Clinical and Genomic-Based Decision Support System to Define the Optimal Timing of Allogeneic Hematopoietic Stem-Cell Transplantation in Patients with Myelodysplastic Syndromes

Oncology Hematology Life Sciences Medicine Genetics
Myelodysplastic Syndromes Allogeneic Hematopoietic Stem-Cell Transplantation Decision Support System Clinical and Genomic Information Prognostic Scoring System Personalized Therapy

Background

Research Flowchart

Myelodysplastic Syndromes (MDS) are a group of heterogeneous diseases originating from bone marrow hematopoietic stem cells, characterized by reduced blood cell production. Although certain progress in treatment has been made in recent years, allogeneic hematopoietic stem-cell transplantation (HSCT) remains the only potentially curative method for MDS. However, due to the non-negligible morbidity and mortality associated with the transplantation procedure, accurate patient selection is crucial. Traditionally, clinical decisions are made based on the Revised International Prognostic Scoring System (IPSS-R), which includes clinical features and cytogenetic abnormalities. High-risk patients are usually recommended for immediate HSCT, while for low-risk patients, the morbidity and mortality associated with transplantation are unacceptably high. However, there exists significant heterogeneity in the disease course among different patients, especially in low-risk cases, which is not effectively captured by the IPSS-R in all instances. It is considered a current research direction to improve clinical decisions by incorporating molecular information, particularly somatic mutations that can serve as prognostic/predictive markers.

Source of the Paper

This paper was co-authored by Dr. Cristina Astrid Tentori and several other researchers from institutions including the Humanitas Clinical and Research Center, University Medical Center Hamburg-Eppendorf, and The University of Texas MD Anderson Cancer Center. The paper was published on May 9, 2024, in the Journal of Clinical Oncology.

Research Process and Methods

Research Process

The primary objective of this study is to develop and validate a Decision Support System (DSS) based on clinical and genomic information to determine the optimal timing for HSCT in MDS patients. The study utilized a retrospective cohort data of 7,118 patients, divided into training and validation cohorts.

  1. Data Collection and Classification:

    • Collect clinical and genomic information of patients and classify using the Molecular International Prognostic Scoring System (IPSS-M).
    • This information includes hematologic parameters, cytogenetic abnormalities, and mutations in 31 MDS-related genes.

  2. Survival Model Construction:

    • Construct cause-specific survival models in the study population to analyze the risk of progression to acute myeloid leukemia (AML), death without HSCT, relapse risk post-HSCT, and non-relapse mortality risk.

  3. Multi-state Decision Model:

    • Develop a semi-Markov multi-state decision model using microsimulation, simulating the effects of different transplant timing strategies based on patients’ age and IPSS-M/IPSS-R risk categories.
    • The simulation includes five states: pre-HSCT MDS, pre-HSCT AML, post-HSCT status, post-HSCT relapse, and death.

  4. Strategy Comparison:

    • Compare transplant strategies based on IPSS-M and IPSS-R to determine the impact of improved IPSS-M on transplantation decisions.

Results

Under low and low-intermediate risk IPSS-M strategies, patients benefited from delayed transplantation, whereas in intermediate-high, high, and very high-risk categories, immediate transplantation was associated with longer restricted mean survival time (RMST). Modeling decision analysis showed that compared to traditional IPSS-R strategies, IPSS-M-based transplantation strategies changed the transplantation decision in a significant proportion of patients, 15% and 19% respectively, with IPSS-M-based strategies bringing significant life expectancy benefits (P = .001).

  • Statistical Analysis:
    • Survival curves were estimated using the Kaplan-Meier method, and multivariable analysis was conducted using the Cox proportional hazards regression model.
    • The decision model results were validated in independent patient cohorts, proving its reliability.

Conclusion

This study provides clinically relevant evidence for incorporating genomic features into the transplantation decision-making process, demonstrating improved personalized risk assessment and HSCT outcomes for MDS patients. Results indicate that IPSS-M outperforms IPSS-R in defining optimal HSCT timing, significantly improving patient life expectancy by including genomic characteristics, particularly for personalized adjustment of MDS treatment strategies.

Research Highlights

  1. Innovation in Decision Support System (DSS):

    • The study utilized an advanced DSS, combining clinical and genomic information to provide personalized HSCT decisions. The large sample data ensures the scientific validity and reliability of the decisions.

  2. Enhancement of Clinical Significance:

    • Introducing IPSS-M significantly improved decision accuracy, especially in distinguishing high-risk patients needing immediate transplantation from low-risk patients who could delay transplantation.

  3. Wide Validation of Results:

    • The study not only drew conclusions in the training cohort but also confirmed the reproducibility and reliability of its results in the independent validation cohort.

Other Valuable Information

  1. Prototype Web Portal:
    • The research team created a publicly available prototype tool (HSCT Optimal Timing Calculator) to define the optimal transplantation timing based on individual patient information, intended for research purposes.

In conclusion, this study provides a solid scientific foundation for personalized decision-making on HSCT timing for MDS patients through the combination of advanced DSS and IPSS-M, significantly improving long-term survival rates and quality of life for patients.