Clinical Efficacy and Safety of Aidi Injection Plus Docetaxel-Based Chemotherapy in Advanced Nonsmall Cell Lung Cancer: A Meta-Analysis of 36 Randomized Controlled Trials

Background. Aidi injection is an important adjuvant anticancer drug commonly used in China. Can Aidi injection plus docetaxel-based chemotherapy improve clinical efficacy with good safety in NSCLC? To further reveal its clinical effectiveness, we systematically evaluated all the related studies. Method. We collected all the studies about Aidi injection plus docetaxel-based chemotherapy for NSCLC on Medline, Embase, Web of Science, CNKI, VIP, Wanfang, CBM, CENTRAL, Chi-CTR, and US-clinical trials. We evaluated their methodological bias risk according to the Cochrane evaluation handbook (5.1.0), extracted data following the predesigned data extraction form according to the PICO principle, and synthesized the data using meta-analysis. Results. We included 36 RCTs with 2837 patients, and most studies had unclear bias risk. The merged RR values and their 95% CI of meta-analysis for ORR, DCR, and QOL were as follows: 1.30 (1.19, 1.42), 1.17, (1.12, 1.22), and 1.73 (1.54, 1.95). The merged RR values for neutropenia, thrombocytopenia, anemia, gastrointestinal toxicity, hepatorenal dysfunctions, and alopecia were as follows: 0.70 (0.61, 0.79), 0.63 (0.53, 0.75), 0.60 (0.48, 0.75), 0.76 (0.65, 0.89), 0.56 (0.36, 0.88), and 0.58 (0.36, 0.93). Compared with chemotherapy alone, all differences were statistically significant. Subgroup analysis showed that, with 100 ml, 80-100 ml, and 50 ml, Aidi injection could increase the tumor response and Aidi injection plus DP, DC, and DO could increase the tumor response. Meta-analysis results had good stability. Conclusions. Aidi injection plus docetaxel-based chemotherapy, especially plus DP, DC, and DO, may significantly improve the clinical efficacy and QOL in NSCLC. It may also have low risk of hematotoxicity, gastrointestinal toxicity, and low risk of inducing hepatorenal dysfunctions. Aidi injection may have attenuation and synergistic efficacy to docetaxel chemotherapy. All these need to have new evidence to be proved.


Introduction
Lung cancer is the leading cause of cancer-related mortality around the world with only 15% of 5-years survival rate [1][2][3].
Approximately 80% of lung cancers are nonsmall cell lung cancer (NSCLC). Nevertheless, over 50% of patients with NSCLC have advanced local invasion and metastasis, when they were admitted to the hospital for diagnosis. They must 2 Evidence-Based Complementary and Alternative Medicine receive the systemic chemotherapy, radiotherapy, or chemoradiotherapy because they missed the opportunity for operation [4][5][6]. As first-or second-line chemotherapy, taxane agents including paclitaxel (taxol) and docetaxel (taxotere) are widely used in NSCLC. But they have different acute/subacute toxicity, which results in poor prognosis with only 15% of 5-years survival rate and substandard quality of life (QOL) [7,8]. Therefore, new effective strategies with attenuation and synergistic efficacy are urgently needed.
As Cantharis and Astragalus-based Chinese herbs, Aidi injection (Z52020236, China Food and Drug Administration) is composed of the extracts of Cantharis, Astragalus, Eleutherococcus senticosus, and Ginseng, which appear to have antitumor efficacy and reduce the toxicity [9][10][11][12][13]. Metaanalysis (Wang, Q. 2010) [14] reported that Aidi injection plus paclitaxel or docetaxel and cisplatin could significantly improve the clinical efficiency and QOL in NSCLC. The combination had low risk of neutropenia, thrombocytopenia, and nausea/vomiting, but unclear risk of anemia, hepatotoxicity, nephrotoxicity, neurotoxicity, and alopecia. However, many studies [15][16][17][18] showed that docetaxel and paclitaxel had different clinical manifestations, especially the acute/subacute toxicity. Docetaxel is one of the important first-or second-line chemotherapeutic agents for NSCLC [19][20][21]. And docetaxel-based chemotherapy refers to docetaxel alone or plus cisplatin, carboplatin, oxaliplatin, lobaplatin, or nedaplatin, which are important chemotherapy regimens in NSCLC. The application of Aidi injection plus docetaxel-based chemotherapy was clinically used in a wide range of treatment. Can Aidi injection plus docetaxel-based chemotherapy improve clinical efficacy with satisfying level of safety in NSCLC? Has Aidi injection attenuated and synergistic efficacy to docetaxel-based chemotherapy in NSCLC? Many studies [22][23][24][25] had shown that Aidi injection plus docetaxel-based chemotherapy might improve the clinical efficacy and QOL with low risk of acute/subacute toxicity in NSCLC. However, these conclusions vary in different studies with limited sample size. At present, there is a lack of strong evidence to prove the efficacy of the treatments. Therefore, to further reveal its real clinical efficacy and provide the best evidence for clinical strategies in NSCLC, we systematically evaluated all the related studies.

