Impact of neonatal protein metabolism and nutrition on screening for phenylketonuria

Development and validation of machine-learning models of diet management for hyperphenylalaninemia: a multicenter retrospective study

BACKGROUND

Assessing dietary phenylalanine (Phe) tolerance is crucial for managing hyperphenylalaninemia (HPA) in children. However, traditionally, adjusting the diet requires significant time from clinicians and parents. This study aims to investigate the development of a machine-learning model that predicts a range of dietary Phe intake tolerance for children with HPA over 10 years following diagnosis.

METHODS

In this multicenter retrospective observational study, we collected the genotypes of phenylalanine hydroxylase (PAH), metabolic profiles at screening and diagnosis, and blood Phe concentrations corresponding to dietary Phe intake from over 10 years of follow-up data for 204 children with HPA. To incorporate genetic information, allelic phenotype value (APV) was input for 2965 missense variants in the PAH gene using a predicted APV (pAPV) model. This model was trained on known pheno-genotype relationships from the BioPKU database, utilizing 31 features. Subsequently, a multiclass classification model was constructed and trained on a dataset featuring metabolic data, genetic data, and follow-up data from 3177 events. The final model was fine-tuned using tenfold validation and validated against three independent datasets.

RESULTS

The pAPV model achieved a good predictive performance with root mean squared error (RMSE) of 1.53 and 2.38 on the training and test datasets, respectively. The variants that cause amino acid changes in the region of 200-300 of PAH tend to exhibit lower pAPV. The final model achieved a sensitivity range of 0.77 to 0.91 and a specificity range of 0.8 to 1 across all validation datasets. Additional assessment metrics including positive predictive value (0.68-1), negative predictive values (0.8-0.98), F1 score (0.71-0.92), and balanced accuracy (0.8-0.92) demonstrated the robust performance of our model.

CONCLUSIONS

Our model integrates metabolic and genetic information to accurately predict age-specific Phe tolerance, aiding in the precision management of patients with HPA. This study provides a potential framework that could be applied to other inborn errors of metabolism.

Phenylalanine Tolerance over Time in Phenylketonuria: A Systematic Review and Meta-Analysis

In phenylketonuria (PKU), natural protein tolerance is defined as the maximum natural protein intake maintaining a blood phenylalanine (Phe) concentration within a target therapeutic range. Tolerance is affected by several factors, and it may differ throughout a person's lifespan. Data on lifelong Phe/natural protein tolerance are limited and mostly reported in studies with low subject numbers. This systematic review aimed to investigate how Phe/natural protein tolerance changes from birth to adulthood in well-controlled patients with PKU on a Phe-restricted diet. Five electronic databases were searched for articles published until July 2020. From a total of 1334 results, 37 articles met the eligibility criteria (n = 2464 patients), and 18 were included in the meta-analysis. The mean Phe (mg/day) and natural protein (g/day) intake gradually increased from birth until 6 y (at the age of 6 months, the mean Phe intake was 267 mg/day, and natural protein intake was 5.4 g/day; at the age of 5 y, the mean Phe intake was 377 mg/day, and the natural protein intake was 8.9 g/day). However, an increase in Phe/natural protein tolerance was more apparent at the beginning of late childhood and was >1.5-fold that of the Phe tolerance in early childhood. During the pubertal growth spurt, the mean natural protein/Phe tolerance was approximately three times higher than in the first year of life, reaching a mean Phe intake of 709 mg/day and a mean natural protein intake of 18 g/day. Post adolescence, a pooled analysis could only be performed for natural protein intake. The mean natural protein tolerance reached its highest (32.4 g/day) point at the age of 17 y and remained consistent (31.6 g/day) in adulthood, but limited data were available. The results of the meta-analysis showed that Phe/natural protein tolerance (expressed as mg or g per day) increases with age, particularly at the beginning of puberty, and reaches its highest level at the end of adolescence. This needs to be interpreted with caution as limited data were available in adult patients. There was also a high degree of heterogeneity between studies due to differences in sample size, the severity of PKU, and target therapeutic levels for blood Phe control.

