Current Treatment Options in Oncology

, Volume 12, Issue 4, pp 341–353

Treatment of Elderly Acute Myeloid Leukemia Patients


  • Gabriela Motyckova
    • Dana-Farber Cancer Institute
    • Dana-Farber Cancer Institute
Leukemia (Janice Dutcher, section editor)

DOI: 10.1007/s11864-011-0162-4

Cite this article as:
Motyckova, G. & Stone, R.M. Curr. Treat. Options in Oncol. (2011) 12: 341. doi:10.1007/s11864-011-0162-4

Opinion statement

Older patients with acute myelogenous leukemia (AML) fare much less well than younger patients with the same disease due to a combination of comorbidities and intrinsic disease resistance. Likely due to aging of the US population, the median age of AML patients at diagnosis has increased from 68 to 72 years. AML is a heterogeneous disease, particularly in older patients, making therapeutic decisions challenging. Older patients who are ‘fit’ for intensive chemotherapy and would have a reasonable chance to benefit based on host and disease characteristics should receive standard induction chemotherapy with 7 days of continuous infusion of cytarabine and at least 60 mg/m2 daunorubicin daily for 3 days. Therapeutic options for patients who are not candidates for or are not likely to respond to intensive therapy include clofarabine, low intensity chemotherapy such as low dose cytarabine, hypomethylating agents, or investigational agents. For older AML patients in complete remission, post-remission or consolidation chemotherapy with repeat induction or modified high dose cytarabine may offer a small chance for long term disease-free survival. Selected older patients who achieve remission by any means should be considered for reduced-intensity stem cell transplantation which may offer improved chances of cure and survival compared with standard post-remission chemotherapy.


AML is a disease of older patients (median age of 72 at diagnosis) [1•]. Older adults with AML fair poorly compared to younger patients with the same disease due to a combination of disease characteristics and medical problems [2•, 3]. The definition of older AML patients varies, but generally would include those above the age of 65. Biological factors accounting for worse outcomes in older patients with AML include prior history of myelodysplastic syndrome (MDS), adverse cytogenetics, increased rate of multidrug resistance protein expression [3] and comorbidities [4, 5]. Fewer older patients are referred to cancer centers for treatment compared to younger patients [6] and many never receive specialized treatment for this disease [7]. As such, older AML patients are usually underrepresented in clinical trials which typically involved analysis of induction, consolidation and transplantation therapies.

Older AML patients respond less well to chemotherapy then younger patients as manifested by lower remission rates, shorter remission duration, increased rate of treatment-related toxicities and shorter overall survival [8]. Three and five year survival in older patients with AML is 10% and 5–8%, compared to 45% and 35% for younger patients, respectively. Thirty-day chemotherapy related mortality in the older patient is higher than in younger patients with AML; about 14–18% for patients with performance status 0–1, and up to 50–82% for those with performance status 2–3 [4]. Older patients who receive induction chemotherapy have complete remission (CR) rates and median survival times between 40% and 60% and 7–12 months, respectively, compared to younger patients whose CR rates are in the 75–80% range [3]. A retrospective study using the Surveillance, Epidemiology and End Results (SEER)-Medicare database in AML patients (probably more reflective of ‘reality’ than results from patients treated on clinical trials) reported a median survival of 3.9 months for all AML patients (with and without chemotherapy treatment) ages 65–74, 2.2 months in 75–84 year olds and 1.4 months in 84 or older patients [7]. Overall less than 10% of patients achieve long term survival [9]. These results indicate that more effective treatment strategies are needed for older AML patients. Moreover, there is no universally accepted approach to the treatment of this patient population.

Increased understanding of both patient heterogeneity regarding host and disease characteristics may inform the choice of most appropriate treatment. In some cases such treatment may include intensive therapy involving induction chemotherapy, consolidation and reduced-intensity stem cell transplantation, and in others less intense chemotherapy with hypomethylating agents or low-dose chemotherapy, investigational agents, or supportive care alone (Tables 1 and 2).
Table 1

Selected studies of older AML patients


Age median (Range)





Induction chemo-therapy



Daunorubicin 90 mg/m2 vs. 45 mg/m2 days 1–3 with cytarabine CI for 7 days, followed by cytarabine 1,000 mg/m2 IV BID days 1–6

