- Open Access
Cancer-related CD15/FUT4 overexpression decreases benefit to agents targeting EGFR or VEGF acting as a novel RAF-MEK-ERK kinase downstream regulator in metastatic colorectal cancer
- Guido Giordano1,
- Antonio Febbraro1,
- Eugenio Tomaselli2,
- Maria Lucia Sarnicola3,
- Pietro Parcesepe4,
- Domenico Parente2,
- Nicola Forte2,
- Alessio Fabozzi1,
- Andrea Remo5,
- Andrea Bonetti5,
- Erminia Manfrin4,
- Somayehsadat Ghasemi4,
- Michele Ceccarelli6, 7,
- Luigi Cerulo6, 7,
- Flavia Bazzoni4 and
- Massimo Pancione7Email author
© Giordano et al. 2015
- Received: 10 July 2015
- Accepted: 24 September 2015
- Published: 1 October 2015
Cancer-related immune antigens in the tumor microenvironment could represent an obstacle to agents targeting EGFR “cetuximab” or VEGF “bevacizumab” in metastatic colorectal cancer (mCRC) patients.
Infiltrating immune cells into tumor tissues, cancer-related expression of immune antigens (CD3, CD8, CD68, CD73, MPO, CD15/FUT4) from 102 mCRC patients receiving first-line Cetuximab or Bevacizumab plus chemotherapy were assessed by immunohistochemistry and validated in an independent tissue microarrays of 140 patients. Genome-wide expression profiles from 436 patients and 60 colon cancer cell lines were investigated using bioinformatics analysis. In vitro kinase assays of target genes activated by chemokines or growth factors were performed.
Here, we report that cancer-related CD15/FUT4 is overexpressed in most of mCRCs patients (43 %) and associates with lower intratumoral CD3+ and CD8+ T cells, higher systemic inflammation (NLR at diagnosis >5) and poorer outcomes, in terms of response and progression-free survival than those CD15/FUT4-low or negative ones (adjusted hazard ratio (HR) = 2.92; 95 % CI = 1.86–4.41; P < 0.001). Overexpression of CD15/FUT4 is induced through RAF-MEK-ERK kinase cascade, suppressed by MEK inhibitors and exhibits a close connection with constitutive oncogenic signalling pathways that respond to ERBB3 or FGFR4 activation (P < 0.001). CD15/FUT4-high expressing colon cancer cells with primary resistance to cetuximab or bevacizumab are significantly more sensitive to MEK inhibitors than CD15/FUT4-low counterparts.
Cancer-related CD15/FUT4 overexpression participates in cetuximab or bevacizumab mechanisms of resistance in mCRC patients. CD15/FUT4 as a potential target of the antitumor immune response requires further evaluation in clinical studies.
- EGFR or VEGF
- ERBB3 or FGFR4
- Metastatic CRC
Cancer is driven by activating mutations and aberrant signal transduction node, most of which (RAS, PTEN, EGFR) play a significant role in the prediction of resistance to epidermal growth factor receptor (EGFR) monoclonal antibodies in mCRC treatment, whereas, equivalent reliable predictors of bevacizumab are currently lacking [1, 2, 6, 7]. Current data provide evidence for extensive autocrine and paracrine EGFR-VEGF (R) cross-talk in both tumor and tumor-associated microenvironment underlining the potential interest in targeting both pathways [1–4]. Although clinically distinct subtypes of CRC are starting to emerge, the factors determining whether a patient will have a response to target-oriented therapy remain still elusive [7, 8].
