`
`CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`APPLICATION NUMBER:
`203284Orig1s000
`
`STATISTICAL REVIEW(S)
`
`
`
`
`
`
`
`

`

`
`
`
`
`U.S. Department of Health and Human Services
`Food and Drug Administration
`Center for Drug Evaluation and Research
`Office of Translational Sciences
`Office of Biostatistics
`
`
`S T A T I S T I C A L R E V I E W A N D E VA L U A T I O N
`CLINICAL STUDIES
`
`NDA/BLA #:
`Supplement #:
`
`Drug Name:
`
`Indication(s):
`
`Applicant:
`
`Date(s):
`
`NDA 203-284
`
`RAVICTITM (glycerol phenylbutyrate; HPN-100); TID, in liquid solution,
`orally with meals
`Adjunctive therapy for chronic management of adults and pediatric patients ≥ 6
`years of age with urea cycle disorders (UCDs)
`Hyperion Therapeutics, Inc.
`Stamp Date: December 23, 2011
`PDUFA Goal date: January 23, 2013
`Standard with Major Amendment (13 month review cycle)
`
`Review Priority:
`
`
`Biometrics Division:
`Division of Biometrics III
`Statistical Reviewer:
`Behrang Vali M.S.
`Concurring Reviewers: Mike Welch Ph.D.
`
`
`Medical Division:
`Division of Gastroenterology and Inborn Errors Products
`MO: Nancy Snow, M.D.
`MOTL and CDTL: Melanie Blank, M.D.
`Project Manager:
`Jessica Benjamin M.P.H.
`
`
`Keywords: NDA review, Clinical Studies, Active Control/Non-Inferiority
`
`
`Clinical Team:
`
`Reference ID: 3247982
`
`

`

`3
`
`TABLE OF CONTENTS
`1 EXECUTIVE SUMMARY .................................................................................................................................5
`2
`INTRODUCTION ...............................................................................................................................................6
`2.1
`OVERVIEW......................................................................................................................................................6
`2.2
`DATA SOURCES ..............................................................................................................................................8
`STATISTICAL EVALUATION ........................................................................................................................9
`3.1
`DATA AND ANALYSIS QUALITY .....................................................................................................................9
`3.2
`EVALUATION OF EFFICACY ............................................................................................................................9
`3.2.1
`Study Design and Endpoints..................................................................................................................9
`3.2.2
`Statistical Methodologies.....................................................................................................................13
`3.2.3
`Patient Disposition, Demographic and Baseline Characteristics........................................................16
`3.2.4
`Results and Conclusions ......................................................................................................................19
`3.3
`EVALUATION OF SAFETY..............................................................................................................................24
`3.4
`BENEFIT-RISK ASSESSMENT.........................................................................................................................24
`4 FINDINGS IN SPECIAL/SUBGROUP POPULATIONS .............................................................................24
`4.1
`GENDER, RACE, AGE, AND GEOGRAPHIC REGION ........................................................................................24
`4.2
`OTHER SPECIAL/SUBGROUP POPULATIONS ..................................................................................................25
`SUMMARY AND CONCLUSIONS ................................................................................................................26
`5.1
`STATISTICAL ISSUES.....................................................................................................................................26
`5.2
`COLLECTIVE EVIDENCE................................................................................................................................26
`5.3
`CONCLUSIONS AND RECOMMENDATIONS .....................................................................................................26
`5.4
`LABELING RECOMMENDATIONS ...................................................................................................................26
`
`5
`
`
`
`
`
`Reference ID: 3247982
`
`2
`
`

`

`
`
`LIST OF TABLES
`Table 1 Summary Information for Relevant Clinical Trials .........................................................................................8
`Table 2 Randomization to Treatment in HPN-100-006..............................................................................................10
`Table 3 Disposition – n (%)........................................................................................................................................17
`Table 4 Demographic and Baseline Characterisitics ..................................................................................................18
`Table 5 Analysis of NH324-hour AUC by Treatment Group ............................................................................................19
`Table 6 Correlation of U-PAGN24-hour Excr and NH324-hour AUC by Treatment Group and Overall ................................20
`Table 7 Summary of Blood Ammonia 24-hour Cmax (µmol/L) by Treatment Sequence and Overall ........................21
`Table 8 Summary of Ammonia Levels above ULN by Treatment Group..................................................................22
`Table 9 Gender Subgroup Analysis of NH324-hour AUC by Treatment Group ...............................................................25
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 3247982
`
`

`

`LIST OF FIGURES
`Figure 1 Disposition ...................................................................................................................................................16
`Figure 2 Mean (±SE) Blood Ammonia by Monthly Visit – HPN-100-106/HPN-100-107 ........................................23
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 3247982
`
`

`

`1 EXECUTIVE SUMMARY
`
`There was a sufficient level of evidence to support an efficacy claim for RAVICTITM (HPN-
`100), and the claims currently reflected within the applicant’s submitted product label are
`supportable as shown in this NDA review. With further motivation under the current public
`health circumstances in which Urea Cycle Disorders are a rare, serious and life-threatening
`condition with a not fully met medical need, this reviewer supports the approval of HPN-100 for
`the treatment of adult and pediatric patients ≥ 6 years of age with this condition.
`
`The efficacy of HPN-100 was principally demonstrated in the single study HPN-100-106. In this
`trial, HPN-100 was determined to be non-inferior to BUPHENYL® (sodium
`phenylbutyrate/NaPBA) with regard to the 24-hour area under the curve (AUC) for blood
`ammonia (i.e. NH324-hour AUC) on Study Days 14 and 28 which were when these drugs were
`expected to be at steady state. In addition, the overall correlation, calculated by pooling all
`patient data across both treatment groups, between urinary phenylacetylglutamine (i.e. U-
`PAGN24-hour Excr) and NH324-hour AUC was the only significant secondary endpoint determined
`through the pre-specified Hochberg’s multiplicity adjustment procedure. This correlation was
`also shown to be positive within each individual treatment group thereby further supporting the
`comparability of the treatments themselves. Although pre-specified by the applicant through
`Hochberg’s multiplicity adjustment procedure, the overall correlation endpoint was really
`exploratory in nature and was tested for possible future utilization of U-PAGN for dose selection
`and dose adjustment purposes. The correlation, whether overall or by individual treatment
`group, only indicates a possible linear association and does not confirm a treatment benefit. As
`such, the applicant never considered this endpoint for labeling purposes and hence the label does
`not reflect these correlation results.
`
`There were no statistical issues that impacted the overall conclusions of trial HPN-100-106. The
`study’s design was adjudicated as being adequate, and the applicant’s corresponding analysis
`plan was deemed appropriate. Consequently, and in addition to the consensus regarding the
`clinical meaningfulness of the NH324-hour AUC endpoint, results from trial HPN-100-106 are
`viewed positively as the formal basis for an efficacy claim to be reflected by the product’s label.
`The apparent sustained efficacy profile during the extension study HPN-100-107 further supports
`the efficacy claim for HPN-100.
`
`This reviewer recommends that careful consideration be made pertaining to the labeling of the
`HPN-100 dose in that the dose for a given patient, as described below in Section 3.2.1, was a
`function of the patient’s NaPBA dose.
`
`
`
`Reference ID: 3247982
`
`5
`
`

