Efficacy and Safety of Acupuncture for Pain in Non-Small Cell Lung Cancer: A Meta-Analysis

Abstract

Background and Objective

Lung cancer remains the primary cause of cancer-related deaths globally, with non-small cell lung cancer (NSCLC) accounting for most cases. Although acupuncture has been used for cancer pain management, its efficacy remains controversial due to varying results from clinical studies. This study aims to evaluate the efficacy and safety of acupuncture for managing pain in patients with non-small cell lung cancer (NSCLC).

Methods

A literature search was conducted across eight major databases up to September 3, 2024. Two reviewers independently screened studies, extracted data, and assessed risk of bias following Cochrane guidelines. RevMan software was employed to calculate risk ratios and standardized mean differences (SMD) with 95% confidence intervals (CI), using fixed or random-effects models based on the heterogeneity observed. Subgroup analyses, sensitivity analyses, publication bias assessment (funnel plots and Egger’s test), evidence quality rating, and trial sequential analysis (TSA) were performed.

Results

Eleven randomized controlled trials incorporating 745 NSCLC patients were included in the meta-analysis. The primary outcomes assessed included pain intensity and adverse events. The meta-analysis revealed that acupuncture significantly outperformed medical treatments in reducing pain intensity (SMD=−1.33, 95% CI [−2.02, −0.63]) and incidence of adverse events (OR = 0.25, 95% CI [0.14, 0.43]). Compared to medical treatment, acupuncture showed benefits in improving total effective rate (OR=3.30, 95% CI [1.15, 9.46]) and reducing symptoms of anorexia (SMD=−0.40, 95% CI [−0.74, −0.06]), insomnia (SMD=−0.67, 95% CI [−1.03, −0.31]) and fatigue (SMD=−1.18, 95% CI [−2.04, −0.33]) compared to medical treatments. Sensitivity analyses validated result robustness. No significant publication bias was detected via funnel plots and Egger tests. TSA confirmed sufficient sample sizes for pain intensity, fatigue, and insomnia outcomes.

Conclusion

Acupuncture has demonstrated positive effects in alleviating cancer pain in patients with NSCLC, contributing to an improved quality of life while maintaining a certain level of safety.

Introduction

Globally, lung cancer tops the list of causes of cancer-related deaths among men and is the runner-up for women, surpassed only by breast cancer.¹ According to the World Health Organization (WHO), lung cancer accounted for 2.09 million new cases and 1.76 million fatalities globally.^1,2 Within this context, non-small cell lung cancer (NSCLC) represents the majority, making up 85 to 90% of lung cancer incidences.³ Throughout their illness and treatment, cancer survivors experience pain as the most challenging symptom to manage.^4,5 Pain markedly affects the quality of life (QOL) of both individuals undergoing cancer treatment and those who have completed it.⁶ Nearly 90% of individuals with cancer experience pain.⁷ Among long-term NSCLC survivors followed up in survivor clinics, 25% have been found to have chronic pain.⁸ Research indicates a genetic link to the intensity of pain experienced, with specific single nucleotide polymorphisms (SNPs) within the PTGS2 and LTA genes potentially forecasting pain levels and QOL among survivors of NSCLC.⁹ Furthermore, cisplatin-based adjuvant chemotherapy regimens are linked to persistent neuropathy and pain, adversely impacting the overall quality of life for patients who have undergone treatment, even when assessed 30 months post-treatment.¹⁰

Post-thoracotomy pain is a distinctive issue among survivors of NSCLC. Characterized by recurrent or persistent neuropathic pain along with surgical incision, which includes sensations of burning, paresthesia, and aching, this type of pain often persists for an extended period beyond two months post-surgery, even in the absence of cancer recurrence or metastasis.¹¹ It is likely associated with damage to the intercostal nerves during surgery, with reported incidence rates varying widely, from 11% to 80%, among NSCLC patients who have had surgery.^12–16 A significant proportion, around 30%, of these patients still report pain up to five years postoperatively, with the severity of early postoperative pain being a key indicator of chronic pain development. This persistent pain significantly impacts the quality of life for lung cancer survivors. Additionally, bone pain is a prevalent complication in advanced-stage lung cancer, with 80% of patients with lung tumors reporting bone pain as the disease progresses, severely impacting their quality of life.¹⁷