A tumor grows in an intricate network of epithelial cells, vascular and lymphatic vessels, cytokines and chemokines, and infiltrating immune cells named the tumor microenvironment [9–11]. Increasing studies underscore the involvement of immune cells, through an emerging hallmark of cancer, evoked as evasion of immune surveillance [9–11]. In keeping with this concept, a higher infiltration of memory cytotoxic Th1 T-lymphocytes and tumor-associated neutrophils (TANs) correlate with a longer survival, evidencing the critical effect of host immune response on tumor evolution and clinical outcome [11–16]. Unfortunately, tumor develops multiple mechanisms of evading immune responses, by forming a compromised microenvironment that contrasts the effects of therapeutic agents [10, 17–19]. We have recently proposed that malignant cells can “express” ectopic immune epitopes, which are typical of immune cells (i.e., CD73, CD68) and that these may serve as tumor antigens to evade immune surveillance facilitating homotypic interactions in distant organs during metastatic process . The connection between immune neoantigens expressed on tumor cells and benefits to target therapies has received little attention so far. In this study, we investigated the relationship between inflammatory response, tumor immune-phenotypic features and patients’ outcome receiving first line cetuximab or bevacizumab plus chemotherapy schedules. Herein, we provide evidence that the “don’t eat me” signal CD15/FUT4 on cancer cells associates with decreased benefit to target therapy. CD15/FUT4 overexpression is driven by constitutive oncogenic signalling pathways in the tumor cells (innate immune resistance) acting as a novel RAF-MEK-ERK kinase downstream regulator through ERBB3 or FGFR4 activation, respectively. The results presented here could help to identify a subset of CD15/FUT4-overexpressing patients who have higher chances of benefiting from MEK inhibitors.
Patient population and samples
To study the relationship between tumor-associated immune infiltration and responses to targeted therapies, between 2010–2014 a retrospective cohort of metastatic CRC patients from two institutions: Medical Oncology Unit of Sacro Cuore di Gesù, Fatebenefratelli Hospital, Benevento (Italy) and Department of Oncology and Pathology, Mater Salutis Hospital, Legnago Verona, (Italy) were recruited. The cohort was partitioned into a discovery and validation set, resulting in a total of (n = 102) patients receiving the target-agents Cetuximab or Bevacizumab plus chemotherapy based schedules (FOLFOX, XELOX or FOLFIRI) first-line therapy. Ethical, legal, and social implications were approved by an ethical review board of Fatebefratelli Hospital Institution. Formalin-fixed, paraffin-embedded (FFPE) tumor tissues were retrieved, anonymized and areas of tumor revaluated on hematoxylin and eosin–stained sections. For systemic inflammatory response, the neutrophil-to-lymphocyte ratio (NLR) was calculated from routine complete blood on the same day of primary surgery. Study endpoints were represented by evaluation of patients’ outcome in terms of objective response (primary endpoint), progression free survival (PFS) and overall survival (OS). The best overall response was evaluated every 8 or 12 weeks. Tumor response was classified in: complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST). Consequently, patients with CR, PR and SD ≥ 6 months were considered responders while the remaining nonresponders. Details of patients’ outcome evaluation are provided in (Additional file 1: Table S1 and Additional file 2).
An independent cohort, named as validation series II consisting of 140 stage I-IV primary CRCs collected consecutively was then evaluated [21, 22]. They represented a continuous, unselected TMAs cohort of patients with molecular, histopathological and clinical findings (OS) recruited during the period 2003–2009. The NLR ratio was collected on the same day of primary surgery for routine laboratory analysis through the full blood count as indicated above. The complete workflow of the study is summarized in (Additional file 3: Figure S1). Details about this data set are provided in (Additional file 1: Table S2 and Additional file 2).
Following pathologic review for diagnostic confirmation and exclusion of highly fibrotic or necrotic tumor sections, whole-blocks 4-μm sections were incubated with antibodies listed in (Additional file 1: Table S3). The 3,3’ diaminobenzidine and haematoxylin were used as chromogen and counterstain, respectively. Tumor-associated infiltrating immune cells and cancer-related expression were identified using staining positivity analysis. Infiltrating immune cells from different tumor locations, were quantified by using ImageJ-based software, while intraepithelial immune cells were counted manually. All the cell counts were expressed as cells mm−2. The data were obtained from two whole sections per tissue and at least 500 epithelial cells and 500 infiltrating immune cells were analyzed per section. The proportion of cancer cells staining was scored in 3 s grades regardless of intensity as follows: 1) Negative staining (Neg) was defined as the complete absence of staining in more than 95 % of tumor cells; 2) Partially positive or (low expression) characterized by a limited number of tumor cells scattered in a background of either negative or positive tumor cells. 3) Diffuse positivity (High expression) corresponding to a homogeneous membrane staining in virtually all tumor cells. Details on immunohistochemistry (IHC) evaluation is provided in (Additional file 2).