`

`2 INTRODUCTION
`
`2.1 Overview
`
`As the regulatory agent on behalf of Ucyclyd Pharma, Inc., a wholly-owned subsidiary of
`Medicis Pharmaceutical Corporation, on December 23, 2011 Hyperion Therapeutics, Inc.
`submitted this New Drug Application (NDA) for RAVICTITM (glycerol phenylbutyrate; HPN-
`100) pursuant to Section 505(b)(1) of the Federal Food, Drug and Cosmetic Act and in
`accordance with Title 21, Part 314 of the Code of Federal Regulations. The active
`pharmaceutical ingredient in HPN-100 (to be orally administered as a liquid solution TID with
`meals) is glycerol phenylbutyrate. This is the first prescription product to have glycerol
`phenylbutyrate as its active pharmaceutical ingredient thereby making it a New Molecular Entity
`(NME). Effective on April 10, 2007, HPN-100 has officially undergone clinical development
`under IND 73,480 in patients with urea cycle disorders (UCDs), and has been developed
`specifically to establish safety and efficacy in a subpopulation of these patients. The official
`proposed indication for HPN-100 is as an adjuvant therapy for chronic management of adult and
`pediatric patients ≥ 6 years of age with UCDs involving deficiencies of the following enzymes;
`carbamyl phosphate synthetase (CPS), ornithine transcarbamylase (OTC), argininosuccinate
`synthetase (ASS), argininosuccinate (ASL) or arginase (ARG) as well as the mitochondrial
`transporter ornithine translocase (HHH deficiency).
`
`The urea cycle is the major route for metabolism of waste nitrogen within the body. UCDs are
`inherited deficiencies of enzymes or transporters, e.g. those previously mentioned in the
`proposed indication, necessary for the synthesis of urea from ammonia. Hence the absence of
`these enzymes or transporters results in the accumulation of toxic levels of waste nitrogen, e.g.
`ammonia, in the blood and brain of all affected patients. Consequently these patients are highly
`sick and encephalopathic. Currently, there are FDA-approved treatment options for patients with
`UCDs e.g. BUPHENYL® (sodium phenylbutyrate/NaPBA) which acts as a nitrogen scavenger in
`the body. The mechanism of action of NaPBA is as follows. After absorption through oral
`administration, the sodium (Na) and the phenylbutyrate (PBA) break apart from each other.
`PBA then undergoes β-oxidation to become phenylacetate (PAA) which is the active metabolite
`utilized for nitrogen scavenging purposes. This is the reason NaPBA is considered a pro-drug of
`PAA. Each mole of PAA subsequently scavenges 2 moles of nitrogen (from existing ammonia
`in the blood) and is then conjugated with glutamine in the liver (and kidney) while ultimately
`being excreted in the form of urine as phenylacetylglutamine (PAGN). The urinary form of
`phenylacetylglutamine is referred to as U-PAGN. The utility of this mechanism of action is that
`the nitrogen content of the resulting U-PAGN in patients with UCDs is equivalent to that of urea
`in patients with a fully functioning urea cycle. Hence PBA (via PAA) provides an alternative
`pathway for nitrogen disposal in patients without a fully functioning urea cycle. The mechanism
`of action of HPN-100 is the same as that of NaPBA except that the initial step in the previously
`described process is slightly different. HPN-100 itself is an inactive compound; however upon
`absorption through oral administration, PBA is extracted from the compound which begins the
`same nitrogen scavenging process as previously described. For this reason, HPN-100 is
`considered a pre-pro-drug of PAA.
`
`
`
`
`Reference ID: 3247982
`
`6
`
`

`

`When administered at the recommended dose levels NaPBA has been shown from clinical
`experience to be safe and effective in improving long-term survival in patients with UCDs (i.e.,
`reducing the incidence of deaths due to hyperammonemic encephalopathy). However,
`compliance with NaPBA is difficult due to a high pill burden (up to 40 pills or 40 mL of
`dissolved powder daily for patients taking 20 g of NaPBA), foul taste, unpleasant odor, and high
`sodium content (approximately 2,300 mg/day for patients taking 20 g). All of these factors
`render NaPBA very difficult to take, and compliance is suboptimal even for UCD patients with
`the most severe deficiency states, whose alternative is life-threatening hyperammonemia.
`Consequently UCDs remains as a rare, serious and life threatening condition with a not fully met
`medical need. HPN-100 is an alternative therapy to NaPBA in patients with UCDs as it is
`expected to provide similar nitrogen-scavenging ability while eliminating the current issues of
`bad taste, odor, sodium content, and pill burden.
`
`Hyperion Therapeutics, Inc. obtained Fast Track designation from the Agency on October 4,
`2010. The review cycle established by the Division of Gastroenterology and Inborn Error
`Products (DGIEP) was a standard 10 month cycle; however this was later amended to being a 13
`month review cycle. The application also qualified for Orphan Exception under Section
`736(a)(1)(E) of the Federal Food, Drug and Cosmetic Act, and Hyperion Therapeutics, Inc. did
`obtain Orphan Designation from the Office or Orphan Products Development (OOPD) on July
`27, 2009.
`
`The clinical efficacy and safety of HPN-100 has been principally evaluated through one study: a
`Phase 3, multicenter, randomized, double-blind, double-dummy, placebo controlled, cross-over
`non-inferiority study (HPN-100-006) which serves as the lone adequate and well controlled
`study of this clinical development program as per 21 CFR 314.126. The design of this study was
`agreed to by DGIEP in the context of a Special Protocol Assessment (SPA) with the agreement
`letter sent to Hyperion Therapeutics, Inc. on June 30, 2009.
`
`Table 1 below presents information on this lone relevant clinical trial contained in the
`submission.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 3247982
`
`7
`
`

