Phenylketonuria (PKU) is an autosomal recessive inborn error of metabolism where high phenylalanine (Phe) concentrations cause irreversible intellectual disability that can be prevented by newborn screening and early treatment. Evidence suggests that PKU subjects not adherent to treatment could be at risk of insulin resistance (IR). We studied how Phe concentrations (PheCs) relate to IR using machine learning (ML) and derived potential biomarkers. In our cross-sectional study, we analyzed subjects with neonatal diagnoses of PKU, grouped as follows: 10 subjects who adhered to treatment (G1); 14 subjects who suspended treatment (G2); and 24 control subjects (G3). We analyzed plasma biochemical variables, as well as profiles of amino acids and acylcarnitines in dried blood spots (DBSs). Higher PheCs and plasma insulin levels were observed in the G2 group compared to the other groups. Additionally, a positive correlation between the PheCs and homeostatic measurement assessments (HOMA-IRs) was found, as well as a negative correlation between the HOMA-Sensitivity (%) and quantitative insulin sensitivity check index (QUICKI) scores. An ML model was then trained to predict abnormal HOMA-IRs using the panel of metabolites measured from DBSs. Notably, ranking the features’ importance placed PheCs as the second most important feature after BMI for predicting abnormal HOMA-IRs. Our results indicate that low adherence to PKU treatment could affect insulin signaling, decrease glucose utilization, and lead to IR.

Hepatocellular carcinoma (HCC) is a major complication of tyrosinemia type 1 (HT-1), an inborn error of metabolism affecting tyrosine catabolism. The risk of HCC is higher in late diagnoses despite treatment. Alpha-fetoprotein (AFP) is widely used to detect liver cancer but has limitations in early-stage HCC detection. This study aimed to implement a machine-learning (ML) approach to identify the most relevant laboratory variables to predict AFP alteration using constrained multidimensional data from Chilean and Italian HT-1 cohorts. A longitudinal retrospective study analyzed 219 records from 35 HT-1 patients, including 8 with HCC and 5 diagnosed through newborn screening. The dataset contained biochemical and demographic variables that were analyzed using the eXtreme Gradient Boosting algorithm, which was trained to predict abnormal AFP levels (>5 ng/mL). Four key variables emerged as significant predictors: alanine transaminase (ALT), alkaline phosphatase, age at diagnosis, and current age. ALT emerged as the most promising indicator of AFP alteration, potentially preceding AFP level changes and improving HCC detection specificity at a cut-off value of 29 UI/L (AUROC = 0.73). Despite limited data from this rare disease, the ML approach successfully analyzed follow-up biomarkers, identifying ALT as an early predictor of AFP elevation and a potential biomarker for HCC progression.