Bioinformatics analysis and independent gene expression profile data sets
The following genome-wide expression data sets were analyzed by using in silico bioinformatics approaches: a) GSE17536/GSE17537 of 226 patients; b) colorectal Cancer Genome Atlas (TCGA) of 210 patients; c) Cancer Cell Line Encyclopedia, Broad Institute/Novartis of 60 CRC cell lines: d) metastatic CRC cell line “SW480” with primary resistance to cetuximab and treated with MEK inhibitor (AZD6244, Selumetinib), GEO Omnibus [7, 23–25]. The IC50, a direct indicator of drug efficacy, for six CRC cell lines, CD15/FUT4-high (HT29, LoVo, SW620) and CD15/FUT4-low (SW480, HCT116, SW48 and GEO) treated with MEKi BAY 86–9766, Selumetinib or Pimasertib was publically available and calculated according to the reported data . Details about in silico analysis is provided in (Additional file 2).
CRC derived cell lines and qRT-PCR validation
A series of 12 representative CRC-derived cell lines, “purchased from American Type Culture Collection (ATCC, Rockville, MD)” were grown in DMEM (Life Technologies, Grand Island, NY, USA) or RPMI 1640 medium plus 10 % FBS (Life Technologies) without antibiotics/antimycotics. All the cell lines were confirmed to be negative for mycoplasma by PCR (Venor GeMkit,Sigma-Aldrich, St. Louis, MO, USA) prior to use. Cells were cultured in a humidified 37 °C incubator at 5 % CO2. Total RNA from cell lines was extracted using miReasy kit (Qiagen, Hombrechtikon, Switzerland) and cDNA was generated using Superscript reverse transcriptase (Life Technologies, Grand Island, NY, USA). The concentration of cDNA was determined (Nanodrop 2000, Thermo Scientific, Asheville, NC, USA) and 25 ng of total cDNA was subjected to quantitative PCR using QI Agility (automated PCR setup, Qiagen), Quanti Tect SYBR Green PCR kit (Qiagen), and Rotor-Gene Q (Qiagen) real-time PCR machine and gene specific primers (Additional file 1: Table S4). The gene-specific copy number was calculated according to the standard curve and normalized to the amount of cDNA (ng) in the reaction. All PCR reactions were performed in triplicate and expression levels were computed as reported [20, 21, 27].
Reagents, transcript induction and kinase assays
CRC cells were then grown to 70 % of confluence, serum starved for 24 h, and stimulated for 8 h with 10 nM EGF (R&D System), 20U/ml IL-1beta (Peprotech), or for 30 min with 200U/ml IL-10 or 50 ng/ml IL-6 (R&D System). Subsequently, the cells were harvested for RNA (qRT-PCR see above) or protein extraction. Western blot was performed according to the published procedures [20, 21, 27]. A ratio of normalized ERK1/2 (pERK/total ERK1/2), Stat3 (pStat3/total Stat3) and stat1 (pstat1/total Stat1) was calculated for monitoring expression and phosphorylation levels. Human polymorphonuclear cells (PMN) and peripheral blood mononuclear cells (PBMC) purified from buffy coats of healthy donors were used as positive control for kinase assays . Details on western-blot and kinase assays are provided in (Additional file 2).
Statistical analyses were conducted by using R statistical software and SPSS version 15 Windows, SPSS Inc, Chicago, IL and GraphPad Prism 5. Data are presented with medians and ranges. Association between IHC expression and clinico-pathological data was assessed using Spearman r correlation or χ2 test. The Wilcoxon–Mann–Whitney or Kruskal-Wallis nonparametric tests were used to identify markers with a significantly differentexpression among patient groups. Kaplan-Meier curves were used to visualize differences between PFS and OS. Significance among patient groups was calculated by using the log-rank test. Prognostic and predictive effects were assessed using PFS and OS as clinical endpoints. The predictive performance of each individual marker was considered alone and together (CD15/FUT4 IHC and NLR at diagnosis). We used Cox proportional hazardsmodel to determine hazard ratios (HRs) with 95 % of confidence interval. The HRs were corrected through multivariate analysis, adjusting for other factors previously shown to be prognostic in the population study. All tests were two-sided, and a P < .05 was considered statistically significant. The (Additional file 2) illustrates further details about the statistical and validation methods used.
