Advances in therapies for acute promyelocytic leukemia
To whom correspondence should be addressed.
Acute promyelocytic leukemia (APL), a distinct subtype of acute myelogenous leukemia (AML), results from the arrest of the maturation of hematopoietic progenitors at the promyelocyte stage. It has been shown that APL is associated with a reciprocal chromosomal translocation, involving chromosomes 15 and 17, which fuses the gene encoding the retinoic acid receptor α (RARα) and the promyelocytic leukemia (PML) gene. The resultant PML-RARα fusion protein plays a critical role in the pathogenesis of APL. Although there are many subtypes of AML, all are typically managed using a standard chemotherapy regimen of an anthracycline plus cytarabine arabinoside (CA). Despite high rates of complete remission following standard chemotherapy, most patients relapse and long-term disease-free survival is only 30–40%. The introduction of drugs such as all-trans retinoic acid (ATRA) that promote progenitor differentiation by directly inhibiting the PML-RARα fusion protein has changed the treatment paradigm for APL and markedly improved patient survival. The purposes of the present review are to provide the latest results and future directions of clinical research into APL and to illustrate how new therapies, such as ATRA plus anthracycline-based induction and consolidation therapy, risk-adapted therapy, salvage therapy containing arsenic trioxide-based regimens, and hematopoietic stem cell transplantation, have improved the treatment outcomes for APL patients. (Cancer Sci 2011; 102: 1929–1937)
Acute promyelocytic leukemia (APL), formerly known as AML-M3, is a biologically and clinically distinct variant of acute myelogenous leukemia (AML), characterized by the clonal expansion of hematopoietic precursors blocked at the promyelocyte stage of differentiation.(1) Acute promyelocytic leukemia is characterized by a balanced reciprocal translocation between chromosomes 15 and 17, resulting in the fusion of the promyelocytic leukemia (PML) and retinoic acid receptor α (RARα) genes.(2) During normal hematopoiesis, retinoic acid (RA) plays a key role in myeloid progenitor differentiation through its nuclear receptors (RARs) and retinoid X receptors (RXRs).(3) The RARs are transcription factors that regulate genes involved in myeloid progenitor development and cell cycle control;(2,4–6) PML-RARα interferes with this process by recruiting abnormal transcrption factors and histone-modifying enzymes to these critical genes, linking the ability of the PML-RARα oncoprotein to promote self-renewal and block differentiation.(6,7) The PML-RARα protein is an aberrant RAR that contributes to leukemogenesis by dominantly antagonizing aspects of cellular differentiation regulated by the RAR/RXR signaling pathway.(4,8,9) Based on these molecular findings about leukemogenesis, APL is now classified in the World Health Organization (WHO) classification system(10) as APL with t(15;17)(q22;q12);PML-RARα.
Clinically, APL represents a medical emergency, with a high rate of early mortality often due to hemorrhage from disseminated intravascular coagulation (DIC) or hyperfibrinolysis.(11,12) Before 1992, the induction therapy for patients with APL was similar to that for all other AML and included anthracycline and cytarabine arabinoside (CA). Compared with other AML subtypes, APL cells are especially sensitive to anthracycline, possibly due to significantly lower expression of P-glycoprotein and other resistance markers.(13–15) Therefore, chemotherapy with anthracycline (daunorubicin [DNR], idarubicin [IDA], or others) and CA was the frontline treatment of APL, achieving complete remission (CR) in 75–80% of patients.(16,17) Unfortunately, this standard treatment regimen was associated with a high early death rate due to the exacerbation of pre-existing DIC, necessitating intensive platelet and fibrinogen support. Despite high sensitivity to anthracycline, only 35–45% of APL patients are cured by standard chemotherapy alone.(18,19)
