GVHD-free with Campath?

Patients with aplastic anemia (AA) derive no benefit from GVHD. In this issue of Blood, Marsh and colleagues report that conditioning with an alemtuzumab-containing regimen almost completely prevented GVHD in AA patients who received a transplant of either bone marrow (BM) or peripheral blood progenitor cells (PBPCs) from HLA-matched siblings or unrelated donors.¹ 

Immunosuppressive therapy (IST) and hematopoietic cell transplantation (HCT) are 2 treatment options for patients with AA, and the debate about the most appropriate use of both modalities continues, particularly when counseling patients without HLA-identical sibling donors. While HCT offers a high probability of cure, those critical of HCT point to the risks of graft rejection and GVHD. Rejection rates have been reduced in recent years to about 10%. However, acute and chronic GVHD have remained major challenges and occur in 30% to 40% of patients.² 

Currently the policy at many centers is this: If HLA-identical siblings are available, patients < 40 or 50 years old proceed to HCT as first-line therapy. Older patients and those without HLA-matched siblings will typically be offered IST and proceed to HCT from unrelated, ideally HLA-matched donors only if they do not respond to IST or AA recurs after an initial response. In the study by Marsh et al, 33 of the 50 patients (including all 29 unrelated transplantation recipients) had previously received at least 1 course of IST.

Over the past decade, several transplant teams and cooperative groups have reported considerable improvement in HCT outcome in patients with AA who received transplants from either related or unrelated donors.^3-5  Those results were achieved, for example, with conditioning regimens combining anti-thymocyte globulin (ATG; 2 or 3 doses) with cyclophosphamide (CY), 50 mg/kg/day for 4 doses ± total body irradiation (TBI), 2-6 Gy,⁶  or with CY, 30 mg/kg/day × 4, plus fludarabine (Flu), 30 mg/m²/day × 4 ± ATG, ± TBI 2 Gy.⁷  Particularly in younger patients, these strategies have resulted in survival rates of ∼ 90% with HLA-identical siblings, and 70% with unrelated donors. However, as stated, GVHD has remained a problem, and there are concerns about late effects of TBI.

Earlier studies suggested that campath anti-CD52 antibodies as part of the conditioning regimen were effective in improving HCT outcome in patients with AA.⁸  Marsh et al now show that conditioning of patients 8 to 62 years of age, with CY at doses of 300 mg/m²/day (∼ 7-8 mg/kg/day) × 4, Flu 30 mg/m² × 4, and alemtuzumab (the humanized anti-CD52 antibody Campath 1-H), given for total doses of 40-100 mg over 4 days, was associated with sustained engraftment in 19 of 21 patients (90%) who received transplants from matched siblings, and 25 of 29 patients (85%) who received transplants from unrelated donors. While this rate of engraftment was comparable with that in several other studies, it was achieved with very low doses of CY in combination with Flu and alemtuzumab, and without the use of TBI. More importantly, the rate of acute GVHD was 13.5% (only grades I-II), and of chronic GVHD 4%, even though in 19 patients the source of stem cells were PBPCs (combined with BM in 7). Two-year overall survival was 95% with HLA-identical siblings and 83% with unrelated donors. Only 1 of the 45 surviving patients appears to have evidence of chronic GVHD. Presumably, the low incidence of GVHD was related to in vivo T-cell depletion with alemtuzumab and the establishment of mixed donor/host chimerism in the CD3 compartment. While alemtuzumab also targets B cells, 2 patients (4%) developed posttransplantation lymphoproliferative disorders, which proved fatal in 1. Factors significantly associated with inferior outcome included older patient age, longer interval from diagnosis to HCT, use of PBPCs as a source of stem cells, and the presence of comorbidities as determined by the HCT-CI,⁹  arguably the first such report in patients with AA.

So, what are the messages? (1) These data confirm that excellent HCT results are achievable in patients with AA not only with related but also with unrelated donors. (2) HCT from an unrelated donor is successful even without the use of TBI. (3) Delay of HCT leads to inferior results, possibly because patients present with more comorbidities. (4) The rate of graft failure was not reduced with PBPCs compared with BM. (5) The use of PBPCs did not increase the incidence of chronic GVHD, but survival was inferior to that obtained with BM. And (6) establishing mixed CD3 chimerism by design may be an approach to prevent GVHD.

The data obtained with this CY/Flu/alemtuzumab regimen are encouraging, even more so because they combine results from 5 institutions. However, this was not a prospective trial, patients received different doses of alemtuzumab, and accrual occurred over a 20-year interval during which supportive care in HCT has changed dramatically. The data do not endorse the use of PBPCs in patients with AA, and we should not all adopt the present regimen without further study. Patient numbers are small (as in most AA trials), and the median follow-up is only 18 months. However, the results should stimulate new discussions among all involved in the design of AA trials, particularly for unrelated donor transplants, to determine whether transplantation success and post-HCT quality of life can be improved further by reducing CY doses in conditioning regimens and by substituting alemtuzumab for TBI. One might then consider carrying out HCT as first-line therapy even from unrelated donors, particularly in young patients and those who are less likely to respond to IST.¹⁰