Relationship among inflammatory cell infiltration, immune-phenotypic traits expressed by cancer cells and therapy response
Next, to distinguish between inflammatory infiltrate and tumor immune-phenotypic traits, we examined the presence of positivity in malignant cells including a larger series of immune markers (Fig. 1b and Additional file 3: Figure S2A,B). Overall, we observed that the neutrophil antigen CD15/FUT4 was positive in the majority of cases. Among the 32 tumors of the discovery set, 23 (72 %) of CD15/FUT4-positive tumors were classified as low or high and had a higher recurrence than other tumor-related immune antigen i.e., CD68, CD73 (Fig. 1b, c and Additional file 2: Figure S2A). Using myeloperoxidase (MPO) as an additional marker of mature neutrophils, we did not detect positivity on malignant cells. In line with this, CD15/FUT4 marked infiltrating granulocytes but not epithelial colonic cells in matched normal adjacent mucosa (Fig. 1c and Additional file 3: Figure S2A, B). Tumors displaying high CD15/FUT4 expression showed a trend towards: a) lower peritumoral immune-cells density and elevated NLR at diagnosis; b) poorer patients’ outcome both in terms of response to first line therapy and progression free survival (Fig. 1b-d and Additional file 3: Figure S2C, D).
Prognostic significance of tumor-related CD15/FUT4 overexpression and inflammatory response
To corroborate CD15/FUT4 cancer-related alteration and its relationship with inflammatory response and patients’ outcome, we analyzed 70 additional samples, resulting in a total of 102 patients. Additional file 1: Table S1 summarizes clinico-pathological features of the entire study cohort.
CD15/FUT4 expression had a significant direct correlation with systemic inflammatory response at diagnosis. The median values of NLR ranged from 5.71 for CD15/FUT4-neg, 6.60 for CD15/FUT4-low and 10.5 for CD15/FUT4-high expressing tumors, respectively (P < 0.05; Fig. 2b and Additional file 3: Figure S3A). This indicated that when CD15/FUT4 increases in malignant cells, CRC inflammatory infiltrate declines and systemic inflammation at diagnosis increases.
We then associated cancer-related expression of CD15/FUT4 with patients’ outcome in terms of Response and Survival (PFS and OS).
Response to treatment according to CD15/FUT4-IHC on primary tumors and NLR at diagnosis
NLR at diagnosis
PFS was significantly different according to CD15/FUT4 expression on malignant cells: mPFS was 5.5 vs 10 and 13 months, in patients with CD15/FUT4-high, low and negative tumors, respectively (HR = 3.37; 95 % CI = 2.14–5.51; P < 0.0001, Fig. 2c). Accordingly, median OS was 13 vs 26 and 38 months in patients with CD15/FUT4-high, low and negative tumors, respectively (HR = 1.95; 95 % CI = 1.37-2.98; P =0.001, Fig. 2c).
Concordance for systemic inflammatory response at time of diagnosis and clinical response was also significant (Additional file 3: Figure S3B). Thirty-eight out of 72 (53 %) with NLR >5 and 2 out of 30 (7 %) patients with NLR ≤ 5 were considered nonresponders, respectively (P = 0.00012545; Table 1). Median PFS was 6.5 vs 12 months in patients with NLR >5 vs ≤ 5, respectively (HR = 2.41; 95 % CI = 1.37–4.32; P = 0.002). Median mOS was 17 vs 35 months in patients with NLR >5 vs ≤ 5, respectively (HR = 2.39; 95 % CI = 1.48–3.85; P < 0.0001; Fig. 2d).
Univariate and multivariate analysis for PFS and OS. Cox’s regression model taking into account the IHC data or NLR at diagnosis
NLR >5 vs NLR ≤5
RAS Mut vs WT
Multiple vs single metastases
CD15/FUT4 high vs low/neg
NLR >5 vs NLR ≤ 5
Stage at diagnosis III and IV vs II
Histotype mucinous vs others
CD15/FUT4 high vs low/neg
Tumor-related CD15/FUT4 expression and associated molecular parameters in an independent TMAs validation set of stages I-IV CRCs
This independent cohort confirmed CD15/FUT4 positivity in the majority of carcinomas (76 %; 107 out of 140). High, low and negative expression was identified in 58 out 107 (54 %), 49 out of 107 (46 %), and 33 out of 107 (24 %) of cases, respectively (Additional file 1: Table S2 and Additional file 3: Figure S4A). Tumors displaying CD15/FUT4-high expression pattern were more frequently associated with advanced tumor stage III and IV and with lower-densities of CD3+ and CD8+ T cells peritumoral infiltrate, confirming the data obtained on the first series (Additional file 3: Figure S4B). Tumors displaying this pattern were also frequently MMR proficient and TP53 and KRAS mutated than CD15/FUT4-negative one (Additional file 1: Table S3). We tested whether the combination CD15/FUT4 and NLR had impact on patients’ prognosis also in this series. Indeed, the combinations CD15NegNLR≤5 and CD15highNLR>5robustly stratify the cohort into two groups with 100 % and 33.1 % of survival rate at 5 years after diagnosis, respectively. (Additional file 3: Figure S4C, D).