Materials and Methods
This article followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PRISMA guidelines). Ethical approval was not required, as materials of this study were published or unpublished studies.

Search Strategy. Two reviewers (Chengqiong Wang and Lianhong Li) independently searched articles in Chinese and
English databases using the search strategy (Aidi OR Aidi injection OR Compound cantharis injection OR Compound disodium cantharidinate injection or Addie injection) and the search strategy (Taxoids OR Docetaxel OR Docetaxel OR Taxotere) and the search strategy ("Lung Neoplasms" [ Ongoing studies were retrieved in Chinese clinical trial registry (Chi-CTR) and US-clinical trials (established to September 2017). All retrievals were implemented by using the Mesh and free word. Finally, all related systematic reviews (SRs) or meta-analysis was evaluated, and studies meeting inclusion criteria were selected from the references.

Inclusion and Exclusion Criteria.
Included studies must meet the following criteria. (1) The patients had NSCLC with stages III to IV being diagnosed and confirmed with the histopathological and cytological diagnostic criteria and TNM staging system. (2) There was no severe damage in liver or kidney function in any of the patients. (3) There were randomized controlled trials (RCTs). (4) The experimental group undergone Aidi injection plus docetaxel-based chemotherapy, and the control group undergone docetaxelbased chemotherapy. Docetaxel-based chemotherapy refers to docetaxel alone or plus platinum such as cisplatin, carboplatin, oxaliplatin, lobaplatin, and nedaplatin (DP, DC, DO, DL, and DN). (5) Patients prior to being included in the study have not accepted the radiotherapy, other chemotherapy, or Chinese herbs. (6) Main outcomes included the clinical efficacy and acute/subacute toxicity. Clinical efficacy was evaluated using tumor responses and QOL. (7) No restrictions were set on the follow-up time or types of hospitals.
Excluded studies must meet the following criteria: (1) duplicates, (2) unrelated studies including studies concerning Aidi injection plus paclitaxel chemotherapy, radiotherapy, additional chemotherapeutic agents, other Chinese herbs and other themes, (3) non-RCTs including case control studies and series case reports, (4) abstracts and reviews without specific data and unrelated SRs, and (5) studies without the clinical efficacy, QOL, and acute/subacute toxicity.

Bias Risk Assessment.
According to the Cochrane evaluation handbook of RCTs (5.1.0) [26], we evaluated the bias risk of all trials using the bias parameters such as the random sequence generation (selection bias), the allocation concealment (selection bias), the blinding of participants and personnel (performance bias), the blinding of outcome assessment (detection bias), the incomplete outcome data (attrition bias), the selective reporting (reporting bias), and the other bias (whether the baseline is comparable). We judged each parameter on three levels ("yes" for a low risk of bias, "no" for a high risk of bias, and "unclear"). Then, we assessed the trials and categorized them into three levels: low risk (all items were "yes"), high risk (at least one item was "no"), and unclear risk (at least one item was "unclear").

