Algorithm predicts NRM, GVHD after HSCT

A biomarker algorithm can identify patients with a high risk of graft-vs-host disease (GVHD) and non-relapse mortality (NRM) after hematopoietic stem cell transplant (HSCT), according to researchers.

The team found evidence to suggest that 2 proteins—ST2 and REG3α—present in blood drawn a week after HSCT can predict the likelihood of GVHD, including lethal GVHD, and NRM in patients with hematologic disorders.

James L.M. Ferrara, MD, of the Icahn School of Medicine at Mount Sinai in New York, New York, and his colleagues reported these findings in JCI Insight.

The researchers analyzed blood samples collected on day 7 after HSCT from 1287 patients. Of these, 620 samples were designated the training set.

The team measured the concentrations of 4 GVHD biomarkers—ST2, REG3α, TNFR1, and IL-2Rα—in the training set and used them to model 6-month NRM in an attempt to identify the best algorithm that defined 2 distinct risk groups.

The researchers applied the resulting algorithm to the test set of samples (n=309) and the validation set of samples (n=358).

The final algorithm used ST2 and REG3α concentrations to identify patients with a high and low risk of NRM at 6 months. Sixteen percent of patients in the training set belonged to the high-risk group, as did 17% of the test set and 20% of the validation set.

In the training set, the cumulative incidence of NRM at 6 months was 28% in the high-risk group and 7% in the low-risk group (P<0.001). The incidence was 33% and 7%, respectively (P<0.001), in the test set and 26% and 10%, respectively (P<0.001), in the validation set.

The high-risk patients were 3 times more likely to die from GVHD than low-risk patients in the overall cohort. The incidence of lethal GVHD was 19% and 6%, respectively (P<0.001).

GVHD-related mortality in the high-risk and low-risk groups, respectively, was 18% and 5% (P<0.001) in the training set, 24% and 4% (P<0.001) in the test set, and 14% and 5% (P<0.001) in the validation set.

The researchers said their algorithm can also be adapted to define 3 distinct risk groups at GVHD onset—Ann Arbor scores 1, 2, and 3.

The team dubbed their algorithm the “MAGIC algorithm,” after the Mount Sinai Acute GVHD International Consortium (MAGIC).

“The MAGIC algorithm gives doctors a roadmap to save many lives in the future,” Dr Ferrara said. “This simple blood test can determine which bone marrow transplant patients are at high risk for a lethal complication before it occurs. It will allow early intervention and potentially save many lives.”

Doctors at Mount Sinai are now designing clinical trials to determine whether immunotherapy drugs would benefit patients if the MAGIC algorithm determines they are at high risk for severe GVHD.

The researchers believe that if patients receive the drugs once the blood test is administered, which is well before symptoms develop, they would be spared the full force of GVHD, and fewer of them would die. 

“This test will make bone marrow transplant safer and more effective for patients because it will guide adjustment of medications to protect against graft-vs-host disease,” said study author John Levine, MD, of the Icahn School of Medicine at Mount Sinai.

“If successful, the early use of the drugs would become a standard of care for bone marrow transplant patients.”

A biomarker algorithm can identify patients with a high risk of graft-vs-host disease (GVHD) and non-relapse mortality (NRM) after hematopoietic stem cell transplant (HSCT), according to researchers.

The team found evidence to suggest that 2 proteins—ST2 and REG3α—present in blood drawn a week after HSCT can predict the likelihood of GVHD, including lethal GVHD, and NRM in patients with hematologic disorders.

James L.M. Ferrara, MD, of the Icahn School of Medicine at Mount Sinai in New York, New York, and his colleagues reported these findings in JCI Insight.

The researchers analyzed blood samples collected on day 7 after HSCT from 1287 patients. Of these, 620 samples were designated the training set.

The team measured the concentrations of 4 GVHD biomarkers—ST2, REG3α, TNFR1, and IL-2Rα—in the training set and used them to model 6-month NRM in an attempt to identify the best algorithm that defined 2 distinct risk groups.

The researchers applied the resulting algorithm to the test set of samples (n=309) and the validation set of samples (n=358).

The final algorithm used ST2 and REG3α concentrations to identify patients with a high and low risk of NRM at 6 months. Sixteen percent of patients in the training set belonged to the high-risk group, as did 17% of the test set and 20% of the validation set.

In the training set, the cumulative incidence of NRM at 6 months was 28% in the high-risk group and 7% in the low-risk group (P<0.001). The incidence was 33% and 7%, respectively (P<0.001), in the test set and 26% and 10%, respectively (P<0.001), in the validation set.

