Background and objectives: A proportion of patients develop poor graft function (PGF) following an allogeneic hemopoietic stem cell transplant (HSCT). It is uncertain whether a boost of donor marrow or blood cells is beneficial in terms of trilineage recovery and non-relapse-related mortality (NRM).
Design and methods: The aim of this study was to compare outcomes in patients with PGF and full donor chimerism following an allogeneic HSCT who did or did not receive a boost of donor stem cells. The study included patients with primary PGF--i.e. those failing to achieve sustained graft function- and secondary PGF--i.e. those developing PGF after complete hematologic recovery. We studied 54 patients with PGF: 20 patients received no further donor cell infusion (group A), 14 received a boost of unmanipulated marrow or blood cells from the original donor, without further conditioning (group B), and 20 received donor cells after CD34 selection without conditioning (group C). The three groups were comparable for disease phase, patients' age, donor type, primary or secondary PGF, full donor chimerism and duration of PGF.
Results: Trilineage recovery was seen in 40%, 36% and 75% of the patients in, respectively, groups A, B and C (p=0.02). In multivariate Cox analysis trilineage recovery was more frequent in patients with secondary PGF (RR of complete recovery 2.82, p=0.01) and in patients receiving CD34+-selected cells (RR of complete recovery 3.0; p=0.007). There was no effect of donor type on hematologic recovery. The rate of NRM was 55%, 64%, 20% in groups A, B and C, respectively (p=0.06) and was highly correlated with trilineage recovery (RR 0.36, p<0.0001). PGF was the primary cause of death in 30%, 21% and 10% of the patients in the three groups, graft-versus-host disease (GVHD) in 5%, 36%, and 10%.
Interpretations and conclusions: In patients with poor graft function (a) a boost of CD34+-selected peripheral blood donor cells is associated with a high chance of trilineage recovery and a low risk of acute GVHD; (b) a boost of unmanipulated donor cells does not seem to offer a survival advantage over no infusion of cells; and (c) NRM is lower when using peripheral blood cells for the boost. These data may be useful when discussing second stem cell donations for patients with poor graft function.