Outcomes of Unrelated Cord Blood Transplantation for Familial Hemophagocytic Lymphohistiocytosis: Risk Factors and Long Term Follow-up. an Analysis on Behalf of Eurocord, Cord Blood Committee of Cellular Therapy and Immunobiology Working Party of the EBMT and Inborn Errors Working Party of the EBMT

Familial Hemophagocytic Lymphohistiocytosis (FHLH) is a systemic hyperinflammatory syndrome caused by excessive activation of lymphocytes and macrophages that produce high levels of cytokines as a result of genetic defects affecting the granule-mediated cytotoxicity pathway. Allogeneic hematopoietic stem cell transplantation (HSCT) is so far the only available curative option for the disease and the majority of studies have been conducted in patients given either bone marrow or peripheral blood transplantation. Previous case series of unrelated umbilical cord blood transplants (UCBT) reported survival rates ranging from 25%-85%. We analyzed the outcomes of 118 children from 43 EBMT centers with HLH undergoing single UCBT and reported to Eurocord. The median year of UCBT was 2010 (range 1996-2014). The median follow-up was 75 months (range 3.3-211). The majority of patients had genetic diagnosis of familial HLH (64%) of which 37.4% were due to perforin defect, 30.6% due to Munc 13-4 defect, 25.3% due to Munc 18-2, 5.4% due to syntaxin 11 and 1.3% due to perforin and syntaxin 11 defects. The median age at diagnosis and at UCBT were 0.3 years (range 0-14) and 1 year (range 0.1-14), respectively. The median time from diagnosis to UCBT was 5.3 months (range 0-57). Amongst 113 patients with known disease status, most patients underwent UCBT in complete remission 68/113 (60%) with 25/113 (22%) in partial remission and 20/113 (18%) having active disease at the time of transplant. Conditioning regimen was Busulfan or Treosulfan-based myeloablative in 78% of the patients while 22% received a reduced intensity conditioning. Seventeen (14%) children did not receive serotherapy as part of conditioning. Thirty-five patients (30%) received a 6/6 HLA matched cord blood unit, 56 (48%) received a 5/6 HLA mismatched cord blood unit, 24 (21%) received a 4/6 (21%) and 1 patient received 3/6 (1%) HLA mismatched graft. The median number of total nucleated cells (TNC) infused was 9.9 x10⁷/kg (range 1.26 - 42.5 x10⁷/kg).

The cumulative incidence (CI) of neutrophil engraftment was 85% (78%-92%) at 60 days with 79% of engrafted patients exhibiting full donor chimerism at day 100.

The CI of grade II to IV acute graft-versus-host disease (aGVHD) was 33% (26%-43%) at day 100, while the 6-year CI of chronic GVHD was 17% (11%-26%). The CI of relapse and non-relapse mortality (NRM) at 6 years was 14% (9%-22%) and 36% (28%-46%), respectively.

The estimated 6-year overall survival (OS) and disease-free survival (DFS) was 55% (45%-64%) and 50% (40%-60%), respectively. Fifty-two patients died: 10 of relapse and 37 of transplant related causes (the most frequent causes being: infections, n=8; GVHD n=8 and ARDS, n=4).

In univariate analysis, children receiving more than 9.9x10⁷/Kg infused TNC had higher rates of engraftment (90% versus 79%, p=0.02) and better OS (65% versus 42%, p=0.04) compared to those patients receiving less cells. Interestingly, a number of TNC more than the median (TNC>9.9x10⁷/Kg) was also associated with decreased incidence of grade II-IV acute (25% versus 43%, p=0.05) and chronic GVHD (3% versus 32%, p<0.01). In multivariate analysis, a number of TNC >9.9x10⁷/Kg was associated with improved OS (HR 0.48, 95%CI 0.27-0.88, p=0.02), greater engraftment rate (HR 1.80, 95%CI 1.16-2.80, p=0.009) and decreased incidence of chronic GVHD (HR 0.10, 95%CI 0.02-0.45, p=0.003).

DFS at 6 years was 69% for a HLA matched (6/6), 49% for 5/6 and 34% for 2 or 3/6 and 6 years NRM was 23%, 39% and 45%, respectively. In multivariate analyses, a graft with more than 2 HLA mismatches was associated with worse DFS and higher NRM (HR 2.81, 95%CI 1.25-6.35, p=0.01 and HR 2.72, 95%CI 1.02-7.23, p=0.04, respectively). We did not observe an impact of disease status at the time of transplant on OS, DFS, relapse incidence and NRM. There was no difference between the use of reduced intensity versus myeloablative conditioning regimen or using serotherapy on the study outcomes. No risk factors were found to be associated with relapse incidence. In conclusion, UCBT for FHLH with a higher cell dose and a well HLA matched cord blood unit (6/6 or 5/6) is a suitable alternative stem cell source for children who lack an HLA- matched BM or PBSC donor. Further studies are warranted to establish the best conditioning regimens to improve outcomes with reduced toxicity and to compare outcomes after UCBT with other alternative approaches such as unrelated donors and haploidentical HSCT.