CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`APPLICATION NUMBER:
`
`203284Orig1s000
`
`SUMMARY REVIEW
`
`
`
`
`

`

`Division Director Review
`
`Summary Review for Regulatory Action
`
`electronic st u u
`
`Donna Griebel, MD
`From
`
`Subject
`Division Director Summary Review
`NBA
`203284
`
`A . ulicant Name
`Date of Submission
`
`PDUFA Goal Date
`
`Proprietary Name /
`Established
`S 1
`
`Name
`
`Dosage Forms / Strength
`
`Proposed Indication(s)
`
`H perion Theta u eutics, Inc.
`December 23, 2011
`
`January 23, 2013
`Action Date: Feb
`
`1, 2013
`
`Ravicti/ glycerol phenylbutyrate
`
`Liquid for oral administration
`1.1 g of glycerol phenylbutyrate (GPB) in 1 ml of
`Ravicti® e uivalent to 1.02 ' hen lbu
`
`Adjunctive therapy for chronic management of adult
`and pediatric patients with urea cycle disorders (UCD)
`involving deficiencies of the following enzymes:
`carbamyl phosphate synthetase (CPS), ornithine
`transcarbamylase (OTC), argininosuccinate synthetase
`(ASS), argininosuccinate lyase (ASL) or arginase
`
`
`
`Action/Recommended Action for Approval
`NME:
`
`—_0ND Action Packa - e, includin :
`
`Names of disci n line reviewers
`
`Ke Zhan, PhD/David Jose h, PhD
`
`DSI
`
`CDTL Review
`
`OSE/DMEPA
`
`OSE/DRISK
`
`K. Malek, MD/Susan Leibenhaut, MD/Susan
`Tho I son, MD
`
`Melanie Blank, MD
`
`Lubna Merchant, PharmD, MS/Kellie Taylor, PhannD,
`MPH/Carol H01 0 uist, RPh
`
`Medication Guide:Latonia Ford, RN, BSN,
`MBA/LaShawn Griffiths, MSHS-PH, BSN, RN/ Barbar
`
`Fuller, RN, MSN, CWOCN
`Pro osed REMS: Yasmin Chou I
`
`Page 1 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`OSE/DPV
`
`PMHS
`
`Worthy, Pharm D./Claudia Manzo, Pharm.D.
`Thang La, PharmD, BCPS/Ann Mackey, RPh,
`MPH/Shewit Bezabeh, MD, MPH/Linda Scarazzini,
`MD, RPh
`Alyson Karesh, MD/Hari Cheryl Sachs, MD/ Jeanine
`Best/Melissa Tassinari, PhD/Lynne Yao, MD
`Michelle Roth-Cline, MD, PhD/Robert Nelson, MD,
`PhD
`Eric Brodsky, MD/Jeanne Delasko/Laurie Burke
`J. Zhang/Q. Dang/D. Marathe/N. Mehrotra/M.
`Fiszman/N.Stockbridge
`
`Pediatric Ethicist/Office of
`Pediatric Therapeutics, OC
`SEALD
`Interdisciplinary Review Team for
`QT Studies
`OND=Office of New Drugs
`OPDP=Office of Prescription Drug Promotion
`OC= Office of the Commissioner
`OSE= Office of Surveillance and Epidemiology
`DMEPA=Division of Medication Error Prevention and Analysis
`DPDP=Division of Professional Drug Promotion
`DSI=Division of Scientific Investigations
`DRISK= Division of Risk Management
`CDTL=Cross-Discipline Team Leader
`PMHS=Pediatric and Maternal Health Staff
`
`
`
`
`
`
`
`Page 2 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`Division Director Summary Review
`
`
`
`1. Introduction
`
`
`Hyperion Therapeutics, Inc. submitted the New Drug Application (NDA) for RAVICTITM
`(glycerol phenylbutyrate) on December 23, 2011 pursuant to Section 505(b)(2) of the Federal
`Food, Drug and Cosmetic Act for the proposed indication:
`
`
`“Adjunctive therapy for chronic management of adult and pediatric patients with urea
`cycle disorders (UCD) involving deficiencies of the following enzymes: carbamyl
`phosphate synthetase (CPS), ornithine transcarbamylase (OTC), argininosuccinate
`synthetase (ASS), argininosuccinate lyase (ASL) or arginase (ARG) as well as the
`mitochondrial transporter ornithine translocase (HHH) deficiency.”
