Rehmannia glutinosa for perimenopausal syndrome in patients with breast cancer: A systematic review and meta-analysis.

This meta-analysis is reported according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. A search was carried out across the PubMed, Cochrane Library, Embase, China National Knowledge Infrastructure, Wanfang, China Science and Technology Journal, and China Biology Medicine databases, from their inception up to June 2024. The search terms used included Rehmannia glutinosa, Dihuang, menopause, hot flashes, breast, and cancer. The detailed search strategy is provided in Appendix 1 . Searches were restricted to full-text articles published in English or Chinese. Clinical trial registers, databases such as the Chinese clinical trial registry and clinical trials were explored for ongoing or unpublished research studies. Inclusion criteria included: I) study design: randomized controlled trials (RCTs); II) study subjects: individuals with a pathological diagnosis of breast cancer and those diagnosed with PMS by a clinician or through any established diagnostic criteria; III) intervention: treatment for PMS with CHM containing Rehmannia glutinosa as the monarch drug or CHM containing Rehmannia glutinosa as the monarch drug combined with Western medicine, and controls that received Western medicine. Chinese herbal medicine containing Rehmannia glutinosa as the monarch drug was characterized as prescriptions containing various herbs, or Chinese patent medicines, in which Rehmannia glutinosa had the largest weight ratio or ranked first among the CHM prescriptions, with no restrictions on dose, preparation, or mode of administration. Western medicine included drugs commonly used to treat PMS in clinical practice, such as oryzanol, remifemin, and vitamin B1, with no restrictions on treatment course (treatment course had to be the same in the intervention and control group); and IV) outcomes: the primary outcome was change in the Kupperman index at the end of follow-up. Secondary outcomes included changes in Karnofsky performance status (KPS), the functional assessment of cancer therapy - breast (FACT-B), percentages of T-cells (CD4, CD8, and CD4/CD8), levels of gonadal hormones (estradiol [E 2 ], luteinizing hormone [LH], and follicle stimulating hormone [FSH]), and tumor markers (CEA, CA125, and CA153) in peripheral blood at the end of follow-up. Safety outcomes were findings on routine blood tests, liver and kidney function, and the occurrence rate of adverse events or side effects. Exclusion criteria included: I) trials that did not explicitly recognize Rehmannia glutinosa as the monarch drug; or 2) trials with incomplete data or duplicates. References were managed using Endnote X9 (Clarivate Analytics). Two authors (DK and FH) independently screened titles and abstracts, assessed full text articles to identify trials that met the inclusion criteria, and extracted data, including general information (title, first author, and publication date); trial characteristics (sample size and sources of bias); subject characteristics (age, race, gender, and duration of disease); intervention details (medication, dosage, formulation, treatment duration, prescription, and follow-up); and outcomes (efficacy and safety). Discrepancies were resolved through discussion with a third author (LM). Two authors (JX and DK) independently evaluated the methodological quality of the included trials using the Cochrane risk of bias-2 (RoB-2) tool. 24 Evaluation criteria encompassed bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in measurement of the outcomes, and bias in selection of the reported results. Risk of bias for each category was classified as high, low, or some concerns. Findings from the risk of bias assessment were presented using a risk of bias chart. Discrepancies were resolved through discussion with a third author (ZS). Statistical analysis was carried out using Review Manager 5.4 and Stata SE 17. The effect measure was mean difference (MD) with corresponding 95% confidence interval (CI). To ensure all measurements were expressed using a single unit, E 2 levels were converted from pg/mL to pmol/L using the formula: pg/ml*3.67=pmol/l. Heterogeneity between trials was investigated with the χ test and the I 2 index. A random-effects model was used in this meta-analysis because there was evidence of high heterogeneity between studies. Heterogeneity sources were identified through subgroup analysis, in which results for the primary outcome were stratified according to preparation of Rehmannia glutinosa (crude Rehmannia glutinosa, Shengdihuang, [CRG] or processed Rehmannia glutinous, Shudihuang, [PRG]), menopausal status, use of endocrine therapy, or treatment duration. Sensitivity analysis was carried out by excluding one study at a time to evaluate the robustness of the results. Publication bias was assessed for outcomes reported by ≥10 trials. A p -value of <0.05 was considered significant.

