Combined large neutral amino acid supplementation for phenylketonuria (PKU)
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To assess the role of LNAA in people with PKU in regard to biochemical control, tolerability of diet and neurocognitive outcomes.
Background
Description of the condition
Phenylketonuria (PKU) is the result of a deficiency of the liver enzyme phenylalanine hydroxylase (PAH) and thus may also be referred to as phenylalanine hydroxylase (PAH) deficiency (Vockley 2014). PAH is necessary to convert the amino acid phenylalanine to the amino acid tyrosine. Deficiency of this enzyme leads to the accumulation of phenylalanine, with persistently raised phenylalanine concentrations causing progressive damage to the central nervous system. Untreated, a child will suffer from seizures, learning difficulties and they will have a small head (microcephaly). In addition there may be a reduction in pigmentation due to decreased melanin production and the classic phenotype is of a child with blonde hair and blue eyes.
Phenylketonuria is inherited in an autosomal recessive manner. The overall birth prevalence of PKU in European, Chinese and Korean populations has been reported as approximately 1 in 10,000 (Hardelid 2008). The birth prevalence of PKU in South‐East England was estimated to be 1.14 (0.96 to 1.33), 0.11 (0.02 to 0.37) and 0.29 (0.10 to 0.63) per 10,000 live births among white, black, and Asian ethnic groups, respectively (Hardelid 2008). In India, the birth prevalence of PKU was reported as approximately 0.5 per 10,000 live births (Rama 2004). The global comparison of incidence of PKU showed variability in various countries and regions, from Turkey as the highest to Finland and Japan as the lowest (Williams 2008) (Table 1).
| Region | Country | Incidence |
| Asian | China | 1: 17,000 |
| Japan | 1: 125,000 | |
| Turkey | 1: 2600 | |
| Israel | 1: 5300 | |
| European | Scotland | 1: 5300 |
| Czech Republic | 1: 7000 | |
| Hungary | 1: 11,000 | |
| Denmark | 1: 12,000 | |
| France | 1: 13,500 | |
| Norway | 1 : 14,500 | |
| United Kingdom | 1 : 14,300 | |
| Italy | 1 : 17,000 | |
| Canada | 1 : 22,000 | |
| Finland | 1 : 200,000 | |
| Arabic | Up to 1: 6000 | |
| Oceania | Australia | 1: 10,000 |
Incidence of phenylketonuria (PKU) by population (Williams 2008).
PKU is considered a treatable disorder and for this reason it is part of many newborn screening programmes around the world. Standard treatment for PKU consists of a phenylalanine‐restricted diet by means of a low‐protein diet and supplementation with a synthetic protein substitute. Treatment is monitored by analysis of blood phenylalanine levels and the diet may be adjusted accordingly (Giovannini 2012; Macleod 2010). There are national and international guidelines which determine the minimum and maximum safe phenylalanine level, which is dependent upon patient age, and varies between countries (Macleod 2010; Pena 2015; Singh 2016). Strict dietary adherence is crucial for the first five years in order to ensure normal brain development. Adherence to the diet often decreases as the individual gets older with the potential consequences being reduced cognitive and executive function and issues relating to foetal health in pregnant women with poorly controlled PKU (Blau 2015).
Description of the intervention
Combined large neutral amino acids (LNAAs) include tyrosine, tryptophan, threonine, methionine, valine, isoleucine, leucine, histidine and phenylalanine (van Spronsen 2009). This review considers the use of LNAA protein substitutes which are phenylalanine‐free for managing PKU in adults and children aged 12 years and over, as an alternative to standard PKU protein substitutes. LNAAs are not recommended for children less than 12 years of age because their safety and effectiveness in this age group are not known and in the early years strict control is crucial to ensure normal neurological outcome (van Calcar 2012).
The rationale for this approach is based on the knowledge that all LNAAs share the same transport system to the brain, therefore, by providing a high concentration of all LNAAS, except phenylalanine, the transport of phenylalanine across the blood‐brain barrier is reduced (Cleary 2013; Moats 2003; Pardridge 1998; Rocha 2009; Schindeler 2007;van Spronsen 2010; Zielke 2002).
How the intervention might work
LNAA transport occurs at both the gut‐blood barrier and the blood‐brain barrier. By using phenylalanine‐free LNAA supplementation, competitive inhibition can reduce phenylalanine transport from the gut into the blood and then from the blood into the brain, thus resulting in reduced cerebral phenylalanine concentrations.