`
`
`Phenylbutyrate, the active pharmaceutical ingredient, is not a new molecular entity (NME).
`Buphenyl (sodium phenylbutyrate) was approved in 1996 and is marketed with the following
`very lengthy indication. I have bolded the words that most clearly reflect an actual indication:
`
`
`“adjunctive therapy in the chronic management of patients with urea cycle
`disorders involving deficiencies of carbamylphosphate synthetase (CPS), ornithine
`transcarbamylase (OTC), or argininosuccinic acid synthetase (AS). It is indicated
`in all patients with neonatal-onset deficiency (complete enzymatic deficiency,
`presenting within the first 28 days of life). It is also indicated in patients with late-
`onset disease (partial enzymatic deficiency, presenting after the first month of life)
`who have a history of hyperammonemic encephalopathy. It is important that the
`diagnosis be made early and treatment initiated immediately to improve survival. Any
`episode of acute hyperammonemia should be treated as a life-threatening emergency.
`BUPHENYL must be combined with dietary protein restriction and, in some cases,
`essential amino acid supplementation. (See Nutritional Supplementation subsection of
`the DOSAGE AND ADMINISTRATION section.) Previously, neonatal-onset disease
`was almost universally fatal within the first year of life, even when treated with
`peritoneal dialysis and essential amino acids or their nitrogen-free analogs. However,
`with hemodialysis, use of alternative waste nitrogen excretion pathways (sodium
`phenylbutyrate, sodium benzoate, and sodium phenylacetate), dietary protein
`restriction, and, in some cases, essential amino acid supplementation, the survival rate
`in newborns diagnosed after birth but within the first month of life is almost 80%. Most
`deaths have occurred during an episode of acute hyperammonemic encephalopathy.
`Patients with neonatal-onset disease have a high incidence of mental retardation. Those
`who had IQ tests administered had an incidence of mental retardation as follows:
`ornithine transcarbamylase deficiency, 100% (14/14 patients tested); argininosuccinic
`acid synthetase deficiency, 88% (15/17 patients tested); and carbamylphosphate
`synthetase deficiency, 57% (4/7 patients tested). Retardation was severe in the majority
`of the retarded patients. In patients diagnosed during gestation and treated prior to any
`
`Page 3 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`episode of hyperammonemic encephalopathy, survival is 100%, but even in these
`patients, most subsequently demonstrate cognitive impairment or other neurologic
`deficits. In late-onset deficiency patients, including females heterozygous for ornithine
`transcarbamylase deficiency, who recover from hyperammonemic encephalopathy and
`are then treated chronically with sodium phenylbutyrate and dietary protein restriction,
`the survival rate is 98%. The two deaths in this group of patients occurred during
`episodes of hyperammonemic encephalopathy. However, compliance with the
`therapeutic regimen has not been adequately documented to allow evaluation of the
`potential for BUPHENYL and dietary protein restriction to prevent mental
`deterioration and recurrence of hyperammonemic encephalopathy if carefully adhered
`to. The majority of these patients tested (30/46 or 65%) have IQ's in the average to low
`average/borderline mentally retarded range. Reversal of pre-existing neurologic
`impairment is not likely to occur with treatment and neurologic deterioration may
`continue in some patients. Even on therapy, acute hyperammonemic encephalopathy
`recurred in the majority of patients for whom the drug is indicated. BUPHENYL may
`be required life-long unless orthotopic liver transplantation is elected.”
