Content uploaded by Jennifer Finnegan-John
Author content
All content in this area was uploaded by Jennifer Finnegan-John on Jan 04, 2016
Content may be subject to copyright.
Acupuncture for Cancer-Related Fatigue in Patients With
Breast Cancer: A Pragmatic Randomized Controlled Trial
Alexander Molassiotis, Joy Bardy, Jennifer Finnegan-John, Peter Mackereth, David W. Ryder,
Jacqueline Filshie, Emma Ream, and Alison Richardson
See accompanying editorial on page 4449
Alex Molassiotis and Joy Bardy, School
of Nursing, Midwifery, and Social Work,
University of Manchester; Peter Mack-
ereth and David W. Ryder, Christie
National Health Service (NHS) Founda-
tion Trust, Manchester; Jennifer
Finnegan-John and Emma Ream, Flor-
ence Nightingale School of Nursing and
Midwifery, King’s College London;
Jacqueline Filshie, The Royal Marsden
Hospital NHS Foundation Trust,
London; and Alison Richardson, Univer-
sity of Southampton, Southampton,
United Kingdom.
Submitted January 4, 2012; accepted
July 26, 2012; published online ahead
of print at www.jco.org on October 29,
2012.
Supported by a grant from Break-
through Breast Cancer.
Authors’ disclosures of potential con-
flicts of interest and author contribu-
tions are found at the end of this
article.
Clinical trial information: NCT00957112.
Corresponding author: Alexander
Molassiotis, RN, PhD, Professor of
Cancer and Supportive Care, University
of Manchester, School of Nursing,
University Place, Manchester M13
9PL, United Kingdom; e-mail: alex
.molassiotis@manchester.ac.uk.
© 2012 by American Society of Clinical
Oncology
0732-183X/12/3036-4470/$20.00
DOI: 10.1200/JCO.2012.41.6222
ABSTRACT
Purpose
We aimed to assess the effectiveness of acupuncture for cancer-related fatigue (CRF) in patients
with breast cancer.
Patients and Methods
We conducted a pragmatic, randomized controlled trial comparing acupuncture with enhanced
usual care. Three hundred two outpatients with breast cancer participated. We randomly
assigned 75 patients to usual care and 227 patients to acupuncture plus usual care (random
assignment of 1:3 respectively) with minimization controlling for baseline general fatigue and
maintenance treatment. Treatment was delivered by acupuncturists once a week for 6 weeks
through needling three pairs of acupoints. The usual care group received a booklet with
information about fatigue and its management. Primary outcome was general fatigue at 6
weeks, measured with the Multidimensional Fatigue Inventory (MFI). Other measurements
included the Hospital Anxiety and Depression Scale, Functional Assessment of Cancer
Therapy–General quality-of-life scale, and expectation of acupuncture effect. Analyses were by
intention to treat.
Results
Two hundred forty-six of 302 patients randomly assigned provided complete data at 6 weeks.
The difference in the mean General Fatigue score, between those who received the
intervention and those who did not, was ⫺3.11 (95% CI, ⫺3.97 to ⫺2.25; P⬍.001). The
intervention also improved all other fatigue aspects measured by MFI, including Physical
Fatigue and Mental Fatigue (acupuncture effect, ⫺2.36 and ⫺1.94, respectively; both at P⬍
.001), anxiety and depression (acupuncture effect, ⫺1.83 and ⫺2.13, respectively; both at P⬍
.001), and quality of life (Physical Well-Being effect, 3.30; Functional Well-Being effect, 3.57;
both at P⬍.001; Emotional Well-Being effect, 1.93; P⫽.001; and Social Functioning
Well-Being effect, 1.05; P⬍.05).
Conclusion
Acupuncture is an effective intervention for managing the symptom of CRF and improving
patients’ quality of life.
J Clin Oncol 30:4470-4476. © 2012 by American Society of Clinical Oncology
INTRODUCTION
Persistent cancer-related fatigue (CRF) is a signifi-
cant problem in as many as 40% of disease-free
patients with breast cancer, who experience moder-
ate to severe levels of fatigue even several years after
treatment,
1,2
impacting quality of life. Evidence to
underpin its management is scarce. Patients with
cancer often display an interest in complementary
therapies,
3,4
and a systematic review showed that
acupuncture is promising in the management of
fatigue,
5
with further supporting evidence from can-
cer
6
and noncancer studies.
7,8
These preliminary
studies had small sample sizes and methodologic
limitations. Our feasibility trial assessed the effects of
acupuncture in a mixed sample of fatigued patients
with cancer (n ⫽47) randomly assigned to receive
six sessions of acupuncture, daily self-acupressure,
or daily sham self-acupressure over 2 weeks.
