Cancer-related fatigue
Evolving concepts in evaluation and treatment
Abstract
BACKGROUND
Although fatigue is one of the most common complaints of patients with cancer, it went unrecognized or overlooked for many years, until clinicians achieved better control over the more acute symptoms of nausea, emesis, and pain. A number of treatment-related and disease-related factors may contribute to the development of fatigue, but its physiologic basis remains poorly understood, and many proposed interventions have not been studied systematically. The lack of a standard of care for the assessment or treatment of fatigue in patients with cancer has limited research in this field. A critical appraisal of these issues is presented in this review.
METHODS
The published literature was reviewed for definition, prevalence, causes, and means of managing cancer-related fatigue (CRF).
RESULTS
Fatigue was reportedly present at the time of diagnosis in approximately 50–75% of cancer patients. The prevalence of CRF increased to 80–96% in patients undergoing chemotherapy and to 60–93% in patients receiving radiotherapy. Two tested interventions that showed consistent effects to alleviate CRF were treatment of cancer-related anemia with erythropoietin agents (recombinant human erythropoietin and darbepotin α) and aerobic exercise.
CONCLUSIONS
Several lines of research are needed to bridge the specific gaps in the current knowledge of CRF. These involve the pathophysiology of the symptom, the validation of diagnostic criteria, and specific therapeutic interventions. Current practice guidelines are based on a combination of research and expert clinical judgment and should be used to guide care with the expectation that they will evolve to incorporate the results of studies currently underway. Cancer 2003. © 2003 American Cancer Society.
Therapeutic improvements in recent years gradually have changed the characteristics of cancer patients, creating an ever-increasing number of individuals with advanced stage disease who need supportive care. One of the consequences has been a greater collective sensitivity to aspects related to the quality of life (QOL) of these patients. The impact of fatigue on QOL is relevant. Carefully constructed surveys have shown that fatigue associated with cancer and chemotherapy not only is the most commonly reported symptom but has a profound effect on the patient's QOL, including physical, psychosocial, and economic/occupational aspects.1, 2 Despite the obvious relevance of the phenomenon, fatigue either was unrecognized or was overlooked until clinicians got better control over the more acute symptoms of nausea, emesis, and pain. Few studies were performed until relatively recently. Therefore, the major problems in dealing with fatigue are the substantial ignorance about the pathogenetic mechanisms and the lack of standards for its assessment or treatment.
Because of its clinical and social impact, we undertook a critical appraisal of the literature evidence in relation to the prevalence, pathophysiology, assessment, impact, and management of cancer-related fatigue (CRF). Our objective was to summarize the current knowledge about CRF, identifying the gray areas that need specific lines of research.
Definition of Fatigue
One of the first challenges in interpreting the reports of the literature is ascertaining whether they really have addressed the same phenomenon, because research on fatigue in patients with cancer has been based on a variety of descriptions of this symptom. According to the definition in the Medical Subject Headings, fatigue is the state of weariness after a period of exertion, mental or physical, characterized by a decreased capacity for work and reduced efficiency to respond to stimuli. However, the temporal relation of becoming tired or weary after an effort, in fact, does not represent the usual experience of cancer patients, who feel tired even at rest. Whereas becoming tired is a common physiologic experience in response to an excessive burden to the organism, cancer patients complain of a reduced capacity to carry out the normal activities of daily living, slow physical recovery from tasks, and diminished concentration. Some authors argue that the term asthenia may be more appropriate to describe the condition of chronic pathologic tiredness in patients with cancer,3 however, the term fatigue has gained a widespread acceptance in the medical literature. An expert panel of the National Comprehensive Cancer Network (NCCN) recently proposed the following definition of CRF: a common, persistent, and subjective sense of tiredness related to cancer or to treatment for cancer that interferes with usual functioning.4 This fatigue differs from the fatigue of everyday life, which usually is temporary and is relieved by rest. CRF is more severe, more distressing, and usually is not relieved by rest.5,6 Finally, it is noteworthy that being fatigued is a subjective experience; therefore, the diagnosis must rely entirely on the patient's description of the symptoms or on what patients say they experience when they talk about fatigue.
Prevalence
An area that shows weaknesses in research methodology is epidemiology, because published studies on CRF are restricted to prevalence data, and there are no reliable incidence studies. Fatigue is a nonspecific symptom that may be found in association with most mental and physical disorders. Indeed, it also can be recorded in the absence of disease in a general medical population, with women affected approximately twice as much as men (23% vs. 13%, respectively).5 However, the prevalence of CRF has been studied systematically only recently. Two national surveys commissioned by The Fatigue Coalition, a multidisciplinary group of medical practitioners, researchers, and patient advocates whose mission is to study the importance of fatigue for patients with cancer and their caregivers, have assessed the prevalence, severity, and QOL consequences of fatigue in patients with cancer. Fatigue-1, which was initiated in 1996, was a three-part investigation aimed at establishing the prevalence, severity, and functional consequences of fatigue in cancer patients. A total of 419 cancer patients who were members of 1700 randomly chosen American families were interviewed by telephone.1 The most commonly represented malignancies were breast cancer (27%), genitourinary tract tumors (16%), gastrointestinal tumors (13%), and leukemias and lymphomas (12%). The treatment received had been chemotherapy in 59% of patients, radiotherapy in 63% of patients, and both chemotherapy and radiotherapy in 24% of patients. Seventy-eight percent of the patients reported that they had suffered from fatigue during the course of their disease and its treatment.
In the Fatigue-2 investigation, 379 patients who had received chemotherapy with or without radiotherapy (40% of them in the previous 2 years) were interviewed.2 These patients, who had received chemotherapy for some form of cancer, were identified from a representative sample of 6125 families in the country, balanced for numerous variables based on data from the United States Bureau of Census. The questionnaire contained approximately 50 questions similar to those in the guide for questionnaires proposed by Cella et al.7 Among this series, 76% of patients had suffered from fatigue at least some days of the month during chemotherapy; whereas, in 30% of patients, fatigue was present every day.
