Molecular Genetics and Metabolism 100 (2010) S65–S71
`
`Contents lists available at ScienceDirect
`
`Molecular Genetics and Metabolism
`
`j o u r n a l h o m e p a g e : w w w . e l s ev i e r . c o m / l o c a t e / y m g m e
`
`Minireview
`Nitrogen sparing therapy revisited 2009
`
`Gregory M. Enns *
`
`Division of Medical Genetics, Department of Pediatrics, Lucile Packard Children’s Hospital, Stanford University, 300 Pasteur Drive, H-315, Stanford, CA 94305-5208, USA
`
`a r t i c l e
`
`i n f o
`
`a b s t r a c t
`
`Article history:
`Received 4 January 2010
`Accepted 8 February 2010
`Available online 13 February 2010
`
`Keywords:
`Hyperammonemia
`Urea cycle disorders
`Alternative pathway therapy
`Survival
`Outcome
`
`Although the protocol that most experienced metabolic centers in the United States follow for treating
`acute hyperammonemia in urea cycle disorders (UCDs) is similar to that proposed by Brusilow and Bat-
`shaw in the early 1980s, over the years a steady evolution has taken place. Continued developments in
`intensive care, surgical and hemodialysis techniques, fluid and electrolyte management, cardiovascular
`support, and emergency transport have contributed to improved management of acute hyperammone-
`mia. Compared to historical data, survival of urea cycle patients has also improved following treatment
`with alternative pathway therapy, in addition to appropriate supportive care, including the provision of
`adequate calories to prevent catabolism and promote anabolism and hemodialysis if needed. However,
`overall neurological outcomes have been suboptimal. There are currently a number of exciting prospec-
`tive new therapies on the horizon, including novel medications or cell-based treatments. Nevertheless,
`the therapeutic expertise that is currently in place at centers specializing in management of metabolic
`emergencies already has the potential to improve survival and outcome in these children significantly.
`The early identification of UCD patients so that transport to a metabolic treatment center may be carried
`out without delay continues to be a major area of focus and challenge.
`Ó 2010 Elsevier Inc. All rights reserved.
`
`Acute hyperammonemia in urea cycle disorders (UCDs) is a
`medical emergency that requires rapid institution of multiple
`treatment modalities, including intravenous ‘‘nitrogen-scaveng-
`ing” medications, provision of adequate calories to prevent catab-
`olism (block production of endogenous ammonium) and promote
`anabolism, and hemodialysis, in order to treat effectively [1]. His-
`torically, mortality and morbidity associated with UCDs have been
`high, with survivors often exhibiting devastating neurological se-
`quelae [2]. Alternative pathway therapy has revolutionized the
`acute and chronic treatment of patients who have UCDs. UCD pa-
`tients have improved survival following therapy with alternative
`pathway therapy [3], but overall neurological outcomes have been
`suboptimal [4,5]. After touching on initial clinical trials using
`alternative pathway therapy, this review concentrates on current
`therapeutic strategies for management of acute and chronic hyper-
`ammonemia secondary to UCDs, as well as newer developments
`that have the potential to improve survival and neurological
`outcome.
`
`Abbreviations: AL, argininosuccinate lyase; AS, argininosuccinate synthetase;
`ASA, argininosuccinic acid; AUC, area under the time concentration curve; CPSI,
`carbamyl phosphate synthetase I; OTC, ornithine transcarbamylase; PA, phenylac-
`etate; PAGN, phenylacetylglutamine; PB, phenylbutyrate; UCD, urea cycle
`disorder.
`* Fax: +1 650 498 4555.
`E-mail address: greg.enns@stanford.edu
`
`1096-7192/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
`doi:10.1016/j.ymgme.2010.02.007
`
`Initial studies
`
`In 1979, Brusilow et al. suggested that patients who have UCDs
`may benefit if the synthesis of non-urea nitrogen-containing
`metabolites whose excretion rates are high could be promoted [6].
`The two main classes of compounds that could utilize alternative
`mechanisms of waste nitrogen disposal are (i) the urea cycle inter-
`mediates arginine and citrulline, and (ii) the amino acid acylation
`products hippuric acid and phenylacetylglutamine (PAGN) [7].
`Brusilow and Batshaw first demonstrated the potential of alter-
`native pathway therapy in two neonates with argininosuccinate
`lyase (AL) deficiency [8]. They hypothesized that argininosuccinic
`acid (ASA) might serve as a vehicle for the excretion of waste nitro-
`gen in patients with AL deficiency. ASA contains two waste nitro-
`gen atoms and has a renal clearance similar to the glomerular
`filtration rate. Because of the location of the enzymatic deficiency
`in AL deficiency, excess arginine would promote ASA synthesis
`and, hence, increased waste nitrogen excretion. Indeed, intrave-
`nous arginine therapy, in combination with peritoneal dialysis in
`one of the patients, resulted in a fall of plasma ammonium and glu-
`tamine to normal [8]. In essence, ASA served as a ‘large urea’
`molecule.
`Soon thereafter, the potential for amino acid acylation products
`to provide a mechanism for the synthesis and excretion of waste
`nitrogen was demonstrated in a teenager with carbamyl phosphate
`synthetase I (CPS I) deficiency and four additional patients in
`
`Horizon Exhibit 2009
`Lupin v. Horizon
`IPR2018-00459
`
`1 of 7
`
`

`

`S66
`
`G.M. Enns / Molecular Genetics and Metabolism 100 (2010) S65–S71
`
`hyperammonemic coma [9]. Treatment with sodium benzoate or
`phenylacetate (PA) resulted in increased urinary nitrogen excre-
`tion and clinical improvement with decreased plasma ammonium
`levels. Hippurate and PAGN, amino acylation products of benzoate
`and PA respectively, were found to account for the increase in uri-
`nary nitrogen [9]. Larger trials confirmed the initial findings.
`Twenty-six infants with urea cycle disorders were treated with
`intravenous benzoate and arginine hydrochloride for periods of
`time ranging from 7 to 62 months. The majority (23/26) also
`underwent peritoneal dialysis. Chronic management consisted of
`sodium benzoate supplementation, a low-protein diet, and provi-
`sion of adequate calories. Overall, 85% (22/26) of the patients sur-
`vived, but neurological impairment was common with 59% (13/22)
`demonstrating normal intelligence [10]. In 1984, Brusilow et al.
`described the response of seven further children with UCDs and
`moderate hyperammonemia (plasma ammonium levels 101–483
`lmol/L) who were treated with a protocol similar to the one still
`in use today. Intravenous sodium benzoate, sodium PA and argi-
`nine hydrochloride were used for treating 12 acute hyperammone-
`mic episodes
`in these children.
`In addition, nitrogen-free
`intravenous fluids containing 10% dextrose were used to provide
`caloric support (at least 40 kcal/kg/day). Two patients had rebound
`hyperammonemia within 7 h after their initial response to therapy
`and required peritoneal dialysis. One patient died despite therapy,
`but the others recovered following the use of this regimen [11].
