Danshen Formulae for Cancer: A Systematic Review and Meta-Analysis of High-Quality Randomized Controlled Trials

Objective Cancer is one of the most dangerous diseases to human life and there is no radical cure for it. In this paper, we compiled quantities of case history to evaluate the current available evidence of herbal Danshen (Radix Salviae Miltiorrhizae, RSM) formulae for the treatment of cancer by means of the high-quality randomized controlled trials (RCTs). Methods English and Chinese electronic databases were searched from PubMed, the Cochrane Library, EMBASE, and the China National Knowledge Infrastructure (CNKI), VIP database, Wanfang database until September 2018. The methodological quality of the included studies was evaluated by using the method of Cocharne evidence-based medicine system evaluation, the quality was evaluated by screening the literature that met the requirements, and the Review Manager 5.3 was used for statistical analysis. The pooled odds ratio (OR) with 95% CIs was used to estimate the correlation between Danshen formulae and therapeutic effects. Results Thirteen RCTs with 1045 participants were identified. The studies investigated the lung cancer (n = 5), leukemia (n = 3), liver cancer (n = 3), breast or colon cancer (n = 1), and gastric cancer (n = 1). A total of 83 traditional Chinese medicines were used in all prescriptions and there were 3 different dosage forms. Meta-analysis suggested that Danshen formulae had a significant effect on RR (response rate) (OR 2.38, 95% CI 1.66-3.42), 1-year survival (OR 1.70 95% CI 1.22-2.36), 3-year survival (OR 2.78, 95% CI 1.62-4.78), and 5-year survival (OR 8.45, 95% CI 2.53-28.27). Conclusion The current research results showed that Danshen formulae combined with chemotherapy for cancer treatment was better than conventional drug treatment plan alone.


Introduction
Cancer, also known as malignant tumors, can destroy the structure and function of tissues and organs and cause necrotic hemorrhage and infection, and patients may eventually die due to organ failure. In recent years, the incidence of malignant tumors has increased. In the European Union, it is estimated that there are about 1.4 million new cancer cases in a year, including 1.2 million women, and about 70,000 men and 55,000 women die of cancer [1]. In the United States, cancer morbidity and mortality will continue to rise, and lung cancer is expected to remain the number one cancer killer [2].
As one of the world's five most intractable diseases, cancer is an incurable disease. If it is not detected and treated in time, it can also be transferred to all parts of the body for growth and reproduction and finally lead to body weight loss, weakness, anemia, loss of appetite fever, damage of viscera function, etc. [3]. The current cancer treatment is mainly surgery, chemotherapy, and radiotherapy, but both of these treatments cannot reduce the recurrence and metastasis after surgery. It is necessary to cooperate with other treatments after the operation [4]. The effects of radiotherapy and chemotherapy are obvious, but the obvious disadvantage is side effects. Cancer patients will be weak because of illness, and their constitution will be worse after chemotherapy and chemotherapy, which will lead to a decline in the quality of life and even make the body weaker and unable to withstand the next treatment [5].
Traditional Chinese medicine (TCM) treatment is a traditional treatment method in China. For cancer, combination of traditional Chinese medicine can promote the rehabilitation of patients and prevent postoperative tumor recurrence and 2 Evidence-Based Complementary and Alternative Medicine metastasis. At the same time, traditional Chinese medicine can reduce side effects by radiotherapy and chemotherapy and improve the quality of life (QOL) of patients and even improve the survival rate [6,7]. The Radix Salviae Miltiorrhizae (Danshen) originated from "Shen Nong's Herbal Classic" is a well-known TCM herb (China Pharmacopoeia Committee, 2005), and it has been used in clinical practice for over 2000 years. However, there was no consensus on the role of Danshen formulae in cancer treatment [8][9][10][11][12][13][14][15][16][17][18][19][20]. To scientifically validate the efficacy and safety of the Danshen formulae, the meta-analysis evaluated the value of Danshen formulae for the treatment of cancer based on high quality randomized controlled trials (RCTs).

