Research Articles

Integrative Cancer Therapies9(2) 158 –167© The Author(s) 2010Reprints and permission: http://www. sagepub.com/journalsPermissions.navDOI: 10.1177/1534735409360666http://ict.sagepub.com

Acupuncture as Palliative Therapy for Physical Symptoms and Quality of Life for Advanced Cancer Patients

Elizabeth Dean-Clower, MD, MPH,1,5 Anne M. Doherty-Gilman, MPH,1 Aparna Keshaviah, ScM,4 Frank Baker, PhD,2 Chiewkwei Kaw, MS,3 Weidong Lu, MB, PhD, MPH, Lic. Ac,1,5 Judith Manola, MS,1 Richard T. Penson, MD, MRCP, 4,5 Ursula A. Matulonis, MD,1,5 and David S. Rosenthal, MD1,5

Abstract

Background. Acupuncture is underutilized as an adjunct cancer therapy. The main study objectives were to determine the feasibility of administering acupuncture as palliative therapy to patients with advanced ovarian or breast cancer and to assess the effect on symptoms and quality of life (QOL). Methods. This study was a pilot, single-armed prospective clinical trial for patients with advanced cancer to receive 12 acupuncture sessions over 8 weeks with follow-up at weeks 9 and 12. Ambulatory patients with advanced ovarian or breast cancer were enrolled to receive treatments at an outpatient academic oncology center. Symptom severity was measured before and after each acupuncture session. A composite QOL assessment tool, consisting of validated instruments, was completed at 5 time points. Results. Forty patients enrolled in the study. Twenty-eight patients (70%; 95% confidence interval [CI] = 53%-83%) completed 4 weeks of treatment, and 26 patients (65%; 95% CI = 48%-79%) completed 8 weeks. Eight patients (20%) withdrew before receiving acupuncture, and 6 patients (15%) discontinued treatment early because of disease progression or scheduling demands. Among all 32 assessed patients, there was self-reported improvement immediately post-treatment in anxiety, fatigue, pain, and depression and significant improvement over time for patients with anxiety (P = .001) and depression(P = .02). Among patients experiencing baseline symptoms, there was improvement in anxiety (P = .001), fatigue (P = .0002), pain (P = .0002), and depression (P = .003). QOL measures of pain severity and interference, physical and psychological distress, life satisfaction, and mood states showed improved scores during treatment, with sustained benefit at 12 weeks. Conclusions. This pilot study demonstrates that an 8-week outpatient acupuncture course is feasible for advanced cancer patients and produces a measurable benefit that should be evaluated in controlled trials.

Keywords

acupuncture, palliative care, symptom management, ovarian cancer, breast cancer

Introduction

In 2009, the estimated number of new cases of ovarian cancer was 21,550 and estimated deaths were 14,600.1 Given the difficulties in early detection, women are often not diagnosed until they have reached an advanced stage.2 Addi-tionally, although chemotherapy is quite effective initially, more than 80% of patients will eventually experience a recurrence of their cancer. In both situations, palliative treat-ment is the goal.3 In 2009, the estimated number of new cases of female breast cancer was 192,370 and estimated deaths were 40,170.1 Although most breast cancer is diag-nosed at an early stage and can be cured, a substantial number of individuals are diagnosed at an advanced stage, when palliative rather than curative treatments are currently

used. Patients with either advanced ovarian or breast cancer may experience difficult disease and treatment-related symptoms and often struggle with diminished quality of life (QOL).4,5

1 Dana-Farber Cancer Institute, Boston, MA, USA2 Center for Medical Outcomes Research, School of Health Sciences & Practice, New York Medical College, Valhalla, NY, USA

3 American Cancer Society, Atlanta, GA, USA4 Massachusetts General Hospital, Boston, MA, USA5 Harvard Medical School, Boston, MA, USA

Corresponding Author:David S. Rosenthal, MD, Leonard P. Zakim Center for Integrative Therapies, Dana-Farber Cancer Institute, 44 Binney St., Boston, MA 02115 USA.E-mail: drose@uhs.harvard.edu

Dean-Clower et al. 159

These patients face daily physical, psychological, spiri-tual, and/or emotional challenges. Though their cancer may be incurable, comprehensive care includes palliative thera-pies that alleviate symptoms and can improve patients’ QOL. Nonpharmacological interventions provided by comple-mentary or integrative therapies, such as acupuncture, may provide some benefit to patients receiving conventional treatment.6-9 Cancer patients and survivors have demonstrated a remarkable interest in complementary and alternative medicine (CAM), with reported use in the United States ranging from 34% to 91%, depending on the particular cancer population studied and whether prayer and exercise are included.10-13 For clinicians, a lack of familiarity with these modalities and with the relevant evidence-based research, safety concerns, and perhaps skepticism may con-tribute to the overall low recommendation rates.14-17

