We recently reported the Spanish Paediatric Haematology Oncology Group (SHOP) experience in treating children and adolescents with Philadelphia-chromosome positive acute lymphoblastic leukaemia (Ph+ ALL) with concurrent imatinib and chemotherapy before stem cell transplantation, and showed an impressive improvement in the early outcome of these children when treated with this approach (Rives et al, 2011). Previously, the Children’s Oncology Group (COG) reported an excellent outcome in Ph+ ALL paediatric patients also treated with imatinib in combination with chemotherapy. Their patients underwent haematopoietic stem cell transplant (HSCT) only when a matched sibling donor was available (Schultz et al, 2009). The early outcome of patients undergoing HSCT was not superior to those that had received only chemotherapy. The major improvement achieved with tyrosine-kinase inhibitors (TKI) in combination with chemotherapy in paediatric Ph+ ALL challenges the need for HSCT in all children with this subgroup of leukaemia. Therefore, it is necessary to identify a good-risk group of patients for whom allogeneic HSCT could be omitted. We retrospectively analysed our pre-imatinib cohort of patients with the aim of identifying prognostic factors that might help to select those patients who could be spared HSCT.
Patients with Ph+ALL treated with pre-imatinib SHOP protocols (SHOP/ALL-94 and SHOP/ALL-99) were analysed. From February 1994 to March 2005, 27 patients were treated with intensive chemotherapy and HSCT from a matched related or unrelated donor (Rives et al, 2011). The disease outcome in our patients was heterogeneous, depending on clinical features at diagnosis and early response to treatment, in line with other series of children with Ph+ALL (Ribeiro et al, 1997; Schrappe et al, 1998; Aricóet al, 2000, 2010; Roy et al, 2005; Gandemer et al, 2009).
Recently, the French Acute Lymphoblastic Leukaemia Study Group (FRALLE) analysed prognostic factors in their pre-imatinib series of paediatric patients with Ph+ ALL and described a prognostic index that could discriminate a subgroup of patients with a better prognosis. This prognostic index included two presenting features (age < 10 years and white blood cell [WBC] count <100 × 109/l) and early response (less than 5% of blasts in bone marrow on day 21 of induction treatment) for good risk. One third of their patients belonged to the good-risk group and had a 5-year event-free survival (EFS) of 79% (Gandemer et al, 2009). In order to test this prognostic index in an independent cohort of patients we analysed the impact of the FRALLE index in our series of paediatric Ph+ ALL patients. With this index it was possible to identify a subgroup of patients with better outcome. Indeed, the analysis of prognostic factors revealed that younger children (1–9 years) with WBC count <100 × 109/l count and a good early response, as per bone marrow blast count on day-15, had a significantly superior outcome (P = 0·018) with around 70% of long-term survivors (Table I, Fig. 1). Our series validated, in an independent cohort of paediatric Ph+ ALL, a strong prognostic factor that can be easily estimated early in the course of the disease and enables a good risk group to be defined among Ph+ALL patients. In this regard, minimal residual disease measurement could also help to refine the identification of good-risk patients who might be considered for treatment de-escalation, e.g. omit allogeneic HSCT in first complete remission.
|All patients||n||5-year EFS||5-year OS|
|27||30 (±9)||P||31 (±9)||P|
|Male||13||31 (±13)||33 (±14)|
|Female||14||29 (±12)||0·8||29 (±12)||0·44|
|<10 years||18||33 (±11)||33 (±11)|
|≥10 years||9||22 (±14)||0·22||25 (±15)||0·56|
|Leucocyte count (×109/l)|
|<25||11||55 (±15)||55 (±15)|
|≥25||16||13 (±8)||0·055||13 (±9)||0·16|
|<50||15||40 (±13)||40 (±13)|
|≥50||12||17 (±11)||0·33||18 (±12)||0·7|
|<100||20||40 (±11)||40 (±11)|
|Blasts on day 15 bone marrow|
|<5%||14||43 (±13)||43 (±13)|
|≥5%||13||15 (±10)||0·09||17 (±11)||0·18|
|Good*||7||71 (±17)||71 (±17)|
|Others||20||15 (±8)||0·018||16 (±8)||0·032|
In conclusion, we have confirmed in our pre-imatinib cohort of patients, a previously described good prognostic index to identify a subset of patients with Ph+ALL with significantly better outcome. This index may help to identify children in whom allogeneic HSCT could be omitted in the present TKI era.