Understanding Treatment Tolerability in Older Adults With Cancer

INTRODUCTION

The treatment cannot be worse than the disease is an oft-cited statement representing the important concept of tolerability; older patients often express concerns about side effects during decision making for cancer treatment.¹ Tolerability, defined by the International Council for Harmonization as “the degree to which overt adverse events can be tolerated by the subject,”² has been historically quantified solely using clinician ratings of adverse events (AEs). However, more recent definitions have recognized the importance of incorporating patients' perspectives.³

CONTEXT

• Key Objective

• What do we know about cancer treatment tolerability in older adults and where are the gaps?

• Knowledge Generated

• Although there is strong evidence that geriatric assessment findings can be helpful in predicting treatment tolerability in older adults with cancer, more research is needed on treatment tolerability assessment, monitoring, and mitigation efforts for older adults during and after completion of treatment. A model is proposed to guide research efforts and clinical care that incorporates multiple tolerability aspects and highlights the patient's experience and functional status as important end points.

• Relevance

• Incorporating patient-reported outcomes, functional evaluation, and adverse event data provides a more comprehensive and accurate understanding of treatment tolerability in older adults.

Understanding tolerability in older adults with cancer is challenging, because the growing population of older patients remains under-represented in clinical research that sets cancer care standards.⁴ Cancer incidence in patients age 65+ is predicted to increase 67% from 2010 to 2030, compared with 11% in younger adults.⁵ Furthermore, more than 25% of new cancer diagnoses in the United States are in patients age 75+, a group for which very limited data exist regarding the safety and efficacy of systemic therapies.⁵ Older patients with aging-related conditions (eg, comorbidity, disability, and geriatric syndromes) are particularly under-represented in cancer clinical trials.^4,6,7 Lack of evidence surrounding tolerability leads to significant variability in treatment decision making.^4,8 Older patients are often given cancer treatments with high toxicity,^9-12 which can lead to early discontinuation, hospitalization, and overall harm greater than benefit.^13-15 Unrelieved symptoms in patients with aging-related conditions are associated with decreased function¹⁶ and poor quality of life (QoL).¹⁷ On the other hand, ageism, or attention to age alone rather than heterogeneous factors such as physiologic resilience and treatment preferences, can promote undertreatment of fit older adults.¹⁸

The Cancer and Aging Research Group (CARG),⁶ ASCO,¹⁹ the US Food and Drug Administration (FDA),^19,20 and the National Cancer Institute (NCI)²¹ have identified research on tolerability of cancer treatment as a priority, including the measurement of tolerability in older adults. The FDA has recognized symptomatic AEs, disease symptoms, and physical function as inter-related core concepts of tolerability.²⁰ Friends of Cancer Research convened a multistakeholder working group to develop a working definition of tolerability that incorporates the patient experience: “The tolerability of a medical product is the degree to which symptomatic and non-symptomatic AEs associated with the product's administration affect the ability or desire of the patient to adhere to the dose or intensity of therapy. A complete understanding of tolerability should include direct measurement from the patient on how they are feeling and functioning while on treatment.”^3(p4)

As part of the Cancer Moonshot initiative in 2018, the NCI selected four research teams to form a tolerability consortium with the goal of developing methods to better analyze and interpret clinician-rated and patient-reported adverse events as well as other clinical and patient-reported outcome (PRO) data to better understand tolerability.²² To help inform this goal, the multidisciplinary team proposed a tolerability model to guide clinical care for older patients with aging-related conditions that confer a high risk of AEs. The plan is to examine constructs in this model using data from a recently completed trial that enrolled patients age 70+ with advanced cancer starting a new treatment regimen (ClinicalTrials.gov identifier: NCT0205474; PI: Mohile).¹³ In this narrative review, the evidence is summarized that supports the integration of both clinician-rated and patient-reported metrics to inform treatment tolerability. Then, a novel operational model for tolerability is described, which includes clinical and PRO end points important to older adults with cancer. Because clinical trials inconsistently report on tolerability outcomes that are essential to inform decision making,²³ the hope is that this review will not only inform clinical care but will also guide the design of clinical trials that evaluate the risks and benefits of systemic treatment for older adults with cancer and aging-related conditions.^6,24

EVALUATING AGING-RELATED CONDITIONS THAT AFFECT TOLERABILITY WITH GERIATRIC ASSESSMENT

The prevalence of aging-related conditions, including comorbidities, disabilities, and geriatric syndromes, is highly variable among older adults with cancer. Older adults at the same chronologic age are likely to have significant heterogeneity in health status; chronologic age alone is insufficient to inform tolerability. More than 80% of older adults with cancer have a comorbidity, 50% have more than two chronic conditions, and almost one third have sensory impairments such as vision and hearing loss.²⁵ Older adults with cancer are at higher risk for disability or interference in activity resulting from a condition or health problem because of a higher prevalence of multimorbidity (ie, simultaneous existence of several medical conditions).²⁶ Comorbidity and disability increase vulnerabilities to stressors including cancer symptoms and toxicities from cancer therapies.¹⁰ Furthermore, older adults with cancer are at high risk for the development or worsening of geriatric syndromes, clinical conditions that are particularly common in older adults with functional and cognitive disability.²⁷ These conditions often cannot be attributed to one underlying disease or etiology and may reflect accumulated impairments in multiple aging-related domains and systems.²⁷ Comorbidities, disabilities, and geriatric syndromes are known to independently increase the risk of adverse outcomes of older patients receiving cancer treatment.¹⁰ For example, the risk of early mortality was shown to be significantly associated with an increasing number of comorbidities in 539 older patients with cancer, and 60% reported a functional limitation related to comorbidity.²⁸ Additionally, the influence of polypharmacy and potentially inappropriate medications, which increase in prevalence with multimorbidity, on cancer treatment tolerability is just starting to be explored.^29,30 In a recent systematic review, polypharmacy (defined as five or more medications in most studies) was found to be associated with chemotherapy toxicities, falls, and functional decline in older adults with cancer.³⁰ Unfortunately, aging-related conditions and syndromes fail to be captured consistently within cancer clinical trials and clinical care.¹⁰

Geriatric assessment (GA) uses validated tools to assess aging-related conditions including function, physical performance (eg, mobility), comorbidities, medications, cognition, nutritional status, psychologic status, and social support. In 2017, a Geriatric Core Dataset was developed with 42 international experts using Delphi consensus methodology, defining a core set of geriatric data to be methodically collected in cancer clinical trials of older adults to enable comparison across trials.³¹ The tools proposed for Geriatric Core Dataset included: (1) function: activities of daily living (ADL) and instrumental ADL (IADL), (2) comorbidity: updated Charlson Comorbidity Index, (3) mobility: Timed Up and Go, (4) cognition: mini-Cog, (5) nutrition: weight loss and body mass index, (6) mood: short Geriatric Depression Scale, and (7) social support: living alone versus support at home. In the United States, the Alliance for Clinical Trials in Oncology has advocated for a standardized geriatric assessment in therapeutic trials enrolling older adults.³² In therapeutic clinical trials, GA has been shown to increase understanding of tolerability through (1) better characterization of the study sample beyond age alone, (2) improved recognition of factors associated with poor tolerability, (3) integration in random assignment to guide treatment decisions, and (4) evaluation of how treatment affects outcomes important to older adults. From 2000 to 2017, only 41.5% of 41 phase II-III trials of systemic therapy exclusively enrolling older adults with cancer included an assessment of comorbidity or frailty and only 36.6% accounted for death from other causes.³³ Table 1 illustrates examples of studies that incorporated GA to improve understanding of cancer treatment tolerability.

TABLE 1.

Examples of Therapeutic Clinical Trials Incorporating GA to Enhance Understanding of Tolerability in Older Adults With Cancer