Main
Outcomes. We measured the tumor response using objective response rate (ORR) and disease control rate (DCR). According to the World Health Organization (WHO) guidelines for solid tumor responses [27] or Response Evaluation Criteria in Solid Tumors (RECIST) [28], indicators were complete response (CR), partial response (PR), no change (NC), progressive disease (PD), ORR being equal to CR plus PR, and DCR being equal to CR plus PR and NC. According to Karnofsky Performance Status scale (KPS scale) [29,30], QOL was considered to be improved if KPS score increased 10 points or higher after treatment. We measured the acute/subacute toxicity using hematotoxicity such as neutropenia (granulocytes < 2 × 10 9 /L), thrombocytopenia (platelets < 100 × 10 9 /L) and anemia (Hemoglobin < 110g/L), liver dysfunction (serum aminotransferase or alkaline phosphatase > 1.25 × N), renal dysfunction (serum urea nitrogen or creatinine > 1.25 × N), hepatorenal dysfunctions, and gastrointestinal toxicity including the gastrointestinal reactions and nausea/vomiting, neurotoxicity (peripheral neuritis), alopecia, rash, phlebitis, and oral mucositis.

Data Extraction.
Two reviewers (Chengqiong Wang and Lianhong Li) independently extracted all the data in a predesigned data extraction form according to the PICO principle. All the data included the first author, the publishing time, the randomization methods, the demographic characteristics, the sample size, the usage of Aidi injection and the types of docetaxel chemotherapy, the evaluation criteria of clinical efficacy and acute or subacute toxicity and the followup information, and main outcomes including the ORR, DCR, QOL, and acute or subacute toxicity. The data were obtained directly from the articles. If insufficient details were reported, authors were contacted for further information.

Statistical Analysis.
Meta-analysis was implemented by two reviewers (Chengqiong Wang and Jing Li) using Review Manager 5.3 (The Cochrane Collaboration, Oxford, UK). The relative risk (RR) and 95% confidence intervals (CI) were calculated. Statistical heterogeneity of the results across trials was assessed by chi-square based Q-statistic test and the consistency was calculated by I 2 . If the homogeneity (P ≥ 0.1, I 2 ≤ 50%) was not rejected, the fixed-effects model (FEM) was used to calculate the summary RR and the 95% CI. Otherwise, the results were calculated by randomeffects model (REM). We performed the subgroup analysis according to different doses of Aidi injection, docetaxelbased chemotherapy and evaluation criteria, which revealed their influence on the tumor responses. Publication bias was evaluated using funnel plots if there were more than 10 included studies. The poor quality studies and studies with over-or underestimated results were important factors that damage the robustness of meta-analysis results. The studies were defined as poor quality studies when they had at least one domain considered as high risk of bias. The overor underestimated studies were identified according to the result of funnel plots and heterogeneity analysis, in which results were statistically different and had positive effects on publication bias or heterogeneity. Therefore, the sensitivity was evaluated through excluding the poor quality studies and studies with overestimated efficacy and underestimated toxicity.

Search
Results. The initial database search identified 286 published studies without ongoing studies using our search strategies (Figure 1). Reading the title and excluding the duplicates, 114 records were included. After reading the abstract, 51 full texts and 2 SRs [14,31]

Methodological Bias Risk.
In 36 studies, nine studies described the random sequence generation using randomized digital table in eight studies [33,43,45,51,55,57,59,61] and lottery in one study [24]. The random allocation concealment was implemented using envelope in one study [34], and other studies did not provide the detailed information about it. None of the studies did provide the detailed information about blinding of participants, personnel, and outcome assessment. All studies had complete outcome data without loss to follow-up. Nine studies [23,36,38,40,47,54,56,57,60] had selective reporting about the acute/subacute toxicity. Except for two studies [37,52], baseline was comparable in other studies. The methodological bias risk of all included studies is presented in Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgments about each risk of bias item for each included study. Bian (Table 2 and Figures S5, 6, and 7).

Subgroup Analysis of ORR and DCR.
Subgroup analysis was performed to reveal the influence of different doses, docetaxel chemotherapy protocols, and evaluation criteria on the ORR and DCR. Drug doses included Aidi injection with 100 ml, 80-100 ml, 80 ml, 60 ml, 50 ml, and 40 ml/time. Subgroup analysis showed that, with 100 ml, 80-100 ml, and 50 ml, Aidi injection could increase the ORR and DCR (Table 3 and Figures S8-9). Docetaxel-based chemotherapy included docetaxel alone, DP, DC, DO, DL, and DN. Subgroup analysis showed that only Aidi injection plus DP, DC, and DO could increase the ORR and DCR (Table 3 and Figures S10-11). Tumor responses were evaluated using WHO or RECIST criteria. Subgroup analysis showed that Aidi injection plus docetaxel-based chemotherapy could increase the ORR and DCR using the WHO or RECIST criteria (Table 3 and Figures  S8-13).

