Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The Role of Myeloid-Derived Suppressor Cells in Patients with Solid Tumors: A Meta-Analysis

  • Shuo Zhang ,

    Contributed equally to this work with: Shuo Zhang, Xuelei Ma, Chenjing Zhu

    Affiliation State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University, Chengdu, PR China

  • Xuelei Ma ,

    Contributed equally to this work with: Shuo Zhang, Xuelei Ma, Chenjing Zhu

    Affiliation State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University, Chengdu, PR China

  • Chenjing Zhu ,

    Contributed equally to this work with: Shuo Zhang, Xuelei Ma, Chenjing Zhu

    Affiliation State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University, Chengdu, PR China

  • Li Liu,

    Affiliation State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University, Chengdu, PR China

  • Guoping Wang,

    Affiliation State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University, Chengdu, PR China

  • Xia Yuan

    Affiliation State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University, Chengdu, PR China

The Role of Myeloid-Derived Suppressor Cells in Patients with Solid Tumors: A Meta-Analysis

  • Shuo Zhang, 
  • Xuelei Ma, 
  • Chenjing Zhu, 
  • Li Liu, 
  • Guoping Wang, 
  • Xia Yuan


Targeting immune cells or factors are effective for patients with solid tumors. Myeloid-derived suppressor cells (MDSCs) are known to have immunosuppressive functions, and the levels of MDSCs in patients with solid tumor are assumed to have prognostic values. This meta-analysis aimed at evaluating the relationship between MDSCs and the prognosis of patients with solid tumors. We searched articles in PUBMED and EMBASE comprehensively, updated to March 2016. Eight studies with 442 patients were included in the meta-analysis. We analyzed pooled hazard ratios (HRs) for overall survival (OS), disease-free survival (DFS) and progression-free survival (PFS). The results showed that MDSCs were associated with poor OS (HR, 1.94; 95% confidence interval [CI], 1.42–2.66; P < 0.0001) in patients with solid tumors. PFS/RFS (HR, 1.85; 95% CI, 1.16–2.97; P = 0.01) also indicated the association between MDSCs and prognosis. The HRs and 95% CIs for OS in Asian and non-Asian patients were 2.53 (95% CI 1.61–3.42, p < 0.00001) and 1.67 (95% CI 1.14–2.46, p < 0.0001), respectively. We further analyzed the data according to tumor types. The combined HRs and 95% CIs for OS were 1.26 (95% CI 1.10–1.44, p = 0.0003) for gastrointestinal (GI) cancer, 2.59 (95% CI 1.69–3.98, p < 0.0001) for hepatocellular carcinoma (HCC) and 1.86 (95% CI 1.26–2.75, p = 0.002) for other tumor types. In conclusion, MDSCs had a fine prognostic value for OS and PFS/RFS in patients with solid tumors. MDSCs could be used as biomarkers to evaluate prognosis in clinical practice.


The incidences of various solid tumors such as gastrointestinal (GI) cancer and breast cancer (BC) are increasing every year [1] and solid tumors are regarded as one of the most frequent causes of death worldwide [26]. Current therapies for different solid tumors include surgical resection, chemotherapy, radiotherapy and immunotherapy [1, 7, 8]. Recently, targeted immunotherapy such as cancer vaccines and monoclonal antibodies has been demonstrated to improve anti-tumor immune responses and may be beneficial for patients with different types of cancers [9, 10]. However, it was reported that cancer-related immune-suppression restricts the effects of immunotherapy [11].

Currently, it has been demonstrated that immune cells in tumor microenvironment help to form an immunosuppressive network, which plays a key role in the suppression of antitumor immune system, and finally leads to tumor invasion [1214]. Myeloid-derived suppressor cells (MDSCs) are regarded as a heterogeneous population of immature myeloid cells with CD11b+CD33+HLA-DR−/low phenotype, which include granulocytic CD14−CD15+ and monocytic CD14+CD15− subtypes [1517]. MDSCs, as immunosuppressive cell subjects, have been reported to play a critical role in mediating immune suppression by inhibiting both the innate and adaptive immunity [1820] and preventing anticancer immunity function of cancer vaccines [13, 14, 21]. MDSCs inhibit the functions of immune cells through activating oxygen species The role of myeloid cells in the promotion of tumour angiogenesis [16, 17] and producing cytokines such as IL-6 and IL-4. Recently, some researches have reported that MDSCs limit the accumulation of T cells in both mice and human models with various cancers such as HCC [22, 23]. Therefore, targeting MDSCs which potentially stimulate anti-tumor immune system [24] may improve the effects of anti-cancer therapies [17, 25].