`
`
`In keeping with multiple interactions with the Division during the clinical development of
`Ravicti, including a SPA agreement, the safety and efficacy data submitted in support of the
`NDA hinge on a trial conducted to establish noninferiority of Ravicti to the approved
`Buphenyl (sodium phenylbutyrate) product in control of venous ammonia level, based on 24-
`hour AUC of ammonia (AUCNH3). This trial (Study 006), which was conducted in adult
`patients with UCD, was essentially designed to demonstrate bioequivalence of the two
`products for the PD marker, AUCNH3, specifically focusing on the upper bound of the
`confidence interval, i.e., the AUCNH3 ratio of the geometric means for Ravicti/Buphenyl must
`not exceed 1.25. Ammonia levels were considered an acceptable endpoint to establish
`efficacy, since high serum ammonia levels are known to cause serious morbidity and mortality
`in patients with urea cycle disorders (UCD). Ammonia was utilized as an endpoint to support
`the 2010 regular approval of Carbaglu for the UCD, N-acetylglutamate synthase (NAGS)
`deficiency.
`
`Phenylbutyrate has been a key component of the armamentarium for managing UCDs for
`decades. Major review issues identified in this NDA for Ravicti were related to knowledge
`gaps also associated with sodium phenylbutyrate at the time of its approval, which are
`reflected in the Buphenyl label. Those issues include:
`
`1) There is an absence of a clear methodology for defining a starting dose in an individual
`patient. Buphenyl product labeling states, “The usual total daily dose of BUPHENYL
`Tablets and Powder for patients with urea cycle disorders is 450 – 600 mg/kg/day in
`patients weighing less than 20 kg, or 9.9 – 13.0 g/m2/day in larger patients.” The key
`efficacy trial submitted in support of the Ravicti NDA (Study 006) evaluated patients who
`were not treatment naïve, and were on a stable dose of sodium phenylbutyrate. Patients
`enrolled in other trials submitted to this NDA were also merely converted from their stable
`dose of Buphenyl, with the exception of only 6 treatment naïve patients (two of whom
`developed neurological treatment emergent adverse events that led to dose reduction and
`discontinuation). The relative absence of data on how to initiate Ravicti in treatment
`
`Page 4 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`naïve patients and the lack of specific instructions for initiating therapy in the Buphenyl
`label (beyond providing a range), was a significant review issue that impacted labeling
`decisions. The investigators for the Ravicti trials have stated in a publication in Molecular
`Genetics and Metabolism [Mokhtarani M, et al. 107 (2012) 308-314], “Although sodium
`phenylbutyrate has been used for the treatment of UCDs since at least 1979, comparatively
`little information is available to guide physicians regarding its optimal dosing.”
`
`2) There were limitations to the strength of evidence provided in the NDA to support
`inclusion of information proposed by the applicant in the product label on how to modify
`dose based on various biomarkers, aside from venous ammonia levels.
`
`3) Ravicti is a pre-pro-drug. The drug must be released from the glycerol backbone to enable
`systemic absorption of therapeutic levels of phenylbutyrate (which is subsequently
`converted to the PAA molecule that binds glutamine to clear nitrogen). Because young
`infants, less than 2 months of age, are known to have immature pancreatic function, there
`is a scientifically known reason to have concern that infants less than 2 months of age will
`not absorb therapeutic levels of phenylbutyrate, due to low levels of pancreatic lipases.
`This is not an issue for the currently marketed sodium phenylbutyrate product. Because
`ineffective treatment of blood ammonia levels in a young infant could result in devastating
`outcomes, there was substantial concern that without a contraindication there would be
`substitution errors of Ravicti for Buphenyl in this age group, since both contain
`phenylbutyrate. This concern resulted in a Contraindication for use in this age group.
`
`
`4) Inadequate data were submitted to establish a safe dose in children between the age of 2
`months and 2 years. There were only 4 children studied in this age range and the data
`collected were inadequate for Clinical Pharmacology reviewers to determine safe dose
`recommendations. The Division had strongly encourage the sponsor during the clinical
`development to obtain adequate clinical data to cover all relevant age groups, recognizing
`that if Ravicti was in fact more palatable, it would be exceedingly important to have
`sufficient data to support labeling a safe and effective dose of Ravicti across all pediatric
`age groups. This gap will be addressed with a PMR under FDAAA, in light of the safety
`issue related to PAA. PREA does not apply since the applicant’s product has orphan
`designation for UCDs. The product label will state that the safety and efficacy have not
`been established in this age range (2 months to less than 2 years).