No difference in OS or EFS; patients ages 60–65 had improvement in CR (73% vs. 51%) and EFS (29% vs. 14%)

Lowenberg et al., 2009 [27••]

Induction vs. palliative approach



Vincristine, daunorubicin, cytarabine, vs. “wait and see”, cytarabine, hydroxyurea

OS 13% vs. 0% (projected) at 2.5 years; median survival 21 vs. 11 weeks; CR 58% vs. 0%

Lowenberg, et al., 1989 [25]

Induction vs. low dose cytarabine



Rubidazone 100 mg/m2 for 4 days, cytarabine 200 mg/m2 for 7 days vs. cytarabine 10 mg/m2 Q12H for 21 days

CR 52% vs. 32%; PR 2% vs. 22%; failure 15% vs. 36%; early death 31% vs. 10%; OS and CR duration not different

Tilly et al., 1990 [26]


70 (50–82)


5-azacitidine 75 mg/m2/day or CCR

OS 24.5 vs. 16 months; 2-year OS 50% vs. 16%, respectively (P = .001).

Fenaux et al., 2010 [31•]


74 (61–87)


Decitabine 20 mg/m2 IV daily days 1–5, 4 week cycle, median 3 cycles (range 1–25 cycles)

Overall response rate 25%; CR 24%; Median OS 7.7 months; 30-day mortality 7%

Cashen et al. [32•]


71 (60–84)


Phase II; 30 mg/m2 IV days 1–5 every 28 or 28–42 days; subsequent dose reduction to 20 mg/m2 days 1–5 in some patients

Response rate 48%; CR 32%; CRi 16%

Burnett et al., 2010 [38]

CI continuous infusion; CR complete remission; PR partial response; CRi complete remission with incomplete count recovery; CRp complete remission with incomplete platelet recovery; EFS event-free survival; DFS disease-free survival; OS overall survival; CCR best supportive care, low-dose cytarabine or intensive chemotherapy

Table 2

Selected studies of reduced-intensity conditioning stem cell transplantation (RIC SCT) in older AML patients

Median age (range)


RIC Regimen



57–58 (17–74)

122 prospective

2 Gy TBI with or without preceding fludarabine

Overall 2-year survival 48%; DFS 44%.

Hegenbart et al., 2006 [48]

50 (18–70)

112 (both RIC and myeloablative) prospective

Busulfan and fludarabine

74% DFS at 3 years for RIC SCT in CR1 [49]

Shimoni et al., 2006 [49, 50]

49, 58, 62, 67 in 4 age categories (40–54; 55–59; 60–64; ≥65)

1,080 (545 with AML) retrospective


36% 2 year survival in the oldest patient category age ≥65

McClune et al., 2010 [51]

63 (60–71)

158 retrospective

Busulfan and fludarabine

For patients over 60, 2 year NRM 10% and relapse 54.6%; OS 46%; PFS 35%

Koreth et al., 2010 [52]


14 prospective

Fludarabine and melphalan

Improved RFS and survival

Estey et al., 2007 [54]

57 (50–76) in a group >50 years old

252 in a group >50 years old (401 RIC SCT total, all ages) retrospective

TBI +/− fludarabine, or fludarabine with either melphalan, busulfan or cy, or cy and thiotepa

For patients over 50, 2 year NRM 25% and relapse 42%

Ringden et al., 2009 [53]

RIC reduced intensity conditioning; OS overall survival; PFS progression-free survival; NRM non-relapse mortality; CR1 first complete remission; N number of patients; TBI total body irradiation; cy cyclophosphamide