The introduction of all-trans retinoic acid (ATRA) in 1987 changed the treatment paradigm of APL.(20,21) High intranuclear concentrations of ATRA increase the expression of normal functioning RARα, which then outcompetes the aberrant PML-RARα, inducing remission through differentiation of neoplastic promyelocytes into mature granulocytes and the re-emergence of normal hematopoietic cells.(4,22,23) Therefore, several clinical trials have established ATRA combined with anthracycline-based chemotherapy for induction and consolidation as standard care.(24–26)
All-trans retinoic acid plus anthracycline-based induction therapy. Complete remission of APL following ATRA treatment was first demonstrated by the Shanghai group,(20,21) who also reported a significant reduction in the duration of coagulopathy compared with standard chemotherapy.(27) Several studies conducted during the early 1990s found that APL patients receiving induction therapy consisting of ATRA followed by chemotherapy fared significantly better than patients treated with chemotherapy alone.(27,28) In these studies, the rates of CR and early death were not significantly different; however, the relapse rate was significantly lower in the ATRA-treated group. The treatment of patients with ATRA was further refined following a randomized trial conducted by the European “French–Belgian–Swiss” APL Group (the European APL Group), showing a benefit of concurrent ATRA and chemotherapy for induction compared with a sequential regimen approach.(29) In addition, ATRA therapy has significant and potentially fatal adverse effects, known as retinoic acid syndrome, which consists of elevated white blood cell (WBC) counts, fever, respiratory distress, interstitial pulmonary infiltration, pleural effusion, and weight gain.(30) Among patients in remission following treatment with ATRA alone, the incidence of retinoic acid syndrome is approximately 25%,(31) whereas that in patients treated with a combination of ATRA plus chemotherapy is reduced to 5%.(24,25) Based on the results of these studies, the combination of ATRA plus chemotherapy became the standard approach for treating newly diagnosed APL.
The Spanish cooperative group Program de Estudio y Trataminento de las Hemopatias Malignas (PETHEMA) took advantage of the sensitivity of leukemic promyelocytes to anthracyclines and performed Phase II studies omitting CA from the induction and consolidation therapies.(25,32) This strategy was supported by a prospective, randomized trial conducted by the Italian cooperative group Gruppo Italiano Malatte Ematologiche dell’Adulto (GIMEMA) prior to the introduction of ATRA, which found no benefit in combining CA with IDA in newly diagnosed APL patients.(33) It is now accepted that CR can be achieved in most APL patients without CA as part of the induction therapy, with potentially less toxicity, and ATRA plus anthracycline combination induction therapy without CA is now the standard induction treatment for APL.(34)
Arsenic trixoide as an alternative approach. After excellent results with arsenic trixoide (ATO) in the treatment of APL relapse,(35,36) several studies explored the role of this agent in front-line therapy.(37–41) The CR rate in these studies was similar to that obtained with the standard ATRA plus anthracycline-based induction therapy, ranging from 86% to 95%. The use of ATO is regarded as a promising option because of its apparently most favorable safety profile compared with conventional chemotherapy, particularly with respect to myelosuppression and some uncommon late complications, as well as its proven therapeutic efficacy. However, the use of ATO would not be exempt of other real or potential disadvantages. For example, ATO is associated with prolongation of the QT interval or ventricular arrhythmias, which can be fatal. Because of the high potential for embryotoxicity, ATO cannot be recommended for use at any stage of pregnancy. Finally, long-term complications of therapy remain uncertain; in particular, there are few data regarding the incidence of secondary neoplasms in APL patients treated with ATO.(42) Currently, some ongoing large studies have been designed to compare ATRA + chemotherapy and ATO-based regimens.(43) However, until the results of these trials are available, the use of ATO for newly diagnosed APL patients as induction therapy should be restricted to patients included in clinical trials or to those in whom chemotherapy is contraindicated.
