Table 3. Acupuncture in Cancer Pain.

Author (Year)	Groups Studied and Intervention	Results and Findings	Conclusions
Chen et al. 201399	Single-center RCT N = 60 stage III-IV pancreatic cancer 2 groups: 30 electroacupuncture; 30 non-penetrative needle Treatment: once daily for 30 minutes for 3 days Outcome measure: Pain intensity measured by a numeric rated scale (NRS) from 0-10 at baseline, after each treatment and 2 days after the last treatment Miscellaneous: * Patient required to have NRS from 3-6, be on stable dose of analgesics at least 72 hrs prior randomization with estimated survival time more than 1 mo and never treated by acupuncture * Blinding was not reported	Pain intensity in the electroacupuncture group decreased compared to baseline after the third day of treatment (-1.67, 95% CI -1.46 to-1.87) with little change in the control group (-0.13, 95% CI 0.08 to -0.35). There was a statistically significant difference between the two groups after the third treatment and the two-day post-treatment follow-up (P-value < 0.001). No adverse effects were observed.	Electroacupuncture can relieve mild to moderate forms of pain in patients with pancreatic cancer but cannot provide evidence of sustained benefit.
Greenlee et al. 2016100	Single-center RCT n = 63 women stage I-III breast cancer receiving adjuvant or new-adjuvant paclitaxel weekly for 12 cycles enrolled, 48 completed week 16 assessment 2 groups: 31 electroacupuncture (EA); 32 sham non-penetrative electroacupuncture (SEA) Treatment: once weekly for 30 minutes for 12 weeks Outcome measure: Neuropathic pain measured by Brief Pain Inventory-Short Form (BPI-SF) and quality of life domains by Functional Assessment of Cancer Therapy-Taxane neurotoxicity sub scale (FACT-NTX) at baseline, 6, 12 and 16 weeks; clinically meaningful change Miscellaneous: * Patient excluded if prior treatment of acupuncture within 12 mo * An change of ≥ 2 points in BPI-SF and ≥ 5 points in FACT-NTX score was considered clinically meaningful	At week 12, both groups reported increase in mean BPI-SF worst pain score but no statistical difference between groups (EA 2.6 vs SEA 2.8, p = 0.86) At week 16, BPI-SF worse pain score of SEA returned to baseline while EA continued to worsen (EA 3.4 vs SEA 1.7, p = 0.03) No difference between groups with respect to the FACT-NTX sub scale score at weeks 6, 12, and 16 weeks. No difference in taxane adherence reported One grade 1 adverse event reported; needle site exhibited minor swelling and bruising after needle removal	Unable to identify a protective effect or an improvement in pain with electroacupuncture on chemotherapy-induced peripheral neuropathy symptoms in women with breast cancer over the course of chemotherapy
Mao et al. 2013101	Single-center Phase II RCT n=67 women stage I-III breast cancer with self-reported arthralgia attributed to aromatase inhibitors 3 groups: 22 electroacupuncture; 22 sham non-penetrative electroacupuncture; 23 waitlist control (WLC) Treatment: 30 minute sessions twice a week for 2 weeks then once a week for 6 weeks for a total of 10 treatments over 8 weeks Outcome measure: Pain measured by BPI at baseline, week 8 and 12	Mean reduction in pain severity was greater in the EA group than in the WLC group at week 8 (2.2 vs 0.2, p = 0.0004) and at week 12 (2.4 vs 0.2, p< 0.0001). The SEA group also showed a statistically significant greater decrease in the BPI severity score compared to the WLC group at both Week 8 (p < 0.001) and Week 12 (p = 0.0036). EA and SEA showed no statistical difference in all study outcomes at Week 8. 18 adverse events reported by 8 subjects which were mild in severity and spontaneously resolved.	Electroacupuncture produces a clinically significant reduction in arthralgias related to aromatase inhibitors in patients with breast cancer, however, sham electroacupuncture elicits a similar effect.