9
Gen-
eral fatigue improved by 36% at the end of the
2-week intervention in the acupuncture group
(compared with 19% in the acupressure group and
0.6% in the sham acupressure group); however, the
effect decreased a month later (22%, 15%, and 7%
improvement, respectively), suggesting that patients
may have been undertreated. The aim of this study
JOURNAL OF CLINICAL ONCOLOGY ORIGINAL REPORT
VOLUME 30 䡠NUMBER 36 䡠DECEMBER 20 2012
4470 © 2012 by American Society of Clinical Oncology
Downloaded from jco.ascopubs.org on December 9, 2015. For personal use only. No other uses without permission.
Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
was to assess the effectiveness of a course of acupuncture for managing
CRF in a sample of patients with breast cancer who had completed
adjuvant chemotherapy.
PATIENTS AND METHODS
Design
A two-group, randomized controlled trial design was used, with partic-
ipants receiving either enhanced usual care or acupuncture. A computer pro-
gram allocated patients to groups and used minimization with a random
element over the margins of the following two factors: baseline Multidimen-
sional Fatigue Inventory (MFI) General Fatigue score (ⱕ16, 17 or 18, or 19 or
20) and maintenance therapy (none, biologic [trastuzumab], and/or hor-
monal [tamoxifen, goserelin]). The allocation ratio was 3:1 to the acupuncture
and control groups, respectively.
Sample and Settings
Patients with breast cancer experiencing persistent fatigue of at least a
moderate level were the focus of the study. Patients were screened for fatigue
through a single-item 10-point scale (where 0 is not fatigued at all and 10 is
extremely fatigued) to identify patients with significant fatigue (ie, those with
score ⱖ5), who were then assessed for eligibility to participate in the study. We
have validated this single item compared with a fatigue scale, and correlations
were r⫽0.75.
9
Recruitment took place at two specialist cancer hospitals in the
United Kingdom, four cancer centers, and three treatment centers of a na-
tional voluntary breast cancer organization.
Eligible patients had a diagnosis of stage I, II, or IIIA breast cancer; had
completed chemotherapy at least 1 month and up to 5 years previously (to
recruit those with persistent/long-term fatigue); had not planned to receive
chemoradiotherapy during the study; had a score ⱖ5ona0to10screening
scale; and were willing to participate and be randomly assigned to one of the
study groups. Patients with previous local recurrence were eligible, but not
patients with distant metastasis.
Exclusion criteria included the following: needle phobia; low platelet
count (⬍50,000/
L); comorbidity with a bleeding disorder or thyroid dys-
function; pregnancy; hemoglobin less than 10 g/dL; hematocrit less than 30%;
active treatment for anemia with erythropoietin or blood transfusions; corti-
costeroid use; and a life expectancy less than 6 months. In addition, the
ipsilateral arm of patients who had undergone axillary dissection was excluded
from needling, as were lymphedematous limbs.
Intervention
All trial patients received usual care. Because there is no recognized
standard care for fatigue, all patients were offered a fatigue information book-
let. In addition, the intervention group was offered six acupuncture sessions
over 6 weeks. On the basis of the Standards for Reporting Interventions in
Clinical Trials of Acupuncture recommendations for reporting acupuncture
trials,
10
patients received a standardized 20-minute acupuncture session nee-
dling bilaterally or unilaterally three points (ST36, SP6, and LI4), with some
flexibility in case points could not be punctured (eg, in case of lymphedema).
Alternative points were selected by therapists at their discretion to maintain an
equal dose of treatment. These points could include GB34 and SP9.
6,9,11
This
approach mimics current acupuncture practice and ensured replicability of
the technique across therapists, which was assessed through regular audits of
the therapists’ records. Patients were observed for 18 weeks. Points were
punctured perpendicularly, with a depth of 0.5 to 1 inch,
11
depending on
patients’ size, sensitivity, and health state. Needles were manufactured by
Seirin (Kyoto, Japan) with guide tubes for single use (size, 36 gauge; point, 16
to 30 mm). No flicking or rotation of the needle took place once inserted. Each
session was based on a strict protocol, and conversation between acupunctur-
ists and patients was kept to minimum. Immediately after each session, ther-
apists completed a form verifying the exact treatment given and any issues that
needed to be reported (eg, adverse effects). These forms were checked regularly
by the investigators for consistency across therapists. No other complementary
therapy use was recommended during the course of acupuncture (although
any such use was documented). Therapists (n ⫽12) were educated in acu-
puncture to degree level, were registered with a professional body, and had a
minimum of 2 years of clinical experience.
The control group received enhanced usual care. In the absence of any
guidelines about how to manage fatigue and the limited available research
evidence, usual care inevitably varied and depended on the clinicians treating
patients with breast cancer at the study sites. To control this current practice,
we chose to enhance usual care by providing all patients with a detailed
information booklet about coping with fatigue developed by Macmillan/Can-
cerBackup (London, United Kingdom), which included details about fatigue
and cancer; causes and effects of fatigue; coping with fatigue at work; diet,
exercise, and sleep; planning the day; and management of fatigue.
