Deep Neural Network Classifier Identifies Pediatric Brain Tumors From Liquid Biopsies
Researchers developed the methylation-based predictive algorithm for central nervous system tumors (M-PACT), a deep neural network classifier, to identify and classify brain tumors in pediatric patients from subnanogram-input cell-free DNA of methylomes in liquid biopsies.
To develop and validate a deep neural network classifier (M-PACT) for identifying and classifying pediatric brain tumors using cell-free DNA from liquid biopsies, which is crucial for improving diagnostic accuracy in this vulnerable population.
Key Findings:
M-PACT achieved 92% accuracy in the benchmarking cohort and 88% in the validation cohort for classifying embryonal CNS tumors.
The classifier can distinguish between primary and secondary malignancies during treatment and follow-up.
M-PACT demonstrated utility in nonembryonal CNS tumors, such as gliomas, and in nonmalignant cerebrospinal fluid.
Interpretation:
M-PACT represents a significant advancement in the use of liquid biopsies for pediatric brain tumor diagnosis, with potential applications beyond pediatric oncology, though challenges remain in broader implementation.
Limitations:
The study primarily focused on pediatric brain tumors, limiting immediate applicability to other cancer types.
Further informatics development is needed to classify a broader range of cancer types, which is essential for expanding M-PACT's utility.
Conclusion:
M-PACT is a powerful tool for pediatric neuro-oncology that could enhance diagnostic accuracy and inform treatment strategies.
ASCO AI in Oncology
is published by Conexiant under a license arrangement with the American Society of Clinical Oncology, Inc. (ASCO®). The ideas and opinions expressed in
ASCO AI in Oncology do not necessarily reflect those of Conexiant or ASCO. For more information, see Policies.