Publication Bias Analysis.
The funnel plots were symmetric in ORR and thrombocytopenia (Figures 6(a) and 6(f)). And there was no publication bias in these studies which objectively reported the results. The funnel plots were asymmetric in DCR, QOL, neutropenia, and gastrointestinal toxicity (Figures 6(b), 6(c), 6(d), and 6(e)). These results indicated that there was publication bias in them. The DCR was underestimated in one study [33]. The QOL was overestimated in one study [49] and underestimated in two studies [52,57]. The neutropenia was overestimated in four studies [33,35,54,59] and the gastrointestinal toxicity was overestimated in four studies [35,39,52,59] and underestimated in one study [41].    (Figures S1-7). high risk of bias and selective reporting about acute/subacute toxicity (Table 4(a)). They had potential effect on robustness of neutropenia, thrombocytopenia, gastrointestinal toxicity, and oral mucositis. Therefore, the sensitivity was evaluated through excluding poor quality studies. After excluding poor quality studies, all results had good consistency. There was statistical heterogeneity in neutropenia, gastrointestinal toxicity and neurotoxicity, and minimal heterogeneity in QOL.

Control groups (Evens/tatol)
There was publication bias in DCR, QOL, neutropenia, and thrombocytopenia. Therefore, the sensitivity was evaluated through excluding the studies with overestimated efficacy or underestimated toxicity. Before and after excluding these studies, results had good consistency (Table 4(b)). In all, this meta-analysis had good stability.

Discussion
Based on previous meta-analysis [14,31]      clinical efficacy in NSCLC? Thirty-four studies [22-24, 32-55, 57-61, 63, 64] involving 2714 patients were included to evaluate the tumor responses. Meta-analysis showed that Aidi injection plus docetaxel-based chemotherapy could significantly improve the ORR and DCR in NSCLC. But there was significant clinical heterogeneity in them. Further subgroup analysis showed that Aidi injection with 100 ml, 80-100 ml, and 50 ml could increase the ORR and DCR and 50 ml was the main dosage. Combined with DP, DC, and DO, Aidi injection could increase the tumor responses. This meta-analysis involved 34 studies with 2714 cases which ensured sufficient sample size for analysis. The DCR was underestimated and the meta-analysis results had good robustness. All these were beneficial to tumor responses. But most studies had unclear bias risk, which weakened the result's reliability. Compared to the previous studies [14,31], this meta-analysis revealed that Aidi injection plus docetaxelbased chemotherapy, especially plus DP, DC, and DO, might significantly improve the ORR and DCR and 50 ml was the main dosage. Our previous meta-analysis [67,68] had shown that Aidi injection plus radiotherapy or gemcitabine and cisplatin (GP) could significantly improve the QOL in patients with lung cancer. Can Aidi injection plus docetaxelbased chemotherapy improve the QOL? To further analyze whether Aidi injection can improve the QOL, 22 studies with 1676 cases were included for analysis. Meta-analysis showed that Aidi injection could significantly improve the QOL. But, QOL was overestimated in one study [49] and underestimated in two studies [52,57]. Sensitivity analysis revealed that QOL had good robustness. But most studies had unclear bias risk. Therefore, we believed that Aidi injection might also improve the QOL. Aidi injection is composed of extracts from Astragalus, Eleutherococcus senticosus, Ginseng, and Cantharis. In vitro studies [69][70][71][72] had shown that cantharidin could induce the tumor cells' apoptosis and inhibit the proliferation, migration, and invasion. Animal studies [73][74][75] had shown that cantharidin or Ginseng could significantly inhibit the growth of malignant tumor cells. Our previous meta-analysis [76] had revealed that Aidi injection could significantly restore the cellular immunity damaged by platinum-based chemotherapy. In addition, many studies [77,78] had shown that Astragalus, senticosus Eleutherococcus, and Ginseng also had antitumor activity and immune regulation functions. These results provided indirect evidence for the above conclusions. In all, we believe that Aidi injection plus docetaxel-based chemotherapy, especially plus DP, DC, and DO, may significantly increase clinical efficacy and improve QOL in patients with NSCLC. The main dose may be 50 ml/time. Results indirectly indicate that Aidi injection may have synergistic efficacy to docetaxel-based chemotherapy. Unfortunately, So far, there was no reliable evidence to confirm the long-term synergistic efficacy.

