Analysis of Risk Factors for Hepatic Sinusoidal Obstruction Syndrome after Allogeneic Hematopoietic Stem Cell Transplantation in Pediatric Patients

Background: Hepatic sinusoidal obstruction syndrome (SOS), which is also called veno-occlusive disease of the liver, remains a serious complication after hematopoietic stem cell transplantation (HSCT). Over the last years, some risk factors have already been identified to be associated with SOS. However, the cause of SOS is still not fully understood and the mortality remains high, especially for SOS leading to multi-organ failure with a mortality rate up to 84%. The aim of our study was to analyze several risk factors of SOS in pediatric patients undergoing allogeneic HSCT. In addition, we investigated new potential risk factors.

Methods: We retrospectively analyzed 105 children who underwent allogeneic HSCT for the first time and did not receive a defibrotide prophylaxis. All transplantations were performed between January 2007 and December 2018 in a single center. The median age was 8.6 years and stem cell source was either bone marrow (n = 74) or peripheral blood (n = 31). Underlying diseases were acute lymphoblastic leukemia (n = 27), acute myeloid leukemia (n = 25), myelodysplastic syndrome (n = 14), lymphoma (n = 2), solid tumor (n = 12) and genetic disease (n = 25). All patients received a myeloablative conditioning regimen. We analyzed the transplantation-related factors graft source, donor-recipient human leukocyte antigen match, donor age, donor sex and conditioning regimen based on busulfan or total body irradiation. Furthermore, we investigated the patient-related factors patient age, patient sex, prior treatment with gemtuzumab ozogamicin as well as the following laboratory parameters: aspartate transaminase, alanine transaminase, cholinesterase, glutamyl transpeptidase, lactate dehydrogenase, alkaline phosphatase, ferritin, albumin, total bilirubin, C-reactive protein and international normalized ratio (INR). All laboratory parameters were measured before HSCT and cutoffs were determined by reference values and receiver operating characteristic (ROC) curves. SOS was defined by modified pediatric Seattle criteria up to day +30 after HSCT because nearly all transplantations were performed before the new pediatric criteria of the European Society for Blood and Marrow Transplantation have been published. In univariate analysis, chi-square test and Fisher's exact test were used. Additionally, the Mann-Whitney U-test was performed to compare the median values of continuous variables. Significant variables (P < .05) were entered in multivariate analysis, which was carried out by using backward stepwise logistic regression.

Results: SOS occurred in 15 out of 105 transplantations (14.3%). The median time of SOS onset was 12 days after HSCT (range, 1 day - 26 days). Three patients died of multi-organ failure following SOS (20%). This mortality rate was very low compared to other studies because our patients were treated with defibrotide immediately after being diagnosed with SOS. In univariate analysis, we found a significant association between patient age <1 year and SOS (Odds Ratio (OR) = 7.25, P = .037). Furthermore, a prior treatment with gemtuzumab ozogamicin (OR = 11.00, P = .020) showed a significant correlation. Patients who developed SOS had a significantly higher median ferritin level (2816.9 ng/mL vs. 1554.0 ng/mL, P = .026). Based on this observation, different ferritin cutoffs were selected by ROC analysis. Ferritin >1500 ng/mL (OR = 4.00, P = .033), ferritin >2000 ng/mL (OR = 4.69, P = .016) as well as ferritin >2400 ng/mL (OR = 5.29, P = .005) revealed significant P values. Besides these results, INR ≥1.3 (OR = 5.91, P = .009) was significantly associated with SOS. In multivariate analysis, the following variables showed P values less than .05: treatment with gemtuzumab ozogamicin (OR = 9.24, P = .048), ferritin >2400 ng/mL (OR = 5.74, P = .023) and INR ≥1.3 (OR = 8.02, P = .007).

Conclusions: Our data confirm the risk factors of young patient age (<1 year), prior treatment with gemtuzumab ozogamicin and high serum ferritin (>2400 ng/mL) in the pediatric population. Moreover, we report for the first time that there is a significant association between high INR (≥1.3) before HSCT and the occurrence of SOS. Especially this new finding could improve the risk stratification of SOS and should be evaluated in further trails.