Traditional Chinese medicines and capecitabine‐based chemotherapy for colorectal cancer treatment: A meta‐analysis

Abstract This meta‐analysis was conducted to evaluate the efficacy and safety of the addition of Traditional Chinese Medicine (TCMs) to capecitabine‐based regimens for colorectal cancer (CRC) in term of tumor. The eight electronic databases including Cochrane Library, PubMed, Web of Science (WOS), Excerpt Medica Database (Embase), Chinese Biomedical Literature Database (CBM), China National Knowledge Infrastructure (CNKI), Chinese Science and Technology Journals (CQVIP), and Wanfang Database were systematically searched for eligible studies from their inception to March 2021. Thirty‐nine randomized controlled trials were involved in this study, and all the data were analyzed by Review Manager 5.3 (Nordic Cochran Centre, Copenhagen, Denmark) and R 4.0.5 software. The meta‐analyses suggested that TCMs in combination with capecitabine‐based regimens increased objective response rate (ORR) in the palliative treatment of CRC (risk ratio [RR], 1.35 [1.17, 1.55], I 2 = 0%), disease control rate (DCR) (RR, 1.22 [1.12, 1.32], I 2 = 3%), and quality of life (QOL) (RR, 1.71 [1.44, 2.03], I 2 = 0%), with decreased risks of myelosuppression, anemia, thrombocytopenia, liver/renal dysfunction, neurotoxicity, nausea/vomiting, neutropenia, diarrhea, leukopenia, improved the peripheral lymphocyte, reduced the expression of tumor markers, and related factors. Further sensitivity analysis of specific plant‐based TCMs found that dangshen, fuling, and gancao had significantly higher contributions to the results of the RR. The results show that capecitabine‐based chemotherapy combined with TCM in the treatment of CRC increases the efficiency of ORR and DCR, reduces chemotherapeutic agents‐associated adverse reactions, and improves their life quality as compared with chemotherapy alone, but further randomized and large sample of studies are needed.


| INTRODUCTION
Globally, the incidence of colorectal cancer (CRC) ranks third among all cancers, second in mortality, 1 and cancer cases and deaths represent 10% of all cancer cases and deaths. 2 Despite radiotherapy and chemotherapy is the main treatment method nowadays, the outcome of advanced CRC remains poor due to tumor recurrence and metastasis and drug resistance. For patients who cannot tolerate surgery, the goal is to minimize the tumor and control its further spread and growth. 3,4 Current first-line chemotherapy approach to treating CRC is fluoropyrimidine (5-FU) 5 or multidrug combination regimen including oxaliplant (OX), irinotecan (IRI), and carbapitabine (CAP). However, the adverse drug reactions during chemotherapy have not been effectively solved, and the treatment outcomes are often unsatisfactory. 6 As a predrug of fluorouracine, capecitabine achieves similar efficacy after oral administration. The incidence of adverse reactions with the capecitabine modified XELIRI (CAP+ IRI) protocol was significantly reduced. 7 However, capecitabine still produces adverse reactions such as hand-foot syndrome, myelosuppression, liver/renal dysfunction, and gastrointestinal reactions during patients. 8 Traditional Chinese Medicines (TCM) has been widely used in China for the supplementary treatment of cancer, including colorectal cancer (CRC). 9,10 As an adjuvant therapy, TCM reduces the side effects of cancer reagents and increases the chemotherapeutic efficacy. 11 However, its substantial evidence is inefficient to prove whether the TCM combined capitabine-based regimen is more effective than capitabine alone.
In this study a systematic review and meta-analysis is performed to compare the clinical efficacy and safety between the capecitabine-based chemotherapy combined with TCM and capecitabine alone in the treatment of CRC. At the same time, the frequencies of combined TCMs are further analyzed to determine which combination methods are efficient to improve objective response rates (ORR) and reduce adverse effects, which will provide evidences for the clinical applications of TCM combinations with capitabine-based regimen in treating CRC.

| MATERIALS AND METHODS
The protocol for this systematic review was registered on INPLASY (Unique ID number) and was available in full on the inpla sy.com (https://doi.org/10.37766/ inpla sy2021.3.0095) and was performed in accordance with the PreferredReporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.

