Objective To systematically review the efficacy and safety of acupuncture for the treatment of tumor-related cognitive dysfunction. Methods The PubMed, The Cochrane Library, EMbase, CNKI, WanFang Data, VIP and CBM databases were electronically searched to collect studies on acupuncture for the treatment of tumor-related cognitive dysfunction from the establishment of the database to February 13th, 2022. Two reviewers independently screened the literature, extracted data, and assessed the risk of bias of the included studies. Meta-analysis was then performed using RevMan 5.4.1 software. Results A total of 16 studies involving 1 361 patients were included. The results of meta-analysis showed that the mini-mental state examination (MD=1.82, 95%CI 1.49 to 2.15, P<0.000 01) and Montreal cognitive assessment (MD=1.56, 95%CI 0.83 to 2.29, P<0.0001) scores of the acupuncture treatment group were superior to those in the control group. Furthermore, the acupuncture treatment group showed a reduced incidence of postoperative cognitive dysfunction (RR=0.50, 95%CI 0.39 to 0.63, P<0.000 01) and decreased levels of interleukin-6 (MD=−10.43, 95%CI −14.91 to −5.95, P<0.000 01), interleukin-1β (MD=−47.14, 95%CI −63.92 to −30.36, P<0.000 01), and tumor necrosis factor-α (MD=−9.13, 95%CI −12.38 to −5.89, P<0.000 01). In contrast, the visual analog scale score of the acupuncture treatment group (MD=−1.26, 95%CI −2.06 to −0.47, P=0.002) was better than that of the control group. No significant difference was found in the level of central nervous system-specific protein (S100β) (MD=−0.06, 95%CI −0.13 to 0.01, P=0.12) between the two groups. Conclusion Acupuncture therapy can improve tumor-related cognitive function in patients. Its curative effect is better than that of non-acupuncture therapy; however, its ability to reduce S100β levels is not significantly different from that of non-acupuncture therapy. Due to the limited quantity and quality of the included studies, more high-quality studies are needed to verify the above conclusion.