Genomic variations related to CD15/FUT4 overexpression
CD15/FUT4 overexpression and MEK inhibitor responses in CRC cell lines
Much effort is currently focused on attempts to target several signaling pathways at the same time. The extensive degree of EGFR-VEGF(R) pathway cross-talk identifies them as particularly promising for joint targeting [3, 29]. In addition to RAS gene family mutations, a number of studies suggest that intracellular downstream effectors of these pathways or immune inflammatory microenvironment could be correlated with primary resistance in metastatic tumors [1–5, 10]. Tumor microenvironment not only plays a pivotal role during cancer progression and metastasis but also has profound effects on therapeutic efficacy [10–13, 17–19]. According to this assumption, immune checkpoint inhibitors are promising new approaches for tackling solid tumors [31–33].
In this study, we investigated candidate determinants of resistance related to the immune tumor microenvironment and inflammatory response, using as proof of principle two mainstay of first line treatment for metastatic CRC anti-EGFR cetuximab, anti-VEGF bevacizumab based therapy. Surprisingly, among a panel of immune markers, we uncovered that a large proportion of tumors overexpressed CD15/FUT4 neuthrophil antigen. Strikingly, this expression pattern associated with short disease control and rapid disease progression. CD15/FUT4-high expressing tumors showed lower intratumoral immune density of both innate (CD68+, CD15+ and MPO+ macrophages and Neutrophil cells) and especially adaptive (CD3+ and CD8+) immune T-cell subsets. The concordance with NLR at diagnosis, suggested that CD15/FUT4 overexpression on malignant cells could connect peritumoral immune suppression and elevated systemic inflammatory response. Patients with tumors harboring CD15/FUT4-high expression were associated with worse PFS both in univariate and multivariate analysis, indicating CD15/FUT4 as a possible marker of decreased therapy response, a finding further strengthened by an elevated blood inflammatory response (NLR > 5). Therefore, the progressive decrease of immune cell densities along with CD15/FUT4 overexpression and increased inflammatory response could provide a clinical context of tumor progression, explainable as a pronounced immune-escape mechanism. Our observations however have a number of limitations in particular because patients had received two treatments targeting different pathways, and are few to support more general conclusions, although the number can be considered large (102) for an IHC study. In this regard, an independent TMAs data set confirmed that high-CD15/FUT4 expressing tumors correlated with reduced density of the immune infiltrates by CD3 and CD8 cells and elevated blood inflammatory response, similarly to those seen in the validation set I. We observed that CD15/FUT4-expressing tumors had generally a well-differentiated gene expression pattern (MMR proficient) and correlated with subtypes enriched for TP53 and KRAS mutations. These results evidenced that CD15/FUT4 could serve as a surrogate marker to identify specific subtypes with chromosomal instability for which has already been proved lack of response to cetuximab and tendency to metastasize . Despite this, such evidence cannot explain lack of response to agents targeting anti-EGFR and even more to anti-VEGF. In this scenario, we performed in-silico analysis across publically available gene-expression data sets along with in vitro assays on derived colon cancer cell lines. Transcriptomic profiling confirmed that high CD15/FUT4 transcript associated with CIN, KRAS mutations and most importantly with ERBB3 and FGFR4 overexpression but not ERBB1 (EGFR). These results supported the hypothesis that consistent activation of CD15/FUT4 transcript acts downstream and/or independently of EGFR or VEGFR pathways. Interestingly, our findings are also consistent with recent studies revealing that transcriptional induction of ERBB3 acts as a prominent “hit” of intrinsic resistance in CRC cell lines, while very little is known about FGFR4 overexpression [34, 35]. Our observations indicate that upregulation of FGFR4 may correlate with intrinsic resistance to bevacizumab, as evidenced in HT29 and SW620 CRC cell lines, respectively.
