Paroxetine

To evaluate the effectiveness of pharmacologic interventions used for fatigue in patients with cancer

Databases searched were PaPaS, Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, Dissertation Abstracts International (DAI), metaRegister of Controlled Trials (mRCT) (January 2007–October 2009). Journals searched were British Journal of Cancer, Journal of Clinical Oncology, Journal of Pain and Symptom Management, and Journal of Palliative Medicine. The reference lists of all articles were checked for additional studies. Conference abstracts also were searched.

Search keywords were neoplasms, bone marrow transplantation, cancer, carcinoma, tumour, adenocarcinoma, leukemia, lymphoma, malignant, radiotherapy, fatigue, tired, weary, weariness, exhausted, lack or loss or lost energy or vigor, apathy or lassitude or lethargy, or feeling drained, sleepy, or sluggish.

Studies were included in the review if they

• Assessed drug therapy for the management of cancer-related fatigue (CRF) compared to placebo, usual care, or a nonpharmacologic intervention.
• Were randomized, controlled trials (single-blind and open-label were allowed).
• Included adult patients with a clinical diagnosis of cancer.

This review was an update of a previous review. The updated search retrieved 647 additional references. Of those, six additional studies met the inclusion criteria. The final sample of studies included was 31.

The review included 7,104 participants who received a drug intervention for CRF.

Psychostimulants

• Four trials examined methylphenidate, and one used dexamphetamine. These included 426 patients total.
• Evidence existed of a significant effect on fatigue with methylphenidate over placebo, and evidence supported the use of psychostimulants in the treatment of CRF.
• The standardized mean difference was positive, with a small effect and narrow confidence interval (CI) (total mean difference = –0.28; 95% CI [-0.48, -0.09]; Z = 2.83; p = 0.005).
• Fatigue was measured with the Functional Assessment of Cancer Therapy-Fatigue (FACT-F) in all studies.

Erythropoietin and Darbepoetin

• Eleven studies were combined in total and demonstrated a positive effect. The weighted mean difference of studies using the FACT-F outcome measure in erythropoietin gave a score of 4.33, which was a clinically significant difference. The conclusion was limited to patients with anemia who were undergoing chemotherapy. Greater improvement was more likely in those with lower hemoglobin levels.
• In placebo-controlled trials of darbepoetin, the mean difference using the FACT-F score was -1.96, which was less than the minimally clinical significant difference.
• Combined analysis for both agents gave a mean difference score of 3.75, which was clinically significant. 
• Erythropoietin and darbepoetin cannot be recommended because of adverse events associated with these drugs.

Antidepressants/Paroxetine

• Two studies using paroxetine and a trial using sertraline were analyzed. Analysis showed no benefit for the treatment of CRF.

Progestational Steroids

• In studies that could be combined, no evidence existed to support continued use for the treatment of fatigue.
• The clinical significance of results of ibandronate were unclear.
• One study of etanercept during chemotherapy had statistically significant results, but the study had a small sample size and poor design. It was suggested that additional trials be conducted.
• One study of donepezil showed no benefit over placebo.

Four trials of methylphenidate provided evidence for use that was supportive but associated with a small effect size in a dose of 10–20 mg per day. Serious adverse events were minimal; however, clinicians need to review contraindications before prescribing. Additional large-scale trials were suggested using methylphenidate to further evaluate use in CRF. Erythropoietin and darbepoetin can no longer be recommended for CRF because of increased adverse events associated with these drugs. No current evidence exists to support the use of steroids.

• Reviewers found major limitations in the reporting of trials and multiple methods of measuring outcomes.
• Some outcomes in trials were not reported due to extensive missing data.
• These findings point to the need for improved research reporting to meet Consolidated Standards of Reporting Trials (CONSORT) guidelines and the benefit that could be derived from use of consistent methods of measuring outcomes.

To determine the efficacy of pharmacological treatment on nonspecific fatigue in palliative care, including patients with advanced cancer and other chronic conditions associated with fatigue.

Databases searched were EMBASE, PsychLit, CENTRAL, and MEDLINE. Reference lists of identified articles were reviewed for inclusion, and textbooks were handsearched. Conference proceedings of the American Society of Clinical Oncology (ASCO) from 2000 to 2008 and the 2005 meeting of the European Cancer Conference were included in the search.

An extensive listing of keywords and specific search methods per database are provided in the article.

Studies were included in the review if

• They were randomized, controlled trials
• The primary outcome was fatigue or related terms, such as asthenia
• Participants were 18 years or older
• The study included evaluation of the effect of pharmacologic treatment of fatigue with psychostimulants, amantadine, corticosteroids, donazepine, and antidepressants if used for the treatment of fatigue. 