Pooling these with other data,8 it can be estimated that fatigue is present at the time of diagnosis in approximately 50% of cancer patients. It occurs in up to 75% of patients if bone metastases already are present. An estimated 60–96% of cancer patients in treatment experience fatigue, including 60–93% of patients on radiotherapy and 80–96% of patients on chemotherapy. Once again, it should be clear that the Fatigue Coalition investigations were based on recall, retrospective, cross-sectional data and used nonvalidated diagnostic criteria. Accordingly, these surveys have a relatively limited reliability and may be biased by the fact that only a minority of potential patients were contacted.
When the main cause of fatigue is treatment, there is a clear temporal correlation between the two. In fact, fatigue is most intense in the first few days after chemotherapy and then gradually abates until the next course of chemotherapy. Nevertheless, fatigue can last sometimes for prolonged periods after the completion of a course of chemotherapy.9-11 Examples of the latter include two studies in patients with ovarian cancer, a disease that frequently has a prolonged evolution and, thus, is suitable for evaluating the effects of treatments on QOL. Portenoy et al.12 reported that 104 of 151 patients (68.9%) undergoing treatment for ovarian cancer suffered from fatigue, whereas a recent study from Holzner et al.13 indicated that 32 of 98 ovarian cancer survivors (32.7%) who had not been treated for at least 6 months were diagnosed with fatigue. It is noteworthy that, in approximately half of the patients from both studies, fatigue was described as an extremely stressful condition.
To our knowledge, literature data regarding the prevalence or incidence of fatigue in patients with specific malignancies are scarce. The data available differ considerably because, until recently, fatigue rarely was reported in the list of therapy toxicities, there was not a standard definition of fatigue, and appropriate instruments for its evaluation were not available.
Pathogenesis
The mechanisms that cause or promote fatigue in patients with cancer still are understood poorly and have been the subject of several theories.14 One of the major obstacles to basic and preclinical research has been the problematic development of animal models of CRF due to the inherent subjectivity of the symptom and the difficulty of establishing objective, behavioral correlates of fatigue. Voluntary, motivated activity has been proposed as one such correlate in animals.15
In addition to the direct effects of cancer and the various modes of cancer treatment, a wide variety of other phenomena contribute to fatigue in cancer patients (Table 1). Proposed tumor-related factors include abnormalities of energy metabolism, decreased availability of metabolic substrates, abnormal production of substances inhibiting metabolism or normal muscle function, neurophysiologic changes of skeletal muscles, chronic stress response, and hormonal changes.14
| Physiologic |
| Underlying neoplastic disease |
| Abnormalities of energy metabolism |
| Decreased availability of metabolic substrates |
| Abnormal production of substances inhibiting metabolism or normal muscle function |
| Neurophysiologic changes of skeletal muscles |
| Chronic stress response |
| Hormonal changes |
| Antineoplastic treatments |
| Chemotherapy |
| Radiotherapy |
| Surgery |
| Biologic response modifiers |
| Concomitant systemic diseases |
| Anemia |
| Infections |
| Lung diseases |
| Liver failure |
| Renal failure |
| Malnutrition |
| Neuromuscular disorders |
| Dehydration or electrolyte imbalances |
| Thyroid disorders |
| Sleep disorders |
| Immobility and lack of exercise |
| Chronic pain |
| Use of centrally acting drugs (e.g., opioids) |
| Psychosocial |
| Anxiety disorders |
| Depressive disorders |
| Associated with stress |
| Associated with different environmental factors |
Although an increased energy expenditure traditionally has been regarded as a characteristic metabolic abnormality of cancer patients, carefully conducted investigations describe a heterogeneous picture, with resting energy expenditure varying between < 60% than predicted and > 150% than predicted.16, 17 It has been hypothesized that, whereas resting energy expenditure is increased, total energy expenditure may be unchanged due to a fall in physical activity.18 This model, however, fails to elucidate how decreased physical activity leads to the perception of fatigue.
Abnormal substrate utilization has been demonstrated in cancer patients with cachexia, a wasting syndrome that has biochemical mediators in common with fatigue. These patients also exhibit relative glucose intolerance and insulin resistance, with an increased rate of glucose production and recycling through lactate (the Cori cycle).19-21 The energy cost of this increased glucose turnover has been estimated as high as 260 kcal per day.21
Loss of muscle mass is one factor likely to be associated with fatigue in cancer patients.14 It has been found that whole-body protein turnover is increased in the majority of patients with advanced stage cancer compared with starved, normal individuals and weight-losing noncancer patients, and it appears to increase further with progression of disease.22, 23 However, skeletal muscle protein breakdown rates do not differ from controls, whereas there is a reduction in the rate of muscle protein synthesis producing net protein breakdown.24, 25 Acute-phase protein response is seen in a significant proportion of patients with a variety of cancers26-29 and has been related to weight loss28, 30, 31 and increased breakdown of skeletal muscle proteins.32 Interest has focused recently on the adenosine triphosphate (ATP) dependent, ubiquitin-mediated proteolytic pathway as a cause of protein breakdown in cachexia.33 In animal models of cachexia, this system appears to be activated preferentially with no change in calcium dependent or lysosomal proteolytic systems.34, 35 Increases in mRNA for elements of the ATP-ubiquitin pathway have been identified in patients with cancer compared with controls; however, those increases were not correlated with weight loss.36, 37 A newly identified gene, atrogin-1 (short for atrophy-related gene), has been involved in muscle loss associated with cancer as well as other conditions, such as fasting, diabetes, and renal failure. Using gene microarray analysis, muscle samples from healthy rats and rats experiencing various muscle-wasting conditions were studied. It was found that atrogin-1 expression was restricted to skeletal muscles; and, on fasting, atrogin-1 mRNA levels increased before atrophy occurred.38 Functional analysis of the atrogin-1 regions suggests that the gene product may be involved in protein breakdown through the ubiquitin-proteasome pathway.
Fatigue has been ascribed to the abnormal production of inflammatory cytokines in the setting of cancer, but the specific evidence for the role of cytokines in CFR is fragmentary. Administration of cytokines, such as tumor necrosis-α, (TNF-α), interleukin-1-β (IL-1-β), IL-6, and interferon-γ (IFN-γ,), leads to anorexia, weight loss, protein and fat breakdown, rises in levels of cortisol and glucagon and falls in insulin levels, insulin resistance, anemia, fever, and elevated energy expenditure in animals.39-44 In addition, elevated circulating concentrations of IL-6 have been associated with weight loss and the acute-phase protein response in some cancer patients.30, 45 Nevertheless, a major methodological difficulty in the study of cytokines is the ability to determine whether elevated levels cause symptoms or merely are associated with them.
