Add-On Therapy with Traditional Chinese Medicine Improves Outcomes and Reduces Adverse Events in Hepatocellular Carcinoma: A Meta-Analysis of Randomized Controlled Trials

Background and Aims Traditional Chinese medicine (TCM) therapy for hepatocellular carcinoma remains controversial. This study aimed to evaluate the efficacy and safety of TCM regimens in HCC treatment. Methods Randomized controlled trials (RCTs) up to June 1, 2016, of the TCM treatment for hepatocellular carcinoma were systematically identified in PubMed, CNKI, Ovid, Embase, Web of Science, Wanfang, VIP, CBM, AMED, and Cochrane Library databases. Results A total of 1010 and 931 patients in 20 RCTs were randomly treated with add-on TCM therapy and conventional therapy, respectively. The additional use of TCM significantly improved six-month, one-year, two-year, and three-year overall survival rates in HCC cases (RR = 1.3, P = 0.01; RR = 1.38, P = 0.0008; RR = 1.44, P < 0.0001; RR = 1.31, P = 0.02, resp.). Add-on TCM therapy significantly increased PR rate and total response rate (tRR) and reduced PD rate compared to those in control group (34.4% versus 26.3%, RR = 1.30, P = 0.002; 41.6% versus 31.0%, RR = 1.30, P < 0.0001; and 16.6% versus 26.5%, RR = 0.64, P < 0.0001, resp.). Additionally, TCM combination therapy significantly increased the quality of life (QOL) improvement rate and reduced adverse events including leukopenia, thrombocytopenia, anemia or erythropenia, liver injury, and gastrointestinal discomfort in HCC patients (all P < 0.05). Conclusion Add-on therapy with TCM could improve overall survival, increase clinical tumor responses, lead to better QOL, and reduce adverse events in hepatocellular carcinoma.


Introduction
Primary liver cancer is the sixth most common cancer and the third most common cause of cancer-related deaths. 70%∼90% primary liver cancers occurring worldwide are hepatocellular carcinoma (HCC), which is the fastest growing cause of cancer-related death globally [1,2]. Recent epidemiology data revealed that liver cancer might account for more cancer-related deaths worldwide [3]. HCC has a 5-year survival rate of only 14% approximately [4]. Most HCCs are diagnosed at an intermediate to advanced stage, at which point surgical treatment and/or chemical embolism are no longer feasible [5]. Therefore, to improve outcome of HCC patients, an alternative or novel approach is required.
Previous report showed a large prevalence of a diversity of traditional Chinese medicine (TCM) clinical application for cancer patients [6]. Sufficient evidence has demonstrated that natural compounds with various types of medicinal ingredients can substantially inhibit tumor formation [7]. Many clinical articles have reported that TCM or TCM plus chemotherapy can significantly alleviate symptoms, stabilize tumor size, reinforce the constitution, enhance therapy tolerance and immunological function, obviously reduce the incidence rate of adverse events, and prolong patients' survival duration for unresectable HCC [8][9][10][11].
Unfortunately, reporting of RCTs on treatment of HCC with TCM is still in low quality, not meeting the CONSORT 2 Evidence-Based Complementary and Alternative Medicine and TREND statement. High quality of evidence based on the existing clinical information is still unavailable [6,12]. A recent meta-analysis also announced that many RCTs of TCM therapy in HCC are not, in fact, randomized [13]. Thus, only RCTs reported randomized methods were included in our current meta-analysis. The purpose of this study is to systematically review and meta-analyze data from RCTs for evidence on the efficacy and safety of add-on therapy with TCM in the treatment of HCC.

Search Strategy and Study Selection.
We searched Pub-Med, Chinese National Knowledge Infrastructure (CNKI) Database, Wanfang Database, Chinese Biomedical (CBM) Database, Chinese Science and Technology Periodical Database (VIP), Allied and Complementary Medicine Database (AMED), Ovid, Embase, Web of Science, and Cochrane Library databases until June 1, 2016. The following medical subject headings were used: "hepatocellular carcinoma;" "primary liver cancer;" "Traditional Chinese Medicine;" "alternative medicine;" "complementary medicine;" "Chinese herbal medicine;" "herb/herbal;" and "decotion/formulation." Electronic searches were supplemented with manual searches of reference lists used in all of the retrieved review articles, primary studies, and abstracts from meetings to identify other studies not found in the electronic searches. Literature was searched by two authors (Z Yang and X Liao) independently.
Two authors independently selected trials and discussed with each other when inconsistencies were found. Articles that satisfy the following criteria were included: (1) for study types, RCTs with randomized method; (2) for participants, HCCs; (3) for interventions, TCMs compared with placebo or no treatment; in addition, any cointervention had to be the same in both groups except for the TCM formulation; (4) for outcome, overall survival and/or solid tumors responses; and (5) available full texts. If the duration and sources of study population recruitment overlapped by more than 30% in two or more papers by the same authors, we only included the most recent study or the study with the larger number of HCC patients. Studies were excluded if they meet the following criteria: (1) studies "so-called" randomized without randomized methods; (2) studies without control subjects or control participants receiving TCM treatment including herbal medicine and acupuncture; (3) studies reporting only laboratory values and/or symptom improvement rather than survival outcomes and clinical responses.

Data Extraction and Methodological
Quality Assessment. Two researchers independently read the full texts and extracted the following contents: publication data; study design; sample size; patient characteristics; treatment protocol; and outcome measures. The methodological qualities of the included RCTs were assessed according to Cochrane Collaboration's Tool described in Handbook version 5.1.0 [14]. Two authors (Z Yang and X Liao) independently assessed quality, and inconsistency was discussed with other reviewerauthors (Y Yu and X Chen) who acted as arbiters.

Definitions.
All the diagnosis should be according to guidelines. The primary outcome overall survival was defined as the time from HCC diagnosis until the death due to any cause. Solid tumor response is categorized as complete response (CR), partial response (PR), stable disease (SD), progressive disease (PD), and CR + PR as a proportion for total response rate (tRR) according to the World Health Organization (WHO) criteria [15] or the Response Evaluation Criteria In Solid Tumors (RECIST) guidelines [16,17]. Karnofsky performance status (KPS) [18] and adverse events were also measured in our study.

Statistical
Methods. The effect measures of interest were risk ratios (RRs) and the corresponding 95% confidence intervals (CIs). Heterogeneity across studies was informally assessed by visually inspecting forest plots and formally estimated by Cochran's test in which chi-square distribution is used to make inferences regarding the null hypothesis of homogeneity (considered significant at < 0.10). A rough guide to our interpretation of 2 was listed as follows: (i) 0% to 40% shows that heterogeneity may not be important. (ii) 30% to 60% corresponds to moderate heterogeneity.
If the eligibility of some studies in the meta-analysis was uncertain because of missing information, a sensitivity analysis was performed by conducting the meta-analysis twice: in the first meta-analysis, all of the studies were included; in the second meta-analysis, only those that were definitely eligible were included. A fixed-effects model was used initially for our meta-analyses; a random-effects model was then used in the presence of heterogeneity. Description analysis was performed when quantitative data could not be pooled. Review Manager version 5.1 software was used for data analysis.

Records screened (n = 895)
Studies included in quantitative synthesis (meta-analysis) (n = 20) Studies included in qualitative synthesis (n = 24) with no presenting of blinding of outcome assessment. Less than 15% of participants were lost to follow-up in the three studies [20,21,25,[35][36][37]39]; these parameters were considered low risk in terms of incomplete outcome data. Selective reporting was found in three studies [21,25,35] because these researches failed to present the clinical data of participants in ITT analysis. Other potential biases were unclear in these trials ( Figure 2).

