Prophylactic administration of imatinib after hematopoietic cell transplantation for high-risk Philadelphia chromosome–positive leukemia

Imatinib has emerged as standard therapy for chronic myeloid leukemia (CML) and has been incorporated into induction and consolidation regimens for Philadelphia chromosome–positive (Ph⁺) acute lymphoblastic leukemia (ALL).^1-3  Since treatment with imatinib cannot cure high-risk Ph⁺ leukemia, most patients with allogeneic donors are offered hematopoietic cell transplantation (HCT). Long-term survival rates after allogeneic HCT are approximately 10% for CML blast crisis, 25% to 40% for accelerated phase or blast crisis in remission, and 21% to 65% for Ph⁺ ALL. Relapse is the most frequent cause of failure.^4-6  Imatinib has been used successfully to treat recurrent CML after HCT, but effects on recurrent Ph⁺ ALL have been limited.^7-9 

Previous studies have shown that minimal residual disease (MRD) is frequently detected in patients at initial engraftment after HCT for Ph⁺ ALL and CML and is associated with increased risk of relapse.^10,11  We postulated that administration of imatinib early after HCT might be an effective approach for preventing recurrent Ph⁺ leukemia. We report the results of a prospective study to determine the safety of imatinib administration early after HCT.

Allogeneic HCT recipients with Ph⁺ ALL or CML beyond first chronic phase treated with myeloablative conditioning were eligible for this study. Leukemia cells had to express p190 or p210 BCR-ABL transcripts. Patients known to be resistant to imatinib before transplantation were excluded from the study. Patients were also excluded if the absolute neutrophil count (ANC) remained below 1.2 ×10⁹/L before treatment with imatinib despite the use of G-CSF, or if leukemia was detected within 4 days after neutrophil engraftment as defined by presence of any blasts in blood or spinal fluid, more than 5% marrow blasts, 1% or higher myeloblasts with aberrant antigen expression in the marrow, any Ph⁺ marrow metaphases, or more than 5% BCR-ABL⁺ marrow interphase nuclei by fluorescence in situ hybridization (FISH).

Institutional review boards of Fred Hutchinson Cancer Research Center, Clinical Research Division, and City of Hope Medical Center approved the study. Informed consent was provided according to the Declaration of Helsinki. Daily dosing of imatinib began at 400 mg (adults) or 260 mg/m² (children) as soon as all eligibility criteria were satisfied, and continued until day 365. Imatinib dose reductions were planned for ANC less than 1.2 × 10⁹/L despite administration of G-CSF, platelet counts less than 10 × 10⁹/L before day 90 or less than 50 × 10⁹/L after day 90 days, serum ALT or AST levels more than 6 times the upper limit of normal, or conjugated bilirubin more than 3 times the upper limit of normal.

The primary end point addressed the safety of imatinib during the first 90 days after HCT. Tolerability was defined as a dose of at least 200 mg/d for adults (100 mg for children younger than 17 years) for an average of at least 6 d/wk until day 90. Anecdotal experience suggested that doses above 100 mg/d were poorly tolerated early after HCT.^12,13  The primary goal of our study was to demonstrate feasibility. Success was predefined as tolerability of imatinib among more than 50% of participants. Secondary end points addressed survival and detection of BCR-ABL transcripts.^11,14 

Twenty-seven patients consented to participate in the study. Two patients died and one had recurrent malignancy before imatinib could be administered, and 2 could not take imatinib because of neutropenia or critical illness. Twenty-two participants were treated with imatinib (Table 1).

During the first 90 days, 17 of 19 adults tolerated imatinib at approximately 400 mg/d, and the 3 children tolerated imatinib at approximately 260 mg/m²/d (Table 2). Two adults did not tolerate imatinib therapy according to the primary end point because of transaminase elevations. One averaged 286 mg/d for 5.0 d/wk because imatinib was held between days 33 and 46. The other averaged 220 mg/d for 5.0 d/wk but permanently discontinued administration of imatinib after day 71. After day 90, imatinib was generally deliverable at full dose to all patients, and 18 subjects have completed therapy for 1 year.

In previous studies, imatinib has been initiated at a median of 4 to 30 months after HCT for treatment of relapse.^8,9,15  Our prospective study provides the first detailed information regarding tolerability of imatinib during the first 90 days after myeloablative allogeneic HCT when the burden of leukemia cells is at a nadir. Anderlini et al¹⁶  reported their anecdotal experience among 15 patients who received a myeloablative preparative regimen for Ph⁺ leukemia. Doses above 200 mg/d were frequently associated with grade 3 or 4 cytopenias. In our patients, 2-fold higher doses of imatinib did not cause significant cytopenias, despite the presence of risk factors (Table 1).

