We conducted a phase I trial to determine the maximum tolerated dose (MTD) of clofarabine with high-dose busulfan followed by allogeneic stem cell transplantation (SCT) in patients with high-risk and refractory acute leukemia. Patients received intravenous busulfan 0.8 mg/kg every 6 h on days −6 to −3 and clofarabine 30–60 mg/m 2 per day on days −6 to −2. Graft-versus-host disease prophylaxis included sirolimus plus tacrolimus (days −2 to +180). A total of 15 patients, median age 48 (30–58) years, with acute leukemia that was relapsed and refractory ( n =8), primary refractory ( n =6), or in CR2 ( n =1), were treated at four clofarabine dose levels: 30 ( n =3), 40 ( n =3), 50 ( n =3) and 60 mg/m 2 per day ( n =6) with busulfan. All engrafted, and the MTD was not reached. Grades 3–4 non-hematological toxicities included vomiting ( n =3), mucositis ( n =9), hand-foot syndrome ( n =1), acute renal failure ( n =1) and reversible elevation of aspartate aminotransferase/alanine aminotransferase ( n =10). The 1-year event-free survival was 53% (95% confidence interval: 33–86%), and the 1-year overall survival was 60% (95% confidence interval: 40–91%). Given the good tolerability and promising results, we recommend clofarabine 60 mg/m 2 per day × 5 days as a phase II dose in combination with busulfan (12.8 mg per kg total dose) for further study as a myeloablative regimen for allogeneic SCT for high-risk acute leukemia.

Viral infections are a significant cause of morbidity after allogeneic stem cell transplantation. Modifying intensity of immunosuppression according to immune function may reduce infections. Currently, no routine functional assays to monitor immune function of the new graft exist. The Cylex ImmuKnow assay is a FDA approved test that quantifies cellular immune function by measuring intracellular ATP activity released by PHA-stimulated CD4 T cells in blood, and has correlated with solid organ graft outcomes. We correlated ATP activity with viral infections in the first 100 days post-transplant. 341 blood samples were collected q1-3 weeks from 32 patients of median age 42 (21-59) years undergoing related (14) or unrelated (18) donor transplants. All received myeloablative doses of chemotherapy±TBI, with ATG in 14 patients. GvHD prophylaxis included FK506 with sirolimus (28) or methotrexate (4). CMV and EBV load was monitored in plasma by PCR q1-2 weeks and preemptive therapy started on reactivation with other viral testing as clinically indicated. 14 patients developed infections while on GvHD prophylaxis, including CMV (10), EBV (3), BK hemorrhagic cystitis (8), adenovirus gastritis (1), and HSV pneumonitis (1). Baseline median ATP activity for all patients was 280 (5-809) ng/ml, compared to 491 (238-842) ng/ml for healthy volunteers (P<0.0001). ATP activity only modestly correlated with lymphocyte (R2=0.21) and WBC (R2=0.45) counts. Median ATP activities were lower for patients who developed infections between days +31 and +60 (31 vs 106 ng/ml, P=0.0007), and days +61 and +100 (34 vs 139 ng/ml, P=0.0006). On logistic regression, low ATP activity was predictive of infections (P<0.0001), but lymphocyte and WBC counts were not. Using 90 ng/ml as cutoff, 11 of 20 patients with ATP activity <100 ng/ml developed infections compared to 1 of 12 with activity ≥100 ng/ml between days +31 and +60 (P=0.01). Between days +60 and +100, 9 of 12 patients with levels <90 ng/ml developed infections, compared to 1 of 14 patients with levels ≥90 ng/ml (P=0.001). Median ATP activity was 251 (97-494) ng/ml at the onset of grade 2-4 acute GvHD that occurred in 4 patients. Our results indicate that low ATP activity in CD4 T cells is strongly predictive of viral infections in the first 100 days after transplant, and suggest that modifying the intensity of immunosuppression to maintain ATP activity around 90 ng/ml should be investigated as a strategy to reduce infections.