Leukoreduction to Prevent Alloimmune Platelet (Plt) Refractoriness in a Dog Transfusion Model: Types of Residual White Blood Cells (WBCs) Directly Affect Transfusion Outcomes.

Abstract Introduction: Current practice assumes that just a quantitative reduction in the number of transfused wbcs to <1 to 5 x 106/transfusion is sufficient to prevent plt alloimmunization. However, our studies indicate that different leukoreduction strategies vary in their ability to remove immunizing wbcs, and this correlates with rates of alloimmune plt refractoriness in immunocompetent recipients. Experimental Design And Methods: Pairs of donor-recipient dogs were selected either at random, as having a shared DLA DR-B epitope, or as being specifically mismatched for this DLA locus. Non-leukoreduced or leukoreduced radiochromium-labeled donor plts were transfused weekly for up to 8 weeks or until the onset of plt refractoriness defined as <5% of the donor dog’s plts circulating in the recipient at 24 hours post-transfusion. Three methods of leukoreduction were evaluated: centrifugation leukoreduction (C-LR); filtration leukoreduction (F-LR) with different types of filters; or combined F-LR/C-LR. Flow cytometry was used to identify the types of residual wbcs following leukoreduction. Table 1 gives the number of residual wbcs and the transfusion outcomes based on the leukoreduction strategy used, while Table 2 gives the relative proportion of the types of residual wbcs after leukoreduction compared to the overall results for the leukoreduction method used. Results: Table 1 Method Of Leukoreduction Filter (Manufacturer) Average Residual WBCs Donor-Recipient DR-B Relationship Non-Refractory Recipients / Recipients Transfused ND-Not done. *Platelets were filtered sequentially using two PLF-1 filters. **Lower limit of detection of the assay. None --- 6.7 x 106 Random 1/3 (33%) Shared Epitope 0/4 (0%) C-LR --- 4.7 x 104 Random 3/21 (14%) F-LR: PL1-B (Pall) 5.0 x 104 ND ND PLF-1 (Pall) 7.9 x 104 Random 3/8 (38%) PLF-1 x2* (Pall) <3 x 103** Random 1/5 (20%) PLS-5A (Fenwal) 3.2 x 104 Mismatched 4/6 (66%) F-LR/C-LR: PL1-B (Pall) <3 x 103** Mismatched 1/9 (11%) <3 x 103** Shared Epitope 5/7 (71%) PLF-1 (Pall) <3 x 103** Random 15/16 (94%) <3 x 103** Mismatched 3/3 (100%) PLS-5A (Fenwal) <3 x 103** Mismatched 9/9 (100%) Table 2 RESIDUAL WBCs Lymphocytes Method of Leukoreduction T CD4 dim T CD4 bright B Monocytes Total Non-Refractory Recipients (%) ND-Not done. *Non-shared DR-B donor-recipient pairs/shared DR-B donor-recipient pairs (p=0.03). None ++ ++ +++ 14% C-LR + ++ + ++ 14% F-LR: PL1-B ++ ++ +++ +++ ND PLF-1 + +++ + 38% PLS-5A ++ + ++ 66% F-LR/C-LR: PL1-B ++ +++ + 11%/71%* PLF-1 ++ +++ 95% PLS-5A ++ + 100% Conclusions: 1) a quantitative reduction in wbcs does not prevent plt refractoriness; 2) the types of residual wbcs correlate directly with transfusion outcomes; 3) even after residual monocytes are removed by F-LR using PLF-1 and PLS-5A filters, residual CD4 bright T lymphocytes are associated with a high percentage of refractory recipients; 4) following F-LR/C-LR with PLF-1 and PLS-5A filters leaving B lymphocytes and CD4 dim T lymphocytes (NKT cells?), refractoriness is prevented even to DR-B mismatched donors; and 5) DLA DR-B matching significantly improves transfusion outcomes when residual monocytes remain following F-LR/C-LR using a PL1-B filter.