`
`Current acute management
`
`Hyperammonemia is a medical emergency that requires imme-
`diate, coordinated action by a trained multi-disciplinary team in
`order to deliver appropriate treatment without delay. Ideally, ther-
`apy should be provided by a specialized center with experience in
`treating acute hyperammonemia. Metabolic centers typically have
`a protocol in place that outlines each therapeutic intervention in
`detail. Placement of intravenous lines, preferably central, is essen-
`tial in order to provide adequate calories through the use of nitro-
`gen-free fluids containing dextrose (e.g. 10% dextrose with
`appropriate electrolytes initially) and intravenous lipids. In addi-
`tion, venous access allows for treatment with alternative pathway
`medications and other support, such as vasopressor agents or buf-
`fering agents, depending on the cardiovascular and acid–base sta-
`tus of the patient [1,12]. At Lucile Packard Children’s Hospital
`(LPCH), we now send our transport team out with the appropriate
`dose of sodium PA and sodium benzoate (AmmonulÒ) with added
`arginine hydrochloride so that a bolus infusion of these medica-
`tions can be delivered en route. Upon arrival at our center, or soon
`thereafter, a post-bolus plasma ammonium level may be checked
`in order to help guide the course of further therapy, especially
`the need for hemodialysis. Often hemodialysis, or at least prepara-
`tion for dialysis, is initiated concomitantly with AmmonulÒ and
`arginine therapy. Hemodialysis (HD) has a very high ammonium
`clearance rate, as compared with other methods such as peritoneal
`dialysis or hemofiltration (HF) [13–15]. Following conventional
`hemodialysis, institution of hemofiltration may prevent rebound
`hyperammonemia [16]. Both sodium PA and sodium benzoate ex-
`hibit high clearance by HD and HF, but this typically does not inter-
`fere with resolution of hyperammonemia [16]. Use of a high-flow
`extracorporeal oxygenation (ECMO) circuit to support HD has also
`been used successfully in neonates with hyperammonemia. The
`addition of the ECMO circuit may further increase ammonium
`clearance while simultaneously improving oxygenation [17].
`The importance of providing calories to prevent catabolism and
`promote anabolism cannot be stressed too highly. A goal of at least
`80 kcal/kg/day is often used in neonates [7], although a higher
`caloric intake (e.g. 120 kcal/kg/day) may be required. We com-
`monly use intravenous fluids with a dextrose concentration of
`
`20% in addition to intravenous lipids (2–3 g/kg/day) in order to
`achieve this caloric goal. An insulin drip may be needed to control
`blood glucose levels. Other supportive care includes intubation and
`ventilation as necessary, correction of anemia, electrolyte and acid/
`base imbalances, maintenance of blood pressure, and treatment of
`any intercurrent illness [1].
`Complete restriction of protein is also a mainstay of acute therapy,
`although this should be limited to only about 24 h after the initiation
`of treatment. Thereafter, protein supplementation is provided via
`parenteral nutrition or nasogastric feeds. In general, transition to ent-
`eral feeds should be done as early as possible. If protein restriction is
`prolonged, depletion of essential amino acids will occur with resul-
`tant further protein catabolism and nitrogen release [1].
`
`Outcome following acute management
`
`In 2005, Nassogne et al. published a report describing the out-
`come of a large cohort of urea cycle patients who were not treated
`using alternative pathway medications [18]. Between 1972 and
`2000, 217 patients were diagnosed with a UCD; 121 had neonatal-
`onset disease and 96 had late-onset disease. Overall, outcome was
`poor with mortality reaching 84% (60% if males with OTC deficiency
`are excluded) in neonatal-onset cases and 28% in those who had
`late-onset forms of disease. There was a high risk of neurological
`impairment in survivors [18]. During the period of study, intrave-
`nous AmmonulÒ and arginine therapy were not readily available
`in Europe. A difference in philosophy regarding the treatment of pa-
`tients in hyperammonemic crisis, compared to some centers in the
`United States, may also have played a role in the choice of manage-
`ment for the UCD patients described in this report. In addition, there
`may have been delays in transporting hyperammonemic patients to
`experienced metabolic centers. The possibility of permanent brain
`damage is high in such cases, so urgent intervention is less likely
`to lead to good outcome, and, therefore, less likely to be used.
`In contrast, an open-label, uncontrolled, non-randomized clini-
`cal trial of AmmonulÒ and arginine hydrochloride therapy was per-
`formed in 118 hospitals in the United States and Canada from
`August, 1980 to March, 2005 [3]. Treating physicians were self-se-
`lected. To enroll a patient, the investigator contacted Dr. Brusilow
`at Johns Hopkins School of Medicine (from 1982 to 1996) or Ucyc-
`lyd Pharma (from 1997 to 2005). During the 25-year period of
`study, 299 urea cycle patients (93 neonates, 237 patients >30 days
`old, including 31 who were treated both as neonates and later)
`were followed. In total, they sustained 1181 episodes of hyperam-
`monemia. The recommended use of alternative pathway therapy
`for this study was similar to the initial treatment protocol pro-
`posed by Brusilow and Batshaw [11]. Neonates and young children
`(weighing up to 20 kg) with CPS I deficiency, ornithine transcar-
`bamylase (OTC) deficiency, and argininosuccinate synthetase (AS)
`deficiency were treated with an intravenous loading dose of
`250 mg/kg AmmonulÒ over a period of 90–120 min. Older children
`and adults (weighing >20 kg) were treated with an intravenous
`loading dose of AmmonulÒ at 5.5 g/m2 or 250 mg/kg over a period
`of 90–120 min. After the loading dose, maintenance infusions of
`the same dose over 24 h were continued until the patient was no
`longer hyperammonemic and oral therapy could be tolerated. Spe-
`cific guidelines for administering sodium PA and sodium benzoate
`were not given for treatment of AL deficiency or arginase defi-
`ciency. Loading and maintenance infusions also contained arginine.
`Dialysis was recommended for any neonate with hyperammone-
`mic encephalopathy or any other patient whose ammonium level
`did not decrease substantially within 8 h after the load infusion [3].
`Overall patient survival was 84% (250 of 299 patients), while
`hyperammonemic episode survival was 96% (1132 of 1181 epi-
`sodes). The episode survival rate was lowest (91%) for males with
`OTC deficiency. The mean number of episodes per patient was
`
`2 of 7
`
`

`

`G.M. Enns / Molecular Genetics and Metabolism 100 (2010) S65–S71
`
`S67
`
`3.3 ± 6.3 (range 1–79). Patients >30 days old were more likely to
`survive a hyperammonemic episode compared to neonates (sur-
`vival rates 98% and 73%, respectively, P < 0.001) (Fig. 1). Patients
`>12 years old were most likely to survive a hyperammonemic epi-
`sode (survival rate 99%, P < 0.001 vs. all other age groups). Among
`patients who were comatose on admission, no coma was present at
`the time of discharge in 97 episodes (81%) while 23 episodes re-
`sulted in death (19%). Survival of hyperammonemic episodes was
`lowest for males with OTC deficiency who were comatose on
`admission (68%). Survival was significantly improved for patients
`who had episodes with a peak plasma ammonium level
`6500 lmol/L compared to those with higher peak ammonium lev-
`els (P < 0.001). Patients <30 days old with a peak ammonium level
`>1000 lmol/L were least likely to survive a hyperammonemic epi-
`sode (38% episode survival) (P < 0.001) (Fig. 2). Dialysis (including
`standard hemodialysis, various combinations of arteriovenous and
`venovenous hemofiltration, and peritoneal dialysis) was used in
`136 of 1181 episodes (12%) and in 105 of 299 patients (35%). Dial-
`ysis was used more commonly in neonates (60% of episodes) than
`in older patients (7% of episodes) [3].
`Most patients experienced adverse events while being treated
`for hyperammonemia; metabolism, nervous system, and respira-
`tory disorders were most frequently reported. Co-morbid features
`were common in patients who died, including seizures (19/49),
`infection (18/49), cerebral edema or increased intracranial pres-
`sure (16/49), disseminated intravascular coagulation (9/49), kid-
`ney failure (6/49), multi-organ system failure (5/49), and cerebral
`hemorrhage (5/49). Although cerebral edema or increased intra-
`cranial pressure was documented in only 16 of 49 death narratives,
`these patients had markedly elevated ammonium levels, so cere-
`bral edema was likely present in nearly all cases. An overdose of
`AmmonulÒ was reported in 13 patients who died. Massive over-
`dose was uncommon, being noted in two episodes (doses between
`9 and 17 times the recommended dose of AmmonulÒ were given in
`these instances). It is likely that the instances of mild overdosing
`(e.g. one or two additional bolus infusions given over several days)
`reflected the severity of the hyperammonemic episode and poor
`clinical status of patients who eventually died [3].