IDENTIFYING OLDER PATIENTS AT HIGHEST RISK OF TOXICITY

Several studies have demonstrated that variables from GA can identify older adults at highest risk of severe toxicity from chemotherapy. The CARG toxicity tool⁹ and the Chemotherapy Risk Assessment Scale for High-Age Patients³⁴ were each developed and validated in almost 1,000 older patients. These tools have been further studied in specific clinical scenarios (such as adjuvant chemotherapy for breast cancer³⁵) and in health care systems outside of the United States,³⁶ with variable results. Brief tools, such as G8³⁷ or the Vulnerable Elders Survey-13,³⁸ have also demonstrated utility for predicting toxicity. In these studies and indeed across the clinical trial landscape, toxicity has conventionally been measured using the NCI's Common Terminology Criteria for Adverse Events (CTCAE) library, which includes > 800 items graded for severity by clinicians, to assess toxicities. Clinician-derived measures of treatment toxicity have the advantage of consistency (assuming that a validated instrument such as NCI's CTCAE is applied as intended) and reliability. Treatment toxicity is routinely captured as part of standard of care, although less rigorously than in clinical trials. Although clinician-rated CTCAE can be used to inform patients about AE trajectories during treatment, it does not account for how trajectories or the cumulative impact of chronic or multiple lower-grade AEs influences the ability to continue treatment.³⁹ In older patients, further considerations for understanding tolerability include the association of AEs with functional decline; in one study of > 200 patients, oncologists modified dosing of treatment in older patients for whom AEs were longer-lasting and affected function (measured by ADLs) but maintained dose intensity for those AEs that resolved quickly.⁴⁰ Although GA-based tools like CARG and Chemotherapy Risk Assessment Scale for High-Age Patients may effectively identify patients most at risk of severe AEs, clinician-rated CTCAE incompletely captures the patient experience, and the NCI, FDA, and others have advocated for complementary approaches that integrate additional metrics including PROs.^19-21

USING CLINICAL END POINTS IN ADDITION TO TOXICITY TO INFORM TOLERABILITY

Indirect measures of treatment tolerability include mortality, hospitalizations, and dose reduction or discontinuation. All of these measures are confounded by the increasing prevalence and breadth of competing risk factors in older adults. Although in younger adults, unplanned hospitalization or dose reduction is more likely to be attributable to treatment-related AEs, in older adults, interconnected factors (such as comorbidities, frailty, polypharmacy, and geriatric syndromes) may be contributory, with no clear way to separate effects. These metrics provide important and useful information, however, about what is likely to happen to an older adult following initiation of treatment.

Although mortality is almost always reported as an outcome in clinical trials, hospitalizations are much less often reported.⁴¹ Moreover, hospitalization rates within randomized clinical trials can vastly underestimate the rates seen in real-world practice. In a systematic review of hospitalizations during treatment for metastatic lung cancer, the patients reported in real-world studies were older than their trial counterparts (71 v 63 years), and hospitalization rates were substantially higher.⁴¹ The adverse consequences of hospitalization for an older adult can include functional decline, delirium (often leading to long-term cognitive impairment), and lower QoL after discharge.⁴² Measuring hospitalizations could permit better extrapolation of the potentially cascading consequences of treatment for an older adult.

Measures of treatment adherence, including dose reductions, discontinuations, and relative dose intensity (RDI), have the advantages of consistency of definition. In cancers treated with curative-intent chemotherapy, RDI is a quality metric, inversely associated with recurrence risk in several solid tumor malignancies and high-grade lymphomas⁴³; if a certain threshold is not reached (≥ 85% has been a typical cutoff in the literature), survival rates are compromised, and patients may even accrue risks in excess of anticipated benefits. A main disadvantage of these measures is that they do not provide context for the decision to stop or reduce treatment. Upfront (primary) dose reductions, which affect overall RDI, are typically in the purview of the treating oncologist; in a study of 500 older patients, primary dose reduction was associated with age but not with GA measures, raising a concern for undertreatment rather than poor tolerability.⁴⁴ Dose adjustments or discontinuation occurring throughout the treatment course may reflect concerns from patients, clinicians, or both and may occur for reasons other than tolerability (eg, dose delays for vacations). It is not fully understood why older patients receive less cancer treatment than younger ones, despite equivalent benefit for fit older adults.⁴⁵ Elucidating these nuances would enhance the interpretability of these metrics; future researchers could extract information from the electronic medical record to identify the reasons for decisions to adjust dosing or discontinue treatment.