MDSCs are thought to have prognostic significance in patients with solid tumors [14, 26, 27]. At present, there is a heated controversy of MDSC on its prognostic significance. Thus a meta-analysis to investigate the prognostic value of MDSCs in patients with solid tumors is urgent.

Materials and Methods

Search Strategy

We performed a comprehensive search in PUBMED and EMBASE databases for all available studies published up to March 2016 to evaluate the prognostic value of MDSCs in patients with solid tumors. We used the following search terms: “Myeloid-derived suppressor cells OR MDSC” and “prognosis OR prognostic OR survival OR outcome” and “cancer OR tumor OR carcinoma OR neoplasm”. We also manually scanned the references of included articles in order to check more relevant studies. Our study was performed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [28].

Inclusion and Exclusion Criteria

The eligible studies in this meta-analysis must meet the following inclusion criteria: (1) published in English, (2) investigated patients with solid tumor, (3) contained information of level of MDSCs, (4) estimated the relationship between MDSCs level and survival outcomes. Articles were excluded with any of the following features: 1) studies without enough information to estimate HR and 95% CI; 2) studies had duplicate or overlapping data; 3) studies were not demonstrated in English.

Data Extraction

Two reviewers extracted the required data from all available studies independently. Titles and abstracts were reviewed to identify potential available articles, and full texts were obtained for more details. We extracted the following contents: the first author’s name and country, publication year, number of patients, subtypes of MDSCs, cut-off value, survival analysis and the HRs of MDSCs for OS and PFS/RFS. If the HR and its 95% CIs could not be obtained directly, they were estimated from the corresponding data or Kaplan-Meier curves extracted from the studies according to the methods reported by Parmar et al [29]. Any discrepancies were resolved by consulting with a third author until consensus was reached.

Statistical Analysis

HRs and the corresponding 95% CIs were calculated to estimate the association between MDSCs and patients’ prognosis according to Tierney’s method [30]. The heterogeneity of combined HRs was evaluated by Cochran’s Q test and Higgin’s I2 statistics [31, 32]. A P value < 0.05 and/or I2 >50% [31] indicated substantial heterogeneity among studies, and a random-effect model was used (DerSimoniane-Laird method) to calculate the combined HR; otherwise, a fixed-effect model (Mantel-Haenszel method) was used [33]. Because elements such as tumor types, region, number of patients and cut-off value may affect outcomes of this meta-analysis, we performed subgroup analyses. In general, if the 95% CI for the combined HR did not overlap one, pooled HR > 1 was thought to suggest a significant relationship with poor prognosis. A sensitivity analysis was conducted by deleting one study at a time to examine its effect on the pooled results. Publication bias was estimated using funnel plots qualitatively with the standard error [34], and evaluated by Begg’s and Egger’s test. All the analyses were carried out by STATA statistical software version 11.0 (StataCorp LP, College Station, TX, USA) and Review Manager Version 5.0 (Copenhagen: The Nordic Cochrane Centre: The Cochrane Collaboration, 2008).

Quality Assessment

According to the nine-star Newcastle-Ottawa scale (NOS) [35], the quality of each study was strictly assessed in three aspects: selection (four points), comparability (two points), and outcome assessment (three points). A nine-point score is regarded as the highest score. Based on quality assessment standards from published meta-analyses [36], a trial with five or more points was identified as high quality. Articles with less than five points will not be retrieved in order to ensure the quality of the meta-analysis. Any ambiguity or differences in quality evaluation were reviewed and solved together by two authors.


Study Selection and Characteristics

As is shown in Fig 1, a total of 181 potentially relevant studies were retrieved according to the search methods. After the evaluation of titles and abstracts manually, forty-seven articles were excluded for the reasons shown in Fig 1. Full-text articles of the remaining 24 were assessed and 17 articles were further excluded due to the lack of essential data for estimating HR. Finally, a total of 7 studies including 442 patients were available for the meta-analysis.

Additionally, two articles [37, 38] studied the prognosis of patients before and after therapy. One [39] investigated two different types of cancer which were colorectal cancer and breast cancer and we retrieved the data of the two tumors separately.

