| Title: |
Can Machine Learning Models Predict Asparaginase-associated Pancreatitis in Childhood Acute Lymphoblastic Leukemia |
| Authors: |
Nielsen, Rikke L.; Wolthers, Benjamin O.; Helenius, Marianne; Albertsen, Birgitte K.; Clemmensen, Line; Nielsen, Kasper; Kanerva, Jukka; Niinimäki, Riitta; Frandsen, Thomas L.; Attarbaschi, Andishe; Barzilai, Shlomit; Colombini, Antonella; Escherich, Gabriele; Aytan-Aktug, Derya; Liu, Hsi Che; Möricke, Anja; Samarasinghe, Sujith; Van Der Sluis, Inge M.; Stanulla, Martin; Tulstrup, Morten; Yadav, Rachita; Zapotocka, Ester; Schmiegelow, Kjeld; Gupta, Ramneek |
| Contributors: |
HUS Children and Adolescents; Children's Hospital |
| Publisher Information: |
Lippincott Williams and Wilkins |
| Publication Year: |
2022 |
| Collection: |
Helsingfors Universitet: HELDA – Helsingin yliopiston digitaalinen arkisto |
| Subject Terms: |
Cancers; acute lymphoblastic leukemia; artificial intelligence; pediatric hematology/oncology; translational research; treatment toxicity; POLYMORPHISMS; RISK; PRSS1-PRSS2; TOXICITY; pediatric hematology; oncology |
| Description: |
Asparaginase-associated pancreatitis (AAP) frequently affects children treated for acute lymphoblastic leukemia (ALL) causing severe acute and persisting complications. Known risk factors such as asparaginase dosing, older age and single nucleotide polymorphisms (SNPs) have insufficient odds ratios to allow personalized asparaginase therapy. In this study, we explored machine learning strategies for prediction of individual AAP risk. We integrated information on age, sex, and SNPs based on Illumina Omni2.5exome-8 arrays of patients with childhood ALL (N=1564, 244 with AAP aged 1.0 to 17.9 y) from 10 international ALL consortia into machine learning models including regression, random forest, AdaBoost and artificial neural networks. A model with only age and sex had area under the receiver operating characteristic curve (ROC-AUC) of 0.62. Inclusion of 6 pancreatitis candidate gene SNPs or 4 validated pancreatitis SNPs boosted ROC-AUC somewhat (0.67) while 30 SNPs, identified through our AAP genome-wide association study cohort, boosted performance (0.80). Most predictive features included rs10273639 (PRSS1-PRSS2), rs10436957 (CTRC), rs13228878 (PRSS1/PRSS2), rs1505495 (GALNTL6), rs4655107 (EPHB2) and age (1 to 7 y). Second AAP following asparaginase re-exposure was predicted with ROC-AUC: 0.65. The machine learning models assist individual-level risk assessment of AAP for future prevention trials, and may legitimize asparaginase re-exposure when AAP risk is predicted to be low. ; Peer reviewed |
| Document Type: |
article in journal/newspaper |
| File Description: |
application/pdf |
| Language: |
English |
| ISBN: |
978-0-00-772033-0; 0-00-772033-5 |
| Relation: |
Funded by the Kirsten and Freddy Johansen Foundation, the Danish Childhood Cancer Foundation, the Swedish Childhood Cancer Foundation, the Danish Cancer Society, The Nordic Cancer Union, The Otto Christensen Foundation, University Hospital Rigshospitalet, the European Union’s Interregional Öresund–Kattegat–Skagerrak interregional Childhood Oncology Precision Medicine (iCOPE) grant and The Novo Nordisk Foundation. R.L.N. was supported by a grant from the Sino-Danish Center for Education and Research and a grant from the Poul V Andersen Foundation.; https://hdl.handle.net/10138/342783; 85117820186; 000772033500014 |
| Availability: |
https://hdl.handle.net/10138/342783 |
| Rights: |
cc_by_nc_nd ; info:eu-repo/semantics/openAccess ; openAccess |
| Accession Number: |
edsbas.E76A7C7F |
| Database: |
BASE |