Studies were excluded if they studied megestrol or focused on physiologic deficiencies, such as lack of hemoglobin and use of erythropoietin.

Initial searching provided 2,000 titles. Of those, 22 met the inclusion criteria. They included data from 11 drugs:  amantadine (6), pemoline (3), methylphenidate (3), dexamphetamine (2), paroxetine (2), acetyl-L-carnitine (2), testosterone (2), fluoxetine (1), donepezil (1), modafinil (1), and acetylsalicylic acid (1). If two or more studies of the same medication could be analyzed in the same subpopulation of patients, meta-analysis was performed. Meta-analysis was performed for amantadine, pemoline, methylphenidate, and modafinil.

• The final sample of 22 studies included 1,632 patients.
• Studies were performed in patients with multiple sclerosis (MS) (10), HIV (4), cancer (6), postpolio (1), and endstage chronic obstructive pulmonary disease (COPD) (1).

Most studies showed some beneficial effect; however, a substantial similar placebo effect was often observed.

Amantadine

• Meta-analysis was conducted for three (n = 154) studies comparing amantadine to placebo in patients with MS.
• Standard mean difference (SMD) favored amantadine (SMD = 1.68; 95% confidence interval [CI] [1.24,1.92]; Z = 12.76; p < 0.00001).
• No statistically significant heterogeneity existed.
• Study samples were generally small, and several methodologic weaknesses were seen.

Pemoline

• Pemoline was used in three studies on MS.
• Meta-analysis demonstrated no benefit (SMD = –0.11; 95% CI [–0.42, 0.2]; Z = 0.71; p = 0.48).
• There was significant heterogeneity among the studies.

Methylphenidate

• Two studies in patients with cancer were included.
• There was a slightly superior effect compared to placebo (SMD = 0.49; 95% CI [0.15, 0.83]; Z = 2.86; p = 0.004).
• There was significant heterogeneity.

Dextroamphetamine

• Two studies compared the drug to placebo in patients with cancer. No significant benefits were seen.

Paroxetine

• There were no significant effects demonstrated in one study in patients with COPD and one in patients with cancer.

Testosterone

• No significant effects were demonstrated in studies on HIV.

Acetyl-L-carnitine

• No significant effects were shown in one study on cancer and one on MS.

Modafinil

• Meta-analysis in two studies on MS showed no significant effect.

Donepezil

• One study in 142 patients with cancer showed no benefit compared to placebo.

Other

• Fluoxetine was inferior to testosterone in one study on HIV.
• Acetylsalicylic acid was associated with relief of fatigue compared to placebo in one study on MS.

Methylphenidate and amantadine showed promise for reducing fatigue in patients with advanced disease. Amantadine has not been studied in patients with cancer-related fatigue, but it has been shown to be effective in patients with MS. The meta-analysis included only a few studies and the evidence was weak, pointing to the need for additional research in this area. It is not clear whether amantadine would be useful for patients with cancer, as this has not been studied.

The analysis was performed only in palliative care populations and did not include studies of methylphenidate in patients with cancer during active treatment, which also have shown some efficacy. However, side effects included insomnia, anorexia, behavior change, and vertigo in studies reviewed with methylphenidate. In addition, although statistically significant, effect sizes were small. These findings suggest that use in patients with cancer, who also may experience anorexia and sleep disorders from other causes, has potential benefits that would need to be balanced with potential adverse effects. Carnitine, acetylsalicylic acid, and modafinil have been used in a few studies with positive results. These drugs warrant additional investigation to confirm efficacy in different patient populations with fatigue.

Patients were given oral paroxetine/placebo 10 mg for one week pre-interferon treatment, paroxetine/placebo 20 mg during the first week of interferon treatment, and then paroxetine/placebo 20 to 40 mg for all subsequent weeks of interferon therapy. Total length of treatment with paroxetine was 12 weeks. The study was based on the hypothesis that, as a selective serotonin reuptake inhibitor (SSRI), paroxetine may improve the neuropsychiatric and neurovegetative symptoms associated with interferon-alpha treatment.

• The study included 38 patients (paroxetine group, n = 18; placebo group, n = 20); there were equal numbers of men and women.
• Mean age was 50 years (25–74).
• Most patients were married, and approximately 50% were college educated.
• Approximately 25% of the sample had a history of major depression.
• All patients were receiving a course of interferon-alpha consisting of 20 million units/m² five days per week for four weeks, followed by eight weeks of interferon-alpha at a dose of 10 million units/m² three days per week.