Because continuous treatment with ATRA alone will cause progressive resistance to the drug resulting in relapse usually within 3–6 months,(12) consolidation chemotherapy is necessary to maintain CR. Since the first successful report of DNR monotherapy,(43) the role of CA in the treatment of APL has remained contentious. Until several years ago, no study had found any benefit in the inclusion of CA compared with high-dose anthracycline monotherapy.(44) Complete molecular remission is achieved in >90% of patients receiving two or three cycles of anthracycline-based chemotherapy following induction, leading to this strategy being adopted as the standard regimen for consolidation.(25,45) The benefit provided by the addition of ATRA to chemotherapy for consolidation has not yet been demonstrated in randomized studies. Nevertheless, historical comparisons of consecutive trials performed independently by the GIMEMA(46) and PETHEMA(32) groups showed a significant improvement in outcomes when a standard dose of ATRA was given for 15 days in conjunction with chemotherapy, suggesting that ATRA contributes to the reduction in the risk of relapse.(34) Combination therapy with current ATRA and anthracycline-based chemotherapy for induction and at least two further cycles of consolidation achieves molecular remission in >90% of patients, resulting in this strategy being adopted as the standard for consolidation.(45) However, approximately 20% of patients will relapse following this treatment, indicating that there are still major obstacles to a cure for APL patients.(24,29,47) Recently, treatments have tended towards risk-adapted strategies to modulate the intensity of treatment during consolidation according to the predefined risk of relapse.(48)
Gore et al.(49) demonstrated the efficacy of a single cycle of ATO-based consolidation therapy in newly diagnosed APL patients following ATRA-plus anthracycline-based induction therapy. They reported that the estimated disease-free survival (DFS) and overall survival (OS) rates were 90% and 88%, respectively, with a median follow-up of 2.7 years. Observations published by the North American Leukemia Intergroup suggest that the addition of ATO consolidation to standard induction and consolidation therapy significantly improves both event-free survival (EFS) and DFS in patients with newly diagnosed APL.(50) These findings suggest that using ATO in the primary management of APL would results in decreased exposure to other cytotoxic agents.
The European APL Group has shown benefits with maintenance chemotherapy using ATRA, which was given intermediately in a randomized study (the APL93 study).(29) That study revealed a lower relapse rate with triple combination therapy with ATRA, methotrexate (MTX), and 6-mercaptopurine (6-MP), which proved particularly effective in patients with an elevated WBC count at presentation who were considered high-risk patients. In addition, long-term follow-up in that study (median follow-up 10 years) demonstrated the benefit of prolonged maintenance with ATRA plus chemotherapy for high-risk patients presenting with initial WBC counts >5 × 109/L.(51) The inclusion of ATRA resulted in the cumulative incidence of relapse declining from 68.4% with no maintenance to 20.6% with combined ATRA and chemotherapy maintenance. In contrast, the APL97 study, conducted by the Japan Adult Leukemia Study Group (JALSG), found no benefit of six courses of intensive maintenance therapy for patients in molecular remission compared with observation only after chemotherapy.(26) Similarly, a recent report from the GIMEMA group based on a very high number of patients with genetically proven APL has questioned the benefit of ATRA-based maintenance therapy.(52) Therefore, it is likely that the usefulness of maintenance therapy in APL may depend on multiple variables, including the type of anthracycline (IDA versus DNR), the intensity of chemotherapy delivered during induction and consolidation, and on the relapse risk of the patient.(26,29,51,52) Despite the uncertainty of the benefit provided by maintenance therapy, it is clear that many patients achieving molecular remission by the end of consolidation will ultimately relapse, especially those presenting with a high WBC (>10 × 109/L). Although the role of ATRA-based maintenance therapy remains contentious, none of these recent studies has included an arm without maintenance therapy, with the exception of JALSG97.(26)
Current Risk-Adapted Therapy
Despite the evidence from large multicenter trials of the benefit of upfront ATRA plus anthracycline-based chemotherapy, relapse still occurs in approximately 20% of patients.(24,25,29,47,51,53) Multivariate analyses have revealed that the single most important factor for relapse is a WBC count of 10 × 109/L or higher at presentation.(25,32,47,48) Therefore, many investigators have tried to classify individual patient risk and modulate the intensity of induction or consolidation therapy accordingly (Table 1).(32,54) A joint analysis by the PETHEMA and the European APL Groups comparing the outcomes of PETHEMA LPA99 and European APL 2000 trials demonstrated that the addition of high-dose of CA during consolidation is of benefit to high-risk patients.(55) Subsequently, through the combined analysis of the APL93 and APL2000 trials, the European APL Group also confirmed a marked improvement in the prognosis of APL patients with high WBC counts treated with escalating doses of CA: in patients with WBC counts between 10 and 50 × 109/L, the remission rate increased from 89% to 93% and the 5-year cumulative incidence of relapse decreased from 40% to 9.5%; in patients with very high WBC counts (50 × 109/L), the remission rate increased from 82% to 91% and the 5-year cumulative incidence of relapse decreased from 59% to 24%.(54) These results demonstrated that upfront ATRA combined with intensified consolidation with high-dose CA has potential benefits for high-risk APL patients. Recently, the GIMEMA (AIDA 2000)(46) and the PETHEMA (LPA 2005)(56) groups demonstrated that the administration of high-dose CA (1 g/m2) for 4 days during consolidation improves the outcome of patients at high risk. In the AIDA 2000 trial, the 6-year OS was 89% in patients with low- and intermediate-risk disease compared with 83% in patients with high-risk disease.(46) Similarly, in the LPA2005 study, the 4-year OS was 96% in low-risk patients, 91% in intermediate-risk patients, and 79% in high-risk patients.(56) These two studies have revealed improved outcome compared with each of their previous risk-adapted protocols using increased anthracycline doses, instead of CA, in high-risk APL patients.