Among complementary and integrative modalities, acu-puncture is relatively less used, despite published evidence of its potential benefits.18,19 Cancer patients’ use of acupunc-ture is estimated to be from 1.7% to 31%.20-22 In addition to some clinicians’ lack of familiarity with this therapy, others may not mention acupuncture because of safety concerns for cancer patients, despite an excellent track record of patient safety.23 Additionally, a clear, straightforward biological mechanism is not yet understood, though acupuncture is an active field of research interest.19 Acupuncture’s benefit for cancer patients with chemotherapy-induced nausea and vomiting was one of the first uses agreed on at the NIH consensus conference in 199724 and has been reconfirmed in other studies.9,25 Acupuncture also benefits patients with other cancer- or treatment-related symptoms, such as pain,7 fatigue,6,26,27 anxiety,27 and overall well-being.27-29 However, the efficacy of acupuncture as an adjunct to palliative che-motherapy is not widely recognized; therefore, it is rarely offered as an adjunct to standard care in the United States.30

The primary aim of this single-arm clinical trial was to evaluate the feasibility and preliminary effects of acupunc-ture for patients with advanced ovarian and breast cancer. Secondary aims were to assess the change in symptom severity over time and to assess the utility of select QOL instruments in measuring symptoms in an acupuncture study.

MethodsParticipants

Patients were recruited from the gynecological and breast oncology programs at 2 academic medical centers: Dana-Farber/Brigham and Women’s Cancer Center and Massachusetts General Hospital, in Boston, Massachusetts. Patients signed consent forms approved by the institutional review board. The acupuncture treatments were provided free of charge to study participants; there was no additional financial incentive provided for patient participation.

The eligibility criteria for this trial were the following: evidence of metastatic ovarian cancer or other advanced can-cers in which there had been a failure of adjuvant therapy and for which there were no curative options; ambulatory status; use of conventional palliative therapy, including chemother-apy; complaints of pain, nausea, and/or inadequate QOL; age > 21years; no acupuncture in the prior 6 months; plate-lets > 25 000 cells/mL; absolute neutrophil count > 500 cells/mL; no history of bleeding disorder; no diagnosis of acute psychosis; Karnofsky performance scale score > 60; and ability to speak and read English. Patients enrolled in other complementary medicine or QOL studies were not eligible.

The target enrollment for the study was 40 patients. The study was open to enrollment in April 2002 and closed to patient accrual in April 2004, after reaching the target accrual.

TreatmentPatients received 8 weeks of acupuncture, given twice weekly during the first 4 weeks and weekly during the last 4 weeks (total of 12 sessions). For twice weekly acupuncture, the sessions were scheduled at least 1 day apart. All acupunc-ture treatments were performed on-site at the Leonard P. Zakim Center for Integrative Therapies at Dana-Farber Cancer Institute. Three licensed acupuncturists, all trained in traditional Chinese medicine with at least 10 years clinical experience, administered the treatments. As part of treatment follow-up, patients were not to receive any additional acupuncture until after completion of the study (week 12).

A standard, predetermined acupuncture protocol (SAP) and symptom-specific acupuncture protocols (SSPs) were developed based on commonly used points from the medi-cal literature as well as acupuncturists’ experiences and consensus. The acupuncture algorithm used was that all patients received the SAP, and if they were experiencing nausea, pain, fatigue, depression, anxiety. insomnia, consti-pation, or cough at the time of the session, they also received a SSP of preselected points based on specific symptom(s).

The base protocol consisted of the following 10 acu-puncture points at 17 sites: ST-36, SP-6, LI-11, LI-4, LV-3, SP-9, Yintang, GV-20, CV-6, and Shenmen. Point selection under the SSP for each category was made from the follow-ing: PC-6, CV-12, ST-37 (nausea); Ashi points (pain); K-3, TW-5, GV-14, LU-9 (fatigue); HT-7, PC-4, LV-5, GV24 (depression/anxiety); GB-20, HT-7, PC-7 (insomnia); ST-25, TW-6, BL-57 (constipation); LU-7, LU-6, LU-10, CV-22, BL13 (cough). A study patient received the full SAP base protocol if there were no specific symptoms. If a patient reported a specific symptom, a minimum of one symptom-specific point from that category would be selected in addition to the SAP. For each additional point used for a specific symptom, an effort was made to deduct the same number of points in the SAP to maintain approxi-mately the same number of acupuncture points for each

160 Integrative Cancer Therapies 9(2)

treatment session. At each session, stainless steel acupunc-ture needles (Vinco, Helio Medical Supplies, USA) with a size 36 gauge, 1 inch (0.20 mm × 25 mm) were inserted into these points and needles were manually manipulated to obtain a “De Qi” sensation—a mild, dull aching sensation. Each session lasted 30 minutes and occurred in a quiet room with dimmed lights. No electrostimulation was applied.

At each session, data were obtained on the presence and severity of the specific symptoms noted above; “other” symptoms reported by patients were recorded, but not treated. Before each acupuncture session (“pre-session”), the acupuncturist asked the patient whether or not each symptom was present, and if so, recorded the symptom severity (scale of 0-10; 10 = most severe). After the session, the acupuncturist completed a second form to record the post-session symptom severity. The patient’s responses were recorded on a standard form. Information about any new medical or complementary treatment initiated since the prior session was also recorded.