The main features of the eligible studies were summarized in Table 1. The total number of patients was 442, ranging from 25 to 131 in each study. In total, OS data were available from 6 trials [13, 37, 39, 40], and 3 studies had data for PFS or RFS [40].

Overall Survival

The association between MDSCs and prognosis was shown in Figs 2 and 3. In total, elevated MDSCs predicted poor outcomes in patients with solid tumors. The combined HR was 1.94 (95%CI: 1.42–2.66, P < 0.0001) for OS with a random-effect model due to the significant heterogeneity (I2 = 59%, P = 0.02). PFS/RFS was 1.85 (95%CI: 1.16–2.97, P = 0.01) calculated using a fixed model (I2 = 5%, P = 0.35).

Fig 2. Meta-analysis of the association between MDSCs and OS in patients with solid tumors.

Results are presented as individual and pooled hazard ratio (HR) with 95% confidence intervals (CIs) using a random-effect model.

Fig 3. Meta-analysis of the association between MDSCs and PFS/RFS.

Results are presented as individual and pooled hazard ratio (HR) with 95% confidence intervals (CIs) using a fixed-effect model.

Subgroup Analysis

To explore the origin of heterogeneity, subgroup analyses for OS were preformed based on tumor types, region, number of patients (number < 50 or ≥50) and cut-off value (cut-off ≥10% or ≤10%). The results (see Table 2) showed that HRs and 95% CIs for OS in GI cancer, HCC and other types of tumors were 1.26 (1.10–1.44), 2.59 (1.69–3.98) and 1.86 (1.26–2.75), respectively. The HRs and 95% CIs for OS in patients <50 group was 2.317 (1.59–3.54) and 1.70 (1.15–2.50) in patients ≥ 50 group. In the cut-off ≥10% group the HR was 2.11 (1.55–2.86) and in cut-off <10% group was 1.72 (1.09–2.71). It showed that MDSCs had a stronger prognostic value for OS in cut-off ≥10% group. In addition, we grouped the studies by patients’ ethnicity, and the HRs and 95% CIs for OS in Asia and non-Asia areas were 2.53 (1.61–3.42) and 1.67 (1.14–2.46), respectively.

Table 2. Stratified analyses of MDSCs on overall survival in patients with solid tumors.

Publication Bias and Sensitivity Analysis

Both Begg’s funnel plot and the Egger’s test were conducted to assess the publication bias of included trials. As is shown in Fig 4, the shape of the funnel plots presented no significant asymmetry. The P values of Begger’s test for OS and PFS/RFS were 0.621 and 0.317, respectively, revealing that no obvious publication bias existed in this study. In order to evaluate the impact of each individual study, we conducted the sensitivity analysis on the pooled HRs for the OS or PFS/RFS. It turned out that there were no significant effects of individual study on the combined HRs. The results indicated that the outcomes of this meta-analysis were reliable.

Fig 4. Summary of Begg’s funnel plots of publication bias for OS in all patients.


Up to now, no meta-analysis has been conducted to assess the prognostic significance of MDSCs. Our combined results demonstrated that elevated MDSCs had a poor outcome in cancer patients.

MDSCs as a significant immune regulator suppress both innate and adaptive immunity and accelerate tumor progression [17, 41]. Due to their remarkable roles in the inhibition of resistance to clinical therapies, it has been assumed that limiting MDSC-mediated immunosuppression may activate antitumor immune response [42]. Some studies have demonstrated that through depletion of some amino acids such as arginine and cysteine, and improve inhibitory cytokines such as IL-10 and IL-12, MDSCs can cause suppression of immune cells and stimulate immune regulators such as tumor-associated macrophages (TAMs) [19, 41, 43, 44].

Considering the function of immunosuppressing, it seems that MDSCs play an important role in tumor growth and contribute to limiting the efficacy of anti-cancer therapies [17, 27, 45, 46]. Thus, it is conceivable that by helping tumor cells against immune system, MDSCs cause higher tumor relapse, deterioration and mortality [47]. There is growing evidence that higher MDSCs are associated with worse outcome. Some studies have evaluated the involvement of MDSCs in the progression of cancer patients, such as CRC [39], HCC [37, 38], GI cancers [13] and so on. What’s more, some experiments have found that targeting MDSCs can actually have positive effect on the angiogenesis of HCC [48, 49].

In our meta-analysis, there was obvious heterogeneity for OS (I2 = 59%, P = 0.02) among available studies. Therefore we adopted a random-effect model to calculate combined subgroup data. The heterogeneity may be caused by different features of included patients. In addition, different cancer types including HCC, CRC and so on may also contribute to the heterogeneity.

In order to find out causes of heterogeneity, subgroup analyses were conducted according to the country of patients, tumor types, number of patients and cut-off value. When we grouped the analysis based on tumor types, the heterogeneity for OS decreased. It implied that tumor types may contribute to heterogeneity. Also the HRs and 95% CIs for OS in patients in Asia and non-Asia were different, indicating that the prognostic role of MDSCs for OS of solid tumors was more significant in Asian group, and MDSCs also had a stronger prognostic value for OS with cut-off ≥10%. All the subgroup analysis had positive consequences.

There were some limits in this meta-analysis. First of all, the sample size of each type of cancers were relatively small, and more studies with large sample size were needed. Secondly, because some HRs could not be extracted directly from articles, we calculated them according to Kaplan-Meier survival curves which may make the results less reliable. Moreover, different cut-off value in the studies may also contribute to inter-study heterogeneity, but we carried out subgroup sensitivity analyses to overcome the shortcoming. Fourth, obvious heterogeneity existed because of different population characteristics or study designs. Finally, there may be some unavoidable bias because positive results tended to be published than negative ones, which may lead to the exaggeration of the correlation between MDSCs and poor prognosis. To avoid this, we examined the bias by excluding one study at a time.

In summary, results of this meta-analysis demonstrated that MDSCs were associated with poor prognosis in patients with solid tumor, and its prognostic role for OS was more significant in asian group. What’s more, MDSCs had a stronger prognostic value for OS with cut-off ≥10%. Therefore, MDSCs could be used as biomarkers to evaluate prognosis in clinical practice.

Supporting Information

S1 Checklist. PRISMA-IPD Checklist of items to include when reporting a systematic review and meta-analysis of individual participant data (IPD) [50].


Author Contributions

  1. Conceptualization: XM.
  2. Data curation: XM.
  3. Formal analysis: SZ.
  4. Funding acquisition: XM.
  5. Investigation: SZ.
  6. Methodology: CZ.
  7. Resources: XY.
  8. Software: LL.
  9. Supervision: XM.
  10. Validation: GW.
  11. Visualization: XM.
  12. Writing – original draft: SZ.
  13. Writing – review & editing: CZ.