Fifty-three percent of nonserious adverse events had a possible or probable relationship to imatinib therapy (Table 3). Self-reported medication compliance for the 22 patients indicated no dose reductions or omissions (n = 9), skipped doses (n = 4), doses omitted for cytopenias (n = 1) or for nausea, vomiting, or diarrhea (n = 5), and doses reduced or held for transaminase elevations with or without intestinal symptoms (n = 7). When emesis and nausea were not associated with graft-versus-host disease (GVHD) or infection, symptoms were generally manageable by divided dosing or by antiemetic administration. Although imatinib may have contributed to transaminase elevations, other temporal associations were noted, including GVHD, infection, triazole antifungal medication, and intrathecal methotrexate in a pediatric patient. CYP3A4-inhibiting effects of triazole medications could have increased imatinib plasma concentrations, thereby increasing the frequency of transaminase elevations.

Administration of imatinib did not affect calcineurin inhibitor levels. Among patients taking cyclosporine (n = 11), the median cyclosporine level within 2 days before administration of imatinib was 371 ng/mL (range, 253-484 ng/mL), and the median change at 3 to 6 days after starting treatment with imatinib was −3% (range, −36% to +48%). Among patients taking tacrolimus (n = 8), the median tacrolimus level was 10.4 ng/mL before administration of imatinib (range, 6.6-21.1 ng/mL), and the median change was 0% (range, −37% to +146%).

Six of the 22 patients had Ph⁺ metaphases and 14 had BCR-ABL transcripts detected before HCT (Table 4). The median follow-up is 1.4 years (range, 0.74-2.7 years). Seventeen patients are alive without detectable BCR-ABL; 15 have been followed for a median of 0.45 years (range, 0-1.6 years) beyond the scheduled discontinuation of imatinib administration at 1 year after HCT. Two patients are scheduled to discontinue imatinib within 3 months. Four had hematologic relapse between 0.3 and 2.0 years after HCT (2 CML, 2 ALL), and 3 have died. Relapses occurred after 89 and 180 days of imatinib therapy, respectively, for 2 patients with CML in CP3 and BC remission. One patient with recurrent ALL received imatinib for only 44 days. Another patient died unexpectedly with acute respiratory distress syndrome (ARDS) at day 368 and BCR-ABL transcripts were last documented at less than 10 copies/μg RNA on day 297.

Given the high risk for hematologic relapse in this cohort, this preliminary evidence of efficacy is encouraging. Our primary assumption was that imatinib would have a direct effect on MRD after HCT. Imatinib might prevent relapse after HCT through immunologic effects. Recent reports have described a new type of immune cell in mice, the interferon-producing killer dendritic cell, which can be activated by imatinib.¹⁷  Experimental results have indicated that these cells mediate antitumor effects in vivo, both by killing target cells and by subsequently presenting antigens from target cells. These cells might provide a pivotal link between innate and adaptive T-cell immunity.^18-20 

We conclude that engrafted recipients can tolerate prophylactic administration of imatinib at standard dose intensities after allogeneic HCT. Our preliminary results encourage a larger efficacy study, particularly in patients with Ph⁺ ALL. Future studies should also investigate how imatinib pharmacokinetics might be affected by the large number of medications typically administered after HCT.

Contributions: P.A.C. was principal investigator, designed and wrote the study, and wrote the manuscript; D.S.S. was a collaborating site investigator with associated responsibilities; M.E.F. participated in study design discussions and treated patients at FHCRC; J.E.S. was part of the study design discussions and treated pediatric patients at FHCRC; T.G. focused on statistical end points in the study design; and P.J.M., F.R.A., and J.P.R. participated in the study design discussion and critical analysis of the results and manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Paul A. Carpenter, Fred Hutchinson Cancer Research Center, Clinical Research Division, 1100 Fairview Ave N, Seattle, WA 98109; e-mail: pcarpent@fhcrc.org.

This work was supported by Adult Leukemia Center (ALC) grant CA18029 from the National Institutes of Health, Bethesda, MD.

The authors are grateful to Lan Beppu and Era Pogosova-Agadjanyan for excellent technical assistance and to Jane Jocom and Penka Ilieva, who contributed to study implementation and careful data collection.