Why it is important to do this review
Phenylketonuria is a rare disease, but is one of the more common inborn errors of metabolism (Gizewska 2016). If untreated in the first five years of life, it results in neurological and cognitive impairment (Kolker 2008). In later life treatment aims to ensure healthy pregnancies in affected females and for many treatment enables adequate concentration and executive function. The PKU diet can be challenging and this review aims to ascertain if LNAA supplementation can be used to ease the specialist diet burden or as an alternative to standard PKU protein substitutes.
Objectives
To assess the role of LNAA in people with PKU in regard to biochemical control, tolerability of diet and neurocognitive outcomes.
Methods
Criteria for considering studies for this review
Types of studies
We will include both published and unpublished randomized controlled trials (RCTs) with no language or date restrictions in our search methods.
Types of participants
Children and adults diagnosed with classical PKU on newborn screening and in whom dietary treatment was initiated at diagnosis. We will exclude individuals with maternal PKU, children under 12 years of age and those treated with a pharmacological treatment such as tetrahydrobiopterin (BH4).
Types of interventions
Diet plus LNAA (any dose) versus diet plus standard protein substitute.
Types of outcome measures
Primary outcomes
-
Blood phenylalanine concentration and phenylalanine/tyrosine ratio
-
Adherence to dietary treatment
Secondary outcomes
-
Quality of life (QoL) (assessed using, e.g. the PKU‐QOL which is designed to specifically assess the impact of PKU on all aspects of the individual's life, including: PKU symptoms; the impact of low‐protein dietary restrictions; and the impact of Phe‐free amino acid supplement intake. Additional detail for the questionnaire can be found at www.proqolid.org).
Search methods for identification of studies
Electronic searches
We will identify relevant studies from the Group's Inborn Errors of Metabolism Trials Register using the term: PKU. There will be no restrictions regarding language or publication status.
The Inborn Errors of Metabolism Trials Register is compiled from electronic searches of the Cochrane Central Register of Controlled Trials (CENTRAL) (updated with each new issue of the Cochrane Library), weekly searches of MEDLINE and the prospective handsearching of one journal ‐ Journal of Inherited Metabolic Disease. Unpublished work is identified by searching through the abstract books of the Society for the Study of Inborn Errors of Metabolism conference and the SHS Inborn Error Review Series. For full details of all searching activities for the register, please see the relevant section of the Cochrane Cystic Fibrosis and Genetic Disorders Group's website.
We will undertake additional searching, including searching the metaRegister of controlled trials (mRCT) (www.controlled‐trials.com/mrct), Clinicaltrials.gov (www.clinicaltrials.gov) and the WHO International Clinical Trials Registry platform (ICTRP) (http://apps.who.int/trialsearch/) (Appendix 1).
Searching other resources
We will attempt to identify additional trials through reference lists. We will contact experts in the field of clinical nutrition for any data from published and unpublished RCTs that they may have on file. We will attempt to identify details of studies which used LNAA supplementation without sufficient evidence of effectiveness by contacting corresponding authors.
Data collection and analysis
Selection of studies
Two authors (FR and MM) will independently select trials for inclusion. These authors will independently undertake the title and abstract screening of retrieved references for inclusion. One author (FR) will obtain the full‐text of all potential eligible studies. In case of disagreements, we aim to reach agreement by consensus.
Data extraction and management
We will obtain full paper manuscripts of any titles or abstracts that appear to be relevant and the relevance of each study will be independently assessed by two authors according to the inclusion and exclusion criteria. Two authors (FR and RT) will independently record information on the studies, including author, journal and year of publication, location of study, selection and characteristics of participants, demographics, ethnicity, dose of LNAA supplement, usual ‘standard’ protein substitute, and type of LNAA supplements. Should there be disagreement, we aim to resolve these by consensus.
Assessment of risk of bias in included studies
Two authors (RF and ASM) will independently assess the risk of bias for each individual trial, using the tool available in the Review Manager software (RevMan 2014). We will consider the risk of bias for each individual trial in relation to several domains, including the generation of the random sequence generation (selection bias), allocation concealment (selection bias), blinding (detection bias), incomplete outcome data (attrition bias), selective reporting and will record any other issues which may cause a risk of bias (Higgins 2011c).