Neonatal phenylalanine wash-out in phenylketonuria

Phenylketonuria (PKU) is the most common inborn error of amino acids metabolism. PKU management aims to keep as soon as possible blood phenylalanine (Phe), a non-acutely neurotoxic metabolite, within safe ranges through a dietary Phe restriction tailored to individual dietary Phe tolerance. Information on initial neonatal management of PKU, when Phe tolerance is still unknown, is scanty. We reviewed the metabolic data from 304 patients with PAH deficiency detected at newborn screening within the last 37 years. In keeping with the general neonatal management of intoxication-type inborn errors of metabolism, initial management consisted in a Phe wash-out through the exclusive administration of normocaloric Phe-free formulas until normalization of blood Phe. Based on genotype and Phe tolerance assessed at follow-up, 55 patients had classic PKU (18%), 50 mild PKU (17%), and 199 non-PKU hyperphenylalaninemia (HPA) (65%). The duration of Phe wash-out amounted to 7 ± 2 days in classic PKU, 4 ± 2 days in mild PKU, and < 24 h in non-PKU HPA (p < 0.001). After the wash-out, dietary Phe re-introduction and its upwardly titration allowed the assessment of individual metabolic phenotype. During the first 6 years of life, Phe tolerance was stable in classic PKU (~ 200 mg/day) but increased in milder forms, allowing unrestricted diet in non-PKU HPA. Neonatal Phe wash-out in PKU ensures the earliest correction of HPA. This metabolic reset also facilitates the prompt definition of individual Phe tolerance, allowing anticipation of dietary personalization and optimization of longitudinal metabolic control.

Neonatal screening for biotinidase deficiency: A 30-year single center experience

We reviewed the outcome of newborn screening for biotinidase deficiency performed at our department since 1987. Among 1,097,894 newborns screened, 461 were recalled, and 18 were identified as affected by complete or partial biotinidase deficiency (incidence 1:61,000, false positive rate 0.04%). The common missense mutation Q456H was found in 80% of patients with profound biotinidase deficiency. Of them, one patient harbored the novel mutation M399I in compound heterozygosity (M399I/Q456H). The complex allele A171T/D444H in cis was found in two patients with profound biotinidase deficiency (in homozygosity and in compound heterozygosity with the R211H mutation, respectively) and in one patient with partial biotinidase deficiency (in compound heterozygosity with the protective allele D444H in trans). All detected patients were treated and followed up at our Center until present. Biotin therapy (10-20 mg/day) allowed the full prevention of clinical symptoms in all patients with no adverse effects. These excellent outcomes confirm that newborn screening for biotinidase deficiency is a very effective secondary prevention program.

Early Screening for Tetrahydrobiopterin Responsiveness in Phenylketonuria

Since 2007, synthetic tetrahydrobiopterin (BH4) has been approved as a therapeutic option in BH4-responsive phenylketonuria (PKU) and since 2015 extended to infants younger than 4 years in Europe. The current definition of BH4 responsiveness relies on the observation of a 20% to 30% blood phenylalanine (Phe) decrease after BH4 administration, under nonstandardized conditions. By this definition, however, patients with the same genotype or even the same patients were alternatively reported as responsive or nonresponsive to the cofactor. These inconsistencies are troubling, as frustrating patient expectations and impairing cost-effectiveness of BH4-therapy. Here we tried a quantitative procedure through the comparison of the outcome of a simple Phe and a combined Phe plus BH4 loading in a series of infants with PKU, most of them harboring genotypes already reported as BH4 responsive. Under these ideal conditions, blood Phe clearance did not significantly differ after the 2 types of loading, and a 20% to 30% decrease of blood Phe occurred irrespective of BH4 administration in milder forms of PKU. Such early screening for BH4 responsiveness, based on a quantitative assay, is essential for warranting an evidence-based and cost-effective therapy in those patients with PKU eventually but definitely diagnosed as responsive to the cofactor.

Long-term follow-up of patients with phenylketonuria treated with tetrahydrobiopterin: a seven years experience

BACKGROUND

Phenylketonuria (PKU) is an autosomal recessive disorder caused by the deficiency of phenylalanine hydroxylase that catalyzes the conversion of phenylalanine to tyrosine, using tetrahydrobiopterin (BH4) as coenzyme. Besides dietary phenylalanine restriction, new therapeutic options are emerging, such as the treatment with BH4 in subgroups of PKU patients responding to a loading test with BH4.

