ORIGINAL ARTICLE Outcome of patients with acute promyelocytic leukemia failing to front-line treatment with all-trans retinoic acid and anthracycline-based chemotherapy (PETHEMA protocols LPA96 and LPA99): benefit of an early intervention J Esteve 1,16 , L Escoda 2,16 , G Martı ´n 3 , V Rubio 4 , J Dı ´az-Mediavilla 5 , M Gonza ´lez 6 , C Rivas 7 ,CA ´ lvarez 8 , JD Gonza ´ lez San Miguel 9 , S Brunet 10 , JF Toma ´s 11 , M Tormo 12 , MJ Sayas 13 , P Sa ´nchez Godoy 14 , D Colomer 1 , P Bolufer 15 and MA Sanz 3 , on behalf of the Spanish Cooperative Group PETHEMA 1 Hospital Clı ´nic, Barcelona, Spain; 2 Hospital Joan XXIII, Tarragona, Spain; 3 Hospital La Fe, Valencia, Spain; 4 Hospital General de Jerez de la Frontera, Jerez de la Frontera, Spain; 5 Hospital Clı ´nico San Carlos, Madrid, Spain; 6 Hospital Universitario, Salamanca, Spain; 7 Hospital General, Alicante, Spain; 8 Hospital de Cruces, Baracaldo, Spain; 9 Hospital Insular de Gran Canaria, Las Palmas de Gran Canaria, Spain; 10 Hospital de Sant Pau, Barcelona, Spain; 11 Fundacio ´ n Jime ´nez Dı ´az, Madrid, Spain; 12 Hospital Clı ´nico, Valencia, Spain; 13 Hosp. Dr Pesset, Valencia, Spain; 14 Hospital Severo Ochoa, Legane ´s, Spain and 15 Department of Medical Biopathology, Hospital La Fe, Valencia, Spain To determine prognosis of acute promyelocytic leukemia (APL) failing to front-line therapy with all-trans retinoic acid (ATRA) and anthracyclines, outcome of 52 patients (32 M/20 F; age: 37, 3–72) included in PETHEMA trials LPA96 and LPA99 who presented with either molecular failure (MOLrel, n ¼ 16) or hematological relapse (HEMrel, n ¼ 36) was analyzed. Salvage therapy consisted of ATRA and high-dose ara-C-based che- motherapy (HDAC) in most cases (83%), followed by stem-cell transplantation (autologous, 18; allogeneic, 10; syngeneic, 1). Fourteen patients with MOLrel (88%) achieved second mole- cular complete response (molCR), whereas 81% HEMrel patients responded to second-line treatment, with 58% molCR. After median follow-up of 45 months, four MOLrel and 18 HEMrel patients, respectively, experienced a second relapse. Outcome after MOLrel compared favorably to HEMrel, with longer survival (5-year survival: 64714 vs 2478%, P ¼ 0.01) and lower relapse risk (5-year relapse risk: 30713 vs 6479%; P ¼ 0.044). Additionally, agep40 and male gender were favor- able variables for survival, whereas molecular response predicted longer leukemia-free survival. In conclusion, early institution of salvage therapy at molecular failure, before onset of hematological relapse, is beneficial in APL. Moreover, given the poor outcome of HEMrel managed with ATRA and HDAC, use of alternative therapeutic strategies in this setting is warranted. Leukemia (2007) 21, 446–452. doi:10.1038/sj.leu.2404501; published online 4 January 2007 Keywords: acute promyelocytic leukemia; molecular relapse; hematological relapse; salvage therapy Introduction The currently established treatment of acute promyelocytic leukemia (APL) relies on the combination of all-trans retinoic acid (ATRA) and chemotherapy and results in cure of more than 70% of newly diagnosed patients. 1–7 Nonetheless, more than 20% of patients presenting with APL still die of their disease. This percentage includes both early death during induction in approximately 10% of patients, mainly owing to hemorrhage and infection, and relapse after appropriate therapy. For these relapsed patients, optimal salvage strategies are not well established and their prognosis is usually considered poor. 8,9 Arsenic trioxide (ATO) provides a high response rate in this setting and is generally considered as the first option of salvage therapy, although its precise role in the management of relapsed APL, owing to its recent introduction, remains to be definitively defined. 10–17 Monitoring of minimal residual disease (MRD) during treat- ment of APL, based on the assessment of the PML/RARa transcript by means of reverse transcription-polymerase chain reaction (RT-PCR), is useful to anticipate overt hematological relapse, as shown by the high predictive value of impending relapse that the persistence or reappearance of this molecular marker after consolidation therapy has. 3,18,19 Therefore, analysis of relapse in patients undergoing treatment for APL should include these molecular failures. Preliminary reports indicate that institution of salvage chemotherapy for molecular disease might result in a significantly improved outcome as compared to that of patients treated for an overt hematological relapse, but this observation has not been subsequently confirmed in larger series. 20,21 With this background, we analyzed the outcome in a series of APL patients treated according to PETHEMA protocols LPA96 and LPA99 4,7 who experienced a therapeutic failure, defined on the basis of molecular or hematological grounds, and treated according to a highly homogenous salvage therapy based on ATRA and high-dose ara-C-based chemotherapy (HDAC). Patients and methods Patients The study included 52 patients, diagnosed with de novo PML/ RARa-positive APL and treated according to PETHEMA proto- cols LPA96 and LPA99, who experienced either molecular failure (MOLrel) or hematological overt relapse (HEMrel) during the period between November 1996 and December 2004. 4,7 Briefly, APL PETHEMA protocols consisted of an induction phase with the AIDA regimen (ATRA 45 mg/m 2 until complete remission (CR) and idarubicin 12 mg/m 2 on days 2–4–6–8), three cycles of anthracycline-based consolidation (idarubicin 5–7 mg/ m 2  4, mitoxantrone 10 mg/m 2  5 and idarubicin 12 mg/ m 2  1–2, respectively), followed by a maintenance therapy Received 20 June 2006; revised 12 September 2006; accepted 13 October 2006; published online 4 January 2007 Correspondence: Dr J Esteve, Department of Hematology, Hospital Clı ´nic, Villarroel 170, 08036 Barcelona, Spain. E-mail: jesteve@clinic.ub.es 16 These two authors contributed equally to this study Leukemia (2007) 21, 446–452 & 2007 Nature Publishing Group All rights reserved 0887-6924/07 $30.00 www.nature.com/leu