Treating acute myeloid leukaemia in the elderly

Acute myeloid leukaemia (AML) is the most common acute leukaemia in adults, with an annual incidence in Europe of five to eight cases per 100,000 persons.(1) It is considered to be primarily a disease of the elderly, with a median age of onset of 67 years and a marked increase in incidence in patients older than 65 years.

Developments in terms of both chemotherapy and supportive care have resulted in steady improvements in survival in younger adults with AML, such that five-year survival rates are now in the region of 40%.(2) However, the prognosis for the majority of older patients remains poor, with the median survival continuing to be measured in months rather than years. A variety of factors are believed to contribute to this marked difference in outcome.

For example, elderly patients are more likely to have adverse cytogenetic abnormalities at presentation, to have developed AML secondary to an antecedent haematological disorder and to overexpress the multidrug resistance 1 (MDR1) gene, leading to increased treatment resistance. In addition, poor performance status, comorbidities and organ dysfunction increase the likelihood of patients being judged unfit for intensive chemotherapy.

However, with the emergence of data demonstrating the benefits of a more intensive approach to treatment in the elderly, and also the introduction of potentially more effective ‘non-intensive’ options, there is room for optimism that outcomes in this population of patients can be improved.

Treatment strategies for elderly AML patients have traditionally been subdivided into intensive and non-intensive options and these will be discussed in turn.

The mainstay of induction treatment for younger patients with AML remains the combination of cytarabine with an anthracycline, usually daunorubicin but sometimes idarubicin. A regimen such as daunorubicin for three days with seven days of cytarabine (the ‘3+7’ schedule) will produce complete remission (CR) rates after two courses of up to 75%.2 Historically, many elderly patients were deemed unfit for such an approach but a recent large Swedish study demonstrated that, in patients up to 80 years of age, giving standard induction chemotherapy actually decreased the early death rate and resulted in an improved overall survival.(3)

Investigators have also focused on whether dose intensification of chemotherapy can improve outcomes in older patients and, although findings have been inconsistent, one group recently showed that increasing the dose of daunorubicin from 45mg/m2 to 90mg/m2 led to an increase in CR from 54% to 64% (p=0.002) and an improvement in two-year overall survival from 23% to 38% (p<0.001) in a subgroup patients aged between 60 and 65 years.(4)

For those elderly patients who achieve a remission after induction therapy, the vast majority will relapse within 12 months, and there remains a lack of randomised trial evidence showing clear benefit from consolidation or maintenance strategies.(5) Although one or two course of cytarabine would be considered a standard approach, attempts to intensify this approach have, to date, met with failure.(6)

Looking forward, the most promising post-remission strategy (for selected patients) is probably allogeneic stem cell transplantation (SCT).

Allogeneic SCT is widely used in the management of younger patients with high-risk AML and might also have a role to play in those with intermediate-risk disease.(2) However, the toxicity of conventional myeloablative conditioning generally precludes its use in patients over the age of 40 years.(2) Recently, reduced intensity conditioning (RIC) regimens have been developed, with the aim of reducing the toxicity of the procedure while still harnessing the graft-versus-leukaemia effect of the donor immune system. Results to date have been encouraging.(7)

For example, a retrospective analysis of international registry data for 1080 patients with AML or myelodysplastic syndrome undergoing RIC allogeneic SCT reported two-year survival rates of between 35% and 50% in patients 55 years and older, and also demonstrated that increasing age was not associated with poorer clinical outcomes.(8) However, concerns have been raised about an increased risk of graft-versus-host disease in older transplant recipients and prospective trials in this area are needed.(2)

For those patients who are not deemed suitable for, or choose to forego, intensive treatment, options are limited. Best supportive care has historically been the most common management strategy but the MRC AML 14 trial clearly demonstrated that this was inferior to low dose cytarabine (at a dose of 20mg twice daily for ten days, given every four-to-six weeks) both in terms of response rate (18% versus 1%; p=0.00006) and overall survival (odds ratio 0.60; p=0.0009).(9) However, the one-year overall survival with cytarabine in this study was only 25%, indicating a need for novel approaches to the management of this difficult to treat population (Table 1). One of the most promising of these agents, decitabine, will be discussed in detail.

DNA methylation is a key regulator of gene transcription and genomic stability and it has been demonstrated that aberrant methylation patterns (leading to the silencing of key tumour suppressor genes) can drive the progression of a number of malignancies, including AML. Decitabine is an analogue of cytidine deoxynucleoside that acts during the S phase of the cell cycle to inhibit the enzyme DNA methyltransferase, leading to reactivation of tumour suppressor genes, induction of differentiation and apoptosis.

Decitabine has been approved by the European Medicines Agency for ‘the treatment of adult patients aged 65 years and above with newly diagnosed de novo or secondary AML….who are not candidates for standard induction chemotherapy’(10) and is administered as a daily IV infusion at a dose of 20mg/m2/day for five consecutive days of a four-week cycle.

Four key clinical trials have investigated the role of decitabine in the upfront management of AML in elderly patients for whom intensive chemotherapy was deemed unsuitable (Table 2). The first of these involved 55 patients (age >60 years) with newly diagnosed AML who received the licensed dosing schedule of decitabine until disease progression or unacceptable toxicity.(11) The mean age of the population was 74 and the median cycles received was three (range 1–25). The CR rate was 24% and the median survival was 7.7 months. Although the 30-day mortality was only 7%, at three months, 25% of patients had died, a rate similar to that seen with more intensive chemotherapy approaches. Toxicity was primarily related to myelosuppression with other grade 3/4 toxicities, including fatigue (26%) and dyspnoea (15%).

An alternative dosing schedule (but identical to that approved in the US for treatment of MDS) was adopted by Lübbert and colleagues in a large Phase II trial of 227 patients (median age 72 years) with newly diagnosed AML.(12) The drug was given at a dose of 15mg/m2 three times daily on three consecutive days every six weeks. The objective response rate (complete and partial) was 26%, although only 13% of patients achieved a CR, and median survival was 5.5 months. Again, toxicity was primarily haematological.

A third phase II study utilised a more prolonged decitabine schedule (20mg/m2/day for ten consecutive days) in 53 elderly patients (median age 74) and demonstrated a higher CR rate of 47% with a median overall survival of 12.7 months.(13) Despite the more intensive decitabine schedule, only one patient died within the first 30 days of treatment.

Most recently, the pivotal phase III study comparing decitabine with treatment choice (best supportive care or low-dose cytarabine) was published in full.(14) A total of 485 patients aged 65 years or older (median 73 years) were randomised to decitabine (20mg/m2/day for five consecutive days every four weeks) or TC, of whom 28 received supportive care and 215 received low-dose cytarabine (20mg/m2/day for ten consecutive days every four weeks). The primary endpoint was overall survival, which demonstrated a non-significant trend in favour of the decitabine arm (7.7 months versus 5 months; p=0.11) and a hazard ratio (HR) for death of 0.85. A subsequent unplanned analysis of more mature data produced an improved HR of 0.82, which was statistically significant (p=0.04). The 30-day mortality was comparable for each chemotherapy schedule (9% for decitabine, 8% for cytarabine) and the incidence of grade 3/4 adverse events and rates of discontinuation due to adverse events was also similar between groups.

It is difficult to directly compare the findings of these studies, given the different trial designs and patient populations. For example, although the best survival outcome was seen with a ten-day schedule of decitabine, that trial included a higher percentage of good-risk patients than seen in the other studies. Furthermore, it is still unclear whether decitabine or azacitidine should be the hypomethylating agent of choice for elderly AML patients. A subgroup analysis of the pivotal AZA-001 trial (which was designed for MDS rather than AML patients but recruited a cohort of patients with low marrow blast count AML) demonstrated a median overall survival for azacitidine-treated patients of 24.5 months despite a CR rate of only 15%.(15) In this trial, patients of 50 years and older were considered elderly and for this, and other methodological reasons, it is inappropriate to directly compare these outcomes with those seen for decitabine. A head-to-head trial of these agents is clearly warranted.

Despite a number of recent developments in treatment options, there remains a lack of consensus on the most appropriate pathway for elderly patients with AML. Clearly, it is crucial to differentiate between those patients who are likely to do well with intensive chemotherapy and those better suited to a non-intensive approach, such that therapeutic benefit can be maximised without exposing patients to unacceptable toxicities. It is encouraging that selected patients can now be offered a potentially curative intervention in the form of RIC allogeneic SCT, while the advent of decitabine and other novel agents is expected to improve the outlook for those patients who are deemed unsuitable for more intensive treatments.