Hershman et al. 2018102	Multi-center RCT; 11 sites n = 226 women stage I-III breast cancer on aromatase inhibitors; 206 completed trial 3 groups: 110 true acupuncture; 59 sham acupuncture; 57 waitlist control (WLC) Treatment: 30-45 minute sessions twice per week for 6 weeks followed by 1 session per week for 6 weeks; total of 18 sessions Outcome measure: Pain measured by BPI-WP at baseline, 6, 12, 16, 20, 24 and 52 weeks Miscellaneous: * Patient excluded if scored under 3 on the Brief Pain Inventory Worst Pain (BPI-WP) * A change of ≥2 points in BPI-SF was considered clinically meaningful	Combined mean baseline BPI-WP score = 6.6 After 6 weeks, the mean BPI-WP score in the true acupuncture group decreased by 2.05 points, in the sham acupuncture group by 1.07 points, and in the waitlist control group by 0.99 points. The adjusted difference for true acupuncture vs sham acupuncture was 0.92 points (95% CI, 0.20-1.65; P = .01) and for true acupuncture vs waitlist control was 0.96 points (95% CI, 0.24-1.67; P = .01). 2 episodes of grade 2 presyncope reported. Bruising at acupuncture sties was the most common adverse effect.	Acupuncture in women with breast cancer and aromatase inhibitor-related arthralgias, true acupuncture compared with sham acupuncture or with waitlist control resulted in a statistically significant reduction in joint pain, however, the change was of uncertain clinical importance.
Crew et al. 2010103	Single-center RCT n = 38 women with history of stage I-III breast cancer on aromatase inhibitors 2 groups: 20 true acupuncture; 18 sham acupuncture Treatment: 30 minute sessions twice a week for 6 weeks Outcome measure: Pain measured by BPI-SF. Joint pain, stiffness and function of the knees and hands measured by Western-Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and Modified Score for the Assessment of Chronic Rheumatoid Affections of the Hands (M-SACRAH) respectively. Patient’s surveyed at baseline, 3 and 6 weeks Miscellaneous: * Patient excluded if scored under 3 on the Brief Pain Inventory Worst Pain (BPI-WP)	Mean BPI-SF worse pain score at 6 weeks was lower in the true acupuncture group than in the sham acupuncture group (3.0 vs 5.5; P=0.001) The true acupuncture group had up to a 70% decrease in the WOMAC and M-SACRAH scores compared with baseline. No adverse effects were observed.	Women with aromatase inhibitor induced arthralgias who were treated with true acupuncture exhibited significant improvement with the pain. Objective functional assessments should in incorporated into the study in addition to the current patient reported outcomes.
Lam et al. 201798	Dual-center RCT n = 42 patients with moderate to severe cancer pain 3 groups: 14 si guan xue acupoint; 12 si guan zue acupoint; 14 common acupoints Treatment: 7 30 minute sessions Outcome measure: Pain measured by NRS from 0-10 measured daily and patient subjective improvement measured by the Patient Global Impression of Change (PGIC) scale measured at baseline and day 7	With a repeated measures design approach, there was no difference in pain score among the three treatment groups. Differences in PGIC scores across the three treatment arms indicate no statistically significant difference. There was a statistically significant improvement in the pain score in treatment using the si guan xue acupoints after the fifth day of treatment (p < 0.05). No adverse effects were observed.	The trial had a limited sample size did not blind or use a non-treatment control. Thus, the results should be considered preliminary.
Lu et al. 201297	Dual-center parallel pilot RCT n=21 women with ovarian cancer receiving chemotherapy 2 groups: 11 electroacupuncture; 10 sham acupuncture Treatment: 10 30 minute sessions 2-3 times per week beginning 1 week prior 2nd cycle of chemotherapy and ending prior patient’s 3rd cycle of chemotherapy Outcome measure: Quality of life measured by the European Organization for Research and Treatment of Cancer-Quality-of-Life Questionnaire-Core 30 Item (EORTC-QLQ-C30) and the Quality of Life Questionnaire–Ovarian Cancer Module-28 Item (QLQ-OV28) was measured at baseline and after the last acupuncture session. Miscellaneous: * A change of ≥10 points was considered clinically meaningful	Original recruitment target was not achieved. No statistically significance between the two treatment groups was found in any subscore. After adjusting for baseline EORTC-QLQ-C30 score differences, only the social function subscore was significantly higher in the electroacupuncture arm, compared with the sham acupuncture arm. And after adjusting for baseline QLQ-OV28 differences, only the score of peripheral neuropathy was still significantly higher in the electroacupuncture group, compared with the sham acupuncture group. No adverse effects were observed.	Electroacupuncture may hav a role in improving quality of life during chemotherapy but no firm conclusions can be drawn due to the limited sample size and high drop out rates.