  1. 1. Vanneman M, Dranoff G. Combining immunotherapy and targeted therapies in cancer treatment. (1474–1768 (Electronic)). doi: D—NLM: NIHMS562329.D - NLM: PMC3967236 EDAT- 2012/03/23 06:00 MHDA- 2012/05/16 06:00 CRDT- 2012/03/23 06:00 AID—nrc3237 [pii] AID—doi: PST—epublish. 22437869
  2. 2. Pourhoseingholi MA, Vahedi M, Baghestani AR. Burden of gastrointestinal cancer in Asia; an overview. (2008–2258 (Print)). doi: D—NLM: PMC4285928 OTO—NOTNLM.
  3. 3. Rasch S, Algul H. A clinical perspective on the role of chronic inflammation in gastrointestinal cancer. (1178–7023 (Electronic)). doi: D—NLM: PMC4134025 OTO—NOTNLM.
  4. 4. Ferlay J, Soerjomataram I Fau—Dikshit R, Dikshit R Fau—Eser S, Eser S Fau—Mathers C, Mathers C Fau—Rebelo M, Rebelo M Fau—Parkin DM, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. (1097–0215 (Electronic)).
  5. 5. Prasad S, Tyagi AK. Ginger and its constituents: role in prevention and treatment of gastrointestinal cancer. (1687–6121 (Print)). doi: D—NLM: PMC4369959 EDAT- 2015/04/04 06:00 MHDA- 2015/04/04 06:01 CRDT- 2015/04/04 06:00 PHST- 2014/10/29 [received] PHST- 2015/01/20 [revised] PHST- 2015/02/16 [accepted] PHST- 2015/03/08 [epublish] AID—doi: PST—ppublish. 25838819
  6. 6. Rozen P. Cancer of the gastrointestinal tract: early detection or early prevention? (0959–8278 (Print)).
  7. 7. Okawa T, Kita-Okawa M. [The role and future of radiotherapy in cancer treatment]. (0385–0684 (Print)).
  8. 8. Romond EH, Perez Ea Fau—Bryant J, Bryant J Fau—Suman VJ, Suman Vj Fau—Geyer CE, Jr., Geyer Ce Jr Fau—Davidson NE, Davidson Ne Fau—Tan-Chiu E, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. (1533–4406 (Electronic)).
  9. 9. Soliman H. Developing an effective breast cancer vaccine. (1526–2359 (Electronic)).
  10. 10. Soliman H. Immunotherapy strategies in the treatment of breast cancer. (1526–2359 (Electronic)).
  11. 11. Stewart TJ, Smyth MJ. Improving cancer immunotherapy by targeting tumor-induced immune suppression. (1573–7233 (Electronic)).
  12. 12. Sun HL, Zhou X Fau—Xue Y-F, Xue Yf Fau—Wang K, Wang K Fau—Shen Y-F, Shen Yf Fau—Mao J-J, Mao Jj Fau—Guo H-F, et al. Increased frequency and clinical significance of myeloid-derived suppressor cells in human colorectal carcinoma. (2219–2840 (Electronic)). doi: D—NLM: PMC3391769 OTO—NOTNLM.
  13. 13. Gabitass RF, Annels Ne Fau—Stocken DD, Stocken Dd Fau—Pandha HA, Pandha Ha Fau—Middleton GW, Middleton GW. Elevated myeloid-derived suppressor cells in pancreatic, esophageal and gastric cancer are an independent prognostic factor and are associated with significant elevation of the Th2 cytokine interleukin-13. (1432–0851 (Electronic)). doi: D—NLM: PMC3176406 EDAT- 2011/06/07 06:00 MHDA- 2011/11/04 06:00 CRDT- 2011/06/07 06:00 PHST- 2010/12/17 [received] PHST- 2011/04/22 [accepted] PHST- 2011/06/05 [aheadofprint] AID—doi: PST—ppublish. 21644036
  14. 14. Lindau D, Gielen P Fau—Kroesen M, Kroesen M Fau—Wesseling P, Wesseling P Fau—Adema GJ, Adema GJ. The immunosuppressive tumour network: myeloid-derived suppressor cells, regulatory T cells and natural killer T cells. (1365–2567 (Electronic)). doi: D—NLM: PMC3575763 EDAT- 2012/12/12 06:00 MHDA- 2013/03/12 06:00 CRDT- 2012/12/11 06:00 PHST- 2012/07/03 [received] PHST- 2012/10/25 [revised] PHST- 2012/10/29 [accepted] AID—doi: PST—ppublish. 23216602
  15. 15. Gabrilovich DI, Ostrand-Rosenberg S Fau—Bronte V, Bronte V. Coordinated regulation of myeloid cells by tumours. (1474–1741 (Electronic)). doi: D—NLM: NIHMS442658 D—NLM: PMC3587148 EDAT- 2012/03/23 06:00 MHDA- 2012/07/10 06:00 CRDT- 2012/03/23 06:00 AID—nri3175 [pii] AID—doi: PST—epublish. 22437938
