SEARCHCOHORTmedical genetics, pediatrics

cohort·medical genetics, pediatrics, metabolic medicine, endocrinology·PMC10000623

Hyperphenylalaninemias genotyping: Results of over 60 years of history in Lombardy, Italy

Endocrinology, Diabetes & Metabolism · 13 authors, 3 centres

AI SUMMARY

FIDELITY 92%

POPULATION826 patients with hyperphenylalaninemia in follow-up at San Paolo Hospital, Milan, Italy (born from late 1970s to 2020)

INTERVENTIONGenotyping of PAH, QDPR, PTS, GCH1, PCBD1, and DNAJC12 genes; APV/GPV phenotype prediction

COMPARISONExpected phenotype based on GPV vs. actual presenting phenotype based on pre-treatment phenylalanine levels

This summary was generated by AI from a single paper. It has not been reviewed by a clinician and is not clinical advice. Verify against the source before acting on it.

AGENT 01 · STRUCTURED SUMMARY

TRACE TO SOURCE

“

TL;DR

This study analyzed PAH gene mutations in 826 hyperphenylalaninemia patients from Lombardy, Italy, identifying 166 variants including 7 novel ones. The most common variant was p.Ala403Val, and the APV/GPV genotype-prediction system matched actual phenotypes in 85% of cases but showed notable discrepancies, particularly for variants like p.Leu48Ser and p.Tyr414Cys. These findings highlight that current predictive tools are insufficient alone and underscore the need for additional factors to improve personalized treatment decisions.

Full summary

3,263 CHARS

**Background:** Hyperphenylalaninemias (HPA) are autosomal recessive disorders caused by defects in phenylalanine metabolism, most commonly due to PAH gene mutations. Genotype-phenotype correlations using the Allelic Phenotype Value (APV) and Genotypic Phenotype Value (GPV) system have reported validity between 80% and 90%, but discrepancies remain. This study aimed to characterize the mutation spectrum in a large Italian cohort and evaluate the reliability of APV/GPV-based predictions.

**Methods:** Medical records of 826 HPA patients in follow-up at the Clinical Department of Paediatrics, San Paolo Hospital, University of Milan, Italy, between 1955 and 2020 were retrospectively reviewed. Genetic analysis included DGGE, direct sequencing, HRM, Sanger sequencing, MLPA, and NGS. Variants were classified using ClinVar, PAHvdb, LOVD, HGMD, and ACMG criteria. APV and GPV were calculated per the Garbade system (APV 0–10; cPKU: 0–2.7, mPKU: 2.8–6.6, mHPA: 6.7–10). Phenotype was classified by pre-treatment Phe levels: cPKU ≥1200 μmol/L, mPKU 600–1199 μmol/L, mHPA <599 μmol/L. Linear discriminant analysis was used to compare expected vs. actual phenotypes.

**Key Results:** Of 826 patients, 814 (99%) had PKU or HPA (mean age 18.75±12.63 years, predominantly Caucasian). Twelve patients had BH4 deficiencies (9 PTPS, 2 PCD, 1 DHPR). A total of 166 different PAH variants were identified across all 13 exons and the 3′UTR. The most frequent variant was p.Ala403Val (149 alleles, 9.81%), followed by p.Arg261Gln (126 alleles, 8.29%), p.Val245Ala (121 alleles, 7.96%), IVS10-11 g>a (103 alleles, 6.78%), p.Tyr414Cys and p.Leu48Ser (84 alleles each, 5.53%). Seven novel variants were identified: p.Leu91Pro, p.Val190Glu, p.Glu316Val, p.Trp326Cys, p.Ser359Pro, p.Lys195Glu, and p.Asp338Ala, all predicted as pathogenic/likely pathogenic. APV/GPV could be calculated for 748 patients (92%). Overall, phenotype prediction matched in 85% of cases (n=638). For GPV ≤2.7 (expected cPKU), reliability was 76% (n=169); 17% presented with mPKU and 7% with mHPA. For GPV 2.7–6.7 (expected mPKU), reliability was only 48% (n=43); 7% presented with cPKU and 45% with mHPA. For GPV ≥6.7 (expected mHPA), reliability was 98%. Among patients with GPV ≥6.7, 90% (n=394) did not require treatment and 10% (n=42) required treatment, with mean Phe levels of 195.80±60.20 μmol/L and 513.85±145.23 μmol/L, respectively. No clinically useful GPV cut-off to discriminate treatment need could be identified via ROC analysis. Variants p.Leu48Ser, p.Tyr414Cys, and p.Glu390Gly were frequently associated with unexpected phenotypes.