`
`
`5) The reviewers considered whether a comprehensive list of UCD subtypes (as proposed by
`the applicant) enrolled in the various trials submitted to the NDA should be included in the
`labeled indication for Ravicti, whether or not the number enrolled with a specific subtype
`was quite small. The Buphenyl label precedent was considered, which on first glance
`appears relatively limited compared to the applicant’s proposal; however, the additional
`text regarding neonatal onset and late onset in that indication seems broad and more
`encompassing. Ultimately, the reviewers considered the variability in the clinical
`presentations of the phenotypes (both among specific UCD subtypes and within specific
`subtypes), how the drug functions biochemically to reduce nitrogen, and how this product
`is clinically used as an adjunct, and determined that more general language was appropriate
`for the Indication section of the label. However, in keeping with the Buphenyl indication,
`
`Page 5 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`the reviewers determined that the Ravicti indication should communicate that the product
`should be reserved for use only in patients who need it to manage serum ammonia. The
`risk of carcinogenesis and the risk of neurotoxicity from PAA cannot be justified if a
`patient’s nitrogen can be managed by other standard measures. Patients with UCDs are
`managed by specialists in treatment of these diseases and the reviewers were confident that
`these limitations outlined in the indication would be adhered to without additional
`measures to assure safe use because the specialists who care for these patients are aware of
`these issues and currently practice within guidelines.
`
`
`6) The reviewers further considered the risk/benefit implications of the nonclinical
`carcinogenicity study results, and how this should be managed in product labeling.
`Ultimately, this (and the neurotoxicity associated with PAA, the active metabolite of
`Ravicti) impacted the Nursing Mothers section of the label and resulted in a PMR to obtain
`levels of the drug and its metabolites in breast milk, since nursing infants (particularly
`those without a diagnosis of UCD) would not have the same risk/benefit ratio for exposure
`to Ravicti as patients with a UCD.
`
` I
`
` will address these issues in the context of this review.
`2. Background
`The urea cycle is the final common pathway for the excretion of waste nitrogen in mammals
`and consists of 6 enzymes: (N-acetyl-glutamate synthetase, carbamyl phosphate synthetase
`[CPS], ornithine transcarbamylase [OTC], argininosuccinate synthetase [AS],
`argininosuccinate lyase [AL], and arginase). Each turn of the cycle results in elimination of
`two nitrogens in the form of urea. (See Figure 1 below). Urea cycle disorders result from a
`deficiency of any of the enzymes involved in the urea cycle. These disorders are autosomal
`recessive diseases, with the exception of ornithine transcarbamylase deficiency, which is an X-
`linked disorder. As stated in the CDTL review, the prevalence of Urea cycle disorders in the
`US is estimated to be 1:8200, with an overall incidence of approximately 1 in 45,000 live
`births.
`
`UCDs are characterized by hyperammonemia, encephalopathy, and respiratory alkalosis.
`Patients with UCDs are at high risk for neurologic deficits and death secondary to
`hyperammonemia. Patients may present with clinical manifestations across the lifespan,
`including as newborn/infants and in early childhood. The CDTL review provides a discussion
`of the variable phenotypic presentations. Partial enzyme deficiencies may present later in life,
`and depending on the level of function of the enzyme affected and the specific enzyme,
`patients may only require dietary management and nutritional supplements for chronic
`management of their disease.
