Abstract

While RIC are presently commonly used, most of the reports suffer from insufficiencies limiting knowledge acquisition: small populations, short follow-up, heterogeneities in the RIC intensity or the donor source. We report here 100 patients with hematological malignancies (HM) treated with geno-identical SCT after the same RIC in a single center from 2000 until 2006. All patients received oral busulfan (8mg/m2), thymoglobulin (2.5 mg/kg) and Fludarabine (150 to 180 mg/m²) (FBT conditioning). All grafts were PBSC from a match sibling donor. All grafts were monitored for CD34, CD3, CD4, CD8, CD19 and CD56+ cells (notably CD34= 5.6 (1.5–22) × 10e6/kg; CD3= 317 (112–887) × 10e6/kg). Median age was 50 (18–64). Hematopoietic cell transplantation comorbidity index (HCT-CI) was 0, 1–2 and > 2 in 31, 39 and 23 of the 93 evaluable patients. All patients received post-graft CSA. Diagnoses included acute leukemia (39%), Myeloid (16%) or Lymphoid (45%) malignancies. 53, 14 and 33 pts were in CR, progression or stable disease. Minimal and median follow-up are respectively 6 and 34 months. All but one engrafted reaching full lymphoid donor chimerism prior to day 100 in 85% of the cases. 55 pts presented aGVHD (G1: 12; G2: 22; G3: 12; G4: 9) for a cumulative incidence (CI) of G2–4 aGVHD of 43% (33–53); 91 patients were evaluable for cGVHD with a 79% (71–87) CI (Lim= 20%; Ext: 80%). In a multivariate analysis cGVHD occurrence could be predicted by 2 independent factors: lower dose of CD34 (but no impact of CD3, CD4, CD8, CD19 or CD56) (odd ratio (OR): 0.79 (0.69–0.90)) and grade 2–4 aGVHD (OR: 1.16 (1.01–1.32)). TRM CI was respectively 5%, 14% and 19% (11–27) at 3, 12 months and overall. TRM was strongly associated with aGVHD occurrence (TRM CI: grade 2–4 aGVHD: 37% (23–51); grade 0–1 aGVHD: 7% (0–14): p<.01). Relapse CI was 15% (8–22) at a median of 169 days (30–769). Disease control was significatively associated with cGVHD occurrence (Relapse CI: cGVHD: 10% (3–17); no cGVHD: 42% (14–70): p=.01). 5 year overall survival (OS) and LFS probability estimates are 62% (50–72) and 60% (48–72) with a plateau starting after 3.5 years. Outcome was similar for the patients above or under 50 and for the different diagnoses. HCT-CI did not show any influence on outcome. In a Cox model analysis, LFS was independently affected by only 2 pre-transplant variables: disease status (CR vs. no CR) (OR=0.45; p=.022) and the dose of infused CD34+ cells (> or < 5.6 × 10e6 CD34+ cells) (OR: 2.04; p=.039). In accordance with above, patients with higher CD34+ cells presented less cGVHD (cGVHD CI: CD34 < 5.6: 98% (94–100); CD34 > 5.6: 76% (64–88) ; p<0.01). These data confirm that FBT RIC, combining myeloablation (Busulfan) and limited Thymoglobulin, is efficient in a wide population in term of age and diagnosis conducting to sustained long term OS and EFS with limited TRM. They suggest also that after RIC, graft composition and function probably in relation with G-CSF mobilization might impact allogeneic effect and invite revisiting this subject in this context.

Author notes

Disclosure:Consultancy: Amgen, Pierre Fabre, Gemzyme. Research Funding: Amgen, Pierre Fabre, Gemzyme.