"An expert panel has recommended expanded adoption of the drug hydroxyurea for the care of people with sickle cell disease, according to a report issued today. The report also suggests that clinicians give periodic blood transfusions to children w"...
Mechanism Of Action
UCDs are inherited deficiencies of enzymes or transporters necessary for the synthesis of urea from ammonia (NH3, NH4+). Absence of these enzymes or transporters results in the accumulation of toxic levels of ammonia in the blood and brain of affected patients. RAVICTI is a triglyceride containing 3 molecules of phenylbutyrate (PBA). PAA, the major metabolite of PBA, is the active moiety of RAVICTI. PAA conjugates with glutamine (which contains 2 molecules of nitrogen) via acetylation in the liver and kidneys to form PAGN, which is excreted by the kidneys (Figure 1). On a molar basis, PAGN, like urea, contains 2 moles of nitrogen and provides an alternate vehicle for waste nitrogen excretion.
Figure 1: RAVICTI Mechanism of Action
In clinical studies, total 24-hour AUC of ammonia concentration was comparable at steady state during the switchover period between RAVICTI and sodium phenylbutyrate [see Clinical Studies].
The effect of multiple doses of RAVICTI 13.2 g/day and 19.8 g/day (approximately 69% and 104% of the maximum recommended daily dosage) on QTc interval was evaluated in a randomized, placebo- and active-controlled (moxifloxacin 400 mg), four-treatment-arm, crossover study in 57 healthy subjects. The upper bound of the one-sided 95% CI for the largest placebo-adjusted, baseline-corrected QTc, based on individual correction method (QTcI) for RAVICTI, was below 10 ms. However, assay sensitivity was not established in this study because the moxifloxacin time-profile was not consistent with expectation. Therefore, an increase in mean QTc interval of 10 ms cannot be ruled out.
RAVICTI is a pro-drug of PBA. Upon oral ingestion, PBA is released from the glycerol backbone in the gastrointestinal tract by lipases. PBA derived from RAVICTI is further converted by β-oxidation to PAA.
In healthy, fasting adult subjects receiving a single oral dose of 2.9 mL/m² of RAVICTI, peak plasma levels of PBA, PAA, and PAGN occurred at 2 h, 4 h, and 4 h, respectively. Upon single-dose administration of RAVICTI, plasma concentrations of PBA were quantifiable in 15 of 22 participants at the first sample time postdose (0.25 h). Mean maximum concentration (Cmax) for PBA, PAA, and PAGN was 37.0 μg/mL, 14.9 μg/mL, and 30.2 μg/mL, respectively. In healthy subjects, intact glycerol phenylbutyrate was detected in plasma. While the study was inconclusive, the incomplete hydrolysis of glycerol phenylbutyrate cannot be ruled out.
In healthy subjects, the systemic exposure to PAA, PBA, and PAGN increased in a dose-dependent manner. Following 4 mL of RAVICTI for 3 days (3 times a day [TID]), mean Cmax and AUC were 66 μg/mL and 930 μg•h/mL for PBA and 28 μg/mL and 942 μg•h/mL for PAA, respectively. In the same study, following 6 mL of RAVICTI for 3 days (TID), mean Cmax and AUC were 100μg/mL and 1400 μg•h/mL for PBA and 65 μg/mL and 2064 μg•h/mL for PAA, respectively.
In adult UCD patients receiving multiple doses of RAVICTI, maximum plasma concentrations at steady state (Cmaxss) of PBA, PAA, and PAGN occurred at 8 h, 12 h, and 10 h, respectively, after the first dose in the day. Intact glycerol phenylbutyrate was not detectable in plasma in UCD patients.
In vitro, the extent of plasma protein binding for 14C-labeled metabolites was 80.6% to 98.0% for PBA (over 1-250 μg/mL), and 37.1% to 65.6% for PAA (over 5-500 μg/mL). The protein binding for PAGN was 7% to 12% and no concentration effects were noted.
Upon oral administration, pancreatic lipases hydrolyze RAVICTI (i.e., glycerol phenylbutyrate), and release PBA. PBA undergoes β-oxidation to PAA, which is conjugated with glutamine in the liver and in the kidney through the enzyme phenylacetyl-CoA: L-glutamine-N-acetyltransferase to form PAGN. PAGN is subsequently eliminated in the urine.
Saturation of conjugation of PAA and glutamine to form PAGN was suggested by increases in the ratio of plasma PAA to PAGN with increasing dose and with increasing severity of hepatic impairment.