The ratio of patients in the experimental and control groups was 3:1,
respectively, to allow us to randomly assign acupuncture patients again at the
end of treatment to one of three groups to test the effectiveness of self-
acupuncture/self-needling as a maintenance therapy (not reported here).
Procedures
The study received ethical approval from a research ethics committee
and all hospitals and centers involved. Patients were referred by clinicians
or health professionals; were self-referred after responding to media pub-
licity, Web site information, or posters displayed within the hospitals and
satellite centers; were referred directly by researchers during outpatient
visits; or responded to mailshot. Patients willing to participate signed a
consent form and were randomly assigned after completion of baseline
data. Researchers telephoned a central service (hospital clinical trials unit),
and operators there used a bespoke computer system to obtain the alloca-
tions. Stratification took place based on main center. Patients in the acu-
puncture group visited the treatment setting to receive the intervention
weekly. Trial participants’ travel expenses were reimbursed. Assessments
were completed at baseline and at the end of 6, 10, and 18 weeks after
baseline (data for 10- and 18-week follow-ups are reported elsewhere). The
trial was monitored by a trial steering/data monitoring committee that
included independent members and a patient representative.
Outcome Measures
The primary outcome measure was the difference in general fatigue, as
self-reported by patients with the MFI
12
at 6 weeks (treatment completion).
The MFI is a brief 20-item well-validated scale measuring general fatigue and
the dimensions of physical and mental fatigue, activity, and motivation. Sec-
ondary outcomes included mental fatigue, activity, and motivation (using the
MFI); anxiety and depression using the 14-item Hospital Anxiety and Depres-
sion Scale (HADS)
13
; and quality of life using the Functional Assessment of
Cancer Therapy–General and the Functional Assessment of Cancer Therapy–
Breast Cancer (FACT-B) module.
14
Sociodemographic and treatment characteristics were obtained at base-
line from patients’ records and patients themselves. Use of complementary
therapies in the past and during the study participation was recorded with a
self-report questionnaire at 6 weeks. Patients were asked about their treatment
expectations, how much they believed this method would help them alleviate
fatigue, and how much faith they have in acupuncture, using three 10-point
scales. Patients completed the questionnaires at home, and these were mailed
back to the researchers using prepaid envelopes. We monitored adverse events
by patient reports and review of therapists’ records.
Sample Size
Our earlier study
9
was used to inform the sample size calculation. The
planned sample size was 320 patients, randomly assigned at a ratio of 3:1 with
240 patients in the experimental group and 80 patients in the control group,
unbalanced to allow for 80% power for the second part of the trial (another
random assignment of acupuncture patients to maintenance therapy with
self-needling, not reported here, for which we needed 192 patients). A 20%
attrition rate was anticipated. With this sample size, there would be in excess of
95% power to detect a half standard deviation (SD) difference in fatigue
change scores. The power was for a two-tailed test of equal mean change scores
at the 5% level of significance. The SD was anticipated to be approximately 4,
and thus, the study was planned to have high power to detect a 2-unit differ-
ence in change scores. Recruitment stopped at 302 patients because the target
Acupuncture for Fatigue in Patients With Breast Cancer
www.jco.org © 2012 by American Society of Clinical Oncology 4471
Downloaded from jco.ascopubs.org on December 9, 2015. For personal use only. No other uses without permission.
Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
numbers for the second phase of the trial (n ⫽192) had been achieved, with
sample power (⬎90%) for the first random assignment.
Data Analysis
Descriptive statistics were used to summarize the data. A simple ttest was
applied to fatigue change scores (week 6 ⫺baseline), but the primary analysis
was an analysis of covariance (ANCOVA) of week 6 fatigue scores with the
baseline fatigue score as a covariate and trial arm as grouping factor. Similar
analyses were performed for secondary outcomes (ie, subscales from the MFI,
HADS, and FACT-B). A last value carried forward approach was used as a
sensitivity analysis for patients missing the primary outcome. Sample sizes
were adequate to justify ttests (the sampling distribution of the mean tends to
normality irrespective of the distribution of the observations by the central
limit theorem). Intent-to-treat analysis was performed. Normality assump-
tions in the data were assessed using residual plots against fitted values and a
quantile-quantile plot of the residuals.
RESULTS
Sociodemographic and Clinical Characteristics
Three hundred two patients were enrolled and randomly as-
signed, 227 to acupuncture and 75 to usual care (Fig 1). The sample
was predominantly white, married, and with college or higher educa-
tion. The mean age of the enhanced usual care group was 53 years
(range, 25 to 80 years), and the mean age of the acupuncture group
was 52 years (range, 30 to 75 years). Time since diagnosis was 20.5
months for both groups, with a range of 6 to 84 months in the usual
care group and 5 to 161 months in the acupuncture group. The mean
duration of fatigue was 18 months (range, 4 to 58 months) in the usual
care arm and 15 months (range, 2 to 69 months) in the acupuncture
arm. Other details are listed in Table 1.