`

`Table 1
`Summary Information for Relevant Clinical Trials
`Study
`Design
`and
`Type of
`Control
`
`Number of
`Dosed
`Patients
`
`Patient
`Diagnosis
`
`Duration
`of
`Treatment
`
`Study
`Status;
`Type of
`Report
`
`Total: 45
`
`UCD Patients
`
`4 weeks
`(2 weeks each
`treatment arm)
`
`Complete;
`Full
`
`Type of
`Study;
`Phase
`
`Study
`Identifier
`
`Efficacy
`and Safety;
`Phase 3
`
`
`HPN-100-006
`
`Objective(s)
`of the Study
`
`To assess the
`non-inferiority
`of HPN-100 to
`NaPBA by
`evaluating
`blood ammonia
`levels in adult
`patients with
`UCDs from
`OTC, CPS, and
`ASS who were
`being treated
`with NaPBA for
`control of their
`UCD
`
`Multicenter,
`randomized,
`double-blind,
`double-
`dummy,
`placebo
`controlled,
`cross-over,
`non-
`inferiority
`
`Test
`Product(s);
`Regimen;
`Route
`
`Treatment Arm
`A:
`NaPBA and
`HPN-100
`placebo followed
`by HPN-100 and
`NaPBA placebo,
`
`Treatment Arm
`B:
`HPN-100 and
`NaPBA placebo
`followed by
`NaPBA and
`HPN-100
`placebo;
`
`TID orally with
`meals;
`
`HPN-100 in
`liquid solution,
`NaPBA in tablet
`or powder form
`
`
`Source: Reviewer’s Table.
`
`
`2.2 Data Sources
`
`This NDA was submitted electronically in eCTD format via the FDA Electronic Submissions
`Gateway (ESG). Its content, including the electronic data sets and labeling information, has
`been stored in the electronic document room (EDR) at this path location:
`\\Cdsesub1\evsprod\NDA203284. Sequences 0000, 0007, and 0009 contain all the contents
`relevant for this review.
`
`For study HPN-100-106, the applicant’s clinical study report (CSR), clinical datasets and
`analysis datasets were reviewed. The clinical datasets were compliant to the CDISC/SDTM
`v.3.1.2 implementation guide standard; however, a non-standardized legacy approach for
`modeling the corresponding analysis data was implemented. This approach was, however,
`somewhat homologous in nature to the CDISC/ADaM v.1.0 implementation guide standard.
`Adequate data definition files, both in Define.XML and Define.PDF format, and software code,
`in .SAS format, were also submitted for the study.
`
`
`
`
`
`
`Reference ID: 3247982
`
`8
`
`

`

`3 STATISTICAL EVALUATION
`
`The statistical evaluation section is written solely for trial HPN-100-006.
`
`3.1 Data and Analysis Quality
`
`This study utilized Electronic Data Capture (EDC), and the submitted data quality and integrity
`appeared to be adequate. There were no issues in reproducing the primary analysis dataset
`(along with the numerical results presented within the CSR), in particular the primary endpoint,
`from the original data source. It was possible to verify the randomized treatment assignments,
`and the applicant submitted documentation of data quality control/assurance procedures within
`Section 9.6 of their ICH E3 compliant CSR. The blinding/unblinding procedures were well
`documented within the protocol and in Section 9.4.6 of their ICH E3 compliant CSR. The
`applicant’s statistical analysis plan (SAP) was finalized on September 23, 2010. The SAP was
`submitted, and all relevant analysis decisions were made before unblinding. Database hard-lock
`was on September 30, 2010 with unblinding one week later on October 7, 2010.
`
`
`3.2 Evaluation of Efficacy
`
`3.2.1 Study Design and Endpoints
`
`As stated previously, the design of this study was agreed to by DGIEP in the context of a SPA
`with the agreement letter sent to Hyperion Therapeutics, Inc. on June 30, 2009. Consequently,
`the finalized protocol for this study was signed off on June 30, 2009. The study was initiated on
`October 12, 2009, and it was completed on September 9, 2010.
`
`This Phase 3 efficacy and safety study serves as the clinical development program’s adequate
`and well-controlled study which makes it the basis for an efficacy claim to be reflected by the
`product label. This was a 4-week, multicenter (with a total of 19 clinical sites), randomized,
`double-blind, double-dummy, placebo controlled, cross-over, non-inferiority trial whose primary
`objective was to assess the non-inferiority of HPN-100 to NaPBA by evaluating blood ammonia
`levels in adult patients with UCDs (with deficiencies of CPS, OTC, or ASS) who were being
`treated with NaPBA for control of their UCD. Patients must have had controlled ammonia levels
`(<100 µmol/L and without signs and symptoms of hyperammonemia) and be on a stable dose of
`NaPBA for at least one week prior to randomization.
`
`After study eligibility was confirmed, patients were randomly assigned, in a blinded fashion, (on
`a 1:1 ratio, in accordance with a computer-generated central randomization schedule) to receive
`either treatment sequence presented in Table 2 below. This was a double blinded study. The
`patients, investigators, study personnel, including the site pharmacist, were all blinded to the
`study drug assignment.
`
`
`
`
`
`
`
`Reference ID: 3247982
`
`9
`
`