Alimi et al. 2003104	Single-centered RCT n = 90 patients with chronic central or peripheral neuropathic pain after treatment of cancer 3 groups: 29 auricular acupuncture;30 sham auricular acupuncture; 31 sham non-penetrative auricular acupuncture Treatment: 2 sessions a month apart averaging of 44 minute each session. Acupuncture points were selected individually for each patient based on electrodermal response; same sites were used for the second treatment Outcome measure: Pain intensity by a VAS scaled from 0 to 100 mm measured at baseline and month 1 and 2. Miscellaneous: * Patients excluded if pain intensity <30 mm on VAS * The sham non-penetrative control group were visually identifiable from the two other groups	More than 2/3 of patients were women treated for breast cancer. A 36% statistically significant decrease (58 mm to 37 mm on VAS) was observed after two months in the acupuncture group compared to the limited change observed in the sham non-penetrative acupuncture group (58 mm to 55 mm on VAS) (p < 0.0001). The decrease in pain intensity correlated with the decrease in the average electrical potential difference at the auricular points in the acupuncture group (R2 > 0.76). No adverse effects were observed.	Auricular acupuncture in adult patients with neuropathic pain after treatment of cancer provides statistically significant pain relief when compared to placebo acupuncture. The study suggests that the reduction in pain may be associated with a decrease in average electrodermal signal on the ear.
Bao et al. 2013105	Dual-center RCT n = 47 women with stage 0-III breast cancer on aromatase inhibitor therapy with associated musculoskeletal symptoms 2 groups: 23 true acupuncture; 24 sham non-penetrating acupuncture Treatment: 8 20 minute weekly sessions Health outcome: Health Assessment Questionnaire Disability Index (HAQ-DI) and VAS measured at baseline and at 8 weeks. Miscellaneous: * Patients were required to have a baseline Health Assessment Questionnaire Disability Index (HAQ-DI) score ≥0.3 and/or pain using a 100 point VAS ≥20 * A change of ≥0.22 points in HAQ-DI was considered clinically meaningful	No statistically significant difference in HAQ-DI (p = 0.30) and VAS (p = 0.31) at week 8 was observed between true acupuncture and sham acupuncture with respect to aromatase inhibitor associated musculoskeletal symptoms (AIMSS). A statistically significant reduction of IL-17 (p = 0.009) was observed in both groups. No adverse effects were observed.	Statistically significant reduction of IL-17 offers preliminary evidence that both real and sham acupuncture may reduce the severity of AIMSS through modulation of IL-17.
Han et al. 2017106	Single-center RCT n = 104 patients with multiple myeloma who developed new onset peripheral neuropathy grade II-IV (as defined by NCI CTCAE neuropathy severity assessment) during chemotherapy 2 groups: 52 acupuncture + methylcobalamin; 52 methylcobalamin only Treatment: 3 cycles of 30 minute sessions daily for 3 days then every other day for 10 days Health outcome: Patient daily activity and degree of neuropathy evaluated by VAS pain score, Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity (Fact/GOG-Ntx) questionnaire and electromyographic (EMG) nerve conduction velocity (NCV) determinations measured at baseline and after completion of final treatment	The VAS pain score decreased more significantly in the treatment group than compared to the control group (p < 0.01) The patients’ daily activity measured by the Fact/GOG-Ntx questionnaire indicated significant improvement in the treatment (p < 0.001). NCV measured by EMG in the treatment group improved significantly. No adverse effects were observed.	Significant decrease in pain and improvement in patient daily activity when acupuncture is combined with methylcobalamin to address chemotherapy induced peripheral neuropathy during treatment of multiple myeloma.