Therefore, the novel role of CD15/FUT4 reflects persistent genetic features of tumor cells rather than differences in immune infiltrates. Indeed, we demonstrate “in vitro” that CD15/FUT4 transcript is induced through MAPK-ERK kinase cascade. This finding was further validated by querying for selective MEK inhibitor modulated genes in cancer cell lines with KRAS or BRAF mutations and primary resistance to cetuximab or bevacizumab, respectively. We observed that the MEK inhibitor “Selumetinib” caused a significant reduction of CD15/FUT4 transcript, as a consequence, CD15/FUT4-high expressing cancer cells were more sensitive to “Selumetinib” and Pimasertib” than CD15/FUT4-low counterparts.
Based on these findings, we proposed a model where drug resistance could synergize on CD15/FUT4 through at least two independent pathways: a) mitogenic intracellular due to ERBB2/ERBB3 overexpression; b) pro-angiogenic and/or mitogenic due to FGFR4 or IL-1β released by an altered tumor microenvironment as suggested by other studies (Fig. 4d) [35–37].
Our results suggest a possible rational for treating CD15/FUT4-overexpressing mCRC through means of IHC. In this subset, MEK inhibitors or dual inhibitors that show strong synergy with MEK inhibition “i.e., cetuximab” could be effective for preventing and/or overcoming primary resistance to cetuximab in CRC patients as recently reported in preclinical studies [34, 35]. CD15/FUT4 could also be dysregulated by tumors as an important immune resistance mechanism similarly to immune-checkpoint proteins. However, the relevance of its molecular interactions to inhibit local antitumor T cell-mediated response in the tumor microenvironment remains obscure .
In conclusion, cancer-related CD15/FUT4 overexpression is associated with decreased benefit to target and chemotherapeutic agents in metastatic tumors. CD15/FUT4 acts as a downstream regulator of MAPK-ERK pathway independently of EGFR or VEGF pathway, by coupling mitogenic signaling cascade and immune-escape mechanisms of metastatic tumors. Upregulation of CD15/FUT4 on the tumor cell surface could represent a potential target to enhance antitumor effector functions in the tumor microenvironment. In addition, IHC assessment of CD15/FUT4 combined with RAS mutation status could be a strategy to identify mCRC patients who have higher chances of benefiting from targeting and MEK inhibitor drugs. However, its role as a potential biomarker susceptible to specific therapeutic agents requires further evaluation in clinical setting.
Part of the study was funded by the (FUR) and Italian Ministry of University and Research (MiUR) to FB and MP.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Argiles G, Dienstmann R, Elez E, et al. Panitumumab: a summary of clinical development in colorectal cancer and future directions. Future Oncol. 2012;8(4):373–89.View ArticlePubMedGoogle Scholar
- Douillard JY, Oliner KS, Siena S, Tabernero J, Burkes R, Barugel M, et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. N Engl J Med. 2013;369(11):1023–34.View ArticlePubMedGoogle Scholar
- Martinelli E, Troiani T, Morgillo F, Orditura M, De Vita F, Belli G, et al. Emerging VEGF receptor inhibitors for colorectal cancer. Expert Opin Emerg Drugs. 2013;18:25–37.View ArticlePubMedGoogle Scholar
- Grothey A, Van Cutsem E, Sobrero A, Siena S, Falcone A, Ychou M, et al. Regorafenib monotherapy for previously treated metastatic colorectal cancer (CORRECT): an international, multicentre, randomised, placebo-controlled, phase 3 trial. Lancet. 2013;381(9863):303–12.View ArticlePubMedGoogle Scholar
