Prognostic role of intratumoral IL-17A expression by immunohistochemistry in solid tumors: a meta-analysis

IL-17A is an important proinflammatory cytokine which is frequently elevated in tumor microenvironment. However, the role of intratumoral IL-17A in solid tumors remains controversial. Herein, we conducted a meta-analysis to assess the prognostic impact of intratumoral IL-17A in patients with solid tumor. PubMed and EBSCO were searched to identify the studies evaluating the associations between intratumoral IL-17A measured by immunohistochemistry (IHC) and overall survival (OS) and disease-free survival (DFS) in solid tumors. A total of 2972 patients with solid tumor from 21 published studies were incorporated into this meta-analysis. We found that high level of intratumoral IL-17A was significantly associated with worse 3-year, 5-year OS and 1-year, 3-year DFS, but not with 1-year OS or 5-year DFS in solid tumors. In addition, in stratified analyses by cancer types, IL-17A overexpression was significantly associated with worse OS in hepatic carcinoma, but with improved OS in esophageal squamous cell carcinoma (ESCC). Furthermore, high IL-17A expression positively correlated with advanced TNM stage. In conclusion, High expression of intratumoral IL-17A leads to an unfavorable clinical outcome in majority of solid tumors, implicating IL-17A is a valuable biomarker for prognostic prediction of human solid malignances and targeting it may have a potential for effective treatment.


INTRODUCTION
Chronic inflammation is closely linked to cancer [1,2], Recent study has demonstrated that cancerrelated inflammation promotes tumor development and progression through various mechanisms, including stimulating angiogenesis, promoting cell proliferation and survival, inducing gene mutation and subverting antitumor immune responses [3,4]. Accumulating evidence has proven that cytokines could promote carcinogenesis by both provoking inflammation [5][6][7] and eliciting immunosuppression [8,9].
Interleukin-17A (IL-17A), which initially termed as cytotoxic T-lymphocyte-associated antigen (CTLA-8), is the founding member of IL-17 cytokine family [10,11]. It is mainly produced by Th17 cells and also by γδ T, natural killer and CD8 + T cells [12,13], relying on STAT3 activation triggered by IL-23 [14]. A large body of evidence suggests that IL-17A is an essential proinflammatory cytokine due to inducing a mass of cytokines and chemokines secretion by distinct cell types, such as mesenchymal cells and myeloid cells, which recruit monocytes and neutrophils into the site of inflammation [15]. Recently, several studies have shown that IL-17A has either a protumor or antitumor role in different cancer models [16]. IL-17A is elevated in several human tumors [17]. In previous studies, high level of IL-17A within tumor was shown to associate with poor survival of several cancers [18][19][20][21]; whereas some other studies reported opposite results [22][23][24][25]. Therefore, an indepth assessment is warranted. Moreover, the potential of IL-17A as an effective biomarker in prognostic prediction and targeted therapy is necessary to be explored.

Meta-Analysis
Oncotarget 66383 www.impactjournals.com/oncotarget Here, we performed this meta-analysis to test overall survival (OS) and disease-free survival (DFS) as outcomes in patients with solid tumor with known IL-17A levels within tumor evaluated by IHC. The aim of this study was to quantitatively summarize the association between intratumoral IL-17A and clinical outcomes in cancer patients, and thereby provided more evidence on the clinical value of IL-17A as a prognostic biomarker and therapeutic target for solid malignances.

Search results and description of studies
Literature searches yield 6219 records and the results are shown in Figure 1. 21 studies containing 2972 cancer patients were identified for the assessment of IL-17A expression within tumor . All the studies were evaluated by the Newcastle-Ottawa Scale (NOS), and were in accordance with the inclusion criteria and suitable for data consolidation. Characteristics of included studies for OS, DFS and clinicopathological features such as TNM stage, tumor differentiation were shown in Table 1 and Supplementary Table 1 respectively.

Sensitivity analysis
Sensitivity analyses were used to determine the influence of individual studies on the overall OR. As a result, the plot showed that all the individual studies had no important impact on the results for OS at 1, 3 and 5-year (Supplementary Figure 3).

Publication bias
Funnel plot and Egger's test were performed to assess the publication bias of this meta-analysis. No publication bias existed between overexpression of intratumoral IL-17A and OS or DFS in solid tumors (data not shown).

