Cancer-related fatigue (CRF) is a symptom frequently experienced by survivors regardless of tumor type or its treatment.1 This report is focused on CRF among the increasing number of breast cancer survivors, which currently number >2.5 M in the United States. CRF is described as a clinical entity characterized by tiredness to exhaustion not precipitated by activity or, if after activity, out of proportion to the level of exertion. It is not relieved by rest and may be worsened with rest. CRF may result from the tumor itself, antedating its treatment. Experts in the field2, 3 have outlined the benefits of using a case-definition approach to CRF. This approach suggests 4 criteria to establish the diagnosis, including: 1) a period of 2 weeks or longer within the preceding month during which significant CRF or diminished energy was experienced each day or almost every day along with additional CRF-related symptoms; 2) the experience of CRF results in significant distress or impairment of function; 3) the presence of clinical evidence suggesting that CRF is a consequence of cancer or cancer therapy; and 4) CRF is not primarily a consequence of a concurrent psychiatric condition, such as major depression.3 Thought leaders recently have suggested that clinicians agree on the definition adopted by the National Comprehensive Cancer Network (NCCN)4 and use the above criteria for assessing clinically significant CRF.5 There is value in using a consistent definition and methods to determine cases of moderate-to-severe CRF and its responsiveness to treatment.
CRF has been documented as the most distressing symptom reported by cancer patients and is more disruptive than pain to normal routines and quality of life (QOL).6 de Jong et al7 reported that 58% to 94% of patients with breast cancer experience CRF during treatment, and between 56% and 95% have CRF after they receive adjuvant chemotherapy. Its overall prevalence is 48%, although it is higher in certain tumors (eg, pancreatic, breast, lymphoma) and during treatment. This is true whether the treatment consists of single or combined regimens, including surgery, chemotherapy, radiation, and/or biologic agents. Symptom onset may be contemporaneous or remote from treatment. CRF is frequently persistent and does not resolve with tumor control or treatment cessation. Cancer survivors, defined here as anyone who has been diagnosed with cancer from the time of diagnosis until death,8, 9 frequently report that they never return to their prediagnosis energy level, and this diminishes their QOL.10
Contributing Factors to Cancer-Related Fatigue
Effective treatment begins by identifying the factors that contribute to this debilitating symptom. Research supports CRF as a multifactorial construct influenced by numerous biopsychosocial factors. Possible causal factors include physical conditions, affective and cognitive states, and distress or spiritual suffering.11 The work of several investigators7, 11 points to the affective dimension as a major contributor; however, affective state does not account for most of the variance among those who report higher CRF.
Investigators have established an important role for proinflammatory cytokines as potential contributors to CRF. Elevated cytokines are among several important findings in fatigue-promoting medical conditions.12, 13 Interleukin-1 receptor antagonists, soluble tumor necrosis factor receptor type II, and neopterin levels were significantly higher in a cohort of fatigued breast cancer survivors compared with the levels in nonfatigued survivors 5 years after diagnosis.14, 15 One hypothesis is that CRF may be the final common pathway of several contributing systems, including the central nervous system, immunoregulatory pathways, and the neuromusculoskeletal and cardiopulmonary systems. Current thinking suggests that both the physiological and immunologic systems, which control inflammatory pathways, and the affective components, which implicate 5-OH tryptamine and catecholamines and regulate the central nervous system, are involved in CRF.16
Other contributors to CRF are metabolic in nature. Data suggest that CRF may result from abnormalities of energy production and use that favor tumor growth.17 CRF, thus, may reflect disordered mitochondrial biogenesis, which upsets the balance between aerobic and anaerobic metabolism. Muscle tissue prefers oxidative phosphorylation as an energy source. However, researchers have observed that tumor cells seem to have an advantage when there is a high rate of conversion of glucose to lactate, favoring the glycolytic pathway.18, 19 This theory would predict the same inadequate production of adenosine triphosphate from oxidative phosphorylation that has been associated with a favorable environment for malignancy and may simultaneously deprive the body of the energy production needed to maintain muscular and neurocognitive function.20 Mitochondria are critical for energy production, which is essential for protein synthesis. Abnormalities in energy generation are associated with diminution of muscle biosynthesis, such as in aging muscle (sarcopenia).21 Muscle is a rich source of mitochondria, and abnormalities have an impact on cardiac performance. Deficits, which may or may not respond to training, are sensitive to neuromuscular measures of skeletal muscle strength and endurance.22
