Multi-Center Phase II Study of CAMPATH-1H Dose De-Escalation Prior to Nonmyeloablative HLA-Identical Sibling Transplantation.

Inclusion of CAMPATH-1H as part of a Fludarabine/Melphalan conditioning regimen is effective at preventing GVHD and reducing non-relapse mortality (NRM)following allogeneic stem cell transplantation. However, these benefits are offset by high rates of infection and potentially a loss of graft-versus-tumor effects. When used at a total dose of 100mg, CAMPATH-1H antibody can still be detected at levels in excess of those required to induce ADCC for several weeks. We reasoned that a reduction in the dose of CAMPATH-1H would permit improved immune reconstitution post-transplantation. We report here the analysis of a national, multi-center trial in which the total dose of CAMPATH-1H was reduced step-wise in separate cohorts from 60mg to 20mg prior to HLA-identical sibling transplantation (n=106). Eligibility criteria included patients with haematological malignancies who were aged 18–65, who had a life expectancy >3 months and who were not suitable for standard myeloablative conditioning. Primary endpoints included PK data, chimerism, NRM and incidence of GVHD or infection. The study received IRB approval and all patients gave informed consent. Four total doses of CAMPATH-1H were tested in consecutive cohorts: group 1, 60mg split between d-2 and d-1 (n=26); group 2, 40mg split between d-2 and d-1 (n=27); group 3, 30mg d-1 (n=28); and group 4, 20mg on day -1 (n=25). 97/106 patients recruited to the study are evaluable with a median follow up of 12 months. Median age was 50 (range 19–64). No major differences were identified in patient characteristics between each cohort. 1-year OS and PFS for the whole population was 80.8% and 67.2% respectively. Peak CAMPATH-IH levels (ug/ml) measured by ELISA on day 0 (n=5 each group) were 7.7 ±1.1 in group 1, 4.3 ±0.7 in group 2, 4.9 ±0.8 in group 3 and 2.7± 0.7 in group 4 (p<0.05 groups 1 vs. each group 2–4). By day 28, CAMPATH-1H levels had fallen substantially in all groups, but especially groups 3/4: 1.1 ±0.4 in group1, 0.6 ±0.1 in group 2, 0.1 ±0.06 in group 3 and 0.1 ±0.06 in group 4 (p<0.05 group 1 vs. each group 3 and 4). In groups 3 and 4, 40% of patients had undetectable CAMPATH-1H levels by day 28. Chimerism data was available in 78 patients and of these, 1 patient showed autologous reconstitution, 52 were mixed chimeras and 25 were full chimeras, with no differences between the groups. Day 100 NRM was 4% in group 1, 8% in group 2, 0% in group 3 and 12% in group 4. Grade III-IV GVHD was 0% in group 1, 4% in group 2, 0% in group 3 and 11% in group 4 (p=0.09 group 1 vs. 4). It is noteworthy that 2 patients in group 4 died of complications secondary to grade IV GVHD, although no patients died of this complication in any of the other cohorts. There were no significant differences in the rates of initial CMV reactivation between the groups, or in the rates of CD4 reconstitution. Cumulative incidences of chronic GVHD at 1-year were 55% and 30% in groups 1 and 2, although further follow up is required in later cohorts. We conclude that significant de-escalation of the CAMPATH-1H dose prior to HLA-identical sibling transplantation is feasible without increasing NRM, although reductions below 30mg are associated with a clinically significant risk of severe acute GVHD. Further studies are warranted to determine whether reductions in CAMPATH-1H dosage will translate into improvements in progression-free survival.