Prognosis: molecular markers in older AML patients

Identification of molecular markers has been important in deciding postremission therapy including consolidation and stem cell transplantation in younger patients. Molecular markers, such as gene mutations or gene expression levels, could be employed as part of a strategy of using disease biology to predict response and thus help decide on the optimal therapy in older patients as well. However, in part because of the pan-resistant phenotype exhibited by older adults regarding response to available agents, molecular markers may have less utility in this patient population. The molecular markers with prognostic impact in all or some AML subgroups include mutations in FLT3, NPM1, MLL, WT1, c-KIT, and expression levels of BAALC, NM1, ERG, and CXCR4 [10, 11]. A recent study showed that FLT3 and NPM1 mutations, two of the best characterized markers, have similar frequencies in younger and older (above age 60) patients with AML [12]. FLT3 activating mutations include FLT3 internal tandem duplications (FLT3-ITDs) in about 30% of AML cases and confer a poor prognosis in all age groups [1113, 14•], including shorter disease-free survival (DFS) and overall survival (OS) in older AML patients [14•]. Mutations in NPM1 confer a good prognosis in adult AML patients [10, 11, 13, 15, 16•], as well as in patients age 60 or older who were found to have a 3-year OS of 35% vs. 8% in older patients without the mutation [16•]. Recently IDH mutations were also found to confer a poor prognosis with worse relapse-free and overall survivals in a study of AML patients ages 16–60 [17], decreased disease-free survival in adult AML patients of all ages with IDH1 mutations [18], and decreased CR rates in older patients with IDH2 mutations [18]. Low BAALC and ERG expression levels were found to be associated with better outcome in older AML patients and distinct gene and microRNA expression profiles [19•]. Low BAALC expression was associated with mutated NPM1, wild-type CEBPA, and low ERG expression level. Patients with low BAALC expression had better CR rate (86% vs. 54%) and longer three-year DFS (19% vs. 12%) and three-year OS (29% vs. 10%) compared to higher BAALC expressing AML patients [19•]. Low ERG expression showed a non-statistically significant trend towards improved CR rates of (76% vs. 65%), but a significant difference in a three-year DFS (18% vs. 14%) and three-year OS (24% vs. 15%) compared to older AML patients with high ERG expression [19•].

That identification of any one or combination of mutations will be used to choose or guide AML therapy in older adults seems likely in the future. More studies and those with targeted therapies will likely be upcoming to help understand and take advantage of the role of molecular markers in AML patients.

Prognosis: cytogenetics in older AML patients

Pretreatment cytogenetics in older AML patients constitute one of the most important prognostic factors [8, 20, 21]. The rate of adverse cytogenetics changes increases with age [4]. Analysis of pretreatment cytogenetics in older AML patients undergoing induction and consolidation chemotherapy (induction with two courses of 12 mg/m2 idarubicin on days 1 and 3, 100 mg/m2 cytarabine continuous infusion on days 1–5 and 100 mg/m2 etoposide on days 1 and 3, with or without all-transretinoic acid (ATRA), followed by consolidation with modified dose of cytarabine, mitoxantrone with or without ATRA, and then a further course of intensive or maintenance therapy) was able to stratify patients into three prognostic subgroups with low risk, standard risk and high risk karyotype with CR rates of 80%, 53%, and 19% and a median overall survival of 26.4, 12.5, and 5.1 months, respectively [20]. The presence of a monosomal karyotype (two or more monosomies, or one monosomy and chromosomal aberrations) is also more common with advancing age and is present in 20% of AML patients over the age of 60. Monosomal karyotype confers an extremely poor prognosis with a reported 4-year OS of only 1% compared to 10% in older AML patients without such cytogenetics [9].

Predictive models: who benefits from intensive induction therapy?

Intensive induction chemotherapy with cytarabine in combination with an anthracycline or anthracycline-like agent, has been and remains the mainstay of initial therapy in younger adults with AML. But is such therapy the optimal choice for older adults who are known to have a lower CR rate, a higher toxic death rate, and a low chance for long term disease-free survival? Several studies presented analyses and models for chemotherapy responsiveness and outcomes in older AML patients [22, 23]. For example a multivariate analysis showed that complex cytogenetics (>5 aberrations) age, male sex, white blood cell count (wbc) and bone marrow blast percentage were significant variables predicting disease-free and overall survival, with increasing age conferring worse prognosis [22]. Another multivariate analysis of AML patients over the age of 60 in two UK trials reported that cytogenetics, wbc, age, AML type and performance status significantly affect survival at 1 year [23]. Factors important in therapy responsiveness and survival were analyzed in a study of older AML patients undergoing induction chemotherapy [24•]. An increased 8-week mortality was associated with age over 80 together with complex karyotype, poor performance status (ECOG 2–4) and elevated creatinine [24•].

Therapy: to induce or not?

