DARWIN EU® - Acute myeloid leukaemia: incidence, patient characteristics, treatments, and survival in the period 2015–2024

Study type

• Study topic:
  Disease /health condition
• Study topic, other:
  Acute myeloid leukaemia
• Study type:
  Non-interventional study
• Scope of the study:
  Disease epidemiology

  Feasibility analysis
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Medical condition to be studied:
  Acute myeloid leukaemia

Population studied

• Short description of the study population:
  For the estimation of incidence, all individuals present in the data source will be included in the denominator according to the requirements described in section 8.4 of the protocol. Only individuals with newly diagnosed AML will be included in the study for objectives 2–4.; Objective 1 inclusion criteria; • Observation in the period 01/01/2015 to 31/12/2024 ; • Minimum 365 days of available history before contributing to person-time at risk (except for children <1 year); Objective 2–4 inclusion criteria; • First diagnosis of AML in the period of 01/01/2015 to 31/12/2024 ; • Minimum 365 days of available history before diagnosis date (except for children <1 year); • No prior history of AML before the study period;
• Age groups:
  • In utero
  • Paediatric Population (< 18 years)

    • Neonate

      • Preterm newborn infants (0 – 27 days)
      • Term newborn infants (0 – 27 days)
    • Infants and toddlers (28 days – 23 months)
    • Children (2 to < 12 years)
    • Adolescents (12 to < 18 years)
  • Adult and elderly population (≥18 years)

    • Adults (18 to < 65 years)

      • Adults (18 to < 46 years)
      • Adults (46 to < 65 years)
    • Elderly (≥ 65 years)

      • Adults (65 to < 75 years)
      • Adults (75 to < 85 years)
      • Adults (85 years and over)

Study design details

• Study design:
  Cohort studies will be conducted using routinely collected health data
• Main study objective:
  1. To estimate the annual incidence of AML in children (<18 years old) and adults (≥18).; 2. To characterise AML patients in terms of demographics, performance status, conditions, disease characteristics, measurements, and procedures.; 3. To describe treatment patterns in AML patients.; 4. To estimate overall survival 1, 3, and 5 years for patients with newly diagnosed AML.;
• Outcomes:
  Objective 1 ; The outcome for this objective was the first diagnosis of AML. ; Objective 4: ; The outcome for this objective was death from any cause (overall survival). ; The concept sets used for identifying outcomes are described in Annex IV. These sets were created during the study execution following the DARWIN EU® phenotyping standard processes, which involved the review of code lists by clinical experts and the review of phenotypes after their execution in the participating data sources.
• Summary results:
  CONCLUSION ; Haemato-oncology is a complex topic to study using observational data due to the complexity of treatment and the requirements for highly specialised data, such as genetic data, procedures, and laboratory information. Most data sources are not equipped for this purpose. ; In the 10 data sources, we observed substantial variability in the completeness of diagnostic, molecular, and treatment-pathway data, highlighting challenges for AML research using observational data. More importantly, a substantial proportion of individuals with AML lacked records of any treatment, or it was impossible to understand the sequence of treatment due to data and analytical limitations. This issue was even more pronounced in paediatric populations due to small sample sizes across different data sources. ; Survival estimates were in line with the literature across most data sources but were relatively high in data sources with selected populations (APL, fit patients, eligible for clinical trials) and in claims data sources. ; The results of this study should not be used to assess the quality of treatment, professional competence, or prognosis in AML in general, because we were describing results from different data sources with varying completeness that represent different AML populations. This study can inform the assessment of the fitness-for-use of available data for future studies on AML and the identification of data sources likely to contribute to such studies, depending on the question of interest. This study also highlights the current need for improvements in data quality, completeness, and analytical tools for treatment patterns in AML, as well as the steps needed to address more advanced research questions, such as those related to relapse/refractory AML.