Prostaglandins may mediate the actions of most proinflammatory cytokines and likely are involved in the metabolic abnormalities of cancer patients.46, 47 In fact, specific inhibitors of prostaglandin synthesis prevent the experimental cachectic effects of TNF-α48 and IL-1.49, 50 It also is noteworthy that release of prostaglandins is a major step in the signaling pathway leading to muscle protein breakdown in normal tissues.51
Tumor-derived catabolic factors, such as proteolysis-inducing factor (PIF), have been isolated in cancer patients.52 PIF activates the ATP-ubiquitin dependent proteolytic pathway and induces the nuclear transcription factors nuclear factor κ-B and STAT3, resulting in cytokine and acute-phase protein synthesis.53-55 It has been found that PIF is expressed in tumor cells from patients with significant weight loss, but not in patients with reasonably stable weight.56 The failure of proinflammatory cytokines, such as IL-6, to induce cachexia reliably in animal models has led to the suggestion that tumor-derived factors like as PIF may act as cofactors with host-derived or tumor-derived cytokines to produce a cachectic state.57
Neurophysiologic changes of skeletal muscles in patients presenting fatigue have been reviewed recently.58 The most typical finding is a long-lasting depression of force production after fatiguing muscle activity, especially at low stimulation frequencies. This low-frequency fatigue appears to be because of structural changes in proteins involved in intracellular Ca2+ handling. Contractions in which the muscle is stretched (eccentric contractions) cause muscle weakness and damage. The initial defect induced by eccentric contractions is overstretched sarcomeres, but these appear to cause localized membrane tears that subsequently contribute to muscle weakness and damage.
Advances in neuroscience have disclosed new perspectives. It has long been believed that chronic stress creates profound physiologic changes that may contribute to fatigue. Among the proposed mechanisms, new emphasis has been given to dysfunction of the hypothalamic-pituitary-adrenal (HPA) axis59, 60 and increases in CRF.61 The secretory end product of the HPA axis, cortisol, is kept within an optimal range through the feedback action of cortisol interacting with neural control mechanisms. Distressing events or situations evoke prominent HPA system activation; and, after long-term exposure, the HPA axis eventually will become dishabituated, resulting in a disruption of central regulatory systems and a net decrease of cortisol output, a flattened diurnal secretory pattern, and inhibition of other endocrine axes.62 The effects of cancer and its treatment on the hypothalamic-pituitary-gonadal axis also may contribute to fatigue. Decreased gonadotropin levels are associated with poor nutritional states; conversely, the administration of progesterone can increase appetite, whereas anabolic steroid use can decrease weight loss in cancer patients.63 Recent data indicate that hormonal ablative therapy for prostate cancer doubles the incidence of reported fatigue, providing a direct correlation between gonadotropin function and fatigue.64
Cancer treatments also can cause or exacerbate fatigue by a number of other mechanisms. Postoperative fatigue frequently is observed in patients who undergo surgery for diagnosis or treatment, but little research exists examining causes and correlates of this fatigue. Factors involved include the effects of anaesthesia, analgesia, and sedation, decreased ventilatory capacity, immobilization, infection, preoperative and postoperative starvation, altered sleep patterns, and anxiety.65 Fatigue after surgery, however, clearly may be compounded by fatigue experienced as a result of other treatment modalities. The mechanisms of chemotherapy-related fatigue are associated with anemia or with an accumulation of cell destruction end products. It has been suggested that chemotherapy drugs that cross the blood-brain barrier can induce neurotoxicities that may produce fatigue.65 Such drugs include methotrexate, ifosphamide, cisplatin, vincristine, fludarabine, and paclitaxel. Newer cytotoxic drugs, such as docetaxel, irinotecan, and raltitrexed, also have fatigue as a major side effect.63
Similarly, the pathogenetic mechanisms of fatigue accompanying radiation therapy are not very clear. However, the severity of fatigue does not seem to be related to either the disease type or to the radiation site and usually declines gradually after treatment is completed. Not all patients return to pretreatment energy levels.66-68 The risk factors for persistent fatigue are older age, more advanced disease, and combination modality therapy.69
Large proportions of patients who are treated with biologic response modifiers and cytokines also experience fatigue. In these patients, fatigue usually occurs as part of a constellation of symptoms called flu-like syndrome, including, fever, chills, myalgias, headache, and malaise.70 The prototypical and, by far, most studied biotherapy agent is IFN-α. IFN-α causes fatigue in 70% of patients; and hypothyroidism, which can lead to fatigue, is identified in up to 20% of patients.71, 72 Neuroendocrine abnormalities with this class of therapy include HPA axis dysfunction; hypothalamic-pituitary-gonadal axis inhibition with decreased levels of estrogen, progesterone, and testosterone; and decreases in growth hormone activity.
Other suggested mechanisms correlate fatigue with mood alterations (depressive disorders) and the closely associated sleep disorders. There is great variation in the reported frequencies of depression in cancer populations. These varying estimates may result from differing definitions of depression, differing assessment tools, and heterogeneous cancer populations with different significant variables, such as timing of the assessment, physical debilitation, and concurrent treatment. A 1995 meta-analysis of studies documenting rates of depression in cancer patients, without regard to cancer type, gender, or disease state, demonstrated mean prevalence rates of 24%, with a range of 15–42%.73 The association between fatigue and depression remains controversial. Whereas, in some studies, fatigue was correlated moderately with depression74 or was an independent predictor of depression,75 other reports concluded that fatigue and depression were unrelated conditions with different patterns over time.76 An interesting investigation was reported recently by a Japanese group, who used the Cancer Fatigue Scale and other standardized questionnaires to study fatigue in 134 women who had undergone surgery for breast cancer and were disease free.77 Multiple regression analysis demonstrated that fatigue, which was present in greater than half of patients (56%), was associated with dyspnea, insomnia, and depression. The other variables investigated, including the time elapsed from surgery and the cancer treatments administered (chemotherapy, radiotherapy, or hormonal therapy), were not linked with fatigue. Those results suggest that fatigue may derive from physical and psychologic distress rather than from the tumor itself or from its treatment. Clinical interventions directed at alleviating dyspnea, insomnia, and depression may be an important strategy for reducing the severity of fatigue.
Contemporary research has focused on the neurobiology of depression in patients with cancer. Recent experimental evidence suggests that depression is associated with activation of some aspects of cellular immunity, resulting in the hypersecretion of proinflammatory cytokines and dysfunction of the HPA axis.78 In addition, attention has been given to treatment-related factors, such as some chemotherapeutic agents that are associated with depression or other secondary neuropsychiatric problems.79, 80
In the most recent version of its guidelines,81 the NCCN panel identified seven factors that often play a significant role in the fatigue experience: pain, emotional distress, sleep disturbance, anemia, nutrition, activity level, and other comorbidities. Comorbidities include medical problems or illnesses that are unrelated to cancer (e.g., congestive heart failure, emphysema) that may contribute to fatigue. Appropriate management of these conditions can help significantly to reduce the level of fatigue.
Because hypothyroidism is very common in the general population,82, 83 cancer patients with the symptom of fatigue should be evaluated for this condition. In individuals with malignancies like head and neck cancer or Hodgkin disease, when the patient's neck or mediastinum is treated with radiation therapy, hypothyroidism develops frequently without being suspected.84, 85
Anemia deserves a particularly thorough analysis, given the possibilities of therapy with this condition. In most patients, cancer-related anemia is associated with normochromic and normocytic red cell indices, a low reticulocyte count, decreased serum iron and transferrin concentrations, and raised ferritin. The pathogenesis of anemia usually is complex, because it can result from impaired use of iron stores; shortened red blood cell survival; blunted erythropoietin (EPO) production in response to anemia; and, obviously, suppression of erythropoiesis by the cancer treatment, whether it is chemotherapy or radiotherapy.86
Because of the high prevalence of anemia and the general efficacy and lack of toxicity of EPO, this treatment has been studied extensively. A systematic review of controlled trials on EPO in patients with cancer-related anemia has been conducted recently.87 Although there seemed to be a clinically obvious association between anemia and fatigue, early reports were unable to demonstrate a clear correlation between hemoglobin (Hb) levels and severity of fatigue in cancer patients.88, 89 However, recent studies using more refined instruments of evaluation, such as the Functional Assessment of Cancer Therapy (FACT), have established a direct relation between Hb levels and fatigue.90, 91 Table 2 shows the correlation between levels of Hb > 12 g/dL and < 12 g/dL and various clinical parameters. It is clear from the table that the general scores for QOL, fatigue, and sensations of physical and functional well being are significantly higher among patients with Hb levels >12 g/dL.
| QOL parameter | P value |
|---|---|
| Higher QOL scores | 0.003 |
| Less fatigue | 0.01 |
| Fewer symptoms of anemia (except fatigue) | 0.02 |
| Better sensation of physical well being | 0.003 |
| Better sensation of functional well being | 0.001 |
- QOL: quality of life.
- Adapted from: Sobrero A, Pugliesi F, Guglielmi A, et al. Fatigue: a main component of anemia symptomatology. Semin Hematol. 2001;28:15–18.
Diagnostic Criteria
Investigators felt that, as the interest and body of research about CRF continued to grow, a working set of diagnostic criteria was essential to yield reproducible results in practice and research environments. Modeled after research diagnostic criteria commonly used in psychiatry, these criteria were created by a multidisciplinary group of cancer treatment and supportive care experts to establish a common diagnostic language for CRF.92 CRF has been proposed as an independent, nosological entity in the International Classification of Diseases 10th Revision—Clinical Modification.7, 92, 93 The criteria that have to be met to make the diagnosis of CRF (at least 6 of 11) certainly are more specific than the simple question, “do you feel tired?,” to which greater than 50% of patients with cancer respond affirmatively, and were used in the Fatigue-2 survey.2 A validation study of these criteria has been carried out recently.94 Three hundred seventy-nine individuals who had been treated with chemotherapy, either alone or in combination with radiation therapy, were interviewed. One hundred forty-one individuals (37%) reported at least 2 weeks of fatigue in the previous month. Of the respondents who had received their last treatment more than 5 years previously, 33% still reported at least a 2-week period of fatigue in the month before the interview. The prevalence of diagnosable CRF in the individuals in this sample, most of whom had completed treatment more than 1 year before the survey, was 17%, a much lower figure than expected based on previous reports that used less stringent criteria. However, the authors of that study acknowledged that methodological factors, such as selection criteria, may have biased the results in the sense of a conservative estimate. They also concluded that studies are needed comparing patients undergoing treatment with patients at various times posttreatment to determine whether differences exist using these proposed stricter criteria.
Assessment Instruments
There is currently a large and increasing number of assessment instruments. Broadly speaking, these instruments can be divided into single-dimension and multidimensional scales. The former include the Profile of Mood States,95 which incorporates a subscale for fatigue or energy, and the validated, reliable Linear Analog Scale Assessment (LASA), which is used frequently in clinical trials.96 The multidimensional scales, which capture various characteristics of fatigue and its impact on function, clearly are more informative instruments. However, from a practical point of view, there often is insufficient time to evaluate the patient thoroughly on one visit. Because of their length, the patients themselves often are too tired to complete overly long questionnaires. Thus, a practical approach involves the use of the following three simple questions:93 1) Do you feel or have you ever felt unusually tired? 2) If yes, can you indicate how tired you feel on average on a scale from 0 to 10? 3) How much does this tiredness affect your daily life activities?
The validated multidimensional questionnaires are usually sophisticated instruments that are used in the setting of research. The first validated multidimensional scale was the Piper Fatigue Self-Report Scale (PFS).97 The PFS was developed initially for patients undergoing radiotherapy but was then applied to cancer patients undergoing other types of treatment. Revised in its current form in 1998,98 this scale investigates the severity, the disorders and the impact of fatigue using a questionnaire with 27 items, 22 of which evaluate perception of current fatigue with an 11-point Likert scale. The remaining 5 questions are open ended, affording patients the opportunity to qualify and quantify fatigue. The PFS can be delineated into four subscales: behavioral/severity (six items), affective meaning (five items), sensory (five items), and cognitive/mood (six items).