  16. 16. Murdoch C, Muthana M Fau—Coffelt SB, Coffelt Sb Fau—Lewis CE, Lewis CE. The role of myeloid cells in the promotion of tumour angiogenesis. (1474–1768 (Electronic)).
  17. 17. Ostrand-Rosenberg S. Myeloid-derived suppressor cells: more mechanisms for inhibiting antitumor immunity. (1432–0851 (Electronic)). doi: D—NLM: NIHMS484892 D—NLM: PMC3706261 EDAT- 2010/04/24 06:00 MHDA- 2010/08/17 06:00 CRDT- 2010/04/24 06:00 PHST- 2010/03/10 [received] PHST- 2010/04/01 [accepted] PHST- 2010/04/23 [aheadofprint] AID—doi: PST—ppublish. 20414655
  18. 18. Mirghorbani M, Van Gool S Fau—Rezaei N, Rezaei N. Myeloid-derived suppressor cells in glioma. (1744–8360 (Electronic)).
  19. 19. Ostrand-Rosenberg S, Sinha P. Myeloid-derived suppressor cells: linking inflammation and cancer. (1550–6606 (Electronic)). doi: D—NLM: NIHMS166179 D—NLM: PMC2810498 EDAT- 2009/04/04 09:00 MHDA- 2009/04/25 09:00 CRDT- 2009/04/04 09:00 AID—182/8/4499 [pii] AID—doi: PST—ppublish. 19342621
  20. 20. Youn JI, Gabrilovich DI. The biology of myeloid-derived suppressor cells: the blessing and the curse of morphological and functional heterogeneity. (1521–4141 (Electronic)). doi: D—NLM: NIHMS354073 D—NLM: PMC3277452 EDAT- 2010/11/10 06:00 MHDA- 2010/12/18 06:00 CRDT- 2010/11/10 06:00 AID—doi: PST—ppublish. 21061430
  21. 21. Duffy A, Zhao F Fau—Haile L, Haile L Fau—Gamrekelashvili J, Gamrekelashvili J Fau—Fioravanti S, Fioravanti S Fau—Ma C, Ma C Fau—Kapanadze T, et al. Comparative analysis of monocytic and granulocytic myeloid-derived suppressor cell subsets in patients with gastrointestinal malignancies. (1432–0851 (Electronic)).
  22. 22. Weinstein EJ, Kitsberg Di Fau—Leder P, Leder P. A mouse model for breast cancer induced by amplification and overexpression of the neu promoter and transgene. (1076–1551 (Print)). doi: D—NLM: PMC1949912 EDAT- 2000/05/10 09:00 MHDA- 2000/06/17 09:00 CRDT- 2000/05/10 09:00 PST—ppublish.
  23. 23. Habibi M, Kmieciak M Fau—Graham L, Graham L Fau—Morales JK, Morales Jk Fau—Bear HD, Bear Hd Fau—Manjili MH, Manjili MH. Radiofrequency thermal ablation of breast tumors combined with intralesional administration of IL-7 and IL-15 augments anti-tumor immune responses and inhibits tumor development and metastasis. (1573–7217 (Electronic)). doi: D—NLM: NIHMS48581 D—NLM: PMC2649692 EDAT- 2008/04/22 09:00 MHDA- 2009/08/26 09:00 CRDT- 2008/04/22 09:00 PHST- 2008/03/05 [received] PHST- 2008/04/10 [accepted] PHST- 2008/04/20 [aheadofprint] AID—doi: PST—ppublish. 18425677
  24. 24. Baek JY, Hur W Fau—Wang JS, Wang Js Fau—Bae SH, Bae Sh Fau—Yoon SK, Yoon SK. Selective COX-2 inhibitor, NS-398, suppresses cellular proliferation in human hepatocellular carcinoma cell lines via cell cycle arrest. (1007–9327 (Print)). doi: D—NLM: PMC4146990 EDAT- 2007/04/25 09:00 MHDA- 2007/07/18 09:00 CRDT- 2007/04/25 09:00 PST—ppublish.
  25. 25. Albeituni SH, Ding C Fau—Yan J, Yan J. Hampering immune suppressors: therapeutic targeting of myeloid-derived suppressor cells in cancer. (1540-336X (Electronic)). doi: D—NLM: NIHMS542799 D—NLM: PMC3902636 EDAT- 2013/11/26 06:00 MHDA- 2014/09/10 06:00 CRDT- 2013/11/26 06:00 AID—doi: AID—00130404-201311000-00006 [pii] PST—ppublish. 24270348