`Survival was clearly improved, compared to historical data, fol-
`lowing treatment with alternative pathway therapy, in addition to
`the provision of appropriate nutrition and, in some cases, dialysis
`
`Fig. 2. Hyperammonemic episodes survived according to peak ammonium levels.
`Patient survival of hyperammonemic episodes depends on peak plasma ammonium
`level, with significantly improved survival for those with peak plasma ammonium
`levels of 500 lM or less (P < 0.001) [3]. Reproduced with permission from Enns
`et al., N. Engl. J. Med. 356(22) (2007) 2282–2292. CopyrightÓ 2007 Massachusestts
`Medical Society. All rights reserved.
`
`[3]. It must be stressed that alternative pathway therapy is only
`a single component of the coordinated therapeutic regimen neces-
`sary for optimal treatment of acute hyperammonemia. Because pa-
`tients were primarily treated in specialized metabolic centers, the
`high survival rate likely reflects, at least in part, the expertise avail-
`able at treating institutions. In addition, these survival statistics
`apply only to patients who received the study drug and may not
`necessarily extrapolate to all UCD patients. Some patients may
`not have been treated because of their poor condition on presenta-
`tion and others may have died before reaching the hospital.
`
`Longitudinal survival data
`
`Summar et al. reported longitudinal survival findings in 260
`UCD patients using a subset of the same database used in the study
`described above [19]. Hyperammonemia presenting in the neona-
`tal period was associated with the worst outcome; only 35% of pa-
`
`Fig. 1. Survival of episodes of hyperammonemia. Hyperammonemic episode survival is shown according to age, diagnosis, and comparison of first episode vs. recurrent
`episodes [3]. Reproduced with permission from Enns et al., N. Engl. J. Med. 356(22) (2007) 2282–2292. CopyrightÓ 2007 Massachusetts Medical Society. All rights reserved.
`
`3 of 7
`
`

`

`S68
`
`G.M. Enns / Molecular Genetics and Metabolism 100 (2010) S65–S71
`
`tients were still alive at the final follow-up time point (approxi-
`mately 11 years after the start of the study period). The great
`majority of deaths in this group (22/26) occurred before age
`2 months. In contrast, 87% of patients who presented later in in-
`fancy were alive at the final follow-up time point. Seventy-eight
`percent of patients presenting with hyperammonemia between
`ages 2 and 16 years survived to the final follow-up time point.
`When specific disorders were studied, males with OTC deficiency
`had the lowest survival (53%), compared to OTC females (74%),
`and those with CPS I deficiency (61%) and AS deficiency (78%) [19].
`
`The Lucile Packard Children’s Hospital experience
`
`Over the past decade, we have treated 25 neonates with hyper-
`ammonemia at Lucile Packard Children’s Hospital at Stanford; 11
`with UCDs, 12 with organic acidemias, and two with long-chain
`fatty acid oxidation defects. Hemodialysis, typically followed by a
`form of hemofiltration, was performed in most of the UCD patients
`(9/11, 82%) and in 58% of those with organic acidemias (7/12), but
`was not required in the fatty acid oxidation patients, even though
`both of these patients had significant hyperammonemia (600 to
`1000 lmol/L). There was one neonatal death (a girl who presented
`with seizures and was diagnosed with CPS I deficiency). Three UCD
`patients died later (two had severe OTC deficiency and the third
`was given a lethal overdose of BuphenylÒ by an outside institu-
`tion). Three organic acidemia patients also died during childhood
`(all three had methylmalonic acidemia mut0 subtype). Both fatty
`acid oxidation patients have survived to date. In this relatively
`small cohort of hyperammonemic patients, neonatal survival was
`96% and overall survival to date is 72%.
`
`Neurological outcomes
`
`Survival has clearly improved compared to historical outcomes
`following alternative pathway therapy, when these medications
`are administered as part of a multi-faceted therapeutic protocol.