Other assessment instruments specific for fatigue are the Brief Fatigue Inventory,99 the Cancer Fatigue Scale,100 the Fatigue Assessment Instrument,101 and the Multidimensional Fatigue Inventory.102 Multidimensional scales, such as the general FACT (FACT-G) and the European Organization for Research and Treatment of Cancer (EORTC) Core QOL Assessment Tool, contain a general evaluation of QOL but include specific subscales for fatigue. The two systems both are self-reported (i.e., the patient completes the questions rather than an interviewer) and are similar conceptually.
The FACT-G is the core component of Functional Assessment of Chronic Illness Therapy (FACIT), a multidimensional questionnaire that is used widely to measure QOL in patients with chronic diseases.103 The questionnaire comprises more than 300 items and has been translated throughout the world into 43 languages. FACT is the version specific for cancer patients.90 FACT-G is the general part, with 27 questions on 4 relevant aspects of QOL: physical (somatic), social and family, emotional, and functional. Moreover, it contains questions specific to the disease and the type of treatment and specific subquestionnaires for the symptoms. The subscale most widely used in oncology to evaluate fatigue is the FACT anemia part (FACT-An). FACT-An includes the general part (FACT-G), a specific questionnaire with 7 questions concerning symptoms linked to anemia (that do not include fatigue), and another fatigue part, FACT-F, for the specific assessment of fatigue. FACT-F consists of 13 topics and is shown in Table 3.
| 1. I feel weary |
| 2. I always feel weak |
| 3. I feel exhausted |
| 4. I feel tired |
| 5. I'm too tired to eat |
| 6. I need to sleep during the day |
| 7. I need help to carry out daily activities |
| 8. I feel frustrated because I'm tired and I can't do the things I want to |
| 9. I have to limit my social activities because I feel tired |
| 10. I have difficulty in starting to do something |
| 11. I have difficulty in finishing something I have started |
| 12. I feel I don't have any energy |
| 13. I am able to carry out normal activities |
When the FACIT questionnaire was administered both to patients with cancer and to healthy individuals, not only did cancer patients report fatigue more often than the normal population, but the average intensity of fatigue was greater in patients than in the general U.S. population (P < 0.001).104 Furthermore, when patients were analyzed on the basis of their Hb levels, it was found that fatigue was less severe in patients with higher Hb levels.
The EORTC-Quality of Life Core 30 (EORTC QLQ-C30) questionnaire is currently the most used in the setting of European clinical trials.105 This is a very complex evaluation system that includes five functional scales (physical, emotional, cognitive, social, role), three scales assessing symptoms (fatigue, pain, vomiting), one scale on general well being and QOL, and six simple questions (e.g., loss of appetite).
To assess similarities and differences between the FACT-G and EORTC QLQ-C30 questionnaires, 244 patients with a diagnosis of breast cancer or Hodgkin disease completed both (German language version) during the same session.106 Questionnaire data were analyzed on a subscale basis using correlation analysis, canonical correlation, and multiple linear regression. Correlation analysis for the two sets of subscales revealed that overall agreement between the two instruments was only moderate. Of the five FACT-G subscales, only one, Physical Well Being, was represented well by the EORTC QLQ-C30 subscales, whereas only three of eight EORTC QLQ-C30 subscales (Physical Functioning, Global QOL, General Symptoms) were represented fairly well by FACT-G subscales. Thus, despite considerable overlap, it was found that the EORTC QLQ-C30 and the FACT-G measured markedly different aspects of QOL. This indicates that neither of the two QOL instruments can be replaced by the other and that a direct comparison of results obtained with the two instruments is not appropriate.
Impact on QOL
Several of the studies cited earlier clearly showed that the prevalence of fatigue is high (and the duration surprisingly long) in patients with a variety of types of cancer. But how much and in what ways does fatigue affect the QOL of patients with cancer? The answer to this question was sought by the Fatigue Coalition in the Fatigue-1 and Fatigue-2 studies. In the Fatigue-1 study, approximately two-thirds of the patients considered that fatigue affected their performance of normal daily activities significantly (31%) or to a fair degree (39%).1 Furthermore, 61% of the patients claimed that fatigue influenced their life more than pain. The Fatigue-2 study was performed between July and August, 1998 to confirm the incidence of fatigue in cancer patients and to establish its socioeconomic impact on the lives of patients and healthcare workers.2 The investigation did not involve the healthcare workers directly but included questions to the patients about the impact of fatigue on those who cared for them. When the patients were asked what was the worst side effect they had suffered during their chemotherapy treatment, 34% of patients reported nausea, and 18% chose fatigue. However, 25% of patients reported fatigue as the main symptom after the completion of chemotherapy, and only 13% reported nausea as the main symptom after chemotherapy. Overall, 79% (n = 301 patients) of patients reported that they felt fatigue, which was defined as a sense of debilitating tiredness or loss of energy, at least once per week. Among those patients with marked fatigue, 91% said that this prevented them from conducting a normal life, whereas 88% stated that it changed their daily routine. A significant percentage of patients reported increasing difficulty in carrying out a variable number of daily activities, including walking (69%), household tasks (69%), and lifting objects (60%). Fatigue had a considerable emotional impact, with the patients reporting loss of emotional control (90%), a feeling of isolation and solitude (74%), and dejection (72%). Furthermore, the patients stated that fatigue affected their participation in social activities, such as the maintenance of interpersonal relationships (37%), spending time with friends (35%), and going out for dinner (34%). Problems were reported with carrying out typical cognitive tasks, including diminished concentration (38%), remembering things (35%), and maintaining temporal order (34%). Fatigue had a marked effect on employment and financial status. Of 177 employed patients, 77% had lost at least 1 day of work per month because of fatigue. Overall, 75% of patients had changed their conditions of employment in some way, accepting less responsibility (35%), decreasing the number of hours worked (34%), taking more days of sick leave or holiday (31%), suddenly interrupting work (28%), or being compelled to ask for disability retirement (23%). The patients also reported the effects on those who cared for them. These caregivers took more working breaks (20%), accepted less responsibility (18%), and reduced their working hours (11%). A further source of financial stress was created by the need to employ someone to help with daily tasks, such as house cleaning (22%), gardening (18%), and food preparation (5%).