METHODS

A no-profit open-label interventional trial with long-term oral BH4 therapy, sponsored by the Italian Medicines Agency (AIFA), was performed in a group of 17 PKU patients resulted as BH4 responders among 46 subjects analyzed for BH4-responsiveness (prot. FARM5MATC7). We report on efficacy and safety data of BH4 therapy and analyze factors predicting BH4-responsiveness and long-term response to BH4. A BH4-withdrawal test was used as a proof of the efficacy of long-term therapy with BH4.

RESULTS

Forty-four percent of the patients responded to the 48 h-long loading test with BH4. All the phenotypic classes were represented. Genotype was the best predictor of responsiveness, along with lower phenylalanine levels at diagnosis, higher tolerance and lower phenylalanine/tyrosine ratio before the test. In BH4 responder patients, long-term BH4 therapy resulted safe and effective in increasing tolerance while maintaining a good metabolic control. The BH4 withdrawal test, performed in a subset of patients, showed that improved tolerance was directly dependent on BH4 assumption. Tolerance to phenylalanine was re-evaluated in 43.5% of patients and was longitudinally analyzed in 5 patients.

CONCLUSIONS

Long-term treatment with BH4 is safe and effective in increasing tolerance to phenylalanine. There is real need to assess the actual tolerance to phenylalanine in PKU patients to ameliorate quality of life, improve nutritional status, avoiding unnecessarily restricted diets, and interpret the effects of new therapies for PKU.

Testing for tetrahydrobiopterin responsiveness in patients with hyperphenylalaninemia due to phenylalanine hydroxylase deficiency

INTRODUCTION

Pharmacological levels of the phenylalanine hydroxylase enzyme cofactor, tetrahydrobiopterin (BH4), reduce plasma phenylalanine levels in some patients with phenylketonuria (PKU), providing the first pharmacological therapy for PKU. Responsiveness to this therapy must be determined empirically through a BH4 loading test or trial. The authors have analyzed the loading tests currently in use in light of the numerous factors that can influence their results. Sapropterin dihydrochloride is a stable, synthetic form of BH4 approved for treatment of PKU in responsive patients.

METHODS

An expert panel identified evidence from published reports of clinical experience. Reports of research involving at least 25 patients and published in English were considered.

RESULTS

In all, 14 studies met both criteria; eight employing the sapropterin dihydrochloride preparation from Schircks Laboratories and six the sapropterin dihydrochloride preparation from Biomarin/Merck Serono.

CONCLUSION

The arbitrary responsiveness definition of a >30% reduction in blood phenylalanine appears to be a good compromise between sensitivity and specificity for the initial screening test. However, individual patient characteristics should be considered when interpreting results, especially in patients with low baseline phenylalanine levels.

Impact of metabolic control on bone quality in phenylketonuria and mild hyperphenylalaninemia

OBJECTIVES

A reduction of bone mineral density of unknown etiology has been reported in phenylketonuria (PKU) by radiological techniques, whereas no data on bone density in mild hyperphenylalaninemia (HPA) are available. We aimed to assess bone condition in PKU and HPA by quantitative ultrasound (QUS), taking into account patients' clinical and biochemical features.

PATIENTS AND METHODS

Phalangeal QUS has been used for bone assessment in 78 patients affected by PKU (n = 42) or mild HPA (n = 36). For each patient, blood phenylalanine concentrations in the 2 years before the study have been recorded and related to bone assessment.

RESULTS

Overall normal bone quality has been observed in the whole study group (AD-SoS standard deviation score [SDS] 0.25 ± 1.29; BTT SDS -0.13 ± 1.08). PKU adolescents (age older than 15 years, AD-SoS SDS -0.54 ± 1.33; BTT SDS -0.85 ± 1.21) and patients with poor compliance with treatment (blood phenylalanine >10 mg/dL, AD-SoS SDS -0.47 ± 1.39; BTT SDS -0.97 ± 1.14) showed lower BTT SDS with respect to normal population (P = 0.003 and P < 0.001, respectively). Patients with PKU with good compliance with treatment (blood phenylalanine < 10 mg/dL, AD-SoS SDS 0.65 ± 1.33; BTT SDS 0.15 ± 0.94) and patients with mild HPA (AD-SoS SDS 0.44 ± 1.06 and BTT SDS 0.19 ± 0.85) showed normal bone mineral density and cortical thickness.

