Haploidentical Transplantation As Salvage Therapy For Disease Relapse After Allogeneic Hematopoietic Cell Transplantation

Therapeutic options and outcome for patients with relapse after allogeneic hematopoietic cell transplantation (HCT) are poor. For individual patients, a second or even third allogeneic HCT can be considered with curative intention. However, treatment-related mortality (TRM) and the risk of relapse after secondary allogeneic HCT are high. Therefore, the use of a haploidentical graft allowing for profound NK- and T-cell-alloreactivity after a reduced-intensity conditioning regimen (RIC) may optimize disease control by enabling potent graft versus leukemia effects and reduction of TRM.

We here retrospectively evaluated 26 consecutive patients undergoing haploidentical HCT as second (n=24) or third (n=2) allogeneic HCT at our center between 2003-2012. Diagnoses comprised relapse of AML (n=17) or ALL (n=7), blast crisis of CML (n=1) and transplant failure (n=1). There were 16 male and 9 female patients with a median age of 36 years (range 18-59).

For RIC, fludarabin, thiotepa and melphalan were used in 16 patients, clofarabin, thiotepa and melphalan in 6 patients and other regimens containing variable combinations of cyclophosphamide, busulfan, TBI and treosulfan in 4 patients. Grafts were manipulated by CD3/CD19 depletion (n=19), TCRαβ depletion (n=1) or CD34 selection (n=6) and consisted of a median of 7.71 x 10⁶ CD34⁺cells/kg bodyweight.

The median interval from first HCT to second HCT was 18 months (range 5-145), and 10 and 16 months from second HCT to third HCT in the two patients undergoing a third HCT. Only 35% (n=9) of the patients receiving a second HCT were in complete remission (CR), while 65% (n=16) were in partial remission (PR). Among the patients receiving a third HCT, one had active disease, while the other was in CR. All patients achieved engraftment of the neutrophils at a median time of 11 days (range 8-26) and platelet engraftment was reached at a median time of 15 days (range 9-35, except one patient at day 375). At present, 5 patients (19%) are alive and in CR with a median follow-up of 1870 days (range 281-3941), while 35% (n=9) died from relapse; non-relapse-mortality was 46% (n=12). Kaplan-Meier estimated overall survival for all patients at 1 year was 33% (52% for patients in CR versus 18% in PR) and at 3 years 17% (26% for patients in CR versus 12% in PR). Causes of death in patients with second HCT included severe infections (n=8), organ failure (n=1), haemorrhage (n=1) and progressive multifocal leukoencephalopathy (n=2). Of the patients with third HCT, one died from respiratory insufficiency due to pulmonary haemorrhage, the other is still alive and in CR.

Acute graft versus host disease (GVHD) occurred in 11 patients with predominantly mild presentation (grade 1: n=9, grade 2: n=2), limited chronic GVHD was apparent in 5 patients with no case of extensive GVHD.

Haploidentical HCT is a feasible salvage concept for patients with relapse after HCT with promising results even in patients not in CR.

No relevant conflicts of interest to declare.