InstrumentsThe QOL assessment tool consisted of individual assess-ment tools used in other studies involving cancer patients or survivors: the Brief Pain Inventory (BPI),31,32 Rotterdam Symptom Checklist (RSCL),4,33 Satisfaction With Life Domains Scale–Cancer (SLDS-C),34-36 Profile of Mood States (POMS),37-40 and Cancer Coping Methods.41

The questionnaires were completed by patients at 5 time points: baseline before treatment (ranging from 1 week to immediately prior to the first treatment session), weeks 4 and 8 of treatment (30 minutes after completion of the acu-puncture session), and weeks 9 and 12 (i.e., 1 week and 4 weeks after the final acupuncture session). Patients were allowed to mail in the questionnaires for the final 2 time points.

Brief Pain Inventory. The BPI consists of 2 subscales that measure (1) the severity of cancer pain (pain severity) and (2) the impact of pain on the patient’s functioning (pain interference) in the past 24 hours on a scale from 0 (no pain or no interference) to 10 (highest level).31,32

Rotterdam Symptom Checklist. The original RSCL included physical symptom and psychological distress sub-scales.4 Stein et al.33 developed a shorter version that omitted the psychological distress items and added some physical symptom items. The version used here added 3 items to the Physical Symptom Distress Scale (Drowsiness, Weak All Over, and Feel Listless [Washed Out]) and dropped 2 items (Low Back Pain and Burning/Sore Eyes). Respondents were asked to indicate the extent to which they have been bothered by each of 32 symptoms during the past week, using a response format of “not at all” to “very much.”

Satisfaction With Life Domains Scale–Cancer. The SLDS-C is a broad measure of QOL that asks respondents to indicate their current satisfaction with 18 different life domains rel-evant to the QOL of cancer patients.34-36 The scale uses a

picture response format in which satisfaction is indicated by choosing 1 of 7 faces, ranging from a “terrible” frowning face to a “delighted” smiling face. Items can be evaluated individually and a total score calculated.

Profile of Mood States–Short Form (POMS-SF). The POMS was originally developed to assess transient distinct mood states using ratings of 65 adjectives that were rated on a 5-point scale from “not at all” to “extremely.” It was scored using 6 factor-based subscales: Depression- Dejection, Tension-Anxiety, Anger-Hostility, Fatigue- Inertia, Confusion-Bewilderment, and Vigor-Activity.37 The POMS-SF that was used in this study is a 37-item ver-sion of the POMS developed and validated on the responses of cancer patients and that retains the 6 subscales.38-40 A seventh score of Total Negative Mood Disturbance is cal-culated by adding together the first 5 subscales and subtracting the score from the one positively scored sub-scale, Vigor-Activity.

Cancer Coping Methods. The Cancer Coping Methods instrument used in this study is from the questionnaires included in the American Cancer Society’s (ACS’s) Studies of Cancer Survivors.41 The Cancer Coping Methods measure consists of a list of complementary therapies plus the addi-tion of some questions regarding talking with other people, which were included based on what patients listed as “other methods” on pilot tests of the ACS questionnaires. In the version used in the large ACS surveys and in our acupunc-ture study, we used a shorter 32-item list from the original 44-item list that both omitted methods that were little used and added the items related to talking.41

Statistical MethodsFor the targeted enrollment of 40 patients, feasibility was defined as 75% of patients completing at least 4 weeks of treatment and/or 50% of patients completing all 8 weeks of treatment. If 30 of 40 patients completed 4 weeks of treat-ment, the 95% confidence interval (CI) for the true 4-week completion rate would be 58.8% to 87.3%. If 20 patients completed 8 weeks of treatment, the 95% CI for the true 8-week completion rate would be 33.8% to 66.2%. A patient was allowed to miss 1 session of treatment at most and still have completed the treatment protocol.

To determine whether any immediate changes after acupuncture treatment had occurred, difference scores (post-treatment minus pre-treatment) were calculated for each patient at each time point. Means and 95% CIs at each time point were computed. Given that acupuncture typically is administered as a course with cumulative benefit, the effect of acupuncture over time (i.e., the change in symptom severity across sessions) was tested. Two separate longitu-dinal analyses were done, one based on all assessed patients (because symptoms could occur at any time over the dura-tion of treatment) and another based on patients who were experiencing a symptom just prior to treatment on day 1.

Dean-Clower et al. 161

Preacupuncture treatment severity score was modeled as the dependent variable, and week was the independent vari-able. If there were 2 sessions in a week, the second session was treated as occurring one half week after the first ses-sion. A compound symmetry covariance matrix was used, and a likelihood ratio test with 11 degrees of freedom was used to determine whether or not there was a statistically significant change in symptom severity over time.

Exploratory analyses tested differences in mean symptom severity by age at study entry (dichotomized as ≤50 vs > 50 years) and years since diagnosis (dichotomized as <2 vs ≥ 2 years) using 2-sample Student’s t tests. To examine whether the longitudinal effect of acupuncture was altered by these characteristics, interaction terms between the characteristic and time were included in the aforementioned repeated-measures, mixed-effects models. The interactions between age and session and between years since diagnosis and ses-sion were tested individually (i.e., in separate models) to conserve power. However, the tests were anticipated to be underpowered given the sample size of this small pilot study.