Rostock et al. 2013107	Single-center RCT n = 60 patients after treatment of chemotherapy exhibiting symptoms of chemotherapy-induced peripheral neuropathy (CIPN) 4 groups: 14 electroacupuncture; 14 hydroelectric baths; 15 vitamin B1/B6; 17 placebo capsule Treatment: 8 ± 1 - 15 minute sessions over 3 weeks for electroacupuncture and hydroelectric baths; 3 capsules per day for 3 weeks for vitamin B1/B6 and placebo Health outcome: Neuropathic pain intensity measured by NRS from 0-10 and quality of life measured by EORTC QLQ-C30 at baseline, day 21 and 84.	CIPN related NRS score improved in electroacupuncture by 0.8 ± 1.2, in hydroelectric baths by 1.7 ± 1.7, in vitamin B1/B6 by 1.6 ± 2.0, and in placebo by 1.3 ± 1.3 points without significant difference between the groups or placebo. By day 21, health-related quality of life metrics improved moderately in all groups but no significant differences were identified by EORTC QLQ-C30 scores between groups.	Electroacupuncture, hydroelectric baths or vitamins B1/B6 was not shown to be superior to placebo in treatment of CIPN in cancer patients
Pfister et al. 2010108	Single-center RCT n = 58 patients with cancer with history of neck dissection associated moderate to severe pain and dysfunction for at least 3 months (Constant-Murley score ≤ 70) 2 groups: 28 acupuncture; 30 usual care Treatment: 4 30 minute acupuncture sessions weekly; usual care includes physical therapy, analgesia and/or anti-inflammatory drugs Health outcome: Pain and function measured by the Constant-Murley score from 0-100 and pain measured by NRS from 0-10 at baseline and after completion of last treatment Miscellaneous: * Randomization stratified by neck procedure type * Baseline NRS score was an average of two surveys 7 days apart within 1 week of starting treatment	Post hoc analysis adjusted for large baseline differences between groups with respect to medication use. Medication use defined by the medication quantification scale decreased in both groups. The differences between the groups were not statistically significant (P=0.4) Acupuncture was significantly superior compared to control for nearly all outcome measures. Patients receiving acupuncture had Constant-Murley scores 11.2 higher than the usual care group (95% CI, 3.0-19.3; p = 0.008) No adverse effects attributed to acupuncture were observed.	Findings suggest significant reductions in pain and dysfunction was observed in patients receiving acupuncture compared to usual care in patients with history of cancer and moderate to severe pain after neck dissection, however, further studies with larger sample size required.
Meng and Feng. 2018109	Single-center retrospective chart review n = 64 patients with stage II-IV cervical cancer 2 groups: 32 pain-specific acupoints; 32 regular acupoints Treatment: 30 minute sessions daily for 14 days Health outcome: Pain measured by NRS from 0-10, quality of life measured by EORTC QLQ-C30, and efficacy by Karnofsky Performance Status (KPS) from 0-100 measured at baseline and day 14	Patients in the pain-specific acupoint group exhibited a -4.2 point decrease in pain after day 14 compared to the -2.2 point decrease in the regular acupoint group and observed to be a statistically significant difference (p = 0.01). No significant difference observed between groups with respect to EORTC QLQ-C30 and KPS scores No adverse effects were observed.	The study suggests that patients with cervical cancer who receive acupuncture at pain specific acupoints exhibited a greater decrease in pain compared to those receiving acupuncture at regular acupoints. However, the findings may be limited by lack of randomization and selection bias.
Miller et al. 2019110	Single-center retrospective chart review n=68 adult patients with cancer or history of cancer referred to palliative care and received acupuncture 1 group: paired t-test Treatment: Acupuncture for median of 28 minute sessions (15 - 40 minute range) Outcome measure: Edmonton Symptom Assessment Scale (ESAS) measured at baseline and after last treatment Miscellaneous: * A change of ≥2 points in ESAS was considered clinically meaningful	Range of 1-13 acupuncture treatments; median = 2; 81% of patients had between 1 - 3 treatments Significant reductions in mean pain scores determined by ESAS were observed after the first treatment (1.9+1.8; p < 0.001) and across all treatments (1.7+1.9; p < 0.001). Multivariable analysis demonstrated clinically meaningful pain improvement in patient with stage III/IV cancer (OR: 3.23, 95% CI 1.11-9.40; p < .001) The percent pain reduction between those with no disease and active cancer showed no significant difference (p = 0.54) No adverse effects were observed.	Patients with advanced cancer and higher baseline pain scores receiving acupuncture are more likely to achieve clinically meaningful pain reduction. However, these findings may be limited by concomitant pain medications confounding the pain scores.