Definition

Cancer-related fatigue (CRF) is experienced by most patients with cancer at any point from the time of diagnosis, during treatment, through survivorship, or at disease progression. The National Comprehensive Cancer Network (NCCN) guidelines consensus definition of CRF is a “distressing persistent, subjective sense of physical, emotional and/or cognitive tiredness or exhaustion related to cancer or cancer treatment that is not proportional to recent activity and interferes with usual functioning.”¹ Fatigue with these characteristics does not appear to occur in the general population. In a large analysis of almost 35,000 patients, fatigue was measured with use of the Functional Assessment of Chronic Illness Therapy (FACIT) questionnaire and comparisons were made between anemic patients with cancer, nonanemic patients with cancer, and a cohort from the general population (with or without medical conditions) (N = 1010). The patients with cancer, regardless of their hemoglobin level, experienced significant fatigue (P > .001) compared with persons without cancer.² CRF can negatively affect functional status, resulting in limitations in ambulation and exercise, oral intake, socialization with supportive friends and families, ability to maintain a job, and quality of life. Of critical importance, fatigue may prevent full doses of anticancer therapy from being delivered, which could potentially affect survival or tumor control.

Incidence

The exact incidence of CRF depends on when in the cancer trajectory it occurs and by what methods patient-reported experience is measured. It does not appear that any time exists on the cancer continuum when patients are not challenged by fatigue (see Fig. 45-1), and CRF occurs with all types of cancers and in all stages. Most likely CRF is underreported, because both patients and health care providers expect fatigue as an inevitable result of a malignancy diagnosis. Discussions between patients and physicians regarding fatigue may be limited, given that few efficacious interventions exist to improve fatigue.

In a review of 27 publications in which CRF was the primary outcome, the incidence ranged from 4% to 91%.³ This wide range probably is the result of significant variation in the patient populations studied, including type of malignancy, stage of disease, receipt of cancer therapy, the type of treatment modality received, and patient gender. In addition, these studies used various survey tools to record patients’ specific description of their fatigue. Currently, 14 validated scales exist to measure CRF, and each has a different construct for evaluating fatigue.⁴ In all, it has been estimated that 80% to 90% of outpatients with cancer who received either chemotherapy or radiation therapy report the presence of various levels of fatigue.⁵ Additionally, severe fatigue has been recorded in 75% of hospitalized patients on a palliative care unit.³ One population studied extensively has been breast cancer survivors. In a study of more than 750 breast cancer survivors, almost 20% of patients who were free of disease for 5 to 10 years after completing therapy had significant residual fatigue that impaired their ability to function.⁶ With new anticancer therapeutic agents causing or associated with fatigue, longer treatment periods, and an increasing number of survivors, CRF is currently and will continue to be a substantial clinical issue.

Etiology

The causes and contributing factors of fatigue are undoubtedly multifactorial, and thus its etiology is not well delineated. Understanding the causative factors is important for finding effective clinical therapies or preventative strategies for CRF. Patients with cancer experience multiple comorbid conditions, in addition to the malignancy itself, that can cause or exacerbate fatigue. This combination of factors may prevent or delay detection of the ultimate contributing disorders. Patients may acquire these medical conditions after receiving therapy or they may have these conditions at the time of presentation with cancer; the term “cancer-related” can encompass all of these factors, because the exact time of onset in the cancer trajectory may be unknown. Defining CRF is quite challenging because many patients with cancer present with fatigue before primary adjuvant therapy is begun. Regardless of the source, clinically recognizing separate causes of fatigue at presentation will help uncover treatable situations, such as anemia, hypothyroidism, depression, concomitant medications, and infection. Remediating potentially reversible conditions is the initial approach to CRF according to NCCN guidelines.¹

