Back

Population-based Characterization of PTEN Hamartoma Tumor Syndrome

Ni, Y.; Yehia, L.; Idumah, G.; Li, L.; Bautista, C.

2025-12-20 genetic and genomic medicine
10.64898/2025.12.19.25342661 medRxiv
Show abstract

PTEN hamartoma tumor syndrome (PHTS) is a cancer predisposition disorder caused by germline PTEN variants, yet its full clinical spectrum remains poorly defined due to reliance on highly selected cohorts. Accordingly, PHTS is underrecognized and its prevalence underestimated. Leveraging genomic and electronic health record data from 414,830 participants in the All of Us (AoU) Research Program, we identified 55 individuals with pathogenic or likely pathogenic PTEN variants, the majority of whom lacked a prior PHTS diagnosis, underscoring underrecognition in the general population. PHTS affects [~]1/7500 individuals in this US cohort, which is about 26-folds higher than historical estimates for PTEN-related disorder. Compared with carriers of other cancer-related gene variants and noncarriers, PTEN variant carriers exhibited the highest cancer prevalence and significantly younger ages at first cancer diagnosis. Phenotype enrichment revealed expected overgrowth-related features as well as previously unreported associations, including adenotonsillar hypertrophy, sleep apnea, acanthosis nigricans, and extreme obesity, suggesting broader systemic involvement than classically appreciated. Variant spectra were consistent across the population-based and clinically-ascertained PHTS cohorts. These findings demonstrate that PHTS is more prevalent, more heterogeneous, and more often undiagnosed than current clinical practice reflects, emphasizing the value of population-scale genomics for comprehensive characterization and earlier detection of PHTS.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.