Guidelines for Assessment and Treatment
The paucity of knowledge about the phenomenon of fatigue among physicians and the patients themselves has lead to a lack of treatment or inappropriate treatment, based most frequently on dietary and vitamin support, in some cases pharmacologic treatment, and in others complete rest. In the Fatigue-2 study, 79% of the patients stated that they had discussed their fatigue with a physician, and only 8% had never mentioned this symptom to any healthcare provider.2 When the patients were asked why they had not discussed their fatigue in greater detail, many replied that they considered that fatigue was caused by the treatment of their cancer (79%), that the symptoms would not last for long (61%), or that they did not believe that their physician would be able to do anything about it (45%). A small percentage of patients reported that they felt uncomfortable about discussing fatigue with their physician (13%) or even that they believed that they would be considered ungrateful (4%). When they were asked whether a physician had ever prescribed or recommended anything to alleviate their fatigue, 40% of patients revealed that no such recommendation or prescription had ever been made; 37% of patients reported that they had been advised to rest or relax, 11% patients were advised to take nutritional or dietary supplements, 9% of patients had been prescribed drugs, 7% of patients had been prescribed vitamins, and 5% of patients had been prescribed exercise.
In a recent study by Stone et al., 52% of the patients interviewed (281 of 538 patients) had never reported symptoms of fatigue to their oncologist.107 Only 75 patients (14%) had received treatment or advice on how to manage fatigue. Thirty-three percent (180 of 358 patients) of the patients with fatigue declared that they had not received adequate treatment. The corresponding figures for pain and nausea/emesis were 9% (46 of 538 patients) and 7% (37 of 538 patients).
These data led the NCCN to draw up guidelines on the diagnostic evaluation and treatment of patients with CRF.4 Based on comparative data from some studies showing that patients with a fatigue intensity ≥ 7 (on a linear scale from 0 to 10) had a dramatic reduction in physical performance and that values between 4 and 6 were associated with a significant reduction in functional capacity, the NCCN considered that a fatigue score ≥ 4 was a practical cut-off score for establishing whether treatment should be commenced or further diagnostic investigations of the fatigue should be performed. The general outline of the algorithm has the following stages: screening, primary evaluation, interventions, and reevaluation.
Screening
The first stage of the algorithm concentrates on the assessment of the presence or absence of fatigue (Fig. 1). If fatigue is present, then a quantitative or semiquantitative estimate of its entity should be made and documented. On a numeric scale from 0 to 10, a score from 0 to 3 is considered mild fatigue, a score from 4 to 6 is considered moderate fatigue, and a score ≥ 7 is considered severe fatigue. If the screening does not identify fatigue or the fatigue is mild, then the patient can be referred for subsequent outpatient reassessments.
National Comprehensive Cancer Network guidelines for the initial evaluation of cancer-related fatigue. aRecommended screen: “Since your last visit, how would you rate your worst fatigue on a scale of 0–10?” bSee Education/counseling and management strategies based on clinical status: Active Treatment (FT-5); Posttreatment Follow-Up (FT-61), End of Life (FT-7). Reproduced with permission from: Mock V, Atkinson A, Barsevick A, et al. NCCN clinical practice guidelines for cancer-related fatigue, version 1.2003. Available from URL: http://www.nccn.org/physician_gls/index.html. To view the most recent version of the guideline, go online to www.nccn.org [access date June 1, 2003].
Primary evaluation
If the screening reveals a fatigue score ≥ 4, then a detailed clinical and instrumental evaluation should be performed. One component of this evaluation, of course, should be the current state of the malignancy, the type and duration of treatment and its capacity to induce fatigue, and the patient's response to the treatment. Another component of the evaluation is documenting the characteristics of the fatigue: when it started, what factors aggravate it, etc. Emotional and psychological components should be identified, and the effect of the fatigue on the performance of normal, daily life activities should be determined. Then, various clinical organic conditions that can cause fatigue should be evaluated. The expert panel of the NCCN, as discussed earlier, considered that there are seven primary clinical conditions associated with fatigue and that these should be evaluated specifically (pain, emotional distress, sleep disturbance, anemia, nutrition, activity level, and other comorbidities). Therefore, a detailed investigation of the various systems and organs is warranted. An evaluation of drugs that can contribute to fatigue also should be made. For example, β-blockers can cause bradycardia and hypotension; whereas antidepressants, antiemetics, and antihistamines can aggravate fatigue by their effects on the central nervous system.
Interventions
A first essential step is that patients are given adequate information on the causes and significance of fatigue. Second, patients should be educated to manage fatigue. Self-management of fatigue means giving priority to certain activities that are considered essential (e.g., personal hygiene) and deferring, postponing, or delegating all those activities that are not essential (e.g., shopping). Patients may find it useful to keep a personal diary to determine which of their activities are associated with the most intense fatigue. Then, activities and related periods of rest can be planned.
Cause specific and pharmacologic interventions.