  26. 26. Markowitz J, Wesolowski R Fau—Papenfuss T, Papenfuss T Fau—Brooks TR, Brooks Tr Fau—Carson WE, 3rd, Carson WE, 3rd. Myeloid-derived suppressor cells in breast cancer. (1573–7217 (Electronic)). doi: D—NLM: NIHMS502408 D—NLM: PMC3773691 EDAT- 2013/07/06 06:00 MHDA- 2014/02/11 06:00 CRDT- 2013/07/06 06:00 PHST- 2013/06/03 [received] PHST- 2013/06/20 [accepted] PHST- 2013/07/05 [aheadofprint] AID—doi: PST—ppublish. 23828498
  27. 27. Quante M, Varga J Fau—Wang TC, Wang Tc Fau—Greten FR, Greten FR. The gastrointestinal tumor microenvironment. (1528–0012 (Electronic)). doi: D—NLM: NIHMS466894 D—NLM: PMC4012393 EDAT- 2013/04/16 06:00 MHDA- 2013/12/16 06:00 CRDT- 2013/04/16 06:00 PHST- 2013/01/29 [received] PHST- 2013/03/27 [revised] PHST- 2013/03/28 [accepted] PHST- 2013/04/10 [aheadofprint] AID - S0016-5085(13)00501-5 [pii] AID—doi: PST—ppublish. 23583733
  28. 28. Moher D, Shamseer L Fau—Clarke M, Clarke M Fau—Ghersi D, Ghersi D Fau—Liberati A, Liberati A Fau—Petticrew M, Petticrew M Fau—Shekelle P, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. (2046–4053 (Electronic)). doi: D—NLM: PMC4320440.
  29. 29. Parmar MK, Torri V Fau—Stewart L, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. (0277–6715 (Print)).
  30. 30. Tierney JF, Stewart La Fau—Ghersi D, Ghersi D Fau—Burdett S, Burdett S Fau—Sydes MR, Sydes MR. Practical methods for incorporating summary time-to-event data into meta-analysis. (1745–6215 (Electronic)). doi: D—NLM: PMC1920534 EDAT- 2007/06/09 09:00 MHDA- 2007/06/09 09:01 CRDT- 2007/06/09 09:00 PHST- 2006/09/25 [received] PHST- 2007/06/07 [accepted] PHST- 2007/06/07 [aheadofprint] AID—1745-6215-8-16 [pii] AID—doi: PST—epublish. 17555582
  31. 31. Higgins JP, Thompson Sg Fau—Deeks JJ, Deeks Jj Fau—Altman DG, Altman DG. Measuring inconsistency in meta-analyses. (1756–1833 (Electronic)). doi: D—NLM: PMC192859 EDAT- 2003/09/06 05:00 MHDA- 2003/09/16 05:00 CRDT- 2003/09/06 05:00 AID—doi: AID—327/7414/557 [pii] PST—ppublish. 12958120
  32. 32. Dickersin K, Berlin JA. Meta-analysis: state-of-the-science. (0193-936X (Print)).
  33. 33. DerSimonian R Fau—Laird N, Laird N. Meta-analysis in clinical trials. (0197–2456 (Print)).
  34. 34. Sterne JA, Egger M. Funnel plots for detecting bias in meta-analysis: guidelines on choice of axis. (0895–4356 (Print)).
  35. 35. Taggart DP, D'Amico R Fau—Altman DG, Altman DG. Effect of arterial revascularisation on survival: a systematic review of studies comparing bilateral and single internal mammary arteries. (0140–6736 (Print)).
  36. 36. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. (1573–7284 (Electronic)).
  37. 37. Wang D, An G, Xie S, Yao Y, Feng G. The clinical and prognostic significance of CD14+HLA-DR-/low myeloid-derived suppressor cells in hepatocellular carcinoma patients receiving radiotherapy. (1423–0380 (Electronic)).
  38. 38. Arihara F, Mizukoshi E Fau—Kitahara M, Kitahara M Fau—Takata Y, Takata Y Fau—Arai K, Arai K Fau—Yamashita T, Yamashita T Fau—Nakamoto Y, et al. Increase in CD14+HLA-DR -/low myeloid-derived suppressor cells in hepatocellular carcinoma patients and its impact on prognosis. (1432–0851 (Electronic)).
  39. 39. Solito S, Falisi E Fau—Diaz-Montero CM, Diaz-Montero Cm Fau—Doni A, Doni A Fau—Pinton L, Pinton L Fau—Rosato A, Rosato A Fau—Francescato S, et al. A human promyelocytic-like population is responsible for the immune suppression mediated by myeloid-derived suppressor cells. (1528–0020 (Electronic)). doi: D—NLM: PMC3709641 EDAT- 2011/07/08 06:00 MHDA- 2011/10/25 06:00 CRDT- 2011/07/08 06:00 PHST- 2011/07/06 [aheadofprint] AID—blood-2010-12-325753 [pii] AID—doi: PST—ppublish. 21734236