In healthy subjects, after administration of 4 mL, 6 mL, and 9 mL 3 times daily for 3 days, the ratio of mean AUC0-23h of PAA to PAGN was 1, 1.25, and 1.6, respectively. In a separate study, in patients with hepatic impairment (Child-Pugh B and C), the ratios of mean Cmax values for PAA to PAGN among all patients dosed with 6 mL and 9 mL twice daily were 3 and 3.7.
In in vitro studies, the specific activity of lipases for glycerol phenylbutyrate was in the following decreasing order: pancreatic triglyceride lipase, carboxyl ester lipase, and pancreatic lipase–related protein 2. Further, glycerol phenylbutyrate was hydrolyzed in vitro by esterases in human plasma. In these in vitro studies, a complete disappearance of glycerol phenylbutyrate did not produce molar equivalent PBA, suggesting the formation of mono- or bis-ester metabolites. However, the formation of mono- or bis-esters was not studied in humans.
The mean (SD) percentage of administered PBA excreted as PAGN was approximately 68.9% (17.2) in adults and 66.4% (23.9) in pediatric UCD patients at steady state. PAA and PBA represented minor urinary metabolites, each accounting for < 1% of the administered dose of PBA.
In healthy adult volunteers, a gender effect was found for all metabolites, with women generally having higher plasma concentrations of all metabolites than men at a given dose level. In healthy female volunteers, mean Cmax for PAA was 51 and 120% higher than in male volunteers after administration of 4 mL and 6 mL 3 times daily for 3 days, respectively. The dose normalized mean AUC0-23h for PAA was 108% higher in females than in males.
Population pharmacokinetic modeling and dosing simulations suggest body surface area to be the most significant covariate explaining the variability of PAA clearance. PAA clearance was 10.9 L/h, 16.4 L/h, and 24.4 L/h, respectively, for UCD patients ages 3 to 5, 6 to 11, and 12 to 17 years.
The effects of hepatic impairment on the pharmacokinetics of RAVICTI were studied in patients with hepatic impairment of Child-Pugh A, B, and C receiving 100 mg/kg of RAVICTI twice daily for 7 days.
Plasma glycerol phenylbutyrate was not measured in patients with hepatic impairment.
After multiple doses of RAVICTI in patients with hepatic impairment of Child-Pugh A, B, and C, geometric mean AUCt of PBA was 42%, 84%, and 50% higher, respectively, while geometric mean AUCt of PAA was 22%, 53%, and 94% higher, respectively, than in healthy subjects.
In patients with hepatic impairment of Child-Pugh A, B, and C, geometric mean AUCt of PAGN was 42%, 27%, and 22% lower, respectively, than that in healthy subjects.
The proportion of PBA excreted as PAGN in the urine in Child-Pugh A, B, and C was 80%, 58%, and 85%, respectively, and, in healthy volunteers, was 67%.
In another study in patients with hepatic impairment (Child-Pugh B and C), mean Cmax of PAA was 144 μg/mL (range: 14-358 μg/mL) after daily dosing of 6 mL of RAVICTI twice daily, while mean Cmax of PAA was 292 μg/mL (range: 57-655 μg/mL) after daily dosing of 9 mL of RAVICTI twice daily. The ratio of mean Cmax values for PAA to PAGN among all patients dosed with 6 mL and 9 mL twice daily were 3 and 3.7, respectively.
After multiple doses, a PAA concentration > 200 μg/mL was associated with a ratio of plasma PAA to PAGN concentrations higher than 2.5.
In vitro studies using human liver microsomes showed that the principle metabolite, phenylbutyrate, at a concentration of 800 μg/mL caused > 60% reversible inhibition of cytochrome P450 isoenzymes CYP2C9, CYP2D6, and CYP3A4/5 (testosterone 6β-hydroxylase activity). Subsequent in vitro studies suggest that in vivo drug interactions with substrates of CYP2C9, CYP2D6, and CYP3A4/5 are possible. No in vivo drug interaction studies were conducted. The inhibition of CYP isoenzymes 1A2, 2C8, 2C19, and 2D6 by PAA at the concentration of 2.8 mg/mL was observed in vitro. Clinical implication of these results is unknown.
In vitro PBA or PAA did not induce CYP1A2 and CYP3A4, suggesting that in vivo drug interactions via induction of CYP1A2 and CYP3A4 are unlikely.