There have been few studies which prospectively compare standard and less intensive treatment in older AML patients (Table 1). A prospective study of patients over 65 year of age compared immediate intensive-induction chemotherapy versus “wait and see” supportive care approach and mild cytoreductive chemotherapy for relief of progressive AML-related symptoms [25]. The study showed poor outcomes for patients on the supportive care or low dose cytoreductive therapy arm compared to patients receiving intensive chemotherapy (overall survival 11 vs. 21 weeks), with a projected survival at 2 years of 0% vs. 13%, respectively. Among patients receiving intensive therapy, 58% entered CR versus none on the supportive care/low intensity therapy arm. There was no significant difference in time spent in a hospital [25]. On the other hand, a randomized trial comparing low dose cytarabine with intensive chemotherapy in patients over 65 years of age showed that the number of complete remissions and also early deaths was higher in the intensive treatment groups of patients and overall survival and CR duration were not significantly different between the two treatment groups [26].

Another strategy examining the value of intensive chemotherapy in older adults with AML was used in a retrospective analysis of the Swedish National Leukemia Registry [1•]. AML patients who lived in counties where the use of intensive induction chemotherapy was more common had decreased early death rates and increased long term survival compared to patients in counties where the treatment was less aggressive [1•], arguing that the option of intensive chemotherapy should be discussed with all patients who are fit and could benefit from intensive chemotherapy.

Once the decision to use induction chemotherapy is made, the treatment in older patients should be given in intensive doses. There is no clear data supporting the use of attenuated dosing of standard drugs [27••]. In a study of patients under the age of 60, daunorubicin at 90 mg/m2 was superior to 45 mg/m2 in terms of rates of complete remission (70.6% vs. 57.3%) and median overall survival (23.7 vs. 15.7 months). However, patients between the age of 55 to 60 did not benefit from increased daunorubicin dose [28]. In a separate study limited to older AML patients, cytarabine 200 mg/m2 continuous infusion for 7 days with daunorubicin 90 mg/m2 versus 45 mg/m2 daily for 3 days (7 + 3 induction chemotherapy) showed that even older patients (60–83 year old) were able to tolerate the higher daunorubicin dose without significant additional toxicities [27••]. The OS or event-free survival (EFS) between the two chemotherapy dose groups was not statistically significant for the entire population of older patients, but the rates of CR were higher (52% vs. 35%) and response was achieved earlier in the higher chemotherapy dose group. Interestingly, when focusing on patients ages 60–65, there was a higher rate of CR (73% vs. 51%), EFS (29% vs. 14%), and OS (38% vs. 23%) [27••] in the higher daunorubicin dose group. This study did not test the more commonly used 60 mg/m2 daunorubicin dose, but if chemotherapy is given to older AML patients, daunorubicin 60–90 mg/m2 should be employed. The achievement of complete remission (standard definition or with incomplete count recovery) alone is a valuable endpoint given the improved quality of life that is derived from having relatively normal blood counts. The relapse-free survival (RFS) of patients achieving CR was longer than that of patients achieving CR with incomplete platelet recovery (CRp). Patients with CRp survived longer than those with resistant disease [29].

Therapy: less intensive treatment for older AML patients

Therapeutic options for AML patients who are not likely to benefit from induction chemotherapy include treatment with supportive care including blood products and antibiotics, less intensive chemotherapy such as hydroxyurea, low dose cytarabine, hypomethylating agents (5-azacitidine and decitabine), clofarabine, kinase inhibitors or phase 1 trials with investigational agents.

In older AML patients who are not candidates for or who would not likely benefit from standard induction chemotherapy, less intense chemotherapy with alternative agents provides a valuable therapeutic option. Treatment with 5-azacitidine prolonged survival compared with conventional care regimens (doctor’s choice of low dose cytarabine, standard AML induction chemotherapy or supportive care alone) in patients with intermediate-2 and high risk myelodysplastic syndromes [30]. In a separate report of this phase III randomized trial with high risk MDS patients, focusing on the one third of patients who would have been classified as having AML based on the WHO scheme with a median follow up of 10.1 months, the median OS for 5-azacitidine vs. conventionally treated patients was 24.5 months compared to 16 months, respectively, and 2 year OS was higher with 5-azacitidine 50% vs. 16%. Moreover, therapy with 5-azacitidine was associated with fewer days spent in a hospital [31•].

Decitabine is approved for the treatment of patients with MDS. Its use in the treatment of AML has been investigated in phase I and II studies with reported responses of 24% [32•] to 44% (with or without valproic acid) [33]. In a recent Phase II study, patients with AML treated with decitabine at 20 mg/m2 daily for consecutive 5 days in a 4 week cycle had a 24% complete remission rate [32•]. Decitabine and 5-azacitidine therefore offer an important treatment option for AML patients [32•, 33] but more information, larger studies and long term follow up are needed to assess the efficacy of this type of chemotherapy in AML, especially in older patients. These two drugs are considered to be DNA hypomethylating agents which by epigenetically altering DNA, regulate transcription of genes important in leukemogenesis. However, study of 5-azacitidine as a potential cytotoxic chemotherapeutic agent predated our understanding of this potential mechanism of action. For example, 5-azacitidine was associated with 24–28% complete remission rates in previously treated AML patients [34, 35].

Low dose cytarabine was able to induce remission in 18% of older AML patients not undergoing intensive induction chemotherapy (compared to 1% rate of remission in patients treated with hydroxyurea), but median survival was less than 6 months [36].

Several studies reported important findings about the activity of clofarabine in treating AML in older patients. Clofarabine is more myelosuppressive compared to hypomethylating agents, but offers a therapeutic option for older patients who are not candidates for intensive induction chemotherapy. A study of clofarabine monotherapy of older AML patients less likely to benefit from induction chemotherapy (patients with at least one of four unfavorable prognostic factors including age 70 or above, Eastern Cooperative Oncology Group (ECOG) performance status 2, presence of antecedent hematologic disorder, or intermediate or unfavorable karyotype) showed overall response rate (ORR) of 46% (38% CR and 8% CRp). The median disease-free survival was 37 weeks, median duration of remission was 56 weeks and the estimated median overall survival was 41 weeks [37•]. A recent European study with clofarabine showed a 48% percent response rate (32% CR, 16% CRi, complete remission with incomplete peripheral blood count recovery, with 18% mortality within 30 days [38]. A study with clofarabine with or without low dose cytarabine in 70 AML patients over the age of 60 reported 56% rate of CR, with CR rate significantly higher in the combination therapy group (63% vs. 31%), with non-significant difference in induction mortality 19% (combination) vs. 31% (clofarabine alone). The combination therapy achieved better EFS (7.1 vs. 1.7 months), but not a significant difference in overall survival (11.4 vs. 5.8 months, P = 0.1) [39]. Clofarabine in combination with cytarabine as induction therapy for AML in older patients had ORR 60% (52% CR, 8% CRp) [40]. An upcoming US Cooperative Group (ECOG) trial will compare the effect of 7 + 3 induction chemotherapy with cytarabine and daunorubicin to clofarabine alone in older AML patients.

Additional therapeutic approaches such as addition of growth factors (G-CSF and GM-CSF), despite decreasing the duration of neutropenia, have not shown consistent improvement in survival or response rates [5, 41]. Drug resistance modulation in addition to induction chemotherapy in older AML patients has also not affected response rates or survival [5, 41].

Post-remission therapy

Optimal postremission therapy in older patients is controversial. In young patients up to the age of 60, the role of consolidation therapy has been established by a CALGB study showing a significant improvement in OS with cytarabine at 3 g/m2 for 6 doses compared to lower doses of cytarabine [42]. The number of cycles of consolidation therapy in younger patients remains unclear and studies reported outcome benefit with one to 3–4 cycles of post remission therapy. There are no clear data guiding the decision concerning consolidation therapy in older AML patients not undergoing stem cell transplantation (SCT), which leads to a great deal of uncertainty [43]. Older patients do not tolerate the same high doses of cytarabine compared to younger patients due to increased cerebellar toxicity and sepsis and therefore doses of chemotherapy are usually decreased [42]. There was also no benefit in DFS in older AML patients receiving high dose cytarabine compared to lower doses of this agent [42]. For older patients not undergoing SCT who are in CR1, post-remission treatment usually includes one or two cycles of modified high dose cytarabine or repeat of 7 + 3 regimen of cytarabine and daunorubicin. Lowenberg and colleagues in 1998 published a study of remission induction and postremission therapy in older AML patients (each of whom had received two induction cycles, with the second induction chemotherapy given in CR) where additional post remission therapy with low dose cytarabine may have prolonged DFS, but did not improve survival [44]. It also appears that additional cycles of consolidation after patients received one such treatment do not affect the relapse risk (81% vs. 73%, difference was not significant), disease-free survival (16% vs. 23%), and overall survival at 5 years (23% vs. 22%) [41]. Favorable AML cytogenetics in older patients may influence the decision to undergo consolidation after CR1 since intensive post-remission therapy may results in 20–30% cure rate in such patients [8, 20].