  40. 40. Romano A, Parrinello Nl Fau—Vetro C, Vetro C Fau—Forte S, Forte S Fau—Chiarenza A, Chiarenza A Fau—Figuera A, Figuera A Fau—Motta G, et al. Circulating myeloid-derived suppressor cells correlate with clinical outcome in Hodgkin Lymphoma patients treated up-front with a risk-adapted strategy. (1365–2141 (Electronic)).
  41. 41. Gabrilovich DI. Editorial: The intricacy of choice: can bacteria decide what type of myeloid cells to stimulate? (1938–3673 (Electronic)). doi: D—NLM: PMC4197560 OTO—NOTNLM.
  42. 42. Zhang QW, Liu L Fau—Gong C-y, Gong Cy Fau—Shi H-s, Shi Hs Fau—Zeng Y-h, Zeng Yh Fau—Wang X-z, Wang Xz Fau—Zhao Y-w, et al. Prognostic significance of tumor-associated macrophages in solid tumor: a meta-analysis of the literature. (1932–6203 (Electronic)). doi: D—NLM: PMC3532403 EDAT- 2013/01/04 06:00 MHDA- 2013/06/19 06:00 CRDT- 2013/01/04 06:00 PHST- 2012/06/19 [received] PHST- 2012/10/29 [accepted] PHST- 2012/12/28 [epublish] AID—doi: AID—PONE-D-12-17794 [pii] PST—ppublish. 23284651
  43. 43. Shen P, Wang A Fau—He M, He M Fau—Wang Q, Wang Q Fau—Zheng S, Zheng S. Increased circulating Lin(-/low) CD33(+) HLA-DR(-) myeloid-derived suppressor cells in hepatocellular carcinoma patients. (1386–6346 (Print)).
  44. 44. Mundy-Bosse BL, Young Gs Fau—Bauer T, Bauer T Fau—Binkley E, Binkley E Fau—Bloomston M, Bloomston M Fau—Bill MA, Bill Ma Fau—Bekaii-Saab T, et al. Distinct myeloid suppressor cell subsets correlate with plasma IL-6 and IL-10 and reduced interferon-alpha signaling in CD4(+) T cells from patients with GI malignancy. (1432–0851 (Electronic)). doi: D—NLM: NIHMS425282D - NLM: PMC3521517 EDAT- 2011/05/24 06:00 MHDA- 2011/12/13 00:00 CRDT- 2011/05/24 06:00 PHST- 2011/02/28 [received] PHST- 2011/04/25 [accepted] PHST- 2011/05/21 [aheadofprint] AID—doi: PST—ppublish. 21604071
  45. 45. Galon J Fau—Pages F, Pages F Fau—Marincola FM, Marincola Fm Fau—Thurin M, Thurin M Fau—Trinchieri G, Trinchieri G Fau—Fox BA, Fox Ba Fau—Gajewski TF, et al. The immune score as a new possible approach for the classification of cancer. (1479–5876 (Electronic)). doi: D—NLM: PMC3269368 EDAT- 2012/01/05 06:00 MHDA- 2012/05/19 06:00 CRDT- 2012/01/05 06:00 PHST- 2011/12/20 [received] PHST- 2012/01/03 [accepted] PHST- 2012/01/03 [aheadofprint] AID—1479-5876-10-1 [pii] AID—doi: PST—epublish. 22214470
  46. 46. Galon J, Mlecnik B Fau—Bindea G, Bindea G Fau—Angell HK, Angell Hk Fau—Berger A, Berger A Fau—Lagorce C, Lagorce C Fau—Lugli A, et al. Towards the introduction of the 'Immunoscore' in the classification of malignant tumours. (1096–9896 (Electronic)). doi: D—NLM: PMC4255306 OTO—NOTNLM.
  47. 47. Fujimura T, Mahnke K Fau—Enk AH, Enk AH. Myeloid derived suppressor cells and their role in tolerance induction in cancer. (1873-569X (Electronic)).
  48. 48. Waldron TJ, Quatromoni Jg Fau—Karakasheva TA, Karakasheva Ta Fau—Singhal S, Singhal S Fau—Rustgi AK, Rustgi AK. Myeloid derived suppressor cells: Targets for therapy. (2162–4011 (Print)).
  49. 49. Yang L, DeBusk Lm Fau—Fukuda K, Fukuda K Fau—Fingleton B, Fingleton B Fau—Green-Jarvis B, Green-Jarvis B Fau—Shyr Y, Shyr Y Fau—Matrisian LM, et al. Expansion of myeloid immune suppressor Gr+CD11b+ cells in tumor-bearing host directly promotes tumor angiogenesis. (1535–6108 (Print)).
  50. 50. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ (Clinical research ed). 2009;339:b2535. Epub 2009/07/23. pmid:19622551; PubMed Central PMCID: PMCPMC2714657.