• Single site
• Major academic cancer center

Patients were undergoing the active treatment phase of care.

The study was a randomized, double-blind, placebo-controlled trial.

Neurotoxicity Rating Scale (NRS) was used to measure various depressive symptoms, cognitive disturbances, and vegetative symptoms, including fatigue.

When compared with the control group, pretreatment with paroxetine was effective in preventing interferon-induced mood and cognitive symptoms, as well improving pain. Paroxetine had less effect on the development of interferon-alpha–related neurovegetative symptoms, including fatigue, as measured by the NRS. Fatigue and somatic symptoms increased in both depressed and nondepressed patients.

Across the twelve weeks of the study, seven patients from the placebo group and one patient from the paroxetine group withdrew due to severe depression or neurotoxicity.

• The study had a small sample size.
• The study used an instrument with a single item to measure fatigue.
• The study used a wide, uncontrolled paroxetine dose range.
• The study excluded patients with a diagnosis of schizophrenia or bipolar disorder; a Mini-Mental State Exam (MMSE) of 24 or less; and uncontrolled neurological, cardiovascular, endocrine, hematologic, hepatic, or renal disease.

No special training is required. There are costs related to drug acquisition.

Patients were given oral paroxetine 20 mg orally daily or placebo for eight weeks.

• In total, 479 patients (paroxetine, n = 244; placebo, n = 235) were included.
• Mean age was 56.5 years (standard deviation = 12.6 years) (range 27–87).
• Most of the patients were female, 90% were Caucasian, 57% had breast cancer, and 14% had lung cancer.
• All patients were receiving cyclic chemotherapy (but not concurrent with radiotherapy or interferon).

18 Community Clinical Oncology Program (CCOP) outpatient centers

Patients were undergoing the active treatment phase of care.

The study was a randomized, double-blind, placebo-controlled trial.

• Fatigue Symptom Checklist (FSCL)
• Multidimensional Assessment of Fatigue (MAF) (Question 1 only)
• Monopolar Profile of Mood States (POMS) Short Form Fatigue/Inertia subscale

The paroxetine group and the placebo control had comparable levels of fatigue and depression at study inception. Paroxetine had neither beneficial nor detrimental effects on fatigue. There was a significantly lower mean level of depression in the paroxetine group compared with the placebo group. Treatment with paroxetine also favorably affected patients’ general moods. There were no differences in the effect of paroxetine on fatigue by gender, age, indication for treatment (adjuvant treatment versus treatment for metastatic disease), or by whether patients were more or less fatigued at baseline.

• Whether the paroxetine group and the placebo control group had comparable hemoglobin levels at study inception is unknown.
• Less than .01% (n = 2) of the patients in the paroxetine arm experienced adverse events (skin rash and pulmonary embolus) that were possibly related to the study medication.

No special training is required to deliver the intervention. There are costs related to drug acquisition.

Patients were given oral paroxetine 20 mg daily beginning seven days after the initiation of the first cycle of chemotherapy for newly diagnosed breast cancer and continued until seven days following their fourth cycle of chemotherapy or placebo. Randomization was stratified by type of chemotherapy regimen to achieve a balanced design.

• The sample was comprised of 94 women with breast cancer initiating at least four cycles of chemotherapy for breast cancer.
• Mean age was 51.2 years (range 31–79). 
• Of the patients, 89% were Caucasian, 28% were significantly depressed at baseline, and 5% had a hemoglobin less than 11 g/dL.
• There were no significant differences between the intervention and placebo control groups in age, mean baseline measures of fatigue, depression, or general health status.
• The difference in baseline hemoglobin between the intervention group (12.3 g/dL) and the placebo control group (12.9 g/dL) was statistically significant (p < 0.05).

Patients were recruited from a university medical center and two of its affiliated hospitals.

Patients were undergoing the active treatment phase of care.

The study was a multicenter, randomized, double-blind, placebo-controlled trial.

• Fatigue Symptom Checklist (FSCL)
• Multidimensional Assessment of Fatigue (MAF) Fatigue Inteference subscale, Question 1
• Profile of Mood States (POMS) Short Form Fatigue/Inertia subscale 

Controlling for baseline fatigue scores, there were no statistically significant differences in fatigue or fatigue interference between the treatment and control groups. There was no relationship between anemia and fatigue at baseline, although changes in anemia over the course of the study were modestly but significantly correlated with one measure of fatigue.

• No power analysis was provided; it was unclear if the study was appropriately powered to detect a difference in fatigue.
• Anemia was not included as a covariate in the analyses.
• Six patients withdrew due to side effects, such as nausea and headache.

There were costs related to drug acquisition.