CONCLUSIONS

Good compliance with treatment in PKU during adolescence and adulthood is desirable because diet discontinuation is associated with bone loss. Mild HPA seems not to be complicated by bone damage.

Sudden death in medium chain acyl-coenzyme a dehydrogenase deficiency (MCADD) despite newborn screening

INTRODUCTION

Medium chain acyl-CoA dehydrogenase deficiency (MCADD) is the most frequent of the fatty acid oxidation disorders (FAOD), a group caused by defects in the mitochondrial B-oxidation of fatty acids. Fatty acid oxidation is critical in supplying energy during periods when glucose is limited or when energy needs are increased beyond the availability of glucose. In MCADD, this energy shortage can result in acute metabolic episodes or sudden death. The prevention of sudden death from MCADD served as the primary impetus to expand newborn screening. However, we have experienced sudden death in four children with MCADD despite their detection by newborn screening. The purpose of this report is to alert others to the danger of sudden death in MCADD even when it is detected by newborn screening, to identify the clinical symptoms that precede sudden death, and to examine the relationship between the newborn screening result and the risk for sudden death.

METHODS

We describe these children and their metabolic findings with emphasis on their newborn screening octanoylcarnitine (C8) level, the primary marker for newborn detection of MCADD. We also performed a literature search of cases of sudden death in MCADD in which the clinical status preceding death is described.

RESULTS

The newborn screening C8 levels in our four cases were markedly elevated, ranging from 8.4 to 24.8micromol/L (cut off<0.8micromol/L). Only two of the children were homozygous for the common c.985A>G MCAD mutation; the other two were heterozygous for this mutation. Similarly, among the eight reported cases which included MCAD genotypes, five were homozygous for the c.985A>G mutation, while two were heterozygous and one was homozygous for a splice site mutation. Vomiting 12-24h before sudden death was present in all four of our cases, and the review of reported cases of sudden death in MCADD disclosed vomiting as a frequent symptom.

CONCLUSION

We suggest that in MCADD (1) a newborn screening C8 level of 6micromol/L or greater represents particular risk of sudden death; (2) that MCAD genotypes other than homozygosity for the c.985A>G mutation are also associated with sudden death; (3) that vomiting is a frequent symptom preceding sudden death; and (4) social support and medical follow-up of these families are crucial in reducing the occurrence of sudden death.

Phenylketonuria management from an European perspective: a commentary

Phenylketonuria is discussed from an European perspective, addressing the need of common definitions of terms commonly used, the need of a world-wide guideline on the diagnosis and treatment of phenylketonuria, the differences between existing European guidelines, and day-to-day care, further directives for the near future, and changing the concept from compliance to concordance, in which patients have a more clearly defined responsibility.

Unresponsiveness to tetrahydrobiopterin of phenylalanine hydroxylase deficiency

Conflicting results have been reported concerning the efficacy of tetrahydrobiopterin (BH4), the cofactor of phenylalanine hydroxylase, for reducing phenylalanine (Phe) concentration in phenylketonuria (PKU). We aimed to test quantitatively the effects of BH4 in PKU patients. Seven fully characterized patients were selected among a population of 130 PKU subjects as harboring PKU mutations predicted as BH4 responsive and previously considered responsive to a cofactor challenge. They received a simple Phe (100 mg/kg) and 2 combined Phe (100 mg/kg) and BH4 (20 mg/kg) oral loading tests. Cofactor was administered either before or after the amino acid. The concentrations of Phe, tyrosine (Tyr), and biopterin were measured over 24 hours after loading. The comparative analysis of the loading tests showed that in all patients plasma Phe concentrations peaked within 3 hours, and fell within 24 hours by about 50% in benign, 20% in mild, and 15% in severe phenylalanine hydroxylase deficiency regardless of BH4 administration. A consistent or moderate increase of plasma Tyr, again independent of the cofactor challenge, was observed only in the less severe forms of PAH deficiency. Mean blood biopterin concentration increased 6 times after simple Phe and 34 to 39 times after combined loading tests. The administration of BH4 does not alter Phe and Tyr metabolism in PKU patients. The clearance of plasma Phe after oral loading and, as well as Tyr production, is not related to cofactor challenge but to patient's phenotype. The assessment of BH4 responsiveness by the methods so far used is not reliable, and the occurrence of BH4-responsive forms of PKU still has to be definitely proven.