If a specific cause of CRF is identified (anemia, insomnia, depression, metabolic disorders, etc.), then this should be treated first. One therapeutic intervention supported by clinical evidence is the use of recombinant human EPO (rHuEPO) in cancer patients with anemia and fatigue. In fact, a series of clinical trials demonstrated the capacity of rHuEPO to raise the Hb levels and reduce transfusion requirements in patients with neoplastic disorders.108-110 The early, placebo-controlled registration studies carried out in patients with anemia secondary to neoplasia showed that treatment with rHuEPO three times per week was associated with a significant increase in hematocrit (P ≤ 0.004), a significant decrease in units of red blood cells transfused, and marked improvements in levels of energy and daily activities that contributed significantly to an overall improvement in QOL (P ≤ 0.05).108 Since the completion of those registration trials, three community studies with an open, nonrandomized design have been carried out examining rHuEPO treatment of anemia secondary to neoplasia. Those studies involved over 7000 patients.111-113 In two of the studies,111, 112 rHuEPO was administered 3 times per week (10,000 IU subcutaneously); whereas, in the trial by Gabrilove et al.,113 single weekly doses (40,000 IU subcutaneously) were used. In addition to establishing the impact of rHuEPO on Hb levels and transfusion requirements, those studies included an evaluation of QOL, exploiting linear analog scale assessment (LASA). The study by Demetri et al.111 also used the FACT, whereas Gabrilove et al.113 incorporated the FACT-An scale. In each study, final Hb levels were significantly higher compared with baseline Hb levels (P < 0.001). The mean increase was 1.8 g/dL (from 9.2 g/dL to 11.0 g/dL) in the study by Glaspy et al.,112 2.0 g/dL (from 9.3 g/dL to 11.3 g/dL) in the study by Demetri et al.,111 and 2.0 g/dL (from 9.5 g/dL to 11.5 g/dL) in the study by Gabrilove et al.113 By the third month of treatment, the percentage of patients requiring transfusional support had decreased from 21.9% to 10.7% in the study by Glaspy et al., from 28.5% to 8.2% in the study by Demetri et al., and from 14.2% to 6.5% in the study by Gabrilove et al. Response to treatment was defined as a 2 g/dL increase in the Hb level or an Hb concentration > 12 g/dL in the absence of transfusions within the preceding 30 days. All of those studies documented significant improvements in energy, activity, and overall QOL, as determined by the LASA, associated with the significant increases of Hb levels observed in each study. Similarly, results of the FACT-An assessment demonstrated that QOL scores were significantly higher after treatment than at baseline and that there was a significant correlation between high levels of Hb and high scores for the domains of physical and functional well being. An incremental analysis applied to the data from the studies by Demetri et al. and Glaspy et al. showed a statistically significant, nonlinear correlation (P < 0.01) between Hb levels and QOL scores.114 The increases in Hb levels consequent to treatment were associated with an improvement in QOL scores for the range 8–14 g/dL. The most substantial improvements in QOL scores, for every 1 g/dL increment in the level of Hb, occurred when the Hb concentration increased from 11 g/dL to 12 g/dL (range, 11–13 g/dL). The study by Demetri et al. included a prospective analysis of changes in QOL and Hb levels in relation to tumor response to chemotherapy. Their results showed statistically significant increases in QOL measures, together with significant increases in the levels of Hb, in patients who achieved a complete or partial response or compared with patients w ho had stationary disease. Multivariate linear regression analysis of the data from that study indicated that both the increase in Hb level and the tumor response were independent predictors of improvement in QOL. The only randomized study that has evaluated QOL difference in EPO-treated and placebo-treated groups demonstrated a significant improvement in the EPO group and a significant worsening of QOL in the placebo group.115 A very interesting study by a Canadian group has been published recently.116 In that trial of 183 anemic cancer patients, rHuEPO was given for up to 16 weeks both to patients who were and were not receiving chemotherapy. In brief, rHuEPO treatment resulted in significantly improved QOL, increased Hb levels, and decreased transfusion use in both cohorts of patients.
More recently, darbepoetin α was investigated in cancer patients receiving chemotherapy. Darbepoetin α is a new erythropoiesis-stimulating protein that has five N-linked carbohydrate chains, compared with three in rHuEPO. Due to its increased carbohydrate content, the terminal half-life of darbepoetin α is two-fold to three-fold greater than that of rHuEPO in patients with chronic kidney disease or cancer. This pharmacokinetic property may allow for less frequent administration of darbepoetin α compared with rHuEPO. Early studies with darbepoetin α in cancer patients were aimed at defining the clinically effective doses.117-120 The effectiveness of this growth factor relative to supportive care has been demonstrated in the pivotal Phase III trial performed in Europe.121 In that multicenter, double-blind, placebo-controlled study, 320 patients with small cell or nonsmall cell lung cancer with at least 12 more weeks of cisplatin-containing chemotherapy scheduled were randomized to receive placebo or once-weekly subcutaneous injections of darbepoetin α 2.25 mg/kg for 12 weeks. The study drug was withheld if Hb levels increased to > 15 g/dL for men or > 14 g/dL for women and was reinstated at 50% of the previous dose if Hb levels decreased to ≤ 13 g/dL. Darbepoetin α significantly decreased the proportion of patients who required transfusions during Weeks 5–12 (21% vs. 51%) and during the entire treatment phase (26% vs. 60%) compared with the patients who received placebo. In addition, patients who were administered darbepoetin α received fewer standard units of blood both during Weeks 5–12 (0.67 units vs. 1.92 units for the placebo group) and during the entire treatment phase (1.14 units vs. 2.64 units for the placebo group). Darbepoetin α had a safety profile similar to that for placebo, appeared to be associated with fewer days in hospital (10.3 days vs. 13.0 days for the placebo group), and showed an increase in FACT-F scores, indicating that fatigue was improved. Because QOL outcomes were demonstrated using patient self-reports, the use of a placebo-controlled study design was important in clarifying the role of darbepoetin α in measures of health-related QOL.
Other specific pharmacologic interventions include antidepressants and hypnotics. Patients suffering from depressed mood, apathy, decreased energy, and poor concentration may benefit from the use of psychostimulants (e.g., caffeine, methylphenidate, and dextroamphetamine) given in low doses. There are no controlled studies of these drugs for CRF, but empiric administration may yield favorable results in some patients. The side effects most commonly described with psychostimulants include insomnia, euphoria, and mood lability. High doses and long-term use may produce anorexia, nightmares, insomnia, euphoria, paranoia, and possible cardiac complications.
Nonspecific pharmacologic treatments tend to increase the overall level of energy by acting on the pathogenetic mechanisms of fatigue. Although their use is supported by limited data from controlled trials, megestrol acetate122 and corticosteroids123, 124 are used extensively in clinical practice.
Nonpharmacologic interventions.
