Effective use of imatinib-mesylate in the treatment of relapsed chronic myeloid leukemia after allogeneic transplantation

Despite recent advances in the treatment of chronic myeloid leukemia (CML), allogeneic stem-cell transplantation (SCT) remains the only curative option. The success of SCT is limited because of relapse in 20–30% of patients.1,2 As the graft-versus-leukemia (GvL) effect contributes to cure, the cessation of immunosuppression3 and use of donor lymphocyte infusions (DLI)4 have become established treatment for relapse. DLI is most effective if given for molecular or cytogenetic relapse. However, GvHD is increased if used in the first year post-transplant. Other options include interferon α(IFN) or a second transplant.

Imatinib mesylate (IM) is a potent inhibitor of the BCR-ABL tyrosine kinase and achieves complete cytogenetic responses (CCR) in 87% of previously untreated patients.5 More recently its role in relapse after SCT has been highlighted.6,7 We describe the use of imatinib in 14 patients who relapsed post-SCT and were subsequently treated with IM.

All patients with Ph-positive CML in first chronic phase (CP) who relapsed (n=14) following an allogeneic transplant performed between 1987 and 2004 are included. There were 7 males and 7 females. The median age at diagnosis was 31 years (range 15–48) (Table 1). Pre-transplant treatment included either hydroxycarbamide (OHU) and/or IFN. All patients were IM naïve at the time of transplant and were transplanted in CP, 13 using a matched sibling donor and one a matched unrelated donor.

The conditioning regimen was Bu/Cy (9 patients) or Cy/TBI (5 patients). All the transplants were T-cell replete and cyclosporine and methotrexate were used as GvHD prophylaxis. Four patients developed acute GvHD limited to the skin and were treated with corticosteroids. Follow-up included clinical evaluation, blood counts, bone marrow examination including morphology and cytogenetics. From 2002, patients were monitored by qualitative, nested BCR-ABL RT-PCR and if positive, had BCR-ABL transcript levels determined by real-time quantitative PCR (RQ-PCR). Patients were deemed to have had a molecular relapse if greater than a five-fold increase in BCR-ABL transcript levels was observed.

Median time to first relapse was 36 months (range 7–180) (Table 2). Prior to the availability of IM, 4 patients received DLI at incremental doses with only one patient showing any durable response. The other 3 patients proceeded to a reduced intensity-conditioning transplant with short responses before relapsing (Patients 2, 7 and 9 in Table 2). At the time of introduction of IM, 10 patients were in their first relapse and 4 patients were in second relapse. Four patients had a hematologic [3 CP and one accelerated phase (AP)] relapse, 4 had a cytogenetic relapse and the remaining 6 had a molecular relapse. Imatinib was started at a dose of 400 mg daily in all patients except the patient with AP disease who received 600 mg daily.

Table 1.Pre-transplant.

Thirteen (93%) patients responded to IM with a median time to response of four months (range 3–15). Of the 4 patients treated in hematologic relapse, 2 achieved a CCR and became nested PCR negative (<1 BCR-ABL transcripts in 105). The other 2 patients had transient responses before developing progressive disease. Of the 10 patients who were treated for cytogenetic or molecular relapses, all achieved a CCR and 9/10 became nested BCR-ABL PCR negative.

When these patients were started on IM, there was no data to indicate whether the molecular remissions achieved would be durable or whether these patients should be maintained indefinitely on therapy. Imatinib was stopped in 7 of the surviving 12 patients. The median duration of treatment for patients who stopped IM was 11 months (range 6–35). No patient stopped the drug because of toxicity. Only 2 of the 7 patients who stopped IM have remained in molecular remission with a median follow-up of 42 months. The other 5 patients all had re-emergence of BCR-ABL transcripts. One patient has stable low levels of transcripts and has not received any further treatment. Four patients were restarted on IM, one has again become PCR negative. The remaining 3 patients received DLI in combination with IM, one remains in molecular remission and 2 have low level stable BCR-ABL transcripts and remain in CCR. Five patients were continued on IM: 4 of these patients remain disease free while one has low level stable BCR-ABL transcripts.

DLI is an effective treatment for patients relapsing after SCT for CML and can restore durable molecular remissions in a high percentage of patients.4,8,9 However, a significant proportion of patients are unresponsive. Toxicity from GvHD remains a concern although it can be reduced by using incremental dose regimens.10 Additionally DLI may not be an option because of unavailability of the original donor, (2/14 patients in our series), or due to GvHD. This report demonstrates the effectiveness of IM in re-establishing leukemia control without serious toxicity and supports the published data.6,7 However the response in most patients was not durable unless IM was continued, as has been previously shown for patients treated with IM as a primary treatment.11 Interestingly, the 2 patients who remain in continued molecular remission had both received IFN therapy (patients 4 and 11, Table 2) pre-transplant. This is analogous to reports of occasional patients treated with IFN and subsequent IM who maintained durable molecular remissions when IM is discontinued.12

Table 2.Post transplant.

What is the best treatment strategy for a patient who relapses post-BMT from the pre-IM era¿ We have shown that IM achieves excellent responses without toxicity although it usually needs to be continued for durable responses. DLI has proven to be effective for early relapses albeit with serious adverse effects in some patients. There may be a role for IM in the induction of responses, with a possibility to consolidate with lower doses of DLI. In the post-IM era, most patients who relapse post-transplant will have already been exposed to IM and deemed unresponsive. Determining the mechanism of resistance in these patients is important and mutational analysis of the BCR/ABL oncogene should be considered. The role of newer tyrosine kinase inhibitors such as nilotinib and dasatinib in these patients is unknown; EBMT is exploring this with the use of dasatinib in a recently opened trial.