For all analyses, a 2-sided significance level of .05 was used. Given that this was an exploratory pilot study, P values were not adjusted for multiple comparisons. Analyses were conducted by statisticians (JM, AK) using SAS V8.2.

QOL Statistical ConsiderationsData analysis on patients’ QOL was conducted (by CK) in SPSS V13.0 and SAS V8.2. If at least half of the items on an instrument were completed, the total score was imputed using the mean of available items. Descriptive statistics were gener-ated to understand the change in QOL scores across different time points. Similar longitudinal models with the consider-ation of demographic and medical characteristic differences (e.g., age, disease stage, treatment, etc.) among the study par-ticipants were tested to further evaluate the association between acupuncture effects and patient’s QOL. Covariance structures with the best-fit statistics were used in the final analyses.

ResultsFeasibility

Twenty-eight of the 40 enrolled patients (70%; 95% CI for treatment completion rate = 53%-83%) completed at least 4 weeks of acupuncture treatment, and 26 patients (65%; 95% CI = 48%-79%) completed all 8 weeks of treatment. The 8-week completion rate met the feasibility criteria, whereas the 4-week completion rate was slightly lower than anticipated.

A total of 8 patients (20%) withdrew from the study before receiving acupuncture because of disease progression (n = 2) or scheduling difficulties/time commitment (n = 6). Six of 32 patients (15%) began study treatment, but withdrew early because of progressive disease (n = 4) or withdrawal of informed consent (n = 2). Therefore, the overall drop-out rate

was 35% (14 of 40 patients). Two acupuncturists adminis-tered 99% of the treatments (WL, Zhi Ping Li). Among the 26 patients who completed treatment, 21 completed treat-ment as scheduled, 4 missed 1 treatment session, and 1 completed all 12 sessions but had her final session 3 weeks later than expected (because of travel). The acupuncture treatments were well-tolerated. No significant side effects or adverse events attributable to acupuncture were observed.

Baseline CharacteristicsTable 1 summarizes baseline patient and disease character-istics for all 40 registered patients. The median age was 52

Table 1. Baseline Characteristics

Characteristic n Percentage

Gender and race Female and White 40 100%

Age at study entry 30-39 3 7.5%40-49 11 27.5%50-59 16 40.0%60-69 6 15.0%70 Or older 4 10.0%

Karnofsky performance status 70-79 6 15.0%80-89 18 45.0%90-99 14 35.0%100 2 5.0%

Disease group Ovarian/Fallopian 24 60.0%Breast 9 22.5%Endometrial/Gynecologicala 7 17.5%

Years since diagnosis <2 Years 17 42.0%≥2 Years 23 57.0%

Prior surgery Yes 35 87.5%No 5 12.5%

Prior radiation Yes 12 30.0%No 28 70.0%

Prior chemotherapy Yes 35 87.5%No 5 12.5%

Prior chemotherapy regimens 0 0 0%1-4 11 27.5%5-8 20 50.0%9 Or more 2 5.0%Missing/Unknown 7 17.5%

Concurrent chemotherapy Yes 25 65.8%No 13 34.2%Missing 2

NOTES: aOvarian cancer is used in the text to include other gynecologic malignancies, specifically Müllarian tumors, which are treated with the same regimens.

162 Integrative Cancer Therapies 9(2)

(range, 31-75). All patients had received prior conventional therapy (surgery, chemotherapy, and/or radiation therapy). At the time of study entry, 25 patients (63%) were receiving concurrent palliative chemotherapy. Although all patients had advanced cancer, 85% had KPS (Karnofsky Perfor-mance Status) scores greater than 80 at the time of receiving acupuncture.

Seven patients (18%) reported initiating a new form of medical treatment and/or complementary therapy over the course of the 8-week acupuncture treatment. These thera-pies were evenly divided between conventional therapy (chemotherapy) and CAM therapy (e.g., Chinese herbal mix-ture); 1 patient began both chemotherapy and vitamins.

Table 2 summarizes the symptom severity scores reported at baseline for the 32 patients who received any acupuncture treatment. Of these, 30 patients had at least one baseline symptom, and many patients were experiencing multiple symptoms. The most prevalent baseline symptoms were fatigue, anxiety, and pain, individually or in combina-tion. For most symptoms, the mean severity experienced was at a moderate level (4-5 out of 10), though some patients had higher severity scores of 8 to 10. Two patients were asymptomatic at baseline, but described experiencing decreased QOL and therefore were eligible to participate. There were no significant differences in baseline symptom severity by age.

Immediate Effects of Acupuncture TreatmentAssessment of change in symptom severity scores (post-treatment score minus pre-treatment score) was made for all assessed patients. There was a significant decrease in sever-ity score at the majority of sessions for anxiety, fatigue, pain, and depression. Additionally, a total of 31 patients (97%) reported, following one or more sessions, that acu-puncture helped their overall sense of well-being.

Figure 1 shows the mean severity score before and after treatment and over the course of treatment for symp-tom categories having 13 or more patients affected at baseline. For insomnia, only pre-treatment scores are clin-ically relevant. For all other symptoms, post-treatment scores were lower than pre-treatment scores across all symptoms and time points, with the exception of depression at weeks 6 and 7.

