectively; Fisher’s exact probability test). No significant difference between groups in the number of patients for other prescribed drugs was observed (data not shown).
The assessment of peripheral edema. None of the patients had experienced edema (defined as grade >1) before OPR treatment, whilst overall, 28 patients (54%) experienced peripheral edema during OPR treatment (Table III). In 10 out of the 11 patients experiencing grade 2 edema, the edema decreased to grade 1 or disappeared. Throughout the observation period, serious edema (i.e., grade >3) was not found during OPR treatment. No significant associations were observed between the occurrence of peripheral edema and fibrosis (p=0.958, Mann–Whitney U-test).
The incidence of peripheral edema in patients treated with and without calcium channel blockers (CCBs) whilst under treatment for hepatitis C virus infection with ombitasvir/paritaprevir/ritonavir.
Regarding the incidence of peripheral edema (Table III), a total of 62% of patients in the CCB group and 48% of patients in the non-CCB group experienced peripheral edema (grade 1 or 2) but this was not significantly different (p=0.403). When examined by the severity of peripheral edema, patients in the CCB group significantly more frequently exhibited grade 2 edema (38%) than their non-CCB counterparts (10%) (p=0.019, Fisher’s exact probability test). As described above, other antihypertensive drugs were frequently used in the CCB group. However, the use of other antihypertensive drugs did not affect the incidence of grade 2 edema (Table IV).
Relationship between the incidence of edema and the use of other antihypertensive drugs in patients under treatment for hepatitis C virus infection with ombitasvir/paritaprevir/ritonavir.
The day of onset of peripheral edema is shown in Figure 1. The median time until experiencing peripheral edema considering all grades was 84 days (interquartile range=37.5-84.5 days) in the non-CCB group and 48 days (interquartile range=16-85 days) in the CCB group, respectively (Figure 1A). There was no between-group difference in the onset day of all-grade peripheral edema after adjusting for age (p=0.247, Figure 1A). For the onset of grade 2 edema, after adjusting for age, the time to onset was significantly shorter for the CCB group (p=0.036, Figure 1B).
Time to onset of peripheral all-grade (A) and grade 2 (B) edema via age-adjusted analysis in patients treated with and without calcium channel blockers (CCBs) whilst under treatment for hepatitis C virus infection with ombitasvir/paritaprevir/ritonavir. There was significant between-group difference in the onset day of grade 2 edema (p=0.036), but not of all-grade edema (p=0.247).
Plasma concentration of amlodipine. As shown in Figure 2A and B, the amlodipine levels were significantly increased on day 7 (203±59%) and on day 14 (245±93%) compared with day 0. This trend continued up to 42 days.
Plasma concentrations of amlodipine in patients treated with ombitasvir/paritaprevir/ritonavir for hepatitis C virus infection. Data represent the mean±standard deviation and are expressed as raw data (A) and as a percentage of that on day 0 (B). The average plasma concentration of amlodipine at day 0 was 10.5±5.7 ng/ml. **Significantly different at p<0.01 vs. day 0 (Student’s t-test).
Discussion
We investigated the occurrence of peripheral edema and the use of CCBs in elderly Japanese patients treated with OPR. OPR administration elevates the levels of plasma CCBs, leading to peripheral edema as an ADR. To our knowledge, this is the first report in Japanese patients with HCV to show the apparent relationship between CYP3A4 inhibition by OPR and a consequent ADR as a drug–drug interaction. In the CCB group, other antihypertensive drugs were frequently used. However, the possibility peripheral edema was caused by these antihypertensive drugs was excluded since no statistical significance between the CCB group and non-CCB group was observed regarding the relationship between the use of these antihypertensive drugs and the incidence of peripheral edema.
In the present study, we found the incidence of mild (grade 2) edema was significantly higher in the CCB group. This result is partially consistent with the GIFT-I study, conducted with Japanese patients (1, 6, 11). In that study, the researchers found that all patients who exhibited peripheral edema had been administered CCBs (1, 11). Another clinical trial in Japan found that fluid retention was the most frequently occurring ADR among patients taking CCBs (6). Moreover, a subgroup analysis of the GIFT-I study found that the incidence of peripheral edema was significantly lower in patients who took the lowest CCB doses (1, 11). We also found a reduced incidence of peripheral edema in patients taking lower doses of CCBs, compared with all patients taking all doses of CCBs [lowest dose, 1/7 cases (14%); other dose, 12 out of 14 cases (86%), p<0.01, Fisher’s exact probability test]. Thus, reducing the dosage of CCBs might be advisable in patients undergoing OPR treatment.
We found that the level of the CCB amlodipine significantly increased within 1 week of initiating OPR treatment. This may have occurred because ritonavir is a CYP3A4 inhibitor, and plasma amlodipine concentrations increase following ritonavir administration (9, 10). Concomitant use of CCBs and CYP3A4 inhibitors such as ritonavir should be carefully monitored. Drug information for OPR also cautions against the concomitant use of CYP3A4 substrates, due to inhibition of CYP3A4 by ritonavir (6).