This theory is supported by observations that either resistance training or aerobic exercise stimulates mitochondrial biogenesis.23-26 Aerobic exercise is an effective strategy for treating CRF and may act through several pathways, including decrease in serum proinflammatory cytokines,27 mobilization of visceral fat, and/or increased insulin sensitivity and glucose uptake.28 Exercise is of specific benefit to women with breast cancer because of the adverse associations between obesity and breast cancer29, 30 and the recurrence and mortality from breast cancer associated with obesity.29, 31 Physical activity reportedly helps regulate sleep, mood,32 and diurnal variation of cortisol.33
Another important contributor to bioenergetics pertains to insulin sensitivity. CRF may be associated with abnormalities of glucose uptake by muscle essential for normal, efficient function. Those with relative insulin resistance have abnormalities of glucose uptake by muscle and increased body fat.34 These abnormalities may be associated with CRF and possibly may be treated with hypoglycemics or agents to increase insulin sensitivity.
On a functional level, CRF has been associated with sleep disturbances.35, 36 Other characteristics of the clinically fatigued breast cancer survivor include: younger age, a body mass index >25 kg/m2, a higher number of positive lymph nodes, an increase in limb volume, an increased in the white blood cell count over baseline, and a low level of physical activity and physical function.37 Some of these factors are remediable. Early intervention for lymphedema may prevent CRF. Increased levels of moderate and strenuous activity may mitigate fatigue through various critical metabolic and inflammatory pathways, including proinflammatory cytokines,27 mobilization of visceral fat, and increased insulin sensitivity and glucose uptake.28
Many problems are commonly experienced by breast cancer survivors and may be associated with CRF.38 The number, type, and incidence of problems may vary by stage of disease, time since diagnosis, treatment, and age. Problems associated with fatigue may be the result of the cancer itself, its treatments, and/or other comorbid conditions.39, 40 Problems, depending on the number and severity, may influence the CRF experience and overall QOL41 (see the articles by Demark-Wahnefried et al, Hayes et al, McNeely et al, Schmitz et al, Schmitz, Speck et al, Stubblefield et al, and Winters-Stone et al in this supplement42-48).
Two large samples of women with early stage breast cancer within 6 months of diagnosis (n = 1219)49 or approximately 7 months after undergoing primary surgery (n = 1372) were surveyed for symptoms and QOL.40 Both samples reported experiencing multiple symptoms, and many women reported that ≥3 symptoms had a negative impact on QOL. Over half the women in the first study reported the following 6 symptoms: systemic treatment side effects (87.7%), fatigue (81.7%), breast symptoms (72.1%), sleep disturbance (57.1%), arm symptoms (55.6%), and pain (50%). Findings were similar in the second sample; almost half of women reported that symptoms interfered with daily functioning or mood and at least 1 of the following symptoms: difficulty sleeping (61.2%), hot flashes (43%), nausea/vomiting (27.4%), vaginal dryness (25.7%), or arm problems (23.9%); fatigue was not included in that study.
Treatment side effects
Cancer treatments are associated with a variety of side effects. Surgery may contribute to pain, lymphedema, and altered upper quadrant function; radiation therapy may contribute to skin changes, pain, and lymphedema; chemotherapy may contribute to peripheral neuropathy and cognitive impairment; and hormone therapy may contribute to arthralgia and hot flashes. Depending on menopausal status and treatment, the most common symptoms from hormone therapy include hot flashes, vaginal dryness, and dyspareunia.50 Hot flashes have been associated with sleep disturbances and CRF and are commonly reported by women who are receiving aromatase inhibitors.51 Women who received both radiation and chemotherapy, compared with those who received radiation alone, were more likely to be fatigued up to 10 years after treatment.52 Later treatment side effects may include osteoporosis and cardiovascular disease. Medications used to manage many of the side effects also contribute to the fatigue experience.
