Comparative Analysis of Ablative and Reduced Intensity Conditioning Hematopoietic Stem Cell Transplantation in Treatment of Adult Patients with Chronic Myeloid Leukemia.

Allogeneic HSCT remains the only curative therapy for chronic myelogenous leukemia (CML) patients. We compared results of myeloablative (MAB) and reduced intensity conditioning (RIC) HSCT from sibling donor in 62 pts with CML, treated between 2000 and 2005. In MAB group, 44 pts (median age 39.5, range 17–54 yrs) were conditioned with BuCy2 regimen. Thirty four pts were in chronic, 7 in acceleration and 3 in 2nd chronic phase. 26 pts received PBSC and 18 were grafted with bone marrow HSC. Transplanted material contained 3.8 (1.24–14.6) x 10⁶/kg CD34+ cells. Prophylaxis of graft versus host disease (GVHD) consisted of cyclosporin (CsA) and methotrexate. Eighteen pts (median age 48, range 37–56 yrs) were treated with fludarabine-based RIC regimen. This group comprised 13 pts in chronic and 5 with accelerated phase. Patients were grafted with PBSC only, containing 4.75 (2.25–10) x 10⁶/kg CD34+. CsA alone was given as GVHD prophylaxis. RIC group differ from MAB group by 2 factors: higher age (p<0.001) and higher EBMT pre-transplant (Gratwohl) risk score (p=0.028).

Results: Hematopoietic recovery occurred in all pts in MAB group, and in 94% in RIC group. Probability of overall survival for all pts was OS=0.57±0.09. 31 pts developed acute GVHD ≥2 grade (21 in MAB and 10 in RIC) and 25 pts extensive chronic GVHD (18 in MAB and 7 in RIC). Mean survival, estimated 5-yrs OS, probabilities of GVHD incidence, relapse and treatment related mortality (TRM), estimated by Kaplan-Meier method, with respect to transplantation method, are shown in Table.

Factors predicting negative outcome by Cox univariate analysis for all pts, were: AGVHD≥2 grade (p=0.0001, HR=2.78 95%CI=1.49–5.26), extensive CGVHD (p=0.0253, HR=3.83, 95%CI=1.01–14.6), female donor (p=0.0551, HR=1.54, 95%CI=0.99–2.40), pre-transplant risk score >2 (p=0.0053, HR=1.89, 95%CI=1.22–2.94) and RIC (p=0.0027, HR=1.96, 95%CI=1.26–3.03). Factors predicting negative outcome by multivariate analysis were: AGVHD≥2 (p=0.0001, HR=3.03, 95%CI=1.56–5.39), extensive CGVHD (p=0.0370, HR=2.75, 95%CI=1.17–6.49), pre-transplant risk score >2 (p=0.0507, HR=1.59, 95%CI=0.99–2.53) and RIC (p=0.0342, HR=1.64, 95%CI=1.17–6.49). Age >45 yrs did not predict for OS. In matched-pair analysis, identified by pre-transplant risk score and age, RIC group had lower OS (0.24±0.13 vs 0.88±0.08 p=0.0010), higher TRM (0.61±0.18 vs 0.06±0.05, p=0.0064), with no difference in AGVHD, CGVHD and relapse incidence, when compared to MAB group. In summary, this experience with sibling donor allogeneic HSCT in CML, demonstrates a similar incidence of AGVHD≥2 and extensive CGVHD in RIC and MAB groups, and more importantly, results of RIC-HSCT in adult CML are not satisfactory, mainly because of high treatment related mortality.