See the source image

Artificial intelligence and machine learning could improve the ability to predict venous thromboembolism (VTE) in cancer patients over the current gold-standard approaches, according to a research grant award presentation at the 2022 annual meeting of the Hematology/Oncology Pharmacy Association (HOPA).

VTE, including both deep vein thrombosis and pulmonary embolism, represents a major source of morbidity and mortality for cancer patients, with an incidence ranging from 1% to 8% and a one-year mortality rate of more than 64%.

“VTE is the third leading cause of death in cancer patients behind relapse and infection, and drastically changes the clinical trajectory of our patients,” Benjamin Andrick, PharmD, BCOP, a hematology/oncology clinical pharmacist at Geisinger Health System, in Danville, Pa., told HOPA attendees. “But identifying the risk factors for VTE is no small feat.”

In addition to the tumor’s direct impact, including release of procoagulants and prothrombotic manipulation of the tumor microenvironment, many other factors play a role in VTE formation, such as type of cancer, type of chemotherapy, surgery, comorbidities, infection and immobility, as well as other unknown heterogeneous risk factors, he noted.