Rates of laboratory adverse events by course in paediatric leukaemia ascertained with automated electronic health record extraction: a retrospective cohort study from the Children’s Oncology Group

Miller TP, Getz KD, Li Y, Demissei BG, Adamson PC, Alonzo TA, Burrows E, Cao L, Castellino SM, Daves MH, Fisher BT, Gerbing R, Grundmeier RW, Krause EM, Lee J, Lupo PJ, Rabin KR, Ramos M, Scheurer ME, Wilkes JJ, Winestone LE, Hawkins DS, Gramatges MM, Aplenc R. Rates of laboratory adverse events by course in paediatric leukaemia ascertained with automated electronic health record extraction: a retrospective cohort study from the Children’s Oncology Group. Lancet Haematol. 2022 Sep;9(9):e678-e688. doi: 10.1016/S2352-3026(22)00168-5. Epub 2022 Jul 20. PubMed PMID: 35870472; PubMed Central PMCID: PMC9444944.

Adverse events (AEs) are often misreported on clinical trials, leading to an incomplete understanding of toxicities. We aimed to test automated laboratory AE ascertainment and grading (ExtractEHR) to demonstrate scalability and define AE rates for children with acute myeloid and acute lymphoblastic leukemia (AML, ALL). Patients aged 0–22 years at Children’s Healthcare of Atlanta (CHOA, 01–01-2010–01-11–2018), Children’s Hospital of Philadelphia (CHOP, 01–01-2011–31-12–2014), and Texas Children’s Hospital (TCH, 01–01-2011–31-12`2014) were included. ExtractEHR acquired, cleaned, and graded laboratory data per CTCAEv5 for 22 commonly evaluated AEs with numerically-based CTCAE definitions. Descriptive statistics tabulated AE frequencies. ExtractEHR-ascertained AEs were compared to manually reported Children’s Oncology Group (COG; AAML1031, NCT01371981; AALL0932, NCT02883049) AEs for trial-enrolled patients.