Versluis lab

The central focus of our lab is on improving outcomes for patients with hematological malignancies, such as myelodysplastic syndrome (MDS), acute myeloid leukemia (AML), and multiple myeloma, by applying personalized treatment approaches using advanced statistical methodologies.

Prediction of outcomes

Predicting outcomes in AML patients remains challenging due to the heterogeneity in aging populations and diverse disease biology. Our previous work has demonstrated the need for continuous reassessment of mortality risk after allogeneic stem cell transplantation, particularly in older patients, as novel treatment strategies evolve. With increasingly complex data structures and a growing number of variables, unbiased approaches are essential for evaluating risk factors. To address this, our lab employs advanced methodologies, including machine learning algorithms, to enhance risk prediction and improve patient outcomes.

Clinical trial efficiency

Randomized clinical trials (RCTs) remain the gold standard for evaluating the efficacy and safety of new treatments. However, the increase of personalized medicine, with targeted therapies for tightly defined patient populations, has made the design and execution of RCTs increasingly complex. Bayesian inference has emerged as an approach to address some limitations of traditional RCTs, such as reliance on implicit prior assumptions, the need for long-term follow-up to accumulate sufficient events, and conservative stopping rules. By incorporating real-world data (RWD) from patients with similar disease characteristics receiving comparable control treatments, Bayesian methods provide a flexible framework that can enhance trial efficiency and adaptability. Our lab applies novel statistical methods to HOVON clinical trials to improve trial design and advance personalized treatment strategies.

Dynamic assessment of events

Detailed, longitudinal clinical data enables a dynamic evaluation of treatment effects over time. Time-dependent analyses, multi-state models and joint modeling approaches account for covariates that evolve throughout the course of treatment, which is particularly relevant to hematological malignancies (e.g., longitudinal monitoring of biomarker after stem cell transplantation, measurable residual disease in MM and AML). Our lab is leveraging these advanced statistical techniques to better understand how intermediate events and biomarker dynamics influence long-term outcomes, with the goal of improving predictive accuracy and informing personalized treatment strategies.

• Valk lab
• Raaijmakers lab

External collaboration

• HOVON Leukemia working group
• HOVON SCT working group
• Harmony Alliance
• EBMT - Acute leukemia working party