Background:

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is one of the most effective treatments for various human blood system diseases. Acute graft-versus-host disease (aGVHD) is a main complication and cause of death after allo-HSCT. Steroids are the first choice for treating aGVHD, but their effectiveness rate is only about 50%. The prognosis of steroid-refractory aGVHD (SR-aGVHD) patients is poor. There is currently no predictive tool available for assessing the long-term survival rates of patients who develop Ⅲ-Ⅵ SR-aGVHD following allo-HSCT. Therefore, we integrated patient clinical characteristics and laboratory test results to construct a survival prediction model for Ⅲ-Ⅵ SR-aGVHD after allo-HSCT.

Patients and methods：

A total of 89 patients of III-IV SR-aGVHD were enrolled in Soochow Hopes Hematology Hospital between June, 2019, and October, 2020. All patients diagnosed with aGVHD received initial treatment with 1.0-2.0 mg/kg/day of methylprednisolone. If aGVHD progressed in 3 days or did not improve within 5-7 days, it was diagnosed as SR-aGVHD. At the same time, the dosages of methylprednisolone were gradually reduced and stopped, with a secondary treatment regimen administered. Multivariate analyses were performed on all factors that could affect the short-term effectiveness and long-term prognosis of patients. A survival prediction model for III-IV SR-aGVHD after allo-HSCT was developed by integrating clinical features, laboratory test results, and statistically obtained short-term treatment effects as well as long-term prognosis of impact factors, and was used to predict the survival rate of patients with III-IV SR-aGVHD after allo-HSCT. The model's accuracy was tested by applying it to our own data.

Results：

We retrospectively studied 89 severe SR-aGVHD patients including 55 males（61.8%） and 34 females（38.2%）. The overall median age was 33 years (range, 14-61). The median follow-up time for all patients was 7.5（0.2-59.3）months. The one-year, two-year and three-year overall survival (OS) rates after treatment were 44.9% (95%CI 39.6%~50.2%), 35.8% (95%CI 30.7%~40.9%), and 34.6% (95%CI 29.5%~39.7%), respectively. While the one-year, two-year and three-year disease-free survival (DFS) rates were 43.8% (95%CI 38.5%~49.1%), 35.8% (95%CI 30.7%~40.9%), 33.5% (95%CI 28.5%~38.5%), and the non-relapse mortality (NRM) were 47.0% (95%CI 41.6%~52.6%), 54.3% (95%CI 48.6%~60.0%) and 58.5% (95%CI 52.6%~64.4%). Multivariate analysis results showed that the occurrence of aGVHD within 30 days（OR=1.985，95%CI 1.125~3.504，P=0.018）, liver involvement（OR=1.867，95%CI 1.106~3.152，P=0.019）, and with transplantation-associated thrombotic microangiopathy (TA-TMA)（OR=0.392，95%CI 0.215~0.714，P=0.002）were independent risk factors affecting overall survival. At the same time，infections occurring before the development of GVHD （OR=0.351，95%CI 0.125~0.985，P=0.047）and liver involvement （OR=3.708，95%CI 1.317~10.438，P=0.013）were independent risk factors for efficacy of the four weeks after initiating second-line therapy. Based on the regression coefficients of independent risk factors, we assigned points to each independent risk factor and combined them with the patient's age and sex to develop a long-term survival prediction model. As the risk score increases, the one-year, two-year, and three-year survival rates of patients also decline. We use our own data for validation, and the consistency index of the predictions is 0.68. The developed model exhibits good concordance between the predicted and actual long-term survival rates of III-IV steroid-refractory aGVHD patients.

Conclusion：

This study is the first exploration of a long-term survival prediction model for III-IV steroid-refractory aGVHD after allo-HSCT. The survival prediction model has been through internal validation and can accurately assess the 3-year survival rate of these patients. It can help early recognize III-IV steroid-refractory aGVHD patients after allo-HSCT who have poor long-term prognoses, and provide guidance on aggressive treatment and to find effective treatment solutions to improve the prognosis.

Key words：

Hematopoietic stem cell transplantation; Acute graft-versus-host disease; Steroid-refractory；Survival prediction

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution