GVHD Risk Assessments Can Predict Mortality, Drug Response

By Cecilia Brown - Last Updated: September 30, 2022

Scores on graft-versus-host disease (GVHD) risk assessments can predict a patient’s risk of mortality and their likelihood of responding to steroid treatment.

Maria Silvina Odstrcil Bobillo, MD, of Hospital Italiano de Buenos Aires in Argentina, and colleagues conducted the research and presented their findings at the Tenth Annual Meeting of the Society of Hematologic Oncology.

The retrospective 2-center cohort study included 261 patients who received allogeneic hematopoietic stem-cell transplantation (HSCT) between 2017 and 2020. The median patient age was 44 years (range, 16-76 years), with a median donor age of 35 years (range, 12-70 years). More than one-third of allogeneic HSCT donors were unrelated, while 31% were histoidentical related donors, and 33% were haploidentical donors. The most common reason for transplant was acute myeloid leukemia, which occurred in 30% of patients. More than half (54.8%) of the 261 patients studied developed acute GVHD.

The cohort’s 2-year overall survival rate was 63%. The median time to starting steroid therapy was 52 days, with an overall response rate (ORR) of 77%, a complete response rate of 69%, and a partial response rate of 8.3%.

The initial GVHD presentation involved skin in more than half the patients (53%) and involved multiple organs in 31%. Less common was GVHD involving the upper gastrointestinal tract (5%) or the liver (1%).

Researchers used the Minnesota GVHD Score to classify 29% of patients as high risk and the remaining 71% of patients as standard risk. Standard-risk patients had a significantly higher ORR (90%) than high-risk patients (49%; P<.001). High-risk patients were also significantly less likely to respond to first-line therapy (odds ratio, 0.03; 95% CI, 0.008-0.114; P<.0001) and had a higher 6-month transplant-related morality rate (27.5%; 95% CI, 14-43) than standard-risk patients (10.8%; 95% CI, 5.5-18.6; P<.03).

The researchers used the Magic Score to classify 46.9% of cases as grade 1 GVHD, while 19.7% were classified as grade 2, 15.5% as grade 3, and 18.3% as grade 4. Patients with GVHD grades 1, 2, and 3 had ORRs of 96%, 100%, and 87%, respectively. However, the ORR in patients with GVHD grade 4 was significantly lower at 27% (P=.0001). Similarly, the instance of transplant-related mortality was higher in patients with GVHD grade 4 (32%) than in those with GVHD grade 1, 2, or 3 (9%, 14%, and 14%, respectively; P=.029)

“Both scores give complementary information,” the researchers concluded. “[Patients with] Minnesota [high-risk] and Magic grade 4 GVHD are less likely to respond to steroids and have higher mortality risks than lower grades.”

Odstrcil Bobillo M, Garcia M, Ferini G, et al. Acute graft-versus-host disease: risk scoring and response to treatment, experience from two centers. Poster CT-347. Presented at the Tenth Annual Meeting of the Society of Hematologic Oncology; September 28-October 1, 2022; Houston, TX.

Post Tags:SOHO 2022: Focus on GVHD