Primary Outcome Analysis
Fifty-six (18.5%) of 302 patients were missing the primary out-
come, 46 (20.3%) of 227 patients in the acupuncture arm and 10
(13.3%) of 75 patients in the control arm. The trial arm effect was
highly significant (P⬍.001), with an estimated difference in week 6
score (acupuncture ⫺control) given equal baseline scores of ⫺3.11
(95% CI, ⫺3.97 to ⫺2.25; Fig 2).
An assumption of no improvement for patients missing the
6-week general fatigue score (ie, baseline value carried forward) was
postulated, providing a conservative sensitivity analysis for the pri-
mary outcome. On fitting such a model, the magnitude of the acu-
puncture effect was reduced to ⫺2.49 (95% CI, ⫺3.29 to ⫺1.69), but
this remained highly significant (P⬍.001; Table 2). ANCOVA for the
primary outcome is presented in Table 3.
Assessment of a center effect was performed with the addition of
two random effects (an intercept and trial arm contrast) within each of
the 11 centers. This extended mixed-effects model was fitted and
found not to significantly improve the fit (P⫽.39). Under the ex-
tended model, the mean treatment contrast was ⫺2.99, with an esti-
mated SD of the random effect of 0.30.
Primary analysis yielded a treatment contrast of ⫺3.11 fitted to
246 patients. Adding additional covariates of baseline FACT-B,
HADS-Anxiety, and HADS-Depression in an ANCOVA model
yielded a treatment contrast of ⫺2.93 fitted to 221 patients (missing
data in 25 patients), which was similar to the estimate without the
covariates and highly significant (P⬍.001).
ANCOVA regression models were fitted for MFI, HADS, and
FACT-B subscales. In each case, the week 6 value was the response and
the baseline value was used as covariate. The key results are listed in
Table 3, showing that the acupuncture intervention had beneficial
effects on all secondary outcomes assessed.
Subgroup analyses were carried out with regard to the screen-
ing fatigue score (5 to 6 [moderate fatigue] v7 to 10 [severe
fatigue]), age (median cutoff point of 52 years), and whether
patients were receiving or not receiving maintenance hormone
therapies. Significance was assessed by adding appropriate interac-
tion terms to ANCOVA. None of the results were significant (test
of heterogeneity, 0.23, 0.23, and 0.34 for fatigue score, age, and
receipt of hormone therapy, respectively).
Participants were asked before random assignment how much
they expected acupuncture to help them alleviate fatigue. Expectation
of effect did not relate to outcome, as the Spearman rank correlations
were r
s
⫽⫺0.18 (n ⫽62, P⫽.15) in the control group and r
s
⫽⫺0.03
(n ⫽175, P⫽.74) in the acupuncture group.
Random Allocation 3:1
(N = 302)
Control arm (enhanced standard care; n = 75)
Received allocated intervention (n = 75)
)722 = n( mra erutcnupucA
Received allocated intervention (n = 218)
Did not receive allocated intervention (n = 9)
Withdrew before treatment (n = 6)
Withdrawn by researcher (n = 3)
Received at least one session but < six (n = 53)
)25 = n( eciohc tneitaP
Withdrawn by researcher (n = 1)
)56 = n( dezylanA
Excluded from analysis (n = 10)
(did not return the primary
outcome data at 6 weeks)
)181 = n( dezylanA
Excluded from analysis (n = 46)
(did not return the primary
outcome data at 6 weeks)
Fig 1. Flow of participants through each
stage of the trial.
Molassiotis et al
4472 © 2012 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on December 9, 2015. For personal use only. No other uses without permission.
Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
DISCUSSION
This study determined that women with breast cancer and CRF re-
ported significant improvements in overall fatigue, physical and men-
tal fatigue, activity, motivation, psychological distress, and all domains
of quality of life after 6 weeks of acupuncture. To our knowledge, this
is the first large multisite trial of its kind and provides some evidence of
the effects of acupuncture.
This trial confirms preliminary and promising evidence obtained
from an uncontrolled single-arm trial
6
and our own pilot study.