Arm/upper quadrant symptoms
After undergoing breast cancer surgery, women may experience problems associated with CRF, including pain, range-of-motion limitations, lymphedema, and functional disability.53, 54 Women are at greater risk for these problems if they undergo axillary lymph node dissection, if they receive radiation to the axilla, if they have a higher body mass index, if they experience a postoperative infection, or if they develop a local recurrence. Upper-quadrant morbidity negatively impacts daytime functioning, activities, and QOL.55-60 Lymphedema in survivors has been linked to worse emotional well being and QOL.61
Sleep problems affect most patients during chemotherapy.62, 63 Many breast cancer survivors experience disturbed sleep, and a subset meets the criteria for insomnia.64, 65 The problem may be pre-existing or may result from the cancer and its treatment. Hot flashes and pain affect sleep efficiency and quality and contribute to CRF and daytime dysfunction.66, 67 Sleep problems have been associated with CRF, depression, and poorer overall QOL.68, 69 Some of the effect of pain on CRF reportedly was mediated by sleep disturbance, but pain also had a direct effect on CRF.70
Mood disturbances and fatigue frequently co-occur, and mood disturbance may amplify the perception of fatigue. Some breast cancer survivors have pre-existing fatigue that may accompany depression or anxiety, whereas others develop these mood disturbances after diagnosis. Among patients with newly diagnosed breast cancer (n = 236), 27% reported a previous depression, 11% currently met criteria for major depression, and 10% met criteria for post-traumatic stress syndrome.71 These diagnoses were associated with marked functional impairment compared with women who did not have a diagnosis of depression or post-traumatic stress disorder (PTSD). Most of the women with PTSD reported that their symptoms were from a combination of the breast cancer and a previous event. It is noteworthy that 56% of the women with major depression who were on antidepressants still met criteria for major depression.
Among another sample of 1083 women with breast cancer, 38% reported moderate-to-high anxiety, 22% had moderate-to-high depression, and 12% had PTSD not associated with time since diagnosis.72 Greater than 40% of these women had 1 psychological diagnosis, and 7.8% met criteria for all 3 diagnoses. A review of studies examining the prevalence of depression in breast cancer survivors indicated that >60% experienced mood disorders, and 13.5% met criteria for major depression. Fatigue almost universally accompanies these conditions.73
Cancer does not occur in a vacuum. Chronic comorbid conditions, such as hypothyroidism, and related treatments may contribute to fatigue, especially in an aging population. One study that included >1500 patients reported that >66% of patients with cancer had at least 1 comorbid condition, and approximately 33% had ≥2 comorbid conditions.74 The most common comorbid conditions included hypertension, cardiovascular diseases, respiratory diseases, diabetes, cerebral-vascular diseases, and arthritis. The most frequent pattern of comorbidities in women, with incidence increasing with age, included hypertension with diabetes, other cardiovascular disease, and arthritis. Hypothyroidism is a frequent comorbidity in postmenopausal women.
