Pretreatment evaluation and prognosis of acute myeloid leukemia in older adults

INTRODUCTION — Prognosis for older patients with acute myeloid leukemia (AML) is generally inferior to that in younger patients. However, there is no consensus definition of "older adult" regarding management of AML. Curative therapy for AML generally requires intensive chemotherapy and is associated with potentially severe complications. While clinical trials have used various age thresholds for eligibility, chronologic age is only one factor to be considered when selecting therapy for older adults with AML.

Management of AML in older patients is challenging because they are more likely to have impaired functional status and/or comorbidities that can limit treatment options. Compared with younger patients, AML in older patients is also more likely to carry adverse cytogenetic features and be refractory to treatment.

This topic review will discuss the pretreatment evaluation and prognosis of older adults with AML.

Related topics include:

●(See "Acute myeloid leukemia: Management of medically unfit adults".)

●(See "Acute myeloid leukemia: Induction therapy in medically fit adults".)

●(See "Systemic chemotherapy for cancer in older adults".)

PROGNOSIS

Overview — Overall survival (OS) rates for AML decrease with increasing age (figure 1). Outcomes might be better if more patients were offered treatment, and the availability of targeted agents and less toxic treatments is expected to increase the percentage of older adults who are treated for AML. (See "Acute myeloid leukemia: Management of medically unfit adults".)

Most series of older patients with newly diagnosed AML have noted complete remission (CR) rates between 40 and 60 percent, with higher rates seen in more recent trials [1-14]. While suitably selected older patients given intensive induction therapy may achieve CR at a rate approximating that of younger patients [5], others may spend a significant proportion of their remaining life in a hospital setting receiving treatment [15,16].

Registry data indicate that the median age for patients with AML is approximately 68 years, which is more than a decade older than the median age of patients commonly reported on clinical trials. Unfortunately, many older adults receive no therapy for this disease. An analysis of Surveillance, Epidemiology and End Results (SEER) data (2000 to 2007) linked to Medicare showed that less than half of newly diagnosed patients ≥65 years of age received any leukemia-directed therapy [17]. An updated analysis (2000 to 2009) showed similar results with older age, comorbidity, and poor performance indicators associated with non-treatment [18].

Older age, variously defined in most studies as over age 55, 60, or 65 years, is a poor prognostic factor in AML. In comparison with younger patients, AML in older patients is generally more challenging to treat, as evidenced by the following characteristics [1,19]:

●Lower percentage of favorable cytogenetics

●Higher percentage of unfavorable cytogenetics

●Higher incidence of multidrug resistance

●Higher incidence of treatment-resistant disease

●Lower CR rates, shorter remission durations, and shorter median OS

Older adults are also more likely to have comorbidities and a poorer performance status that increase treatment-related morbidity and mortality and limit intensive treatments such as allogeneic hematopoietic cell transplantation (HCT). Although age is one of the most important predictors of treatment-related mortality (TRM, eg, 30-day mortality), it is likely not an independent variable and largely acts as an indirect measure of other prognostic factors. This was best shown in an analysis of 3365 adults with newly diagnosed AML that created a scoring system to estimate TRM based on age, sex, race/ethnicity, performance status, laboratory features, and the presence or absence of secondary AML [20]. The ability of the scoring system to predict TRM was similar whether or not age was incorporated; as such, older age by itself is not a contraindication to treatment.

Outcomes might be better if more patients were offered chemotherapy. An analysis of Medicare claims for 2657 older patients with AML diagnosed between 1991 and 1996 underscored the grim prognosis for AML in the older patient [21]:

●Median survival for all patients was two months, with a two-year OS of 6 percent. For patients ≥85 years of age, median survival was only one month.

●Only 30 percent of patients received chemotherapy; when compared with those not receiving chemotherapy, they tended to be younger (average age 73 versus 78 years) and live longer (median survival seven versus one month).

