Better pass boldly into that other world, in the full glory of some passion, than fade and wither dismally with age. — James Joyce
Idioms describing the perils of old age are plentiful. Bernard Baruch remarked, “To me, old age is always 10 years older than I am.” That aging catches people unaware is a universal experience. This is often the premise when older adults are diagnosed with acute myeloid leukemia (AML) and are suddenly forced to consider how age will impact their prognosis and treatment options. As the median age at diagnosis rests at 67 years, AML is largely a disease of older adults, with 70% of new cases arising in patients ≥55 years.1 Patients >60 years diagnosed with AML tend to have 5-year survival rates around 5% to 15%.2 Because long-term survival is typically dependent on tolerance of an intensive treatment regimen, decisions about whether to proceed with treatment of AML in older adults and which treatments are most likely to balance benefit and harm can be complex. The situation is compounded by the aggressive nature of the disease, which does not afford the luxury of time to make decisions. This article seeks to explore the challenges that confront older adults with AML and their healthcare providers in determining how to treat the disease in this population.
AML is a cancer of the blood cells that arises in the bone marrow. An aggressive disease, it spurs complications related to bone marrow failure. Patients may present routinely with low blood counts or more emergently with fevers, infections, or bleeding. AML is considered an emergency and is one of the few cancers that requires urgent treatment to prevent the catastrophic effects of organ infiltration by leukemic cells.1 When a patient with AML has an adequate performance status and wishes to accept intensive treatment, induction chemotherapy is initiated quickly. While prognosis is connected to the cytogenetic and molecular characteristics of the leukemia, in general, the 5-year survival rate ranges from 35% to 40% in patients <60 years. The prognosis remains bleaker among older adults and is impacted by various patient- and disease-related factors.2
Standard Treatment Modalities for AML
Standard induction treatment for AML combines cytarabine given continuously over 7 days with an anthracycline given for 3 days. As such, it has been coined the “7 and 3” regimen. This regimen requires hospitalization during treatment and for several weeks afterward due to prolonged myelosuppression and the need for supportive care. When remission is achieved with induction chemotherapy, it is most often followed by consolidation chemotherapy, which includes several cycles of intermediate- to high-dose cytarabine. Complete responses in patients >60 years are 40% to 60%.2 This mode of AML treatment has been in use for decades with little change until recently, when certain targeted therapies for AML became available. FLT3 and IDH inhibitors are novel therapies only recently available for patients with documented FLT3 and IDH1 or IDH2 mutations.1 Prognostic classification of AML into low-, intermediate-, or high-risk categories, along with patient’s age and performance status, influence whether a hematopoietic stem cell transplantation is recommended.
Lower-Intensity Treatment for AML
For those patients unlikely to tolerate intensive therapy, lower-intensity options exist. Low-dose cytarabine yields a complete response rate of 5% to 20% but lacks durability of response.2 Use of hypomethylating agents such as azacitidine and decitabine is prevalent in patients with high-risk myelodysplastic syndrome and has also shown benefit in the AML population, in which they boast a response rate of 20%. The median survival with these agents, however, is a mere 5 to 6 months.1 To date, there has not been a randomized clinical trial directly comparing azacitidine (given 7 days each 28-day cycle either intravenously or subcutaneously) with decitabine (given 5 days in each 28-day cycle intravenously); however, decitabine has conventionally yielded higher response rates.3 In clinical trials, the median time to best response with azacitidine was 5 to 6 months, an important consideration depending on the pace of the leukemia.1 Some patients opt for best supportive care, which may or may not include hydroxyurea to lower the white blood cell count in conjunction with transfusion of blood products.2
Prognosis of AML
Prognostic factors can be separated into those associated with the patient and those associated with the disease. Age has long been touted as a reasonable predictor of suitability for treatment and outcome in AML, but use of this measure alone has limitations, and other factors warrant consideration.4 Such factors include comorbidities and performance status. Prognostic factors related to the disease take into account white blood cell count at diagnosis, antecedent hematologic disorder, prior cytotoxic therapy for another condition, and genetic changes in the leukemia cell. Research is ongoing about the prognosis particular molecular markers or combinations of markers portend. At present, NPM1, CEBPA, and FLT3 mutations are applied in clinical practice. The prognostic implications of other mutations such as DNMT3A, IDH1, and IDH2 remain less clear.2 Older adults with AML are more likely to have qualities associated with resistance to therapy, such as adverse genetic mutations, an antecedent hematologic disorder, or treatment with previous chemotherapy or radiotherapy. Furthermore, older adults in general are less apt to be able to tolerate intensive chemotherapy and are more susceptible to treatment-related mortality. This presents a complex quandary as patients and providers struggle to consider all approaches to the patient, the leukemia, and the treatment and to optimize survival and quality of life.5
Introduction to Literature Review
A robust body of literature exists exploring approaches to treatment in older adults with AML; however, few authors claim to have the panacea. Instead, many articles highlight prognostic factors in this patient population and discuss challenges in making treatment decisions. A literature review was completed to delve into existing research, current practice, and gaps in the evidence.
Sanford and Ravandi1 discuss the heterogeneous nature of AML and the aforementioned prognostic factors. They concede that, in general, the prognosis of older adults with AML is poor but mention that a small portion of this population can attain durable remissions. The authors reference a study of 998 older adults treated with intensive induction chemotherapy. Various unfavorable elements were isolated, including age ≥75 years, unfavorable karyotype, Eastern Cooperative Oncology Group (ECOG) performance status >2, lactate dehydrogenase >600 IU/L, creatinine >1.3 mg/dL, and not being treated in a laminar airflow room. This research illustrated that despite the overall poor prognosis of older adults with AML, variability in prognosis exists within the population. Sanford and Ravandi1 highlight the importance of assessing fitness for therapy and encourage consideration of age as well as other factors. They support this with retrospective data from the Southwest Oncology Group documenting that worse performance status correlated with an increased rate of early death in patients ≥56 years. Sanford and Ravandi warn of the subjectivity of the performance status scales and the fact that they do not synthesize the key domains of health, such as functional ability, cognition, mental health, physical function, and socioenvironmental situation, as a formal geriatric assessment does.1
Sanford and Ravandi1 described another tool, the AML score, as an estimated calculation of the likelihood of death within 60 days and complete response, factoring in clinical information, laboratory values, and cytogenetic characteristics, thus allowing patients and providers to glimpse the potential balance between risk and benefit with intensive induction chemotherapy. The authors discuss the limitation of using early death as a measure of treatment-related toxicity, as death is often a consequence of the underlying leukemia. Despite these comprehensive assessment approaches, the characteristics of the AML, the alternate treatment options, and the patient’s wishes are equally important to consider.1 Ultimately, Sanford and Ravandi cite research showing that most patients with AML will likely benefit from some form of therapy, and they recommend that cytogenetic and molecular testing be completed in all elderly patients with this disease to serve as a guiding force in treatment planning.1
Walter and Estey5 convey the benefits of intensive therapy in patients with AML who are aged ≤80 years. They advise assessing individual factors regardless of presumed benefits and caution that documented benefits from intensive treatment cannot be isolated from the accompanying benefits of diligent supportive care in patients undergoing chemotherapy. While they acknowledge data supporting the benefits of treatment in older adults with AML, they discuss the ambiguity related to identification of which patients should be treated intensively and which should not. The authors present 3 studies that portray the challenges in precisely eliciting the balance between treatment efficacy and treatment toxicity. Walter and Estey also endorse the incorporation of other factors in addition to age in fitness for treatment assessments and highlight the value of comprehensive geriatric assessment in predicting treatment-related mortality.5 The authors mention the significant decline in early death rates after intensive induction therapy in recent decades, even though the treatment regimen remains largely unchanged. While not specific to older adults, such data imply that improved supportive care has made intensive therapy safer. Walter and Estey support assignment to intensive or nonintensive therapy based on the probability of treatment-related mortality and the risk of toxicity deemed acceptable.5
Mamdani and colleagues3 reviewed intensive and lower-intensity treatment options for older adults with AML. Based on the current research, they highlight the common approach of using hypomethylating agents in those older adults who are not fit for intensive therapy. They underscore the lack of representation of older adults in clinical trials, despite AML largely being a disease of the elderly. The authors recognize that the variability in tolerance of treatment and the absence of reliable ways to measure specific patient features that portend intolerance of treatment and outcomes makes treatment planning difficult. They urge careful attention to comorbidities, performance status, and cytogenetic and molecular properties as well as the patient’s preference in decision-making. In addition, the authors outline that patients with disease that is considered high-risk, or those who have multiple comorbid conditions or a poor performance status, are likely best served by supportive care or lower-intensity therapy.3
In a candid editorial, Sekeres and Gerds6 discuss the plight of the older adult with AML, denoting both those who exit gracefully from life and those who meet their demise reluctantly. They critically describe the common practice of asking patients to choose from the lesser of 3 evils (intensive therapy, lower-intensity therapy, or supportive care) within hours of diagnosis. Despite well-intended guidance, the authors point out the lack of knowledge about how each option will translate to time spent in the hospital or receiving treatment of some sort. They emphasize the healthcare providers’ responsibility to temper their optimism and prepare older adults with AML for the unfortunate realities of the disease.6 Citing the retrospective analysis by El Jawahri et al of 330 older adults with AML exploring their use of healthcare at the end of life, Sekeres and Gerds describe barriers to this patient population accessing palliative care and hospice. This gap in utilizing such resources represents an opportunity for further research and improvement.
Wass and colleagues4 completed a retrospective study of 194 adults with AML aiming to understand how comorbidities existing prior to treatment impacted survival. The study employed the Hematopoietic Cell Transplantation-specific Comorbidity Index, the Adult Comorbidity Evaluation-27 (ACE-27), as well as the Cumulative Illness Rating Scale for Geriatrics. Demographic information along with cytogenetics, record of treatment, and outcomes were analyzed. The study revealed that among patients receiving intensive therapy, more comorbid conditions per the ACE-27 tool independently impacted survival. Patients in this particular study most frequently had coexisting cardiovascular disease, infection, hepatic disorders, and prior cancers. The authors found that the ECOG performance status remained a better short-term predictor of early death than comorbidity status. Long-term survival, however, was more accurately reflected by comorbidities. The article contends that numerical age should always be considered in conjunction with assessment of comorbid conditions and other patient- and disease-related circumstances.4
Wang and colleagues7 highlighted patterns of healthcare use in older adults with AML at the end of life. The authors acknowledge that a small segment of older adults with AML may have positive long-term outcomes, which could inspire patient and provider decisions in the hope of being an outlier. The article portrays the initial dilemma as patients and providers weigh the knowledge that intensive chemotherapy offers the best possibility of long-term survival with the older adults’ vulnerabilities to the toxic effects of treatment. To add insult to injury, research has indicated that older adults may experience the greatest functional decline a short time after induction therapy and are unlikely to recover as thoroughly as their younger counterparts. This functional decline can preclude older adults from tolerating standard postremission therapy, thereby impacting their ultimate outcome.7
Klepin8 describes the unique intensity of induction chemotherapy with its protracted myelosuppression and several weeks of hospitalization to underscore the need for reliable ways to measure and prevent the negative impact of functional decline in patients receiving treatment for AML. She outlines the damaging effects of such decline, including altered quality of life, increased dependence, and risk of secondary medical events such as falls, increased use of healthcare, and inability to receive subsequent treatment to maintain remission. Klepin suggests the ultimate challenge lies in determining who is fit for intensive therapy, as the definition of fit is fluid and subjective. While ECOG performance status assessment may identify the frailty of older adults with AML, it is not a reliable measure of vulnerability that captures subclinical deficiencies that could impair recovery when the patient is stressed by intensive therapy. Klepin also endorses the use of a geriatric assessment to predict tolerance of treatment as well as to address vulnerabilities.8 Although more research is needed, Klepin advocates for the use of a comprehensive assessment of the characteristics of the older adult with AML to help discern the patient’s fitness, frailty, and/or vulnerability.8
Age considers, youth ventures. — Rabindranath Tagore
Implications for Practice
While proverbs abound heralding age as a mere number, something that can be overcome by mind over matter, the challenges advanced age poses in an illness such as AML are unparalleled. Few expect to be stricken in their golden years with such a sudden, immediately life-threatening illness that quickly derails normal life and is often accompanied by long and frequent hospitalizations.9 An older adult with AML can easily change from active and independent to debilitated overnight. Although no clear consensus exists about how best to treat AML in the older adult population, much of the literature supports a comprehensive assessment of patient- and disease-related characteristics as particularly helpful in determining suitability for treatment, and at what intensity. The research of Klepin and colleagues suggests that there may be a role for strategizing to minimize functional decline in patients who are considering intensive therapy.10 Older adults are woefully underrepresented in clinical trials, and efforts to encourage participation of this population in clinical trials when possible are important. At least 90% of older adults with AML will die of their disease.11 If individual patient and disease characteristics are not carefully considered, this patient population is at high risk for poor quality of life and suffering. When given the diagnosis of AML, patients enter a strange new world in which they are asked to make hefty decisions regarding treatments for which they have no frame of reference. Healthcare providers are well positioned to guide patients through the arduous process of facilitating an understanding of the nature of treatment options and weighing each option. Reducing barriers to palliative care and hospice for this population has the potential to facilitate shared decision-making and preserve quality of life. While outcomes are uncertain, being able to make an informed decision and choose an option that is compatible with the patient’s wishes can ease the burden of this devastating disease.
Although treatment for AML has not changed significantly in recent decades, progress continues to be made. With the advent of molecular testing and targeted therapy, the landscape of the disease may change in coming years as new therapies are developed.12 Support of patients’ participation in clinical trials whenever feasible will also aid improvements in the future. In the interim, patients and providers are left to navigate the uncertain terrain of this disease and embark on a treatment plan that is most compatible with the patient’s priorities and performance status. Use of formal geriatric assessments and development of more sophisticated risk stratification tools can help support the process. In the face of no standard solution to treating older adults with AML, acknowledgment of this dilemma and subsequent comprehensive evaluation of patients and their disease offer the best hope of informed, personalized decision-making.
- Sanford D, Ravandi F. Management of newly diagnosed acute myeloid leukemia in the elderly: current strategies and future directions. Drugs Aging. 2015;32:983-997.
- Dohner H, Weisdorf DJ, Bloomfield CD. Acute myeloid leukemia. N Engl J Med. 2015;373:1136-1152.
- Mamdani H, Santos CD, Konig H. Treatment of acute myeloid leukemia in elderly patients – a therapeutic dilemma. J Am Med Dir Assoc. 2016;17:581-587.
- Wass M, Hitz F, Schaffrath J, et al. Value of different comorbidity indices for predicting outcome in patients with acute myeloid leukemia. PloS One. 2016;11:e0164587.
- Walter RB, Estey EH. Management of older or unfit patients with acute myeloid leukemia. Leukemia. 2015;29:770-775.
- Sekeres MA, Gerds AT. The graceful exit or reluctant demise of the older adult with acute myeloid leukemia. Cancer. 2015;121:2678-2680.
- Wang R, Zeidan AM, Halene S, et al. Health care use by older adults with acute myeloid leukemia at the end of life. J Clin Oncol. 2017;35:3417-3424.
- Klepin HD. Definition of unfit for standard acute myeloid leukemia therapy. Curr Hematol Malig Rep. 2016;11:537-544.
- Rodin G, Yuen D, Mischitelle A, et al. Traumatic stress in acute leukemia. Psychooncology. 2013;22:299-307.
- Klepin HD, Tooze JA, Pardee TS, et al. Effect of intensive chemotherapy on physical, cognitive, and emotional health of older adults with acute myeloid leukemia. J Am Geriatr Soc. 2016;64:1988-1995.
- Brown G, Marcinkowska E. Acute myeloid leukaemia: new targets and therapies. Intl J Mol Sci. 2017;18:2577.
- Doria-Rose VP, Harlan LC, Stevens J, Little RF. Treatment of de novo acute myeloid leukemia in the United States: a report from the Patterns of Care program. Leuk Lymphoma. 2014;55:2549-2555.