Methods
This systematic review and meta-analysis are based on the Preferred Reporting Items for Systematic Reviews and Meta Analyses: PRISMA statement search strategy [21].
. . Search Strategy. We conducted literature retrieval through 5 database systems including PubMed, the Cochrane Library, EMBASE, and the China National Knowledge Infrastructure (CNKI), VIP database, Wanfang database. The retrieval deadline was September 2018. The main search terms of this paper included the following three parts: ( . There was no restriction on the type of language. In addition, we used manual references to previously published system reviews to manually search for additional related research. The specific herb name "Danshen" has not been specifically searched to ensure that as many herbal formulae as possible were included. . . Inclusion Criteria. (1) Type of participants: researches involving patients with any type of cancer (2) Type of study: only RCTs that assessed the efficacy and safety of cancer treatment were eligible (3) Type of intervention: Danshen must be included in the herbal formula used in the experimental group. There were no restrictions on the form of the drug (e.g., decoction, injection, pill, and capsule), dosage, frequency, or treatment time. Control group medications include placebo or conventional medication (4) Types of results: the efficacy of cancer treatment was evaluated through OS (overall survival), duration of overall response, duration of stable disease, DFS (diseasefree survival), PFS (progression-free survival), TTP (time to progression), TTF (time to failure), DCR (disease control rate), ORR (overall response rate), RR (response rate), CBR (clinical benefit rate), and ORR (objective response rate). Secondary outcome measures were quality of life (QOL) or side effects (such as fatigue, pain, infection/fever, anemia, diarrhea, nausea and vomiting, hair loss, and myelosuppression) (5) RoB scored≥4 points . . Exclusion Criteria. If the above conditions were not met, the study was excluded. In addition, the following documents were also excluded: (1) duplicate publications and (2) case series, review, animal studies, and pharmacological experiments, (3) other TCM therapies, such as acupuncture, massage, qigong, moxibustion, and Taiji were included in the study.
. . Study Selection. The two reviewers selected the study by screening the title and summary of the selected qualifying RCTs independently. Obtain and read the full text of studies that may meet predefined criteria. When the data were duplicated, only the most recent information is selected. The differences in the study choices were resolved by discussion with the latter corresponding author.
. . Quality Assessment. The methodological quality of the included studies was assessed using the risk of bias (RoB) tools, according to Cochran's Systematic Review Handbook on interventions [22]. Including the following seven contents: (A) random sequence generation (selection bias); (B) allocation concealment (selection bias); (C) blinding of participants and personnel (performance bias); (D) blinding of outcome assessment (detection bias); (E) incomplete outcome data (attrition bias); (F) selective reporting (reporting bias); (G) other bias.
. . Data Extraction. Two reviewers used predesigned standard data extraction forms to independently extract data from eligible trials. The excerpts were as follows: (1) the year of publication and the name of the first author, the language of publication, the type of cancer; (2) the characteristics of the participants, including the number, gender, age; (3) treatment information, including intervention management details, treatment process, and side effects; (4) measurement of results.
. . Danshen Formulae Composition. In each study, components of Danshen formula were documented, with a frequency analysis of the types of cancer it treated and common drugs combined with it.
. . Data Analysis. Dates from eligible researches were aggregated and a quantitative summary was generated by Review Manager 5.3. Egger's test was carried out by Stata 12.0.
. . Effect Size. Dichotomous data were reported as odds ratio (OR) with 95% confidence intervals (CI). The OR was deliberate significant at the P < 0.05 level when value 1 was not taken into the 95% CI. The purpose of this article is to explore the effectiveness of the Danshen formulae; therefore, we take the number of positive events as weights.  is no absence of heterogeneity or moderate heterogeneity (P > 0.1, 2 < 50%) [23], a fixed effect model (FEM) will be used; otherwise, a random effects model (REM) will be applied.
. . Publication Bias. If a research project contained more than 10 studies, the funnel plot were used to test publication bias.
. . Sensitivity Analysis. Sensitivity analysis was performed to test the effect of a single study on the combined effect by removing the individual survey. If the estimated value of the point after deleting a study fell beyond the 95% CI of the total effect amount (or was significantly different from the combined effect amount), we considered the study in question to have exerted a great influence on the combined effect amount and that this study required further review.

Results
. . Description of Studies. A total of 3209 studies were searched by 4 electronic databases and other sources, and 2664 records were retained after deduplication. Of these, 366 studies were unrelated to cancer, 434 were animal experiments, 462 were mechanical experiments, and 728 were reviews, protocols, experiences, or case reports. By reading the full text, 762 studies, including 103 control interventions were inappropriate, lacking 50 of the control group and 70 items without full text, 200 items were not really RCTs, 99 items did not use the Danshen formula, and 142 included other CAM treatments, such as Acupuncture, Qigong, and 90 methodology that is of low quality. In the end, the study included 13 studies with 1045 patients that met the Cochrane Robb score of 4 and conducted a meta-analysis. A PRISMA flowed graph describe the search process and research options (Figure 1).
. . RoB Assessment. The RoB evaluation is shown in Table 4. All studies are described as random. Twelve RCTs mentioned random allocation methods, including random sampling, picking method, hospitalization time, completely randomized digital table, and stratified permuted block method, and the rest of the study had only the words "randomized grouping". 1 RCT explicitly proposed that the study was conducted by a single blind method [9], 2 RCT explicitly proposed the use of double-blind method in the title or abstract [10,13], and the rest of the studies were relatively vague about the blind method, and we need to get relevant information by reading the full text. As shown in the table, among the 13 studies, 3 articles scored 5 points [9,10,13], and the rest were scored 4 points [8,11,12,[14][15][16][17][18][19][20] (Table 5) [8,[10][11][12][14][15][16][17][18][19] analyzed RR, indicating the RR of the experimental group was higher than that of the control group (OR 2.38, 95% CI 1.66-3.42) (Figure 2). Heterogeneity test P = 0.21, I = 25% showed 13 included articles with no heterogeneity, so the statistical analysis with fixed effects model. Pooled OR with 95% CIs showed Z = 4.71, P<0.00001 (Figure 2), suggesting that the difference was statistically significant. It can be considered that the RR of Danshen formulae with the general treatment regimen was higher to the control scheme without the Danshen formulae.
. . Publication Bias. The funnel plot and further Egger's test were used to evaluate publication bias for RR of two groups of cancer patients. As the two results, though, showed a leftright asymmetry, but both P>0.05, suggesting that there was no publication bias (Figures 6 and 7).
. . Sensitivity Analysis. Our sensitivity analysis did not indicate that the results of any individual study would change the final outcome, indicating that none of the studies significantly affected the pooled OR and 95% CI.  (Figure 8).

Discussion
. . Summary of Evidence. In the past decades, much work has been reported in Chinese Herbal Medicine (HCM) in the treatment of cancer [8-20, 24, 25], and Zhang's review provided evidence for the effectiveness of Danshen in the treatment of cancer [26]. However, there has not been a meta-analysis to study the value of Danshen formulae in cancer treatment. This paper was a systematic review of 13 high-quality RCTs, including 1045 participants, to determine the efficacy and safety of the Danshen formulae for cancer  . . Implications for Practice. Modern pharmacological studies were performed on more than 10 tanshinone monomers, including tanshinone I (TNI), tanshinone A (TNIIA), tanshinone B, and cryptotanshinone (CPT), from Danshen root. Tanshinone TNI are the main bioactive components, TNIIA, and implicit tanshinone (CPT); TNIIA activity in salvia miltiorrhiza is the strongest diterpene quinine pigment; TNI and CPT are effective cytotoxic agent and can induce apoptosis and the stagnation of the cell cycle; potential mechanisms involved include raised to promote apoptosis proteins such as p53, Bax, and p21 and inhibit antiapoptotic proteins, including the Bcl-2, survivin, and c-Myc and activated caspase protein to trigger apoptosis, by activating AMP activated protein kinase and extracellular signal regulating kinase (ERK) and suppress the target of pakamycin and 70 kDa ribosomal protein S6 kinase signaling pathways; TNIIA induces autophagic cell death in various cancer cells. Furthermore, TNIIA and TNI can inhibit the migration, invasion, and metastasis of cancer cells by changing the tissue inhibitors of matrix metalloproteinase and/or metalloproteinase [26,27]. In addition, TNIIA can also promote the differentiation of several cancer cell types and regulate the CCAAT/enhancer binding protein (C/EBP) and C/EBP homologous protein. Besides, in animal models, the side effects of TNIIA, TNI, and CPT were minimal [28].
In addition, Danshen root also has anti-inflammatory effects. TNIIA inhibits the NF-kB induced kinase/IkappaB alpha kinase (NIK/IKKalpha), while ERK1 suppress NF-B induced by LPS and c-Jun n-terminal kinase (JNK) pathway. The anti-inflammatory effects of TNIIA may be related to the inhibition of the Toll-like receptor (TLR) signaling pathway by TNF receptor-associated factor (TRAF) 2/3/6. TNI significantly inhibits the activity of IIA secreting phospholipase A2 (GIIA), thereby blocking the formation of prostaglandin E2 (PGE 2 ) in LPS-activated macrophages [27,29]. TNI and CPT also significantly inhibit IL-12 production in LPS-activated macrophages and interferonproduction in lymphocytes. Recent studies have shown that Salvia miltiorrhiza extract inhibits the production of iNOS and COX-2 by regulating NF-B and MAPKs, thereby inhibiting the secretion of inflammatory cytokines. In LPS induced RAW264.7 macrophages, salvia miltiorrhiza extract reduced the secretion of nitric oxide (NO), tumor necrosis factor-(TNF-) , and interleukin 6 (IL-6) and decreased the expression of inducible nitric oxide synthase (iNOS), cyclooxygenase-2 (COX-2), and NF-B. Moreover, salvia miltiorrhiza extract can significantly inhibit the activation of JNK1/2 and ERK1/2 induced by LPS and disrupt the TLR4 dimerization in LPS-induced RAW264.7 macrophages [29]. Therefore, the anti-inflammatory effect of Danshen root is partly due to the blocking of TLR4 dimerization, which can be used in clinical treatment of liver injury and infection during the anticancer strategies.
Therefore, Chinese medicine practitioners usually use traditional Chinese medicines similar to Danshen root, such as Scutellaria barbata and Hedyotis diffusa in the treatment of cancer [8][9][10][11][12][13][14][15][16][17][18][19][20]. It should be mentioned that the current metaanalysis is the first systematic review of the application of Danshen formulae in cancer-assisted treatment. The current meta-analysis found that Danshen formulae can improve the clinical efficiency in cancer treatment. After the addition of Danshen formulae, RR and survival rates were significantly improved. However, it is not clear which components of Danshen formulae have anticancer effects during the treatment, and what role Danshen root plays, which should be the goal of further research.
. . Limitations. All literatures for this meta-analysis were from China, among which 4 were in English and the rest were in Chinese, and only one was multicenter study [10]. Blind methods have been described vaguely in many studies and most of the references were scored 4 points. Moreover, all types of researches were single-center studies with a small sample size and lack of data support for multicenter, large RCTs. The funnel plot analysis also found obvious asymmetry between the left and the right of funnel plot, therefore, the effect of Danshen formulae in assisting cancer treatment may be exaggerated. In addition to all of the above, the cycle of cancer treatment is extremely long and an army of patients died as their condition worsens during the treatment process, which leads to follow-up work is difficult, resulting in the inability to obtain valid data for many studies, and some literature related to this study cannot be included. Therefore, to better explore the contribution of Danshen formulae in cancer treatment, more large-scale and higher standard studies are needed.
. . Implications for Further Studies. Nearly 50% of the RCTs included in our study are related to lung cancer, indicating that Danshen formulae may be more widely used in lung cancer. Lung cancer is one of the most dangerous diseases, and it has a variety of treatment options, but the death rate is stubbornly high [2]. Except lung cancer, the included studies also cover leukemia, liver cancer, breast, colon cancer, and gastric cancer, and our study found that 9 kinds of herbs were used in combination with Danshen root in cancer treatment, suggesting that the pharmacological effects of these drugs together may be a mechanism to improve clinical efficacy and reduce side effects. Therefore, this study provides the basis for the clinical treatment and scientific research of Danshen for cancer. In terms of gender ratio, we found that men have a higher risk of cancer than women (605/440), suggesting that men should pour more attention into prevention of cancer, which was also a limitation of this paper, indicating that we should avoid gender selection bias by recruiting women to a certain extent in future studies. By inputting the dose form of Danshen formulae to statistical analysis, we found that there were 1 used granule preparation, 1 used tablet, and the other 11 used decoctions. Ling proved that TCM preparations more safer, effective, and easier to use than decoctions of traditional Chinese medicine [30], and the present study showed that the clinical curative effect and dosage forms of CHM were interconnected, interdependent, and mutually reinforcing with each other; drug application shall be familiar with drug characteristics on the premise of fully considering disease characteristics and age constitution and patient and choose the appropriate dosage forms, through the appropriate method to give full play to the effect, and make the drug in patients with optimal clinical curative effect [31,32]. Therefore, rational selection of drug dosage forms is beneficial to enlighten and create new drugs, which can better promote the further development and research of new dosage forms with higher drug absorption rate. In terms of the treatment of cancer, in traditional western medicine, chemotherapy and radiotherapy are the main treatments for cancer. The purpose of these therapies is to kill or destroy cancer cells. Unfortunately, for most cancer treatments, it is difficult to distinguish between cancer cells and normal healthy cells, which leads to damage to normal cells [33,34]. The results of this injury are known as complications and side effects of cancer treatment. There were 10 included studies showed that the formulae of Danshen had significant effect in reducing the side effect of vomiting and blood toxicity, which suggested that we could cooperate with the Danshen formulae in the treatment of cancer in the future to reduce the gastrointestinal reaction of patients. The results of subgroup analysis suggested that Danshen formulae might be taken advantage of for cancer treatment. At last, the exact pathologic and clinical pharmacological mechanisms of cancer are still largely unknown and should be studied further.

Conclusion
Current findings suggested that Danshen formulae offered statistically significant benefits for cancer, which we generally considered safe. Thus, evidence from the existing study supported the use of Danshen formulae as a treatment for cancer. However, this study was based on several small-sample studies. Therefore, studies with rigorous, large-scale RCTs of Danshen formulae in treating of cancer were needed to further confirm its efficacy.

Consent
The study did not involve human participants and/or animals.

Conflicts of Interest
The authors declare that they have no conflicts of interest.