Our group has published encouraging results in adult patients who underwent full haplotype mismatched related graft after a novel chemotherapy-alone conditioning regimen (

Lacerda et al,
Biol Blood Marrow Transplant
Biol Blood Marrow Transplant
). With a longer follow-up, patients with acute myeloid leukemia transplanted in 1st, 2nd, or 3rd remission have a 57% probability of disease free survival at 8 years. In the current study, we assessed the degree of long-term immune reconstitution in patients submitted to a haploidentical HSCT more than 5 years ago. A total of 5 patients (mean age 25 years) at a median of 7 years post-transplant were compared with their parental donors (mean age 48 years) and 5 age-matched controls (AMC) (mean age 29.5 years). There were no significant differences in absolute counts of CD3+, CD4+, CD8+, CD19+ and CD56+CD16+ cells. Within the CD4+ population, there was a trend for an increased percentage of naïve cells (CD45RA+CD62L+CD27+) in patients (57.8%) as compared to donors (46.1%) and AMC (53.4%). This was associated with a decreased percentage in late memory CD4+ cells (CD45RO+CD45RA-CD27-) in the patients (4.3%) as compared to donors (9.1%) and AMC (7.1%). In the CD8+ population, patients had a significantly higher percentage of naïve (CD45RA+CD45RO-CD27+) cells than their donors (67.9% versus 29.6%, p=0.002) (also significant in absolute number counts) and a lower percentage of late memory CD8+ (CD45RA+CD45RO-CD27-) cells than both their donors (2.2% versus 28.6%, p=0.0001) (also significant in absolute number counts) and AMC (2.2% versus 9.2%, p=0.05). With respect to the evaluation of the relative contributions of de novo thymopoiesis and peripheral expansion to the maintenance of the naïve T cell pool, patients had a significantly greater proportion of recent thymic emigrants within the CD4+ T cell population (CD45RA+CD62L+CD31+) than both their donors (40.8 versus 20.7%, p=0.019) and AMC (40.8% versus 17.5%, p=0.0006). In agreement with these observations, patients had higher levels of TRECs than both their donors and AMC, although this did not reach statistical significance. Telomeres of naïve CD4+ and CD8 + cells were shorter in patients as compared to AMC, but longer than their older donors. In CD4+ naïve T cells, the difference in telomere length reached statistical significance when patients and AMC (p=0.03), and when donors and AMC were compared (p=0.03); whole differences in CD8+ T cells did not reach statistical significance. We also found decreased serum IL-7 levels in patients as compared to donors (5 pg/ml versus 6.8 pg/ml, p=0.04) and AMC (5 pg/ml versus 9.5 pg/ml, p=0.005). Furthermore, our patients also had normal proportions of regulatory T cells and within this population we found a significantly higher percentage of cells with a naïve phenotype (CD45RO-CD62L+) when compared to donors (37.9% versus 18.8%, p=0.02) and AMC (37.9% versus 18.4%, p=0.01). Together, our data suggest that despite the HLA mismatch, there is a major contribution of thymic output to the reconstitution of the T cell pool, resulting in a normal balance of naïve and memory T cell subsets in these patients.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author