Graft-Specific HLA-Antibodies Do Not Influence Unit Dominance and Do Not Prevent High Rates of Sustained Donor Engraftment in Recipients of Double-Unit Cord Blood Transplantation

Pre-transplant human leukocyte antigen (HLA)-antibodies have an established role in failed engraftment after single-unit cord blood (CB) transplantation (CBT). Their role in double-unit CBT (DCBT) engraftment, however, is more controversial. Some transplant centers advocate pre-transplant screening of the recipient for HLA-antibodies, and avoidance of units with the corresponding antigens if antibodies are detected.

We evaluated the influence of pre-transplant HLA-antibodies on sustained donor neutrophil engraftment and unit dominance in 82 recipients (median 48 years, range 2–69) of 4–6/6 HLA-A, B antigen, DRB1 allele matched double-unit CB grafts, and transplanted for hematologic malignancies at MSKCC from July 2008 to July 2012. HLA-antibodies were measured by the HLA Laboratory of the American Red Cross Blood Service using Single Antigen Luminex beads and analyzed by HLA Fusion software. The cut-off for positive results was a normalized mean fluorescence intensity value > 1000.

Overall, 28/82 (34%) patients were positive for HLA-antibodies [16 (19.5%) had antibodies without graft specificity, and 12 (14.5%) had antibodies with graft specificity]. These patients were more likely to be female with acute leukemia and cytomegalovirus seropositive. All patients with graft-specific antibodies received myeloablative conditioning; 5 had antibodies against class I HLA, 6 against class II, and one patient had antibodies to both Class I/II. Moreover, of these 12 patients, 6 had antibodies against one unit, and 6 had antibodies against both units. Neutrophil engraftment according to antibody presence is summarized in the Table. 64 of 66 (97%) evaluable myeloablative DCBT recipients (Table 1A) engrafted at a median of 24 days (range 12–40); one patient without antibodies and one with antibodies against both units had primary graft failure, both in the setting of early onset multi-organ failure. Both patients were 100% donor with one unit but did not recover counts. Of the 6 patients with antibodies to one unit, 3 engrafted with that unit and 3 with the opposite unit. Of the 6 patients with antibodies against both units, one had clinical graft failure as described above, and the 5 others had sustained donor engraftment (4 with one unit and one with both). In engrafting myeloablative recipients, the median time to neutrophil recovery was 8 days slower in patients with graft-specific antibodies, but the engrafting unit in these patients also had the lowest infused dose: median CD34+ cell dose/kg if no antibodies 0.83 × 10⁵/kg, non-specific antibodies 1.09 × 10⁵/kg, and antibodies specific to graft 0.65 × 10⁵/kg. In non-myeloablative recipients, 15/16 (94%) engrafted (Table 1B). The single patient with graft rejection and autologous recovery had HLA-antibodies that were not graft specific.

11 of 12 double-unit CBT recipients with graft-specific antibodies engrafted successfully. While myeloablative recipients with graft-specific antibodies engrafted more slowly, this may be explained by a lower infused CD34+ cell dose of their engrafting units. Multivariate analysis of larger patient numbers will be required to further evaluate the effect of graft-specific antibodies on engraftment speed. At this time, however, the presence of graft-specific antibodies should not preclude DCBT, and whether their presence should influence graft selection is not clear. Furthermore, there is no suggestion that the presence of HLA-antibodies influences unit dominance after DCBT.

Giralt:Celgene: Honoraria, Research Funding.