Longitudinal Changes in Patients’ SymptomsFigure 1 also demonstrates that over time, there were statis-tically significant improvements in 4 of the 5 predominant symptoms at baseline: anxiety, fatigue, pain, and depression. The longitudinal changes in anxiety severity differed sig-nificantly by age at study entry (P = .01), with a greater decrease in anxiety for patients 50 years or younger.

Quality of LifeQOL questionnaire results generally indicated that acu-puncture had a positive impact on patients’ QOL. Arithmetic means and standard deviations for the QOL scores are pre-sented in Table 3. Model means also were generated and tested for differences across time points. Patients had better QOL during the 8-week acupuncture treatment than either before treatment was initiated or shortly after it was stopped. Symptom severity decreased over time from the pre-treatment baseline through the last week of acupuncture treatment (week 8). Symptoms or other qualitative param-eters then showed some temporary worsening 1 week later (week 9). However, the results 1 month after acupuncture was completed (week 12) continued to show improvement of QOL compared with baseline values.

Patients’ pain severity and pain interference scores were significantly improved by 1.26 (P ≤ .0001) and 1.51

Table 2. Baseline Symptom Severity Scores (0-10)

Among Patients with Symptoms

No. With Data No. With Symptoms Percentage With Symptoms Mean Median Range

Fatigue 32 21 66 4.5 4 1-10Anxiety 32 19 59 5.1 5 1-8Pain 32 17 53 5 5 1-9Depression 32 13 41 4.8 5 2-9Insomnia 32 13 41 4.5 4 2-10Constipation 31 12 39 3.7 3 1-9Nausea 32 8 25 2.9 2.5 1-8Cough 32 3 9 4.3 4 3-6Othera 30 11 37 4.3 4 1-8

NOTES: aIn all, 11 patients had severity scores >0 for the following: diarrhea (2), numbness (2), itching/rash, hot flashes, heart burn, neuropathy in feet, shortness of breath, stiffness, and weakness in arms and hands.

Dean-Clower et al. 163

Figure 1. Mean symptom severity scores before and after acupuncture treatments for symptom categories having 13 or more affected patients at baseline among only patients with symptoms just prior to treatment on day 1. Mean severity scores before treatment were generally higher than after treatment. For insomnia, only pre-treatment scores are applicable. Vertical lines represent standard errors; dotted line shows model-based estimate of slope; P values indicate significance of change over time among patients with symptoms just prior to treatment on day 1. Pre-tx = pre-treatment; Post-tx = post-treatment

(P = .0006) units, respectively, at week 8 of treatment compared to the baseline data (pain severity at baseline = 2.64; pain interference at baseline = 2.30). The study also

revealed that the pain status was better one month after the treatment was stopped compared with the time before treatment was given (pain severity and interference one

164 Integrative Cancer Therapies 9(2)

month after treatment completion were 1.59 and P = .0010, and 1.32 and P = .0055, respectively).

We also observed significant reduction in the RSCL physical (PHY) and psychological (PSY) burden by 11.53 units (P ≤ .0001) and 13.95 units (P = .0004) units, respectively, at week 8 compared with baseline data. Lower levels of physical and psychological distress were also reported 1 month after the treatment was completed as compared with the baseline values (PHY P = .0001; PSY P = .0004).

A similar trend of improvement was also observed in the SLDS-C, in which patients reported significantly higher life satisfaction at the completion of the acupunc-ture treatments (week 8) than either before the first session (14.32, P = .0002) or 1 week after the last treat-ment (6.52; P = .0013). The improvement in the SLDS-C score 1 month after the treatment versus the baseline was 7.81 units (P = .0071). For the POMS-SF, significant dif-ferences were detected for all subscales and for total mood disturbance, except for the Vigor-Activity subscale (P = .73). Patients reported 16.10 (P = .0001) and 18.39 (P ≤ .0001) improvement in their mood state at weeks 8 and 12, respectively, compared with baseline.

Regarding patients’ use of various Cancer Coping Meth-ods at baseline, the most frequent choices encompassed talking to family or friends (78%), music (68%), prayer or spiritual practice (65%), and exercise (47%), among other choices (data not shown). Nutritional supplements were used by 59% of patients. Other CAM therapies ranged from 3% to 40%, for hypnosis and meditation, respectively. Only 3 patients (10%) indicated that they had used acupuncture or acupressure as a coping strategy prior to this study.

Discussion

This study reports that acupuncture is a feasible, safe, and well-tolerated intervention for patients with advanced ovar-ian or breast cancer; therefore, the primary study aim was achieved. There was also a measurable improvement for certain physical symptoms and subjective QOL. Given that acupuncture usually requires successive treatments to be effective, it was important to learn that patients with advanced cancer, who were naïve to acupuncture, could be compliant with scheduling demands in an outpatient setting and could be followed longitudinally.