| Eligibility criteria and outcome measures
According to the PICOS acronym, 12 the inclusion criteria were as follows: Participants (P): All included cases must be confirmed to be CRC after histopathological examination. No restriction on gender, race, or nation was found. Patients with non-primary CRC or other tumors were excluded.
Intervention (I): The random clinical trials (RCTs) with TCMs combined with capecitabine-based chemotherapy were included. No restrictions were in the types of TCM.
Comparison (C): In the control groups, the patients with CRC were treated with the capitabine-based regime.
Outcomes (O): efficacy and safety of TCM. Study design (S): RCTs. Exclusion criteria: (i) no capecitabine-based chemotherapy and (ii) non RCTs and (iii) with incomplete outcomes and (iv) lack in sufficient data. Primary outcomes included three efficacy measurements: short-and long-term clinical efficacy, and adverse drug reactions (ADRs) according to world health organization (WHO) criteria and response evaluation criteria in solid tumors (RECIST). (I) Short-term clinical efficacy: the short-term tumor response included complete response (CR), partial response (PR), response rates in stable disease (SD), response rates in progressive disease (PD), ORR, and disease control rate (DCR). ORR was defined as the sum of CR and PR, and DCR was the sum of CR, PR, and SD; (II) Long-term clinical efficacy: 1-5 year overall survival rate (OS); (III) quality of life (QOL), QOL is considered to be improved when Karnofsky performance status (KPS) score is higher than 10 points after treated. Secondary outcomes included ADRs, peripheral blood lymphocytes, tumor markers and related cytokines, transfer rate, and time to progress (TTP). According to WHO recommendations for grading of acute and subacute toxicity or NCI common terminology criteria for adverse events (CTCAE), ADRs are evaluated by testing hematotoxicity (neutropenia, anemia, thrombocytopenia, and leukopenia), gastrointestinal reaction (nausea and vomiting, diarrhea), liver/renal dysfunction, neurotoxicity, myelosuppression, and hand-foot syndrome. T-lymphocyte subsets such as the proportion of CD3 + , CD4+, and CD8 + T cells, the ratio of CD4 + /CD8 + T cells, and the proportion of natural killer cells (NK cells) are measured. Tumor markers and related factors tested include CEA, CA199, CA125, CA724, and TNF-α.

| Search strategy and study selection
Literature search in both international (Cochrane Library, PubMed, EMBASE, and Web of Science) and Chinese (CBM, CNKI, CQVIP, and Wanfang Database) databases will be systematically searched for eligible studies from their inception to March 2021, were independently conducted by two researchers (Hui-zhong Jiang and Ya-li Jiang). The retrieved keywords included TCM, CRC, capecitabine, and ADRs. The titles and abstracts were independently screened and then full texts of relevant publications for eligibility were read. Any discrepancy was discussed with a third researcher (Dong-xin Tang). In addition, the references listed in original reports and previous reviews were reviewed, and manually selected for other available publications.

| Data extraction
The following study and participant characteristics were extracted, including first author, year of publication, sample size, type of medications, mean age of participants, cancer staging system (TNM stage), Karnofsky performance status (KPS), TCM intervention (dosage and duration), drug delivery, capecitabine regimen (dose and cycles), and outcome measurements. Any disagreement was resolved by consensus.

| Quality assessment and evidence level
The quality of studies were assessed by Cochrane risk of bias tool Review Manager 5.3 (Nordic Cochran aa). The review criteria cover seven areas included random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, other sources of bias. The included studies were evaluated to three degrees including low, unclear, and high risk of bias.

| Statistical Analyses
Statistical analyses were performed using Review Manager 5.3 and R 4.0.5 software. The outcomes were mainly represented by risk ratio (RR) and standardized mean difference (SMD) with its 95% CIs. A two-tailed p < 0.05 is considered to be statistically significant. Cochrane's Qtest and I 2 statistics were used to assess heterogeneity between studies; p ≤ 0.1 or I 2 > 50% indicates statistical heterogeneity. A fixed-effects model was used to calculate the outcomes when statistical heterogeneity was absent. Otherwise, the random-effects model was used according to the DerSimonian and Laird method. Studies with zero events were included to avoid overestimation of effect. 13 When the same outcome was reported by more than 10 studies, publication bias was tested using funnel plots, Egger's regression test, and Begg's rank test. Sensitivity analysis was conducted to explore an individual study's influence on the pooled results by deleting one single study each time from pooled analysis.
Subgroup analyses were carried out based on the methods of the TCM administration. Meanwhile, only the TCMs with significant tumor responses were included in our analyses. Pooled ORRs were calculated for each group of studies that contained the same TCM. The same pairs of TCMs in three or more studies were identified. The pooled RRs were calculated. They were listed in descending order and any significant was highlighted.

| Literature search and study characteristics
A total of 313 articles were initially identified. After screening the titles and abstracts, 119 articles were retrieved for full-text review. Finally, 39 studies 14-52 with 1384 patients in the TCM combined with capecitabine group and 1367 patients in the capecitabine group were included for metaanalyses ( Figure 1). All the 39 studies were RCTs, and the characteristics are summarized in Table 1. All the 39 studies were conducted in China. Thirty-five studies used the oral TCM, two studies used external TCM, and two studies used commercially available TCM injections (Table 1).