`
`Current treatments for UCDs include restriction of nitrogen load by a low protein diet, oral
`neomycin to decrease bacterial ammonia production, and the use of NH3 scavengers. There are
`two nitrogen scavengers approved in the US for treatment of hyperammonemia in patients with
`UCDs: Buphenyl [(oral sodium phenylbutyrate (NaPBA)], Ammonul (intravenous mixture of
`sodium benzoate and sodium phenylacetate (NaPAA). Compounding pharmacies provide
`sodium phenylacetate and sodium benzoate for oral use. Carbaglu (carglumic acid), a specific
`
`Page 6 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`treatment for the UCD called NAGS deficiency, is a structural analogue of NAG, which is the
`essential allosteric activator of the enzyme CPS. Carbaglu does not function as a nitrogen
`scavenger, but instead serves to activate the key enzyme “at the top” of the urea cycle (see
`Figure below).
`
`Ravicti contains three molecules of 4-phenylbutyric acid (PBA) bonded via ester linkages to a
`glycerol backbone. It is converted into one glycerol and three phenylbutyrate (PBA) molecules
`either in the gut or during trans-enteric transport, and is converted by β-oxidation into PAA,
`which binds to glutamine. The resulting conjugate, PAGN (phenylacetylglutamine, see Figure
`below), is excreted via kidney. Since glutamine contains two nitrogens, this results in
`elimination of two nitrogens from the body.
`
`Figure 1: Metabolic pathways for nitrogen disposal
`
`
`Adapted from: http://www.drugs.com/pro/ammonul.html (12-March-2009)
`
`As discussed in the Introduction, one of the major review issues considered during product
`labeling was whether the indication should be limited to only those disorders in which there
`had been adequate characterization of safety and efficacy in the NDA trials. Ultimately,
`considering that the alternative pathway of nitrogen disposal that PAA provides, outside of the
`cycle, the review team determined that a general indication could be justified, as long as the
`indication clearly stated that the product was to be used only as an adjunct to other standard
`
`
`
`Page 7 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`interventions, such as diet, and that it should be used only if those other standard interventions
`were inadequate by themselves to manage the patient’s nitrogen.
`
`Regulatory background. Key goals for the interactions between FDA and the applicant
`during the clinical development of Ravicti included defining the appropriate endpoint for
`establishing efficacy and assuring that adequate numbers of children were studied to support
`defining a safe and effective dose across the full age range of patients who are affected by
`UCDs. I have summarized content from specific interactions between FDA and the applicant
`that impacted the content and review of the submitted NDA and/or labeling below.
`
`Pre-IND Meeting December 12, 2005:
`1) FDA told IND sponsor that because available pharmacokinetic data did not
`establish that Ravicti was bioequivalent to sodium phenylbutyrate, based on
`phenylbutyrate levels, PAA and PAGN levels, an efficacy trial would be required
`to support a marketing application.
`2) FDA suggested that the primary efficacy objective for an efficacy trial should
`include AUCNH3 and 24-hour urinary excretion of glutamine—related compormds.
`
`End of Phase 2 Meeting Janggg 14: 2009:
`1)
`FDA recommended that the primary efficacy objective for phase 3 trial(s)
`intended to support registration should be a co-primary of AUCNH3 and AUC
`of PAGN. FDA recommended that the sponsor consider a bioequivalence
`approach to analyses of the primary endpoint, and that the definition of success
`should also include that the AUCNH3 does not exceed 100 micromol/L.
`The sponsor proposed
`
`(DNA)
`
`2)
`
`The
`
`FDA recommended that the pediatric trial should be completed prior to
`initiating the proposed “pivotal” efficacy trial (Study 006), to inform
`assumptions used to power the trial. The FDA recommended that if the sponsor
`initiated Study 006 before the completion of Study 005, that children should be
`excluded from Study 006.
`The FDA stated that adult efficacy data might be “extrapolatable” to the
`pediatric population; however, the dose and safety in children is not.
`The FDA stated that the safety database for an NDA should include at least 35-
`40 patients who have been evaluated for at least 12 months on treatment.
`
`3)
`
`4)
`
`SPA No Agreement Letter issued to sponsor on April 3, 2009:
`1)
`Sponsor proposed that the primary efficacy endpoint of Study 006 would be
`m4) would be evaluated as a secondary
`
`2)
`
`endpoint.