The literature regarding possible CRF mechanisms contains a few small clinical trials and little preclinical work. The published data suggest multiple nonspecific interactions between cancer, the effects from cancer therapy, and the patients’ responses to both. Unfortunately, the consequences of these interactions have not been convincingly correlated to patient report of fatigue. As well, findings of putative mechanism-directed therapies have not been successful to date. One possible mechanism currently being studied, however, involves the inflammatory response to cancer and its therapy, especially cytokine activation. Clinical studies have revealed increases in interleukin-1 (IL-1) receptor antagonist and C-reactive protein in fatigued patients with cancer who are undergoing radiation therapy.⁷ CFR has also been correlated with tumor necrosis factor (sTNF-RII), immunoregulatory CD4 lymphocytes, and white blood cell subsets.⁸ Proinflammatory proteins have been noted in breast cancer survivors with no evidence of disease who have experienced fatigue 3 to 5 years after cancer therapy.⁹ The reasons why inflammatory responses are still maintained after treatment and resolution of the cancer have not been fully explained. In a study of a group of breast cancer survivors with or without fatigue, at least one cytokine polymorphism at IL-1β–511 (P = .007) was associated with reported fatigue.¹⁰

Regardless of the quality or quantity of the pathophysiology involved, patients consistently note multiple sensations that often occur not in isolation but in clusters of complaints. Given the interwoven nature of the facets of CRF, causative perturbations may exert a systemic effect on multiple organs and tissues to obtain the phenotype of fatigue. Clinically, this cluster appears similar to inflammatory-induced symptoms referred to as “sickness behavior” and may encompass depression, sleep disturbances, and cognitive dysfunction.¹¹ A sickness behavior animal model induced by lipopolysaccharide injection has been associated with behavioral changes (e.g., listlessness, decreased oral intake, and sleep disturbances) and increased levels of circulating proinflammatory cytokines.¹² Although this preclinical model has been widely used for CRF studies, a lack of valid preclinical models of fatigue remains. The discovery of new animal models of fatigue continues to be a challenging yet key aspect of CRF research.¹³ Because CFR is subjective, the patients’ description of fatigue (at the bedside) should guide the generation of hypotheses (at the bench) to begin to uncover the etiology or etiologies of CRF, even though some questions surrounding CRF mechanisms do not easily lend themselves to preclinical assessment. Modeling human situations such as lack of motivation to perform, comorbid conditions, and concomitant medications cannot yet be recapitulated. Valid preclinical models can help to accelerate an improved mechanistic understanding of fatigue. This understanding will allow for more precise interventions to be brought into clinical testing. Several grants and research projects are currently exploring inflammation as the primary driver of CRF.

A second potential mechanism under active study is cardiovascular fitness and physical performance. For usual strength and function, a fundamental requirement is the ability to deliver sufficient oxygen (O₂) to active skeletal muscles so that adenosine triphosphate can be resynthesized for energy. Given that patients with cancer often describe fatigue as a feeling of muscle weakness or lack of energy, it appears reasonable that impairments in the cardiopulmonary system may contribute to a decrease in either the delivery or utilization of O₂. The delivery or transport of O₂ in organ systems may be significantly impaired from the cancer or the cancer therapy; treatments may unintentionally target the vasculature, cellular mitochondria, and other tissues critical to the delivery of O₂.

Patients with cancer are often older and commonly present with a diverse range of age-related changes, similar to the general adult population, that limit cardiorespiratory fitness. However, these normal aging consequences are dramatically compounded by the consequences of anticancer therapies, tumor burden in advanced disease, comorbid conditions such as antecedent cardiac disease and pulmonary insufficiency, and decreased activity/deconditioning.¹⁴ These situations simultaneously affect the O₂ cascade.