Clinical Studies In Adult Patients With UCDs
Active-Controlled, 4-Week, Noninferiority Study (Study 1)
A randomized, double-blind, active-controlled, crossover, noninferiority study (Study 1) compared RAVICTI to sodium phenylbutyrate by evaluating venous ammonia levels in patients with UCDs who had been on sodium phenylbutyrate prior to enrollment for control of their UCD. Patients were required to have a confirmed diagnosis of UCD involving deficiencies of CPS, OTC, or ASS, confirmed via enzymatic, biochemical, or genetic testing. Patients had to have no clinical evidence of hyperammonemia at enrollment and were not allowed to receive drugs known to increase ammonia levels (e.g., valproate), increase protein catabolism (e.g., corticosteroids), or significantly affect renal clearance (e.g., probenecid).
The primary endpoint was the 24-hour AUC (a measure of exposure to ammonia over 24 hours) for venous ammonia on days 14 and 28 when the drugs were expected to be at steady state. Statistical noninferiority would be established if the upper limit of the 2-sided 95% CI for the ratio of the geometric means (RAVICTI/sodium phenylbutyrate) for the endpoint was ≤ 1.25.
Forty-five patients were randomized 1:1 to 1 of 2 treatment arms to receive either
- Sodium phenylbutyrate for 2 weeks → RAVICTI for 2 weeks; or
- RAVICTI for 2 weeks → sodium phenylbutyrate for 2 weeks.
Sodium phenylbutyrate or RAVICTI were administered TID with meals. The dose of RAVICTI was calculated to deliver the same amount of PBA as the sodium phenylbutyrate dose the patients were taking when they entered the trial. Forty-four patients received at least 1 dose of RAVICTI in the trial.
Patients adhered to a low-protein diet and received amino acid supplements throughout the study. After 2 weeks of dosing, by which time patients had reached steady state on each treatment, all patients had 24 hours of ammonia measurements.
Demographic characteristics of the 45 patients enrolled in Study 1 were as follows: mean age at enrollment was 33 years (range: 18-75); 69% were female; 33% had adult-onset disease; 89% had OTC deficiency; 7% had ASS deficiency; 4% had CPS deficiency.
RAVICTI was noninferior to sodium phenylbutyrate with respect to the 24-hour AUC for ammonia. Forty-four patients were evaluated in this analysis. Mean 24-hour AUCs for venous ammonia during steady-state dosing were 866 μmol•h/L and 977 μmol•h/L with RAVICTI and sodium phenylbutyrate, respectively. The ratio of geometric means = 0.91 (95% CIs = 0.8-1.04).
The mean venous ammonia levels over 24-hours after 2 weeks of dosing (on day 14 and 28) in the double-blind short-term study (Study 1) are displayed in Figure 2 below. The mean and median maximum venous ammonia concentration (Cmax) over 24 hours and 24-hour AUC for venous ammonia are summarized in Table 2. Ammonia values across different laboratories were normalized to a common normal range of 9 to 35 μmol/L using the following formula after standardization of the units to μmol/L:
Normalized ammonia (μmol/L) = ammonia readout in μmol/L x (35/ULN of a laboratory reference range specified for each assay)
Figure 2: Venous Ammonia Response in Adult UCD
Patients in Short-Term Treatment Study 1
Table 2: Venous Ammonia Levels in Adult UCD Patients
in Short-Term Treatment Study 1
|Mean (SD)||Median (min, max)|
|Daily Cmax (μmol/L)|
|RAVICTI||61 (46)||51 (12, 245)|
|Sodium phenylbutyrate||71 (67)||46 (14, 303)|
|24-Hour AUC (μmol&bul;h/L)|
|RAVICTI||866 (661)||673 (206, 3351)|
|Sodium phenylbutyrate||977 (865)||653 (302, 4666)|
Open-Label, Uncontrolled, Extension Study in Adults
A long-term (12-month), uncontrolled, open-label study (Study 2) was conducted to assess monthly ammonia control and hyperammonemic crisis over a 12-month period. A total of 51 adults were in the study and all but 6 had been converted from sodium phenylbutyrate to RAVICTI. Venous ammonia levels were monitored monthly. Mean fasting venous ammonia values in adults in Study 2 were within normal limits during long-term treatment with RAVICTI (range: 6-30 μmol/L). Of 51 adult patients participating in the 12-month, open-label treatment with RAVICTI, 7 patients (14%) reported a total of 10 hyperammonemic crises. The fasting venous ammonia measured during Study 2 is displayed in Figure 3. Ammonia values across different laboratories were normalized to a common normal range of 9 to 35 μmol/L.
