Preliminary Data from a Multi-Center Study of Clofarabine / Busulfan Conditioning for Acute Myeloid Leukemia Not in Remission at Allogeneic Transplantation

Acute Myeloid Leukemia (AML) not in remission at time of HSCT carries a very poor prognosis.  We previously reported a single institution experience indicating significant activity for myeloablative clofarabine / busulfan conditioning (CloBu4) in refractory hematologic malignancies, particularly in AML (Blood. 2011 Oct 13; 118(15)). In an effort to better establish efficacy of HSCT with CloBu4 for AML not in remission we initiated a multi-center phase II study with a target accrual of 75 pts. Only pts with active AML defined as >5% myeloblasts on bone marrow examination within 14 days of registration were eligible. AML was required to be refractory to two lines of intensive induction, one line of induction following relapse (CR > 6 month), or untreated relapse (CR < 6 months). Pts receiving previous HSCT were excluded. Busulfan was administered as a single daily dose of 3.2 mg/kg IV x 4d (days -5 to -2) targeting a plasma AUC of 4500-5500 µM.min/day and clofarabine as a single daily dose of 40 mg/m2 IV x 5d (days -6 to -2). An 8/8 HLA matched related or unrelated donor was required. GVHD prophylaxis was per institutional preference, but included calcineurin inhibitor together with either mycophenolate mofetil or methotrexate. Twelve centers across US and Canada are participating. At time of interim analysis there were 60 pts with 56 evaluable for toxicity and response with additional pt accrual expected at time of abstract submission. One pt enrolled but withdrew before conditioning. Pts had median age of 56 yrs (range: 3-65 yrs), received a median of 2 inductions and had a median of 26% bone marrow blasts (range: 5-94%) within 14 days of enrollment. Matched unrelated donors were utilized in 46% of pts. Pts engrafted at a median of 13 days (range: 9-25) for neutrophils and 16 days (range: 8-72) for platelets. The most frequent grade 3-4 non-hematologic toxicities reported by day + 100 as related to conditioning were infections (N=6, 11%) and aGVHD (N=4, 7%). Grade 5 non-hematologic toxicities possibly related to conditioning include neutropenic enterocolitis (N=1), aGVHD (N=1), pneumonia (N=1), alveolar hemorrhage (N=1), ascites/veno-occlusive disease (N=1), and metabolic acidosis / sepsis (N=1).  Morphologic remission was observed 92% of pts with response assessment at day + 30 (N=53).  The cumulative incidence of aGVHD grades II-IV by day + 100, accounting for relapse and death as competing risks, was 37% [95% CI: 24, 50]. Non-relapse mortality and relapse at one year was 25% [95% CI: 11%, 38%)] and 43% [95% CI: 27, 58), respectively. The estimated one year OS and DFS are 39% [95% CI: 24%, 64%] and 32% [95% CI: 20%, 52%)], respectively. In summary, preliminary analysis of a multi-center study of CloBu4 conditioning in AML not in remission exhibits impressive early responses with NRM and relapse incidence similar to previous trials utilizing this regimen.