Reduced-intensity allogeneic hematopoietic stem cell transplantation with alemtuzumab conditioning regimens: survival does not plateau until after day 200

Reduced-intensity conditioning has in the last few years revolutionized allogeneic hematopoietic stem cell transplantation (allo-HSCT).1-6 The reduced toxicity has extended its availability to groups of patients hitherto ineligible because of advanced age or comorbidity. Alemtuzumab (CAMPATH-1H) has been used successfully for in vivo T-cell depletion, reducing the incidence of graft-versus-host-disease (GVHD) and preventing graft rejection.7-10 

We have performed 60 reduced-intensity allo-HSCT procedures in myelodysplastic syndromes and acute myeloid leukemia, using conditioning with 30 mg/m² fludarabine administered intraveneously each day, days −9 to −5; 4 mg/kg/d busulphan administered orally days −3 and −2; and alemtuzumab 20 mg daily intravenously days −5 to−1 (Table1). Immunosuppression was composed of intravenously then orally administered cyclosporine from day −1, titrated to plasma trough levels of 150-200 ng/L. In the absence of GVHD, this was weaned off from day 56. At a median follow-up of 204 days (range, 37-1196 days) for volunteer unrelated donors (VUDs) and 253 days (range, 50-1093 days) for sibling recipients, 57% (21 of 37 patients) and 61% (14 of 23 patients) were alive, respectively. The median times to all-cause mortality were 195 days (range, 37-579 days) and 120 days (range, 50-969 days), to relapse 168 days (range, 28-412 days) and 106.5 days (range, 30-288 days), and to transplant-related mortality (TRM) 117 days (range, 52-220 days) and 121 days (range, 94-969 days), respectively. Initial cytomegalovirus (CMV) antigenemia/DNAemia occurred at a median of 30 days (range, 10-96 days) for VUDs and 43 days (range, 30-356 days) for sibling recipients. There was no statistical significance between the 2 groups.

Kaplan-Meier analysis predicts that to day 300, of the 32.8% overall mortality, 43% would occur between day 100 and day 200. Similarly, of the TRM of 24.7% and disease-free survival (DFS) of 46%, 52%, and 40%, respectively, would occur within the second hundred days (Figure1A).

The incidence of GVHD (before donor lymphocyte infusion) in sibling and VUD recipients was 26.1% (6 of 23 patients) and 21.6% (8 of 37 patients), with a median time to onset of 43 days (range, 39-100 days) and 19 days (range, 14-131 days), respectively. There were no grades III-IV GVHD. All sibling recipient-donor pairs were fully human leukocyte antigen (HLA) matched at 6 loci, while 12 of 44 VUD recipient-donor pairs were disparate (10 HLA class I mismatches; 1 class II; 1 class I and II). Four of these patients received a graft with a major HLA mismatch.

Although the median time of onset of GVHD and CMV viremia occurred within the first 100 days, the major events marking overall survival, TRM, and relapse all have a median onset at day 100 in the second 100 days after transplantation, with no plateau in the survival curves.

In conventional myeloablative HSCT, day 100 has traditionally been a temporal landmark, with the plateau in survival curves (Figure 1B). Although the conventional and reduced-intensity HSCT groups are not comparable because of age and other major differences, especially concerning comorbid medical conditions, with reduced-intensity allografts the plateaus appear to occur much later (Figure 1A), accounted for by delayed TRM and relapse (Tables2-3). Although day 100 will remain a useful comparator, is day 200 or day 300 a more appropriate landmark in reduced-intensity allografts? We suggest authors consider reporting these time points in reduced-intensity studies, as we believe that this would then include the highest risk period and reflect more accurately the eventual outcome.