`

`Table 2
`Randomization to Treatment in HPN-100-006
`Treatment Arm Period 1 (2 weeks)
`Period 2 (2 weeks)
`A
`NaPBA + HPN-100 placebo HPN-100 + NaPBA placebo
`B
`HPN-100 + NaPBA placebo NaPBA + HPN-100 placebo
`Source: Reviewer’s Table.
`
`
`It is to be noted that all patients who were randomized into and completed this study were later
`given the opportunity to roll over into a 12-month, multicenter, open-label, un-controlled
`extension study (trial HPN-100-007) which assesses the long term efficacy and safety of HPN-
`100.
`
`
`HPN-100 is a colorless to pale yellow, odorless, nearly tasteless
` orally via mouth, gastrostomy or nasogastric tube. NaPBA tablets were the preferred
`formulation for administration of NaPBA, and every reasonable effort was made to convert
`medically eligible adult patients who may have been on NaPBA powder to tablets at least one
`week prior to study randomization. However, patients who could not comply with taking
`NaPBA tablets (e.g., had difficulty swallowing tablets) were able to receive NaPBA powder
`upon consultation with the applicant and documentation of the reason the patient was not able to
`take NaPBA tablets. Upon approval by the applicant, a supply of NaPBA powder was provided
`for that patient throughout the study duration. No patient was allowed to switch from NaPBA
`tablet to NaPBA powder, or vice versa, while the study was ongoing. NaPBA tablets were orally
`administered while oral, nasogastric, or gastrostomy tube administration was utilized for NaPBA
`powder. The motivation for the double-dummy design was due to the fact that NaPBA was in
`tablet or powder form while HPN-100 was a liquid. HPN-100 and NaPBA (regardless of
`formulation) have short half-lives, and, consequently, were administered TID with meals. The
`short half-lives of these treatments were also the reason why a wash-out period was not instituted
`in the cross-over design.
`
`At the screening visit, the investigator determined the trial dose of NaPBA for each patient.
`Because the dose of NaPBA was chosen based on the severity of the enzyme deficiency, on the
`content of the patient’s diet, and on the intake of amino acids or other supplements, the dose of
`NaPBA did vary among study patients. And because a patient’s HPN-100 dose was dependent
`on their NaPBA dose, the HPN-100 dose also varied among study patients. The 100% HPN-100
`dose-equivalent to the 100% NaPBA dose was calculated as follows:
`
`Total daily NaPBA dose (g) × 0.95/1.1 = Total daily HPN-100 dose (mL)
`
`
`This equivalent HPN-100 total daily dose, which was derived to match the corresponding
`NaPBA total daily dose, was administered to ensure consistent metabolic control for each
`patient. The total daily dose of NaPBA (g) and HPN-100 (mL) divided by three was the dose to
`be taken during each meal. No adjustment to the NaPBA or HPN-100 dose (or schedule) was
`allowed during the study.
`
`
`
`
`
`Reference ID: 3247982
`
`10
`
`(b) (4)
`
`