`The goal, however, is not only survival, but also optimization of neu-
`rological outcome. Unfortunately, reported IQ data following sur-
`vival of an episode of neonatal hyperammonemic coma have been
`sobering. In 24 UCD children tested at ages 12–74 months following
`treatment with alternative pathway therapy, the mean IQ was
`43 ± 6, with only 21% having an IQ > 70 [4]. Moreover, 79% of pa-
`tients had at least one developmental disability. The depth of coma
`was found to correlate inversely with IQ, with patients with a his-
`tory of stage III or IV coma having the worst neurological outcome.
`Although prolonged hyperammonemic coma was found to be asso-
`ciated with brain damage and impaired intellectual function, the
`investigators speculated that poor outcomes may be prevented by
`early diagnosis and institution of appropriate therapy [4].
`
`Chronic therapy
`
`After the initial metabolic crisis has been treated, surviving pa-
`tients are started on a chronic therapeutic regimen. This consists of
`protein restriction (along with provision of approximately 50% of
`protein intake from an essential amino acid formula), supplemen-
`tation with citrulline or arginine depending on the precise diagno-
`sis, use of alternative pathway medications, and prompt
`recognition and treatment of any intercurrent illness [20]. The
`treatment regimen must be individualized. Many centers, includ-
`ing ours, initially administer BuphenylÒ (sodium phenylbutyrate)
`as the primary nitrogen-scavenging medication once AmmonulÒ
`has been stopped. In some cases, the addition of sodium benzoate
`may be required to help control hyperammonemia. The tendency
`to hyperammonemia may be controlled in other patients by using
`
`sodium benzoate alone, although in our experience this is usually
`possible only in the more mild forms of disease, such as manifest-
`ing heterozygote females with OTC deficiency.
`Poor adherence to the prescribed doses of alternative pathway
`medications has been a major difficulty in treating patients with
`urea cycle defects. BuphenylÒ requires frequent dosing and has
`an unpleasant taste and odor. Patients may be faced with taking
`up to 40 large tablets of BuphenylÒ daily in order to manage their
`underlying disease. Some patients may prefer sodium benzoate,
`although this medication also has a noxious taste. Another poten-
`tial problem has been documented more recently; alternative
`pathway medications may cause a selective deficiency of
`branched-chain amino acids [21]. Branched-chain amino acid sup-
`plementation may improve protein titration in UCD patients and is
`discussed in detail by Scaglia in this supplement.
`
`Missing metabolites
`
`After an oral dose of sodium phenylbutyrate, the concentration
`of phenylbutyrate (PB) peaks at 50–90 min and PA and PAGN peak
`at 150–180 min [22]. Although PAGN is the major product of PA
`metabolism, by conjugation of phenylacetyl-CoA with glutamine,
`a small amount of unchanged PA and PB is also excreted in the ur-
`ine. However, taken together the urinary excretion of PA, PB, and
`PAGN accounts for less than 50% of the ingested PB. The mystery
`of the ‘‘missing” PB has been solved only partially. In 2002, the no-
`vel metabolite phenylbutyrylglutamine was identified, although
`only about half of ingested PB was accounted for by the addition
`of this compound [22]. More recently, a number of other PB metab-
`olites, formed by interaction of PB with lipids and carbohydrates,
`have been identified. These new metabolites are either glucuron-
`ides (e.g. phenylbutyryl- or phenylacetyl-b-glucuronate) or b-oxi-
`dation
`side
`products
`(e.g.
`R-3-hydroxy-4-phenylbutyrate,
`phenylisopropanol). Nevertheless, even with the addition of these
`carbohydrate and lipid derivatives, the total recovery of PB and
`its metabolites accounted for only 62% of the ingested PB dose
`[23]. Therefore, it is likely that other metabolites of PB remain to
`be discovered. In addition, some of the ingested PB, metabolites
`of PB, or both may be excreted in feces. The fate of the PB glucuro-
`nide metabolites excreted in bile is unknown [23].
