Can Planned CD34+ Stem Cell Boost Prevent Poor Graft Function after Peripheral Blood Haploidentical Hematopoietic Transplantation?

Abstract

<h3>Introduction</h3> Traditional outcome measurements for hematopoietic recovery after Haploidentical HCT (haplo-HCT) with post-transplant cyclophosphamide (PTCy) may not capture a subset of patients who meet engraftment milestones, and have full donor chimerism, but subsequently suffer poor graft function (PGF) resulting in prolonged cytopenias and requiring transfusion support. This is termed secondary PGF, occurs in 5-26% of patients, and is associated with significant morbidity and mortality. CD34+ selected donor cell infusions have been employed in this setting without causing graft versus host disease (GVHD). <h3>Objectives</h3> We hypothesized that planned CD34+ stem cell boost (SCB) after PTCy could be used to prevent PGF after haplo-HCT. <h3>Methods</h3> Full donor chimerism was defined as ≥ 95% donor chimerism from bone marrow on day +30. Poor graft function was defined as transfusion dependence with ≥ two cytopenic lines (ANC < 1500 cells/mm3, platelets < 30 000/mm3 and Hgb < 8.5 g/dL) in the presence of full donor chimerism and in the absence of severe GVHD, CMV infection, and relapse. Patients underwent haplo-HCT per institutional protocol. Part of the donor product underwent CD34+ selection and cells were cryopreserved. On day 5-6, CD34+ selected donor product was infused. <h3>Results</h3> Between February 1, 2018 and August 30, 2018, nine patients underwent planned SCB after haplo-HCT. All patients received GVHD prophylaxis with PtCY, tacrolimus, and mycophenolate per institutional guidelines. The comparison cohort consisted of all patients who have undergone haplo-HCT at our institution between July 7^th, 2009 and October 8^th, 2018. All nine SCB patients had neutrophil engraftment (NE) with a median time of 16 days (13 – 25) compared with 17 days (12 – 78) in the comparison cohort (p = 0.40). Six of eight evaluable SCB patients had platelet engraftment (PE) with a median time of 22.5 days (12 – 55) compared with 26 days (9 – 134) in the comparison cohort (p = 0.675). One patient was censored due to death prior to PE on day +19. None of the 8 surviving patients developed secondary PGF. Two patients in the planned SCB group developed grade III acute GVHD involving the GI tract. The remaining seven had no GVHD. <h3>Conclusion</h3> These patients were selected on the basis of being high risk for graft failure or poor graft function. Two patients had active disease at the time of transplant and five of the nine patients were undergoing their second HCT. Time to NE and PE in the SCB group were 1 and 3.5 days faster than in the comparison group. This equivalency is notable given that the SCB patients were specifically selected for being high risk or graft failure. None of the SCB patients had secondary graft failure, compared with published rates between 5.6% and 26%. This approach appears to be safe with acceptable incidence of severe acute GVHD.