The role of consolidation chemotherapy in older AML patients prior to reduced-intensity conditioning stem cell transplantation (RIC SCT) is unknown, but studies focusing on elderly AML population of patients usually use some form of consolidation prior to RIC SCT such as modified high dose cytarabine [27••]. In younger patients (<53–56 years old) studies showed that additional consolidation chemotherapy in first remission prior to transplantation does not offer survival benefit compared to immediate transplantation in first remission [45, 46].

Reduced-intensity stem cell transplantation

Older AML patients have increased morbidity and mortality from full myeloablative stem cell transplantation [47]. RIC SCT is increasingly used for older patients, up to age 75 or above in some transplant centers and can result in long term survival and apparent cures (Table 2) [4853], but further long term follow up studies are needed. The 2 year survival in patients ≥65 years undergoing RIC SCT old was found to be 36% [51, 52]. Nevertheless, even when RIC SCT is planned for older AML patients, only a small fraction of older adults undergo the procedure [54].

The use of haploidentical stem cell transplant or double cord stem cell transplant is usually reserved for younger AML population and therefore is not commonly used for older patients.

New agents

Non-intensive chemotherapy approaches remain an important strategy in older patients with AML who cannot tolerate and would not benefit from intensive chemotherapy. Agents that are being investigated in older patients with AML include both new chemotherapy agents (e.g. CPX-351, a novel liposomal encapsulated version of daunorubicin and cytarabine), older agents used in a new fashion (e.g. laromustine, lenalidomide) and so-called targeted agents (inhibitors of tyrosine kinases, cyclin-dependent kinases, MEK kinase, PI3 kinase, or FLT3, etc.) [55] .

A novel alkylating agent laromustine was reported to have significant single agent activity in older patients with AML with various risk factors and comorbidities showing overall response rate of 32% with 23% CR and 8% CRp [56]. However, because there was no comparison treatment arm, it is not possible to know how patients would have fared compare to other treatments.

Single agent high dose lenalidomide has been employed as a treatment strategy in AML because of the success with this agent in the treatment of MDS patients, some with higher risk disease, and MDS with 5q- abnormalities. The agent at high doses was found to have a CR rate of 30% and CRi rate of 53% in older AML patients who managed to complete the treatment, suggesting a promising clinical activity [57].

Other novel agents include CPX-351 (a liposomal formulation of a fixed ratio of cytarabine and daunorubicin), CDK inhibitor alvocidib, and farnesyltransferase inhibitor tipifarnib [55]. Although tipifarnib induced CR in some older adults with AML [58], a phase III trial failed to show survival difference compared to supportive care [59].

Tyrosine kinase activation plays an important role in pathogenesis of AML [60], therefore the use of tyrosine kinase inhibitors (TKIs) offers a unique strategy to interfere with signaling pathways involved in AML pathogenesis. In a study of 95 patients with either FLT3-mutated (46% were age 65 or older) or FLT3 wild type (75% were age 65 or older) AML or MDS, Midostaurin (PKC412) decreased bone marrow blast percentage in 71% (out of 92) of patients with FLT3-mutant and in 42% of patients with FLT3 wild-type disease, with one partial response occurring in a patient with FLT3 mutation [61]. The results suggest that midostaurin has activity in both FLT3 mutated and wild type disease. Clinical trials using a combination of standard induction chemotherapy and the FLT3-inhibitor midostaurin in patients with newly diagnosed AML are currently underway. Sorafenib treatment did not improve EFS or OS in older patients (age above 60) with AML treated with standard induction chemotherapy and sorafenib compared to standard induction therapy alone [62].


R. Stone: Consultancy for Genzyme and Novartis; G. Motyckova: none.

Copyright information

© Springer Science+Business Media, LLC 2011