Figure 3: Venous Ammonia Response in Adult UCD
Patients in Long-Term Treatment Study 2
Clinical Studies In Pediatric Patients With UCDs
The efficacy of RAVICTI in pediatric patients 2 to 17 years of age was evaluated in 2 fixed-sequence, open-label, sodium phenylbutyrate to RAVICTI switchover studies (Studies 3 and 4). Study 3 was 7 days in duration and Study 4 was 10 days in duration.
These studies compared blood ammonia levels of patients on RAVICTI to venous ammonia levels of patients on sodium phenylbutyrate in 26 pediatric UCD patients between 2 months and 17 years of age. Four patients < 2 years of age are excluded for this analysis due to insufficient data. The dose of RAVICTI was calculated to deliver the same amount of PBA as the dose of sodium phenylbutyrate patients were taking when they entered the trial. Sodium phenylbutyrate or RAVICTI were administered in divided doses with meals. Patients adhered to a low-protein diet throughout the study. After a dosing period with each treatment, all patients underwent 24 hours of venous ammonia measurements, as well as blood and urine PK assessments.
UCD subtypes included OTC (n=12), argininosuccinate lyase (ASL) (n=8), and ASS deficiency (n=2), and patients received a mean RAVICTI dose of 7.9 mL/day (8 mL/m²/day, 8.8 g/m²/day), with doses ranging from 1.4 to 17.4 mL/day (1.5 to 14.4 g/m²/day). Doses in these patients were based on previous dosing of sodium phenylbutyrate.
The 24-hour AUCs for blood ammonia (AUC0-24h) in 11 pediatric UCD patients 6 to 17 years of age (Study 3) and 11 pediatric UCD patients 2 years to 5 years of age (Study 4) were similar between treatments. In children 6 to 17 years of age, the ammonia AUC0-24h was 604 μmol•h/L vs 815 μmol•h/L on RAVICTI vs sodium phenylbutyrate. In the UCD patients between 2 years and 5 years of age, the ammonia AUC0-24h was 632 μmol•h/L vs 720 μmol•h/L on RAVICTI vs sodium phenylbutyrate.
The mean venous ammonia levels over 24 hours in open-label, short-term Studies 3 and 4 at common time points are displayed in Figure 4. Ammonia values across different laboratories were normalized to a common normal range of 9 to 35 μmol/L using the following formula after standardization of the units to μmol/L:
Normalized ammonia (μmol/L) = ammonia readout in μmol/L x (35/ULN of a laboratory reference range specified for each assay)
Figure 4: Venous Ammonia Response in Pediatric UCD
Patients in Short-Term Treatment Studies 3 and 4
Open-Label, Uncontrolled, Extension Studies in Children
Long-term (12-month), uncontrolled, open-label studies were conducted to assess monthly ammonia control and hyperammonemic crisis over a 12-month period. In two studies (Study 2, which also enrolled adults, and an extension of Study 3, referred to here as Study 3E), a total of 26 children ages 6 to 17, were enrolled, and all but 1 had been converted from sodium phenylbutyrate to RAVICTI. Mean fasting venous ammonia values were within normal limits during long-term treatment with RAVICTI (range: 17-23 μmol/L). Of the 26 pediatric patients 6 to 17 years of age participating in these two trials, 5 patients (19%) reported a total of 5 hyperammonemic crises. The fasting venous ammonia measured during these two extension studies in patients 6 to 17 years is displayed in Figure 5. Ammonia values across different laboratories were normalized to a common normal range of 9 to 35 μmol/L.
Figure 5: Venous Ammonia Response in Pediatric UCD
Patients in Long-Term Treatment Studies 2 and 3E
In an extension of Study 4, after a median time on study of 4.5 months (range: 1.0-5.7 months), 2 of 16 pediatric patients ages 2 to 5 years had experienced three hyperammonemic crises.
1. Brusilow SW. Phenylacetylglutamine may replace urea as a vehicle for waste nitrogen excretion. Pediatr Res. 1991;29(2):147-150.
Last reviewed on RxList: 1/26/2016
This monograph has been modified to include the generic and brand name in many instances.
Additional Ravicti Information
Report Problems to the Food and Drug Administration
Find out what women really need.