●Of those older patients dying from AML during the follow-up period of the study (94 percent of the sample), 31 percent of their remaining days had been spent in an inpatient facility.

Modest improvements were seen in an updated analysis using SEER/Medicare data for 5480 patients with newly diagnosed AML from 2000 to 2007 [17]. In this cohort approximately 40 percent of patients received anti-leukemia therapy, with a median survival of two months in untreated patients versus six months in those who received any therapy. The largest improvements were seen in those aged 65 to 69 years (median survival 10 months, if treated). Analysis of 8336 patients using SEER/Medicare data with newly diagnosed AML from 2000 to 2009 again showed approximately 40 percent received any therapy; the rate increased to 50 percent by 2009 [18]. Unadjusted survival was 2.5 months overall; compared with untreated patients, survival was longer for patients who were treated (5 months versus 1.5 months, respectively). In multivariate analysis, treated patients had a 33 percent lower risk of death compared with untreated. Analysis of 3637 patients ≥60 years in the PETHEMA registry reported 4.7 month OS, 29 percent one-year OS, and 7 percent five-year OS [22].

A retrospective analysis of 2767 patients with AML (excluding those with acute promyelocytic leukemia) from the Swedish acute leukemia registry reported that rates of early death (ie, death within 30 days of diagnosis) were lower in patients receiving intensive induction chemotherapy when compared with those who received palliative therapy, even when stratified for performance status; however, it is possible that patients with a better prognosis were more likely to be offered induction chemotherapy [2,23]. The difference between 30-day mortality rates for the two groups ranged from 16 to 35 percent. Patients who had de novo AML, were "fit" for intensive chemotherapy, had an Eastern Cooperative Oncology Group (ECOG) performance status of zero to 2, and were age 16 to 55, 56 to 65, 66 to 75, and 76 to 89 years had median OS times of 7 years, 18 months, 14 months, and 6 months, respectively.

In addition, a study of 103 newly diagnosed patients with AML who received intensive chemotherapy suggested that the impact of treatment on quality of life and physical function (both self-reported and objectively measured) was similar between younger (range 21 to 59 years) and older (60 to 80 years) patients [24].

Prognosis

Risk factors — A number of risk factors have been identified that occur more frequently in the older patient with AML and appear to contribute to the worse outcomes (table 1). The major independent prognostic factors in older adults with AML are [3,5,25-29]:

●Age

●Cytogenetic and genetic abnormalities

●Poor performance status

●Secondary leukemia (ie, after an antecedent hematologic disorder, such as myelodysplastic or myeloproliferative neoplasm)

●Therapy-related leukemia

●White blood cell count at diagnosis

●Multidrug resistance-1 (P-glycoprotein) expression

●CD34 expression

●Mutation spectrum (eg, TP53, SRSF2, SF3B1, U2AF1, ZRSR2, ASXL1, EZH2, BCOR, or STAG2) [30]

The effects of age, performance status, and comorbid conditions are discussed below, while prognostic factors common to all age groups (eg, cytogenetics, secondary leukemia) are presented in more detail separately. (See 'Physical functioning' below and 'Comorbid conditions' below and "Acute myeloid leukemia: Risk factors and prognosis".)

The cytogenetic abnormalities most often associated with treatment failure in young patients with AML (eg, abnormalities of chromosomes 5 or 7 or complex karyotypes) are considerably more common in older patients, occurring in 32 to 57 percent of patients in two series [27,31-34]. Conversely, all of the "favorable" cytogenetic abnormalities, such as t(8;21), t(15;17), or inv(16), are more common in younger subjects and are responsible in part for their better disease-free survival (figure 2) [33-39]. (See "Acute myeloid leukemia: Risk factors and prognosis".)