PATIENT-REPORTED MEASURES OF TOLERABILITY

The importance of the patient's perspective has been recognized in the ASCO Value Framework, which recommends clarifying patient goals and preferences during decision making for treatment; the framework for the palliative-intent setting includes symptom palliation and QoL metrics.⁴⁶ Elucidating what matters most to older patients has been identified as a key component of age-friendly cancer care by the Institute for Health Improvement.⁴⁷ Tolerability, like QoL, is influenced by an individual's worldview,⁴⁸ and thus, its meaning may not be fully shared by patients, caregivers, and health care providers or even between individuals within those groups. Reductionist attempts to develop tolerability metrics solely based on clinician-rated assessments are likely to be poorly generalizable. Tolerability includes an individual's determination of tradeoffs between the anticipated benefits of treatment and the risks, entailing careful consideration and discussion about the goals of treatment (curative v palliative) and what is known about treatment efficacy and tolerability as applied to the specific patient. Older adults often prioritize maintaining or preserving their independence, their cognitive abilities, and their day-to-day functioning. Furthermore, older adults may value maintaining QoL over length of life.^10,49 A revealing component mostly overlooked in these discussions is the patient's assessment of their usual level of tolerability. For example, does the patient generally have a high or low threshold for pain or discomfort, or has the patient been supersensitive or tolerant of medication or medication side effects in the past? Overall, the most dependable measure of tolerability derives from asking the patients directly—not only about what they are experiencing but also how they situate that experience within their preferences, goals, values, and previous experiences.

Symptomatic Toxicities as Assessed by Patient Report

Research to date has shown that patients often report more symptomatic toxicities, with earlier onset and at higher levels of severity and interference compared with clinician ratings,⁵⁰ and that PRO measures better reflect daily health status.⁵¹ The PRO version of the CTCAE (PRO-CTCAE) was developed to capture symptomatic AEs by patient self-report as a complement to clinician-rated CTCAE.^20,21 PRO-CTCAE, like CTCAE, is reported and analyzed descriptively by item; the two approaches provide different information about toxicity.^52,53 In some settings, PRO-CTCAE may better differentiate side effects between treatment arms in clinical trials.⁵²

In a trial of 718 older adults with aging-related conditions and advanced cancer starting a new treatment regimen,¹³ PRO-CTCAE data were collected for 27 symptomatic toxicities across 49 items. Data collection was feasible; 98% of patients provided complete data on all items. Symptom burden was high at baseline (Fig 1), with only 2% of patients reporting no symptoms. Multiple co-occurring symptoms were common; patients reported on average 9.2 symptoms. When asked about severity, 86% of patients reported at least one symptom with moderate or high severity and 19% endorsed at least four co-occurring grade ≥ 3 symptoms (severe or very severe) symptoms (Fig 2).

FIG 1.

Severity of PRO-CTCAE symptoms reported by older adults with advanced cancer starting a new treatment regimen. On the basis of data from a geriatric assessment intervention trial for patients age 70 years and older receiving chemotherapy or similar agents for advanced cancer: Reducing Toxicity in Older Adults Study (n = 718).¹³ Symptoms were captured using the National Cancer Institute PRO-CTCAEs. PRO-CTCAE, Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events.

FIG 2.

Distribution of the number of co-occurring PRO-CTCAE symptoms reported by older adults with advanced cancer starting a new treatment regimen. On the basis of data from a geriatric assessment intervention trial for patients age 70 years and older receiving chemotherapy or similar agents for advanced cancer: Reducing Toxicity in Older Adults Study (n = 718).¹³ Symptoms were captured using the National Cancer Institute PRO-CTCAEs. PRO-CTCAE, Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events.

In addition to PRO-CTCAE, other patient-reported symptom questionnaires can inform symptomatic toxicities (Table 2). Older patients report a different pattern of symptoms and more interference with function and QoL than younger patients with cancer.⁵⁴ PROs inform clinicians about patients' symptom burden, offering the potential to improve patient-doctor communication about symptomatic toxicities.²⁰ In a study (n = 766; median age = 61 years) of an electronic Symptom Tracking and Reporting System, the symptom intervention improved survival and QoL and decreased emergency room visits and hospitalizations; however, an examination of age revealed that older adults receiving the intervention did not have improved survival or a lower risk of emergency room visits.^55,56 With the FDA's recent launch of the Project Patient Voice online platform, patients, their caregivers, and clinicians can examine patient-reported symptom data submitted to the FDA from clinical trials.⁵⁷ Studies examining PRO and tolerability specifically in older adults with cancer remain scarce, and additional research is required to determine when and how these measures can be used in these patients and how they can be used to trigger interventions intended to improve tolerability in older adults.

TABLE 2.

