Prognostic value of measuring LSC frequency in AML: a systematic review and meta-analysis

Leukemia stem cells (LSC) are thought to be responsible for relapse in patient with acute myeloid leukemia (AML). LSC can be quantified at diagnosis to improve risk stratification, and during follow-up to monitor residual disease. To systematically evaluate the prognostic relevance of measuring LSC at diagnosis and in remission in patients with AML, we conducted a systematic and reconstructed individual patient data meta-analysis. We performed a comprehensive search for studies that reported overall survival (OS) and/or event-free survival (EFS) in relation to LSC measurements at diagnosis or in remission. We reconstructed individual patient data based on Kaplan Meier curves. Fifty-six studies were included, including a total of 44 cohorts (n=7781) for OS and 37 cohorts (n=5032) for EFS at diagnosis, and 19 cohorts (n=2006) for OS and 22 cohorts (n=1342) for EFS in remission. Positive LSC status was significantly associated with inferior OS and EFS at diagnosis (OS: hazard ratio [HR] 1.96 (1.85-2.08); EFS: HR 2.32 (2.15-2.49)) and in remission (OS: HR 2.59 (2.23-3.03); EFS: HR 2.53 (2.18-2.94)). To conclude, we found that measuring LSC has prognostic relevance both at diagnosis and in remission. This supports LSC frequency as a key prognostic factor that warrants implementation into clinical practice.