Data are available on the impact of recurring gene mutations in younger and older adults with AML. Mutations of NPM1 and both alleles of CEBPA are favorable for outcome [40,41], whereas internal tandem duplications of FLT3 (FLT3-ITD) and mutations of ASXL1 and DNMT3A are associated with poor outcomes in patients over age 60 years [42-46]. In one study, older adults without mutations in genes recurrently mutated in myelodysplastic syndrome (SRSF2, SF3B1, U2AF1, ZRSR2, ASXL1, EZH2, BCOR, or STAG2) had response rates similar to those of younger de novo AML when treated with standard AML therapy [30]. The role of genomic biomarkers in AML is discussed in greater detail separately. (See "Acute myeloid leukemia: Risk factors and prognosis".)

Prognostic models — Prognostic models for older adults with AML incorporate chronologic age, cytogenetic information, and clinical variables. These models can provide useful information for risk stratification, although all rely on chronologic age as a surrogate for certain measurable patient-specific characteristics (ie, comorbidity, functional limitations) that may differ among individuals of similarly age. These models were primarily developed for consideration of intensive therapy.

Data from a retrospective cohort of 1100 patients with AML aged 20 to 89 were used to develop an AML model to estimate the risk of mortality among patients who received treatment [47]. The model included comorbidity burden (measured by the augmented HCT-CI, see below), age and cytogenetic/molecular risks and was updated using the 2017 European LeukemiaNet Risk Classification [48]. Analysis of 2483 patients used data from two treatment trials of adults ≥60 years identified cytogenetic category, white blood cell (WBC) count, performance status, age, and secondary AML as significant factors for defining risk groups associated with OS [26]. Other models also found association of age, WBC count, cytogenetics, and mutation status with survival [49-51].

PRETREATMENT EVALUATION

General — The assessment of an older adult with AML includes evaluation of functional status and comorbid conditions.

Detailed pretreatment evaluation of all patients with AML is presented separately, as is an overview of the comprehensive geriatric assessment of cancer patients. (See "Acute myeloid leukemia: Induction therapy in medically fit adults", section on 'Pretreatment' and "Comprehensive geriatric assessment for patients with cancer".)

Specific testing for older adults is presented in the following sections.

Physical functioning — The patient's performance status, self-reported ability to perform basic and instrumental activities of daily living, and physical performance are assessments of physical function that can help predict the ability to withstand rigorous chemotherapy regimens.

Performance status — Several studies have demonstrated that measures of performance status provide prognostic information in older adults with AML, particularly for those with poor performance status at diagnosis. Both the Eastern Cooperative Oncology Group (ECOG) and Karnofsky performance status have been used as measures of physical functioning to inform prognosis in patients with AML (table 2A-B). (See "Acute myeloid leukemia: Risk factors and prognosis", section on 'Performance status'.)

●A retrospective study of data from five Southwestern Oncology Group (SWOG) trials that included 968 patients with AML found that the mortality rate within 30 days of initiation of induction therapy is dependent on both the patient's age and ECOG performance status (PS) at diagnosis [1].

•Thirty-day mortality rates were 2 to 3 percent for patients under the age of 55 years regardless of the PS.

•For patients over age 55 years, mortality rates ranged from 5 to 18 percent for patients with a PS of 0 or 1. Patients 55 to 65 years old with a PS of 2 had a similar mortality rate (18 percent).

•Patients over age 55 years with a PS of 3 and those over age 65 with a PS of 2 or 3 had much higher mortality rates that ranged from 29 to 82 percent.

•The proportion of patients with poorer performance status increased with age. PS of 2 or 3 was observed in 15, 24, 26, and 32 percent in those under age 56, 56 to 65, 66 to 75, and >75 years of age, respectively.

•The prognostic importance of poor performance status increased with age. Patients with ECOG PS 3 at the time of diagnosis had 30-day mortality rates of 29, 47, and 82 percent for age groups 56 to 65, 65 to 75, and >75 years, respectively.

