Abstract 1983

Background:

Acute graft versus host disease (aGvHD) is a significant cause of morbidity and mortality after allogeneic Hematopoietic cell transplantation (HCT) with multiple risk factors identified, mostly in adult patients' cohorts. Few studies addressed this issue in children. This study was designed to determine aGvHD incidence and risk factors in a large cohort of children with acute leukemia who underwent related HCT in a single center.

Patients and Methods: This is a retrospective study to evaluate aGvHD incidence and risk factors after related, myeloablative, first HCT in 225 children (age ≤ 14 years) with acute myeloid leukemia (AML; n=116), or acute lymphoid leukemia (ALL; n= 109) from 1993 to 2008, all treated at King Faisal Specialist Hospital & Research Center (KFSHRC). There were 76 females and 149 males. Conditioning regimens were chemo-based (busulfan, cyclophosphamide +/− etoposide) or radiation-based (total body irradiation -TBI- and cyclophosphamide), and GvHD prophylaxis consisted of cyclosporine +/− methotrexate (MTX). At the time of HCT, 120 patients were in CR 1, 84 patients in CR2, 18 patients in CR 3, and 3 patients had progressive disease. Donors included: HLA-identical siblings in 208 patients, HLA-identical parents in 10 patients, and HLA class I, one-antigen-mismatched relatives (siblings/parents) in 7 patients. Acute GvHD was defined using standard published criteria.

Results:

Cumulative incidence of aGvHD was 42.2%; 73 patients had mild grade I-II aGvHD, and 22 had severe, grade III-IV aGvHD. In univariate analysis, factors significantly associated with increased aGvHD risk were patient's age under 5 years vs. older (P= 0.002), higher dose of CD34 (median of 6.2 vs. 5.15 × 106/kg of recipient body weight; P= 0.031), female donor vs. male donor (P= 0.004), HLA-identical parent/HLA 1-antigen mismatched parent/sibling donor vs. HLA-identical sibling donor (P < 0.0001), ALL vs. AML (P < 0.0001), TBI vs. chemo-based conditioning (P < 0.0001), no MTX vs. MTX in the prophylaxis regimen (P= 0.004) and lower recipient body weight at time of HCT (median of 20 vs. 24.35 Kg; P=0.008). In subgroup analysis, female gender was associated with more aGvHD among matched donor-recipient gender (P=0.041) vs. mismatched (P=0.075). Age at HCT significantly correlated with recipient body weight at HCT and CD34 dose/Kg (Correlation co-efficient: 0.893; p-value <0.001 and - 0.39, p-value <0.001). Similarly, TBI was significantly correlated with primary disease ALL vs. AML (p-value<0.001). Therefore, only patient's age at HCT, donor's gender, HLA typing, use of TBI and use of MTX were tested in the multivariate analysis; younger patient's age (OR: 5.565), female donor (OR:2.312), TBI-conditioning (OR: 4.916) and use of HLA-identical parent, and one-antigen mismatch parent/sibling (OR: 15.818), and the omission of MTX from the GvHD prophylaxis (OR:2.917), all conferred a significantly increased risk of aGvHD (all p-values < 0.05). No significant survival difference was noted between patients with aGvHD and those without. However, the grade of aGvHD seemed to impact on overall survival; patients with mild aGvHD (grade I, and II) had a significantly improved survival (57%) vs. (50%; 63.433 months) in patients with no aGvHD; conversely, patients with severe aGvHD (grade III, IV) had a significantly worse survival (14%) with 7.00±1.309 months (P< 0.001, Tarone Ware).

Conclusion:

Our analysis demonstrates the importance of particular risk factors for aGvHD in pediatric HCT such as use of TBI, younger age, and higher CD-34 dose; these results should help the transplanter to better customize GvHD prophylaxis when risk factors are identified pre-HCT. This analysis also demonstrates the positive effect of mild acute GVHD on survival of patients with acute leukemia inferring a favorable graft vs. leukemia effect.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution