Peripheral blood DNA methylation predicts the early onset of primary tumor in TP53 mutation carriers

Vallijah Subasri ^1,2,3, Benjamin Brew ¹, Brianne Laverty ^1,2

¹Genetics and Genome Biology Program, The Hospital for Sick Children, Toronto, ON, Canada.
²Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada.
³Vector Institute, Toronto, ON, Canada.
⁴Department of Computer Science, University of Toronto, Toronto, ON, Canada.
⁵Institute of Medical Science, University of Toronto, Toronto, ON, Canada.
⁶Children's Cancer Centre, Royal Children's Hospital, Melbourne, VIC, Australia.
⁷Murdoch Children's Research Institute, Melbourne, VIC, Australia.
⁸Department of Pediatrics, University of Melbourne, Melbourne, VIC, Australia.
⁹Michael Rice Centre for Haematology and Oncology, Women's and Children's Hospital, Adelaide, SA, Australia.
¹⁰South Australia Health and Medical Research Institute, Adelaide, SA, Australia.
¹¹South Australia Immunogenomics Cancer Institute, University of Adelaide, Adelaide, SA, Australia.
¹²Department of Pediatrics, Children's Hospital at London Health Sciences Centre, London, ON, Canada.
¹³Department of Pediatrics, Western University, London, ON, Canada.
¹⁴Department of Pediatrics, McMaster University, Hamilton, ON, Canada.
¹⁵Faculty of Medicine, University of Toronto, Toronto, ON, Canada.
¹⁶Department of Medical Oncology, Princess Margaret Hospital, Toronto, ON, Canada.
¹⁷Department of Medical Oncology, Mount Sinai Hospital, Toronto, ON, Canada.
¹⁸Pediatrics and Radiation Oncology, The Ohio State University College of Medicine, Columbus, OH, USA.
¹⁹Department of Oncology, St Jude Children's Research Hospital, Memphis, TN, USA.
²⁰Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, USA.
²¹Division of Hematology/Oncology, The Hospital for Sick Children, Toronto, ON, Canada.
²²Department of Pediatrics, University of Toronto, Toronto, ON, Canada.
²³Clinical Genetics Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.
²⁴Department of Pediatrics, University of Utah, Salt Lake City, UT, USA.
²⁵Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada.
²⁶Ontario Institute for Cancer Research, Toronto, OH, Canada.
²⁷Genetics and Genome Biology Program, The Hospital for Sick Children, Toronto, ON, Canada. david.malkin@sickkids.ca.
²⁸Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada. david.malkin@sickkids.ca.
²⁹Institute of Medical Science, University of Toronto, Toronto, ON, Canada. david.malkin@sickkids.ca.
³⁰Division of Hematology/Oncology, The Hospital for Sick Children, Toronto, ON, Canada. david.malkin@sickkids.ca.
³¹Department of Pediatrics, University of Toronto, Toronto, ON, Canada. david.malkin@sickkids.ca.
³²Genetics and Genome Biology Program, The Hospital for Sick Children, Toronto, ON, Canada. anna.goldenberg@sickkids.ca.
³³Vector Institute, Toronto, ON, Canada. anna.goldenberg@sickkids.ca.
³⁴Department of Computer Science, University of Toronto, Toronto, ON, Canada. anna.goldenberg@sickkids.ca.
³⁵CIFAR, Toronto, ON, Canada. anna.goldenberg@sickkids.ca.

⁰Genetics and Genome Biology Program, The Hospital for Sick Children, Toronto, ON, Canada.

Nature Communications +

|

August 26, 2025

View abstract on PubMed

Summary

Li-Fraumeni syndrome (LFS) patients with TP53 pathogenic variants (PV) can now benefit from personalized cancer risk prediction. A new machine learning model identifies children at high risk for early-onset tumors, improving surveillance and patient outcomes.

Area Of Science

Genetics and Genomics
Oncology
Bioinformatics

Background

Li-Fraumeni syndrome (LFS) is a hereditary cancer predisposition syndrome caused by germline TP53 pathogenic variants (PV).
Current surveillance protocols, like the Toronto Protocol, offer comprehensive screening but lack personalization for individual cancer risks.
There is a need for tailored screening strategies to optimize early tumor detection in TP53 PV carriers.

Purpose Of The Study

To develop and validate a predictive model for early-onset primary tumors (age < 6) in TP53 PV carriers.
To personalize cancer surveillance by stratifying individual risk based on peripheral blood methylation data.
To improve clinical management and patient outcomes through accurate risk assessment.

Main Methods

A support vector machine model was developed using peripheral blood methylation data from 237 TP53 PV carriers.
The model was validated on a cohort of 64 patients and externally tested on 79 patients.
Performance metrics including AUROC, F1-score, and Negative Predictive Value (NPV) were evaluated.

Main Results

The model achieved high performance with an AUROC of 0.928, F1-score of 0.692, and NPV of 0.984.
Overall accuracy was 91%, correctly identifying 90% of patients with early-onset cancer and 87% of cancer-free individuals in the external test set.
The model demonstrated significant potential for risk stratification of early-onset malignancies.

Conclusions

A novel machine learning tool effectively predicts early-onset tumors in Li-Fraumeni syndrome patients.
This personalized risk stratification can optimize clinical surveillance protocols.
The developed model holds promise for improving patient outcomes by enabling timely and targeted interventions.

Related Concept Videos

Epigenetic Regulation

Epigenetic mechanisms play an essential role in healthy development. Conversely, precisely regulated epigenetic mechanisms are disrupted in diseases like cancer.

In most mammals, females have two X chromosomes (XX) while males have an X and a Y chromosome (XY). The X chromosome contains significantly more genes than the Y chromosome. Therefore, to prevent an excess of X chromosome-linked gene expression in females, one of the two X chromosomes is randomly silenced during early development....

Abnormal Proliferation

Under normal conditions, most adult cells remain in a non-proliferative state unless stimulated by internal or external factors to replace lost cells. Abnormal cell proliferation is a condition in which the cell's growth exceeds and is uncoordinated with normal cells. In such situations, cell division persists in the same excessive manner even after cessation of the stimuli, leading to persistent tumors. The tumor arises from the damaged cells that replicate to pass the damage to the...