●A retrospective analysis of 998 patients age 65 or greater (range 65 to 89; median 71 years) who underwent intensive induction chemotherapy reported eight-week mortality rates of 23, 40, and 72 percent for patients with ECOG PS of 0 to 1, 2, and 3 to 4, respectively [15]. One-year overall survival (OS) rates for the same groups were 35, 25, and 7 percent, respectively.

●A retrospective study of 2767 patients with AML (excluding those with acute promyelocytic leukemia) from the Swedish acute leukemia registry also reported that older patients with an ECOG PS of 0 to 1 had 30-day death rates after intensive chemotherapy of less than 15 percent, while patients with a PS of 3 or 4 had higher early death rates regardless of patient age ranging from 26 to 36 percent [2]. Seventy percent of patients up to age 80 years had a PS of 0 to 2. Of note, there were some long-term survivors in all age and PS groups.

●A prospective trial of induction chemotherapy with cytarabine plus daunorubicin in 813 older adults (median age 67 years, range 60 to 83 years) with ECOG PS of 0 to 2 reported a 30-day mortality rate of 11 percent [3].

Older adults may have impairments that are not adequately reflected in the performance status. This was best demonstrated in a single-institution feasibility study of an inpatient geriatric assessment in 54 patients with newly diagnosed AML >60 years (median age 70.8 years) [52]. Among the 38 patients with an ECOG PS ≤1, additional testing revealed the following impairments in at least one-quarter of patients: depression, distress, impairment in instrumental activities of daily living, impairment in activities of daily living, objectively measured physical impairment, cognitive impairment, and comorbidity; importantly, nearly two-thirds of patients were impaired in more than one domain. Studies investigating the prognostic significance of individual geriatric assessment domains are described below. The prognostic value of comprehensive geriatric assessment among older adults with AML is under investigation.

Activities of daily living — Geriatricians commonly measure functional status by evaluating basic activities of daily living (ADLs) and instrumental activities of daily living (IADLs). ADLs are the skills that are necessary for basic living, and include feeding, grooming, transferring, and toileting. IADLs are required to live independently in the community and include activities such as shopping, managing finances, housekeeping, preparing meals, and taking medications. Assessment of ADLs and IADLs add to the functional status obtained from the ECOG or Karnofsky performance status [52].

●Among older adults with ECOG PS <2 treated intensively for AML in a single institution study, limitations in ADLs and IADLs were prevalent (34.2 and 23.7 percent, respectively) [52].

●A prospective trial of 63 adults (age 19 to 85) with newly diagnosed AML measured Karnofsky performance status (KPS) and IADL at diagnosis [53]. Impaired KPS (<80 percent) and impaired IADLs were noted in 33 and 25 percent of patients, respectively. Impairment in IADLs was associated with decreased rates of OS independent of age and KPS. However, fewer patients with IADL limitations received intensive chemotherapy compared with those without IADL limitations (46 versus 93 percent, respectively).

●A multi-center study investigating pretreatment geriatric assessment among older adults with newly diagnosed myelodysplastic syndrome (MDS) or AML found that ADL impairment was associated with worse OS among non-intensively treated patients independent of age, cytogenetics, or KPS [54].

●A retrospective single institution study of 101 adults ≥65 years of age with newly diagnosed AML utilized questionnaire data from a quality of life survey (EORTC-QLQ C30) to determine whether specific screening questions that focused on physical, social, cognitive, psychologic, nutritional status, or pain were predictive of survival [55]. Patients who reported difficulty with strenuous activities (eg, lifting a heavy shopping bag) had a twofold increased risk of death compared with those reporting less difficulty independent of tumor biology, ECOG score, comorbidity, or treatment allocation.

Physical performance tests — The Short Physical Performance Battery (SPPB) is a validated objective measure of lower extremity function that has been shown to predict future disability, hospitalizations, and mortality among older adult patients without AML [56-64]. It includes measures of standing balance (timing of tandem, semi-tandem, and side-by-side stands), four-meter walking speed and ability, and time to rise from a chair five times. Total scores range from 0 to 12 and total scores <9 are indicative of physical frailty and worse clinical outcome among older adults without cancer.

While validation studies in AML are underway, it is reasonable to utilize the SPPB to objectively assess functional status when making treatment decisions in older adults with AML, particularly among those with ECOG PS <3. This test can be performed in approximately five minutes by trained staff in the inpatient or outpatient setting using publically available training modules to facilitate standardization of testing.

The following studies have evaluated physical performance tests in this setting:

●A prospective single center study evaluated the prognostic value of the SPPB in 74 older adults (≥60 years) with newly diagnosed AML undergoing induction chemotherapy [65]. The mean age was 70 years and 78 percent had an ECOG PS ≤1. The median OS was 11 months, and the 30-day mortality rate was 15 percent. The median SPPB score was 8.5. When compared with those with higher scores, patients with an SPPB score <9 had a shorter median survival (6.0 versus 16.8 months; hazard ratio [HR] 2.2, 95% CI 1.1-4.6). When analyzed as a continuous variable, each two-point increase in the SPPB score was associated with a 15 percent decrease in the HR for death (HR 0.85; 95% CI 0.72-1.01) controlling for age, sex, ECOG PS, cytogenetic risk, prior myelodysplastic syndrome, and hemoglobin.

●A single center study evaluated patients ≥60 years old with the Wheatley index, a composite prognostic tool that includes pretreatment quality of life (QOL) and physical performance measures (PPMs; ie, hand grip strength, a two-minute walk test, and timed chair stands) [66]. QOL and PPMs at diagnosis were not good predictors of one-year mortality, but patients with a poor Wheatley risk score had higher one-year mortality than patients who had good or standard scores.

Differences between the two studies' results may relate to differences in the patient populations and/or the distinct measures of physical performance tests, which may not be equivalent.

●Use of the four-meter walk test (a subcomponent of SPPB) is a reasonable substitute for evaluating objective physical function. Usual gait speed is a consistent predictor of outcomes among older patients in varied settings [67,68]. A registry analysis evaluated the predictive utility of four-meter usual gait speed among 448 adults aged 75 years or older seen at a single institution with a new diagnosis of hematologic malignancy [69]. In this study, every decrease of 0.1 meters per second (m/s) in gait speed was associated with higher mortality (HR 1.2, 95% CI 1.12-1.29), odds of unplanned hospitalization (OR 1.33, 95% CI 1.16-1.51), and emergency department visits (OR 1.34, 95% CI 1.17-1.53). Those with the best survival had a gait speed faster than 0.8 m/s, while those with gait speed slower than 0.6 m/s had worse survival, regardless of ECOG performance status. These thresholds can be useful in clinical practice to identify patients with physical frailty.

Comorbid conditions — Comorbid conditions are common among older adults with AML; analysis of over 5000 adults in the United States with newly diagnosed AML (median age 78 years, range 65 to 93) showed that approximately half had a claims-based diagnosis of at least one major comorbid condition [17]. The presence of major comorbidity influences the treatment plan and is frequently associated with increased treatment risk and worse outcomes [2,5,15,20,70,71].

Many studies have shown a relationship between increased comorbidity burden and decreased remission rates, increased early mortality, and decreased OS [17,54,55,70,72-77]. However, other studies have not shown a clear relationship between comorbidity burden and outcomes [65,78,79]. Cross study comparisons can be difficult due to differences in treatment and patient selection. The implications of comorbidity may also differ by treatment type and by characterization of the comorbid conditions themselves. Because of a lack of uniform collection of comorbidity data in many multi-center trials of AML, much of the evidence supporting assessment and management of older patients with comorbid illnesses is limited to smaller studies and population-based studies.

Major barriers to successful treatment in older patients include decreased bone marrow regenerative capacity (even after successful leukemia cytoreduction), an inability to tolerate long periods of pancytopenia, malnutrition, and the nephrotoxicity of certain drugs (eg, aminoglycosides, amphotericin). Evidence is limited regarding the implications of individual comorbid conditions on treatment tolerance beyond those conditions that are known contraindications for specific chemotherapeutic agents. As an example, the presence of diabetes and hyperglycemia may increase the risk of mortality for older adults with AML [80,81], but further research is needed to determine if changes in management of comorbid conditions will influence outcomes.

Measures of comorbidity — A modified Charlson comorbidity index (CCI) and the hematopoietic cell transplantation-specific comorbidity index (HCT-CI) are frequently used to assess comorbidity in the setting of AML, even though neither test was originally designed for use in older adults with AML. Use of either the CCI or HCT-CI to assess comorbidity burden can inform decisions and management. Studies utilizing the CCI and HCT-CI in AML are represented below.

Charlson comorbidity index — The original Charlson comorbidity index (CCI) was devised as a measure of comorbidities in older adults. A revised version has been utilized in studies of older adults with AML with mixed results (table 3). Small retrospective studies and large population-based studies have evaluated CCI in older adults with AML [17,70,78,79].

●In an analysis of SEER/Medicare data representing 5480 people ≥65 years of age with newly diagnosed AML, 45 percent had at least one major comorbid condition reflected by CCI ≥1 [17]. Higher scores on the claims-based CCI were independently associated with higher 30-day mortality and shorter survival, after adjusting for age and receipt of therapy. Performance status was not included in the analysis.

●A Danish study representing 2792 patients diagnosed with AML between 2000 and 2012 identified 40 percent with at least one major comorbid condition (modified CCI ≥1); only 25 percent of those receiving intensive therapy (n = 1467) had CCI ≥1 [79]. Among patients receiving intensive therapy, poor performance status but not comorbidity was associated with remission and early mortality. These results suggest that comorbidity impacted treatment decisions.

HCT comorbidity index — The hematopoietic cell transplantation specific comorbidity index (HCT-CI) was designed to predict outcomes in younger adults undergoing HCT (table 4). It has had mixed results in predicting outcome in older adults with AML. More details on the HCT-CI are presented separately. (See "Hematopoietic support after hematopoietic cell transplantation".)

Moderate-sized retrospective studies have evaluated the impact of HCT-CI in older adults with AML [55,72,73,75,76,78]. Analysis of data from the multi-center ALFA-9803 trial, in which patients ≥65 years of age (n = 416) with newly diagnosed AML were treated with intensive induction therapy, showed that an HCT-CI ≥3 was independently associated with worse OS. The prevalence of HCT ≥3 in the clinical trial was low (<10 percent) [76].

An augmented HCT-CI has been developed and tested in the setting of AML. This augmented index includes the comorbidities from the HCT-CI plus the addition of hypoalbuminemia, thrombocytopenia, and lactate dehydrogenase (LDH) [47]. Compared with the original HCT-CI, the augmented index better predicted survival in combination with age and cytogenetic/molecular risk among a large cohort of adults treated for AML.

Cognitive impairment — The incidence and prevalence of cognitive impairment increases with age. Pretreatment cognitive impairment may increase the risk of complications during and after intensive therapy for AML. Cognitive screening may identify patients who are at higher risk for complications, including delirium. (See "Assessment of decision-making capacity in adults".)

Research in this area is still limited. A study of 54 patients with AML/MDS documented impaired performance on a battery of cognitive tests in up to 40 percent of the study population before treatment [82]. Similarly, a second small prospective study detected cognitive impairment in 30.5 percent of patients (mean age 70.8) prior to treatment with intensive induction chemotherapy [52]. In further analysis of this population, patients with cognitive impairment had a shorter median survival (5.2 versus 15.6 months). Results remained significant after controlling for age, sex, ECOG performance status, cytogenetic risk, prior MDS, and hemoglobin (HR 2.5; 95% CI 1.2-5.5) [65]. Multi-site studies are on-going to determine whether brief cognitive screening can identify older adults at risk for poor treatment tolerance. It is reasonable, however, to incorporate a brief cognitive screen into pretreatment evaluations as results can inform management (with a focus on prevention of delirium) and have implications for considerations of capacity during informed decision-making.

A large single institution study evaluated adults aged ≥75 years who were seen in consultation for myeloma, leukemia or lymphoma. This study identified a high prevalence of cognitive impairment using simple screening tools for executive dysfunction (35 percent impaired using the Clock-in-the-Box test) and working memory (17 percent impaired using the five-word delayed recall) [83]. Impaired working memory was associated with worse survival, including when patients were stratified by aggressive versus indolent malignancy. In multivariate analysis, impaired working memory (≤2 out of 5 words correct) was associated with worse survival controlled for age, comorbidity and disease aggressiveness (OR 0.26, 95% CI 0.13-0.5).

Symptoms — Specific symptoms present at the time of pretreatment assessment may be prognostic for older adults with AML. A multi-site study of pretreatment geriatric assessment for older adults with MDS or AML identified a high level of fatigue (score >50 on the EORTC-QLQ C30 fatigue subscale) as independently predictive of worse OS [54]. A single institution study of older adults with newly diagnosed AML identified pain (reported as more often versus less often) as an independent prognostic factor for increased risk of death [55].

Comprehensive geriatric assessment — Pretreatment assessment for older adults with AML should include factors that may differ between adults with the same chronologic age and tumor biology. A comprehensive and systematic approach should be used for evaluation since many patients present with more than one impairment or vulnerability. Geriatric assessment is recommended for adults ≥65 years of age who are considered for chemotherapy [84]. Both single institution and multi-site studies indicate that geriatric assessment is feasible in the context of AML therapeutic decision-making [52,54,85]. A feasibility study showed that a primarily self-reported geriatric assessment could be administered in approximately 30 minutes (approximately 10 minutes of staff time) [86]. As discussed above, geriatric assessment can detect multiple vulnerabilities among older adults with AML that may not be captured in routine clinical practice (eg, depression, distress, impairment in basic and instrumental activities of daily living, objectively measured physical impairment cognitive impairment and comorbidity). (See 'Performance status' above.)

At present, the evidence is strongest to support the addition of more refined measures of physical function to standard oncology performance scales. We recommend the assessment of ADLs and IADLs for all patients and consideration of physical performance testing (SPPB or gait speed) for patients with good performance status to identify vulnerability. We also recommend the use of a standardized measure of comorbidity burden (modified CCI, HCT-CI or augmented HCT-CI). Consideration should be given to screening for cognitive impairment. We suggest considering the five-word delayed recall as a practical screening test.

At present there is no validated definition of "fitness" or "unfitness" in the context of AML therapy. Available evidence supports poor performance status, ADL impairment, and high comorbidity burden as vulnerabilities that can characterize individuals at greatest risk for treatment-related toxicity and shorter survival. Studies are underway to characterize "fitness" and "frailty" in the context of more intensive and less intensive therapies [87,88]. In addition, pretreatment geriatric assessment can also identify vulnerabilities that can been intervened upon during therapy to improve treatment tolerance and survivorship.

TREATMENT CHOICES — A discussion with the patient, with or without loved ones and caregivers, should include a review of the following [89]:

●Prognostic information to enable informed decisions regarding choice of treatment. Regardless of treatment choice, patients and their family members often report not being offered alternative treatment options and tend to overestimate the chance of cure [90]. Written consent forms required for clinical trials may serve an educational role, even for those who do not desire to enter into a formal study.

●Implications of the diagnosis and treatment on quality of life [91,92], functional independence [93-95], and social support.

●Who has durable power of attorney for health issues if the patient becomes unable to make decisions?

●Does the patient have an updated will? Do other members of the family know where this information is kept? Will the family have access to adequate funds while the patient is hospitalized?