`The FDA did not agree and stated that the primary endpoint should be AUCNH3.
`AUC of blood PAGN and 24 hour urinary PAGN excretion should be a
`secondary endpoints. Other secondary endpoints of interest were number of
`hyperammonemic crises and severity of hyperammonemic crises.
`
`Page 8 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`Ravicti 1 glycerol phenylbuggate) received Orphan Drug designation for maintenance of
`treatment of patients with deficiencies in egmes of the urea cycle on April 27, 2009.
`
`Meeting to discuss SPA No Agreement Letter on May 7, 2009:
`1) Sponsor agreed to a primary endpoint of 24-hour AUCNH3.
`2) Sponsor proposed, for the primary efficacy analysis of ratio of AUCNm of
`Ravicti/sodium phenylbutyrate, that the upper bound for the confidence interval to
`define success would be (m4), instead of 1.25.
`3) The FDA disagreed with the proposed upper bound, stating it should be 1.25.
`
`SPA Agreement Letter issued on June 30, 2009:
`1) The trial would only enroll adults.
`2) The primary endpoint was 24-hour AUC for venous NH3 at the end of treatment
`with each drug (Days 14 and 28). The primary efficacy analysis was the ratio of
`the AUCNH3 geometric means of Ravicti/sodium phenylbutyrate, with an upper
`bound of the confidence interval not exceeding 1.25 (utilizing a l-sided alpha of
`0.025) constituting evidence of efficacy.
`
`SPA Protocol Amendment submitted on March 19, 2010 (review filed on June 30, 2010):
`1) The sponsor propose
`
`(m4)
`
`2) The FDA did not agree with the amendment and said that it would result in
`nullification of the SPA. The pediatric data from Study 005 revealed substantive
`differences in PK profiles between pediatric patients and the adults in Study 003
`(adult PK study). Of particular concern, from a safety standpoint, was the apparent
`higher PAA exposure in children relative to adults. In addition, Study 005 showed
`differences in the PK profile between Ravicti and sodium phenylbutyrate. The
`table below summarizes the data that were bases for these concerns (reproduced
`fiom the clinical review in the regulatory file, dated June 30, 2010):
`
`Page 9 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`Table 1: PK Comparison of Adult (UP 1204-003) vs. Pediatrics (HPN-100-005) UCD Subjects
`
`PK Parameter
`
`NaPBA ,
`
`llPN—JOO
`
`Adults
`UP 1204-003
`
`Peds
`HPN-lOO-OOS
`
`Adults
`UP 1204-003
`
`Pcds
`HPN-lOO-OOS
`
`
`
`(N=l I)
`(N=10)
`(N=l l)
`(N-IO)
`————
`739(492)
`540(60J)
`63l(44.9)
`
`
`
`
`0.588(255)
`
`141(443)
`
`95.6(42.0)
`
`1.so<99.s>
`
`57460689)
`
`964 (63.6)
`
`40-50476)"
`
`90.5 (69.!)
`
`71066107)
`
`2.99 (122.1)
`
`IO98(44.2)
`
`1378(40.2)
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`——
`1133 31.1)
`
`
`Cmaxssfllg/mL)
`
`71.9(56.0)
`
`
`
`12-1034-4)
`
`
`105(335)
`
`
`13-1649)
`
`
`Pre-NDA Meeting December 7I 2010:
`1) FDA expressed concern that the sponsor’s NDA package, as outlined in the
`meeting backgrounder, would not provide adequate pediatric information,
`specifically information to support dosing in children under the age of 6 years and
`limited characterization of PAA levels in patients 6 years to 17 years of age.
`2) The sponsor proposed
`
`(b) (4)
`
`The FDA could not agree with this proposal in light of the
`EOP2 recormnendation of a safety data base that included 35-40 patients with 12
`months of safety data at the time of NDA submission.