Studies that have investigated the use of acupuncture for specific symptom management in palliative care patients30,42,43 indicate that acupuncture has not yet had an impact on clinical practice, though efficacy has been dem-onstrated in 24 of 27 randomized controlled trials.30

Regarding the potential benefit of acupuncture as adjunct therapy, patients reported positive experiences both during the treatment phase and at the study’s completion, as indi-cated by improved symptom scores recorded by the acupuncturists and patients’ QOL scores from a composite self-assessment tool. In general, mean symptom scores were mild to moderate, though some more significant changes occurred pre-treatment to post-treatment or over time for specific patients and symptoms.

For the 8-week treatment duration, measures of QOL showed continued improvement, with sustained improve-ment at week 12 relative to baseline values. The decrease in QOL scores 1 week after treatment ended, followed by improvement 3 weeks later to levels comparable to the

Table 3. Quality of Life Scale Score Arithmetic Means and Standard Deviations (SDs)

Baseline Week 4 Week 8 Week 9 Week 12

Quality of Life Scalea Mean SD Mean SD Mean SD Mean SD Mean SD

Brief Pain Inventory Pain severity 2.62 2.416 1.58 1.622 0.83 1.346 1.22 1.567 0.98 1.806Pain interference 2.49 2.843 1.49 1.557 0.68 1.124 1.28 1.766 0.63 1.705

Rotterdam Symptom Checklist Physical symptoms distress 26.69 14.18 18.26 13.41 14.67 11.07 17.57 12.97 15.62 12.66Psychological distress 36.46 22.42 28.04 22.85 21.90 17.98 24.22 19.63 18.81 17.85

SLDS-C total 83.44 21.40 93.00 16.12 98.56 17.34 91.70 18.93 98.15 20.43Profile of Mood State

Total negative mood 24.91 27.46 12.82 20.47 9.31 15.56 13.08 15.83 7.13 16.95Depression 8.06 7.46 5.07 5.64 4.15 4.52 5.46 5.33 3.13 4.12Vigor 9.31 5.49 10.00 5.45 9.38 5.95 8.75 5.89 9.65 7.24Confusion 4.75 3.09 3.82 2.55 2.85 2.43 3.21 2.08 3.13 2.85Tension 7.44 5.98 4.46 3.81 3.08 3.05 4.79 3.88 3.65 4.07Anger 5.56 5.94 3.39 4.24 3.73 4.90 3.42 3.27 3.22 4.82Fatigue 8.41 6.05 6.07 5.58 4.88 3.93 4.96 3.96 3.65 3.90

NOTES: SLDS-C = Satisfaction With Life Domains Scale–Cancer.aEstimated marginal means and P values adjusted for the repeated-measure study design are included in the text.

Dean-Clower et al. 165

last acupuncture session, could reflect loss of continuity with the practitioner or simply an independent time trend.

There are several limitations to this study. Without a control group in this feasibility study, the significant find-ings cannot be directly attributed to acupuncture treatments, and all conclusions should be guarded. It is possible that patients improved because of the attentive presence of the acupuncturists, the relaxing environment, patients’ desire to please the practitioners, concurrent cancer treatment, pla-cebo effect, or a combination of these factors. It is recognized that expectations of the acupuncture procedure can influ-ence outcome measurements provided by patients in a positive way44 and may possibly reflect the Hawthorne effect,45 that is, patients performing better because they are participating in a trial.

From a statistical perspective, caution should be exer-cised in assessing symptoms and QOL over time. Some of the comparisons were based on fewer than 10 patients and are therefore likely to be underpowered. It is also possible that statistically significant differences arose merely by chance, given the large number of comparisons made. Additionally, although it is reasonable to evaluate the group of patients with symptoms at the beginning of the study, this subset analysis has a small sample size.

Given the positive changes in symptom management and improved QOL for patients that occurred over the duration of this acupuncture study, some considerations for designing future studies are warranted. Performing randomized trials, which include a control arm for stan-dard of care, is fundamentally necessary before any firm conclusions can be drawn about the utility of acupuncture

A. Rotterdam Physical Distress B. Rotterdam Psychological Distress

0

5

10

15

20

25

30

35

40

45

0 4 8 9 12Week

Mea

n

0

5

10

15

20

25

30

35

40

45

0 4 8 9 12Week

Mea

n

C. Satisfaction with Life Domain Scale - Cancer D. Profile of Mood States - Total Mood Disturbance

60

65

70

75

80

85

90

95

100

105

0 4 8 9 12

Week

Mea

n

0

5

10

15

20

25

30

35

40

45

0 4 8 9 12

Week

Mea

n

Figure 2. Model-based marginal mean estimates and standard errors over time: higher scores indicate a higher level of symptomatology for RSCL measures (A and B) and Profile of Mood States–Total Mood Disturbance (POMS-TMD, D), whereas higher scores for Satisfaction of Life Domain Scale–Cancer (SLDS-C) indicate a higher level of life satisfaction (C). Positive QOL change was observed especially during treatment.

166 Integrative Cancer Therapies 9(2)

in cancer care. Regarding the acupuncture protocol, a fully standardized protocol would minimize variability versus this protocol, which had some individualization. The choice of standardized versus individualized acupuncture remains an active topic of interest and debate in research studies. The required travel to a single urban clinic and the frequency of sessions were the biggest reported problems for patients. In future studies, to increase enrollment and retention, we could investigate the use of satellite loca-tions and increase patient awareness about the availability and potential benefits of acupuncture.

