Skip to Main Content


Skip Nav Destination

Simple Risk Model Predicts Grade 4 Bleeding in Patients With AML Who Are Receiving Intensive Induction Chemotherapy

April 27, 2022

May 2022

A new risk model was found to predict grade 4 bleeding in patients with acute myeloid leukemia (AML) who are undergoing intensive induction chemotherapy, according to a study published in Blood Advances.

Bleeding can be a significant concern that causes early morbidity and mortality in this patient group, so finding a way to identify those who are at the highest risk could aid in the development of strategies to improve overall patient outcomes. With this goal in mind, researchers led by Jurjen Versluis, MD, PhD, of Dana-Farber Cancer Institute, used retrospective patient data to create a simple predictive tool to identify patients with AML who are at high risk for grade 4 bleeding events.

The development cohort included data from 341 adult patients who had been newly diagnosed with non-acute promyelocytic leukemia (APL) AML at Dana-Farber/Brigham and Women’s Cancer Center between August 2014 and March 2020. To be included in the study, patients must have received intensive induction chemotherapy, which was most often a regimen of anthracycline and cytarabine for three and seven days, and could not have received prior AML treatment outside of hydroxyurea or all-trans retinoic acid. Patients were excluded from the study if they were previously transplanted with an allogeneic donor for myelodysplastic syndromes or myeloproliferative neoplasms.

“We identified pre-treatment risk factors consistent with DIC-like [disseminated intravascular coagulation] coagulopathy, which predicted grade 4 bleeding in patients with AML receiving intensive induction chemotherapy,” Dr. Versluis told ASH Clinical News. “These risk factors were used in a novel and simple tool, allowing us to identify patients, prior to the start of treatment, who were at high risk for severe and life-threatening bleeding events during AML induction chemotherapy.”

A validation cohort with similar exclusion criteria included 143 adult patients with non-APL AML who were newly diagnosed between 2008 and 2020 at the Roswell Park Comprehensive Cancer Center in Buffalo, New York.

Researchers performed a retrospective analysis of bleeding events that occurred between 14 days before and 60 days after the start of induction chemotherapy. They identified grade 4 bleeding events using the World Health Organization bleeding assessment scale, which defines grade 4 events as including one of the following criteria: bleeding with severe hemodynamic instability requiring red blood cell transfusion over routine transfusion needs, fatal bleeding, retinal bleeding with visual impairment, or central nervous system bleeding with or without neurologic dysfunction.

The investigators did not find any significant difference in the primary outcome of grade 4 bleeding between the two cohorts, noting the outcome occurred in 5.9% of those in the development cohort and 9.8% of the validation cohort (p=0.123).

In the development cohort, only two factors were independently associated with grade 4 bleeding. These factors included a baseline platelet count of ≤40×109/L and an increased international normalized ratio (INR) of prothrombin time.

Points were assigned to each of the risk factors (platelet count ≤40×109/L = 1 point, INR 1.3-1.5 = 1 point, INR >1.5 = 2 points), and a risk model that stratified patients as either being low-risk (0-1 point) or high-risk (2-3 points) for grade 4 bleeding was developed. They found that those patients deemed high-risk had higher rates of grade 4 bleeding at day +60 compared to those in the low-risk group (31±7% vs. 2±1%; p<0.001).

The same trend was observed among the validation cohort, in which patients at high risk of grade 4 bleeding events at day +60 had a cumulative incidence of 25±9% compared to 7±2% in the low-risk group (p=0.008).

High risk of bleeding was associated in the development and validation cohort with DIC and proliferative disease.

A limitation of the study was its retrospective design, which required researchers to rely on data from individual patient charts that could have excluded relevant data, such as whether a patient had major bleeding before. Researchers also did not have data on AML with monocytic differentiation and noted that the validation cohort was made up of a non-consecutive group of patients with genetic analysis available, leading to possible selection bias.

Any conflicts of interest declared by the authors can be found in the original article.


Versluis J, Pandey M, Flamand F, et al. Prediction of life-threatening and disabling bleeding in patients with AML receiving intensive induction chemotherapy [published online ahead of print, 2022 Jan 28]. Blood Adv. doi: 10.1182/blood​advances.2021006166.


Connect with us:

May 2022


Close Modal

or Create an Account

Close Modal
Close Modal