| Methodological bias of the included studies
In 39 trials, the methods of random allocation were described clearly in only 18 trials. 14,15,16,22,26,30,37,38,39,40,42,44,45,47,48,49,50,52 This indicated that there was selectivity bias in the included studies. The random allocation concealment was unclear. Not all the included studies were described as blinding to patients and doctor. Therefore, it indicated that there were selective bias and implementation bias. All data were complete and selective report did not appear in all of the studies. Other bias was not clear. Characteristics and quality of all included studies are presented in Figure 2
One study and 11 studies were included to evaluate the DCR in the non-oral and oral groups. Similarly, compared with capecitabine alone, the combined treatment significantly improved the pooled DCR in the non-oral and oral groups (n = 62, RR,

| The effects of multi-ingredient TCM in the oral administration group
The multi-ingredient TCM formulae had similarity in their main ingredients and functional approximation. In order to identify the most comparable subgroups of studies and potential synergistic effects, a series of planned sensitivity analyses were made. Only the TCMs with significant ORR results have been reported in our analyses. In Table 6, all significant RR results (excluding those with heterogeneity >30%) were ranked in order according to descending RR. Level 1: Single TCM. Sixty ingredients in the formulae have been included in this review. Among them, there are 11 ingredients that have been used in three or more formulae. The Chinese name in pin yin of each ingredient was used to represent the TCMs. According to their pooled RR values were divided into two groups. The RR values in the first group were equal to or greater than the total pool. In the second group, the RR values were less than the total pool.
Level 3: Combinations of 3 TCMs. At this level, there were two significant pairs from level 2 that were combined with other TCMs that showed significant RRs compared with single TCM group. At this level, the RR values of all pairs including dangshen+fuling+gancao (n = 4), huangqi+banxia+baizhu (n = 3) were greater than the total pool (Table 6). Levels 4 to 7: Combinations of 4 to 7 TCMs. There were no combinations of 4, 5, 6 TCMs, and there was one combination of 7 which showed an RR equal to the pool:  huangqi+banxia+baizhu+sheshecao+dangshen+ful-ing+gancao (n = 2) ( Table 6).

| DISCUSSION
TCM's essential components are being studied constantly, and more and more research has proven that TCM may assist with tumor treatment. 55,56 Capecitabine, a 5-FU prodrug, is an effective first-line therapy for CRC due to its ease of use and low frequency of ADRs. 57 Though oxaliplatin-or 5-FU-based chemotherapy combined with TCM was shown to be more effective than TCM alone in two studies, 58,59 the efficiency of capecitabine-based chemotherapy combined with TCM in CRC is yet unknown. Thirty-nine studies including 2751 patients were included in meta-analyses to evaluate the therapeutic CRC regimen capitabine-based coupled with TCMs clinical effectiveness and ADRs. As a consequence, capecitabinebased chemotherapy regimens were shown to be more effective when combined with TCM. The ORR and DCR of the oral TCM or non-oral group (e.g., injection, enema) were shown to be substantially greater than those utilizing capitabine alone, as a consequence of which we exhibited. Improving immunological function and overall well-being  Figure S3. is critical for cancer patients undergoing treatment. We compared the QOL and the number of peripheral blood lymphocytes in each group as part of our research. According to the findings, combining capecitabine-based chemotherapy with TCM improved CD3 + T cells, CD4 + T cells, CD4 + / CD8 + T cells ratio, and NK cells, as well as overall QOL. In addition, it has the potential to decrease tumor marker expression levels (CEA, CA199, and CA125). This will assist the patient's immune system, allowing him or her to fight off tumor recurrence and metastasis in the future. T-lymphocyte expression is linked to poor prognosis and tumor metastasis, 60 such as CD3 +,61 CD4 +,62 CD4 + /CD8 + T cell ratio, 63 and NK cells, 64 increasing immune function to inhibit tumor growth. 65 An important clinical biomarker for gastrointestinal malignancies is a cell surface glycoprotein called carcinoembryonic antigen (CEA) 66 . There has been an increase in CEA overexpression in 90% of gastrointestinal cancers, including CRC tumor recurrence is predicted by an increase in postoperative CA125 and CA199 levels, and this information is critical for the diagnosis of digestive system cancer. 67,68 TCMs used orally or intravenously showed promise in the treatment of CRC. Specific plant-based TCMs were further analyzed and shown to have substantially greater contributions to the RR value, including yiyiren, fuling, gancao, baizhu, sheshecao, dangshen, banxia, and nvzhenzi. Dangshen, fuling, and gancao all contributed considerably more to the RR value than the others at all levels. Capecitabine-based chemotherapy for CRC may benefit from the addition of TCMs.

| CONCLUSION
Our research found that the combination of TCM and capecitabine-based chemotherapy was more effective than the capecitabine-only regimen. Additionally, it has the potential to decrease adverse responses in patients,  enhance survival rates, and the body's capacity to fight off infection, lower tumor marker expression levels, and even slow tumor development. Specific TCMs may have the potential to improve the efficacy of capecitabine-based chemotherapy for CRC.

AUTHOR CONTRIBUTIONS
Hui-Zhong Jiang and Ya-Li Jiang conceived of and designed the study. They had full access to all data in the study and took responsibility for the integrity of the data, the accuracy of the data analysis, and the writing of the report. Yang Bing, Feng-Xi Long, Zhu Yang, and Dong-Xin Tang critically revised the report. Hui-Zhong Jiang and Ya-Li Jiang performed the statistical analyses. All the authors contributed to the data acquisition and analyses. The authors have reviewed and approved the final version of the manuscript.