Among 1800 postmenopausal breast cancer survivors, 93% had 1 or more comorbid conditions, and 43% had ≥4 comorbid conditions.75 Hypertension was the most common condition followed by arthritis, and aging was associated with a higher number of conditions. A history of 1 or more previous cancers was reported in 15% of these women. During the 30-month postdiagnosis period, 14.6% of patients died (51.3% of breast cancer, 17.1% of heart disease, and 8.4% of another cancer). Hypertension, heart disease, and diabetes were identified as the 3 most common comorbid conditions in 799 women when diagnosed with breast cancer.76 Hypertension predicted long-term fatigue in women with breast cancer up to 10 years after diagnosis.52
It is difficult to isolate the multiple contributors to fatigue in breast cancer survivors with comorbid conditions. It is important to obtain a thorough history of the diagnosis and treatment of comorbid conditions preoperatively and to review their management regularly, especially in older women. This population benefits from coordinated care between the oncology team and the primary care provider, and referrals to rehabilitation specialists should be made as needed.77-79
Describing important characteristics of CRF has posed significant practical and methodologic problems. First, the primary source of information comes from patients' self-reports, and what they mean by “fatigue” varies. Self-report is the cornerstone of assessing CRF but may represent only 1 approach to measuring the consequences and corollaries of this symptom. Self-reports of CRF, biologic measures, and physical and cognitive performance do not correlate well, perhaps because each measure is capturing different aspects of the phenomena of fatigue.80
Second, universal agreement is lacking about a conceptual framework for CRF, and this hinders progress in defining and measuring the phenomenon. One approach has been to postulate that there are 2 types of CRF: central and peripheral. Central fatigue is the failure to initiate and/or sustain attentional tasks and activities requiring self-motivation. Peripheral fatigue is a decline in physical performance resulting from inadequate cardiorespiratory, muscle-metabolic, and physiologic function. The value of these distinctions is that they potentially can be differentiated using a combination of patient-reported and clinician-reported measures. It has been demonstrated that central fatigue and peripheral fatigue are distinct and definable in objective terms.81 Each is associated with different biologic correlates and physical performance indicators and is described using different adjectives (eg, “I am exhausted all the time” vs “I get tired running”). Care in selecting fatigue measures will assist in providing a comprehensive evaluation. Good reviews of this literature are available.82-84
Health care providers do not routinely assess patients to determine causes of fatigue or prescribe effective treatments for it.85 Establishing a standard, reliable, and clinically feasible instrument to measure CRF will increase the likelihood that health care providers will use it. Collecting data in a systematic way is likely to assist in identifying correlates of, and possibly contributors to, CRF and may lead to the selection of effective interventions.
The NCCN guidelines for CRF recommend screening all patients at regular intervals.4 All patients/families need to receive education, counseling, and general strategies for managing CRF. When patients report moderate or severe CRF, a focused history and assessment need to be performed of treatable, contributing factors (medications/side effects, pain, emotional distress, anemia, sleep disturbance, nutritional deficit/imbalance, decreased functional status, comorbidities). Evidence-based interventions to reduce CRF should be selected based on individual needs and desires.
A range of nonpharmacologic and pharmacologic interventions has been tested to reduce CRF. The widely disseminated NCCN Cancer-Related Fatigue Guidelines in Oncology4 and Oncology Nursing Society (ONS) Putting Evidence into Practice (PEP) fatigue guidelines86 reflect evidence of the efficacy of interventions to modify CRF. The NCCN CRF guidelines are written for all diagnosed cancer patients according to their phase on the cancer trajectory: on active treatment, post-treatment, or at the end of life. Most nonpharmacologic interventions for reducing CRF have been tested in patients with local/regional disease, whereas pharmacologic treatments have been tested in patients with advanced/metastatic disease. In the paragraphs below, we summarize current knowledge regarding interventions to treat CRF in mixed types of cancers, with a focus on breast cancer patients.
Nonpharmacologic interventions to decrease cancer-related fatigue
No gold standard exists for the nonpharmacologic treatment of CRF in patients with breast cancer. Several nonpharmacologic interventions have been tested and have demonstrated positive effects in treating CRF. Numerous recent systematic reviews and meta-analyses have evaluated the efficacy of interventions to reduce CRF in adults with mixed types of cancer.86-100 Breast cancer patients have participated in most studies. These interventions have been categorized in 2 major, clinically applicable domains: physical activity enhancement and psychosocial therapies.
