Risk Stratification in Acute Promyelocytic Leukemia Based on Elevated White Cell Count and Reduced Platelet Count Does Not Identify Subsequent Relapses: A Retrospective Review of 60 Patients.

Acute promyelocytic leukemia (APL) has become one of the more readily treatable subtypes of acute myeloid leukemia on the basis of its particular sensitivity to anthracyclines and the introduction of the differentiating agent all-trans retinoic acid (ATRA). Unfortunately some patients will ultimately relapse after achievement of complete remission. A PETHEMA and GIMEMA (PETH/GIM) cooperative group joint study sought to identify risk factors associated with relapse and developed a predictive model based on white cell count (WCC) and platelet count (Plt) at diagnosis. This model for relapse free survival was capable of segregating patients into low risk (WCC < 10 x 10⁹/L / Plt > 40 x 10⁹/L), intermediate risk (WCC < 10 x 10⁹/L / Plt < 40 x 10⁹/L), and high risk (WCC > 10 x 10⁹/L). A subsequent follow up study demonstrated improved outcomes for intermediate and high risk group patients treated with an anthracycline and ATRA based risk adapted strategy. We reviewed 60 cases of adult de novo APL treated at Vancouver General Hospital from August 1995 to December 2003. Thirty five females (58%) and 25 males (42%) were treated with a standard induction and consolidation protocol consisting of 3 cycles of Daunorubicin (60 mg/m² OD for 3 days), Cytosine arabinoside (100 mg/m² BID for 7 days), and ATRA (45 mg/m²/day until complete remission or a maximum of 60 days). In 2000 maintenance therapy was introduced and ATRA and/or varying doses of Methotrexate and 6-Mercaptopurine were then instituted for up to 2 years duration. At the completion of induction chemotherapy complete morphological and cytogenetic remission was achieved in 55/60 patients (92%) with 4 induction failures and only 1 death during induction treatment. Subsequently during and after consolidation and maintenance therapy at a median follow up of 34 months there were a total of 10 episodes of relapse (18%) in these 55 patients. When the 55 patients in complete remission were stratified at diagnosis by the PETH/GIM risk group there were 18 low risk patients (33%), 26 intermediate risk patients (47%), and 11 high risk patients (20%). The number of relapses was 6 in the low risk group (33%), 4 in the intermediate risk group (15%) and no relapses were identified in the high risk group. In our experience the predictive model for relapse free survival developed by the PETH/GIM cooperative group does not identify those at most likelihood for subsequent relapse after achieving an initial complete remission. The greatest number of relapse was seen in the low risk group with no relapses in the high risk group. These variations may be explained in part by the different chemotherapy treatment protocols with variable maintenance therapy. Future investigation will focus on the prognostic role of immunophenotype, additional cytogenetic abnormalities and molecular isoforms in determining outcome in APL.