A machine learning model of response to hypomethylating agents in myelodysplastic syndromes
Nathan Radakovich,
David A. Sallman,
Rena Buckstein,
Andrew Brunner,
Amy Dezern,
Sudipto Mukerjee,
Rami Komrokji,
Najla Al-Ali,
Jacob Shreve,
Yazan Rouphail,
Anne Parmentier,
Alexandre Mamedov,
Mohammed Siddiqui,
Yihong Guan,
Teodora Kuzmanovic,
Metis Hasipek,
Babal Jha,
Jaroslaw P. Maciejewski,
Mikkael A. Sekeres,
Aziz Nazha
Affiliations
Nathan Radakovich
Leukemia Program, Department of Hematology and Medical Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, USA; Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH, USA
David A. Sallman
Department of Malignant Hematology, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, USA
Rena Buckstein
Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, ON, Canada
Andrew Brunner
Massachusetts General Hospital, Boston, MA, USA
Amy Dezern
Johns Hopkins Sidney Kimmel Comprehensive Cancer Center, Baltimore, MD, USA
Sudipto Mukerjee
Department of Hematology and Medical Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, USA
Rami Komrokji
Department of Malignant Hematology, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, USA
Najla Al-Ali
Department of Malignant Hematology, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, USA
Jacob Shreve
Department of Internal Medicine, Cleveland Clinic, Cleveland, OH, USA
Yazan Rouphail
Ohio State University, Columbus, OH, USA
Anne Parmentier
Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, ON, Canada
Alexandre Mamedov
Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, ON, Canada
Mohammed Siddiqui
Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, ON, Canada
Yihong Guan
Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, USA
Teodora Kuzmanovic
Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, USA
Metis Hasipek
Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, USA
Babal Jha
Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, USA
Jaroslaw P. Maciejewski
Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, USA
Mikkael A. Sekeres
Division of Hematology, Sylvester Comprehensive Cancer Center, University of Miami, Miami, FL, USA
Aziz Nazha
Department of Oncology, Sidney Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA, USA; Corresponding author
Summary: Hypomethylating agents (HMA) prolong survival and improve cytopenias in individuals with higher-risk myelodysplastic syndrome (MDS). Only 30-40% of patients, however, respond to HMAs, and responses may not occur for more than 6 months after HMA initiation. We developed a model to more rapidly assess HMA response by analyzing early changes in patients’ blood counts. Three institutions’ data were used to develop a model that assessed patients’ response to therapy 90 days after the initiation using serial blood counts. The model was developed with a training cohort of 424 patients from 2 institutions and validated on an independent cohort of 90 patients. The final model achieved an area under the receiver operating characteristic curve (AUROC) of 0.79 in the train/test group and 0.84 in the validation group. The model provides cohort-wide and individual-level explanations for model predictions, and model certainty can be interrogated to gauge the reliability of a given prediction.