The high-risk patients were 3 times more likely to die from GVHD than low-risk patients in the overall cohort. The incidence of lethal GVHD was 19% and 6%, respectively (P<0.001).

GVHD-related mortality in the high-risk and low-risk groups, respectively, was 18% and 5% (P<0.001) in the training set, 24% and 4% (P<0.001) in the test set, and 14% and 5% (P<0.001) in the validation set.

The researchers said their algorithm can also be adapted to define 3 distinct risk groups at GVHD onset—Ann Arbor scores 1, 2, and 3.

The team dubbed their algorithm the “MAGIC algorithm,” after the Mount Sinai Acute GVHD International Consortium (MAGIC).

“The MAGIC algorithm gives doctors a roadmap to save many lives in the future,” Dr Ferrara said. “This simple blood test can determine which bone marrow transplant patients are at high risk for a lethal complication before it occurs. It will allow early intervention and potentially save many lives.”

Doctors at Mount Sinai are now designing clinical trials to determine whether immunotherapy drugs would benefit patients if the MAGIC algorithm determines they are at high risk for severe GVHD.

The researchers believe that if patients receive the drugs once the blood test is administered, which is well before symptoms develop, they would be spared the full force of GVHD, and fewer of them would die. 

“This test will make bone marrow transplant safer and more effective for patients because it will guide adjustment of medications to protect against graft-vs-host disease,” said study author John Levine, MD, of the Icahn School of Medicine at Mount Sinai.

“If successful, the early use of the drugs would become a standard of care for bone marrow transplant patients.”

A biomarker algorithm can identify patients with a high risk of graft-vs-host disease (GVHD) and non-relapse mortality (NRM) after hematopoietic stem cell transplant (HSCT), according to researchers.

The team found evidence to suggest that 2 proteins—ST2 and REG3α—present in blood drawn a week after HSCT can predict the likelihood of GVHD, including lethal GVHD, and NRM in patients with hematologic disorders.

James L.M. Ferrara, MD, of the Icahn School of Medicine at Mount Sinai in New York, New York, and his colleagues reported these findings in JCI Insight.

The researchers analyzed blood samples collected on day 7 after HSCT from 1287 patients. Of these, 620 samples were designated the training set.

The team measured the concentrations of 4 GVHD biomarkers—ST2, REG3α, TNFR1, and IL-2Rα—in the training set and used them to model 6-month NRM in an attempt to identify the best algorithm that defined 2 distinct risk groups.

The researchers applied the resulting algorithm to the test set of samples (n=309) and the validation set of samples (n=358).

The final algorithm used ST2 and REG3α concentrations to identify patients with a high and low risk of NRM at 6 months. Sixteen percent of patients in the training set belonged to the high-risk group, as did 17% of the test set and 20% of the validation set.

In the training set, the cumulative incidence of NRM at 6 months was 28% in the high-risk group and 7% in the low-risk group (P<0.001). The incidence was 33% and 7%, respectively (P<0.001), in the test set and 26% and 10%, respectively (P<0.001), in the validation set.

The high-risk patients were 3 times more likely to die from GVHD than low-risk patients in the overall cohort. The incidence of lethal GVHD was 19% and 6%, respectively (P<0.001).

GVHD-related mortality in the high-risk and low-risk groups, respectively, was 18% and 5% (P<0.001) in the training set, 24% and 4% (P<0.001) in the test set, and 14% and 5% (P<0.001) in the validation set.

The researchers said their algorithm can also be adapted to define 3 distinct risk groups at GVHD onset—Ann Arbor scores 1, 2, and 3.

The team dubbed their algorithm the “MAGIC algorithm,” after the Mount Sinai Acute GVHD International Consortium (MAGIC).

“The MAGIC algorithm gives doctors a roadmap to save many lives in the future,” Dr Ferrara said. “This simple blood test can determine which bone marrow transplant patients are at high risk for a lethal complication before it occurs. It will allow early intervention and potentially save many lives.”

Doctors at Mount Sinai are now designing clinical trials to determine whether immunotherapy drugs would benefit patients if the MAGIC algorithm determines they are at high risk for severe GVHD.

The researchers believe that if patients receive the drugs once the blood test is administered, which is well before symptoms develop, they would be spared the full force of GVHD, and fewer of them would die. 

“This test will make bone marrow transplant safer and more effective for patients because it will guide adjustment of medications to protect against graft-vs-host disease,” said study author John Levine, MD, of the Icahn School of Medicine at Mount Sinai.

“If successful, the early use of the drugs would become a standard of care for bone marrow transplant patients.”