9
Although the magnitude of the improvement in the current trial in
absolute numbers was lower than previous preliminary studies, it is
nevertheless both statistically and clinically important. The approach
used in this study not only improved physical fatigue and physical and
functional quality of life, but also improved mental fatigue, which has
Table 1. Sociodemographic and Clinical Characteristics of the
Patient Sample
Characteristic
Usual Care
Group
(n ⫽75)
Acupuncture
Group
(n ⫽227)
No. of
Patients %
No. of
Patients %
Marital status
Single 17 23 47 21
Married 46 61 124 55
Divorced or separated 8 11 49 22
Widowed 4 5 7 3
Living arrangements
Living alone 15 20 51 22
Living with husband or partner 51 68 140 63
Other 9 12 36 16
Educational attainment
Secondary school 13 17 64 28
College/diploma 25 33 81 36
University/degree 23 31 45 20
Postgraduate 14 19 36 16
Missing 0 0 1 ⬍1
Race or ethnicity
White 71 95 212 93
Black 1 1 3 1
Asian/Chinese 0 0 9 4
Mixed 3 4 2 1
Missing 0 0 1 ⬍1
Religious affiliation
Christian 57 76 173 76
Muslim 0 0 5 2
Hindu 0 0 2 1
None 16 21 36 16
Prefer not to say 0 0 3 1
Other 2 3 8 4
Occupational status
Employed full-time 19 25 59 26
Employed part-time 21 28 74 33
Retired 19 25 30 13
Unemployed 0 0 5 2
Casual worker 1 1 0 0
Not working because of ill health 10 13 39 17
Housewife 3 4 12 5
Other 0 0 8 4
Missing 2 3 0 0
Cancer treatment
Surgery and chemotherapy 12 16 45 20
Chemotherapy and radiotherapy 1 1 2 1
Surgery, chemotherapy, and radiotherapy 62 83 180 79
Chemotherapy type
CMF 1 1 4 2
Anthracyclines 32 43 99 44
Taxanes 8 11 16 7
CMF and anthracyclines 10 13 44 19
Anthracyclines and taxanes 20 27 48 21
CMF, anthracyclines, and taxanes 0 0 2 1
Missing 4 5 14 6
Baseline General Fatigue score
ⱕ16 36 48 113 50
17 or 18 24 32 68 30
19 or 20 15 20 46 20
(continued in next column)
Table 1. Sociodemographic and Clinical Characteristics of the Patient
Sample (continued)
Characteristic
Usual Care
Group
(n ⫽75)
Acupuncture
Group
(n ⫽227)
No. of
Patients %
No. of
Patients %
Question rating
Faith in acupuncture
Median 7 7
Range 0-10 0-10
No. of patients with missing data 2 8
Expectations of acupuncture in general
Median 7 7
Range 0-10 0-10
No. of patients with missing data 2 7
Expectations of acupuncture with fatigue
Median 7 7
Range 0-10 0-10
No. of patients with missing data 3 8
Abbreviation: CMF, cyclophosphamide, methotrexate, and fluorouracil.
AcupunctureControl
MFI: General Fatigue (week 6-baseline)
-5-10-15 0 105
Fig 2. Change in Multidimensional Fatigue Inventory (MFI): General Fatigue
scores from baseline to 6 weeks.
Acupuncture for Fatigue in Patients With Breast Cancer
www.jco.org © 2012 by American Society of Clinical Oncology 4473
Downloaded from jco.ascopubs.org on December 9, 2015. For personal use only. No other uses without permission.
Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
been a difficult area in the past in which to demonstrate an improve-
ment.
9
The improvements in psychological distress and emotional
adjustment are also interesting, particularly because fatigue has been
shown to be part of symptom clusters alongside anxiety and depres-
sion,
15
and acupuncture could be an approach to manage clusters of
symptoms rather than only single symptoms.
The evidence around the role of acupuncture for supportive care
in cancer is increasing, and positive trials exist for the management of
hot flushes, aromatase inhibitor–induced arthralgia, chemotherapy-
induced nausea and vomiting, and xerostomia.
16,17
Why acupuncture
helps to reduce fatigue is not known, and this should be the focus of
future work. However, we know acupuncture has an effect on inflam-
matory cytokines, T lymphocytes, and various peptides,
18
and recent
results suggest that cytokines and tumor necrosis factor
␣
signaling are
contributing factors in the development of fatigue.
19,20
Although it is acknowledged that a placebo effect in an acupunc-
ture trial may be possible, a sham acupuncture needling method was
not included because of debate surrounding sham methods. It is
increasingly believed a sham controlled trial is only appropriate when
two acupuncture interventions are compared. In addition, sham acu-
puncture designs cannot detect the whole placebo effect, may generate
false-negative results,
21-24
and may introduce ethical and practical
dilemmas.
25
Experimental and clinical studies have also shown that
minimal or sham acupuncture used as placebo control is not neces-
sarily inert from a physiologic perspective, and its relevance as placebo
acupuncture must be questioned,
26,27
even when it is not used as per
acupuncture principles.
22,28
Instead of reducing bias, sham designs
may introduce bias against the treatment being tested.
29
To regard
placebo acupuncture as a universally effective super placebo would be
inappropriate, and results should be interpreted with care.
30,31
We followed a pragmatic design focusing on effectiveness rather
than efficacy that best reflects the likely clinical response in practice,
but we acknowledge that acupuncture may have a placebo effect.