`
`HPN-100 (glyceryl tri-[4-phenylbutyrate])
`
`HPN-100 (glyceryl tri-[4-phenylbutyrate], glycerol phenylbuty-
`rate) is a new oral medication being developed for hyperammone-
`mia control in UCDs. Clinical trials in adult patients are currently
`underway. This investigational agent is a pro-drug of PB; glycerol
`phenylbutyrate is a triglyceride that contains three molecules of
`4-phenylbutyric acid joined via ester linkage to glycerol. Impor-
`tantly, glycerol phenylbutyrate is an organic liquid (oil) with little
`odor or taste that has the potential to deliver the same amount of
`phenylbutyrate in a much reduced amount of medication. The
`maximum anticipated dose of glycerol phenylbutyrate is 5.8 ml
`three times daily, which provides the equivalent to 20 g of PB
`(i.e. a little over three teaspoons daily provides the equivalent of
`40 tablets of BuphenylÒ). Glycerol phenylbutyrate also does not
`contain sodium. Greater patient acceptance and compliance may
`result from the combination of improved ease of administration
`and minimal odor.
`Pharmacology data from Cynomolgus monkeys, which have the
`capacity to convert PA to PAGN, suggest that glycerol phenylbuty-
`rate may act as a ‘slow release’ compound that may be metabolized
`to PAGN more efficiently than PB [23]. Ester hydrolysis of glycerol
`phenylbutyrate results in release of PB, which then can undergo b-
`oxidation to PA, the active compound. A Phase 2 study compared
`
`4 of 7
`
`

`

`G.M. Enns / Molecular Genetics and Metabolism 100 (2010) S65–S71
`
`S69
`
`plasma ammonium values, plasma and urine PAGN, and amino
`acid levels in 10 adult urea cycle patients who switched from their
`prescribed dose of BuphenylÒ to an equivalent dose of glycerol
`phenylbutyrate based on amount of PB. After seven days of treat-
`ment with three times daily glycerol phenylbutyrate, 24-h phar-
`macokinetic data was obtained and serial plasma ammonium
`levels were measured. Treatment with glycerol phenylbutyrate
`resulted in 30% lower plasma ammonium values as assessed
`by time-normalized AUC (mean NH3 on BuphenylÒ 38.4 ±
`19.6 lmol/L vs. 26.1 ± 10.3 lmol/L on glycerol phenylbutyrate,
`not statistically significant), similar plasma PAGN and amino acid
`levels, and similar urinary excretion of PAGN. Urinary PAGN ac-
`counted for 55% of the PB administered as BuphenylÒ or glycerol
`phenylbutyrate. Interestingly, there was a trend towards lower
`nocturnal ammonium values in subjects while being treated with
`glycerol phenylbutyrate. Somewhat fewer adverse events were re-
`ported during the glycerol phenylbutyrate period of this trial (21
`adverse events in seven subjects during BuphenylÒ treatment com-
`pared to 15 adverse events in five subjects during glycerol phen-
`ylbutyrate treatment). The only two hyperammonemic events
`during this study occurred in subjects being treated with Buphe-
`nylÒ and both episodes were attributed to medication non-adher-
`ence
`[24].
`In conclusion, glycerol phenylbutyrate plasma
`ammonium control appears to be at least comparable to Buphe-
`nylÒ. The trend toward lower overnight ammonium values may re-
`flect delayed release of PB following ester hydrolysis. This study
`also noted that eight of 10 subjects were prescribed BuphenylÒ
`in doses below the recommended treatment guidelines according
`to the package insert. Because 30% to 40% of plasma ammonium
`values were abnormal in these patients, higher medication doses
`may be beneficial. Further development of glycerol phenylbutyrate
`is warranted, especially given the difficulties facing urea cycle pa-
`tients with respect to daily pill burden and the noxious taste of
`BuphenylÒ. If the promise of the initial Phase 2 results is confirmed
`in larger-scale trials, glycerol phenylbutyrate has the potential to
`improve both medication adherence and, therefore, control of plas-
`ma ammonium levels.