IL-17A shows its bidirectional functions both in pro-tumor and anti-tumor effect. Previous meta-analyses
Oncotarget 66384 www.impactjournals.com/oncotarget reported that IL-17A wasn't significantly associated with OS in cancers [39,40]. However, the studies included in these meta-analyses reported the different sources of IL-17A involving tumor tissue, peripheral blood or peritoneal lavage and different detecting methods such as Flow Cytometry, ELISA, IHC and RT-PCR. Thus, the results were not accurate even wrong when they were from the combinations of all these studies. In this study, we found high level of intratumoral IL-17A was significantly correlated with worse 3-year, 5-year OS and 1-year, 3-year DFS, but not with 1-year OS or 5-year DFS in majority of solid tumors, which differing markedly from the previous studies [39,40]. Furthermore, IL-17A in tumor tissue positively correlated with TNM stage. Therefore, we believe our study provides meaningful statistical evidence for the prognostic value of intratumoral IL-17A for the first time.
Actually, IL-17A, which is secreted by Th17, γδ T, Tc17 and mast cells in tumor tissue, could promote cancer-elicited inflammation and prevent tumor cells from immune surveillance [17], through subverting T cell mediated anti-tumor immune responses by polarization of myeloid-derived suppressor cells (MDSCs) [41]. In stratified analyses by cancer types, we found the IL-17A overexpression was significantly associated with decreased OS in hepatic carcinoma and NSCLC, but with improved OS in ESCC. The possible explanation might be that the number of intratumoral IL-17A-producing cells positively correlated with infiltrating effector CD8 + T and CD57 + NK cells in human ESCC as previous reported [42].
Some limitations should be noted from this metaanalysis. First, significant heterogeneity observed across studies cannot be completely accounted. The possible sources of heterogeneity were the inconsistent cut-off values for assessing IL-17A expression and the IL-17A IHC methods taken by different researchers. Therefore, we suggest researchers to use IL-17A mAbs with the same clone number and employ similar cut-off value when they assess IL-17A in the future study. In addition, most of the included studies were performed in China, and studies with negative results or small sample size may not be published, which can cause potential publication bias.
In conclusion, high level of intratumoral IL-17A leads to an unfavorable clinical outcome in majority of

Search strategy
We searched PubMed and EBSCO for studies assessing the expression of IL-17A in tumor tissue and survival in patients with solid tumor from 1996 to August 2016. The searching keywords were ("Interleukin-17A " OR "IL-17A" OR "IL17A") AND ("neoplasms" OR "cancer" OR "tumor" OR "carcinoma"). A total of 1028 and 5191 entries were identified in PubMed and EBSCO respectively.

Inclusion and exclusion criteria
Inclusion criteria of the study were the measurement of IL-17A by immunohistochemistry (IHC), provided the survival data (OS and /or DFS) and publication in English. Exclusion criteria included studies that evaluated IL-17A with ELISA or Flow Cytometry or RT-PCR, detected IL-17A not in tissues, and studies on animals or in lab.

Endpoints
OS was recorded as the primary endpoint, while the second endpoint was DFS. Cut-offs of intratumoral IL-17A expression level defined by individual studies classified patients with solid tumor into high-and lowgroups.

Data extraction
Two authors (GM.H. and ZA.L.) independently reviewed and extracted data using predefined data abstraction forms from each eligible studies. Extracted information included first author's name, publication year, country, type of cancer, number of patients, median age, gender, Tumor, Lymph Node, Metastasis (TNM) stage, tumor differentiation, time of follow-up, technique used to quantify IL-17A, and cut-off value to determine IL-17A positivity. OS, DFS and clinicopathological data were extracted from the text, tables, or Kaplan -Meier curves for both IL-17A-high and -low group.

Quality assessment
The studies included in the meta-analysis were cohort studies. The quality of each included study was assessed using Newcastle-Ottawa Scale (NOS) (Supplementary Table 2) by two independent authors [43]. (Supplementary Table 3) The studies with 6 scores or more were classified as high quality studies. A consensus NOS score for each item was achieved.

Statistical analysis
Extracted data were combined into a meta-analysis using STATA 12.0 analysis software (Stata Corporation, College Station, TX, USA). Statistical heterogeneity was assessed using the chi-squared based Q-test or the I 2 method [44]. Data were combined according to the random-effect model in the presence of heterogeneity [45], otherwise, the fixed-effect model was performed [46]). Sensitivity analysis was employed to assess the influence of each study on the pooled result. Begg's funnel plot and Egger's test [47] were calculated to investigate potential publication bias. All P values were two-sided and less than 0.05 are considered statistically significant.