The WHO advocates a stepwise analgesic approach for managing cancer pain, recommending opioids as the primary treatment for moderate to severe pain associated with cancer.¹⁸ Commonly utilized opioids include oxycodone, morphine, hydromorphone, and fentanyl.¹⁹ However, these pharmacological interventions are associated with significant side effects, such as constipation, vomiting, urinary retention, delirium, and dizziness.^20,21

Non-pharmacological interventions for cancer pain encompass psychological education, cognitive-behavioral therapy, complementary and alternative medicine, and integrative non-pharmacological strategies.²⁰ Acupuncture, a traditional treatment with deep roots in China, is broadly recognized and accepted for its efficacy.²² The analgesic effects of acupuncture are increasingly understood through modern neurobiological research. Evidence suggests that acupuncture stimulates the release of endogenous opioids such as endorphins and enkephalins in the central nervous system, activates descending pain inhibitory pathways, and modulates neuroinflammatory responses via the hypothalamic-pituitary-adrenal axis. These mechanisms align with the gate control theory of pain, providing a plausible rationale for its application in cancer pain, which often involves both nociceptive and neuropathic components. Given the growing interest in integrative oncology and the limitations of opioid-centric pain management, an updated synthesis of evidence specific to NSCLC is warranted.

Since the initiation of acupuncture for cancer pain in 1985, numerous clinical trials have confirmed its efficacy in alleviating various types of pain, including that associated with cancer.^23,24 In a clinical study, lung cancer patients were administered a combination therapy of electroacupuncture (EA) and syndrome-based herbal treatment along with oxycodone controlled-release tablets, and it was observed that the pain relief experienced by the treatment group surpassed that of the monotherapy group.²⁵ In another study involving lung cancer patients with bone metastases-related pain, the group treated with a combination of traditional Chinese medicine and electromagnetic wave therapy targeted at specific acupoints, while the control group was administered morphine controlled-release tablets.²⁶ The findings revealed that the treatment group not only had a longer duration of pain relief but also did not exhibit drug resistance or addiction.

In recent years, a growing body of randomized controlled trials (RCTs) has focused on the efficacy of acupuncture in reducing pain related to NSCLC. Despite this, a meta-analysis addressing this specific topic has yet to be performed. Consequently, our study aims to undertake a systematic meta-analysis assessing the efficacy and safety of acupuncture in alleviating pain associated with NSCLC, while also evaluating the methodological rigor and reporting the quality of clinical trials focused on acupuncture for NSCLC pain. This study is the first to specifically focus on non-small cell lung cancer (NSCLC) as a distinct population. By narrowing the inclusion criteria, we effectively reduce the clinical heterogeneity caused by different types of cancer, making the evaluation of acupuncture’s interventional effects more targeted and accurate.

Methods

Literature Search

The study encompassed thorough research of eight major databases, including PubMed, EMbase, Cochrane, Web of science, China National Knowledge Infrastructure (CNKI), China Biology Medicine disc (CBM), Wanfang and Chongqing VIP (CQVIP), from the time of inception up to 2024/9/3 (Supplementary Table 1). No language restriction was set. According to title, abstract and full text, two researchers included RCTs which met the criteria. This article was registered in PROSPERO, with the ID CRD42024593147 (https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=593147).

Inclusion Criteria

Type of Participants

Adults with a diagnosis of NSCLC with pain were included, regardless of their current stage, surgical method, or past treatment. In this review, the background, gender and nationality of the participants were not considered.