Table 1. Risk-adapted therapy for patients with acute promyelocytic leukemia
There is great expectation that minimal residual disease (MRD) monitoring will enable precise identification of patients who will relapse.(57,58) Lee et al.(59) reported that MRD by real-time quantitative polymerase chain reaction (RQ-PCR) analysis after upfront ATRA and anthracycline-based induction chemotherapy was detectable in half of the patients, but was undetectable in the remaining half. All patients negative for RQ-PCR after induction had a favorable clinical course thereafter, without relapse. In contrast, after the first consolidation, MRD was still detectable exclusively in approximately 30% of patients positive for RQ-PCR after induction, who were highly susceptible to subsequent hematologic relapses despite additional consolidation.(59) Recently, Grimwade et al.(60) demonstrated that MRD monitoring by RQ-PCR successfully identified the majority of APL patients subject to relapse and provided the most powerful predictor of relapse-free survival (RFS), and was far superior to presenting WBC counts. In that study, early treatment intervention with ATO prevented progression to overt relapse in most patients and the RFS rate at 1 year from molecular relapse was 73%. The results of these studies suggest that careful follow-up with serial quantification of MRD by RQ-PCR would be needed to assess the value of an individualized, response-oriented treatment strategy in the treatment of APL patients.
Salvage Therapy for Relapsed and Refractory APL Patients
The standard salvage therapy for patients with relapsed or refractory APL used to consist of the readministration of ATRA and chemotherapy for induction, generally containing a high dose of CA, followed by additional chemotherapy and/or hematopoietic stem cell transplantation (HSCT).(61) However, the identification of ATO as an effective salvage therapy has changed the approach significantly. After the initial report of the use of ATO from the Shanghai group,(62) several groups confirmed the high efficacy of ATO for patients with APL who relapsed after ATRA-containing regimens. Recent studies have shown that one to three treatment cycles with ATO achieve a second molecular remission in nearly 80% of relapsed or refractory patients, with overall 2-year survival rates of 50–60% after repeated cycles of ATO combined with chemotherapy (Table 2).(63–68) Given the high efficacy of ATO in this patient group, ATO is now regarded as the best treatment option in these settings;(36,69) however, the best consolidation strategy following the ATO-induced second molecular remission remains unknown.