32
We
enhanced usual care by providing participants with a booklet about
fatigue management options, therapists in the acupuncture group
communicated and stayed with patients only to facilitate treatment,
and there was no evidence of expectancy in the results; these facts
partly control for placebo effects as a result of increased contact and
expectation. Future acupuncture trials should include an active con-
trol arm, such as attention control, educational intervention, or exer-
cise, alongside a no treatment arm or wait-list arm, so nonspecific and
placebo effects of acupuncture can be assessed more appropriately.
Findings need to be set in context of the study’s limitations.
Because of self-referral and the multiplicity of sites involved in recruit-
ment, it was not possible to obtain data on how many patients were
approached for participation; hence, we do not know whether our
Table 3. Difference in Outcomes at Week 6 for Fatigue, Quality of Life, Anxiety, and Depression
Scale Score Range No. of Patients Missing Data Equal Slopes PTrial Arm PAcupuncture Effect
ⴱ
SE 95% CI
MFI
GF 0-20 56 .23 ⬍.001 ⫺3.11 0.44 ⫺3.97 to ⫺2.25
GF (LVCF) 0-20 0 .35 ⬍.001 ⫺2.49 0.44 ⫺3.29 to ⫺1.69
PF 0-20 56 .40 ⬍.001 ⫺2.36 0.45 ⫺3.25 to ⫺1.47
RA 0-20 61 .68 ⬍.001 ⫺2.29 0.41 ⫺3.10 to ⫺1.48
RM 0-20 60 .93 ⬍.001 ⫺2.02 0.40 ⫺2.82 to ⫺1.22
MF 0-20 61 .14 ⬍.001 ⫺1.94 0.44 ⫺2.81 to ⫺1.07
HADS
Anxiety 0-21 70 .94 ⬍.001 ⫺1.83 0.44 ⫺2.69 to ⫺0.97
Depression 0-21 67 .67 ⬍.001 ⫺2.13 0.36 ⫺2.85 to ⫺1.41
FACT-B
PWB 0-28 63 .006 ⬍.001 3.30 0.57 2.17 to 4.43
SFWB 0-28 63 .08 .05 1.05 0.54 ⫺0.01 to 2.11
EWB 0-24 62 .04 ⬍.001 1.93 0.49 0.96 to 2.90
FWB 0-28 85 .29 ⬍.001 3.57 0.61 2.38 to 4.76
NOTE. Analysis of covariance was used with baseline values as covariate and trial arm as grouping factor. For conciseness, only the trial arm estimates are
given here.
Abbreviations: EWB, Emotional Well-Being; FACT-B, Functional Assessment of Cancer Therapy–Breast Cancer; FWB, Functional Well-Being; GF, General Fatigue;
HADS, Hospital Anxiety and Depression Scale; LVCF, last value carried forward; MF, Mental Fatigue; MFI, Multidimensional Fatigue Inventory; PF, Physical Fatigue;
PWB, Physical Well-Being; RA, Reduced Activity; RM, Reduced Motivation; SFWB, Social/Family Well-Being.
ⴱ
Estimated difference in week 6 score (acupuncture– control) given equal baseline scores.
Table 2. Fatigue Change Scores Analyses
Measure
Complete Patients
(n ⫽246) LVCF (n ⫽302)
Mean score (GF.6–GF.0), SC ⫺0.62 ⫺0.53
Mean score (GF.6–GF.0), Acu ⫺3.72 ⫺2.96
Difference (Acu–SC) in mean change in
GF (GF.6–GF.0) ⫺3.10 ⫺2.43
95% CI ⫺3.98 to ⫺2.23 ⫺3.19 to ⫺1.67
P⬍.001 ⬍.001
Covariance model estimates for GF
Baseline GF
Estimate 0.47
SE 0.07
Acupuncture
Estimate ⫺3.11
SE 0.44
Abbreviations: Acu, acupuncture arm; GF, General Fatigue; GF.0, General
Fatigue at baseline; GF.6, General Fatigue at 6 weeks; LVCF, last value carried
forward; SC, standard care.
Molassiotis et al
4474 © 2012 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on December 9, 2015. For personal use only. No other uses without permission.
Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
sample is highly selected or whether it differs from those who were
ineligible. One of the difficulties in recruitment was to identify women
with moderate or severe fatigue levels; despite the literature showing
high incidence of CRF, much of the fatigue reported by patients in our
study was of low severity, and we needed to screen many hundreds of
patients to identify those with at least moderate CRF. It is also impor-
tant to emphasize that acupuncture is not just a simple needling
technique eliciting specific physiologic effects, but wide improve-
ments in patient-reported outcomes may indicate nonspecific placebo
effects. We also need to recognize that improvements in outcome
variables in an unblinded trial may be susceptible to assessment bias,
although the secondary patient-reported outcomes we have used are
well known to co-occur with fatigue and impact on quality of life.