`

`The following prohibited medications were not to be used during the study:
`• Drugs known to cause hyperammonemia, such as valproate
`• Drugs known to increase protein catabolism, such as corticosteroids
`• Drugs known to significantly affect renal clearance, such as probenecid
`• Drugs known to lower blood ammonia, including sodium benzoate
`
`
`In addition, patients also followed a stable diet throughout the study as prescribed by the
`investigator and dietician. All patients were to adhere to the low-protein diet and amino acid
`supplements prescribed for them. The diet chosen for each individual depended on age and
`residual enzyme activity and should not have been altered for this study.
`
`The primary study objective was to establish the non-inferiority of HPN-100 to NaPBA as
`assessed by venous ammonia. Blood samples were collected for assessment of venous ammonia
`levels, and at each designated time point indicated by the schedule of assessments; 2 mL of
`venous blood was drawn and processed by the laboratory at the investigator site per the facility
`standard operating procedures. This is consequently a local laboratory study (not a central
`laboratory study). There were different in vitro diagnostic methods administered at each site
`which measured plasma ammonia concentration. Two general types of methods were employed
`across the trial; an indirect, colorimetric method and a direct, enzymatic method. In addition,
`each laboratory used a slightly different normal reference range. Patients were ultimately
`admitted to the research unit for 24 hours of venous ammonia, PK blood and urine sampling
`(including an overnight stay) at the end of each treatment period by which time the study drug
`would have reached steady state.
`
`The following primary and secondary endpoints were pre-specified by the applicant.
`
`Primary Endpoint: The primary endpoint was the 24-hour area under the curve for blood
`ammonia (NH324-hour AUC) on Days 14 and 28 are when the drugs were expected to be at steady
`state. The AUC was calculated, using the trapezoidal rule, on a sequence of ammonia level
`concentrations obtained at pre-dose, and 2, 4, 8, 12, 16, 20, and 24 hours post-dose on Days 14
`and 28.
`
`Secondary Endpoints:
`• Overall correlation between 24-hour urinary PAGN excretion (i.e. U-PAGN24-hour Excr)
`and venous ammonia AUC0-24 (i.e. NH324-hour AUC) observed on steady state. The overall
`correlation pertains to pooling all patient data across both treatment groups.
`• Maximum venous ammonia values (i.e. Cmax) observed on steady state NaPBA versus
`HPN-100.
`• Rate (percentage) of ammonia values above upper limit of normal (ULN) observed on
`steady state NaPBA versus HPN-100.
`• Number and severity of symptomatic hyperammonemic crises.
`• PK parameters including Cmax for major metabolites of NaPBA and HPN-100 (including
`plasma PAA, PBA, PAGN and U-PAGN24-hour Excr).
`• Rate of adverse events in each treatment group.
`
`11
`
`
`
`
`Reference ID: 3247982
`
`

`

`A sample size of 44 evaluable patients will have provided 90% power to demonstrate that the
`ratio of the means of NH324-hour AUC between HPN-100 and NaPBA did not exceed 1.25. This
`assumed a one-sided α of 0.025, a standard deviation of the within-patient differences (natural
`log scale) of 0.225 and an expected ratio of the group means of 1.
`
`Statistical non-inferiority is shown if the upper limit of the two-sided 95% CI for the ratio of the
`treatment means (µHPN-100 / µNaPBA) is less than or equal to 1.25 Although NH324-hour AUC is a
`pharmacodynamic endpoint, the upper margin of 1.25 was based on standard bioequivalence
`rules (i.e. the 95% CI for the ratio of mean AUCs being within 0.80 and 1.25) and deemed
`appropriate by the clinical and pharmacology teams. It is to be noted that meeting the 0.80 lower
`limit was not made a requirement because it was not clinically relevant. The aforementioned
`ratio would be allowed to go as low as possible with a lower ratio being deemed as more
`clinically meaningful/effective. Once non-inferiority is established, the first three secondary
`endpoints, i.e. the key secondary endpoints, are evaluated using Hochberg’s multiplicity
`adjustment procedure in order to control the overall type I error.
`
`Statistical significance of the following key secondary efficacy endpoints were to be evaluated
`using
`Hochberg’s procedure:
`• S1. Overall correlation between U-PAGN24-hour Excr and NH324-hour AUC at steady state i.e.
`H0: ρ=0 vs. H1: ρ≠0
`• S2. Maximum venous ammonia values (i.e. Cmax) observed on steady state NaPBA
`versus HPN-100 i.e. H0: μ1=μ2 vs. H1: μ1≠μ2
`• S3. Rate (percentage) of ammonia values above upper limit of normal (ULN) observed
`on steady state NaPBA versus HPN-100 i.e. H0: μ1=μ2 vs. H1: μ1≠μ2
`
`
`Under Hochberg’s procedure, the resulting three p-values, each corresponding to one of the three
`aforementioned key secondary endpoints, were to be ordered from highest to lowest. If the
`largest p-value is less than or equal to α (i.e. 0.05), then the null hypotheses for all three key
`secondary endpoints were to be rejected. If, however, this largest p-value is greater than 0.05,
`then the two other endpoints were to be assessed at α/2 (i.e. 0.025). If the largest remaining p-
`value is less than or equal to 0.025, then the null hypotheses of the two remaining endpoints were
`to be rejected. If, however, this largest remaining p-value is greater than 0.025, then the sole
`remaining endpoint was to be assessed at α/3 (i.e. 0.0167). If the p-value for this third endpoint
`is less than or equal to 0.0167, then the corresponding null hypothesis was to be rejected.
`
`Reviewer Comments:
`The primary endpoint and non-inferiority margin were deemed by the review team as clinically
`meaningful, and the estimated sample size was validated and confirmed as appropriate. Overall,
`the design of study HPN-100-006 was deemed adequate. It would have been ideal if this study
`utilized a central laboratory. However this was not possible because clinical studies with very
`sick patients, such as those with UCDs, require laboratory assessment results to be reviewed as
`soon as possible for the overall welfare of the patient. There typically would not be enough time
`to send laboratory vials to a central laboratory for analysis, and to wait for these results to be
`sent back to the clinical sites for investigator review and potential subsequent action.
`
`
`
`Reference ID: 3247982
`
`12
`
`