In summary, the results of this pilot study indicate that an 8-week treatment course of acupuncture is feasible for advanced cancer patients. Patients reported positive changes in QOL by validated scales, and specific symptoms showed sustained statistically significant changes, warranting fur-ther evaluation in controlled trials.

Acknowledgments

We would like to acknowledge the assistance of the following people: acupuncturists—Zhi Ping Li, MB, Lic Ac, Grant Hou, MB, Lic Ac; clinical research coordinators—Kelechi Ohiri, MD, MPH, Aravind Sugumar, MD, MPH; data management—Alyssa DellaCroce, MPH, Emily Collins, Donoria Wilkerson, Yajaira Pagan; statistics—Sabrina Khan, Qian Wang; background—Janet Abrahm, MD; protocol development—Amy Geng, MD; manu-script preparation—Katherine Connolly.

Declaration of Conflicting Interests

The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding

This study was supported by the National Center for Complemen-tary and Alternative Medicine (NCCAM) [grant # 5 R21 AT01010-02]. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NCCAM or the National Institutes of Health.

References

1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer J Clin. 2009;59:225-249.

2. Cannistra SA. Cancer of the ovary. N Engl J Med. 2004;351: 2519-2529.

3. du Bois A, Quinn M, Thigpen T, Vermorken J, Avall-Lun-dqvist E, Bookman M, et al., 2004 consensus statements on the management of ovarian cancer: final document of the 3rd International Gynecological Cancer Intergroup Ovarian Can-cer Consensus Conference (GCIG OCCC 2004). Ann Oncol. 2005;16(suppl 8):viii7-viii12.

4. de Haes J, van Knippenberg F, Neijt J. Measuring psycho-logical and physical distress in cancer patients: structure and application of the Rotterdam Symptom Checklist. Br J Cancer. 1990;62:1034-1038.

5. Penson RT, Wenzel LB, Vergote I, Cella D. Quality of life considerations in gynecologic cancer: FIGO 6th Annual Report on the Results of Treatment in Gynecological Cancer. Int J Gynaecol Obstet. 2006;95(suppl 1):S247-S257.

6. Vickers A, Straus D, Fearon B, Cassileth B. Acupuncture for postchemotherapy fatigue: a phase II study. J Clin Oncol. 2004;22:1731-1735.

7. Alimi D, Rubino C, Pichard-Léandri E, Fermand-Brulé S, Dubreuil-Lemaire ML, Hill C. Analgesic effect of auricular acupuncture for cancer pain: a randomized, blinded, con-trolled trial. J Clin Oncol. 2003;21:4120-4126.

8. Vickers A, Cassileth B. Unconventional therapies for cancer and cancer-related symptoms. Lancet Oncol. 2001;2:226-232.

9. Shen J, Wenger N, Glaspy J, et al. Electroacupuncture for control of myeloablative chemotherapy-induced emesis: a randomized controlled trial. JAMA. 2000;284:2755-2761.

10. Yates JS, Mustian KM, Morrow GR, et al. Prevalence of com-plementary and alternative medicine use in cancer patients during treatment. Support Care Cancer. 2005;13:806-811.

11. Armstrong T, Cohen MZ, Hess KR, et al. Complementary and alternative medicine use and quality of life in patients with primary brain tumors. J Pain Symptom Manage. 2006;32: 148-154.

12. Gansler T, Kaw C, Crammer C, Smith T. A population-based study of prevalence of complementary methods use by cancer survivors: a report from the American Cancer Society’s studies of cancer survivors. Cancer. 2008;113:1048-1057.

13. Mao JJ, Farrar JT, Xie SX, Bowman MA, Armstrong K. Use of complementary and alternative medicine and prayer among a national sample of cancer survivors compared to other populations without cancer. Complement Ther Med. 2007;15:21-29.

14. Wahner-Roedler DL, Vincent A, Elkin PL, Loehrer LL, Cha SS, Bauer BA. Physicians’ attitudes toward complementary and alternative medicine and their knowledge of specific therapies: a survey at an academic medical center. Evid Based Complement Alternat Med. 2006;3:495-501.

15. Milden SP, Stokols D. Physicians’ attitudes and practices regarding complementary and alternative medicine. Behav Med. 2004;30:73-82.

16. Weiger WA, Smith M, Boon H, Richardson MA, Kaptchuk TJ, Eisenberg DM. Advising patients who seek complementary and alternative medical therapies for cancer. Ann Intern Med. 2002;137(11):889-903.

17. Tasaki K, Maskarinec G, Shumay DM, Tatsumura Y, Kakai H. Communication between physicians and cancer patients about complementary and alternative medicine: exploring patients’ perspectives. Psychooncology. 2002;11:212-220.

18. Lu W, Dean-Clower E, Doherty-Gilman A, Rosenthal D. The value of acupuncture in cancer care. Hematol Oncol Clin North Am. 2008;22:631-648, viii.