`
`3)
`
`(m4)
`
`identification of an appropriate
`pediatric dose and evaluation of safety is necessary in light of the number of
`pediatric UCD patients who would be administered the drug once it is approved.
`
`Written answers issued to guestions (submitted by sponsor in a Febm 2011 meeting
`reguest) on Auggt 3, 201 l:
`1)
`FDA provided comments on the pharmacokinetic model.
`2)
`Sponsor proposed to include a
`
`mm
`
`Page 10 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`(D) (4)
`
`3)
`
`FDA reiterated its recommendation that the sponsor provide data to support
`dosing in children under the age of 6 years of age in the NDA submission,
`stating, “Submission of a Risk Evaluation and Mitigation Strategy (REMS)
`would not satisfy nor replace the need for information in this patient population.
`Again, as stated in the meeting, this population constitutes a significant portion
`of the UCD population and would likely use your product, if approved.
`Therefore, we again strongly recommend that you provide this information at
`the time of your NDA submission. It may be acceptable for you to submit the
`PK results from Study HPN-100-012 for review at the time of the NDA
`submission to provide information on dosing in patients younger than 6 years of
`age. However, if these data suggest substantially different exposures compared
`to adults, additional safety data from the 12-month open label extension study
`may also be required.”
`
`3. CMC/Device
`
`I concur with the conclusions reached by the chemistry reviewer this NDA provided
`“sufficient information to assure identity, strength, purity, and quality of the drug product,
`Ravicti liquid for oral administration.” The manufacturing site inspections were acceptable. I
`concur with the CMC reviewers that the product should be described as an “oral liquid” in the
`product label, instead of an “oral solution”. The product is not a substance that has been
`dissolved into a solution. There are no outstanding issues.
`
`4. Nonclinical Pharmacology/Toxicology
`
`I concur with the conclusions reached by the Pharmacology/Toxicology reviewer that there
`are no outstanding pharmacology/toxicology issues that preclude approval.
`
`I concur that the nonclinical studies support labeling consistent with the requirements for
`Pregnancy Category C.
`
`Carcinogenicity. The results of a 2-year carcinogenicity study in rats (administered glycerol
`phenylbutyrate) were presented to the CAC on July 17, 2012, and the Committee concluded
`that tumors observed in the study were drug related. Multiple tumor types were observed,
`arising in the pancreas (acinar cell adenoma, carcinoma and combined adenoma/carcinoma) in
`males and females, Zymbal’s gland (carcinoma) in males and females, adrenal cortex
`(combined adenoma/carcinoma) in females, uterus (endometrial stromal polyp and combined
`polyp/sarcoma), and thyroid (follicular cell adenoma, carcinoma and combined
`adenoma/carcinoma) in females. As noted in the Nonclinical Pharmacology review and the
`CDTL review, the doses administered in this carcinogenicity study ranged 3-8 times the
`exposure expected in human patients being treated for underlying UCD [range depends on sex
`and age (adult/pediatric)].
`
`Page 11 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`The Pharmacology reviewers concluded that even with the positive carcinogenicity study, the
`risk/benefit of Ravicti favored its approval. The CDTL concurred after considering: 1) an OSE
`consult review, 2) whether carcinogenicity would only be expected with the glycerol
`phenylbutyrate product (and not the currently marketed sodium phenylbutyrate product), and
`3) the benefit associated with phenylbutyrate in managing ammonia levels in patients with
`UCD.
`
`OSE’s Division of Pharmacovigilance was consulted to evaluate whether there have been
`spontaneous reports of malignancy associated with Buphenyl (sodium phenylbutyrate). No
`reports were identified in the AERS database and in an NIH PubMed search; however, a signal
`of malignancy would be difficult to detect from these sources. My PubMed search for
`published evidence of increased risk of tumors in patients with UCDs, which had not
`necessarily been linked to their medications, found only limited information. A Japanese
`publication pointed to a case series of 8 cases of hepatocellular carcinoma in 56 adult patients
`with citrullenemia due to argininosuccinate synthetase deficiency (Nakayama M, et al.