- Xu JM, Liu XJ, Ge FJ, Lin L, Wang Y, Sharma MR, et al. KRAS mutations in tumor tissue and plasma by different assays predict survival of patients with metastatic colorectal cancer. J Exp Clin Cancer Res. 2014;10:33–104.Google Scholar
- Bardelli A, Siena S. Molecular mechanisms of resistance to cetuximab and panitumumab incolorectal cancer. J Clin Oncol. 2010;28:1254–61.View ArticlePubMedGoogle Scholar
- Cancer Genome Atlas Network. Comprehensive molecular characterization of human colon and rectal cancer. Nature. 2012;487:330–7.View ArticleGoogle Scholar
- Sadanandam A, Lyssiotis CA, Homicsko K, Collisson EA, Gibb WJ, Wullschleger S, et al. A colorectal cancer classification system that associates cellular phenotype and responses to therapy. Nature Med. 2013;19(5):619–25.PubMed CentralView ArticlePubMedGoogle Scholar
- Mantovani A, Allavena P, Sica A, Balkwill F. Cancer related inflammation. Nature. 2008;454:436–44.View ArticlePubMedGoogle Scholar
- Trédan O, Galmarini CM, Patel K, Tannock IF. Drug resistence and the solid tumor microenvironment. J Natl Cancer Inst. 2007;99:1441–54.View ArticlePubMedGoogle Scholar
- Giraldo NA, Becht E, Remark R, Damotte D, Sautès-Fridman C, Fridman WH. The immune contexture of primary and metastatic human tumours. Curr Opin Immunol. 2014;27:8–15.View ArticlePubMedGoogle Scholar
- Fridman WH, Pages F, Sautes-Fridman C, Galon J. The immune contexture in human tumours: impact on clinical outcome. Nat RevCancer. 2010;12:298–306.View ArticleGoogle Scholar
- Vayrynen JP, Tuomisto A, Klintrup K, Mäkelä J, Karttunen TJ, Mäkinen MJ. Detailed analysis of inflammatory cell infiltration in colorectal cancer. Br J Canc. 2013;109:1839–47.View ArticleGoogle Scholar
- Donskov F. Immunomonitoring and prognostic relevance of neutrophils in clinical trials. Semin Cancer Biol. 2013;23(3):200–7.View ArticlePubMedGoogle Scholar
- Droeser RA, Hirt C, Eppenberger-Castori S, Zlobec I, Viehl CT, Frey DM, et al. High myeloperoxidase positive cell infiltration in colorectal cancer is an independentfavorable prognostic factor. PLoS One. 2013;8(5):e64814.PubMed CentralView ArticlePubMedGoogle Scholar
- Tecchio C, Scapini P, Pizzolo G, Cassatella MA. On the cytokines produced by human neutrophils in tumors. Semin Cancer Biol. 2013;3:159–70.View ArticleGoogle Scholar
- Malietzis G, Giacometti M, Kennedy RH, Athanasiou T, Aziz O, Jenkins JT. The emerging role of neutrophil to lymphocyte ratio in determining colorectal cancer treatment outcomes: a systematic review and meta-analysis. Ann Surg Oncol. 2014;21(12):3938–46.View ArticlePubMedGoogle Scholar
- Sussman DA, Santaolalla R, Bejarano PA, Garcia-Buitrago MT, Perez MT, Abreu MT, et al. In silico and Ex vivo approaches identify a role for toll-like receptor 4 in colorectal cancer. J Exp Clin Cancer Res. 2014;33:45.PubMed CentralView ArticlePubMedGoogle Scholar
- McMillan DC. Systemic inflammation, nutritional status and survival in patients with cancer. Curr Opin Clin Nutr MetabCare. 2009;12(3):223–6.View ArticleGoogle Scholar
- Pancione M, Giordano G, Remo A, Febbraro A, Sabatino L, Manfrin E, et al. Immune escape mechanisms in colorectal cancer pathogenesis and liver metastasis. J Immunol Res. 2014;2014:686879.PubMed CentralView ArticlePubMedGoogle Scholar
- Pancione M, Remo A, Zanella C, Sabatino L, Di Blasi A, Laudanna C, et al. The chromatin remodelling component SMARCB1/INI1 influences the metastatic behavior of colorectal cancer through a gene signature mapping to chromosome 22. J Transl Med. 2013;11:297S.View ArticleGoogle Scholar
- Pagnotta SM, Laudanna C, Pancione M, Sabatino L, Votino C, Remo A, et al. Ensemble of gene signatures identifies novel biomarkers in colorectal Cancer activated through PPARγ and TNFα signaling. PLoSOne. 2013;8(8):e72638.View ArticleGoogle Scholar
