When Less Is More: Reevaluating the Role of Intensive Chemotherapy for Older Adults With Acute Myeloid Leukemia in the Modern Era

Acute myeloid leukemia (AML) is predominantly a disease of older age, with a median age of 68 years at diagnosis.¹ Older and unfit adults with AML (herein defined as age ≥ 60 years) represent a particularly challenging population to treat, and their outcomes are significantly worse than their younger counterparts, due in part to worse disease biology, age-associated comorbidities, and poorer tolerance of conventional chemotherapy. Historically, intensive chemotherapy was the only therapy capable of inducing a meaningful rate of response in this population. However, most older patients are not suitable candidates for an intensive approach because of unacceptable anticipated rates of toxicity and treatment-related mortality (TRM).² Until recently, options for these unfit patients included supportive care, low-dose cytarabine (LDAC), or hypomethylating agents (HMAs) with azacitidine or decitabine. The dearth of tolerable yet effective options for older and unfit patients is highlighted by a Medicare analysis of patients age > 66 years diagnosed with AML between 2000 and 2009 in which 60% of patients did not receive antileukemia therapy.³ Given the historical lack of effective, lower-intensity therapies for these patients, many AML experts have supported the use of intensive chemotherapy for older adults who are deemed adequately fit, specifically those between age 60 and 74 years without a significant comorbidity. However, with the recent development of safe and effective lower-intensity venetoclax-based therapies,⁴ the wisdom of treating older and unfit adults with AML with intensive chemotherapy should be revisited.

Intensive Chemotherapy in Older AML: The Best Therapy or Just Better Than Nothing?

A major limiting factor concerning the use of intensive chemotherapy in older and more frail populations is the potential for TRM, primarily because of myelosuppression-related infectious complications. In fact, even in large, well-controlled studies of intensive chemotherapy in highly selected older patients with AML, early death rates of 10%-25% have been reported.^5-7 In one recent population-based study, the early mortality rate among patients age 65-89 years treated with intensive chemotherapy was 24%; 30% of these older adults required intensive care unit transfer and 14% required mechanical ventilation during induction.⁸ Despite the risk of TRM with intensive chemotherapy, several studies have suggested that its use is associated with superior overall survival (OS) in older adults with AML.^3,9,10 In one population-based registry study from Sweden, the use of intensive therapy was associated with lower rates of early death and superior OS compared with palliation only (including LDAC or hydroxyurea); this finding was observed across age groups and performance status.⁹ Despite the superiority of intensive chemotherapy, the 5-year OS rate in patients age 65-79 years was still < 20%, including among patients with excellent baseline performance status. On the basis of the superiority of intensive chemotherapy, the authors concluded that “most AML patients up to 80 years of age should be considered fit for intensive chemotherapy.”^9(p4179) Because of this and other similar analyses, the use of intensive chemotherapy for fit older patients has been endorsed by major consensus guidelines.^11-13

Putting aside the numerous challenges and uncertainties in accurately assessing a patient's fitness for intensive chemotherapy (reviewed in detail by others¹⁴), it is important to note that these recommendations favoring intensive chemotherapy are almost universally based on retrospective analyses from an era before the wide use of HMAs—and certainly before the development of more effective regimens such as an HMA plus venetoclax. Thus, these studies support the use of intensive chemotherapy rather than LDAC or supportive care alone but have not addressed the issue of whether an intensive approach is still appropriate in the context of the contemporary therapeutic landscape. In fact, some retrospective analyses have shown similar outcomes in older adults who receive an HMA or intensive chemotherapy as frontline AML therapy.^15-17 In the randomized, controlled trial of azacitidine versus conventional care regimens in older adults age ≥ 65 years with newly diagnosed AML, patients who received azacitidine or intensive chemotherapy had comparable median OS (13.3 v 12.2 months, respectively) and 1-year OS rates (55.8% v 50.9%, respectively), despite the azacitidine arm having more than twice as many patients age ≥ 75 years (57% v 27% with intensive chemotherapy).¹⁸ These more recent analyses therefore raise questions about recommendations favoring intensive chemotherapy over HMA-based regimens in older AML.

Some subsets of very carefully selected older patients, however, can indeed have relatively favorable outcomes with an intensive approach. A randomized phase III study compared CPX-351 (a liposomal formulation of cytarabine and daunorubicin) with 7 plus 3 chemotherapy in patients age 60-75 years with newly diagnosed secondary AML.¹⁹ In this relatively fit population of older patients with high-risk AML, CPX-351 improved the median OS and 2-year OS rate (median OS: 9.6 v 6.0 months and 2-year OS: 31% v 12%, respectively). The 60-day mortality rate with CPX-351 was 14%, which is in line with other studies of highly selected older fit patients treated intensively in the context of a clinical trial. Among those who responded to CPX-351 and subsequently underwent hematopoietic stem-cell transplantation (HSCT) in first remission (26% of CPX-351-treated patients overall), the 2-year OS rate was 63%. These results suggest that favorable outcomes can indeed be achieved in some older adults when treated intensively, although identifying this subset of patients at the time of treatment selection remains a challenge.

Lower-Intensity Venetoclax-Based Regimens: A New Era in the Treatment of Older and Unfit AML

Today, with the availability of safe and effective HMA- and venetoclax-based regimens, the argument for preferential use of intensive chemotherapy in older adults with newly diagnosed AML is even more tenuous. Table 1 shows the results from three large prospective studies in older adults with newly diagnosed AML using these different approaches—7 plus 3 in the HOVON-SAKK-AMLSG study,⁶ CPX-351 in secondary AML,¹⁹ and azacitidine plus venetoclax in the VIALE-A study.⁴ Notably, patients in VIALE-A were required to be considered unfit for intensive chemotherapy because of either age ≥ 75 years or a significant comorbidity that precluded safe delivery of intensive therapy, whereas the HOVON-SAKK-AMLSG and CPX-351 studies required patients to be fit to receive intensive chemotherapy. Compared with those in the HOVON-SAKK-AMLSG and CPX-351 studies, patients in VIALE-A were older (median age 76 v 67 years) and had worse performance status (European Cooperative Oncology Group > 1 in 45% v 10%). Patients treated with azacitidine plus venetoclax in VIALE-A had higher rates of poor-risk cytogenetics compared with those in HOVON-SAKK-AMLSG (36% v 21%, respectively) but lower rates compared with those in the CPX-351 study (57%). Despite more adverse risk features in patients treated with azacitidine plus venetoclax compared with 7 plus 3 in the HOVON-SAKK-AMLSG study, these patients had similar (or even superior) survival rates compared with patients who received intensive chemotherapy. Comparisons between azacitidine plus venetoclax and CPX-351 are a bit more challenging as the CPX-351 trial population was enriched for those with poor-risk cytogenetics and/or secondary AML, approximately one third of whom had prior exposure to an HMA. Nevertheless, it is notable that the subgroup of patients in the VIALE-A study with secondary AML who received azacitidine plus venetoclax had a median OS of 16.4 months, which compares favorably with the median OS of 9.6 months reported for the global CPX-351 population and is similar to the median OS of 15.7 months in patients with preceding myelodysplastic syndrome without prior HMA exposure who received CPX-351.^4,19

TABLE 1.

Comparison of Clinical Characteristics and Outcomes of Older Patients Treated With Intensive Chemotherapy or Azacitidine Plus Venetoclax in Three Randomized Prospective Studies

Although such cross-trial comparisons require cautious interpretation, a propensity score–matched analysis has been performed that suggests the benefit of HMA plus venetoclax over intensive chemotherapy in older AML.²⁰ In this retrospective analysis, a 10-day schedule of decitabine (DAC10) plus venetoclax was associated with higher rates of complete remission (62% v 42%; P = .01), lower rates of relapse (34% v 56%; P = .01), and lower 30-day mortality (1% v 24%; P < .01) compared with intensive chemotherapy. Notably, the significant benefit of DAC10 plus venetoclax was observed in patients at both low and high expected risk of TRM.²¹ Even among patients at low risk for TRM who received intensive chemotherapy, 30-day mortality was 16% and median OS was 6.8 months; in contrast, among patients expected to be at high risk for TRM who received DAC10 plus venetoclax, 30-day mortality was 0% and median OS was 9.1 months. These data therefore support consideration of HMA plus venetoclax combinations, rather than intensive chemotherapy, for older adults with newly diagnosed AML, including those with expected low risk of TRM. However, although the safety profile of HMA plus venetoclax appears superior to that of intensive chemotherapy, it is important to note that significant myelosuppression is still very common even with this relatively low-intensity regimen. Strategies that mitigate the hematologic toxicities of venetoclax-based combinations, including appropriate modifications to venetoclax dosing and schedule in consolidation cycles, have been proposed by some experts and are imperative to ensure safe delivery of this regimen in older adults.²²

A pressing yet unanswered question is which frontline approach is optimal for the population of older adults who are HSCT-eligible. There are relatively fewer data to directly support the use of HMA plus venetoclax in this context, although some small analyses provide guidance. In one small retrospective analysis, among 21 patients who received azacitidine plus venetoclax and subsequently underwent HSCT in first remission (the majority of whom were adverse-risk), the 3-year OS rate was 70%-75%.²³ In another analysis of 12 patients treated with DAC10 plus venetoclax followed by HSCT, the estimated 18-month OS rate after HSCT was 60%-70%.²⁴ Together, these studies suggest that HSCT may be an effective consolidation regimen for patients who receive frontline HMA plus venetoclax and can be associated with favorable survival.

A critical parameter that drives post-HSCT outcomes is the ability of these induction and consolidation regimens to induce measurable residual disease (MRD) negativity, as dozens of studies in AML have shown decreased risk of relapse and improved OS when MRD negativity is achieved, particularly before HSCT.^25,26 Retrospective studies suggest that intensive chemotherapy and HMA plus venetoclax may lead to similar rates of MRD negativity in older adults.^27,28 The similar MRD negativity rates achieved with these two approaches suggest that either might be a suitable bridge to HSCT and could be considered for HSCT-eligible older patients, although longer-term data are still needed to confirm the durability of responses with HMA plus venetoclax, particularly when subsequent HSCT is performed.

Should Low-Intensity Therapy Even Be Considered for Some Younger Patients With AML?

The discussion of intensive versus low-intensity therapy has so far concerned older and unfit patients with AML. However, intensive chemotherapy may be largely futile even in some younger patients with very poor–risk disease features, challenging the notion that these patients should, by default, receive intensive therapy solely because of their younger age. We propose that the use of low-intensity HMA plus venetoclax–based therapy should be considered not only in situations where an unacceptably high rate of TRM is expected (as in older and unfit adults) but also when intensive chemotherapy is associated with low expected rates of complete remission or complete remission with incomplete hematologic recovery (eg, < 40%) and poor OS (eg, median OS < 12 months). This triad definition of high early TRM, low response rate, and poor OS can be applied not only to older and unfit patients with AML but also to additional poor-risk AML subsets regardless of age, including therapy-related AML, AML arising from prior antecedent hematologic malignancy (particularly after failure of HMAs), AML with complex karyotype, TP53-mutated AML, AML with 11q23 translocations, and MDS1 and EVI1 complex gene (MECOM) rearranged AML.¹¹ Thus, rather than dichotomizing patients as either younger and fit or older and unfit, the optimal frontline treatment approach may be determined according to an individual's expected outcomes with intensive chemotherapy (eg, favorable or intermediate outcomes with intensive chemotherapy versus poor outcomes with intensive chemotherapy) on the basis of both age and disease-related characteristics. Instead of continuing to treat patients in this latter group with potentially toxic and futile intensive chemotherapy, alternative strategies involving a backbone of low-intensity therapy, venetoclax, and/or additional novel, targeted therapies should be explored in clinical trials.

Selecting Frontline AML Therapy in Older and Unfit Patients

Despite several methods of estimating TRM and suitability for intensive chemotherapy in older patients, such tools are still imperfect and can therefore lead to unacceptable cases of early deaths when intensive chemotherapy is used. Because of the high rates of early mortality with intensive chemotherapy in older patients and the development of new, highly effective lower-intensity regimens, the use of lower-intensity HMA plus venetoclax–based therapy for all patients age ≥ 60 years with newly diagnosed AML is favored at several centers (including our own), regardless of a patient's perceived fitness for an intensive approach. A randomized, phase II study comparing intensive chemotherapy plus HMA plus venetoclax in older fit patients is ongoing and may help to settle this pressing clinical question (NCT04801797). For very high–risk AML subgroups such as those with TP53 mutations or MECOM rearrangements, lower-intensity therapies may even be reasonably considered for younger patients as intensive chemotherapy has been historically futile in these AML subsets.^29,30 Prospective, randomized studies evaluating these approaches in a younger population with poor-risk AML are needed.

As we attempt to optimize the HMA plus venetoclax backbone with the addition of other agents (eg, FLT3 inhibitors, IDH1 or IDH2 inhibitors, and other promising novel therapies, which are all being explored in ongoing phase II clinical trials), it is important to continually reevaluate the appropriateness of intensive versus less-intensive therapy across age groups and AML subtypes. In our own practice, we favor lower-intensity HMA plus venetoclax–based regimens even in older patients who otherwise appear fit to receive intensive chemotherapy as this strategy may be associated with lower rates of TRM yet similar (or even superior) efficacy. This paradigm is also currently being explored in acute lymphoblastic leukemia, where the development of novel monoclonal antibodies and more potent tyrosine kinase inhibitors has led to reduced reliance on intensive chemotherapy even for younger patients.³¹ Similarly, in the rapidly evolving AML treatment landscape, it is not impossible to imagine a future where these more tolerable lower-intensity therapies may actually be the preferred treatment for patients across all ages.

AUTHOR CONTRIBUTIONS

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

When Less Is More: Reevaluating the Role of Intensive Chemotherapy for Older Adults With Acute Myeloid Leukemia in the Modern Era

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).