Examples of PRO Measures With Descriptions and Key Findings in Older Adults With Cancer

QoL and Global Measures of Burden and Satisfaction

Identifying metrics that can globally encapsulate treatment burden remains a focus of researchers; both single items and composite measures are being investigated. For example, the single item “I was bothered by side effects of treatment” has demonstrated a consistent association with AE severity.⁵⁸ An International Society of Geriatric Oncology task force recommended longitudinal QoL monitoring in older adults with cancer in both trials and clinical practice.⁵⁹ Multiple valid scales are available for use in older patients, including the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 and its supplement for older adults (QLQ ELD15)⁶⁰ and the Functional Assessment of Cancer Therapy-General.⁶¹ QoL scales assess physical, functional, psychologic, and social well-being and thus may be well-suited to inform tolerability across multiple dimensions.

Decisional regret and its obverse, treatment satisfaction, are two metrics that may capture an overall gestalt of whether the treatment was worth it. The Satisfaction with Cancer Therapy scale measures the individual's satisfaction with treatment relative to their expectations of benefits and side effects.⁶² Older adults with improved treatment outcomes and higher QoL scores reported higher satisfaction scores.⁶³ Decisional regret, defined as “distress or remorse after a (health care) decision,” has been assessed using the validated Decisional Regret scale.⁶⁴ Older age is associated with increased levels of decisional regret following treatment for prostate cancer,⁶⁵ but data for other clinical situations are limited.

THE IMPORTANCE OF FUNCTION FOR UNDERSTANDING TOLERABILITY IN OLDER ADULTS

Clinical trials routinely include clinician-rated performance status (PS) as a global assessment of functioning. PS measures, however, do not adequately capture functional status in older adults. ASCO guidelines¹⁰ recommend collection of patient-reported ADL (ability to care for one's basic needs) and IADL (ability to live independently) before treatment. Requiring only minutes to complete, these PRO measures could easily be incorporated longitudinally in trial designs. Functional status (as measured by ADL or IADL) is associated with change in treatment plan by oncologists, toxicity from treatment, and mortality.⁶⁶ In addition to questionnaires, the results of objective tests to assess physical performance are associated with outcomes in older patients with cancer receiving treatment.¹⁰ These tests are brief, performed in the clinic, and accurately assess functional status (eg, Timed Get Up and Go, handgrip strength, and gait speed).⁶⁷ The Short Physical Performance Battery combines three physical tests of standing and walking and uses a summary score to capture lower extremity function in older adults.⁶⁷ Diminished handgrip strength has been shown to be a significant predictor of treatment toxicity⁶⁸ and dosing modifications.⁶⁹ Objective measures can be used both to identify those at highest risk of poor tolerability when measured before treatment and to track the impact of toxicities on function (when measured longitudinally).

METHODOLOGIC AND ANALYTIC PROGRESS AND CHALLENGES

Multiple efforts are currently underway to guide standardization of PRO and QoL reporting in cancer clinical trials in the United States.⁷⁰ The FDA is working to standardize patient-reported tolerability data collection by establishing a core set of measures for health care utilization (eg, hospitalizations) and inclusion of patient-reported symptomatic toxicities and functional measures.⁷⁰ How to best encode patient interpretations of tolerability within discrete item measures is an understudied question. However, these data can be analyzed longitudinally and in aggregate to define an overall shape and spectrum of the tolerability trajectory. For example, for any treatment, it is important to know the proportion of patients who reported certain toxicities, what their experiences were with those toxicities, at what point toxicities arose and/or dissipated, and how toxicities informed treatment modifications, discontinuation, and outcomes.

Several analytic concerns are paramount in older adult populations. The first is related to symptoms present at initiation of treatment. Causal attribution of symptoms in older adults receiving treatment may be particularly complex because of the competing contributions of multimorbidity, cancer burden, and treatment. Statistical techniques for estimating burden attributable to treatment toxicity include baseline-adjusted scores⁵² or constrained longitudinal data analysis.⁷¹ Since older adults may discontinue treatment early because of poor tolerability, appropriate statistical methods must be applied to account for dropout. Several statistical models have been suggested, including joint modeling of longitudinal and time to event outcomes^72,73 or QoL-adjusted survival.⁵⁵ Development of novel analytic methods may enhance understanding of early dropout because of unacceptable tolerability and QoL. Finally, statistical methods are needed to analyze and report cumulative toxicity burden^74,75 and link these measures with intuitive longitudinal data visualization to better convey the information to patients and clinicians.⁷⁶

