According to a story from Science and Technology Research News, a group of researchers from the Cleveland Clinic led by Dr. Aziz Nazha will soon present a new myelodysplastic syndromes model that predicts outcomes for patients more accurately and in a more personalized fashion. The model has demonstrable advantages over currently available tools.
About Myelodysplastic Syndromes
Myelodysplastic syndromes are a type of blood cancer in which developing blood cells remain immature and fail to transform into usable blood cells. Risk factors for this disease include exposure to radiation, chemotherapy, benzene, xylene, and Agent Orange. Family history is also a risk, as are certain genetic disorders such as Down syndrome. In an individual case, it is rare for the direct cause to be identified. Myelodysplastic syndromes rarely present with symptoms initially, but it can eventually present with anemia, neutropenia, thrombocytopenia, cell abnormalities, chromosome abnormalities, enlarged spleen and/or liver, easy bleeding and bruising, and infections. The disease also has the potential to evolve into acute myeloid leukemia. Treatment may include bone marrow transplant, stem cell transplant, blood products, and certain chemotherapy agents. Outcomes in this disease ranges widely and can depend on a number of factors. To learn more about myelodysplastic syndromes, click here.
An Unpredictable Cancer
The various factors that impact prognosis for patients with myelodysplastic syndromes makes it exceptionally challenging to predict outcomes. These outcomes also vary widely; some patients do not survive for more than several months, but there are other patients that can live for decades with the disease. Even though there are multiple scoring systems for myelodysplastic syndromes, they do a poor job of predicting survival, which means that many patients may be over or under-treated.
The new model utilizes an algorithm which analyzes each individual patient’s health data. This data can simply be entered in a web application which issues predictions about overall survival and the likelihood of the disease transforming into acute myeloid leukemia. The model was able to predict survival in a given time period with 74 percent accuracy compared to just 67 percent accuracy with earlier models. It could also predict leukemia with 80 percent accuracy compared to only 73 percent accuracy with the old model.