The initial search yielded 1,883 articles, and screening resulted in 1,787 articles. After exclusion of 659 duplicates and 1128 articles that did not meet the inclusion criteria, 96 full text articles were examined. After exclusion of 79 articles because of study design, incomplete data or duplicate data, 17 RCTs were included in the meta-analysis ( Figure 1 ). - Inclusion screening flow chart. The included RCTs are presented in Table 1 . All RCTs were published in Chinese and carried out between 2008-2024. The sample size of the RCTs varied between 42-132 patients that were aged 20-80 years. Treatment ranged from 2 weeks to 3 months. - Characteristics of 17 articles. Xiao et al 16 reported the outcome indicators at 4 weeks and 8 weeks, and finally included the outcome indicators at 8 weeks for meta-analysis. TAM: tamoxifen, PRG: processed Rehmannia glutinosa (Shudihuang), CRG: crude Rehmannia glutinosa (Shengdihuang), CHM: Chinese herbal medicine, KPS: Karnofsky performance status, FACT-B: functional assessment of cancer therapy - breast, E 2 : estradiol, LH: luteinizing hormone, FSH: follicle stimulating hormone There were some concerns on the randomization process in 12 trials, deviations from the intended interventions in 15 trials and missing outcome data in 4 trials. All RCTs exhibited a low risk of bias regarding measurement of the outcomes and selective reporting ( Figure 2 ). - Risk of bias summary. The Kupperman index was reported in 14 RCTs (n=951). 5 - 18 The meta-analysis revealed a notable enhancement in the Kupperman index in patients with breast cancer and PMS treated with CHM with Rehmannia glutinosa as the monarch drug compared to Western medicine (MD= -12.79, 95% CI: [-15.76 to -9.82]; p <0.01; I 2 =91.7%, Figure 3 ). There was evidence of high heterogeneity between studies, but the treatment effect was maintained in subgroup analyses ( Table 2 ), and the sensitivity analysis showed that the heterogeneity could not be attributed to any one RCT ( Figure 4 ). - Forest plot of the influence of Chinese herbal medicine with Rehmannia glutinosa as the monarch drug in the Kupperman index. - Forest plot of sensitivity analysis. - Subgroup analysis of Kupperman index between the Chinese herbal medicines group and the control group. PRG: processed Rehmannia glutinosa (Shudihuang), CRG: crude Rehmannia glutinosa (Shengdihuang) Karnofsky performance status was reported in 6 RCTs (n=358). 5 , 12 , 14 , 16 , 18 The meta-analysis demonstrated a significant improvement in KPS in patients with breast cancer and PMS treated with CHM with Rehmannia glutinosa as the monarch drug compared to patients treated with Western medicine (MD=7.61; 95% CI: [3.79-11.44]; p <0.01; I 2 =80.9%, Table 3 ). - Meta-analysis of Karnofsky performance status, functional assessment of cancer therapy - breast, T-cell subset, gonadal hormone concentrations, and tumor markers. PWB: physical well-being, SWB: social/family well-being, EWB: emotional well-being, FWB: functional well-being, AC: additional concerns, E2: estradiol, LH: luteinizing hormone, FSH: follicle stimulating hormone The FACT-B was reported in 3 RCTs (n=244). 6 , 12 , 15 The meta-analysis demonstrated no significant improvements in the physical well-being (MD=6.11; p =0.07), social/family well-being (MD=3.00; p =0.11), emotional well-being (MD=3.63; p =0.16), functional well-being (MD=3.80; p =0.05), and additional concerns (MD=4.25; p =0.19) domains of the FACT-B in patients with breast cancer and PMS treated with CHM with Rehmannia glutinosa as the monarch drug compared to Western medicine. ( Table 3 ). T-cell subset percentages in peripheral blood were reported in 3 RCTs (n=241). 13 , 15 , 16 The meta-analysis demonstrated an increase in the percentage of CD4 cells (MD=8.84; p <0.01), a decrease in the percentage of CD8 cells (MD= -2.15; p =0.02), and an improvement in the CD4/CD8 ratio (MD=0.37; p =0.03) in patients with breast cancer and PMS treated with CHM with Rehmannia glutinosa as the monarch drug compared to Western medicine ( Table 3 ). The E 2 levels in peripheral blood were reported in 12 RCTs (n=703). 6 , 7 , 9 - 11 , 13 , 14 , 16 - 21 The LH levels in peripheral blood were reported in 5 RCTs (n=324). 6 , 9 , 17 , 19 , 20 The FSH levels in peripheral blood were reported in 10 RCTs (n=612). 6 , 9 , 11 , 13 , 14 , 16 , 17 , 19 - 21 The meta-analysis demonstrated no significant difference in E 2 levels (MD=1.14; p =0.56), LH levels (MD= -0.90; p =0.11), or FSH levels (MD=0.58; p =0.25) in patients with breast cancer and PMS treated with CHM with Rehmannia glutinosa as the monarch drug compared to Western medicine ( Table 3 ). The CA-125 levels in peripheral blood were reported in 4 RCTs (n=239). 12 , 16 , 18 , 20 The CA-153 levels in peripheral blood were reported in 5 RCTs (n=289). 12 , 16 - 18 , 20 The CEA levels in peripheral blood were reported in 3 RCTs (n=161). 12 , 16 , 18 The meta-analysis demonstrated no significant difference in CA-125 levels (MD= -0.16; p =0.68), CA-153 levels (MD=0.22; p =0.77), or CEA levels (MD= -0.28; p =0.33) in patients with breast cancer and PMS treated with CHM with Rehmannia glutinosa as the monarch drug compared to Western medicine ( Table 3 ). A total of 13 RCTs described that adverse reactions (namely, gastrointestinal reactions, liver dysfunction, and so on) were not observed. 5 - 7 , 10 - 18 , 21 A total of 4 RCTs did not describe adverse reactions. 8 , 9 , 19 , 20 Publication bias was assessed for the 14 RCTs reporting the Kupperman index. The funnel plot and Egger regression method ( p =0.77) implied there was no publication bias ( Figure 5 ). - Publication bias funnel plot.

This systematic literature review and meta-analysis quantitatively evaluated the effectiveness of CHM with Rehmannia glutinosa as the monarch drug for the treatment of PMS in patients with breast cancer. The meta-analysis included 17 RCTs involving 1,139 patients with breast cancer and PMS. Findings showed that CHM with Rehmannia glutinosa as the monarch drug significantly improved the severity of PMS symptoms (Kupperman index), functional performance (KPS score), and quality of life (FACT-B score) in individuals with breast cancer and PMS compared to Western medicine. The CHM with Rehmannia glutinosa as the monarch drug also increased the percentage of CD4 cells, reduced the percentage of CD8 cells, and improved the CD4/CD8 ratio in peripheral blood, improving the body’s immune function. There were no significant differences in gonadal hormone levels or tumor markers in peripheral blood and no serious adverse reactions were observed. These results align with prior systematic reviews and meta-analyses examining the efficacy of CHM in managing PMS among breast cancer patients. 25 , 26 This systematic literature review and meta-analysis investigated the effectiveness of CHM with Rehmannia glutinosa as the monarch drug in this patient population. The findings of this meta-analysis indicate that the use of CHM with Rehmannia glutinosa as the monarch drug may benefit patients with breast cancer and PMS and is generally safe. Other herbs frequently used with Rehmannia glutinosa included Poria cocos (Fuling) (n=12 RCTs), Cornus officinalis Siebold & Zucc. (shanzhuyu) (n=11 RCTs), and Bupleurum alpinum Nyman (Chaihu) (n=10 RCTs). The present meta-analysis implies that Rehmannia glutinosa as the monarch drug and these other herbs represent a core herbal prescription that warrants investigation in clinical trials for patients with breast cancer and PMS. Modern scientific techniques and methodologies are required to clarify the potential therapeutic mechanisms of CHM with Rehmannia glutinosa as the monarch drug may in patients with breast cancer and PMS. Catalpol is an active ingredient in Rehmannia glutinosa. 27 Catalpol can induce apoptosis in breast cancer in vitro and in vivo and inhibit apoptosis and oxidative stress in glucose-deprived H9c2 cells by enhancing cell mitophagy and modulating the estrogen receptor, thereby protecting cardiomyocytes from myocardial ischemia. 28 , 29 Dried root of Rehmannia glutinosa can prevent bone loss caused by ovariectomy without affecting hormones like estrogen. 30 The present study showed that CHM with Rehmannia glutinosa as the monarch drug significantly improved the severity of PMS symptoms compared to Western medicine in patients with breast cancer; however, E 2 levels in peripheral blood were not significantly different. The pharmacological mechanism of Rehmannia glutinosa in patients with breast cancer and PMS warrants further investigation. Tamoxifen, an endocrine therapy used to treat breast cancer, has the dual effect of stimulating and antagonizing estrogen in different tissues. Estrogen antagonism occurs in breast tissue and estrogen-like effects occur in uterine, heart, and bone tissue; therefore, estrogen levels in peripheral blood are unlikely to generate meaningful information on the mechanism of action of Rehmannia glutinosa. Future research may explore the effect of CHM with Rehmannia glutinosa as the monarch drug on the myocardium, bone tissue, and uterus (endometrium, uterine fibroids, and adenomyosis) in patients with breast cancer and PMS. The RCTs included in this meta-analysis used multi-herb formulae, most frequently Rehmannia glutinosa combined with Poria cocos (Fuling) (n=12 RCTs), Cornus officinalis Siebold & Zucc. (shanzhuyu) (n=11 RCTs), and Bupleurum alpinum Nyman (Chaihu) (n=10 RCTs). These formulae appeared to be beneficial in patients with breast cancer and PMS but the active components and underlying mechanisms of action remain unclear. Pachymic acid (PA), the most active compound in Poria cocos (Fuling), can induce cytotoxicity in the human breast cancer cell line MDA-MB-231, but not the human normal breast cell line MCF-10A. 31 Purified triterpenes dehydropachymic acid and polyporenic acid C (PPAC) extracted from Poria cocos(Fuling) can dose dependently downregulate the expression of CDC20 in the human pancreatic cancer cell line PANC 1 to block tumor progression. 32 Secoiridoids from Cornus officinalis Siebold & Zucc. (shanzhuyu) can potentiate progesterone signaling to relieve PMS. 33 Saikosaponin A (SSA), a primary triterpenoid saponin derived from Bupleurum alpinum Nyman (Chaihu), can suppress tumor growth in an orthotopic 4T1 breast cancer model by inhibition of tumor angiogenesis and reverse P-glycoprotein-mediated multidrug resistance in breast cancer therapy. 34 , 35 First, the risk of bias assessment indicated the methodological quality of the 17 included RCTs was low; none of the RCTs calculated optimal sample size to ensure sufficient power for detecting statistical significance; and 4 RCTs did not report adverse reactions. Second, PMS was diagnosed according to TCM theory, but there are no concise and precise standards in TCM practice. Third, although Rehmannia glutinosa was the monarch drug, included RCTs used many different multi-herb formulae, and the duration of the intervention was short, ranging from 2 weeks to 3 months. In conclusion, findings from this meta-analysis suggest that CHM with Rehmannia glutinosa as the monarch drug may be used to treat PMS in patients with breast cancer. These results should be considered in the context of the methodological flaws and high heterogeneity between included RCTs. Large scale RCTs with longer follow-up are required to verify these data.