A growing body of evidence shows that patients with cancer have significant impairments in cardiorespiratory fitness.¹⁵ The data have been collected by methods using objective determination of cardiopulmonary function,¹⁶ thus providing reproducible, standardized information. International guidelines defining the proper conduct of cardiorespiratory fitness testing have also been developed.^19–19 An incremental cardiopulmonary exercise test with gas exchange measurement to assess peak oxygen consumption (VO_2peak) constitutes the gold standard for assessment of cardiorespiratory fitness. Although routine use of cardiopulmonary exercise testing is not feasible, it is a highly informative research tool to measure physiological processes that may deepen the understanding of CRF. One study reported that the mean VO_2peak in 346 presurgical patients with non–small cell lung cancer was 25% to 44% below that for age- and sex-matched normative data for sedentary persons.²⁰ As well, the mean VO_2peak was 27% below that of age-matched sedentary healthy persons compared with patients with stages I to IV breast cancer, with the lowest mean VO_2peak seen in women with metastatic cancer. Additionally, it was determined that a 40-year-old woman with breast cancer has a cardiopulmonary capacity similar to that of a healthy sedentary 70-year-old woman.²¹ It is not clear if these observed marked reductions in cardiorespiratory fitness are due to limitations in pulmonary, cardiovascular, or peripheral tissues. However, the impairment is probably multifactorial with no one single organ component of O₂ transport/utilization yet implicated as being solely responsible.¹⁴ It may be that a combination of exercise plus pharmacologic and psychosocial interventions could be tested for their role in initiating and maintaining CFR.

If damage to some aspect of the cardiorespiratory system by cancer or its treatment were contributory, then it follows that either measure of the individual organ components and/or the total functional capacity of these organs (cardiorespiratory fitness) should be correlated with patient-reported fatigue. Jones et al.²² recently reported that in patients with newly diagnosed malignant glioma, prior to therapy, decreased peak VO_2peak was correlated with patient-reported fatigue (measured by the FACIT fatigue scale). Also, measures of body composition such as lean body mass, as well as total muscle cross-sectional area by magnetic resonance imaging, were inversely correlated with reported fatigue. In a subsequent study, functional capacity, as measured by a 6-minute walk test, was inversely correlated with fatigue in 171 patients with recurrent malignant glioma.²³ Similar findings have been reported in other cancer populations with different measures of cardiorespiratory fitness and functional capacity, including patients with advanced lung cancer²⁴ and early-stage breast cancer.²⁵ These studies support the correlation between CRF and diminished cardiorespiratory activity and strengthen the hypothesis that increasing physical function during or immediately after therapy (primarily through physical exercise) may have a significant effect on reducing CRF.

Evaluation

Fatigue is an experienced sensation. According to NCCN guidance, patients are asked to assess their current level of CRF on a 0 to 10 numeric rating scale (NRS) that is portioned as follows: 0, no fatigue; 1-3, mild fatigue; 4-7, moderate fatigue; and 8-10, severe fatigue.¹ This NRS is a screening tool that has also been utilized as an eligibility criteria measurement for fatigue in patients prior to clinical trial enrollment. Based on the results of this screening tool, NCCN has published guidelines for evaluating patients with fatigue that includes reassessment for any changes in fatigue.¹ The approach to CRF assessment is similar to the evaluation of pain, a common pathophysiological condition that is defined by patient report. Pain assessment is a high priority in cancer management and offers an opportunity to ask the patient questions regarding fatigue at the same time. Like the sensation of pain, the severity of CRF is influenced by physical and psychosocial factors. As some of these factors fluctuate, constant monitoring with changes in the management plan is essential for both conditions. “Early detection” and intervention for modifiable fatigue-associated conditions is a sound strategy and has clear potential for patient benefit.

The NRS screening tool, unfortunately, does not convey the multiple dimensions of either fatigue or pain. For example, it is not clear how functional capabilities relate to the numeric values associated with CRF levels. Distinguishing these other factors can be clinically relevant, depending on the interventions available to improve that particular element. For example, fatigue that is not specifically cancer related, such as from psychiatric and mood disorders, needs to be managed per the accepted standard of care.