`
`The future of hyperammonemia therapy
`
`In a sense, the future of UCD therapy is now. Specialized meta-
`bolic centers already have the clinical and technical expertise in
`place to cope with acute hyperammonemia, although more of such
`centers and improved access to those already in place are clearly
`needed [25]. Experienced metabolic centers typically have hyper-
`ammonemia protocols in place so that a coordinated response by
`multiple care providers (especially specialists in biochemical
`genetics, neonatology, surgery and nephrology) can proceed
`smoothly. Electronic order sets may further improve the delivery
`of multi-disciplinary care. Improvement is still needed with re-
`spect to the timeliness of identification of patients with elevated
`ammonium levels so that care can be instituted without delay. Ex-
`panded newborn screening has the potential to detect some UCDs
`based on elevations of citrulline or arginine, but at present proxi-
`mal defects (CPSI and OTC deficiencies) are not identified by tan-
`dem mass spectrometry, although techniques for doing so are
`under development. Even so, patients with neonatal onset forms
`of disease typically become symptomatic before results of new-
`born screening are known. Therefore, although further expanding
`newborn screening to include the proximal defects and improving
`the turn around time for obtaining test results may have some ef-
`fect on early identification of affected neonates, in practice out-
`comes may not be changed significantly unless there is increased
`awareness of these disorders by primary medical care providers.
`The connection between community hospitals and specialized
`treatment centers needs further strengthening.
`
`Rapid institution of alternative pathway therapy and adequate
`caloric support has the potential to improve outcome. For example,
`our institution strives to send transport teams out with a prepared
`supply of AmmonulÒ and arginine hydrochloride so that the pa-
`tient receives the initial bolus during transport. If the initial bolus
`is provided in such a manner, the treating physician may be able to
`make a decision on the need to institute hemodialysis upon arrival
`to the hospital. Of course, in some instances plasma ammonium
`will be at levels that clearly would not respond to alternative path-
`way therapy regardless of the time course of the intervention. In
`such instances, a decision about whether or not to proceed with
`dialysis will have to be made after careful evaluation of potential
`risks, benefits, and likely outcome. Hemodialysis, hemofiltration,
`and hemodiafiltration have been extremely effective treatments
`for acute management of hyperammonemia and techniques con-
`tinue to improve [26,27]. Mild systemic hypothermia, in addition
`to hemodialysis or hemofiltration, is another novel technique that
`may be useful
`in correcting hyperammonemia [28], but this
`modality has yet to be studied in UCD patients in a controlled clin-
`ical trial.
`Prenatal delivery of nitrogen-scavenging medications may also
`play a role in the initial care of UCD patients if a diagnosis is known
`before birth. Two fetuses with a prenatal diagnosis of a UCD were
`treated by infusing their mothers with intravenous benzoate. Ther-
`apeutic benzoate concentrations were detected in umbilical cord
`blood and in the blood of the neonates. Plasma ammonium and
`glutamine concentrations were normal in the neonatal period for
`these children [29]. Novel medications, such as glycerol phenylbu-
`tyrate, may also prove useful in the chronic management of UCD
`patients.
`Liver transplantation has become an increasingly utilized ther-
`apeutic option for definitive correction of the underlying defect in
`UCDs, if the patient survives the initial hyperammonemic episode
`[30]. Advances in immunosuppression, operative techniques, and
`post-operative intensive care have contributed to improved out-
`come following transplantation [31]. Post-transplantation survival
`rates of patients who have inborn errors of metabolism appear to
`be higher, approaching 100%, when compared to survival following
`transplantation for other indications, such as acute liver failure,
`extrahepatic biliary atresia, or postnecrotic liver cirrhosis [32]. Im-
`proved survival following liver transplantation in metabolic disor-
`ders may be related to the presence of normal anatomy, the ability
`to perform elective surgery during periods of relative clinical sta-
`bility, or other factors. Preservation of intelligence is possible if
`transplantation is performed early, before neurological damage
`has been sustained [30]. Although hepatocyte transplantation is a
`promising new approach for the treatment of liver-based meta-
`bolic disorders, including UCDs, only limited success has been re-
`ported to date [33]. Gene therapeutic approaches also hold
`promise for the future, but significant hurdles need to be over-
`come. This topic is discussed more fully elsewhere in th