Advertisement

Lower levels of vitamin A are associated with increased gastrointestinal graft-versus-host disease in children

Dana T. Lounder, Pooja Khandelwal, Christopher E. Dandoy, Sonata Jodele, Michael S. Grimley, Gregory Wallace, Adam Lane, Cynthia Taggart, Ashley C. Teusink-Cross, Kelly E. Lake and Stella M. Davies

Article Figures & Data

Figures

  • Figure 1.

    Vitamin A levels and transplant outcomes. (A) Free vitamin A levels in patient plasma 30 days posttransplant are in a narrow range and normally distributed with a median value of 1.296 ng/mL. (B) Vitamin A levels below the median are associated with increased incidence of grades 2-4 GVHD (38.6% vs 12.4% at 100 days, P = .0008). (C) Vitamin A levels below the median are associated with increased incidence of GI GVHD (30.4% vs 7% at 100 days, P = .002). (D) TRM was also increased in patients with vitamin A levels below the median compared with patients with vitamin A levels above the median (17.7% vs 7.4% at 1 year, P = .03).

  • Figure 2.

    RBP levels and transplant outcomes. RBP4 levels measured in plasma day 30 posttransplant had no effect on transplant outcomes including cumulative incidence of GI GVHD (A) and TRM (B).

  • Figure 3.

    Cumulative incidence of MBI-LCBIs. Incidence is higher in patients with vitamin A levels below the median at day 30 posttransplant compared with those with vitamin A levels above the median (24% vs 8% at 1 year, P = .03).

  • Figure 4.

    Mucosal damage leads to increased incidence of GI GVHD and is increased in patients receiving a myeloablative conditioning regimen (MAC) compared with those receiving a reduced intensity conditioning regimen (RIC). (A) Cumulative incidence of GI GVHD is increased in patients with more mucosal damage as measured by I-FABP levels below the median (27% vs 18% at 100 days, P = .0024). (B) I-FABP levels are lower at day 7 posttransplant in patients who received MAC vs RIC, which correlates with increased mucosal damage.

Tables

  • Table 1.

    Patient and transplant demographics (n = 124)

    CharacteristicValue
    Age, y
     Median8
     Range0.4-32
    Diagnosis, n (%)
     Malignancy33 (26)
     Immune deficiency44 (36)
     Bone marrow failure47 (38)
    Conditioning regimens, n (%)
     Myeloablative72 (58)
     Reduced intensity52 (42)
    Stem cell source, n (%)
     BM104 (84)
     PBSC14 (11)
     CB6 (5)
    Cell dose
     Median6.8 × 108 cells/kg
     Range15.4 × 106-32.75 × 108 cells/kg
    Donor, n (%)
     Related47 (38)
     Unrelated77 (62)
    Match, n (%)
     BM and PBSC
      10/1092 (74)
      9/1023 (19)
      8/103 (2)
     CB
      6/62 (1)
      5/62 (1)
      4/62 (1)
    GVHD prophylaxis, n (%)
     CSA ± other108 (87)
     Tacrolimus ± other5 (4)
     Ex-vivo T-cell depletion9 (7)
     Other2 (2)
    • BM, bone marrow; CB, cord blood; CSA, cyclosporine; PBSC, peripheral blood stem cell.

  • Table 2.

    Multivariate analysis of development of GVHD and TRM

    GVHD OR (95% CI)P valueTRM OR (95% CI)P value
    Vitamin A level3.32 (1.5-6.7).0033.07 (0.9-10.1).06
    Malignancy1.9 (1-3.7).061.89 (0.6-5.8).27
    HLA match1.42 (0.5-3.9).51.11 (0.2-6.7).9
    Stem cell source
     BM1.8 (0.5-6).31.74 (0.2-13.7).6
     CB1.7 (0.1-24.2).73.73 (0.08-164.9).5
     PBSCBVNABVNA
    • BV, baseline value; NA, not available.

  • Table 3.

    Correlation of vitamin A levels and CCR9 expression on TEM cells (n = 42)

    Vitamin A
    CCR9 expressionCorrelationP value
    CD8 TEM−0.34.03
    CD8 Naive−0.21.19
    CD4 TEM−0.11.58
    CD4 Naive−0.27.09