Clinical evaluation of Fufangkushen combined with oxaliplatin in the treatment of advanced colorectal cancer

ZHOU Tong, WANG Shuo, HU Shuai-hang, LI Zheng, FAN Bing-jie, LI Jing-lei, HOU Wei?

1. Guang' anmen Hospital Affiliated to China Academy of Chinese Medical Sciences, Beijing 100053, China

2. China Academy of Chinese Medical Sciences, Beijing 100700, China

Keywords:Fufangkushen Chemotherapy Advanced colorectal cancer Curative effect Security Meta analysis

ABSTRACT Objective: To systematically evaluate the efficacy and safety of compound Kushen injection combined with oxaliplatin chemotherapy in the treatment of advanced colorectal cancer. Methods: We searched PubMed, EMbase, the Cochrane Library, CNKI, VIP and Wan Fang database, SinoMed to collect compound Kushen injection combined with chemotherapy oxaliplatin into treatment of advanced colorectal cancer in randomised controlled trials; the databases weresearched from inception to December 2020. Meta-analysis of the included studies was performed using RevMan 5.4. Results: A total of 34 randomized controlled trials involving 2 664 patients with colorectal cancer were included. Results of Meta-analysis showed that compound Kushen injection combined with oxaliplatin chemotherapy regimen improved the objective response rate of tumor [RR=1.40, 95%CI(1.29,1.51), P＜0.000 01] and disease control rate [RR=1.12, 95%CI(1.08,1,16), P 0.000 01] improved the quality of life[RR=1.24, 95%CI(1.14,1.36), P＜0.000 01], and significantly reduced the incidence of leukopenia[RR=0.35,95%CI(0.23,0.52), P＜0.000 01] and the incidence of diarrhea [RR=0.36, 95%CI(0.19,0.70),P=0.003], and improved the immune function of patients (CD3+, CD4+, CD4+/CD8+, NK cell levels).However, compared to the control group, the levels of CD8+ cells were decreased in the experimental group. Conclusion: Compound Kushen injection combined with oxaliplatin chemotherapy regimen can significantly improve the clinical efficacy of advanced colorectal cancer patients, improve the quality of life of patients, reduce the occurrence of adverse reactions, and has good efficacy and safety comparison with oxaliplatin chemotherapy regimen alone.?Corresponding author: HOU Wei, Professer, Doctoral Supervisor.E-mail: houwei1964@163.com

1. Introduction

Colorectal cancer (CRC) is one of the most common malignant tumors in the world[1]. According to statistics, the incidence and mortality of CRC ranked the third among all malignant tumors in the world in 2020, and the incidence and mortality of CRC ranked the fifth among all malignant tumors in China[2-3]. Patients with early colorectal cancer can choose the treatment method of colonoscopy combined with surgery with a high cure rate, but about 30-50% of them will have recurrence and metastasis after radical resection [4], and most of them are diagnosed as advanced stage, in which case chemotherapy becomes the first treatment. Oxaliplatin chemotherapy regimen is currently the standard chemotherapy regimen for patients with advanced colorectal cancer [5], which

often causes multiple adverse reactions, including bone marrow suppression, digestive tract reaction and peripheral neurotoxicity,which seriously affects the quality of life of patients. Many patients are forced to give up chemotherapy, and the clinical efficacy is greatly reduced [6].

In recent years, compound Kushen Injection (CKI) with sophora and cocos as the main active components has been widely used in clinical practice, and has played a good role in inducing tumor cell apoptosis and promoting host anti-tumor immune response[7].Most of the previous studies comprehensively reviewed colorectal cancer patients at all pathological stages, and did not strictly limit the chemotherapy regimen, which could not provide guidance for the clinical treatment of advanced colorectal cancer patients with infeasible surgery or postoperative recurrence and metastasis. This article aims to fill this gap and conduct a rigorous meta-analysis of the Randomized controlled trial (RCT) of compound matrine injection combined with oxaliplatin in the treatment of advanced colorectal cancer, in order to evaluate its efficacy and safety and guide clinical application.

2. Materials and methods

2.1 Inclusion and exclusion criteria

2.1.1 Study typesRandomized controlled trial (RCT) of compound Matrine injection combined with oxaliplatin in the treatment of advanced colorectal cancer.

2.1.2 Types of ParticipantsAll enrolled patients met the 8th American Joint Committee on Cancer (AJCC) diagnostic criteria for advanced colorectal cancer [8]or Dukes stage. Stage Ⅲ-Ⅳ patients with pathology, cytology and imaging diagnosis of infeasible surgical treatment or postoperative recurrence and metastasis.

2.1.3 Types of InterventionControl group: oxaliplatin chemotherapy regimen group (including FOLFOX regimen and XELOX regimen); Experimental group:On the basis of control group, compound Matrine injection was combined.

2.1.4 Outcomes

① Clinical efficacy: Objective response rate (ORR) and disease control rate (DCR) were used to judge the clinical efficacy:According to the World Health Organization (WHO) efficacy evaluation criteria for solid tumors [9] and RECIST efficacy evaluation criteria for solid tumors [10], the patients were divided into complete response (CR), partial response (PR), stable response (SD)and progressive response (PD). Objective response rate = (CR +PR)/total number × 100%. Disease control rate = (CR cases +PR cases+SD cases)/total cases ×100%. ② Quality of life: Karnofsky(KPS score) improvement rate was used to evaluate the quality of life.After treatment, KPS score increased by 10 was considered as improved, decreased by 10 was considered as decreased,and changed by ＜10 was considered as stable. Improvement rate=(improved cases + stable cases)/total cases × 100%. ③ Safety:The incidence of adverse reactions was used as the outcome index of safety evaluation. According to the Criteria for Acute and subacute Toxicity of Chemotherapy Drugs formulated by WHO in 1981 [11],adverse reactions include bone marrow suppression (hemoglobin toxicity, leukocyte toxicity, platelet toxicity), digestive tract reactions(nausea and vomiting, diarrhea), peripheral neurotoxicity, and abnormal liver function. ④ Immune function: The proportion of immune cell subsets after treatment was used as the evaluation index of immune function. Including CD3+, CD4+, CD8+, CD4+/CD8+, NK.

2.1.5 Exclusion criteria

① Non-chemotherapy means such as radiotherapy and targeted drugs were used. ② Use other traditional Chinese medicine preparations except compound Matrine injection. ③ Repeatedly published literature (the first paper was retained), or the research data of different literatures were duplicated (the paper with the most complete data retained). ④Complete research data could not be obtained.

2.2 Research Strategy

Chinese literature databases include China National Knowledge Infrastructure (CNKI), Chongqing VIP Chinese Science and Technology Journal Database (VIP), WanFang Database and China Biomedical Literature Service (SinoMed). The English databases are PubMed, EMBASE and Cochrane Library. Databases such as CLINICALTRIALS.GOV and WHO REGISTRY were manually searched for ongoing clinical trials or unpublished grey literature data. Retrieval date Self-built database to December 31, 2020.According to the RCT search strategy listed in Cochrane Systematic Review manual, the search terms were divided into two parts: target disease and intervention, and the combination of subject terms and free terms was used. The Chinese search terms were "colon cancer","rectal cancer", "colorectal cancer", "colorectal cancer", "compound matrine injection", "compound Matrine injection", "compound Matrine injection", "Yanshu injection", "Yanshu". The English search words are "Colorectal Neoplasms", "Colonic Neoplasms", "Rectal Neoplasms", "Compound Kushen Injection", "Fufangkushen","yanshu".

2.3 Literature screening and data extraction

By reading the title and abstract, the papers that obviously do not meet the inclusion criteria are excluded, and the full text is analyzed to determine whether to be included. The screening was performed by two investigators independently and cross-checked. In case of disagreement, the third investigator evaluated and handled the screening. In case of missing documents, contact the author by email or telephone to supplement them. The main contents of the extracted data include: ① basic information; ② Study population data; ③Intervention measures; ④ Control measures; ⑤ Outcome indicators.

2.4 Literature quality evaluation

The methodological quality of the included studies was assessed using the Cochrane Manual of Systematic Reviews 5.1.0 Risk of Bias Assessment tool. The evaluation included 7 aspects: random sequence generation, random concealment, patient/investigator blinding, outcome rater blinding, outcome data integrity, selective outcome reporting, and other sources of bias. Each evaluation Angle determines the risk type according to the risk of bias evaluation standard [12].

2.5 Data synthesis

Revman 5.4 software was used for statistical analysis. Clinical efficacy, quality of life and safety outcome indicators were dichotomous data, and relative risk (RR) was used as the effect indicator. The outcome index of immune function was continuous variable data, and the mean deviation (MD) was used as the effect index, and 95% confidence interval (CI) and P value were calculated. If P＞0.1 and I250%, there was no significant statistical heterogeneity, and fixed effects model was used for analysis. If P 0.1 and I2＞50%, the heterogeneity between studies was significant.Random effects model was used for analysis, and the source of heterogeneity was determined by sensitivity analysis or subgroup analysis. Funnel plots were used to analyze the publication bias of the included studies.

3. Results

3.1 Literature search results

A total of 810 articles were retrieved from the database according to the retrieval strategy. The obtained articles were imported into NoteExpress software, and 503 duplicated articles were eliminated.By reading the title and abstract, 247 articles that did not meet the inclusion and exclusion criteria were excluded, and 26 articles with inconsistent baseline data or low literature quality were excluded from the full text of the remaining 60 articles. Finally, 34 RCTS were included. See Figure 1 for the literature screening process.

3.2 Study Characteristics

34 RCTs[13-46] were included, with a total sample size of 2 664 cases, including 1 343 cases in the experimental group and 1 321 cases in the control group. The minimum sample size was 21 cases and the maximum was 125 cases. The intervention measures in the control group included FOLFOX and XELOX chemotherapy regimens, while those in the experimental group were compound Matrine injection +FOLFOX/XELOX chemotherapy regimens.4 studies [30,31,38,40] course of treatment > 4 cycles, 23 studies courses 4 cycles. 16 [13,16,18,24,27,28,35,37,38-43,45,46] on the basis of KPS score grading evaluation of quality of life improved, 15 research [15,17,18,21,2428,34-38 9] are described, what is happening to adverse reactions In one study [13], the number of hemoglobin, white blood cells and platelets was detected by flow cytometry without grading. Specific literature characteristics are shown in Table 1.

3.3 Quality evaluation of included studies

Fig 1 Flowchart of literature screening

Tab1 Document characteristic information table

The 34 RCTS [13-46] included in the evaluation showed no statistically significant differences in baseline levels. Nine studies described stochastic methods, among which 7 studies [15,18-20,22,23,41] used random number table method,1 study [37] used stratified random equalization method, and 1 study [31] used computerized random blinding method. If the outcome indicators only included objective outcome indicators such as ORR and DCR, the implementation of blinding method would not affect the outcome bias, so it was evaluated as "low risk". If objective outcome indicators were not included, the risk of bias was evaluated as"high risk". If the outcome indicators included subjective outcome indicators such as quality of life and adverse reactions, the risk of bias was evaluated as "unclear". 34 RCTS [13-46] were not terminated early, and all of them had complete outcome reports. As shown in Figure 2.

Fig 2 Assessment of risk of bias of the included studies

3.4 Results of Meta-analysis

3.4.1 Objective response rate

3.4.1.1 WHO efficacy criteria for solid tumors 24 study [14,1622,24-28,30-32,34,36,38,40,41,43,44,46]using the WHO solid tumor curative effect evaluation standard for objective response rate statistics, statistical heterogeneity between studies smaller (P=0.42,I2= 3%), it USES the fixed effect model combined effect quantity.The results of meta-analysis showed that the experimental group could increase the ORR to 1.43 times of the control group, and the difference was statistically significant. (RR=1.43, 95%CI(1.31,1.55),P＜0.000 01)

3.4.1.2 RECIST Efficacy evaluation criteria for solid tumors

The objective response rate of 7 studies [13,23,29,33,35,42,45] was calculated using the RECIST efficacy evaluation criteria for solid tumors. The statistical heterogeneity among the studies was small(P=0.96, I2=0%), so the effect size was pooled using the fixedeffect model. The results of meta-analysis showed that the ORR of the experimental group was better than that of the control group, and the difference was statistically significant. (RR=1.27,95%CI(1.051,1.54), P=0.01)

Comprehensive data analysis showed that the objective remission rate of the experimental group was better than that of the control group, and the difference was statistically significant. (RR=1.40,95%CI(1.29,1.51), P＜0.000 01). As shown in Figure 3.

2.4.2 Disease control rate

2.4.2.1 WHO efficacy criteria for solid tumors

24 study[14,1622,24-28,30-32,34,36,38,40,41,43,44,46] using solid tumor WHO curative effect evaluation standard of disease control statistics,statistical heterogeneity between studies smaller (P = 0.39, I2=5%), it USES the fixed effect model combined effect quantity. The results of meta-analysis showed that DCR of the experimental group was better than that of the control group, and the difference was statistically significant. (RR=1.11, 95%CI(1.07,1.16), P ＜ 0.000 01)

3.4.2.2 Efficacy evaluation criteria of RECIST solid tumors

The objective response rate of 7 studies[13,23,29,33,35,42,45] was calculated using the RECIST efficacy evaluation criteria for solid tumors. The statistical heterogeneity among the studies was small(P=0.79, I2=0%), so the effect size was pooled using the fixedeffect model. The results of meta-analysis showed that DCR of the experimental group was better than that of the control group, and the difference was statistically significant. (RR=1.13, 95%CI(1.03,1.24),P=0.01)

Comprehensive data analysis showed that the disease control rate of the experimental group was better than that of the control group, and the difference was statistically significant.(RR=1.12,95%CI(1.08,1,16),P ＜ 0.000 01).

3.4.3 Quality of life evaluation

16 [13,16,18,24,27,28,35,37,38-43,45,46] original data are stable and the number of cases in Karnofsky score growth or ratio of the total sum of the cases take statistics as the evaluation standard of quality of life improved, the statistical heterogeneity (P=0.006, I2=54%),so the random effect model was used to combine the effect sizes.The results of meta-analysis showed that the quality of life in the experimental group was better than that in the control group, and the difference was statistically significant. (RR=1.24, 95%CI(1.14,1.36),P ＜ 0.000 01). As shown in Figure 5.

Due to the moderate heterogeneity, the literature was excluded one by one for analysis. When the study of Gao Wei et al. [18] was excluded, the heterogeneity decreased to 23%, indicating that this study was the main source of heterogeneity. The oxaliplatin in the chemotherapy regimen of this study was 130 mg/m2, while the other studies were mostly 85 mg/m2or 100 mg/m2. Therefore, the source of heterogeneity was considered to be related to the high dose of chemotherapy drugs in this study.

Fig 3 Forest plot of objective response rate

3.4.4 Adverse reactions

A total of 15 studies [15,17,18,21,2428,34-38,41] reported adverse reactions after treatment, and the detailed analysis results are shown in Table 2. Compared with the control group, experimental group to reduce the incidence of hemoglobin incidence, the incidence of leukopenia, thrombocytopenia, incidence of nausea and vomiting,diarrhea incidence, the incidence of peripheral nerve toxicity,reduced incidence of liver function damage, the results showed that compound sophora injection to reduce leukopenia and the incidence of adverse reactions of diarrhea curative effect is distinct,The difference was statistically significant(P＜0.05). There was no significant difference in hemoglobin reduction rate between the experimental group and the control group(P＞0.05).

Due to heterogeneity in the rate of hemoglobin reduction, the incidence of nausea and vomiting, and the incidence of peripheral neurotoxicity, all indicators were removed from the study one by one for analysis. It was found that all the above indicators included this study, and the heterogeneity index of each indicator I2decreased to 0% after removal. And one by one to eliminate the remaining literature, statistical result of combined effect quantity and total amount effect were no significant difference, and statistical heterogeneity, there was no apparent change, illustrate Xi Rui’s research is the main source of heterogeneity, after the comprehensive analysis of research shows that the study intervention time about 6 cycles, as included in the study period is the elder, Therefore, the main reason for considering large heterogeneity is related to the long duration of treatment.

Fig 4 Forest map of disease control rate

3.4.5 Immune functionA total of 3 studies [15,20,24] have evaluated the immune function of patients, and the specific analysis results are shown in Table 3. Compared with the control group, the levels of CD3+, CD4+,CD4+/CD8+and NK cells in the experimental group were higher than those in the control group after treatment, and the differences were statistically significant(P＜0.05). The level of CD8+cells in experimental group was lower than that in control group, and the difference was statistically significant(P＜0.05). The heterogeneity of CD4+and CD8+ index was significant, which was considered to be related to the small number of studies and samples. Table 3.

3.5 Publication bias

Funnel plots were drawn for objective response rate and disease control rate, and it was found that the research distribution was relatively symmetrical and there was no obvious publication bias.See Figure 6 and 7.

Tab 2 Meta-analysis of adverse reactions of CKI combined with oxaliplatin chemotherapy in the treatment of colorectal cancer

Fig 5 Forest plot of quality of life

Tab 3 Meta-analysis of immunologic function of CKI combined with oxaliplatin chemotherapy in the treatment of colorectal cancer

Fig 6 Funnel plot of objective response rate

Fig 7 Funnel plot of disease control rate

4. Disscussion

Past for CKI system evaluation is for the treatment of colorectal cancer are relationship, but more is the early, middle and late curative effect evaluation on the colorectal cancer study mixed together, did not give specific clinical staging of into the object of study, lead to clinical efficacy evaluation conclusion heterogeneity is bigger, can't give late infeasible the post-operation relapse and metastasis of CRC surgery or clinical practice provides guidance significance. In addition, previous studies have not restricted the chemotherapy intervention program, and the efficacy evaluation of FOLFOX, XELOX, FOLFIRL, chemotherapy combined with targeted therapy and immunotherapy is mixed, which is difficult to reflect the specific clinical efficacy of the first-line chemotherapy regimen of compound matrine injection combined with oxaliplatin alone. This study systematically evaluated the efficacy and safety of CKI combined with oxaliplatin in the treatment of advanced CRC from four aspects: clinical efficacy, quality of life, adverse reactions and immune function.

In our country, the number of death cases caused by CRC is 187 100, accounting for 8.01% of all malignant tumor deaths [47]. The 5-year relative survival rate of stage Ⅰ colorectal cancer can reach 90%, while the 5-year relative survival rate of stage Ⅳ colorectal cancer with distant metastasis is only 14%[48]. At present, FOLFOX and XELOX chemotherapy regimens are the most commonly used first-line chemotherapy regimens in clinical practice [49]. Oxaliplatin is a third-generation platinum-based broad-spectrum cytotoxic drug,which belongs to DNA-damaging chemotherapy drugs containing 1,2-diaminocyclohexane group. Platinum atoms antagonize replication and transcription of DNA by cross-binding with DNA, and have obvious anti-cancer activity [5,50]. How to combine drugs to reduce the toxic and side effects of oxaliplatin chemotherapy regimen,improve the clinical efficiency of drug-resistant patients, and improve the quality of life of patients is an urgent clinical problem to be solved.

CKI is the relationship of sophora and clay three flavour drug extracting effective ingredients of traditional Chinese medicine injection, has clear heat cool blood, fights the efficacy of detoxification, through the joint action of targets, mechanism, the comprehensive control proto-oncogenes and tumor suppressor gene,inducing tumor cell apoptosis, promote the host antitumor immune response, inhibit the formation of tumor angiogenesis [51] has good effect. Molecular biological studies have confirmed that CKI can affect the expression of BCL-2 proto-oncogene and telomerase,reduce the level of survivin, and induce apoptosis of tumor cells, so as to achieve anti-tumor effect [52]. Through in vitro experiments,Wang Hongqian et al. found that the CKI group, oxaliplatin group and the combination group had certain killing effects on colon cancer cell lines SW620, HT29 and LOVO, while the inhibition effect of the combination group on cell line proliferation was significantly stronger than that of the oxaliplatin group and CKI group, showing a dose-dependent effect [53].

A total of 34 studies [13-46] were included in this study, including 2 664 patients, 1 343 in the experimental group and 1 321 in the control group. Results compared with oxaliplatin, CKI combined with oxaliplatin could improve the objective response rate of advanced CRC patients by about 17% and the disease control rate by about 9%, and the differences were statistically significant (P＜0.05).At the same time, it can reduce the occurrence of adverse reactions such as bone marrow suppression, digestive tract reaction, liver function injury, peripheral neurotoxicity, especially in reducing the adverse reactions of leukopenia and diarrhea. In terms of immune function, CD3+reflects the level of cellular immunity of the body,while CD4+expands the immune response mainly by secreting lymphokines and plays a synergistic role with other immunizations.CD8+can inhibit CD4+cells and B cells [54]. The decrease of CD4+/CD8+ratio represents the decrease of cellular immune function in patients. As one of the evaluation criteria of anti-tumor, NK cells play an important role in immune surveillance and killing mutant cancer cells [55]. In the included study, the levels of CD3+, CD4+,CD4+/CD8+and NK in the experimental group were higher than those in the control group, while the levels of CD8+were lower than those in the control group, suggesting that CKI may improve the immune function of the body by enhancing CD3+, CD4+, CD4+/CD8+and NK.

Limitations of this study: (1) The included study was a singlecenter study: all the included literatures were single-center studies,so the test efficacy was insufficient; (2) The methodological quality of the included studies was poor: most of the studies did not mention the evaluation report on the concealment of the randomized protocol and the degree of blinding implementation. In the study of tumor diseases, considering the side effects of chemotherapy and the long treatment cycle, it is difficult for patients to completely complete the treatment according to the predetermined protocol. However,34 studies [13-46] did not mention the situation of loss to follow-up or dropping off, so there is potential bias in the selective outcome report. (3) Lack of long-term outcome indicators in the included studies: Most of the outcome indicators of each study were based on clinical efficacy and quality of life. For example, there was only one study that analyzed the end points of clinical trials, such as progression-free survival, disease-free survival, and time to disease progression. There were too few data available for analysis,suggesting that other RCTS should pay attention to long-term survival indicators in the future. To provide research data for longterm clinical benefit of compound Matrine injection. (4) Different drug courses of included chemotherapy regimens: Although the drugs of specific chemotherapy regimens were strictly limited,there were differences in chemotherapy cycles among different studies, which would lead to potential bias in research conclusions.Therefore, large-sample, multicenter RCT is still needed for further verification.

Author’s contribution:

The first author, Zhou Tong, was responsible for the design and writing of the paper, and completed the literature screening and data extraction; Wang Shuo and Hu Shuai-hang were responsible for independently screening the literature and extracting data; Li Zheng and Fan Bing-jie checked and consulted and assisted in revising the format of the paper and related charts; Li Jing-lei was responsible for revising the paper; and corresponding author Hou Wei was responsible for topic selection, design and revision.

All authors declare no conflict of interest.