I
Docetaxel-based chemotherapy has varying degrees of blood, liver, kidney, and gastrointestinal toxicity due to docetaxel plus platinum [79][80][81]. However, can Aidi injection plus docetaxel-based chemotherapy increase the risk of acute/subacute toxicity? To answer this question, 31 studies [22, 23, 32-43, 45-47, 49-55, 57, 59-64] involving 2434 patients were included to reveal the risk of toxicity. Metaanalysis showed that Aidi injection plus docetaxel-based chemotherapy had lower risk of the neutropenia, thrombocytopenia, anemia and gastrointestinal toxicity, hepatorenal dysfunctions, and alopecia compared to that of docetaxelbased chemotherapy alone. And there were no significant differences in liver dysfunction, renal dysfunction, neurotoxicity, rash, phlebitis, and oral mucositis between the two groups. The meta-analysis of neutropenia, thrombocytopenia, and gastrointestinal toxicity had sufficient studies and sample size. But there were limited studies and sample size in other meta-analysis, especially in the meta-analysis of liver and renal dysfunction, which might lead to insufficient assessment. Sensitivity analysis showed that the merged value of neutropenia, thrombocytopenia, and gastrointestinal toxicity had good robustness. Compared to the previous meta-analysis [14,31], this study further revealed that Aidi injection plus docetaxel-based chemotherapy had low risk of the neutropenia, thrombocytopenia, and gastrointestinal toxicity. In addition, we found that it also had low risk of anemia, hepatorenal dysfunctions, and alopecia. Our previous study [67] had shown that Aidi injection plus GP had low risk of hematological and gastrointestinal toxicity and neurotoxicity in NSCLC. Furthermore, Aidi injection could alleviate the radiotherapy related toxicity, such as myelosuppression, radiation pneumonitis, and esophagitis [68]. These results provided indirect clinical evidence for the above conclusions. Zhu X and et al. [82,83] had reported that Astragalus membranaceus injection (AMI) could promote myelopoiesis through improving the hematopoietic microenvironment and relieving the bone marrow cells apoptosis in mice. Hu, W et al. [84][85][86][87] had revealed that ginsenoside Rg1 also had antimyelotoxicity activity and promotion of myelopoiesis through enhancing the antioxidant and antiinflammatory capacities of bone marrow mesenchymal stem cells (BMSCs) in vivo. Liu L and et.al [88] had shown that Astragalus injection ameliorated the cisplatin-induced nephrotoxicity through regulating the Bax and Bcl-2 expression in mice. Other study [89] had shown that ginsenoside Rg1 also had antioxidant activities which ameliorated the cisplatin-induced hepatic injury through Nrf2 signaling pathway in mice. All these revealed that Astragalus and Ginseng could ameliorate chemotherapy related toxicity through enhancing the antimyelotoxicity activity, antiapoptotic, and antioxidant activities. These results provided the basic and mechanism evidence for the above conclusions. In summary, Aidi injection plus docetaxel-based chemotherapy may have low risk of hematotoxicity, gastrointestinal toxicity, and hepatorenal dysfunctions. Based on the optimization of efficacy and safety, results indicated that the optimal dose might be 50 ml/time. These results indirectly reveal that Aidi injection may have attenuation effect to docetaxel related toxicity.
There were some limitations in this study. Firstly, Chinese and English databases were retrieved but not Japanese and Korean databases. All included studies were published in China, which may lead to ethnical bias. Secondly, only 9 studies reported the random allocation method. No studies provided the detailed information about the random allocation concealment and the binding. Nine studies had selective reporting about the acute/subacute toxicity. Third, longterm efficacy had not been evaluated. Fourth, most studies reported the acute/subacute toxicity using WHO standards [27] or NCI-CTC [66]. And there were limited studies and sample size in liver and renal dysfunction, neurotoxicity, and alopecia. All these limitations might lead to an inadequate assessment of the clinical efficacy and safety.

