Prospective Evaluation Of Alternative Donor Availability In 708 Patients: Improved Allograft Access With Enlarging CB Inventory For All Patients Including Racial and Ethnic Minorities

Availability of suitably HLA-matched adult donors is a major barrier to allogeneic transplantation. Cord blood (CB) extends transplant access to many patients including minorities. However, some patients have no suitable unrelated donors (URDs) or CB grafts. Moreover, HLA-match is a critical determinant of URD & CB transplantation (CBT) outcome, & some patients will only have access to mismatched URDs &/or highly HLA-mismatched CB units. We evaluated if URD/ CB access has improved with increased size of URD registries & global CB inventories.

We prospectively collected ancestry data on patients undergoing searches between 10/2005-6/2013, & analyzed the availability of 9-10/10 HLA-matched URDs or 4-6/6 HLA-A,-B antigen, -DRB1 allele matched CB units by recipient ancestry in patients with hematologic malignancies. 10 HLA-allele matched URDs were given priority if available; otherwise HLA-mismatched URDs or double-unit CB grafts were chosen. Given our minimum CB total nucleated cell (TNC) dose criteria was increased from 1.5 to 2.0 x 10⁷/kg/unit in 2010, this analysis was restricted to units with a cryo. TNC dose ≥ 2.0. Also, as high-resolution typing was performed in all CB units & considered in unit selection since 2012, availability of CB units by 10-allele match was evaluated. Finally, given the global public CB inventory reached 500,000 units in 2011, URD/CB availability was compared between early (10/2005-12/2010) & recent (1/2011-6/2013) periods.

708 patients were evaluated including 223 (31%) of non-European origins. 511 (72%) patients received an URD transplant, 159 (22%) received a CBT, & 38 (5%) had no graft. URD recipients were predominantly of European ancestry (n = 400, 78%) whereas only 77 (48%) of CBT recipients had European origins. 30/38 (79%) patients without any graft were non-European with over half (n = 20) being of African ancestry. “No graft” patients had a higher weight (median 93 kg, range 50-151) than CBT recipients (median 68 kg, range 8-133), p < 0.001. The availability of grafts for Europeans & non-Europeans during the early versus recent time periods is shown (Table). Patient ancestry distribution was equal during the periods. The majority of Europeans received a 10/10 URD transplant in both periods & very few had no graft (< 1% recently). However, only approximately half of non-Europeans undergoing URD transplant received a 10/10 match. Moreover, of non-Europeans overall, only 29/129 (22%) had a 10/10 URD in the early period, & this has not improved recently [28/94 (30%), p = 0.28]. The percentage of non-Europeans receiving CB units with a ≥ 5/6 donor-recipient HLA-match has not improved (data not shown). However, while only 28% of non-Europeans received a CB graft consisting of ≥ 6/10 HLA-matched units in the early period, this donor-recipient HLA-match rate improved to 69% recently (p < 0.001), a rate similar to that of European patients. The percentage of non-Europeans with no graft has decreased from 17% to 8.5% (p = 0.08). Furthermore, while African patients were the largest group among no graft patients (68% in early period & 63% recently), of African patients without an URD, 16/31 (52%) had a CB graft in the early period versus 15/20 (75%) recently, p = 0.14.

Very few Europeans do not have an URD &/or CB graft. However, URDs are a poor donor source for patients with non-European ancestry. By contrast, CB is a critically important stem cell source for minorities, & transplantation of units with higher rates of 10 HLA-allele match is now frequently possible likely due to a bigger inventory & the use of high-resolution matching. Our analysis suggests CB availability in non-Europeans has improved. However, additional measures are required to extend graft access for African ancestry patients.

No relevant conflicts of interest to declare.