Type Studies

This article especially included RCTs on the efficacy and safety of acupuncture in NSCLC patients with cancer pain. Blinding and publication dates were not concerned.

Type of Intervention Measures

Participants in the control groups of the studies under consideration should receive either sham acupuncture, conventional pharmaceutical therapy, or routine medical treatment. In contrast, the trial groups should be subjected to acupuncture as their intervention.

Outcome Type

The dominant outcome indicators reported in this article are pain intensity and adverse events, while the secondary outcome indicators were total effective rate, degree of insomnia, fatigue degree and anorexia score.

Exclusion Criteria

To ensure the uniqueness of the data and avoid duplication of labor, duplicate publications were screened out. Moreover, non-human studies and clinical studies were not considered. Besides, studies in which the experimental groups did not incorporate acupuncture as part of their treatment regimen were not considered for inclusion, whereas the control group could not include acupuncture. In the end, any studies lacking outcome measures were also not included in this meta-analysis. The adoption of exclusion criteria could ensure the reliability and quality of the study data and thus the credibility and overall quality of the meta-analysis results.

Screening for Eligibility and Data Pulling

Two researchers separately reviewed the retrieved literature against the established criteria. Following this, they compared their findings. Disagreements would be resolved through discussion, and if required, consultation with a third party would be sought. The data extracted encompassed the principal investigator’s name, the date of publication, the age of the participants, the intervention strategies utilized and the resulting study findings.

Evaluation of Bias Risk in Quality

Two authors (JM Liu and JM Chang) independently assessed the risk of bias using the Cochrane Manual on Systematic Review of Interventions (version 5.1.0). Each area of risk is categorized into three levels: “high risk” “low risk” and “unclear”. The third author (YL Ye) was tasked with resolving any disagreements between two reviewers.

Statistical Analysis

RevMan (version 5.4)²⁷ software was utilized for statistical analysis. Risk ratio (RR) were used for dichotomous outcomes, Standardized mean difference (SMD) were used for Continuous variables. The 95% confidence interval (CI) for the corresponding data was determined. According to the Cochrane Handbook for Systematic Reviews of Interventions (Version 5.1.0), P < 0.05 indicated substantial differences. In every analysis, the I² statistic was employed to assess the degree of statistical heterogeneity across the trials. I² < 50% indicated remarkable homogeneity, so the fixed effects model was conducted, while I² ≥ 50% used the random effects model.

Subgroup Analysis

Subgroup analyses were conducted relying on two criteria: (1) the type of acupuncture intervention, specifically EA versus manual acupuncture (MA); (2) the duration of treatment, distinguishing between less than 14 days and 14 days or more. Given the limited number of studies included, only primary outcomes were utilized for analysis.

Sensitivity Analysis

To ensure the reliability of the heterogeneity results, sensitivity analysis was carried out by sequentially removing individual studies.

Assessment of Publication Bias

The assessment of publication bias comprised a visual representation in the form of a funnel plot and an objective Egger test. The former served to identify any potential publication bias, while the latter offered a more rigorous approach to analysis. When the P>|t| value exceeded 0.05, it signified the absence of significant publication bias. In contrast, the trim-and-fill test was used to determine if any bias was introduced into the results, particularly with regard to the number of articles selected for trimming and the impact of the trimming procedure on the P-value.

Evidence Quality Assessment on GRADE

Utilizing the GRADEpro software tool,²⁸ the quality of evidence was assessed following the structured Grade of Recommendations Assessment, Development and Evaluation (GRADE) guidelines. The studies were categorized into four levels: high, moderate, low, or very low. Evaluation criteria included assessing the risk of bias, consistency, directness of evidence, precision, and possibility of publication bias.