`

`Consequently, local laboratories were utilized at the clinical sites themselves. The resulting
`drawback, of course, pertains to having the aforementioned two different laboratory assay
`methods (i.e. indirect and direct) in addition to having slightly different normal reference ranges
`across these clinical sites. The potential impact of the two different laboratory assay methods
`was assessed by an exploratory subgroup analysis (see results in Table 5 below in Section 3.2.4).
`To combat the slightly different normal reference ranges, the measured ammonia values
`themselves were normalized to a standard laboratory reference range before conducting the
`primary efficacy analyses. The method of normalizing these values is presented below in Section
`3.2.2.2. Although pre-specified by the applicant through Hochberg’s multiplicity adjustment
`procedure, the overall correlation endpoint was really exploratory in nature and tested for
`possible future utilization of U-PAGN for dose selection and dose adjustment purposes. The
`correlation, whether overall or by individual treatment group, only indicates a possible linear
`association and does not confirm a treatment benefit. As such, the applicant never considered
`this endpoint for labeling purposes and hence the label does not reflect these correlation results.
`
`
`3.2.2 Statistical Methodologies
`
`3.2.2.1 Analysis Sets
`The primary analysis set, i.e. the analysis set used for all primary and key secondary endpoint
`analyses, was the Intent-to-Treat (ITT) analysis set which includes all randomized patients who
`receive at least one dose of either study drug (HPN-100 or NaPBA). In this analysis set, patients
`were included in the treatment group, based on period, within the treatment sequence that they
`were randomized to receive regardless of actual treatment sequence received.
`
`All efficacy analyses were confirmed by utilizing the Per-Protocol (PP) analysis set which
`includes all patients in the ITT set who received both study medications (HPN-100 and NaPBA)
`and met all of the following criteria:
`• Completed the study
`• Actually received the treatment sequence that they were randomized to receive
`• Had a calculable NH3 AUC for both treatment periods
`• Had at least 4 ammonia samples one of which must be at either the 8 or 12 hour post-dose
`time point on Days 14 and 28
`• On Days 14 and 28, time zero (i.e. pre-dose) ammonia sample drawn not more than 60
`minutes after drug dosing and breakfast, and the 24 hour post-dose ammonia sample
`drawn not more than 60 minutes after drug dosing and breakfast
`• Had been compliant with study medication ≥80% on Days 14 and 28
`• Had not used sodium benzoate on either Day 14 or Day 28
`In this analysis set, patients were included in the treatment group, based on period, within the
`treatment sequence that they actually received.
`
`All analyses were re-conducted, for sensitivity analysis purposes, utilizing an All-Randomized
`analysis set which includes all patients who were randomized into the study. In this analysis set,
`patients were included in the treatment group, based on period, within the treatment sequence
`that they were randomized to receive regardless of actual treatment sequence received.