33-35
There was a 20% missing data rate (which was expected and incorpo-
rated at the time of sample size calculations), which was partly ad-
dressed through sensitivity analysis; however, a 7% difference in
missingness between the two arms, although not statistically signifi-
cant, needs to be considered when results are interpreted. Also, the
trial was powered for the primary outcome analysis; all other analyses
on secondary outcomes need to be considered corroborative or hy-
pothesis generating rather than definitive. Finally, lack of blinding
may be a concern, but in the absence of credible controls for an
acupuncture trial, we chose a pragmatic trial design. Inclusion of
objective physiologic data in future trials will provide stronger evi-
dence of effect in acupuncture trials.
The trial shows that patients in the intervention group reported
better fatigue, anxiety, depression, and quality-of-life scores, but the
design of the trial does not allow the specific effect of acupuncture to
be distinguished from other elements and effects of the intervention.
Although this trial provides some evidence of effectiveness, further
effectiveness trials using appropriate controls, where hypotheses and
study design are formulated based on conditions in routine practice,
and outcomes essential for clinical decisions are warranted before
recommending widespread use of acupuncture in patients with breast
cancer. Acupuncture could be a treatment option for CRF, although
its availability may be an issue in many health care services and patients
may not be able to afford private costs. Future research should con-
centrate on cost-effectiveness of acupuncture, because potentially ex-
pensive acupuncture treatments may be offset by reduced societal
costs and health care utilization in those suffering significant and
protracted fatigue after treatment. Future studies should collect long-
term outcome data for fatigue that could make results more convinc-
ing. The use of acupuncture should also be tested in other cancer
diagnostic groups.
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS
OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Alexander Molassiotis, Peter Mackereth, David
W. Ryder, Jacqueline Filshie, Alison Richardson
Administrative support: David W. Ryder
Provision of study materials or patients: Peter Mackereth,
Jacqueline Filshie
Collection and assembly of data: Joy Bardy, Jennifer Finnegan-John,
Peter Mackereth, Emma Ream
Data analysis and interpretation: Alexander Molassiotis,
David W. Ryder
Manuscript writing: All authors
Final approval of manuscript: All authors
REFERENCES
1. Servaes P, Gielissen MF, Verhagen S, et al:
The course of severe fatigue in disease-free breast
cancer patients: A longitudinal study. Psychooncol-
ogy 16:787-795, 2007
2. Servaes P, Verhagen S, Bleijenberg G: Deter-
minants of chronic fatigue in disease-free breast
cancer patients: A cross-sectional study. Ann Oncol
13:589-598, 2002
3. Molassiotis A, Scott JA, Kearney N, et al:
Complementary and alternative medicine use in
breast cancer patients in Europe. Support Care
Cancer 14:260-267, 2006
4. Molassiotis A, Fernadez-Ortega P, Pud D, et
al: Use of complementary and alternative medicine
in cancer patients: A European survey. Ann Oncol
16:655-663, 2005
5. Sood A, Barton DL, Bauer BA, et al: A critical
review of complementary therapies for cancer-
related fatigue. Integr Cancer Ther 6:8-13, 2007
6. Vickers AJ, Straus DJ, Fearon B, et al: Acu-
puncture for postchemotherapy fatigue: A phase II
study. J Clin Oncol 22:1731-1735, 2004
7. Tsay SL, Cho YC, Chen ML: Acupressure and
Transcutaneous Electrical Acupoint Stimulation in im-
proving fatigue, sleep quality and depression in hemodi-
alysis patients. Am J Chin Med 32:407-416, 2004
8. Yuemei L, Hongping L, Shulan F, et al: The
therapeutic effects of electrical acupuncture and
auricular-plaster in 32 cases of chronic fatigue syn-
drome. J Tradit Chin Med 26:163-164, 2006
9. Molassiotis A, Sylt P, Diggins H: The manage-
ment of cancer-related fatigue with acupuncture and
acupressure: A randomised controlled trial. Comple-
ment Ther Med 15:228-237, 2007
10. MacPherson H, Altman DG, Hammerschlag
R, et al: Revised STandards for Reporting Interven-
tions in Clinical Trials of Acupuncture (STRICTA):
Extending the CONSORT statement. PLoS Med
7:e1000261, 2010
11. Xin-Nong C: Chinese Acupuncture and Moxi-
bustion (revised ed). Beijing, China, Foreign Lan-
guage Press, 1999
12. Smets EM, Garssen B, Bonke B, et al: The
Multidimensional Fatigue Inventory (MFI) psycho-
metric qualities of an instrument to assess fatigue.