Trial Sequential Analysis (TSA) Analysis

TSA 0.9.5.10 beta https://ctu.dk/tsa/ was utilized for analysis. RIS estimation relies on specific statistical measures: type I error probability (α=0.05) and type II error probability (β=0.2). For dichotomous data, relative risk reduction (RRR) is set at 35%, with a control group event rate (Pc) of 3%, based on meta-analysis findings. In the case of continuous data, we calculated the mean difference and variance from software-generated empirical assumptions. Fleming boundary is commonly used as the significance threshold for cumulative data in many randomized controlled trials. The Z-value is determined by dividing the combined intervention effect by its standard error. Certain markers were omitted because of inadequate research volume, such as efficiency, adverse event rate, anorexia score.

Results

Fundamental Details of Literature

We retrieved 680 studies across various databases. Following a stringent screening process, 11 studies^29–39 were selected for the final analysis, with 1 in English and 10 in Chinese, as detailed in Figure 1.

Figure 1.

Literature Screening Flowchart.

Studies Overview

A synthesis of the literature was conducted, encompassing 745 patients diagnosed with NSCLC pain. The experimental group consisted of 374 patients, while 371 were treated within the control group. Details are displayed in Table 1, the control group received either sham acupuncture or standard pharmacological treatments. Conversely, those in the trial group underwent treatments with acupuncture.

Table 1.

Basic Information of Included Literature

Included Studies	Sample/Case	Age/Year		Intervention		Diagnosis	Course of Treatment	Outcome
	T/C	T	C	T	C
Huang L F 202132	30/30	NA	NA	Fu’s Acupuncture with Continuous Thoracic Paravertebral Block	Ropivacaine Hydrochloride Injection/100mL/qd	Pathological Diagnosis of Non-small cell lung cancer	3 days	①③
Lai Z 201936	32/32	62.44±8.77	64.94±10.21	Warming acupuncture with Burnet root leukopoietic tablets /0.2g/tid	Burnet root leukopoietic tablets /0.2g/tid	<Guidelines for Diagnosis and Treatment of non-small cell lung cancer>	10 days	①②⑤⑥
Miao B 202033	76/75	55±7	56±7	Conventional Chemotherapy	Conventional Chemotherapy	<Chinese Classification and Diagnostic Criteria for Mental Disorders>	56days	①③④
				Acupuncture with Traditional Chinese Medicine Five Elements Music Therapy	Ganmai Dazao Decoction with Traditional Chinese Medicine Five Elements Music Therapy
Randolph H L 200639	13/12	64.6±8.0	64.5±8.5	Electroacupuncture with Morphine	Sham Acupuncture with Morphine	Operable non–small cell lung cancer with primary tumor equal or larger than 4 cm in diameter precluding a video assisted thoracic surgery approach	7 days	①
Tian W P 202034	24/23	60.54±9.24	62.13±7.59	Conventional Chemotherapy with Acupuncture	Conventional Chemotherapy	<Primary Lung Cancer Diagnosis and Treatment Guidelines (2018) >	42 days	①②④⑤⑥
Wang J C 202429	48/48	60.0±2.9	64.1±3.0	Electroacupuncture with Morphine	Morphine	<Modern Oncology>	14 days	①②③⑤
Wang Y 201637	30/30	64.8±9.4	62.1±14.7	Electroacupuncture with Oxycodone extended-release tabletst/40mg/bid	Oxycodone extended-release tabletst/40mg/bid	Pathological /Cytohistologic Diagnosis of Non-small cell lung cancer	14days	①②③④⑤⑥
Wen X H 201935	25/25	63.96±7.09	61.52±6.79	Sanjiao acupuncture	Conventional Western Medicine Therapy	International Classification of Diseases	14 days	①②④⑤⑥
Yao J L 202231	28/28	18~80	18~80	Knife Acupuncture with Functional Training	Functional Training	<Chinese Guidelines for Diagnosis and Treatment of Primary lung Cancer (2015)>	28 days	①④
Yuan L 202330	35/35	45~75	45~75	Conventional Western Medicine Therapy with Acupuncture-embedding	Conventional Western Medicine Therapy	<Chinese Medical Association Lung Cancer Clinical Guidelines (2018)> and <Cancer pain diagnosis and treatment code (2018)>	21 days	①②③
Zhu S H 201038	30/30	31~69	31~69	Conventional Chemotherapy with Acupuncture	Conventional Chemotherapy	<New Chinese Common Malignant Tumor Diagnosis and Treatment Standards>	21 days	①④⑤⑥