Discussion
Most newly diagnosed HCC cases are at an intermediate advanced stage, and the therapeutic options are limited to palliative approaches using TACE or chemotherapeutic agents [5,40]. Even worse, many patients poorly respond to TACE or suffer from poor outcomes and side effects with conventional systemic cytotoxic chemotherapy [40], leading to disappointing results of systemic chemotherapies and a poor prognosis. Therefore, novel therapeutic strategies are essential to improve the clinical management of patients with HCC.
With a long history of clinical use, essential components of TCM have gradually become a common used treatment for cancer in China [41]. In particular, TCM has been used to treat HCC extensively and it can be used throughout the whole course of HCC [42]. In the past decades, many compounds derived from Chinese herbals of both preclinical and clinical researches have shown promising potentials in novel anti-HCC natural product development [43]. Previous studies indicated that the effect of TCM has targeted the stimulation of the host immune response for cytotoxic activity against liver cancer by inhibiting proliferation and promoting apoptosis of tumor cells [7,44], thereby improving survival and alleviating palliative approaches-related side effects in HCC patients [45][46][47].
This meta-analysis summarized evidence on the effects of TCM therapy for HCC patients, on top of conventional treatment. For survival, it is observed that the additional use of TCM significantly improved six-month, one-year, two-year, and three-year survival rates in HCC cases. Additionally, TCM combination therapy could increase PR rate and tRR and reduce PD rate in this population. Given above, results from our study demonstrated add-on benefits of TCM in improving outcomes of HCC patients. As the molecular pathogenesis of HCC is highly associated with multigene, multifactor, and multistep processes and is quite complicated, add-on TCM therapy combined with other therapeutic options has a promising potential for its multilevel, multitarget, and coordinated intervention effects against HCC [43]. Many active compounds from TCM have shown their noticeable potentials in inhibiting the promotion, proliferation, angiogenesis, and metastasis of HCC [43,44] practice. Although the mechanisms of TCM components in anti-HCC were well reviewed before [43], further in-depth mechanistic studies and well-designed clinical trials are warranted. Previous work has suggested that QOL is an important predictor of survival for cancer patients [48]. Although more sophisticated approaches of QOL measurement were developed, the KPS scores are still widely recognized as a tool for the assessment of the functional status of cancer patients and highly reliable [49]. Based on the evidence we identified, TCM combination therapy may be considered as an alternative option to improve QOL in HCC patients. Previously, KPS as a predictor of survival has been demonstrated in patients with different kind of cancers [48,49], and few studies focused on the relationship between KPS scores and HCC survival. Whether KPS has a role in predicting HCC outcomes should be focused on in future studies.
Evidence of this meta-analysis also showed that the combination of TCM and chemotherapy significantly reduced adverse events including leukopenia, thrombocytopenia, anemia or erythropenia, liver injury, and gastrointestinal discomfort in HCC patients. However, because of the toxic effects of chemotherapy and anticancer drugs on normal cells and tissues, anticancer drugs and approaches cause many side effects and adverse events with various symptoms, including hematocytopenia, gastrointestinal discomfort (nausea, vomiting, anorexia, and diarrhea), and liver injury. These side effects often influence patients' QOL and sometimes make the chemotherapy discontinued [50,51]. Consistent with our results, growing evidences suggest that TCM appears to have beneficial effects for prevention and improvement of several chemotherapy-induced side effects [52,53], leading to better outcomes in this population.
This meta-analysis had the following limitations. First, majority of the included studies had small samples, with midto low-quality designs. Second, all included studies were conducted in China. According to our experience, only positive results are published in Chinese medical journals. We cautiously drew the conclusion that publication bias might have been present in this meta-analysis. Third, most included studies failed to address blinding assessment, which may influence the objectivity of HCC outcomes. High-quality, welldesigned, large sample trials focused on the efficacy and safety of TCM therapy for HCC should be performed in the future.
In conclusion, add-on therapy with TCM could improve overall survival, increase clinical tumor responses, and reduce adverse events in hepatocellular carcinoma. Previous surveys indicated that the trend of TCM use in patients with cancer is on the rise. Surveys have also found that many cancer patients were more inclined to use TCM therapies in combination with conventional therapy rather than in lieu of conventional therapy [54]. Thus, investigating the combined use of TCM and conventional therapy in the oncology setting is urgently essential for practitioners. Evidence-based approaches in the clinic have to be supplemented by experimental studies to unravel cellular and molecular modes of action of TCM treatments [45].

Conflicts of Interest
There are no conflicts of interest.

Authors' Contributions
Zongguo Yang and Xian Liao have contributed equally to this work.