Physical activity enhancement involves the use of some form of exercise. Interventions include various modes, intensity, and timing of exercise activities, such as walking, cardiovascular and/or flexibility training, and resistance training. Exercise interventions have included both supervised and home-based programs.89
Two recent meta-analytic reviews concluded that exercise interventions had a near moderate effect size (ES) in reducing CRF compared with the control condition, usually “standard care.”89, 92 Kangas and colleagues reported ES in 16 randomized controlled trials (RCTs) (N = 1001) with a weighted, pooled mean ES of −0.42 (z = −4.41; P < .001). A Cochrane meta-analysis of 28 RCTs89 reported a smaller overall ES with a standard mean difference of −0.23 (95% confidence interval [CI], −0.33 to −0.13). Benefits favored programs with multiple exercise components, at least partially home-based, individualized, and >8 weeks long.89, 92
The evidence suggests that physical activity enhancement, and specifically exercise, has benefit in managing CRF both during and after cancer treatment. Current evidence does not clarify which elements of exercise are most effective, including type (aerobic and/or resistance), mode, and dose,(frequency, intensity, length of sessions). Physical activity interventions need to be individualized based on the level of physical capacity and pre-existing conditions and should begin at low intensity levels with gradual progression. Most studies prescribed less than moderate intensity aerobic activity, as recommended for healthy adults, probably because participants were advised to start at a lower level and to progress according to their physical capacity. Few adverse events from exercise, including back, wrist, lower-leg, and rotator cuff injury, were reported.25 The NCCN CRF guidelines4 call attention to specific indications for referral to physical therapy and instances in which exercise may be contraindicated or used cautiously. There is insufficient evidence to evaluate the efficacy of physical activity as a supportive care intervention for patients with cancer who are receiving palliative care.101
Recent systematic reviews and meta-analyses have focused specifically on exercise in patients with breast cancer.25, 100, 102, 103 Duijts and colleagues observed statistically significant results with physical exercise interventions for CRF in patients with breast cancer (ES, −0.315; 95% CI, −0.532 to 0.098; P = .004). Those findings were consistent with results from other studies in breast cancer.24, 25, 100, 103, 104 In addition to effectively reducing CRF, most studies report that exercise is feasible for women with stage I through IIIA breast cancer during and after cancer-directed treatment.
Psychosocial therapies comprise a diverse set of educational, supportive, and behavioral approaches grouped for clinical application in the NCCN CRF guidelines. Kangas and colleagues92 conducted a systematic review of 113 studies that included a CRF outcome. The studies included RCTs and non-RCTs. That meta-analysis included findings from 41 RCTs that tested the efficacy of psychosocial interventions administered during and after treatment in adults with mixed cancer diagnoses. The effect of psychosocial interventions was in the small-to-moderate range and was clinically meaningful. The individual ES for the 41 psychosocial RCTs (N = 3620) had a significant, weighted, pooled mean ES of −0.31 (z = −9.62; P < .001).
Interventions were grouped in categories of cognitive behavioral therapy (CBT), supportive-expressive therapy, education/counseling, behavioral/relaxation therapy, massage, and mental restorative treatments. Psychosocial interventions that exhibited the most benefit across several RCTs included CBT, education/counseling, and supportive/expressive therapy. Behavioral interventions, such as relaxation and imagery and general fatigue education, produced smaller and sometimes nonsignificant effects.92 Massage and therapies for mental rest and restoration also demonstrated benefit. Effective psychosocial interventions for reducing CRF in adults during cancer treatment contained certain specific elements, namely, tailored education about CRF, self-care and coping techniques, activity management, and energy-conservation activities.90
A meta-analysis focused on the effectiveness of behavioral techniques on CRF in breast cancer102 included 14 studies and concluded that behavioral techniques had a positive effect (ES, −0.158; 95% CI, −.233 to −0.082; P < .001). The majority of interventions were delivered during active treatment. The format for delivery varied from 2 to 12 sessions and included individual, group, and combined intervention approaches.
Less evidence is available to support other types of interventions for reducing CRF. CBT for insomnia was rated as “likely to be effective” by ONS-PEP86 and was included in the 2011 NCCN guidelines, but more evidence is needed.105 Patients with breast cancer comprise most or all of the patients who experience reduced fatigue after a CBT intervention for insomnia.106-109
Since the most recent meta-analysis, several feasibility and RCT studies have been published that tested types of physical activity/exercise, physically based therapies, CBT, and psychosocial interventions to lower CRF in patients with cancer. Most findings were consistent with prior studies, and the results did not alter the evidence-based guidelines for CRF. It is noteworthy that 2 of those studies tested single versus combined physical activity and/or psychosocial interventions to reduce fatigue in cancer patients with mixed diagnoses. During curative cancer treatment, reduced fatigue occurred after CBT without a lasting increase in physical activity.110 Approximately 1 year after completing treatment, survivors reported significantly reduced fatigue from both physical training (PT) alone and PT + CBT compared with a wait-list control group, but CBT did not add to the benefits from PT.