The management of fatigue with physical exercise is supported by convincing clinical evidence, but diet, sleep, and rest are all strategies used with a certain degree of success. Physical exercise was proposed as treatment for fatigue based on the concept that the toxic effects of treatment and the reduced level of activity during treatment cause a reduction in physical performance capacity. For this reason, patients must force themselves more and use more energy to perform normal activities. Aerobic exercises increase the individual's functional capacity and reduce fatigue. Obviously, the specific programs of physical exercise must be preceded by a thorough evaluation of concomitant diseases and contraindications to such a program. Furthermore, the programs must be tailored individually, taking into consideration the patient's age and gender, the type of cancer, the treatment received, and the patient's current level of physical capacity. Relief of fatigue by exercise has been studied in patients undergoing a variety of treatments, including high-dose chemotherapy and autologous stem cell transplantation, adjuvant chemotherapy, radiation therapy, and IFN-α (Table 4). Most of these studies have been performed in women with breast cancer; thus, the validity of extending the findings to other cancer sites is dubious. Samples usually are limited numerically, because the idea that physical exercise may be beneficial during cancer treatment is a novelty that has struggled to gain acceptance. Despite these considerations and the different types of treatments and programs of physical exercise, all of those studies showed a significant reduction in fatigue, emotional disorders, and sleep disorders in patients undertaking physical exercise during treatment.125-131 Consistent findings in patients with breast cancer have been reported by Mock et al., who used the Levine Conservation Model as the theoretical basis for their research.130 It is an adaptation model that supports maintenance of life by conserving integrity in four domains: conservation of energy, structural integrity, personal integrity, and social integrity. This multicenter, randomized, controlled clinical trial was performed with the goal of determining the effect of a home-based, walking exercise program on fatigue, emotional distress, and QOL. Four outcomes were measured in the study: level of fatigue, emotional distress, sleep disturbance, and physical functioning. These variables were measured pretreatment, midtreatment, and posttreatment in an experimental group that participated in the walking program and a control group that did not exercise. The exercise program was a progressive, moderate-intensity walking program that allowed women to reach 60–80% of their maximum heart rate. This was comprised of 20–30 minutes of walking in their neighborhood 5–6 times per week and began concurrently with chemotherapy or radiation therapy. The investigators gave women in the exercise group an exercise prescription with specific instructions that included warm-up and cool-down periods. Women also were taught to take their pulse before and after walking. The results of that study clearly indicated that women who exercised at least 90 minutes per week on 3 or more days reported significantly less fatigue (assessed with the Revised Piper Fatigue Scale) and emotional distress as well as higher functional ability and QOL compared with women who were less active during treatment.
| Reference | Type of tumor | No. of patients | Treatment received | Study design | Type of exercise | Results |
|---|---|---|---|---|---|---|
| Mock et al.125 | Breast cancer | 14 | Chemotherapy | Phase II | Home-based walking | ↑ Walking ability in exercises; ↓ psychosocial distress vs. controls; less fatigue in exercisers |
| Mock et al.126 | Breast cancer | 46 | RT | Phase III | Home-based walking | ↑ Walking ability in exercisers; ↓ fatigue and other symptoms vs. controls |
| Dimeo et al.127 | Various | 59 | HSCT | Phase III | Bed cycle ergometer | ↓ Fatigue and psych distress in exercisers |
| Schwartz128 | Breast cancer | 27 | Chemotherapy | Phase II (pretest/posttest) | Home-based walking or patient's choice | ↑ Pretest to posttest walking ability; ↑ QOL and less fatigue in active exercisers vs. noncompliers |
| Schwartz et al.129 | Breast cancer | 72 | Chemotherapy | Phase II (pretest/posttest) | Home-based walking or patient's choice | ↓ Fatigue at all levels; ↑ functional ability |
| Mock et al.130 | Breast cancer | 52 | Chemotherapy and RT | Phase III | Home-based walking | ↓ Fatigue and psych distress in exercisers; ↑ functional ability and QOL in exercisers |
| Schwartz et al.131 | Melanoma | 12 | INF-α | Phase I/II (pretest/posttest) | Patient's choice plus methylphenidate 20 mg/day | ↓ Fatigue and cognitive dysfunction; ↑ functional ability |
- RT: radiation therapy; HCST: hematopoietic stem cell transplantation; psych: psychological; INF-α: interferon-α; QOL: quality of life; ↑: increased; ↓, reduced.
- a Only articles indexed in PubMed® (MEDLINE®) are reported.
Reevaluation
Fatigue can appear at different times over the course of a patient's disease and treatment, thus requiring a periodic reassessment. The same diagnostic-therapeutic algorithm used for primary evaluation can be applied.
Conclusions
CRF is an extremely prevalent symptom. The results of the Fatigue-2 study indicate that the effect of fatigue on the QOL of patients is more marked and lasts longer than the effects of nausea, depression, or pain. In addition to its physical consequences, fatigue is associated with considerable emotional, psychological, and social effects and has appreciable financial consequences for both the patients and their caregivers. Various studies indicate that a substantial proportion of patients do not discuss their fatigue with their own physician or that, often, the whole nature of the patient's concerns are not expressed. In any case, the physican's therapeutic response to fatigue as a disease symptom often is nonspecific. Further research will be necessary to clarify the pathogenesis of fatigue and to obtain a better understanding of its effects on daily life and those of any treatment. To facilitate recognition and swift treatment of fatigue, an immediate priority should be the improvement of diagnostic and treatment algorithms for CRF. Progress in this field has already been made, and the NCCN recently published a practical algorithm for the evaluation and treatment of CRF. Careful clinical history taking and thorough clinical examinations are necessary to establish both physiological factors (e.g., anemia, infections, electrolyte imbalances, sleep disorders, pharmacologic treatment) and psychologic factors (e.g., mood disorders, stress) that contribute to fatigue. The specific causes of CRF can be tackled individually. Antidepressants and analgesics should be prescribed for patients with depression and pain. After common causes of anemia have been excluded, patients with low levels of Hb should receive treatment with erythropoietic agents. Clinical studies have shown that both rHuEPO and darbepoetin-α are effective in correcting anemia, reducing transfusion requirements, and improving QOL in cancer patients. Patients with sleep disorders should receive appropriate instructions for improving sleep quality or carefully prescribed drugs. Metabolic disorders, such as alterations in calcium, magnesium, and phosphorus, should be corrected. In the absence of a physiologic etiology for the fatigue, symptomatic treatment can be effective. Educating patients about fatigue is extremely useful. Moderate exercise can be more effective than continuous rest. Aerobic exercises can increase overall muscle tone, and it may be wiser to advise such exercise than rest. The dulling effect of narcotic drugs can be counteracted by careful use of psychostimulators. Patients who are refractory to such management can be considered for an empiric trial of corticosteroids, although established criteria for the use of these drugs are not currently available. Clearly, more work is necessary to gain a better understanding of the physiopathology and treatment of CRF. Fatigue is currently the least treated symptom in cancer patients but is being transformed into an important and appropriate area of research. Nonetheless, tools for improving QOL in patients with fatigue already are available.
Acknowledgements
The authors thank Dr. James B. Bussel from the Division of Hematology/Oncology, New York Presbyterian Hospital, Weill Medical College of Cornell University, New York for carefully reviewing the article and for helpful suggestions. They also acknowledge Dr. Mercedes Bradaschia for the scientific editing of the article.