Notes Outcome: ① Pain intensity. ② Adverse events. ③Total effective rate. ④ Degree of insomnia. ⑤ Fatigue degree. ⑥ Anorexia score.

Risk of Bias Assessments

Figure 2 illustrates the percentage of risk assessment outcomes across each domain. Among the 11 studies, a high risk of bias was noted in the execution and detection domains, occurring in the same publication,³¹ as it explicitly stated that it was a non-blind study. Additionally, eight other publications^{29,30,32–34,36–38} did not describe these two domains, leaving their risk levels unclear. The remaining two studies^35,39 utilized blinding methods and were consequently classified as being at “low risk.” Regarding the generation of random sequences, aside from one publication³² that did not explicitly mention it, the rest^{29–31,33–39} employed a random number, indicating low risk. Except for three studies^30,32,37 that did not mention it, the remaining eight studies^{29,31,33–36,38,39} had allocation concealment rated as “low risk.” For follow-up bias, one publication³⁸ reported two cases of withdrawal, and the authors did not disclose the reasons for it. All other publications^29–37,39 had no missing data or clearly reported the number of dropouts and withdrawals, along with the reasons and handling methods, and were considered to be at low risk. Insufficient information was available in one article³⁹ to determine the risk level for reporting bias; the rest of the studies^29–38 reported all pre-specified outcomes. For more detailed judgments, refer to Supplementary Table 2.

Figure 2.

Quality Assessment Chart of Studies. (A) Risk of bias graph. (B) Risk of bias summary. Green represents low risk of bias, yellow represents unclear risk of bias, and red represents high risk of bias.

Meta-Analysis

Primary Outcomes

Pain Intensity

Eleven studies utilized pain scores (VAS score, NRS score, degree of pain) to make the assessment of acupuncture’s efficacy in managing pain associated with NSCLC patients.^29–39 The results of meta-analysis indicated that the acupuncture group experienced better pain relief compared to the MT group (SMD=−1.33, 95% CI [−2.02, −0.63], P=0.0002; Figure 3A).

Figure 3.

Meta-analysis Forest Plots: (A) Pain Intensity Outcomes. (B) Adverse Events Outcomes.

Adverse Events

Seven studies reported adverse events, and the analysis concluded that the disparities between the acupuncture and MT groups were statistically significant.^{29,30,32,34–37} The incidence of adverse events was lower in the acupuncture group (OR=0.25, 95% CI [0.14, 0.43], P<0.00001; Figure 3B).

Secondary Outcomes

Total Effective Rate

Five studies compared the overall response rate of acupuncture versus MT group in treating pain associated with non-small cell lung cancer.^{29,30,32,33,37} The results indicated that the acupuncture group demonstrated greater overall effectiveness than the MT group (OR=3.30, 95% CI [1.15, 9.46], P=0.03; Supplementary Figure 1A).

Degree of Insomnia

Degree of insomnia contained 6 studies.^{31,33–35,37,38} The combined results showed that the acupuncture group had better improvement in insomnia severity as opposed to the treatment group (SMD=−0.67, 95% CI [−1.03, −0.31], P=0.0002; Supplementary Figure 1B).

Fatigue Degree

Six studies reported fatigue degree.^29,34–38 The comprehensive analysis of the findings indicated that the disparity between acupuncture and MT groups in alleviating fatigue was statistically significant (SMD=−1.18, 95% CI [−2.04, −0.33], P=0.007; Supplementary Figure 1C).