Table 2. Arsenic trixoide therapy for relapsed/refractory acute promyelocytic leukemia
|Shen et al.(62)||15||40||ATRA + CHT or CHT||93||NA||NA||85% 1-year OS 76% 1-year DFS|
|Soignet et al.(64)||12||34||ATRA + CHT||91.7||66.7||Five to six cycles of ATO||63% 2-year OS† 49% 2-year DFS†|
|Niu et al.(63)||47||38||ATRA + CHT||85.1||NA||ATO or CHT or ATO + CHT||50% 2-year OS 41% 2-year DFS|
|Soignet et al.(65)||40||NA||ATRA or 9-cis RA + CHT||85||77.5||ATO or CHT or auto/allo-HSCT||66% 1.5-year OS 56% 1.5-year RFS|
|Raffoux et al.(66)||20||46||ATRA + CHT||80||NA||One to two cycles of ATO ± ATRA ± auto/allo-HSCT||59% 2-year OS 59% 2-year RFS|
|Thomas et al.(67)||28||52||NA||86||NA||ATO + CHT ± auto/allo-HSCT||73% 2-year OS 84% 2-year DFS|
|Shigeno et al.(70)||34||47||CHT ± ATRA||91||72||ATO or CHT or auto/allo-HSCT||56% 2-year OS 17% 2-year EFS|
Hematopoietic Stem Cell Transplantation
The risk-adapted protocols described above clearly illustrate the benefits, in terms of both efficacy and safety, of matching the intensity of post-remission treatment to the risk group.(32,46,54) To maximize treatment intensity for very high-risk patients, chemotherapy can be combined with some form of HSCT. Autologous HSCT, such as autologous peripheral stem cell transplantation (auto-PBSCT), would be the ideal therapy because this would maximize the antileukemic effects while keeping transplant-related mortality (TRM) to a minimum. The Fukuoka BMT group has used upfront auto-PBSCT in 20 APL patients in CR and after consolidation, including five high-risk patients.(70) Pretransplant conditioning consisted of the BEA regimen: busulfan 1 mg/kg every 6 h, p.o., on Days -8 to -5, etoposide 20 mg/kg, i.v., on Days -4 and -3, and CA 3 g/m2, i.v., every 12 h on Days -3 and -2. Granulocyte colony-stimulating factor (G-CSF) was also included (G-CSF combined BEA regimen), given i.v., as follows: 5 μg/kg on Days -14 to -8, 10 μg/kg on Days -7 and -6, and 20 μg/kg on Days -5 and -4 in combination with continuous infusion of CA 100 mg/m2 on Days -12 to -6.(71,72) Rapid engraftment was obtained in all patients with no TRM. Remarkably, leukemia-free survival (LFS) in the absence of maintenance therapy was 100% at 10 years (Table 3). All peripheral blood stem cells (PBSC) that were infused were negative for MRD in both low- and high-risk patients who were examined. Thus, high-dose chemotherapy followed by auto-PBSCT as well as deep remission status, reflecting negative MRD in PBSC, would have benefited our patients, because several reports indicate that molecular remission would be an important prognostic factor for patients undergoing autotransplantation during the second molecular remission. In contrast, according to a multicenter retrospective survey from the European Cooperative Group for Blood and Marrow Transplantation (EBMT), autologous HSCT (auto-HSCT) does not lead to any improvement in 5-year LFS (∼70% in 149 patients in first remission) compared with ATRA-combining risk-adapted chemotherapy.(55) Therefore, auto-HSCT is not routinely recommended for APL patients in first remission,(34,73) although these treatments have yet to be compared directly in a randomized trial.(74–78)
Table 3. Autotransplantation for acute promyelocytic leukemia
|Mandelli et al.(75)||187||30||CR1: 129 CR2: 58||BM||NA||NA||CR1: 42% 7-year LFS CR2: 22% 4-year LFS|
|Meloni et al.(81)||15||38||CR2: 15||BM||NA||Negative 8 Positive 7||Median duration of CR PCR positive: 5 months PCR negative: 28.5 months|
|Ferrant et al.(76)||36||NA||CR1: 36||BM||NA||NA||70% 3-year LFS 83% 3-year OS|
|Roman et al.(78)||10||47||CR1: 8 CR2: 1 PR: 1||BM 4||NA||Negative 2||Median survival: 41 months|
|Lo Coco et al.(57)||8||40||CR2: 8||BM||NA||Negative 8||Median duration of CR: 11 months|
|Ottaviani et al.(77)||16||30||CR1: 13 PR: 1 CR2: 1 CR3: 1||BM||NA||Negative 12 Posirtive 3||Median survival CR1: 55 months CR2: 16 months|
|Thomas et al.(67)||22||NA||CR2: 22||BM 5 PB 17||Negative 2 Positive 4||Negative 9 Positive 1||77% 3-year DFS|
|Ferrara et al.(82)||6||38||CR2: 6||BM or PB||Negative 6||Negative 6||Median duration of CR: 36 months|
|de Botton et al.(80)||50||45||CR2: 50||BM 43 PB 7||Negative 20 Positive 2||Negative 28 Positive 2||79% 7-year RFS 61% 7-year EFS|
|Sanz et al.(74)||344||50 (CR1) 38 (CR2)||CR1: 149 CR2: 195||CR1: BM 92, PB 57 CR2: BM 91, PB 104||NA||NA||CR1: 70% 5-year LFS CR2: 51% 5-year LFS|
|Thirugnanam et al.(83)||14||33||CR2: 12 CR3: 2||PB||NA||Negative 14||83% 5-year EFS|