J Psychosom Res 39:315-325, 1995
13. Zigmond AS, Snaith RP: The hospital anxiety
and depression scale. Acta Psychiatr Scand 67:361-
370, 1983
14. Fairclough DL, Cella DF: Functional Assess-
ment of Cancer Therapy (FACT-G). Qual Life Res
5:321-329, 1996
15. So WK, Marsh G, Ling WM, et al: The symp-
tom cluster of fatigue, pain, anxiety, and depression
and the effect on the quality of life of women
receiving treatment for breast cancer: A multicenter
study. Oncol Nurs Forum 36:E205-E214, 2009
16. Chao LF, Zhang AL, Liu HE, et al: The efficacy
of acupoint stimulation for the management of
therapy-related adverse events in patients with
breast cancer: A systematic review. Breast Cancer
Res Treat 118:255-267, 2009
17. O’Regan D, Filshie J: Acupuncture and can-
cer. Auton Neurosci 157:96-100, 2010
18. Stone JA, Johnstone PA: Mechanisms of ac-
tion for acupuncture in the oncology setting. Curr
Treat Options Oncol 11:118-127, 2010
19. Bower JE, Ganz PA, Irwin MR, et al: Inflam-
mation and behavioral symptoms after breast can-
cer treatment: Do fatigue, depression, and sleep
disturbance share a common underlying mecha-
nism? J Clin Oncol 29:3517-3522, 2011
20. Bower JE: Cancer-related fatigue: Links with
inflammation in cancer patients and survivors. Brain
Behav Immun 21:863-871, 2007
21. Paterson C, Dieppe P: Characteristic and inci-
dental (placebo) effects in complex interventions
such as acupuncture. BMJ 330:1202-1205, 2005
22. Kaptchuk TJ, Stason WB, Davis RB, et al:
Sham device v inert pill: Randomised controlled
trial of two placebo treatments. BMJ 332:391-
397, 2006
23. White AR, Filshie J, Cummings TM, et al:
Clinical trials of acupuncture: Consensus recom-
mendations for optimal treatment, sham controls
and blinding. Complement Ther Med 9:237-245,
2001
24. Mason S, Tovey P, Long AF: Evaluating com-
plementary medicine: Methodological challenges of
randomised controlled trials. BMJ 325:832-834,
2002
Acupuncture for Fatigue in Patients With Breast Cancer
www.jco.org © 2012 by American Society of Clinical Oncology 4475
Downloaded from jco.ascopubs.org on December 9, 2015. For personal use only. No other uses without permission.
Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
25. Park JJ, Bang H: Methodological advances
needed in analysis and interpretation of sham acupunc-
ture validation studies. Acupunct Med 29:168-169, 2011
26. Campbell A: Point specificity of acupuncture
in the light of recent clinical and imaging studies.
Acupunct Med 24:118-122, 2006
27. Tsukayama H, Yamashita H, Kimura T, et al:
Factors that influence the applicability of sham nee-
dle in acupuncture trials: Two randomized, single-
blind, crossover trials with acupuncture-experienced
subjects. Clin J Pain 22:346-349, 2006
28. McManus CA, Schnyer RN, Kong J, et al:
Sham acupuncture devices: Practical advice for re-
searchers. Acupunct Med 25:36-40, 2007
29. Birch S: A review and analysis of placebo
treatments, placebo effects, and placebo controls in
trials of medical procedures when sham is not inert.
J Altern Complement Med 12:303-310, 2006
30. Madsen MV, Gøtzsche PC, Hro´ bjartsson A:
Acupuncture treatment for pain: Systematic review
of randomised clinical trials with acupuncture, pla-
cebo acupuncture, and no acupuncture groups. BMJ
338:a3115, 2009
31. Streitberger K, Kleinhenz J: Introducing a pla-
cebo needle into acupuncture research. Lancet 352:
364-365, 1998
32. Linde K, Witt CM, Streng A, et al: The impact
of patient expectations on outcomes in four random-
ized controlled trials of acupuncture in patients with
chronic pain. Pain 128:264-271, 2007
33. Kim SH, Son BH, Hwang SY, et al: Fatigue and
depression in disease-free breast cancer survivors:
Prevalence, correlates, and association with quality
of life. J Pain Symptom Manage 35:644-655, 2008
34. Bower JE: Behavioral symptoms in patients
with breast cancer and survivors. J Clin Oncol 26:
768-777, 2008
35. Fiorentino L, Rissling M, Liu L, et al: The
symptom cluster of sleep, fatigue and depressive
symptoms in breast cancer patients: Severity of the
problem and treatment options. Drug Discov Today
Dis Models 8:167-173, 2011
■■■
2013 ASCO Annual Meeting
Each year, ASCO organizes a wide array of high-quality meetings that provide educational and scientific programs to
advance our understanding of cancer. At each of ASCO’s meetings, you can expect an engaging and interactive agenda
featuring high-level scientific or clinical abstracts and educational sessions led by world-class faculty. Join us to earn CME
credit, network with colleagues, and interact with cancer experts.
Join more than 25,000 oncology professionals from a wide range of specialties at the world’s premier oncology event,
May 31-June 4, 2013, in Chicago, Illinois.
Molassiotis et al
4476 © 2012 by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on December 9, 2015. For personal use only. No other uses without permission.
Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