- Smith JJ, Deane NG, Wu F, Merchant NB, Zhang B, Jiang A, et al. Experimentally derived metastasis gene expression profile predicts recurrence and death in patients with colon cancer. Gastroenterology. 2010;138:958–68.PubMed CentralView ArticlePubMedGoogle Scholar
- Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, et al. The cancer cell line encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature. 2012;483(7391):603–7.PubMed CentralView ArticlePubMedGoogle Scholar
- Schoumacher M, Hurov KE, Lehár J, Yan-Neale Y, Mishina Y, Sonkin D, et al. Inhibiting Tankyrases sensitizes KRAS-mutant cancer cells to MEK inhibitors via FGFR2 feedback signaling. Cancer Res. 2014;74(12):3294–305.View ArticlePubMedGoogle Scholar
- Troiani T, Napolitano S, Vitagliano D, Morgillo F, Capasso A, Sforza V, et al. Primary and acquired resistance of colorectal cancer cells to anti-EGFR antibodies converge on MEK/ERK pathway activation and can be overcome by combined MEK/EGFR inhibition. Clin Cancer Res. 2014;20(14):3775–86.View ArticlePubMedGoogle Scholar
- Curtale G, Mirolo M, Renzi TA, Rossato M, Bazzoni F, Locati M. Negative regulation of Toll-like receptor 4 signaling by IL-10-dependent microRNA-146b. Proc Natl Acad Sci U S A. 2013;110(28):11499–504.PubMed CentralView ArticlePubMedGoogle Scholar
- Bordon Y. Anticancer drugs need bugs. Nature Rev Immunol. 2014;14:1.View ArticleGoogle Scholar
- Kitamura T, Qian BZ, Pollard JW. Immune cell promotion of metastasis. Nature Rev Immunol. 2015;15:73.View ArticleGoogle Scholar
- Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating immunity’s roles in cancer suppression and promotion. Science. 2011;331:1565–70.View ArticlePubMedGoogle Scholar
- Steinert G, Schölch S, Niemietz T, Iwata N, García SA, Behrens B, et al. Immune escape and survival mechanisms in circulating tumor cells of colorectal cancer. Cancer Res. 2014;74:1694–704.View ArticlePubMedGoogle Scholar
- Snyder A, Makarov V, Merghoub T, Yuan J, Zaretsky JM, Desrichard A, et al. Genetic basis for clinical response to CTLA-4 Blockade in Melanoma. N Engl J Med. 2014;371(23):2189–99.PubMed CentralView ArticlePubMedGoogle Scholar
- Yadav M, Jhunjhunwala S, Phung QT, Lupardus P, Tanguay J, Bumbaca S, et al. Predicting immunogenic tumour mutations by combining mass spectrometry and exomesequencing. Nature. 2014;515(7528):572–6.View ArticlePubMedGoogle Scholar
- Misale S, Arena S, Lamba G, Siravegna G, Lallo A, Hobor S, et al. Blockade of EGFR and MEK intercepts heterogeneous mechanisms of acquired resistance to anti-EGFR therapies in colorectal cancer. Sci Transl Med. 2014;6(224):224ra26.View ArticlePubMedGoogle Scholar
- Sun S, Hobor A, Bertotti D, Zecchin D, Huang S, Galimi F. Intrinsic resistance to MEK inhibition in KRAS mutant lung and colon cancer through transcriptional induction of ERBB3. Cell Reports. 2014;7:86–93.View ArticlePubMedGoogle Scholar
- Carmi Y, Dotan S, Rider P, Kaplanov I, White MR, Baron R, et al. The role of IL-1b in the early tumor cell–induced angiogenic response. J Immunol. 2013;190(7):3500–9.View ArticlePubMedGoogle Scholar
- Cheng L, Luo S, Jin C, Ma H, Zhou H, Jia L. FUT family mediates the multidrug resistance of human hepatocellular carcinoma via the PI3K/Akt signaling pathway. Cell Death Dis. 2013;4:923.View ArticleGoogle Scholar
- Warren HS, Altin JG, Waldron JC, Kinnear BF, Parish CR. A carbohydrate structure associated with CD15 (Lewis x) on myeloid cells is a novel ligand for human CD2. J Immunol. 1996;156:2866–73.PubMedGoogle Scholar