Anorexia Score

Five studies reported anorexia scores,^34–38 and the overall results showed that the acupuncture group was better than the MT group in enhancing appetite among patients with non-small cell lung cancer pain (SMD=−0.40, 95% CI [−0.74, −0.06], P=0.02; Supplementary Figure 1D).

Subgroup Analysis

Pain Intensity

For the kind of acupuncture, MA group (SMD=−1.25, 95% CI [−1.96, −0.54], P<0.05) was better than MT group in decreasing pain intensity. However, EA group had no statistically significant difference with MT group, as indicated by an SMD of −1.45 and a 95% CI of [−3.48, 0.58] (P=0.16, Figure 4A). As for the treatment duration, both the groups treated for ≤14 days (SMD=−1.77, 95% CI [−3.13, −0.41], P<0.05) and those treated for more than 14 days (SMD=−0.85, 95% CI [−1.42, −0.28], P<0.05) were more advantageous than MT group in decreasing pain intensity (Figure 4B).

Figure 4.

Subgroup of pain intensity. (A) The type of acupuncture. (B) The treatment duration.

Adverse Events

For the kind of acupuncture, EA group (OR=0.21, 95% CI [0.10, 0.44], P<0.05) exceeded MT group in lowering the frequency of adverse events. However, MA group (OR=0.31, 95% CI 0.31 [0.05, 2.03], P=0.22) had no statistically significant difference with MT group (Supplementary Figure 2A). As for the treatment duration, both the groups treated for ≤14 days (OR=0.25, 95% CI [0.14, 0.45], P<0.05) and those treated for more than 14 days (OR=0.23, 95% CI [0.06, 0.94], P<0.05) were better than MT group (Supplementary Figure 2B).

Sensitivity Analysis

For six different outcomes, heterogeneity was assessed. To assess the robustness of results, a sensitivity analysis was conducted by excluding included references one by one to determine if any individual study significantly influenced the results. Both the outcomes pain intensity (I²=94%) and Fatigue Degree (I²=93%) exhibited high heterogeneity. Sensitivity analysis did not significantly alter the effect value or heterogeneity, indicating good stability of results. The high heterogeneity observed for pain intensity and fatigue degree may reflect clinical diversity among the included studies, including variations in acupuncture protocols and differences in control group interventions. In the heterogeneity test for Adverse Events, it was found to be within a reasonable range (P=0.22, I²=29%). Surprisingly, the heterogeneity decreased to 0% upon the exclusion of a specific study.³⁵ The outcome Degree of Insomnia displayed very high heterogeneity (I²=67%). Upon exclusion of a specific study³⁴ in the sensitivity analysis, the I² value decreased to 27%, showing low heterogeneity. In the test for Total Effective Rate, it showed substantial heterogeneity (P = 0.008, I² = 71%). The I² value decreased to 48% when a study was excluded.³³ In the heterogeneity test for Adverse Events, it was found to be within an acceptable range (P = 0.16, I² = 41%). The heterogeneity decreased to 0% when a study was excluded.³⁵ The results of sensitivity analysis are shown in Supplementary Table 3.

Publication Bias

The funnel plot of the total effective rate exhibited a symmetrical distribution, while Egger test revealed no signs of publication bias (P=0.205, Supplementary Figure 3A). In contrast, the funnel plots for degree of insomnia, fatigue degree, anorexia score and adverse events were asymmetrically distributed. However, the Egger test still showed no potential publish bias (P=0.121, P=0.629, P=0.357, P=0.176, Supplementary Figure 3B–E). A combination of the funnel plots and the Egger test revealed that the pain intensity exhibited potential publication bias (P=0.012, Supplementary Figure 3F). The process of Egger test is shown in Supplementary Table 4. However, examination using the trim-and-fill test indicated that this bias had no impact on the estimation process, that is, as the data set remained unchanged, there was no need for trimming (Supplementary Table 5).

