ISSN 1866-8836
Клеточная терапия и трансплантация

PC-04. Risk factors for the development of graft dysfunction after allogeneic hematopoietic stem cell transplantation

Ulyana V. Maslikova1, Ekaterina D. Mikhaltsova1, Alexey A. Glazkov2, Olga S. Karavaeva1, Luiza A. Karaseva1, Feruza A. Omarova1, Elmira I. Kolgaeva1, Inara S. Saidullayeva1, Darya A. Mironova1, Zoya V. Konova1, Henrik Grigoryan2, Vera A. Vasilieva1, Larisa A. Kuzmina1, Mikhail Y. Drokov1, Ekaterina G. Khamaganova1, Elena N. Parovichnikova1

1 National Medical Research Center for Hematology, Moscow, Russia
2 M. F. Vladimirsky Moscow Regional Memorial Medical Center, Moscow, Russia

Contact: Dr. Ulyana V. Maslikova, phone: +7 (937) 208-86-68, e-mail:

doi 10.18620/ctt-1866-8836-2023-12-3-1-176


Hematopoietic stem cell engraftment is the key moment in the success of allogeneic hematopoietic stem cell transplantation (allo-HSCT). Despite the increase in transplantation activity, graft failure (GF) remains one of the key problems reducing overall survival of patients and significantly impairing the quality of life. The search for new modeled risk factors to reduce the incidence of GF continues. Natural killer (NK) cells are the first lymphocyte subpopulation to recover after allo-HSCT. After allo-HSCT, natural killer cells play an important role in controlling the growth and dissemination of cancer cells. They can detect and destroy remaining tumor cells, which reduces the likelihood of disease recurrence. However, some NK cells remain to be of host origin early after HSCT, thus potentially interacting with graft cells, mediating the development of graft failure. Alloreactivity of NK cells may occur when there is a mismatch between the immunoglobulin-like receptor (KIR) of recipient NK cells and appropriate ligand of the graft cells. The aim of this work was to evaluate the influence of KIR receptors on the development of graft failure and to build a prognostic model based on risk factors.

Materials and methods

We analyzed 69 first allo-HSCT, which were performed on patients in remission of acute leukemia (AML N=43, ALL N=26), with reduced intensity conditioning. Risk factor analysis was performed with logistic regression and Cox regression models.


Significant factors in the prediction of graft failure in our patient cohort were recipient age (HR [95% CI] = 1.04 [1.01; 1.07]), absence of KIR2DS4f receptor in the recipient (HR [95% CI] = 5. 5 [1.32; 22.9]), use of bone marrow (HR [95% CI] = 3.79 [1.64; 8.79]) as a graft source, absence of a mismatch towards GvH by HLA at locus A (HR [95% CI] = 7.38 [1.01; 54]). Fig. 1 shows the results of the ROC analysis for the main significant indexes, with the best performance demonstrated by the regression model including all identified factors.

Using Cox regression model construction, it was evaluated how the obtained model can be used to predict the complication of bone marrow and HSC transplantation (Fig. 2).


The presence of KIR2DS4f in the recipient significantly increased the risks of developing GF. The graft source, age of the recipient and the presence of HLA mismatch were also significant in the development of GF. The developed model may have a prognostic value, therefore it is planned to continue this work.


Graft failure, allogeneic stem cell transplantation, allo-HSCT.

Supplement 12-3

Download PDF version

doi 10.18620/ctt-1866-8836-2023-12-3-1-176

Back to the list