Factors Affecting Platelet Transfusion Response Following Haploidentical Stem Cell Transplant

A mainstay of therapy for certain hematologic malignancies and non-malignant disorders is allogeneic stem cell transplant. One known complication is post-transplant thrombocytopenia which results in a significant bleeding risk. Although prophylactic platelet transfusions reduce the severity of thrombocytopenia, refractoriness to transfusions often develops. Several studies have examined clinical and laboratory factors that are associated with poor platelet response in the general hematologic malignancy and hematopoietic stem cell transplant population, which include lymphocytotoxic antibodies, bleeding, medications and infection. Only a few detailed studies have examined factors which specifically affect platelet transfusion response in patients who have undergone haploidentical stem cell transplant (HI SCT).

This was a single center retrospective analysis of 721 platelet transfusions from 110 HI SCT patients to determine risk factors associated with platelet refractoriness. During the Jefferson 2-step HI SCT, the recipient receives myeloablative or reduced intensity conditioning followed by donor lymphocyte product infusion (DLI). Cyclophosphamide is given 2-3 days after DLI for tolerization, followed in 2 days by CD34+ stem cell product infusion. Patient-specific, transplant-specific and transfusion-specific data were collected and analyzed within 28 days following the 2-step HI SCT. Patient- and transplant-specific data included patient age, gender, body surface area, indication for transplant, conditioning regimen, date of transplant, pre-transplant CMV antibody status, pre-transplant anti-HLA antibodies, number of prior pregnancies, history of splenectomy or splenomegaly, and patient co-morbidities according to the Hematopoietic Cell Transplantation-Specific Co-morbidity Index. Transfusion-specific data included major ABO or Rh mismatch between bone marrow transplant (BMT) recipient and platelet product, major ABO or Rh mismatch between BMT donor and platelet product, age of platelet product at the time of transfusion, pre- and post-transfusion platelet count, and time from BMT to transfusion.

Corrected count increments (CCI) were calculated according to standard formula, CCI=(post-transfusion - pre-transfusion platelet count) x body surface area / number of platelet transfused. The date and time of each platelet transfusion were recorded along with pre- and post-transfusion counts. CCI measures were modeled as a function of time from BMT using a linear mixed effects model allowing for the patient-specific trajectories by incorporating random effects in the intercept and slope for time from BMT.

Our analysis identified significant associations with recipient gender, splenomegaly, coronary artery disease/congestive heart failure (CAD/CHF) and time from BMT to transfusion. CCI was on average lower by 23% in male compared to female recipients (95%CI: 13%, 33%; p<0.001), and by 40% in patients with splenomegaly (95%CI: 29%, 49%; p<0.001). When assessing CCI trends over time, transfusions further out from BMT were shown to have improved CCI versus those closer to BMT with significant improvement in patients without CAD/CHF (average 2.8% CCI improvement per day in non-CAD/CHF patients versus 0.8% in CAD/CHF patients; p=0.034). The effect of pre-transplant class 1 anti-HLA antibodies was not significant (p=0.510).

This study confirms results from prior studies which indicate that male gender and splenomegaly are important contributors to post-transplant platelet transfusion refractoriness. Previous findings indicating the impact of anti-HLA antibodies in the pre-transplant setting were not confirmed by this study. We hypothesize that the marked increase in sensitivity that characterizes current-generation anti-HLA antibody testing may have lowered the predictive value of this test. Associations between platelet responsivity, CAD/CHF and time from transplant are novel findings and will require confirmation in future studies.