To further define CRF and to highlight assessment of other conditions, the Fatigue Coalition has established the International Classification of Diseases–10 (ICD-10) criteria for cancer-related fatigue²⁶ (see Box 45-1). These criteria include four classifications of symptoms (A, B, C, and D). Patients indicate the A1 symptom of significant fatigue, decreased energy, or increased need to rest without activity level change, plus six or more A-level symptoms that have been present every day or nearly every day in the same 2-week period during the past month. The symptoms the patient must have to fulfill the diagnosis for CRF by the ICD-10 criteria are common patient-reported sensations associated with fatigue, such as “…experience of sleep as unrefreshing or nonrestorative.” Additionally, some of the diagnostic criteria could be consistent with concomitant or causative conditions such as depression. Given the overlap of some of the qualities of CRF with mood and other psychiatric disorders and cognitive dysfunction, the extraction of differences is problematic, yet quite clinically significant.

Utilization of the ICD-10 list of criteria in its entirety has not been widely adopted. When applying the full set of criteria (A-D), the incidence of CRF dramatically decreases (2%), as opposed to 37% noted with only the A1 criteria, ten other A-level symptoms, and B-level criteria combined.²⁷ Use of this assessment tool, however, can eliminate known conditions and improve diagnostic accuracy and clinical classification. A harmonious definition of CRF can refine the incidence and define populations/subpopulations for specific clinical trials and for routine clinical practice. The science of CRF will be aided by the development of accurate diagnostic tools with pertinent physiologic assessments and common outcome measures.

Measuring Fatigue

Fatigue is described in different ways by patients and physicians. Patient-reported outcomes (PROs) are the patients’ perspective of their health status or benefit of any interventions. Patient-reported information is requisite for appraisal of subjective data such as experienced symptoms and their impact on function and quality of life. PRO instruments are surveys or questionnaires targeting specific information obtained from the patient. As with objective data, PROs must be interpreted and validated. These instruments, which are more specific than Health-Related Quality of Life instruments, must accurately reflect the patients’ perspective of their condition. Once CRF has been diagnosed, PROs can be used to guide subsequent assessments.

It has been reported that three one-dimensional (focus on physical functioning) patient-reported measures for fatigue have been validated in patients with cancer⁴: the Brief Fatigue Inventory (BFI),²⁸ the Functional Assessment of Cancer Therapy-Fatigue,²⁹ and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire–C30.^30,31 Although these tools are used frequently in placebo-controlled intervention clinical trials for CRF, their utility in the clinic has not been established. As well, there is not one agreed-upon measure used in the majority of trials that could allow for comparison between various interventions for fatigue, in part because of the difficulty in integrating the cognitive and emotional aspects of CRF with the physical and functional fatigue in a single validated measure. For specificity, these measures must be validated in the actual cancer population to be studied (e.g., patients with solid tumors who are receiving chemotherapy vs. postoperative immobilized patients without cancer). A future area of research should include devising outcome measures in CRF for clinical practice. Therefore applied research methodology is a critical part of the multidisciplinary approach required to ensure accurate measurement of the patient experience and to advance the science of CRF.

Therapy for CRF

Evidence-based treatments for CRF are scarce. Even after specific conditions for which known therapies exist are identified and properly managed, many patients with cancer still note fatigue. Various interventions have been studied, including pharmacologic agents for symptom palliation (i.e., psychostimulants) and those that target possible etiologies (i.e., serotonin reuptake inhibitors). The complete list of consensus interventions is found in the NCCN guidelines.¹ Until improved evidence-based therapies are developed, the best intervention may be enrollment in a fatigue-ameliorating clinical trial when such a trial is available.

Conclusion

CRF is a complex, multicomponent problem that affects most patients with cancer at any time point in their cancer care. Because the mechanisms leading to CRF have not been well defined, few evidence-based interventions are available for CRF. Research efforts need to be expanded to identify potential molecular signatures, patient genotypes and phenotypes, validation of other patient reported measures, and components of the cardiopulmonary system involved in CRF. ICD-10 and NCCN guidelines provide the best assessment methods. Currently, the strongest evidence favors some type of exercise for patients who are able to exercise, recognition of remediable conditions by standard intervention, and educational/psychosocial interventions. Patients reporting fatigue should be enrolled in longitudinal and interventional research studies for CRF whenever possible.

Acknowledgment

We appreciate the administrative assistance of Ms. Pamela Maxwell.