●A discussion concerning "code" status and the possibility that the patient might need to be transferred to an intensive care unit, with its attendant morbidity and mortality [96]. This should include issues related to "do not resuscitate" and "do not intubate" orders, such that the patient and family can make properly informed decisions on these matters. (See "Advance care planning and advance directives".)

●The effect on the patient's employment. Most patients will not be able to return to even part-time work until the completion of induction and consolidation chemotherapy.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Acute myeloid leukemia".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient education" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Acute myeloid leukemia (AML) (The Basics)" and "Patient education: Leukemia in adults (The Basics)")

●Beyond the Basics topics (see "Patient education: Acute myeloid leukemia (AML) treatment in adults (Beyond the Basics)")

SUMMARY

●Description – Prognosis for older patients with acute myeloid leukemia (AML) is generally inferior to that of younger patients. There is no consensus definition of "older adult" regarding management of AML.

Age is only one factor to be considered when selecting treatment for older adults with AML. We favor functional assessment of the individual rather than chronologic age, per se. Treatment decisions are individualized and should consider age, functional status, comorbid conditions, and clinical and cytogenetic features of the leukemia.

●Prognosis – Survival for patients with AML decreases with increasing age (figure 1). (See 'Prognosis' above.)

●Risk factors – Both patient-specific factors and biologic features of the leukemia contribute to inferior outcomes in older patients. Important risk factors include (see 'Risk factors' above):

•Age

•Impaired performance status

•Cytogenetic and genetic abnormalities – Adverse features (eg, abnormalities of chromosomes 5 or 7 or complex karyotypes) and mutations (eg, mutated TP53) are more common in older patients

•White blood cell count at diagnosis

•Secondary leukemia (ie, AML arising from an antecedent myelodysplastic or myeloproliferative neoplasm)

•Therapy-related myeloid neoplasms

●Evaluation. (See 'Pretreatment evaluation' above.)

•General evaluation – All patients with AML, regardless of age, require a general pretreatment evaluation, as discussed separately. (See "Acute myeloid leukemia: Induction therapy in medically fit adults", section on 'Pretreatment'.)

•Physical functioning – Assessment of older individuals should include:

-Performance status – Eastern Cooperative Oncology Group (ECOG) and Karnofsky performance status (table 2A-B). (See 'Performance status' above.)

-Activities of daily living (ADLs) – Basic ADLs (eg, feeding, grooming, transferring, toileting) and instrumental ADLs (IADLs; activities required to live independently in the community, such as shopping, managing finances, housekeeping, preparing meals, taking medications). (See 'Activities of daily living' above.)

-Physical performance tests – The Short Physical Performance Battery (SPPB), a measure of lower extremity functioning. (See 'Physical performance tests' above.)

•Comorbidities – Patients with chronic cardiac, pulmonary, hepatic, or renal disorders or diabetes experience greater acute toxicity from chemotherapy. (See 'Comorbid conditions' above.)

Either of the following measures of comorbidities (table 4) should be applied to assess severity of comorbidities:

-Charlson comorbidity index (CCI). (See 'Charlson comorbidity index' above.)

-Hematopoietic cell transplantation specific comorbidity index (HCT-CI). (See 'HCT comorbidity index' above.)

•Cognitive screening – Cognitive screening is discussed separately. (See "Assessment of decision-making capacity in adults".)

•Geriatric assessment – For selected patients, a comprehensive geriatric evaluation may be useful to assess impairment and/or vulnerability. (See 'Comprehensive geriatric assessment' above.)

●Treatment choices – Discussions with the patient, with or without loved ones and caregivers, should include frank conversation to enable informed decision-making. (See 'Treatment choices' above.)

Important topics include the individual's overall prognosis and the impact of treatments on quality of life, functional independence, and social support. Choices regarding the patient's willingness to be resuscitated, intubated, and/or transferred to an intensive care unit should be addressed.