Appropriate CRF management requires optimal cancer treatment and reduction of potentially reversible contributors, such as mood disorders, hypothyroidism, anemia, infection, and comorbid diseases.87 The effects on CRF from a wide variety of pharmacologic agents have been examined in single-arm studies and, to a lesser extent, in RCTs, but no consensus has been reached regarding which agents are efficacious4 The NCCN guidelines recommend treatment for pain, emotional distress, and anemia as well as optimizing treatment for sleep dysfunction, nutritional deficits/imbalance, and comorbidities.
A meta-analysis of 2 studies in which CRF was treated indicated that methylphenidate was superior to placebo (standardized mean difference in change in CRF score, −0.30; 95% CI, −0.54 to −0.05; P = .02); however, progestational steroids (medroxyprogesterone acetate or megestrol acetate) and paroxetine were no better than placebo.82 A meta-analysis of psychostimulant trials (only 1 included exclusively breast cancer patients) lead to similar findings (standardized mean difference in change in fatigue score, −0.28; 95% CI, −0.48 to −0.09; P = .005) but also reported that several trials did not demonstrate a benefit over placebo. Two Cochrane reviews111, 112 concluded that methylphenidate produced a small but significant improvement in CRF over placebo in patients with mixed tumors who had advanced disease, but those results need to be confirmed in a single, large RCT.
Future interventions need to focus on patients who have moderate-to-severe CRF that has an impact on physical and mood status. Consensus on the definition for the symptom/syndrome of CRF is recommended. Theoretical models to determine contributors that initiate and sustain CRF or that influence the efficacy of interventions for CRF require testing. Contributing factors and symptoms that cluster with CRF need to be included when designing studies. Metrics for both subjective and objective measures, including inflammatory and genetic markers, and longitudinal designs are recommended. Advanced statistical procedures, such as structural equation modeling and growth curve modeling, are recommended when appropriate.
We need to learn when to offer single or combined pharmacologic and nonpharmacologic treatments. Placebo and nocebo effects have been observed in double-blind RCTs that tested agents for CRF in advanced cancer patients and warrant further exploration.113 We need to determine the most appropriate recommendations for exercise based on CRF level. In addition, we must develop and test interventions to increase physical activity in survivors who have clinically significant CRF. Further studies are indicated to determine the effectiveness of evidenced-based interventions in community settings: What works in the short and long term, for whom, and in which settings?
Assessment and management of CRF are essential components of a surveillance model of physical rehabilitation for women with breast cancer (see Stout et al114 in this supplement). During the preoperative evaluation, identification and management of baseline CRF and co-occurring symptoms, such as pain from a biopsy, and an assessment of activity limitations are essential. The preoperative CRF level may reflect the presence and impact on function of premorbid conditions and may serve as a prognostic indicator of subsequent risk for impairment. In the first few months of the postoperative rehabilitation period, reassessment of CRF, with particular attention to its correlation with pain and function, can assist in identifying women who are and are not performing exercise and other health promoting behaviors at the recommended levels and can help in selecting an intervention.
Ongoing surveillance provides an opportunity to identify long-term or persistent, moderate-to-severe CRF. A moderate-to-severe CRF level (≥4 on a scale from 0 to 10), particularly in survivors with lower premorbid physical function, indicates the need for a comprehensive primary fatigue evaluation4 and prompt referral to an appropriate rehabilitation program. When CRF is detected, pain, other co-occurring symptoms, and function need to be assessed and managed. Every ongoing surveillance encounter provides an opportunity to promptly initiate a referral for rehabilitation to an appropriate symptom-management expert. Every encounter also provides time to discuss health-promoting skills and behaviors, including exercise, which lead to better QOL and may help reduce the risk of breast cancer reoccurrence.