Importantly, for 12 weeks, no patients discontinued OPR treatment, consistent with the GIFT-I study (1, 11). Dose reduction or discontinuation of CCBs was observed in one patient each. However, in both cases, these changes were not because of ADRs. Moreover, and consistent with the GIFT-I study (1, 11), no patients experienced grade 3 or higher edema. However, 54% of patients (regardless of CCB administration, or not) experienced grade 1 (33%) or 2 (21%) peripheral edema. This rate differs from several other reports (1, 11, 19, 20). In the GIFT-I study, only 5.4% of patients experienced peripheral edema (3.9% at grade 1 and 1.6% at grade 2) (11). In a study to evaluate the safety and effect of the experimental drugs ABT-450/ritonavir/ABT-267 and ABT-333 in patients with chronic HCV infection (PEARL-II), the most common ADRs associated with OPR/dasabuvir (the so-called 3D regimen) were headache (23.2%) and fatigue (15.8%), on the other hand, peripheral edema was not reported as a common ADR (present in ≤10%) (19). Moreover, in the AMBER multicentre, open-label, investigator-initiated study in Poland to investigate the effectiveness and safety of OPR ± dasabuvir ± ribavirin under ‘real-world’ conditions, mild peripheral edema occurred in 11.3% of patients taking the 3D regimen, without CCBs, although the most common ADR was headache (18.9%) (20). Moreover, they also demonstrated that the instances of edema were not serious (20). These studies suggested that there were few serious ADRs associated with the 3D regimen, and that peripheral edema was not always closely related to administration of the 3D regimen (11).
There are several possible reasons for the higher rate of peripheral edema in the present study compared to the previous reports (1, 11, 19, 20). Firstly, compared to these previous studies (1, 11, 19), our patients were older, and had hypoalbuminemia and mild hepatic/renal dysfunction. In the AMBER study, the patients were relatively young, although their hepatic function was worse than that observed in the present study (20). Aging, hypoalbuminemia, and mild hepatic/renal dysfunction may all contribute to the onset of peripheral edema (21–23). Thus, our patients were, by definition, at high-risk for peripheral edema. Secondary conditions such as hypoalbuminemia and mild hepatic/renal dysfunction might alter the pharmacokinetic properties of ombitasvir, paritaprevir, or ritonavir, potentially causing peripheral edema.
Indeed, the area under the concentration–time curve (AUC) of paritaprevir significantly was shown to be increased in patients with liver dysfunction (6), whereas that of ritonavir significantly increased in patients with renal impairments (6). Pharmacokinetically, ombitasvir shows few changes in patients with liver dysfunction or renal impairment (6). Ombitasvir, paritaprevir and ritonavir bind at high rates to plasma protein (99.9%, 97-98.6%, and more than 99%, respectively) (6). Thus, hypoalbuminemia might also increase the effect of these drugs by increasing their free fraction. Considering the low risk (<2%) of peripheral edema secondary to administration of ritonavir alone (24), it is possible that peripheral edema can be caused by paritaprevir owing to the increase in the AUC. Therefore, it is likely that, among the three drugs, paritaprevir was the most likely to cause peripheral edema, although no evidence examining solo use of paritaprevir is available.
This study has some limitations. Compared with the non-CCB group, patients in the CCB group were significantly older and had significantly higher creatinine levels. As mentioned above, the AUC of ritonavir was increased in patients with renal dysfunction (6) although amlodipine levels were shown to be unaltered in such patients (13). Significant differences were still noted in the occurrence of grade 2 edema between the two groups after adjusting for age. These results suggest that the use of CCBs is a factor in increasing the incidence rate of edema, although these between-group differences in patients’ backgrounds may also affect the onset and the severity of peripheral edema.
Finally, we note that our patients generally exhibited mild edema (grade 1 or 2). Consequently, no patients taking CCBs discontinued the 12-week OPR treatment because of peripheral edema. Thus, it is not necessary for patients to discontinue amlodipine while they are treated with OPR.
In conclusion, OPR worsened peripheral edema in elderly Japanese who took CCBs. Elevation of the amlodipine level by ritonavir appeared to be involved in this deterioration. However, CCBs did not markedly worsen peripheral edema, and no patients discontinued treatment secondary to peripheral edema. Thus, appropriate precautions may be helpful for treating HCV with OPR, particularly in Japanese patients concomitantly treated with CCBs.
Footnotes
↵* These Authors contributed equally to this study.
Authors’ Contributions
YS and MS equally contributed as first authors. AH, TM, CG, and KK designed the study. AH collected the clinical data. YN and ET contributed to clinical analysis, and YS, YM, and MS measured plasma concentrations of amlodipine. YS, MS, and KK performed statistical analyses and drafted the article. All Authors read and approved the final article.
Supplementary Material
Available at: <https://drive.google.com/file/d/1r2tovApHC2wHRrLU2wgU7e2t0GV1LpVK/view>
Conflicts of Interest
The Authors declare that they have no conflicts of interests in relation to this study.
- Received January 14, 2022.
- Revision received February 16, 2022.
- Accepted February 21, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.