`Hepatology, 1990; 11(5):819-23). In addition, Wilson, et al. (Molecular Genetics and
`Metabolism 105, 2012: 263-265) reported a possible association of UCDs with liver
`dysfunction, which they linked to an increased risk of developing hepatocellular carcinoma.
`The authors examined charts from the Children’s Hospital Colorado longitudinal study site for
`the NIH-funded Rare Diseases Clinical Research Center longitudinal study of UCDs and
`found that more than 50% of patients at that site with symptomatic OTCD had liver
`dysfunction or failure. The authors cited prior publications that had documented acute liver
`dysfunction as a clinical presentation of ornithine transcarbamylase deficiency (OTCD). The
`authors linked these liver dysfunction signals and hepatocellular carcinoma to the underlying
`disease, and not the patients’ medications. In addition, the medical history was reported in
`detail in some of the patients, which indicated they had not been exposed to phenylbutyrate.
`
`The Buphenyl label does not contain carcinogenicity study information, and it appears it was
`approved in the absence of the existence of such information. Buphenyl is sodium
`phenylbutyrate, and while it is unlikely that the carcinogenicity study of sodium
`phenylbutyrate would differ from Ravicti, based on the presence of the glycerol component in
`Ravicti, that possibility cannot be completely excluded without actual data from a sodium
`phenylbutyrate carcinogenicity study. However, the phenylbutyrate is released from the
`glycerol backbone primarily within the gut lumen, so the drug to which both patients and rats
`are primarily systemically exposed is the same between Buphenyl and Ravicti (like UCD
`patients, intact glycerol phenylbutyrate was not detected in blood in rat PK studies). In
`addition, the Clinical Pharmacology review found that intact Ravicti was not detected in
`pharmacokinetic analyses of samples taken from UCD patients. (Although it was detected in
`normal volunteers, the applicant attributed the difference in presence of intact drug between
`populations to contamination during processing of the samples in the healthy volunteer study.
`Refer to the Clinical Pharmacology review for more detailed information.)
`
` concur with the reviewers that the risk/benefit of Ravicti for this issue still favors its
`approval. However, the same risk/benefit assessment does not apply to a breastfeeding infant
`who doesn’t have UCD (if the infant’s mother is taking Ravicti for her own UCD). The
`Maternal Health team was consulted regarding this issue and they contributed to evaluating the
`
` I
`
`Page 12 of 33
`
`Reference ID: 3254202
`
`

`

`Division Director Review
`
`label to assure that appropriate language was included in the Nursing Mother section. In
`addition, the Maternal Health team and Pediatric Ethics consultants worked with the Division
`to develop a PMR to evaluate breast milk for levels of exposure to phenylbutyrate and its
`metabolites.
`
`Nonclinical hepatic histopathology review. In light of the literature search results regarding
`liver dysfunction, cited above, I examined the Pharmacology/Toxicology review for evidence
`of hepatotoxicity associated with glycerol phenylbutyrate in the submitted nonclinical studies.
`The Pharmacology/Toxicology review states histopathology examinations in nonclinical
`studies “revealed hepatocellular hypertrophy in 13-week oral toxicity studies in mice and
`monkeys and in a 52-week oral toxicity study in monkeys” (in which the hypertrophy
`increased with increasing doses). Mild mixed cellular infiltrates in liver were observed in a
`neonatal rat toxicity study. The histological description in the 12 month monkey study was
`“hypertrophy was characterized by enlarged hepatocytes with stippled to granular eosinophilic
`cytoplasm that compressed and constricted sinusoidal spaces without evidence of passive
`congestion or ischemia.” I discussed those findings with the reviewers, and the Nonclinical
`Pharmacology team leader advised me that hepatocellular hypertrophy is assumed to be
`indicative of enzyme induction, and not classic evidence of a strong potential that the drug is a
`hepatotoxin in humans
`
`Buphenly label subcutaneous PAA study. Another issue raised in the
`Pharmacology/Toxicology and CDTL reviews, and addressed in labeling discussions, is the
`presence o