**Clinical Implications:** The APV/GPV system is helpful but not sufficient alone for phenotype prediction, with notable discrepancies especially in the mPKU category (only 48% accuracy). Variants such as p.Leu48Ser and p.Tyr414Cys show variable expressiveness, possibly due to allosteric mechanisms or interallelic complementation. Clinicians should interpret GPV predictions cautiously and consider additional clinical and biochemical data. The identification of 7 novel PAH variants expands the known mutation spectrum. Further research is needed to identify modifying factors and improve predictive accuracy for personalized treatment decisions.

PICO

PPOPULATION

826 patients with hyperphenylalaninemia in follow-up at San Paolo Hospital, Milan, Italy (born from late 1970s to 2020)

IINTERVENTION

Genotyping of PAH, QDPR, PTS, GCH1, PCBD1, and DNAJC12 genes; APV/GPV phenotype prediction

OOUTCOME

Genotype-phenotype correlation accuracy; identification of novel variants; discrepancies between predicted and actual phenotypes

STUDY TYPE

cohort

SPECIALTY

medical genetics

SUMMARISED BY

AI pipeline

FIDELITY CHECK

92% · A

ORIGINAL PAPER

PMC10000623

Hyperphenylalaninemias genotyping: Results of over 60 years of history in Lombardy, Italy

Endocrinology, Diabetes & Metabolism

READ THE FULL PAPER

ABSTRACT

Abstract Background Hyperphenylalaninemias (HPA) are due to several gene mutations, of which the PAH gene is the most frequently involved. Prevalence and incidence of disease vary between populations, with genotype/phenotype correlations not always capable to correctly predict disease severity. The aim of this study was to give an overview of PAH mutations among one of the largest cohort of patients among Europe, born in Lombardy (Italy) starting from late 1970 s and including over a 60 years of activity; furthermore, to evaluate and discuss identified genotype/phenotype correlations and related reliability. Patients/Methods Eight hundred and twenty‐six HPA patients in current follow‐up at the San Paolo Hospital in Milan (Italy) were retrospectively reviewed, including molecular results and allelic phenotype and genotype values (attributed on the basis of the APV/GPV system) to verify genotype–phenotype correlations. Results A total of 166 different PAH variants were reviewed; of those, seven variants were identified as not previously described in literature. Most frequently reported variant was p.Ala403Val, followed by p.Arg261Gln, p.Val245Ala, IVS10‐11 g>a, p.Tyr414Cys and p.Leu48Ser. Phenotype prediction, based on APV/GPV, matched the actual phenotype in most cases, but not always. Conclusion/Discussion The cohort of patients included in this study constitute a representative sample of the HPA population worldwide. Studies on this sample may allow to improve clinical and genetic evaluation performances for affected patients, consequently to develop personalized medicine interventions and provide more precise indications on the correct treatment approach based on the accumulated evidence, also in light of a prognostically reliable but not always conclusive APV/GPV system.

CITATION EXPORT

Valentina Rovelli, Graziella Cefalo, Vittoria Ercoli et al.. Hyperphenylalaninemias genotyping: Results of over 60 years of history in Lombardy, Italy. Endocrinology, Diabetes & Metabolism. (2022). https://doi.org/10.1002/edm2.396

AUTHORS

13 · 3 CENTRES

Valentina Rovelli, Graziella Cefalo, Vittoria Ercoli, Juri Zuvadelli, Turri Olivia, Daniela Graziani, Alberti Luisella, Davide Bassi, Alice Re Dionigi, Raed Selmi, Sabrina Paci, Elisabetta Salvatici, Giuseppe Banderali

1 Clinical Department of Pediatrics, Inborn Errors of Metabolism Unit, San Paolo Hospital, ASST Santi Paolo e Carlo University of Milan Milan Italy
2 Department of Laboratory Diagnostic Technologies San Paolo Hospital, ASST Santi Paolo e Carlo Milan Italy
3 Regional Laboratory of Newborn Screening, Department of Women, Mothers and Neonatal Care, Children's Hospital “V. Buzzi” ASST Fatebenefratelli Sacco Milan Italy