191eP Machine learning assisted rapid, point-of care sarcoma classification based on demographic, clinical and molecular landscape

ESMO rare cancers. 2026 Open access

Read full text
Authors: A. Zick, Myriam Maoz, B. Ben-Kuzari, A. Avisrur, D. Sheril-Rofe, Daniel Burg, A. Hilu, I. Lavon, Omri Or, Noam Olshinka, J.A. Demma, M. Brakha
+8 more Adi Nitzan–Luques, Assaf Grunwald, J. Moss, Tamar Peretz‐Yablonski, Gilad Wolf Vainer, A. Darwish, B.P. Berman, Amir Eden
DOI: 10.1016/j.esmorc.2026.100397
Cited by: 0

Abstract

Pathological diagnosis of sarcoma based on histopathological features and molecular workup correctly classifies 80–85% of cases. Using low coverage whole genome DNA Nanopore sequencing, we identify copy number alterations and methylation patterns. Using methylation-based classification, NanoDx, we correctly identified 78% of 28 sarcoma cases. In this work, we expand our cohort and develop machine-learning models using clinical and molecular data.

Fields of Study & Subjects

Sarcoma Diagnosis and Treatment Cancer Genomics and Diagnostics Genomics and Rare Diseases

Publication details

Journal / source
ESMO rare cancers.
Volume
5
Pages
100397–100397
Publication date
1 March 2026
Language
EN