Assessment of Evidence Reliability

The results of the GRADE assessments are presented in Supplementary Table 6, indicating that the pain intensity, degree of insomnia, fatigue degree was assessed as “moderate quality.” In contrast, the evaluations for total effective rate, anorexia score, and adverse events were categorized as “low quality.”

TSA Analysis

Regarding pain intensity and fatigue degree, the Z-curve not only crosses the conventional statistical significance level, but also exceeds the RIS threshold, which means that the accumulated information volume has exceeded the expected data volume, and the conclusion is positive (Figure 5A and B). As for degree of insomnia, results indicate that the Z-curve surpasses both the conventional statistical significance level and the TSA threshold (Figure 5C). This suggests that, despite the cumulative information not meeting the anticipated level, additional experiments are unnecessary to reach a positive conclusion at this stage.

Figure 5.

Trial sequential analysis (TSA). (A) TSA of pain intensity. (B) TSA of fatigue degree. (C) TSA of degree of insomnia.

Discussion

The findings of our study indicate that acupuncture represents an efficacious and safe intervention for patients with non-small cell lung cancer (NSCLC) who are experiencing pain related to their condition. The results indicate that acupuncture not only significantly reduces pain sensations but also lowers the rate of adverse events. Moreover, acupuncture plays a crucial role in controlling tumor progression, enhancing appetite, ameliorating insomnia, and decreasing the severity of fatigue. Furthermore, our subgroup analysis reveals that electroacupuncture outperforms manual acupuncture in reducing pain sensations, and shorter treatment durations correlates with greater pain relief. The results of publication bias indicate that although potential publication bias was identified for pain intensity, it did not impact the overall estimations, suggesting the findings remain robust despite the asymmetry. Our evidence quality analysis reveals that the pain intensity was supported by moderate-quality evidence, suggesting a reasonable level of confidence in these findings. The TSA analysis indicates that the findings for pain scores and fatigue are robust and significant, while the results for insomnia scores suggest that further experimentation is unnecessary to confirm a positive conclusion at this stage.

Research has indicated that the evidence supporting the use of opioids for cancer pain is both limited in quantity and quality.⁴⁰ In clinical practice, most cancer patients achieve adequate pain relief with opioid therapy. However, responses to opioids vary. Studies indicate between 10% and 15% of cancer pain patients are non-responders.⁴¹ Moreover, as time progresses, patients may require escalating doses of analgesics to combat pain due to the development of drug tolerance, which can exacerbate the financial burden on patients due to the high cost of medication.²⁰ The mechanisms by which acupuncture exerts its effects are intricate and involve interactions across various physiological systems.⁴² Evidence from recent functional MRI (fMRI) investigations suggests that the stimulation of specific acupoints can lead to changes in the levels of neurotransmitters and the modulation of pain regulatory mechanisms throughout the central nervous system.⁴³ Additionally, it had been observed that the brain regions that participate in the acupuncture response are part of an extensive network that encompasses areas responsible for emotional, cognitive, and sensory functions. These included the somatosensory cortex, the limbic system, the basal ganglia, the brainstem, and the cerebellum.⁴⁴ The findings of a prospective observational study indicated that acupuncture is an effective method for alleviating symptoms experienced by patients with lung cancer, particularly pain and overall well-being.⁴⁵ As a safe and minimally invasive intervention, acupuncture shows promise as a beneficial option for those receiving cancer treatment. A systematic review and meta-analysis studied the association of acupuncture with reduction in cancer pain, which concluded that acupuncture is significantly linked to alleviating cancer pain and reducing the reliance on analgesics, albeit with moderate evidence quality.⁴⁶ This geographic concentration is consistent with previous meta-analyses, including the comprehensive review by He (2020), which also identified a predominance of Chinese studies in acupuncture research for cancer pain.⁴⁶ While our findings align with He in demonstrating beneficial effects of acupuncture, our study extends the literature by focusing specifically on NSCLC—a population with distinct pain etiologies—and by including recent RCTs published up to 2024.⁴⁶ A network comparative effectiveness meta-analysis has been conducted to evaluate the efficacy of acupuncture in conjunction with chemotherapy for the therapeutic management of non-small cell lung cancer.⁴⁷ The results demonstrate that acupuncture in combination with chemotherapy treatment is superior to chemotherapy in terms of efficacy, improved quality of life and safety. These findings are consistent with our own and serve to demonstrate once again that acupuncture represents an effective and safe intervention. Nevertheless, the scarcity of experimental data that directly contrasts the effectiveness and safety profiles of diverse acupuncture modalities for pain management in patients with non-small cell lung cancer remains a notable gap in the literature. Our study is the first to directly compare the effects of electroacupuncture and manual acupuncture on pain perception. Moreover, additional evidence suggests that acupuncture is an efficacious method of pain relief, which improves overall health-related quality of life, including appetite, insomnia, and fatigue levels.