Conclusions
The available evidence indicates that Aidi injection plus docetaxel-based chemotherapy, especially plus DP, DC, and DO, may significantly improve the clinical efficacy and QOL in patients with NSCLC. It may have low risk of hematotoxicity, gastrointestinal toxicity, and hepatorenal dysfunctions.
Results indirectly indicate that Aidi injection may have attenuation and synergistic efficacy to docetaxel chemotherapy. Based on the optimization of efficacy and safety, the results indicated that the optimal dose may be 50 ml/time. Unfortunately, whether Aidi injection can improve long-term efficacy is still unclear. Furthermore, many limitations might lead to an inadequate assessment of the clinical efficacy and safety. Therefore, we look forward to larger scale RCTs or real-world studies for a more thorough review in future publications. Consequently, we hope that this study will provide valuable evidence for Aidi injection as an important supplementary therapy for malignant tumors.

Supplementary Materials
Meta-analysis results of acute or chronic toxicity ( Figures  S1-7). Figure S1: the analysis of neutropenia between the two groups. Meta-analysis showed that Aidi injection plus docetaxel-based chemotherapy had low risk of neutropenia [RR = 0.70, 95% CI (0.61, 0.79), and P < 0.00001] using random-effects model. Figure S2: the analysis of thrombocytopenia between the two groups. Meta-analysis showed that Aidi injection plus docetaxel-based chemotherapy had low risk of thrombocytopenia [RR = 0.63, 95% CI (0.53, 0.75), and P < 0.00001] using fixed-effects model. Figure S3: the analysis of anemia between the two groups. Meta-analysis showed that Aidi injection plus docetaxel-based chemotherapy had low risk of anemia [RR = 0.60, 95% CI (0.48, 0.75), and P < 0.00001] using fixed-effects model. Figure S4: the analysis of gastrointestinal toxicity between the two groups. Metaanalysis showed that Aidi injection plus docetaxel-based chemotherapy had low risk of gastrointestinal toxicity [RR = 0.76, 95% CI (0.65, 0.89), and P = 0.0006] using randomeffects model. Figure S5: the analysis of hepatorenal dysfunctions between the two groups. Meta-analysis showed that Aidi injection plus docetaxel-based chemotherapy had low risk of hepatorenal dysfunctions [RR = 0.56, 95% CI (0.36, 0.88), and P = 0.01] using fixed-effects model. But there were no statistically significant differences in liver dysfunction [RR = 0.69, 95% CI (0.47, 1.01), and P = 0.05], renal dysfunction [RR = 0.56, 95% CI (0.31, 1.00), and P = 0.05] between two groups. Figure S6: the analysis of neurotoxicity between the two groups. There were no statistically significant differences in neurotoxicity [RR = 0.65, 95% CI (0.35, 1.18), and P = 0.16] between two groups. Figure S7: the analysis of other toxicity between the two groups. Meta-analysis showed that Aidi injection plus docetaxel-based chemotherapy had low risk of alopecia [RR = 0.58, 95% CI (0.36, 0.93), and P = 0.02] using fixed-effects model. But there were no statistically significant differences in rash [RR = 0.75, 95% CI (0.38, 1.49), and P = 0.42], phlebitis [RR = 1.00, 95% CI (0.63, 1.59), and P = 1.00], and oral mucositis [RR = 0.64, 95% CI (0.38, 1.09), and P = 0.10] between two groups. Subgroup analysis results of ORR and DCR (Figures S8-13). Figure S8: subgroup analysis of ORR via drug doses. Subgroup analysis showed that with 100 ml, 80-100 ml, and 50 ml, Aidi injection could all increase the ORR. Figure S9: subgroup analysis of DCR via drug doses. Subgroup analysis showed that with 100 ml, 80-100 ml and 50 ml, Aidi injection could all increase the DCR. Figure S10: subgroup analysis of ORR via docetaxelbased chemotherapy. Subgroup analysis showed that only Aidi injection plus DP, DC, and DO could increase the DCR. Figure S11: subgroup analysis of DCR via docetaxelbased chemotherapy. Subgroup analysis showed that only Aidi injection plus DP, DC, and DO could increase the DCR. Figure S12: subgroup analysis of ORR via evaluation criteria. Subgroup analysis showed that Aidi injection plus docetaxel-based chemotherapy could increase the ORR using the WHO or RECIST criteria. Figure S13: subgroup analysis of DCR via evaluation criteria. Subgroup analysis showed that Aidi injection plus docetaxel-based chemotherapy could increase the DCR using the WHO or RECIST criteria. (Supplementary Materials)