19. Sagar SM. Acupuncture as an evidence-based option for symptom control in cancer patients. Curr Treat Options Oncol. 2008;9:117-126.

Dean-Clower et al. 167

20. Lafferty WE, Bellas A, Corage Baden A, Tyree PT, Standish LJ, Patterson R. The use of complementary and alternative medi-cal providers by insured cancer patients in Washington State. Cancer. 2004;100:1522-1530.

21. Cui Y, Shu XO, Gao Y, et al. Use of complementary and alter-native medicine by Chinese women with breast cancer. Breast Cancer Res Treat. 2004;85:263-270.

22. Morris KT, Johnson N, Homer L, Walts D. A comparison of complementary therapy use between breast cancer patients and patients with other primary tumor sites. Am J Surg. 2000;179:407-411.

23. MacPherson H, Thomas K, Walters S, Fitter M. The York acupuncture safety study: prospective survey of 34,000 treat-ments by traditional acupuncturists. BMJ. 2001;323:486-487.

24. NIH Consensus Conference. Acupuncture. JAMA. 1998;280:1518-1524.

25. Ezzo JM, Richardson MA, Vickers A, et al. Acupuncture-point stimulation for chemotherapy-induced nausea or vomit-ing. Cochrane Database Syst Rev. 2006;(2):CD002285.

26. Molassiotis A, Fernandez-Ortega P, Pud D, et al. Comple-mentary and alternative medicine use in colorectal cancer patients in seven European countries. Complement Ther Med. 2005;13:251-257.

27. Wang SM, Gaal D, Maranets I, Caldwell-Andrews A, Kain ZN. Acupressure and preoperative parental anxiety: a pilot study. Anesth Analg. 2005;101:666-669, table of contents.

28. Nedstrand E, Wyon Y, Hammar M, Wijma K. Psychologi-cal well-being improves in women with breast cancer after treatment with applied relaxation or electro-acupuncture for vasomotor symptom. J Psychosom Obstet Gynaecol. 2006; 27:193-199.

29. Gejervall AL, Stener-Victorin E, Möller A, Janson PO, Werner C, Bergh C. Electro-acupuncture versus conventional analgesia: a comparison of pain levels during oocyte aspira-tion and patients’ experiences of well-being after surgery. Hum Reprod. 2005;20:728-735.

30. Standish LJ, Kozak L, Congdon S. Acupuncture is underuti-lized in hospice and palliative medicine. Am J Hosp Palliat Care. 2008;25:298-308.

31. Daut RL, Cleeland CS, Flanery RC. Development of the Wis-consin Brief Pain Questionnaire to assess pain in cancer and other diseases. Pain. 1983;17:197-210.

32. Serlin R, Mendoza T, Nakamura YE, Edwards KR, Cleeland C. When is cancer pain mild, moderate or severe? Grading pain severity by its interference with function. Pain. 1995;61: 277-284.

33. Stein KD, Denniston M, Baker F, et al. Validation of a modi-fied Rotterdam Symptom Checklist for use with cancer patients in the United States. J Pain Symptom Manage. 2003; 26:975-989.

34. Baker F, Curbow B, Wingard JR. Role retention and qual-ity of life of bone marrow transplant survivors. Soc Sci Med. 1991;32(6):697-704.

35. Baker F, Curbow B, Wingard JR. Development of the satisfac-tion with life domains scale for cancer. J Psychosoc Oncol. 1992;10(3):75-90.

36. Baker F, Denniston M, Hann DM, Gesme DH, Reding D, Kennedy J. Factor structure and concurrent validity of the satisfaction with life domains scale for cancer (SLDS-C). J Psychosoc Oncol. 2007;25:1-17.

37. McNair D, Lorr M, Droppleman LF. EITS Manual for Pro-file of Mood States. San Diego, CA: Education and Industrial Training Service, 1971.

38. Shacham S. A shortened version of the Profile of Mood States. J Pers Assess. 1983;47:305-396.

39. Curran S, Andrykowski M, Studts J. Short form of the Profile of Mood States (POMS-SF): psychometric information. Psy-chol Assess. 1995;7:80-83.

40. Baker F, Denniston M, Zabora J, Polland A, Dudley W. A POMS short form for cancer patients: psychometric and struc-tural evaluation. Psychooncology. 2002;11:273-281.

41. Hann DM, Baker F, Roberts CS, et al. Use of complemen-tary therapies among breast and prostate cancer patients during treatment: a multisite study. Integr Cancer Ther. 2005;4: 294-300.

42. Faily J, Oneschuk D. Acupuncture in palliative care. Support Care Cancer. 2007;15:1003-1007.

43. Mansky PJ, Wallerstedt DB. Complementary medicine in palliative care and cancer symptom management. Cancer J. 2006;12:425-431.

44. Kaptchuk TJ, Stason WB, Davis RB, et al. Sham device v inert pill: randomised controlled trial of two placebo treat-ments. BMJ. 2006;332:391-397.

45. Braunholtz DA, Edwards SJ, Lilford RJ. Are randomized clin-ical trials good for us (in the short term)? Evidence for a “trial effect.” J Clin Epidemiol. 2001;54:217-224.