Several limitations should be considered when interpreting our findings. Firstly, among the included studies, only one employed sham acupuncture control, while the majority compared acupuncture with active medical treatments without blinding of participants or personnel. This lack of blinding, particularly in studies using patient-reported outcomes such as pain scores, may introduce performance bias and detection bias, potentially influencing the effect estimates. Therefore, although our findings suggest that acupuncture may offer clinical benefits when added to standard care, these results should be interpreted with caution, and the specific efficacy of acupuncture beyond placebo remains to be confirmed in rigorously designed sham-controlled trials. Additionally, heterogeneity in treatment protocols, outcome measures, and follow-up durations across the included studies may limit the robustness of our pooled estimates. The relatively small sample sizes of some included trials further constrain the generalizability of our findings.

Based on the findings of our study, which provide evidence supporting acupuncture as an effective and safe adjunctive or alternative approach to pharmacologic pain management in NSCLC patients, several avenues for future research emerge. Firstly, exploring the effects of electroacupuncture might enhance understanding of their efficacy in pain management. Longitudinal studies are also fundamental for the assessment of the long-term effects of acupuncture on pain control and patient survival. Additionally, investigating the dose–response relationship between treatment frequency and duration with pain relief outcomes could help establish optimized protocols. Expanding research to include other symptoms and side effects of cancer treatment, like nausea and vomiting, would provide a broader perspective on acupuncture’s role in supportive care. Finally, the therapeutic effect of acupuncture on patients with cachexia could not be ignored. Future research could design RCTS to explore this question and yield groundbreaking results.

Conclusion

This meta-analysis suggests that acupuncture may offer beneficial effects for pain management in patients with NSCLC, including reduced pain intensity, fewer adverse events, and improvements in fatigue, insomnia, and anorexia. It is important to note that while most included studies compared acupuncture to active medical treatments, only one study employed a sham acupuncture control, reflecting the current paucity of sham-controlled trials in this specific population. Therefore, our findings primarily inform the comparative effectiveness of acupuncture versus standard treatments, rather than its specific efficacy beyond placebo. Future research should prioritize well-designed sham-controlled trials to further elucidate the specific therapeutic mechanisms of acupuncture in NSCLC pain management.

Funding Statement

This article was supported by the National Key Research and Development Program of China (2023YFC3503303), Guangzhou Science and Technology Planning Project (2025A03J4133), the Major Innovation Technology Construction Project of Synergistic Chinese Medicine and Western Medicine of Guangzhou (Grant number 2023-2318) and Guangzhou Municipal Science and Technology Bureau, China (grant number 2025A03J2557).

Data Sharing Statement

All data included in this study are available upon request by contact with the corresponding author.

Author Contributions

All authors made a significant contribution to the work reported, whether in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas. All authors have agreed on the journal to which the article has been submitted and agreed to be accountable for all aspects of the work.

Disclosure

The authors declare no conflicts of interest in this work.