Distinct metabolic profiles associated with autism spectrum disorder versus cancer in individuals with germline PTEN mutations

Ying Ni,Tammy Sadler,Lamis Yehia,Thomas W Frazier,Charis Eng

doi:10.1038/s41525-022-00289-x

Abstract

PTEN hamartoma tumor syndrome (PHTS), caused by germline PTEN mutations, has been associated with organ-specific cancers and autism spectrum disorder (ASD) and/or developmental delay (DD). Predicting precise clinical phenotypes in any one PHTS individual remains impossible. We conducted an untargeted metabolomics study on an age- and sex-matched series of PHTS individuals with ASD/DD, cancer, or both phenotypes. Using agnostic metabolomic-analyses from patient-derived lymphoblastoid cells and their spent media, we found 52 differentially abundant individual metabolites, 69 cell/media metabolite ratios, and 327 pair-wise metabotype (shared metabolic phenotype) ratios clearly distinguishing PHTS individuals based on phenotype. Network analysis based on significant metabolites pointed to hubs converging on PTEN-related insulin, MAPK, AMPK, and mTOR signaling cascades. Internal cross-validation of significant metabolites showed optimal overall accuracy in distinguishing PHTS individuals with ASD/DD versus those with cancer. Such metabolomic markers may enable more accurate risk predictions and prevention in individual PHTS patients at highest risk.

Highlights

Hereditary cancer predisposition syndromes and neurodevelopmental disorders account for a large subset of individuals in the medical genetics clinic[1–3]
The first group consisted of 10 PTEN hamartoma tumor syndrome (PHTS) individuals diagnosed with autism spectrum disorder (ASD) and/or developmental delay (DD) without cancer identified to date, with a median age at consent of 24 years
The third group consisted of 10 PHTS individuals who in addition to ASD/DD, had a cancer diagnosis, with a median age at onset of 18 years

Summary

Introduction

Hereditary cancer predisposition syndromes and neurodevelopmental disorders account for a large subset of individuals in the medical genetics clinic[1–3]. While PTEN germline mutations were originally identified in a relatively rare subset of disorders predisposing to breast, thyroid, and other cancers[7], subsequent studies have shown that PTEN germline mutation is amongst the most common causes of autism spectrum disorder (ASD)[8,9]. This PTENrelated phenotypic dichotomy poses a challenge for more timely and precise medical management of individuals with germline PTEN mutations[6,10]. Deciphering this dichotomy may have value in identifying the etiology of a subset of ASD/DD

Methods

Results

Conclusion