Long‐term follow‐up of Philadelphia chromosome‐positive (Ph+) chronic myeloid leukaemia (CML) in children and adolescents managed at a single institution over a 20‐year period

Abstract

Chronic myeloid leukaemia (CML) rarely affects children. Allogeneic haematopoietic stem cell transplant (allo-HSCT) is feasible only for a minority of patients. Although clinical research on alpha-interferon (IFN) in CML began two decades ago, the few published series of childhood CML reported cytogenetic response (CyR) rates but no long-term treatment results (Dow et al, 1991; Millot et al, 2002). Recently, imatinib has shown efficacy in Philadelphia chromosome-positive (Ph+) CML patients, also in those previously treated with IFN (O'Brien et al, 2003; Champagne et al, 2004; Kantarjian et al, 2004). The treatment results, updated at December 2004, of 30 Ph+ CML children and adolescents (16 males and 14 females; median age of 12·17 years), diagnosed at our Institution between June 1980 and September 2001, are reported (Table I). Allo-HSCT was performed in patients with a matched related donor (MRD), while those lacking a MRD received different treatments. Before 1989, patients without a MRD were treated with hydroxyurea; during that period, two patients underwent an autologous stem cell transplant (ASCT) and then low-dose IFN. Starting from 1990, 19 patients received IFN at a dosage of 2·5–9 MU/day (median 6 MU/day). When patients did not respond to IFN, a search was started for a human leucocyte antigen (HLA)-matched unrelated donor (MUD) and, from 1995, for umbilical cord blood (UCB) stem cells. Recently, patients who failed IFN were treated with imatinib. A CyR was achieved in 11 of 17 evaluable patients treated with IFN (65%): complete (CCyR) in four and partial in seven; the median time to achieve maximal CyR was 12 months (range: 4–96 months). The CCyR persisted in three of the four complete responders, in whom the BCR-ABL transcript subsequently disappeared. Of the 14 patients who failed IFN treatment, five underwent allo-HSCT, while five were switched to imatinib and obtained a CCyR. The projected 8-year survival of all patients treated with IFN was 63% [95% confidence interval (CI): 39·6–87·3]; censoring patients at the start of imatinib or at the date of allo-HSCT, the projected 8-year survival was 62% (95% CI: 31·6–92·7). Thirteen patients underwent an allo-HSCT, seven of them had previously received IFN and one had also received an ASCT. Four patients that were allografted from a MRD are alive with no evidence of the BCR-ABL transcript, two of them after IFN dose escalation combined with a single donor lymphocyte infusion (DLI) because of disease recurrence. Three MUD allografted patients and one patient submitted to a mismatched related donor transplant are alive with no evidence of the BCR-ABL transcript. The projected 8-year survival from the date of allo-HSCT for all transplanted patients, independent of the type of transplant, disease status, interval from diagnosis to transplant and prior therapy was 61% (95% CI: 33·5–87·7). In our experience, which reflects the therapeutic changes that have occurred over the two last decades, the survival probability of patients treated with high-dose IFN is similar to that of patients submitted to allo-HSCT. The prolonged use of IFN did not impair the outcome of allo-HSCT and induced a CcyR, even after 8 years. Furthermore, it led to a BCR-ABL transcript disappearance in three of four CCy responders to high-dose IFN and in two children who had relapsed after transplant and subsequently treated with IFN combined with DLI. Disappearance of the BCR-ABL transcript after IFN has been recorded in adults, also in those who relapsed after allo-HSCT (Steegmann et al, 1999), but so far not in children. Furthermore, in agreement with reported data (Champagne et al, 2004; Kantarjian et al, 2004) imatinib induced CCyR in our children that had previously been treated with IFN. In conclusion, our results indicate that IFN may still have a role in the future treatment strategies for childhood CML, combined or in sequential treatment with imatinib.