Composite measures such as Overall Treatment Utility (OTU)⁷⁷ integrate both clinician-rated and patient-reported tolerability assessments and can be further evaluated in trials. OTU, developed as an end point for the FOCUS2 trial⁷⁷ (Table 1), incorporates a simplified clinician-assigned metric for benefit (based on the absence of clinical and radiologic disease progression and no general health deterioration), treatment toxicity grading, and PROs assessing treatment satisfaction and interference of side effects with daily activities. These measures are combined to determine good, intermediate, or poor OTU. Although this approach combines clinical and patient values of treatment, OTU has not been widely adopted to date. Additional composite measures include QoL-adjusted survival⁵⁵ or quality-adjusted time without symptoms or toxicity, a measure used in a study of older patients with lung cancer.⁷⁸ The selection of optimal composite measures to inform tolerability will require more research to facilitate wider implementation within clinical trials and integration into practice.

PROPOSED MODEL OF TOLERABILITY IN OLDER ADULTS

Based on the available evidence, a conceptual model was developed that integrates measures to identify older adults at risk for poor tolerability and outcome measures to evaluate tolerability. The model identifies seven constructs for consideration of tolerability in the older adult (Fig 3). Social determinants of health (any nonbiologic determinant) have been prioritized by ASCO⁷⁹ and the American Cancer Society.⁸⁰ Specific social determinants, which have been shown to affect outcomes in older adults, include financial insecurity, health literacy, transportation concerns, food insecurity, and social isolation.⁸¹ Ageism, or negative stereotyping and discrimination of older adults, is known to have deleterious effects on health outcomes and is a pervasive issue in health care institutions and the clinical trial infrastructure.⁸²

FIG 3.

Proposed model of treatment tolerability in older adults with cancer. ADL, activities of daily living; CTCAE, Common Terminology Criteria for Adverse Events; IADL, instrumental activities of daily living; PRO-CTCAE, Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events.

Tolerability is proposed as a multidimensional construct, persisting throughout the cancer care trajectory with three distinct windows (pretreatment, treatment, and survivorship) that offer opportunities for interventions. Projecting how tolerable a treatment may be for a specific older adult requires review of clinical trial data on safety and tolerability, which will (ideally, although not consistently) provide guidance relevant to older adults. GA¹⁰ may provide insight into anticipated tolerability. Patient values, goals, and concerns about desired treatment benefits and unacceptable risks are of paramount importance; in pretreatment discussions, it is essential to learn how the patient perceives tolerability and what matters to them, especially in the context of uncertainty about benefits and risks. For the clinical trial results to be generalizable to older adults, increased enrollment of older adults with varying health statuses and integration of PRO and functional assessments of tolerability will be necessary. Research on management of symptomatic toxicities and maintenance of function during active treatment can improve tolerability; however, adaptation of supportive care interventions may be necessary for older adults.⁸³ Understanding tolerability in the older adult requires a recognition that receiving treatment requires ongoing work⁸⁴ (ie, transportation to and from appointments, taking medications as prescribed, engaging in self-management recommendations, and reporting health changes) by the patient and caregivers. For older adults, these efforts may exceed their capabilities and resources, and caregivers may not be available or able to meet these demands.⁸⁵ Research is also needed to address long-term tolerability concerns that may persist and/or develop after treatment ends and into survivorship, especially with regard to changes in living situation and emergence of frailty characteristics.

In conclusion, understanding treatment tolerability in the older adult with cancer requires conversations to understand what matters to older patients, what they are experiencing, and how they feel about it. To date, research has identified higher toxicity in older patients with cancer and it has established mechanisms for assessing risk. Continuous research efforts are needed, however, to expand understanding of tolerability for specific treatment regimens, to identify persistent long-term tolerability concerns, and to implement management efforts in clinical practice. We offer a model that incorporates a multidimensional perspective on the components of tolerability to stimulate research and guide clinical implementation efforts. A widespread understanding of tolerability that expands beyond clinician-rated assessments is necessary to foster the routine integration of patient-reported symptomatic AEs and function in clinical research and care.

AUTHOR CONTRIBUTIONS

Conception and design: All authors

Administrative support: Supriya G. Mohile

Financial support: Supriya G. Mohile

Collection and assembly of data: All authors

Data analysis and interpretation: All authors

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

Understanding Treatment Tolerability in Older Adults With Cancer

The following represents disclosure information provided by the 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).