Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

A Personalised Prediction Model Based on Clinical and Genomic Data Outperforms Established Prognostic Models in MDS

New prediction model can provide probability of survival and leukaemia transformation at diagnosis of myelodysplastic syndrome and throughout a patient's disease course
02 Sep 2021
Haematologic malignancies;  Personalised medicine

A new model based on clinical and genomic data that can be used as a stand-alone model or in conjunction with other established models to improve the predictability of outcomes in patients with myelodysplastic syndromes (MDS) is built and validated by Dr. Aziz Nazha of the Leukemia Program, Department of Hematology and Medical Oncology and Center for Clinical Artificial Intelligence of the Cleveland Clinic in Cleveland, OH, US and colleagues. This new model is dynamic, predicting survival and leukaemia transformation probabilities at different time points, and can upstage and downstage patients into more appropriate risk categories. It can also be used as a risk stratification tool for enrolment in clinical studies. The authors published their findings on 18 August 2021 in the Journal of Clinical Oncology.

MDS are clonal haematopoietic disorders that lead to bone marrow failure and a risk of progression to acute myeloid leukaemia. The outcomes of patients with MDS are heterogeneous with survival ranging from months to decades. Accurately predicting outcome can help patients manage expectations for their disease trajectory and help physicians identify appropriate therapies.

Established prognostic models rely primarily on clinical variables that are derived from bone marrow pathology and peripheral blood counts and divide patients into a handful of risk categories. The most commonly used models in clinical practice and for eligibility into clinical studies are the International Prognostic Scoring System (IPSS) and the revised IPSS (IPSS-R). The recent addition of molecular data to these scoring systems has enhanced the accuracy of these models. These prognostic models, many of which were developed in untreated patients, may underestimate or overestimate the actual survival of a patient, affecting treatment recommendations and prediction of disease course.

In this study, the investigators took advantage of a machine learning algorithm by taking into account clinical, pathologic, and molecular variables, as well as their interactions with each other, and developed and validated a prediction model that can provide a personalised prognosis that is specific for a given patient.

In total, 1,471 patients with MDS with comprehensively annotated clinical and molecular data were included in a training cohort and analyzed by using machine learning techniques. A random survival algorithm was used to build a prognostic model, which was then validated in external cohorts. The accuracy of the proposed model, compared with other established models, was assessed using a concordance index.

The median age for the training cohort was 71 years. Commonly mutated genes included SF3B1TET2, and ASXL1. The algorithm identified chromosomal karyotype, platelet, haemoglobin levels, bone marrow blast percentage, age, other clinical variables, seven discrete gene mutations, and mutation number as having prognostic impact on overall and leukaemia-free survivals.

The model was validated in an independent external cohort of 465 patients, a cohort of patients with MDS treated in a prospective clinical study, a cohort of patients with paired samples at different time points during the disease course, and a cohort of patients who underwent haematopoietic stem-cell transplantation.

The authors concluded that they built and validated a personalised prediction model that can provide probability of survival and leukaemia transformation at MDS diagnosis and throughout a patient's disease course. The model outperformed other existing models that are used in clinical practice, for eligibility into clinical study, and the timing of transplant. This model can be used as a stand-alone model or in conjunction with the IPSS/IPSS-R scoring systems to improve their accuracy.

Prognostic systems that incorporate advanced analytics of clinical, pathologic, and genomic data have the potential to more accurately and dynamically predict survival in patients receiving various therapies.

Reference

Nazha A, Komrokji R, Meggendorfer M, et al. Personalized Prediction Model to Risk Stratify Patients With Myelodysplastic Syndromes. JCO, Published online 18 August 2021. DOI: 10.1200/JCO.20.02810.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings