Induction Therapy

The evolution of treatment strategies for APL based on clinical observation and well-constructed clinical trials is one of the most rewarding sagas of modern hematology. In 1988, a group in Shanghai reported that ATRA given as single agent induced complete remission rates of 85%.¹⁹ The first North American Intergroup study confirmed a 70% complete remission rate with single-agent ATRA, which was equivalent to rates seen with conventional doses of cytarabine (Ara-C) and daunorubicin.^20,21

The French APL 93 trial compared ATRA plus chemotherapy (Ara-C and daunorubicin) and ATRA followed by chemotherapy. Complete remission rates were 92% in both arms, but the relapse rate at 2 years was 6% in the combined group versus 16% for the sequential group.^22,23

Induction regimens were pared down to ATRA and idarubicin in the Italian GIMEMA 93 and Spanish PETHEMA LPA 94 trials, which produced complete remission rates of 95%, raising the question of whether Ara-C was needed in APL induction.^24,25 Patients with elevated WBC counts were commonly observed to have high-risk disease, based on higher deaths during induction and increased rates of relapse.

As an outgrowth of the PETHEMA LPA 94 trials, Sanz et al.^26,27 developed a risk stratification based solely on WBC and platelet counts at presentation. The induction regimen remained the same (idarubicin and ATRA), but ATRA was added to consolidation cycles 1 through 3 for all but low-risk patients (WBC ≤ 10,000/mcL and platelets > 40,000/mcL). The complete remission rate in this trial was 91%, with almost all failures attributed to hemorrhage, infection, or differentiation syndrome. Factors predictive of death during induction were WBC count greater than 10,000/mcL, age older than 60 years, creatinine level of 1.4 or greater, and male sex.

In 2006, Ades et al.²⁸ reported the outcome of the French APL 2000 trial, in which 340 patients younger than 60 years with a WBC count less than 10,000/mcL were randomized to receive ATRA (45 mg/m²) and daunorubicin (60 mg/m²/d for 3 days) as induction therapy with or without Ara-C (200 mg/m²/d for 7 days). Those randomized to Ara-C in induction also received Ara-C in consolidation. Patients with WBC counts greater than 10,000/mcL or age older than 60 years all received Ara-C. Although the complete remission rates were similar in the randomized groups (99% with and 94% without Ara-C), those receiving Ara-C had a lower 2-year cumulative incidence of relapse that translated into an improved event-free survival rate (93% vs. 79%, respectively, at 2 years). In patients with WBC counts greater than 10,000/mcL, the complete remission rate was 97% and the 2-year disease-free survival rates were 89% for those younger than 60 years and 79% for those older than 60 years. A report of a joint analysis of the outcomes in the PETHEMA 99 and French APL 2000 trials in patients younger than 65 years showed that in patients with WBC counts less than 10,000/mcL, complete remission rates were similar, but the relapse rates at 3 years were lower (4.2%) in the PETHEMA trial, which used ATRA during consolidation, than in the APL 2000 Ara-C–containing regimen (14.3%; P = .03).²⁹ However, for patients with WBC counts greater than 10,000/mcL, the Ara-C–containing protocol had better rates of complete remission (95% vs. 84%) and 3-year survival (92% vs. 81%; P = .18).²⁹ The second North American Intergroup Trial (NAIT) also used ATRA (45 mg/m²), daunorubicin (50 mg/m²/d for 4 days), and Ara-C (200 mg/m²/d for 7 days), and showed similar initial complete remission rates.³⁰ Consolidation in this trial differed in that 2 cycles of a novel agent, arsenic trioxide, were given after induction and before the final 2 cycles of anthracycline.

All 3 regimens offer excellent outcomes, with the choices influenced by risk group, age, and cardiovascular risks. The panel recommends that patients with APL be treated with one of the regimens established from the clinical trials and that the regimen be used consistently through all components, rather than mixing induction from one trial with consolidation from a second trial. The NCCN Guidelines are broken down by ability to tolerate anthracyclines and the PETHEMA risk classification based on WBC count at diagnosis. Patients with APL who can tolerate anthracyclines are stratified into risk groups based on WBC counts. In general, the panel has preferentially listed regimens of equivalent efficacy for a given risk group based on ease of administration. For low- or intermediate-risk patients (WBC < 10,000/mcL), the panel recommends initial induction with ATRA plus idarubicin alone³¹ (category 1); ATRA plus daunorubicin and Ara-C^21,29,30 (category 1 for those on French APL 2000 protocol²⁹); or enrollment in a clinical trial.

For high-risk patients (WBC > 10,000/mcL), the panel prefers regimens that include Ara-C along with ATRA plus daunorubicin over ATRA plus idarubicin because of higher disease-free survival rates at 2 years, although the complete remission rates reported in the trials were very comparable.^21,29,31

Arsenic trioxide (ATO) has also been found to be a potent promoter of apoptosis in APL cells.^32,33 In 2004, Shen et al.³⁴ published outcomes on single-agent ATRA, single-agent ATO, or the combination of both. Although complete remission rates exceed 90% in all 3 arms, the decline in quantity of PML/RAR alpha fusion protein as measured by quantitative PCR was significantly higher with the combination. Hematologic recovery was quicker and relapse-free survival was improved at 18 months.

Subsequently, Estey et al.³⁵ treated 25 patients with low- and intermediate-risk APL using a similar combination of ATRA and ATO, and 19 high-risk patients were treated using ATRA and ATO combined with gemtuzumab ozogamicin, 9 mg/m², on day 1 of induction therapy. Complete remission occurred in 24 patients with low-risk disease and 15 patients with high-risk disease. The authors suggested that ATRA plus ATO may be an alternative to chemotherapy in patients with low- and intermediate-risk untreated APL and, when combined with gemtuzumab ozogamicin, may improve outcomes in high-risk patients. However, in October 2010, the FDA withdrew the prior approval of gemtuzumab ozogamicin for the treatment of older patients with relapsed AML because of a lack of clinical benefit, and therefore this drug is no longer commercially available in the United States. These NCCN Guidelines indicate that ATRA plus ATO is an alternative for patients who cannot tolerate anthracycline therapy.

Consolidation Therapy

Because the differentiating action of ATRA occurs over a longer period than the cytoreduction of conventional chemotherapy, early marrow evaluations for hematologic response at day 7 to 14 after induction are misleading and may lead to overtreatment. Marrow evaluation is not recommended until recovery of blood counts, which usually occurs 4 to 6 weeks after therapy. Cytogenetics is usually normal by this point, but molecular remission often requires at least 2 cycles of consolidation. All consolidation regimens have high cumulative doses of cardiotoxic agents. Therefore, cardiac function of the patient should be assessed before initiating each anthracycline- or mitoxantrone-containing consolidation cycle.

The goal of consolidation therapy for APL is the conversion of a morphologic and cytogenetic remission into a durable molecular remission. Data from the 2 sequential PETHEMA trials,^25–27 which produced the current risk model, were used to construct subsequent trials that intensify therapy for the high-risk groups. In the second PETHEMA trial (LPA 99), 15 days of ATRA (45 mg/m²) were added to each of 3 cycles of anthracycline consolidation. In the low-risk group, no difference was seen in rates of relapse (3%–6%) or 3-year disease-free survival (93%–97%) with the use of ATRA compared with a similar consolidation without ATRA in trial LPA 94.²⁷ For the intermediate-risk group, the relapse rate was 2.5% versus 14% in the historic control, with a 3-year disease-free survival of 97% versus 82%. Although the addition of ATRA to the high-risk group improved rates of relapse and disease-free survival, the rates of 21% and 77%, respectively, indicate that there is room for improvement. The new AIDA-2000 trial of the Italian GIMEMA group has confirmed that inclusion of ATRA in consolidation significantly improves outcome.³⁶

In the French APL 2000 trial, which randomized Ara-C with daunorubicin consolidation without ATRA, the low- and intermediate-risk groups had a 2-year disease-free survival rate of 93% with Ara-C versus 77% for the group without.²⁸ Thus the outcomes for consolidation with anthracycline plus ATRA or anthracycline plus Ara-C are comparable for patients with intermediate-risk APL. For all high-risk patients younger than 60 years, the addition of Ara-C seems to offer some benefit, with a 2-year disease-free survival rate of 89%.²⁹

In the new nonrandomized Spanish trial (LPA 2005), Ara-C (1 g/m²/d for 4 days) was added to the combination of ATRA (45 mg/m²/d for 15 days) and idarubicin (7 mg/m²/d for 4 days) for consolidation in high-risk patients younger than 60 years. The high-risk patients treated with Ara-C had a significant lower relapse rate at 3 years (11%) than those treated without Ara-C (21%).^36,37 The LPA 2005 trial also began to investigate reducing toxicity during consolidation therapy in low- and intermediate-risk patients through a dose reduction of mitoxantrone (from 10 mg/m²/d for 5 days to 10 mg/m²/d for 3 days) and a small dose reduction of idarubicin (from 7 mg/m²/d for 4 days to 5 mg/m²/d for 4 days). Results showed that lowering the dose of mitoxantrone reduced toxicity and decreased length of hospital stay while maintaining antileukemic activity.

NAIT also investigated decreasing toxicity during consolidation through incorporating ATO into the consolidation schema directly after remission occurred.^30,38 This trial randomized patients to receive either 2 courses of 25 days of ATO (5 days per week for 5 weeks) immediately after experiencing complete remission followed by a standard postremission regimen with 2 more courses of ATRA plus daunorubicin, or only 2 courses of ATRA plus chemotherapy. Those who were treated with ATO, particularly those with high-risk disease, showed significantly better event-free and overall survival rates.^30,38 The overall outcomes do not seem to be superior to the less-complex consolidation schedules in either of the 2 most recent European trials for patients in the low- and intermediate-risk groups, but inclusion of ATO appeared to offer improved survival for high-risk disease. The consolidation period in the NAIT protocol is longer and maybe difficult for some patients. The potential benefits of ATO use in consolidation may be that it confers a lesser risk of long-term cardiovascular complications and perhaps a lower risk of secondary myelodysplasia.

For patients with high-risk disease, the panel suggests including Ara-C as used in the French APL 2000 trial²⁸ and PETHEMA LPA 2005,³¹ or 2 cycles of ATO followed by 2 additional cycles of standard chemotherapy as used in the U.S. Intergroup trial for consolidation.^30,38 When using an Ara-C containing regimen, dose adjustments of Ara-C may be needed for older patients or those with renal dysfunction.^29,30 For low-risk patients, the panel has positioned the revised PETHEMA LPA 2005 regimen slightly higher than either the French APL 2000 or the NAIT because of ease of administration and decreasing toxicity. However, all 3 regimens yield excellent results. For the intermediate-risk group, outcomes are very similar as long as one regimen is followed consistently from induction through consolidation.

In patients who received ATRA and ATO for induction therapy but could not tolerate anthracyclines, the reported trials continued with repeated cycles of ATRA and ATO after induction. Based on these reports, the panel recommends consolidation therapy with ATO (0.15 mg/kg intravenous daily for 5 days per week every other month for 4 cycles) and ATRA (45 mg/m² in divided doses daily orally during 1 week monthly for 7 cycles).

Postconsolidation or Maintenance Therapy

After consolidation therapy, patients are assessed for molecular remission using RT-PCR techniques on bone marrow samples. For those who are PCR-negative, a 1- to 2-year course of ATRA maintenance therapy has been recommended, which may be combined with 6-mercaptopurine (6MP) and methotrexate.

The recommendations for maintenance ATRA arose from several early trials that showed superior relapse-free survival for patients receiving ATRA alone or in combination as maintenance. The French APL 93 trial showed decreased relapse rates at 2 years for ATRA (21%), 6MP, and methotrexate (13%), and ATRA plus 6MP and methotrexate (8%), versus no maintenance (35%).²² Results of a longer follow-up of the APL 93 trial show a beneficial effect of maintenance treatment with intermittent ATRA and continuous 6MP and methotrexate, with an additive effect of the 2 modalities.³⁹ The study data showed that maintenance treatment reduced the incidence of early relapses without increasing the incidence late relapses. Patients with a WBC count higher than 5000/μL benefitted most from maintenance therapy. The relapse rate dropped from 68.4% with no maintenance to 20.6% with combined ATRA plus chemotherapy maintenance. In patients with WBC counts lower than 5000/μL, the 10-year relapse rate decreased from 29.2% without maintenance to 11.5% with combined maintenance.³⁹

The first U.S. Intergroup trial showed superior disease-free survival among patients receiving maintenance ATRA compared with no maintenance.^21,22 However, data from the All-Trans Retinoic Acid and Idarubicin (AIDA) trial suggest that patients who are molecularly negative at end of consolidation experience no benefit from maintenance.⁴⁰

As consolidation regimens have evolved to incorporate ATRA or ATO into consolidation, the role of maintenance chemotherapy is less clear, particularly for patients with low-risk disease who achieve a molecular remission at the end of consolidation. An international cooperative group trial (SWOG 0521) examined the need for maintenance in low-/intermediate-risk patients. The studies showing benefit for maintenance were performed before ATRA was used for consolidation.^21,22 The conflicting data indicate that benefit of maintenance depends on prior induction and consolidation therapy. Therefore, it would be appropriate to use maintenance in conjunction with protocols in which it has been shown to confer benefit.

At completion of consolidation, RT-PCR should be performed on a marrow sample to document molecular remission. Subsequent monitoring of patients with PCR can be performed using peripheral blood, although the technique is more sensitive and may show earlier signs of relapse when performed on a marrow sample. Monitoring is recommended at a minimum of every 3 months for 2 years during maintenance to detect molecular relapse in patients with intermediate- and high-risk disease. Clinical experience indicates that risk of relapse is low in patients with low-risk disease experiencing molecular remission at completion of consolidation, and that monitoring may not be necessary outside the setting of a clinical trial. At the current level of resolution, a change from PCR-negative to -positive should be confirmed with bone marrow in a reliable laboratory within 4 weeks. If molecular relapse is confirmed by a second positive test, intervention should be strongly considered. If the second test was negative, maintenance therapy and frequent monitoring every 2 to 3 months for an additional 2 years is strongly suggested to assure that the patient remains negative. Testing should be performed in the same laboratory to maintain a consistent level of sensitivity. In patients who develop cytopenias and have a negative RT-PCR, a marrow analysis is recommended to evaluate for new cytogenetic abnormalities, because secondary MDS and AML have occurred after APL therapy.

Management of Relapse

ATO is the recommended therapy for patients who remain molecularly positive at completion of consolidation or subsequently show molecular relapse. As a single agent, it produced complete remission rates of 80% to 90% in patients with hematologic relapse, and molecular remissions in 70% to 80% of those patients. ATRA and ATO seem to be synergistic, and this combination could be considered in patients who did not receive ATRA during consolidation.^32–34 However, a small randomized study of 20 patients with relapsed APL all previously treated with ATRA-containing chemotherapy showed no improvement in response after ATRA was added to ATO compared with ATO alone.⁴¹

A small percentage of relapsed APL has a CNS component.^42,43 Therefore, for patients who are in second morphologic remission, the panel strongly recommends intrathecal therapy as CNS prophylaxis.

Patients who experience a molecular remission after second-line therapy should be considered for autologous HSCT if they have no contraindications to high-dose therapy. In a recent retrospective study by the European APL Group, patients who received a PCR-negative autograft had a 7-year overall survival rate of 75%, compared with 52% in those receiving an allogeneic HSCT.⁴⁴ The differences in survival are accounted for by a higher treatment-related mortality in the allogeneic group, which influences the guideline recommendations to reserve allogeneic transplantation for those who have persistent disease despite salvage therapy. For patients in second complete remission who have contraindications to HSCT, continued therapy with ATO for 6 cycles is recommended in the absence of a clinical trial.

Supportive Care

Specific supportive care issues should be considered when treating patients with APL. Therapy for APL is often associated with a constellation of symptoms and physiologic abnormalities, including fluid retention, dyspnea, episodic hypotension, pulmonary infiltrates, and pulmonary or pericardial effusions now referred to as differentiation syndrome. Early in treatment involving either ATRA or arsenic as single agents or in combination, patients may begin to develop evidence of differentiation syndrome. These patients develop fever, fluid retention, and rapidly rising WBC counts to greater than 10,000/mcL. These patients should be closely monitored for hypoxia and the development of pulmonary infiltrates or pleural effusion. Differentiation syndrome and hemorrhage are the leading causes of death during induction. Early recognition and prompt initiation of steroids are the keys to dealing with this complication. In some studies, low mortality and morbidity rates were seen when corticosteroids were administered prophylactically in patients presenting with high WBC counts.^27,45

Kelaidi et al.⁴⁶ assessed the outcomes of patients with high WBC counts (> 10,000/mcL) enrolled in the APL 93 and APL 2000 trials. The factor in supportive care that had an effect on early induction deaths was shown to be the use of dexamethasone, 10 mg every 12 hours, beginning on day 1 for patients on APL 2000. The early death rate from differentiation syndrome dropped from 8 of 139 in APL 93 to 2 of 133 in APL 2000. The panel recommends giving prophylactic steroids to patients with a WBC count greater than 10,000/mcL to prevent differentiation syndrome. The panel has provided management recommendations in the Supportive Care section of the algorithm. For patients with WBC counts greater than 10,000/mcL or symptoms of differentiation syndrome, the panel recommends treatment with dexamethasone, 10 mg, twice daily for 3 to 5 days, and then tapering the dose over 2 weeks. ATRA may need to be suspended during the initial acute symptomatic period but may be restarted when symptoms improve. Other factors that have been reported to increase the risk of differentiation syndrome are a high body mass index and age older than 40 years.

Leukapheresis is not recommended in the routine management of patients with APL with high WBC counts because of the difference in leukemia biology. However, in a life-threatening case with leukostasis that is not responsive to other modalities, leukapheresis can be considered with caution.

Because coagulopathy is common in patients with APL, it is good clinical practice to screen for this problem with prothrombin time (PT), partial thromboplastin time (PTT), and fibrinogen as part of the initial workup and before any invasive procedure. Clinical coagulopathy is managed with aggressive transfusion support to maintain platelets at 50,000/mcL or greater, fibrinogen replacement with cryoprecipitate and frozen plasma to maintain a level of 150 mg/dL, and maintaining levels of PT and PTT close to normal. Patients with clinical coagulopathy must be monitored daily until the condition resolves.

ATO therapy may prolong the QT interval, making patients susceptible to ventricular arrhythmias. Therefore, before each cycle of therapy, an electrocardiogram (ECG or EKG) is recommended to assess prolonged QT interval. Routine monitoring weekly during therapy is also suggested for older patients. Serum electrolytes should also be monitored before and during therapy to maintain electrolytes (calcium ≥ 9.0, potassium ≥ 4.0, magnesium ≥ 1.8) in the upper normal range. Other drugs that prolong the QT interval should be avoided during ATO to minimize the risk of cardiac arrhythmias. Patients with an absolute QT interval greater than 500 m/s should be reassessed.

A 4% incidence of CNS relapse was seen in patients with a WBC count greater than 10,000/mcL in the French APL 93 trial. In APL 2000, the high-risk population received 5 doses of intrathecal chemotherapy using a combination of methotrexate, Ara-C, and steroids on count recovery after induction. They also received a higher dose of Ara-C (2 g/m²) during the second consolidation cycle compared with 1 g/m² in APL 93. No CNS relapses occurred in APL 2000 versus 5 cases in APL 93. Although the original trial used high-dose Ara-C as the second consolidation, some investigators suggest using it earlier, particularly in patients who are not receiving intrathecal therapy for prophylaxis.

Patients Younger Than 60 Years

Induction Therapy: Standard induction regimens are appropriate for patients younger than 60 years who have no antecedent hematologic disease, such as MDS or treatment-related secondary AML. These regimens are based on a backbone of Ara-C and an anthracycline and have changed little in the past 25 years. Historically, in most large cooperative group trials, daunorubicin has been the most commonly used anthracycline at doses of 45 to 60 mg/m². However, idarubicin, which has a longer intracellular retention time, is used at doses of 12 mg/m² for 3 days and has had comparable remission rates, with fewer patients requiring additional therapy at day 15 to experience remission. Complete remission rates for patients aged 50 years or younger have consistently ranged from 60% to 70% in most large cooperative group trials of infusional Ara-C and anthracycline. A recent ECOG study reported a significant increase in complete remission rates (71% vs. 57%) and overall survival using daunorubicin, 90 mg/m², for 3 days versus 45 mg/m² for 3 days in patients younger than 60 years.⁴⁸ The survival benefit was restricted to patients with favorable- and intermediate-risk cytogenetic profiles and those younger than 50 years.

A European trial comparing idarubicin, 12 mg/m², for either 3 or 4 days and daunorubicin, 80 mg/m², for 3 days in patients between ages 50 and 70, showed complete remission rates of 83% in patients treated with idarubicin, 12 mg/m², for 3 days versus 70% in those treated with daunorubicin (P = .024).⁴⁹ No difference was seen in relapse rate, event-free survival, or overall survival among the arms. Infusional Ara-C for 7 days combined with either idarubicin or escalated daunorubicin is a category 1 recommendation.

For patients with impaired cardiac function, other regimens that combine a nonanthracycline (e.g., fludarabine⁵⁰ or topotecan⁵¹) with Ara-C have been published.

Two large cooperative clinical trials explored dose-intensive Ara-C therapy during induction. An Australian Leukemia Study Group trial⁵² randomized 301 patients younger than 60 years to receive either high-dose Ara-C (3 g/m² every 12 hours on days 1, 3, 5, and 7 for a total of 24 g/m²) or standard Ara-C therapy (100 mg/m²/d for 7 days through continuous infusion). Both arms received daunorubicin (50 mg/m² on days 1–3) and etoposide (75 mg/m²/d for 7 days). The complete remission rates were equivalent in both arms (71% and 74%, respectively), although treatment-related morbidity and mortality were higher in the high-dose arm. Patients in both arms received only 2 cycles of standard-dose Ara-C, daunorubicin, and etoposide for consolidation. Median remission duration was 45 months for the dose-intensive arm compared with 12 months for the standard treatment arm.

A SWOG study⁵³ randomized patients to receive either high-dose Ara-C (2 g/m² every 12 hours for 6 days for a total of 24 g/m²) or standard-dose Ara-C (200 mg/m²/d for 7 days). Patients in both arms also received daunorubicin (45 mg/m²/d for 3 days). Patients receiving high-dose Ara-C induction therapy received a second high-dose cycle for consolidation, and patients in the standard-dose treatment arm were randomized to receive either 2 cycles of standard-dose Ara-C consolidation or 1 cycle of high-dose Ara-C plus daunorubicin consolidation. The complete response rates were again equivalent: 55% for the high-dose Ara-C treatment arm compared with 58% for the standard-dose arm in patients younger than 50 years; and 45% for high-dose Ara-C versus 53% for standard-dose therapy for patients 50 to 65 years of age. Patients in the high-dose Ara-C arm experienced higher treatment-related mortality (12% vs. 5%) and neurologic toxicity.

Younger patients who received both high-dose Ara-C induction and consolidation in the SWOG trial had the best survival and disease-free survival rates (52% and 34%, respectively) at 4 years compared with those receiving standard induction and consolidation (34% and 24%, respectively) or standard induction with high-dose consolidation (23% and 14%, respectively). However, the percentage of patients experiencing a complete remission who did not proceed to consolidation was twice as high in the high-dose Ara-C induction arm. Risks for neurotoxicity and renal insufficiency are increased with high-dose Ara-C, and therefore renal and neurologic function should be closely monitored. In a CALGB trial,⁵⁴ patients who received induction therapy with standard-dose Ara-C and daunorubicin or 3 to 4 courses of high-dose Ara-C consolidation also had a 44% probability of a 4-year disease-free interval, with similar rates of neurotoxicity and treatment-related mortality.

Because the overall survival for the high-dose arm in the SWOG trial (high-dose Ara-C induction and 2 cycles of high-dose Ara-C consolidation) is comparable to the CALGB trial with standard-dose infusional Ara-C induction and 4 cycles of high-dose Ara-C consolidation, the use of high-dose Ara-C in induction outside a clinical trial remains controversial. The decision to use high-dose Ara-C versus standard-dose Ara-C for induction might be influenced by consolidation strategies; fewer high-dose consolidation cycles may be needed for patients induced with high-dose Ara-C or for patients who will undergo early autologous stem cell transplantation. Although the remission rates are the same for standard- and high-dose Ara-C, 2 studies have shown more rapid marrow blast clearance after 1 cycle of high-dose therapy and a disease-free survival advantage for patients aged 50 years or younger who received the high-dose therapy.⁵⁵ No data are available for doses of more than 60 mg of daunorubicin or 12 mg of idarubicin with high-dose Ara-C. High-dose Ara-C plus an anthracycline as induction therapy is considered a category 2B recommendation for patients younger than 60 years.

With either high- or standard-dose Ara-C–based induction for younger patients, between 20% and 45% of these patients will not experience remission. In a recent report of 122 patients treated with high-dose Ara-C and daunorubicin, the remission rates were strongly influenced by cytogenetics, with complete remission rates of 87%, 79%, and 62% for favorable-, intermediate-, and poor-risk groups, respectively.⁵⁶

Patients with antecedent hematologic disease or treatment-related secondary leukemia are considered poor-risk unless they have favorable cytogenetics, such as t(8;21), inv(16), t(16;16), or t(15;17). In addition, patients with unfavorable karyotypes such as –7, –5, 11q23 abnormalities or complex cytogenetic abnormalities are also considered poor-risk and treated similarly. This group of patients should be enrolled in a clinical trial (incorporating either chemotherapy or low-intensity therapy), if available, because only 40% to 50% of these patients experience complete remission with standard induction therapy, and response durations are short. In addition, HLA testing should be performed promptly in those who may be candidates for either fully ablative or reduced-intensity allogeneic HSCT from a sibling or an unrelated donor, which constitutes the best option for long-term disease control. Because patients with antecedent myelodysplasia or treatment-related leukemia have a decreased probability of experiencing remission through induction chemotherapy, transplantation without induction chemotherapy may be considered for those who have an available sibling donor and a relatively low percentage marrow involvement. In an EBMT (European Blood and Marrow Transplantation) trial,⁵⁷ patients with high-risk myelodysplasia or AML evolving from myelodysplasia who underwent allogeneic transplantation without prior Ara-C–based chemotherapy had a 34% 3-year disease-free survival rate. Patients who underwent antecedent chemotherapy and experienced a complete remission had a 45% disease-free survival rate, compared with 10% for patients whose disease did not respond to chemotherapy before transplantation. An alternative strategy for patients with antecedent myelodysplasia who have not received a hypomethylating agent would be a trial of either decitabine or azacytidine while a rapid donor search is initiated.

Postinduction Therapy: To judge the efficacy of the induction therapy, a bone marrow aspirate and biopsy should be performed 7 to 10 days after completion of induction therapy. In patients who have received standard-dose Ara-C induction and have residual blasts without hypoplasia, additional therapy with standard-dose Ara-C and anthracycline should be considered. For those with significant residual blasts or clear-cut induction failure, escalation to high-dose Ara-C with or without an anthracycline is the most common salvage strategy. Other options include an allogeneic HSCT if a matched sibling or an alternative donor has been identified, or participation in a clinical trial. For patients whose clinical condition has deteriorated to the point that active treatment is no longer appropriate, best supportive care should be continued. If the marrow is hypoplastic (defined as cellularity < 10%–20% and residual blasts < 5%–10%), additional treatment selection may be deferred until marrow recovery when the remission status can be assessed.

Patients initially treated with high-dose Ara-C and who have significant residual blasts 7 to 10 days after completion of chemotherapy are considered to have induction failure. Additional high-dose Ara-C is unlikely to induce remission. If a sibling or HLA-matched unrelated donor has been identified, an allogeneic HSCT may salvage 25% to 30% of patients with induction failure. If no donor is immediately available, patients should be considered for a clinical trial.

Occasionally, ALL therapy may be effective in patients with both myeloid and lymphoid markers at diagnosis (biphenotypic leukemia) if an AML induction regimen failed. Treatment decisions for patients with significant reduction without hypoplasia or those with hypoplasia are deferred until the blood counts recover and a repeat marrow is performed to document remission status. Response is then categorized as complete response or induction failure.

Postremission or Consolidation Therapy: Although successful induction therapy clears the visible signs of leukemia in the marrow and restores normal hematopoiesis in patients with de novo AML, additional therapy (consolidation) is needed to first reduce the residual abnormal cells to a level that can be contained by immune surveillance.

Since 1994, multiple (3–4) cycles of high-dose Ara-C therapy have been the nonprotocol standard consolidation regimen for patients younger than 60 years with either good- or intermediate-risk cytogenetics. This therapy is based on a CALGB trial comparing 100 mg/m², 400 mg/m², and 3 g/m² doses.⁵⁴ The 4-year disease-free survival rate for patients receiving 3 g/m² was 44%, with a 5% treatment-related mortality rate and a 12% incidence of severe neurologic toxicity. Although the initial report did not breakdown disease-free survival rates according to cytogenetic subgroups, subsequent analysis showed disease-free survival rates of 60%, 30%, and 12% for good-, intermediate-, and poor-risk cytogenetics, respectively, in patients treated with high-dose Ara-C consolidation. These outcomes are similar to those seen in the high-dose Ara-C treatment arm of the SWOG trial, in which patients received high-dose Ara-C induction and 2 cycles of high-dose Ara-C consolidation.⁵³ The CALGB trial also included maintenance chemotherapy after consolidation; however, only a small fraction of patients received maintenance, and subsequent trials have not included this aspect.

Other options for consolidation strategies include one or more cycles of high-dose Ara-C followed by autologous HSCT or allogeneic HSCT from sibling or unrelated donors. When choosing among these options, decisions are influenced by the 1) expected relapse rate with high-dose Ara-C consolidation chemotherapy, 2) the additional morbidity and mortality associated with the transplant procedure, which in turn are strongly influenced by patient-specific comorbidity, and 3) salvage options. Factors such as patient age, comorbid conditions, and features of the disease at diagnosis, including elevated leukocyte counts (≥ 50,000/mcL) or number of cycles of induction to achieve remission, should play a role in choosing a consolidation strategy, as should issues regarding fertility and salvage options. Patients in whom 2 cycles of chemotherapy are required before remission are at very high risk for relapse and should be considered for either clinical trial or allogeneic transplant as initial consolidation whenever possible.

Previous guidelines have used cytogenetics as the major defining criteria for risk of relapse. This version of the NCCN Guidelines attempts to incorporate emerging data on the influence of mutations in specific genes, such as c-KIT, FLT3, CEBPA, and NPM1, on subsets of patients within a cytogenetic category. When the data on the CALGB trial was analyzed by cytogenetic groups, the dose-intensive Ara-C consolidation produced relapse-free survival rates of 50% to 60% overall in patients with inv(16) or t(8;21), with cumulative treatment-related mortality rates of 8% to 10%.^6,54,56 The data were subsequently reanalyzed to include information on the effects of c-KIT mutations, which occur in 20% to 30% of these favorable-risk patients, on relapse rates and outcomes. In patients with inv(16), the relapse rate increased from 29% for wild-type c-KIT to 56% for patients with c-KIT mutations, which also translated into a decreased overall survival at 2 years of 56% versus 76%. In patients with t(8;21), the risk of relapse increased from 30% to 70% and the overall survival decreased to 42% at 2 years.⁴⁸

In the EORTC/GIMEMA trial comparing autologous and allogeneic HSCT, disease-free survival rates for patients with good-risk cytogenetics (t(8;21) or inv(16)) were 66% and 62%, respectively.⁵⁸ Treatment-related mortality rates were 6% and 17%, respectively. Small single-institution studies have reported a disease-free survival rate of 88% for patients undergoing autologous HSCT.⁵⁹ These data suggest that in this subgroup of patients with AML, allogeneic HSCT may be restricted to salvage therapy or those with c-KIT mutations.

Therefore, the panel recommends consolidation therapy for patients with better-risk cytogenetics involve either 3 to 4 cycles of high-dose Ara-C (category 1) followed by maintenance therapy (category 2B), or 1 to 2 cycles of dose-intensive Ara-C followed by autologous HSCT (category 2B) for patients with core binding factor leukemia lacking a c-KIT mutation. However, patients who have c-KIT mutations have a more similar outcome to those with intermediate-risk karyotype and should be considered for either clinical trials targeted toward the molecular abnormality, or consolidation strategies similar to those used in the intermediate-risk group. A carefully considered salvage plan with either sibling or unrelated donor HSCT should be an important part of the decision for these patients.

Panel members agreed that transplant-based options (either matched sibling or alternate donor HSCT, or 1 to 2 cycles of dose-intensive Ara-C followed by autologous stem cell transplantation) afforded a lesser risk of relapse and a somewhat higher disease-free survival compared with consolidation for most patients with intermediate-risk cytogenetics. In the EORTC/GIMEMA trial, the 4-year disease-free survival rates were 48.5% for allogeneic and 45% for autologous HSCT in patients with normal cytogenetics (with NN and –Y only).⁵⁸ Other options for this group include clinical trials or multiple courses (3–4) of high-dose Ara-C consolidation.⁶⁰ Alternative regimens incorporating intermediate doses of Ara-C (1.5–2.0 g/m²) are also acceptable in this group. Comparable 5-year disease-free survival rates were reported in patients younger than 60 years with normal karyotype after either 4 cycles of intermediate- or high-dose Ara-C (41%) or autologous HSCT (45%).⁶⁰

However, in the past 3 to 5 years, experts have learned that “normal” cytogenetics encompasses several molecular mutations with divergent risk behaviors. A large German trial¹⁶ showed additional molecular prognostic markers for patients with “normal” karyotype. The presence of an isolated mutant NPM1 cytoplasmic shuttle protein improves prognosis to a level comparable to that of patients with better-risk cytogenetics. For this subset of patients, therapy with multiple cycles of high-dose Ara-C is a reasonable option, and transplantation should be reserved until relapse. In contrast, patients with an isolated FLT3-ITD mutation and normal karyotype have a similar outlook to those with poor-risk cytogenetics¹⁴ and should be considered for a clinical trial or early allogeneic transplantation. Preliminary trials incorporating FLT3 inhibitors have been disappointing. The panel strongly recommends clinical trials as standard therapy for patients with poor prognostic features, which include FLT3 abnormalities in the setting of otherwise normal karyotype, high WBC count (> 50,000/mcL) at diagnosis, or 2 cycles of induction therapy needed to produce complete remission.

Sibling allogeneic HSCT produced a 43% disease-free survival rate in a group of patients with poor-risk cytogenetics in the EORTC/GIMEMA trial, with the International Bone Marrow Transplant Registry reporting similar outcomes for unrelated donor recipients. The outcome for autologous HSCT was comparable to chemotherapy, with an 18% disease-free survival rate.⁵⁸ The panel uniformly endorsed allogeneic sibling HSCT or HLA-matched unrelated donors (including cord blood) or clinical trial as consolidation therapy for patients with poor-risk cytogenetics or molecular abnormalities, or those with therapy-related AML or prior myelodysplasia. Another option for this group is 1 to 2 cycles of high-dose Ara-C–based consolidation followed by autologous HSCT if no allogeneic transplant option is available (category 2B).

Patients Aged 60 Years and Older

Induction Therapy: The creation of separate algorithms for patients older than 60 years recognizes the poor outcomes seen with standard Ara-C and an anthracycline, particularly in those aged 75 years or older, or those aged 60 to 75 years with significant comorbidities or an ECOG performance status greater than 2.^47,51 Treatment-related mortality frequently exceeds any expected transient response in this group. The British MRC AML 14 trial randomized 217 older patients unfit for chemotherapy to receive either 20 mg of Ara-C through subcutaneous injection twice daily for 10 consecutive days each month, or hydroxyurea.⁶¹ Ara-C produced a complete remission in 18% of the patients and a survival benefit in those with favorable or normal karyotype. Even this “low-intensity” approach had a 30-day mortality rate of 26%.

Most patients with leukemia seen by hematologists and oncologists consist of the “robust” baby boomers aged 60 to 75 years. Many of these patients will present with pancytopenia with modest marrow infiltration (20%–40%). If they are clinically stable, their cytogenetic prognostic group is helpful to know when deciding treatment.⁶²

For older patients (> 60 years) with AML, the panel recommends using performance status, adverse features (e.g., unfavorable cytogenetics and therapy-related AML or prior MDS), and comorbid conditions when selecting treatment, rather than age alone.

Older adults with good functional status (ECOG score 0–2), minimal comorbidity, and good-risk cytogenetic or molecular mutations, may benefit from standard therapies regardless of age. A reasonable treatment regimen for these patients is 7 days of continuous infusion standard-dose Ara-C (100–200 mg/m²/d) along with 3 days of anthracycline. Patients older than 75 years with significant comorbidities usually do not benefit from conventional chemotherapy treatment. However, the rare patient with good or normal karyotype and no significant comorbidities may benefit from conventional chemotherapy treatment. For patients with normal karyotype, the remission rates are 40% to 50% with Ara-C combined with idarubicin, daunorubicin, or mitoxantrone. The study by Pautas et al.⁴⁹ published as a 2007 ASH Annual Meeting abstract showed that idarubicin treatment, compared with high doses of daunorubicin up to 80 mg/m², yields higher complete response rates and more complete responses after one course. Löwenberg et al.⁶³ showed that the complete remission rates and 2-year overall survival in patients between 60 and 65 years of age treated with daunorubicin, 90 mg/m², are also comparable to outcomes in patients treated with idarubicin (12 mg/m²),⁶⁴ and that the benefit in overall survival in the high-dose daunorubicin group was observed only in patients younger than 65 years and those with core binding factor leukemia.⁶³

Several promising therapeutic leads have emerged for the treatment of elderly patients with AML deemed unfit to undergo intensive chemotherapy.

The study by Burnett et al.⁶¹ established low-dose Ara-C as an accepted standard of care in elderly patients with AML who are unfit for chemotherapy. The study showed that low-dose Ara-C therapy was associated with a higher complete remission rate (18% vs. 1%) and longer overall survival compared with hydroxyurea.

Fenaux et al.⁶⁵ compared 5-azacytidine with conventional care (e.g., best supportive care, low-dose Ara-C, intensive chemotherapy) in a study evaluating treatment options in patients with high-risk MDS using FAB criteria. However, using the 2008 WHO classification, 113 study patients (32%) fulfilled criteria for AML with a percentage of marrow blasts between 20% and 30%. In the subgroup of patients with RAEB-T, a significant survival benefit was seen with 5-azacytidine compared with best supportive care or low-dose Ara-C, which was exceeded by 9.5 months in both groups. A better overall survival was also seen with 5-azacytadine compared with standard induction therapy, but the difference was not significant because of small sample size.

Decitabine (20 mg/m²/d for 5 days) has been used as remission induction therapy for older patients with AML.⁶⁶ The complete remission rate was 29% (including 3 of 10 patients with poor-risk cytogenetics). Both azacytidine and decitabine are approved by the FDA as therapy for myelodysplasia. The purine nucleoside analog clofarabine, which was approved by the FDA for relapsed pediatric ALL, has also received recent attention as a possible induction therapy for older patients. Results of a single-agent phase II trial showed that single-agent clofarabine was associated with an overall remission rate of 45%, with a 30-day mortality of 10%. In a group of 115 patients with a median age of 71 years, median overall survival was 59 weeks for responders versus 41 weeks for nonresponders. Confirmatory phase III trials are in progress for clofarabine.⁶⁷

The panel has included subcutaneous Ara-C, 5-azacytidine, and decitabine as low-intensity treatment options, and clofarabine as an intermediate-intensity treatment option for patients with AML aged 60 years or older. Best supportive care includes red cell and platelet transfusions to alleviate symptoms of anemia and thrombocytopenia; prophylactic antibiotic and antifungal drugs to reduce the risk of infection; and hydroxyurea for management of leucocytosis.

Older adults with newly diagnosed AML with an ECOG score between 0 and 2 with or without adverse features (e.g., therapy-related AML/prior MDS or unfavorable cytogenic or molecular markers) may be managed through either enrollment in a clinical trial, treatment with standard infusional Ara-C and anthracycline, or low-/intermediate-intensity therapy.

Patients with an ECOG score greater than 2, significant comorbidity, or poor-risk cytogenetics are more likely to experience toxicity and less likely to benefit from standard induction chemotherapy. Therefore, the panel believes it is reasonable to offer these poor-risk patients either supportive care or enrollment in a clinical trial investigating novel agents or low-intensity therapy.

Postinduction Therapy: Similar to younger patients, older patients who undergo standard Ara-C/anthracycline induction therapy are evaluated with a bone marrow analysis 7 to 10 days after completion of chemotherapy, and categorized according to the presence of blasts or hypoplasia. Patients with significant cytoreduction without hypoplasia may receive standard-dose Ara-C with an anthracycline or anthracenedione. A repeat bone marrow analysis is performed in these patients and those with hypoplasia after induction to document the remission status. Because many older patients have some evidence of antecedent myelodysplasia, full normalization of peripheral blood counts often does not occur even if therapy clears the marrow blasts. Thus many phase I/II trials for AML in older patients include categories such as incomplete complete remission (CRi) for patients who have fewer than 5% marrow blasts but have mild residual cytopenia.

Many of the newer treatment strategies are designed to work more gradually, using agents that may allow expression of tumor suppressor genes (e.g., a methyltransferase inhibitor such as decitabine or 5-azacytidine) or through increasing apoptosis (e.g., histone deacetylators). Thus, success in these trials may be assessed using indirect measures such as hematologic improvement or decreased transfusion requirements, and survival without actually experiencing complete remission. In these trials, marrow examination frequently is not performed until completion of 1 or 2 cycles of therapy.

