Abstract
Over 7 years ago, we developed a 2 step myeloablative approach to haploidentical HSCT in which patients, after conditioning with 12 Gy of total body irradiation (TBI days -9 to -6), receive a fixed dose of 2 x 108/kg of donor T cells (DLI-step 1 of HSCT) immediately after the last fraction of TBI. This large dose of haploidentical lymphocytes results in high fever, and in some cases, diarrhea and rash, developing on days-5 and -4. Cyclophosphamide (CY) 60 mg/kg, given on days -3 and -2 for T cell tolerization, results in complete resolution of this “alloreaction.” Tacrolimus and MMF are begun on day-1, followed by a CD 34 selected donor product infused on day 0 (Step 2 of HSCT). The separation of the lymphoid and myeloid portions of the graft (1) avoids the exposure of allogeneic stem cells to CY, (2) avoids the polarization of T cells to a TH2 phenotype because the donor T cells are collected prior to the initiation of G-CSF, (3) allows the administration of a fixed dose of T cells establishing a consistent platform from which to compare outcomes, and (4) provides a method to deliver higher doses of T cells. Recent data suggests that T cell doses > 1.1 x108/kg are associated with superior GVT effects as compared to lower T cell doses (Guo et al. JCO, 2012;30:4084 and Colvin et al. BBMT, 2009;15:421). To date, over 180 patients have undergone 2 step haploidentical myeloablative and reduced intensity HSCT at our institution. In the initial myeloablative haploidentical trial (2006-2009), twenty-seven patients underwent treatment using this approach. Immune reconstitution was brisk with median CD3/4 and CD3/8 counts at day +28 of 33.6 and 28.7 cells/ul respectively Cumulative incidences of grades III-IV graft versus host disease (GVHD), NRM, and relapse-related mortality were 7.4%, 22.2%, and 29.8% respectively. OS for the 12 patients without disease at HSCT was 75% and 48% for the whole cohort with 51-79 (median 63) months of follow-up.
A 2nd generation myeloablative 2 step trial specific to patients without morphologic evidence of disease at the time of their haploidentical HSCT completed accrual in 2013. Twenty-eight patients with AML (15), ph+ ALL (4), B cell ALL (4), T cell ALL (2), MDS (1), mantle cell NHL (1), and hepatosplenic NHL (1), were treated and are 3-32 (median 14) months post HSCT. All patients engrafted and no patients died of GVHD or infection. Median CD3/4 and CD3/8 counts at 28 days were 74 and 47 cells/ul respectively. The only death related to treatment occurred in a patient who suffered a subdural hematoma after an LP performed for fever. Four patients have relapsed (14%), 3 of whom have died of their disease. The probability of OS of the patients treated on the current trial is 85% at 15 months. The combination of the most recently treated 28 patients with the 12 patients treated on the initial trial who were in morphologic CR at the time of HSCT, has resulted in a probable OS rate of 78% with 3 to 79 (median 18 months of follow-up).
For patients without morphologic evidence of their disease at the time of HSCT, the 2 step myeloablative approach to haploidentical HSCT has been associated with robust immune reconstitution resulting in low rates of infectious death. There have been no deaths from GVHD, and very low rates of treatment-related mortality. OS rates of the good risk patients treated on this approach are very high due to this low degree of NRM. The results of this 2nd generation trial have confirmed that the 2 step myeloablative approach to haploidentical HSCT is safe and efficacious and produces outcomes that are comparable to those reported by CIBMTR (2000-2010) for recipients of matched related or other alternative donor grafts. This approach has become our alternative HSCT strategy of choice allowing virtually every patient to be treated with HSCT rapidly at our institution.
Kasner: Roche: Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.