|Kamimura et al.(70)||26||45||CR1: 20 CR2: 6||PB||CR1: Negative 15||CR2: Nevative 6||CR1: 100% 11-year LFS CR2: 100% 3-year LFS|
Regarding successive consolidation, there is some evidence that treatment intensification with HSCT may improve the outcomes for patients in second remission following ATO therapy.(79) The consolidation strategy after ATO-induced second remission generally consists of HSCT, with the choice of transplant modality based mainly on PCR status. Although allogeneic HSCT (allo-HSCT) offers the greatest antileukemic activity due to its graft-versus-leukemia effect, it is only recommended for patients who have not achieved second molecular remission because of the risk of TRM.(68) Autologous HSCT has a lower risk of TRM than allo-HSCT and is a reasonable option for consolidation in relapsed APL patients (Table 3).(34,68,73,74)
The EBMT surveyed the outcome of APL patients in second remission who underwent auto-HSCT between 1993 and 2003 and found a 5-year LFS of 51%.(74) De Botton et al.(80) also documented the benefit of auto-HSCT in 50 APL patients who relapsed after an ATRA-containing treatment in 2004: the RFS at 7 years was 79.4%, with a TRM of only 6% after auto-HSCT. In that study, two of 30 PBSC samples were positive for MRD by PCR; one of these patients relapsed following autograft compared with three of the 28 PCR-negative patients (11%) who relapsed.(80) Of the patients autografted with PCR-negative PBSC, the RFS at 7 years increased to 87.3%, indicating that auto-HSCT would be effective for the treatment of relapsed APL if performed during the second molecular remission, consistent with previous studies(57,81,82) (Table 3).
Thomas et al.(67) used ATO as reinduction therapy for 28 relapsed APL patients. Nine of 24 patients achieving molecular remission underwent auto-HSCT, all of whom achieved second remission with 2-year LFS and OS rates of 100%. In our study,(70) six low-risk patients relapsed after cessation of maintenance therapy, three of whom were treated with chemotherapy as reinduction for relapsed APL prior to the use of ATO, whereas the remainder underwent ATO-containing chemotherapy. All six patients achieved second molecular remission after a single course of reinduction chemotherapy and PBSC were harvested at this time. Six patients received auto-HSCT during their second remission and remained in CR without maintenance for a median of 41 months (range 2–187 months). Furthermore, Thirugnanam et al.(83) have recently shown that, for patients in second remission following ATO-based regimens, consolidation with auto-HSCT is associated with a significantly superior clinical outcome compared with ATO-based maintenance: 5-year EFS was 83.3% in patients receiving auto-HSCT compared with 34.4% in those who did not. Because ATO therapy achieves molecular remission in nearly 80% of relapsed patients when given for at least two cycles,(68) auto-HSCT may be adopted as standard therapy for this patient group (Table 3).
The outcomes of APL have improved over the past 20 years due to the incorporation of ATRA into chemotherapy regimens, which has reduced both the incidence of early death and relapse. Large multicenter studies have shown that treatment regimens combining upfront ATRA and anthracycline-based chemotherapy without CA for induction and consolidation, as well as ATRA-based maintenance therapy, are sufficient for the majority of low-risk patients. In addition, escalating doses of CA during consolidation therapy benefit high-risk patients. There is a trend towards designing risk-adapted treatment strategies, which modulate treatment intensity during consolidation according to the risk of relapse, as indicated by the WBC count at presentation. These strategies are proving to be very effective in improving survival even in high-risk groups, where survival rates of 85% are possible. Because of these successes, HSCT is not routinely recommended for APL patients in first molecular remission. In contrast, auto-HSCT can be considered as consolidation therapy for relapsed and refractory patients following ATRA-containing regimens, provided that the patient achieves a second molecular remission and the harvested PBSC or bone marrow cells are PCR negative. Autologous HSCT may play a more prominent role in the treatment of relapsed and refractory APL patients in the future, based on the high incidence of second remissions following ATO treatment.
The authors declare no potential conflict of interest.