Measures of treatment effect
For continuous outcomes we will record either the mean change from baseline for each group or mean post‐treatment values and standard deviation (SD) or standard error (SE) for each group. We plan to calculate a pooled estimate of the treatment effect by calculating the mean difference (MD) or standardized mean difference (SMD) and their 95% confidence intervals (CIs). For dichotomous outcomes, we will calculate the odds ratio (OR) and the corresponding 95% CIs as a pooled estimate of the treatment effect of supplementation across trials.
Unit of analysis issues
If any cross‐over trials are included we will follow advice as recommend by Elbourne (Elbourne 2002). The preferred method of analysis for cross‐over trials will be to use the results of a paired analysis, which allows a within‐individual comparison of the treatment effect. However, if this is not possible, we will aim to use a second approach, which will involve taking data from the first cross‐over period of the trial only. A third, and less preferable approach, will be to ignore the cross‐over design and use the combined results.
Dealing with missing data
If there are missing data, in the first instance, we will contact original authors to request the relevant data or information. If we receive no response, then we will attempt to impute the missing data (according to the type of data). As per the recommendations by Higgins, we will use 'informative missingness differences in means' for continuous outcomes, and for binary outcomes the 'informative missingness odds ratio' (IMOR) to impute the missing data (Higgins 2008).
Assessment of heterogeneity
To evaluate the between‐trial heterogeneity, we will use both the Chi²‐based Q‐statistic and the I‐squared (I²) statistic. We will interpret the I² statistic as follows (Deeks 2011):
-
0% to 40% might not be important;
-
30% to 60% may represent moderate heterogeneity;
-
50% to 90% may represent substantial heterogeneity;
-
75% to 100% considerable heterogeneity.
Assessment of reporting biases
We will assess publication bias by a funnel plot based on Egger's test and will use a t‐test to determine the significance of the asymmetry. An asymmetric plot suggests possible publication bias (P value greater than or equal to 0.05 suggests no bias). We will also apply Egger’s test, in which a regression model will identify any bias using the standardized estimate of size effect as a dependent variable and the inverse of the SE as an independent variable.
Data synthesis
Based on between‐trial heterogeneity, we will use the fixed‐effects model if the studies are assumed to be homogenous and a random‐effects model when they are heterogeneous (i.e. where the P value is less than or equal to 0.10 and where the I² is less than or equal to 40% we will use a fixed‐effect model, if these values are higher, we will use the random‐effects model).
Subgroup analysis and investigation of heterogeneity
If we are able to combine a number of trials and identify a large or extreme amount of heterogeneity (as defined above), we plan to undertake subgroup analyses and stratify participants according to:
-
severity of PKU (mild or moderate ‐ phenylalanine level at diagnosis ‐ 600 to 1200 μmol/L; versus classical ‐ phenylalanine level at diagnosis ‐ over 1200 μmol/L) (Bosch 2015);
-
dose (prescribed large neutral amino acids intake (g/day) on different treatment regimens) (van Spronsen 2010).
Sensitivity analysis
If there are sufficient comparable trials, i.e. 10 or more, we will perform sensitivity analyses excluding trials with clearly inadequate allocation of concealment, blinding, randomisation method or dropouts.
Summary of findings table
We will prepare a summary of findings table to present the results for all three outcomes. We will convert results into absolute effects when possible, and provide a source and rationale for each assumed risk cited in the table(s) when presented, and use the GRADE system to rank the quality of the evidence based on the methods described in chapter 11 of the Cochrane Handbook for Systematic Reviews of Interventions. We will assess and report the quality of the evidence, using GRADEpro software and GRADE criteria to assess the quality of the evidence for each outcome: risk of bias, inconsistency, imprecision, indirectness and publication bias. Two authors (FR and ASM) independently assessed the quality of the evidence (Schünemann 2011a; Schünemann 2011b).
| Region | Country | Incidence |
| Asian | China | 1: 17,000 |
| Japan | 1: 125,000 | |
| Turkey | 1: 2600 | |
| Israel | 1: 5300 | |
| European | Scotland | 1: 5300 |
| Czech Republic | 1: 7000 | |
| Hungary | 1: 11,000 | |
| Denmark | 1: 12,000 | |
| France | 1: 13,500 | |
| Norway | 1 : 14,500 | |
| United Kingdom | 1 : 14,300 | |
| Italy | 1 : 17,000 | |
| Canada | 1 : 22,000 | |
| Finland | 1 : 200,000 | |
| Arabic | Up to 1: 6000 | |
| Oceania | Australia | 1: 10,000 |
| Incidence of phenylketonuria (PKU) by population (Williams 2008). | ||