Patients who experience a complete remission (including CRi) with standard induction chemotherapy may receive further consolidation with these agents. The French Acute Leukemia (ALFA) 98 trial randomized patients experiencing remission to consolidation with either one additional course of standard Ara-C, 200 mg/m², for 7 days plus the anthracycline to which they had been randomized for induction (idarubicin, 9 mg/m², for 4 days, or daunorubicin, 45 mg/m², for 4 days) or 6 monthly courses at 1 day of anthracycline at the above doses and Ara-C, 60 mg/m², every 12 hours as a subcutaneous infusion at home for 5 days each month. Patients in the ambulatory arm had a better 2-year disease-free survival rate (28%) than those in the single intense consolidation arm (17%; P = .03). Overall survival, transfusion requirement, and days in hospital all favored the ambulatory arm.⁶⁸ Although the CALGB trial did not show an overall benefit for higher doses of Ara-C consolidation in older patients, a subset of patients with a good performance status, normal renal function, and a normal or low-risk karyotype might be considered for a single cycle of Ara-C (1.0–1.5 g/m²/d for 4–6 doses) without an anthracycline.

The role of myeloablative allogeneic HSCT is limited in older patients because of significant comorbidities, but interest in reduced-intensity allogeneic HSCT as consolidation therapy continues.^69,70 Case series and analysis of registry data have reported encouraging results, with 2-year overall survival rates of 40% to 60% and a nonrelapse mortality rate of 20% for patients undergoing transplant in remission. However, Estey et al.⁷¹ prospectively evaluated a protocol in which patients older than 50 years with unfavorable cytogenetics would be assessed for a reduced-intensity allogeneic HSCT. Of the 259 initial patients, only 14 ultimately underwent transplantation, because of illness, lack of donor, refusal, or unspecified reasons. The authors compared the results with matched cases receiving conventional-dose chemotherapy. Results suggested that reduced-intensity allogeneic HSCT was associated with improved relapse-free survival, and the authors concluded that this approach remains of interest. For this strategy to be better used, potential candidacy should be considered during induction, and unrelated donor options explored early.

The guidelines note that reduced-intensity allogeneic HSCT is considered an additional option for patients aged 60 years and older either as a postremission therapy for those experiencing a complete response to induction therapy or to treat induction failure only in patients with low-volume disease.

Postremission Surveillance and Salvage Therapy

CBC counts, including platelets, should be monitored every 1 to 3 months for the first 2 years after patients have completed consolidation, then every 3 to 6 months for a total of 5 years. Bone marrow evaluation is recommended only if the hemogram becomes abnormal, rather than as routine surveillance at fixed intervals, unless it is part of a research protocol.

A matched unrelated donor search (including cord blood) should be initiated for high-risk patients experiencing first complete remission who would be candidates for HSCT, or be considered concomitant with initiation of reinduction therapy during first relapse in appropriate patients.

Treatment strategies for relapse are categorized according to patient age. For patients younger than 60 years who have experienced a relapse, enrollment in a clinical trial is considered appropriate and is strongly preferred by the panel. If the relapse occurs after a “long” (> 12 months) remission, retreatment with the previously successful regimen is an option. If the relapse is detected when the tumor burden is low and the patient has a previously identified sibling or unrelated donor, salvage chemotherapy followed by allogeneic HSCT can be considered. Transplant should be considered only if the patient has experienced remission or in the context of a clinical trial.

The algorithm provides a list of some commonly used salvage regimens (page 298). The regimens included represent purine analog (e.g., fludarabine, cladribine) regimens that have shown remission rates of 30% to 40% in several clinical trials and those that have been used in the comparator arms in U.S. cooperative group trials in the past decade. The representative regimens include 1) cladribine, Ara-C, and granulocyte colony-stimulating factor (G-CSF) with or without mitoxantrone or idarubicin^72,73; 2) fludarabine, Ara-C, G-CSF with or without idarubicin^74,75; and 3) mitoxantrone, etoposide, and Ara-C (MEC).⁷⁶

Patients aged 60 years or older who are robust and wish to pursue treatment after relapse may be offered 1) therapy on clinical trial (strongly preferred option by the panel); 2) salvage chemotherapy followed by match sibling or alternate donor HSCT (transplant should be considered only if the patient has experienced remission or in the context of a clinical trial); or 3) repetition of the initial successful induction therapy only if they had a long initial remission (i.e., relapse > 12 months of induction therapy). Best supportive care is always an option for those who do not wish to pursue intensive treatment.

Supportive Care

Based on results of an ECOG study, growth-factor support may be considered for older patients after chemotherapy administration.⁷⁷ Recommendations on the use of cytokines for infection or slow marrow recovery is left to institutional policy. G-CSF should be discontinued for a minimum 7 days before assessing marrow because it may affect interpretation of pathology.

Leukocyte-depleted blood products should be used for transfusion. Cytomegalovirus (CMV) screening for potential HSCT candidates is left to institutional policies regarding provision of CMV-negative blood products to patients who are CMV-negative at diagnosis. Radiation of all blood products is advised to reduce the risk of graft-versus-host disease in all immunosuppressed patients.

Standard tumor lysis prophylaxis is hydration with alkalinization of the urine, allopurinol administration, or rasburicase treatment. Rasburicase is a genetically engineered recombinant form of urate oxidase enzyme, and should be considered as initial treatment in patients with rapidly increasing blast counts, high uric acid, or evidence of impaired renal function.

Patients who receive high-dose Ara-C must be closely monitored for changes in renal function. Renal dysfunction is highly correlated with increased risk of cerebellar toxicity. Patients must be monitored for nystagmus, dysmetria, and ataxia before each dose of high-dose Ara-C; patients exhibiting any neurologic signs should discontinue high-dose Ara-C, and all subsequent Ara-C therapy must be standard-dose. Patients who develop cerebellar toxicity should not be rechallenged with high-dose Ara-C in future treatment cycles.⁷⁸ High-dose Ara-C should also be discontinued in patients with rapidly rising creatinine caused by tumor lysis.

Decisions regarding use and choice of antibiotics to prevent and treat infections should be made by individual institutions based on the prevailing organisms and their drug resistance patterns. Cornely et al.⁷⁹ showed that in patients undergoing chemotherapy for AML or MDS, posaconazole significantly prevented invasive fungal infections compared with fluconazole or itraconazole, and improved overall survival of the patients. Outcomes with other azoles, such as voriconazole, echinocandins, or amphotericin B, may produce equivalent results. Azoles should not be given during anthracycline chemotherapy because they impair drug metabolism and can increase toxicity.