Mortality from relapsed disease remains a significant barrier to long term survival (OS) after HI HSCT despite presumed heightened alloreactivity from the mismatched graft. The Jefferson group uses a 2 step approach to HI HSCT where patients are typically conditioned with 12 Gy total body irradiation (TBI) in 8 fractions of 1.5 Gy over 4 days, immediately followed by an infusion of 2 x 108/kg donor CD3+ cells (DLI). After 2 rest days, cyclophosphamide (CY) 60 mg/kg daily x 2 is given for bidirectional tolerization, followed a day later by a CD34 selected stem cell infusion. In an attempt to maximize graft versus tumor (GVT) effects, we changed the timing of the TBI in the 2 step approach. In this updated regimen, TBI was given in 6 fractions of 2.0 Gy over 3 days, resulting in a 24 hour delay between conditioning and DLI. Theoretically, the extra time reduced residual disease burden prior to the introduction of the DLI, in turn reducing the number of donor T cells activated by tumor, thus avoiding their elimination by CY.

Major HSCT endpoints of OS (Kaplan Meier), relapse and non-relapse mortality (NRM), acute and chronic graft-versus-host disease (GVHD) [cumulative incidence (CI)-EZR Software v 1.37] were assessed using this updated 2 step HI HSCT approach.

Forty patients, median age 51 (range 19-65) years with AML (13), MDS (7), B ALL (7), PH+ ALL (4), T cell ALL (2), Burkitt (2), and other heme malignancy (5) with revised disease risk assessment scores of intermediate (21), high (17), and very high (2) were treated from 2013 to 2017. Median follow-up is 36 (range 15 to 56) months.

At 24 months, probability of OS was 59%, CI NRM and relapse were 34% and 15% respectively. CI aGVHD (grades 2-4), (grades 3-4), and cGVHD were 38%, 5%, and 20%. Median d+28 T cell chimerism of 36 patients engrafted and alive was 100% (range 97% to 100%). Median CD3/4 and CD3/8 counts at d +28 were 49 (range 9-417) and 54 (8-1329) cells/ul. Of the 4 remaining patients, two without donor specific antibodies rejected their graft, one with a large burden of CMML at the time of HSCT. An additional patient relapsed prior to the attainment of sustained donor T cell chimerism and one patient died of sinusoidal obstructive syndrome prior to d+28. Causes of death were infection (7), regimen toxicity (4), GVHD (2), and disease (3).

This updated 2 step regimen was associated with a highly acceptable 2 year OS rate and low rates of disease recurrence. Of the patients that died, cause of death was primarily due to NRM and not relapsed disease suggesting that the added extra day may have enhanced GVT effects. In the absence of donor specific antibodies, the 2 early and 1 late graft rejections are atypical for a 2 step MA HI HSCT approach and were potentially caused by a rebound recipient hematopoiesis allowed by the delay in the DLI in a minority of patients. While formal comparison to prior patient cohorts is not feasible, this relapse rate compares favorably to the 2 year 27% relapse rate in similar patients treated on our initial trial.(Grosso, et al., Blood, 2011, 118:47320). The concept of allowing time for malignancy burden to decline in high risk patients prior to introduction of DLI warrants further evaluation going forward in efforts to reduce relapse after HSCT.

Disclosures

Porcu:Innate Pharma: Consultancy.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution