O

C

12

3

45

A

5ssatr©

K

H

h1

ARTICLE IN PRESSNCH-1961; No. of Pages 9

Critical Reviews in Oncology/Hematology xxx (2015) xxx–xxx

Biology and immunology of cancer stem(-like) cells in headand neck cancer

Xu Qian a,b, Chenming Ma b, Xiaobo Nie a, Jianxin Lu a, Minoo Lenarz b,Andreas M. Kaufmann c, Andreas E. Albers b,∗

a Key Laboratory of Laboratory Medicine, Ministry of Education, Zhejiang Provincial Key Laboratory of Medical Genetics, Wenzhou Medical University,Zhejiang, PR China

b Department of Otorhinolaryngology, Head and Neck Surgery, Charité-Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germanyc Clinic for Gynecology, Charité-Universitätsmedizin Berlin, Campus Mitte and Benjamin Franklin, Berlin, Germany

Accepted 30 March 2015

ontents

. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

. The CSC hypothesis in HNSCC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.1. CSCs in cancer progression and metastasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.2. Interactions between human papillomavirus (HPV) and CSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.3. Resistance to current therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

. CSC-induced immune-responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.1. Recognition of CSCs by the immune system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2. Immune escape of CSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.3. Immune suppression by CSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

. CSC-based vaccination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Reviewers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Biographies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

bstract

Immunological approaches against tumors including head and neck squamous cell carcinoma (HNSCC) have been investigated for about0 years. Such immunotherapeutic treatments are still not sufficiently effective for therapy of HNSCC. Despite the existence of immuno-urveillance tumor cells may escape from the host immune system by a variety of mechanisms. Recent findings have indicated that cancertem(-like) cells (CSCs) in HNSCC have the ability to reconstitute the heterogeneity of the bulk tumor and contribute to immunosuppression

nd resistance to current therapies. With regard to the CSC model, future immunotherapy possibly in combination with other modes ofreatment should target this subpopulation specifically to reduce local recurrence and metastasis. In this review, we will summarize recentesearch findings on immunological features of CSCs and the potential of immune targeting of CSCs.

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

2015 Elsevier Ireland Ltd. All rights reserved.

eywords: Cancer stem cells; ALDH1; Epithelial–mesenchymal transition; Human

∗ Corresponding author at: Department of Otorhinolaryngology, Head and Neckindenburgdamm 30, D-12200 Berlin, Germany. Tel.: +49 30 84454586.

E-mail address: andreas.albers@charite.de (A.E. Albers).

ttp://dx.doi.org/10.1016/j.critrevonc.2015.03.009040-8428/© 2015 Elsevier Ireland Ltd. All rights reserved.

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

papillomavirus; Immunotherapy; Vaccination

Surgery, Charité-Universitätsmedizin Berlin, Campus Benjamin Franklin,

ARTICLE IN PRESSONCH-1961; No. of Pages 9

2 Oncolog

1

csrriosttcfgwtata

atrgecc[apEiai[tiwp

2

tr(thHualgs

fiocc

2

tfmaicprctcE(

(sdaCstCewaAdrofi[CRgtGaYivwuat

X. Qian et al. / Critical Reviews in

. Introduction

The current treatments for head and neck squamous cellarcinoma (HNSCC) have been challenged by the cancertem(-like) cell (CSC) hypothesis. These cells play a centralole in initiation, progression, invasion, metastasis, recur-ence of tumors and resistance to therapies [1]. In vitro andn vivo studies of HNSCC have shown that putative CSCsr CSC-enriched non-adherent spheroid cells present withtem cell-like self-renewal properties, invasion capacity andherapy resistance [2–5]. The CSC model is closely relatedo the phenomenon that HNSCC initially respond well toonventional treatments, but local and distant relapses occurrequently. It is interesting to note that phenotypic hetero-eneity and plasticity of CSCs was observed to be associatedith epithelial-mesenchymal transition (EMT), which collec-

ively promotes metastasis [6]. Subsequently, CSCs require special microenvironment to regulate their stemness, ando initiate and promote cancer development by recruiting andctivating special cell types [7–10].

The development and the introduction of immunother-py for HNSCC holds promise as an attractive supplemento traditional treatments such as surgery, chemotherapy, andadiation therapy. Since immunotherapies are designed to tar-et directly the tumor cells the incidence of side effects isxpected to be low. Many approaches based on bulk tumorells have been developed and successfully monitored, but aorrelation with good clinical responses has been sparse so far11,12]. The main issues in developing cancer immunother-py are the strengthening of cytotoxic T cell responses andrevention or reversal of tumor-induced immune-escape.merging evidence indicates that the host immune system

s able to recognize CSCs and mount an effector responsegainst them, but CSCs may also play a role in mediat-ng immunosuppression within the tumor microenvironment13,14]. Therefore, it is necessary to gain further insight intohe immunological features of CSCs and explore potentialmmunotherapeutic approaches against CSCs. In this review,e discuss the biology of CSC in HNSCC with regard to theirotential as targets for future immunotherapy.

. The CSC hypothesis in HNSCC

Accumulating evidence suggests that in a heterogeneicumor, a subpopulation of tumor cells with stem cell-like self-enewal capacity, known as CSCs or tumor-initiating cellsTICs) have the ability to give rise to a proliferative bulkumor cell mass and to survive systemic treatments [1]. CSCsave been identified in many types of solid tumors includingNSCC [15,16]. One of the first studies of CSCs in HNSCCsing an immunodeficient mouse as model demonstrated that

+

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

minor population of CD44 cancer cells, which account foress than 10% of cells in a HNSCC primary tumor, couldive rise to new tumors in vivo and displayed the ability ofelf-renewal and differentiation [2]. In consistency with this

oTnm

y/Hematology xxx (2015) xxx–xxx

nding, important advances have been achieved in the studyf the role of HNSCC CSCs in the progression of malignan-ies in in vitro or in vivo mouse models and patient-derivedlinical samples.

.1. CSCs in cancer progression and metastasis

Once initiated, CSCs may generate macroscopic tumorshrough the stem cell processes of self-renewal and dif-erentiation into multiple cell variants. Furthermore, CSCsay undergo EMT, a process involved in embryogenesis

nd considered also to be involved in metastatic dissem-nation [17]. During EMT, cells of epithelial phenotypeonvert to migratory and invasive cells with mesenchymalhenotype. When the migrating mesenchymal cells haveeached their destination, they may undergo a reverse pro-ess, a mesenchymal–epithelial transition (MET), to regainhe epithelial phenotype. Recent studies highlight that tumorells undergoing EMT acquire stem cell-like properties, andMT can also induce non-CSC to acquire a CSC-like state

Fig. 1) [18–20].We previously showed that aldehyde dehydrogenase 1

ALDH1)+-CSC enriched cell populations from 3 dimen-ional spheroid cultures generated from HNSCC cell linesisplayed EMT characteristics with enhanced colony formingbility and invasiveness [4]. Further, the presence of HNSCC-SCs with the ability to undergo both EMT and MET by

witching between their epithelial and mesenchymal pheno-ypes has been discovered by Biddle et al. [6]. MigratoryD44high epithelial-specific antigen (ESA)low EMT-CSCxpressed EMT markers and a mesenchymal phenotype,hile CD44highESAhigh non-EMT-CSC had epithelial char-

cteristics. Importantly, EMT-CSC thereby required anLDH+ phenotype to switch to non-EMT-CSC and toevelop metastasis successfully. More recently, a CD44-egulated signaling pathway mediated by phosphorylationf glycogen synthase kinase 3� (GSK3�) has been identi-ed and has shown the potential to affect CSC phenotypes21]. Inhibition of GSK3� could reduce the formation ofSCs-enriched tumor spheres and “holoclone” colonies.eduction of the expression of stem cell markers and upre-ulation of the differentiation markers were also found inhe CD44highESAhigh cell fraction by GSK3� inhibition.SK3� knockdown could induce CSCs reversing from EMT

nd back to the epithelial CSC phenotype. In another study,ang et al. identified a mechanism in which the EMT

nducer Twist1 elicits cancer cell movement through acti-ation of RAC1 [22]. They found that Twist1 cooperatesith BMI1 to suppress let-7i expression, which results inp-regulation of NEDD9 and DOCK3, leading to RAC1ctivation and enabling mesenchymal-mode movement inhree-dimensional environments. Moreover, the suppression

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

f let-7i contributes to Twist1-induced stem-like properties.hese tumor cells expressing a stem-like cancer cell phe-otype could transit from non-motile, epithelial-like cells tootile mesenchymal cells. Reversing EMT in prostate cancer

ARTICLE IN PRESSONCH-1961; No. of Pages 9

X. Qian et al. / Critical Reviews in Oncology/Hematology xxx (2015) xxx–xxx 3

Fig. 1. Cancer-stem(-like)-cells-induced therapy resistance and its interaction with tumor microenvironments contribute to tumor recurrence and the developmentof distant metastasis. Abbreviations: cancer stem(-like) cell (CSC); circulating tumor cell (CTC); cytotoxic T lymphocyte (CTL); infiltrating T regulatory cell(

ciKpH

smfriafndwtadgb

2a

s

fHodsHcgpisd

HiimlEZloac

Treg); tumor-associated macrophage (TAM).

ells by forced re-expression of miR-200, which also signif-cantly inhibited the self-renewal capacity, was reported byong et al. [23]. Together, inhibition or reversal of the EMTrocess appears to be an attractive therapeutic strategy inNSCC and other human cancers.Although there is accumulating experimental evidence

upporting the role of CSCs in driving tumor growth andetastasis, clinical evidence for their driving role in the

ormation and progression of cancer is still sparse. In aecent study, we investigated the ALDH1+ CSCs frequencyn primary oropharyngeal squamous cell carcinoma (OSCC)nd its corresponding metastases [24]. OSCC with higherrequency of ALDH1+ cells present a more aggressive phe-otype characterized by higher nodal classification and lowerifferentiation. A higher number of ALDH1-expressing cellsere found more frequently present in lymph node metas-

ases than its corresponding primary tumors indicating thebility of CSCs to complete the metastatic cascade andevelop metastases. Studies from carcinomas of different ori-ins also presented traits of high-grade malignancy that coulde specifically traced back to the presence of CSCs [25–27].

.2. Interactions between human papillomavirus (HPV)

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

nd CSCs

A subgroup of head and neck cancer is associated with per-istent HPV infection thus rendering HPV infection a target

cwC

or therapy and prevention. The incidence of HPV+ cases inNSCC is rising, in particular in OSCC, while the incidencef HPV− cases induced by alcohol- or tobacco-abuse hasecreased in recent years with successfully established anti-moking campaigns [28]. Besides the epidemiological role ofPV in HNSCC, studies also addressed the role of HPV in

ancer progression and the differences of age, genetic back-round and prognosis between HPV+ and HPV− HNSCCatients [28]. With regard to these two different etiologies,nvestigations on the role of HPV with CSCs in initiating andpreading HNSCC are emerging but still have not been wellocumented.

In a recent study, Jung et al. investigated the role ofPV16-E6 and E7 oncogenes during the process of EMT

n vitro [29]. A stable expression of HPV16-E6 or E7nduced morphological conversion of epithelial cells to aesenchyme-like phenotype. In addition to these morpho-

ogical changes, both E6 and E7 induced expression of theMT-activating transcription factors Slug, Twist, ZEB1 andEB2. This study suggests that HPV16 could induce an EMT-

ike process. Tang et al. investigated whether the behaviorf CSCs is affected by the HPV-status or not in in vitrond in vivo studies [30]. In this study, the HPV+ cells andells transduced with HPV E6/E7 demonstrated a greater

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

lone forming capacity than HPV− cells. The HNSCC-CSCsere shown to be more resistant to cisplatin than the non-SCs. However, there was no difference of the proportion

ARTICLE IN PRESSONCH-1961; No. of Pages 9

4 Oncolog

oHtshteSmmsahwagH

2

scdiarafclfiteahstpwBAaMiiAgcuawc

3

stat[tseCTtsi

3

ioirctpAeeHtoiIecCAprHaCalssCaa

X. Qian et al. / Critical Reviews in

f CSCs between HPV+ and HPV− HNSCC. Moreover, thePV status did not affect the response of CSCs to cisplatin

herapy. Our previous study on human primary OSCC demon-trated that fewer ALDH1-expressing CSCs were found inigh-risk (HR)-HPV-DNA+/p16+ primary OSCC comparedo HR-HPV-DNA−/p16− primaries [24]. There were, how-ver, no similar findings in the corresponding metastases.imilarly, Rietbergen et al. has shown that HPV+ OSCC pri-ary tumors harbored a lower percentage of CSCs with thearkers of CD44 and CD98 than HPV− tumors. A better

urvival was found in HPV+ patients with a lower percent-ge of CD98+ tumor cells compared to HPV+ patients withigh fraction of CD98+ tumor cells [31]. Further studies arearranted to further elucidate the interactions between HPV

nd CSCs to identify a possible use of immunological tar-ets specific to the HPV-related etiology of this subgroup ofNSCC.

.3. Resistance to current therapy

In patients with locally advanced HNSCC treated byurgery and/or chemo-radiotherapy, loco-regional controlan be gained for some time but patients will frequentlyevelop a recurrence and/or metastasis. Despite intensivenvestigation of resistance mechanisms to chemo- and radi-tion therapy, our understanding of the relationship betweenesistance mechanisms and therapeutic success is still limitednd cannot be predicted. The CSC model offers explanationsor these treatment failures [1,32]. The tumor cells in a stemell-like state could escape from therapies acting against pro-iferating cells as “dormant” cells, a feature characteristicor CSC, and resume proliferation once the toxic pressures removed. Radioresistance of CSCs has been attributedo their self-renewal capacity, DNA repair capacity andnhanced reactive oxygen species (ROS) defence [15,33]. In

study, CD44+ALDH1+ cells isolated from HNSCC showedigher tumorigenicity, radioresistance, and high expres-ion of stemness-related genes (Bmi-1/Oct-4/Nanog). Whenreated with Cucurbitacin I additionally, the induction of apo-tosis and sensitivity to radiation of this CSC-populationas enhanced [34]. Chen et al. reported that silencing ofmi-1 significantly enhanced the sensitivity of HNSCC-LDH1+ CSCs to chemo-radiation that may be explained be

n increased degree of chemo-radiation mediated apoptosis.oreover, the experiment showed that knockdown of Bmi-1

ncreased the effectiveness of radiotherapy and resulted innhibition of tumor growth in nude mice transplanted withLDH1+ CSCs [35]. Recently, in HNSCC cell lines, Bour-uignon et al. [36] observed that CD44v3highALDH1high

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

ells were more resistant to cisplatin, the most commonlysed chemotherapy drugs for treatment of HNSCC. Theylso discovered a new HA-CD44v3-mediated signaling path-ay leading to the stimulation of apoptosis and enhanced

hemosensitivity in CSCs.

pampe

y/Hematology xxx (2015) xxx–xxx

. CSC-induced immune-responses

Current immunotherapy is mainly based on antigens pre-ented to effector T cells by dendritic cells (DCs). Generally,hese antigens are selected and derived from bulk tumor cellsnd target non-CSC tumor cells. They are not derived of CSCshat may not express immunogenic differentiation antigens37]. CSCs also may be defective in antigen presentation dueo the downregulation of human leukocyte antigen (HLA)urface expression [38]. Therefore, in a heterogeneous tumorntity, escaping from the attack of current immunotherapy,SCs may lead to a treatment failure and disease progression.hus, a better knowledge of the cross-talk between CSCs and

he immune system (Table 1, Fig. 1) and a development ofpecific therapies targeted at CSCs are of interest for furthermprovement of cancer immunotherapies.

.1. Recognition of CSCs by the immune system

One important issue for future immune therapy target-ng CSCs is the recognition and distinction of CSCs fromther cells by the host immune system. The immunogenic-ty of HNSCC-CSCs has been observed recently. Amongeported CSC markers, ALDH1 is considered a more spe-ific CSC marker than any of the other phenotypes usedo identify the small population of highly tumorigenic cellsresent in HNSCC and other carcinomas, as well [5,39–41].LDH1 has already been recognized as an antigen-source

liciting a humoral immune response in HNSCC. Visust al. [42] showed that ALDH1A188–96 peptide was anLA-A2-restricted, naturally presented, CD8+ T cell-defined

umor-antigen. ALDH1 peptide-specific CD8+ T cells couldnly recognize HLA-A2+ HNSCC cell lines overexpress-ng ALDH1 but not a human MRC fibroblast cell line.mportantly, potential toxicity against CD34+ hematopoi-tic progenitor cells by ALDH1A1 peptide-specific CD8+ Tells was excluded. By ELISPOT interferon (IFN)-� assays,D34+ hematopoietic progenitor cells isolated from HLA-2+ bone marrow cells were not recognized by ALDH1A1eptide-specific CD8+ T cells. Most importantly, targetecognition was blocked by anti-HLA class I and anti-LA-A2 antibodies. In addition to this defined CSC-tumor

ntigen, we recently compared the susceptibility of putativeSC-enriched spheroid culture-derived cells (SDC) gener-ted from two HNSCC cell lines and one cervical cancer celline to immunological recognition and killing by alloantigen-pecific CD8+ cytotoxic T lymphocytes (CTL) [13]. Thistudy was important to investigate immune suppression bySC and its reversal by IFN-� treatment. In order to avoidntigen-specific bias we used alloantigen-specific responsess strong and generic T cell targets. We found that CSCopulations were less sensitive to MHC class I-restricted

+

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

lloantigen-specific CD8 CTL lysis, as compared to theironolayer-derived cells (MDC). However, an additional

retreatment with IFN-� resulted in over-proportionallynhanced lysis of SDC. We also observed that the subset

ARTICLE IN PRESSONCH-1961; No. of Pages 9

X. Qian et al. / Critical Reviews in Oncology/Hematology xxx (2015) xxx–xxx 5

Table 1The relationship between CSCs and the immune system.

Immunological molecules expressed by CSCs Model Ref.

Similar expression of MHC class I between CSCs and bulk tumor cells HNSCC [13]Low expression of MHC class I Cervical cancer [13]No expression of MHC class I, low expression of MHC class II Glioma [14]Cancer/testis antigens Acute myeloid leukemia,

renal cell carcinoma[43,44]

Cancer/testis genes Lung adenocarcinoma, colonadenocarcinoma, breastadenocarcinoma

[45]

Defect in antigen presentation due to downregulation of HLA surface expression Colon cancer, pancreaticcancer, breast cancer

[38]

Immune responses induced by CSCsALDH1-specific CD8+ T cells recognize and eliminate CSCs HNSCC [42,67]Inhibition of the effector function of T lymphocytes including T cell proliferation and T cell

apoptosis by CSCsGlioma [14]

EMT CSCs inhibit CTL-mediated tumor cell lysis Breast cancer [54]Induction of FoxP3+ Tregs by CSCs Glioma [14]TAMs increase the tumorigenicity and drug resistance of CSCs by activating STAT3 and the Sonic

Hedgehog pathwayColon and lung cancer [66]

T pressio

ockoiatCbtp

wAlrMGSidn[t

3

iiafT

pIcsrshaLaaDmdCwwIaccdtwnuMsIei

AMs induce CSCs by activating STAT3; In turn, CSCs induce immunosupCD8+ T cell responses

f ALDHhigh expressing SDC was more sensitive than theirounterpart of ALDHlow SDC toward cognate CD8+ CTLilling. Moreover, the potentially different immunogenicityf CSCs and the differentiated bulk tumor cells was observedn our study. We found SDC derived of the cell line CaSkippeared to be more resistant to the recognition and destruc-ion by MHC class I-restricted alloantigen-specific CD8+

TL than the matched MDC. Together, the data presentedy the groups above have shown that the host immune sys-em is able to recognize and distinguish CSCs with ALDH1henotype from non-CSC cells.

In addition to ALDH1, some cancer/testis (CT) antigensere found to be preferentially expressed in CSCs. Cyclin-1 was reported in leukemic stem cells of acute myeloid

eukemia [43]. DNAJB8 was identified as novel CT antigen inenal CSCs [44]. Eighteen CT genes (MAGEA2, MAGEA3,

AGEA4, MAGEA6, MAGEA12, MAGEB2, GAGE1,AGE8, SPANXA1, SPANXB1, SPANXC, XAGE2,PA17, BORIS, PLU-1, SGY-1, TEX15 and CT45A1) exhib-

ted higher expression levels in CSCs than in non-CSCserived from LHK2 lung adenocarcinoma, SW480 colon ade-ocarcinoma and MCF-7 breast adenocarcinoma cells lines45]. The specific expression of CT antigens may enable uso target CSCs specifically in this manner.

.2. Immune escape of CSCs

It has been well documented that tumor cells can evade themmune system by altering their phenotype or by suppress-

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

ng immunity [46]. In HNSCC, defects in the MHC class Intigen processing machinery are observed and responsibleor the escape of tumor cells from recognition by cytotoxic

lymphocytes (CTL) [47,48]. This correlates with poor

edo

n in TAM and Pancreatic ductaladenocarcinoma

[8]

rognosis in patients with HNSCC [49]. Cytokines such asFN-� can restore antigen processing [47]. In addition, tumorells may downregulate MHC expression to escape immuneurveillance [50] as has also been shown for CSC [12]. Aecent study in melanoma and epithelial cancer cell lines hashown that SDC containing putative CSCs showed equal origher mRNA expression levels of molecules involved inntigen processing and presentation (APM) such as LMP2,MP7, and MECL-1, of APM molecules transporters-ssociated with antigen presentation (TAP) 1 and TAP2nd, also of TAA including differentiation antigens [38].ownregulation or loss of MHC class I and MHC class IIolecules in SDC was observed which was associated with

ecreased recognition of peptide-loaded disrupted SDC byD8+ T cells. Interestingly, MHC expression on tumor SDCas not responsive to stimulation with IFN-�. Previously,e investigated the expression of MHC class I, MHC class

I, and of immune recognition associated molecules on SDCnd MDC generated from two HNSCC cell lines and oneervical cancer cell line CaSki [13]. We found that MHClass I was expressed on SDC and MDC by all cell lines withifferent levels. Its expression on CaSki SDC was nearlywo times lower than on corresponding MDC, but thereere no difference in HNSCC cell lines. MHC class II wasegative on SDC and MDC in all cell lines. IFN-� treatmentpregulated the expression of MHC class I and inducedHC class II. In the above two studies, IFN-� treatment

howed different effects on the upregulation of MHC class and MHC class II. This may be due to different immunevasion mechanisms selected for in different tumor entities,ndividual patients and tumors or the isolated cell lines

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

mployed for the experimental research. Therefore thesesata can demonstrate the diversity of potential mechanismsf immune evasion. These findings, nevertheless, revealed a

ARTICLE IN PRESSONCH-1961; No. of Pages 9

6 Oncolog

pm

ewEnTsofoeeiftdepcpCcb

3

cttTsgolHspr

llsri[rlltHptm

ItBwaptCip[ilEilt(twMrciTTtmii

CvC

4

CpcC

gcclavA

X. Qian et al. / Critical Reviews in

otential defect of antigen-presenting functions of CSCs thatight lead to the protection of CSCs from T cell rejection.Some studies suggest that immunoedited tumor cells may

vade the immunosurveillance through EMT [51–53]. Ase described above, CSCs can metamorphose between anMT-state and non-EMT-state. The plasticity of the CSC phe-otype enables migration and tumor formation at distant sites.hus, one can hypothesize that CSCs undergoing EMT mayurvive the initial immune response and drive the regrowthf the tumor and development of metastasis. An importanteature of CSCs with their contribution to tumor evasionf the immunosurveillance has been observed recently. Weit al. reported that glioma CSCs could inhibit T cell prolif-ration and effector responses, trigger T cell apoptosis andnduce FoxP3+ regulatory T cells [14]. Later, Akalay et al.ound that the EMT phenotype acquired in various deriva-ives of MCF-7 human breast cancer cells was associated withramatic morphologic changes and actin cytoskeleton remod-ling, with CD24−/CD44+/ALDH+ stem cell populationsresent exhibiting a higher degree of EMT relative to parentalells [54]. In the same study, acquisition of an EMT-CSChenotype was found to be associated with an inhibition ofTL-mediated tumor cell lysis. In conclusion, these data indi-ate that the acquisition of an EMT phenotype by CSCs maye an additional mechanism of immune-escape by the tumor.

.3. Immune suppression by CSCs

Immunotherapeutic approaches for HNSCC are compli-ated due to the profound immune suppression induced byhis disease. Mechanisms such as increased apoptosis ofumor-specific CD8+ T cells and increased tumor-infiltrating

regulatory cells (Tregs) in peripheral blood and at the tumorite have been demonstrated [55–57]. HPV-encoded onco-enic proteins have been reported to downregulate expressionf MHC class I [58] and a higher level of tumor-infiltratingymphocytes was observed in HPV+ OSCC patients thanPV− OSCC patients [59]. While the majority of these

tudies performed in regard to bulk tumor cells, immunosup-ressive properties mediated by CSCs have been investigatedecently.

Work by Krishnamurthy et al. showed that CSCs areocated in close proximity to blood vessels and that endothe-ial cell-initiated signaling could enhance survival andelf-renewal of HNSCC-CSCs [7]. Clinically, patients withecurrent HNSCC showed an increased concentration of IL-6n serum, in comparison with patients with primary HNSCC60]. Elevated IL-6 levels could independently predict tumorecurrence, poor survival, and tumor metastasis [61,62]. Inine with this finding, Yu et al. demonstrated that secretionevels of IL-6 and sIL-6R from CSCs were vital to main-ain the self-renewal and tumorigenic properties of CSCs in

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

NSCC [60]. Tumor-associated macrophages (TAMs) maylay a critical role in tumor progression by interacting withhe tumor microenvironment [63]. Tregs are thought to pro-

ote tumor progression and link to a worse prognosis [64].

pOst

y/Hematology xxx (2015) xxx–xxx

t was reported that glioma TICs could induce FoxP3+ Tregshat were mediated by the costimulatory inhibitory molecule7-H1 and soluble Galectin-3 [14]. Further, the inductionas diminished by altering the differentiation of the TICs. In

study of primary human gliomas and a orthotopically trans-lanted syngeneic glioma model, the distribution of TAM athe invasive tumor front was correlated with the presence ofD133+ glioma CSCs. TAM could significantly enhance the

nvasive capability of glioma stem cells through paracrineroduction of TGF-�1 [65]. In another study of Jinushi et al.66], the role of TAMs in the regulation of CSC-activityn relation to drug resistance was identified. They found aarge amount of TAM-derived milk-fat globule EGF-8 (MFG-8) increased tumorigenicity and anticancer drug resistance

n CD44+ALDH+ colon tumor cells and CD133+ALDH+

ung cancer cells. Furthermore, MFG-E8 was found mainlyo activate signal transducer and activator of transcription-3STAT3) and Sonic Hedgehog pathways in CSCs and to fur-her amplify their anticancer drug resistance in cooperationith IL-6. More recently, a similar finding was addressed byitchem et al. [8] in pancreatic ductal adenocarcinoma. They

eported that TAMs directly induce TICs properties in pan-reatic cancer cells by activating STAT3. In turn, TICs inducemmunosuppression in TAMs, and block antitumor CD8+

-lymphocyte responses during chemotherapeutic treatment.argeting TAMs by inhibiting either the myeloid cell recep-

ors colony-stimulating factor-1 receptor or chemokine (C-Cotif) receptor 2 decreases the numbers of pancreatic TICs,

mproves chemotherapeutic efficacy, inhibits metastasis andncreases antitumor T cell responses.

Together, all these findings validated the interplay betweenSCs and the tumor immune microenvironment. However,ery little is known about the immune suppressive role ofSCs of HNSCC and other solid tumors so far.

. CSC-based vaccination

Previous investigations conducted on the immunity ofSC with ALDH1A1 phenotype presented an attractiveotential immunotherapeutic approach by targeting theseells. Studies on vaccination against antigen ALDH1A1+ ofSCs have been performed and achieved significant progress.

Visus et al. [67] have demonstrated the ability of in vivoenerated ALDH1A1-specific CTLs to eliminate ALDHbright

ells present in HLA-A2+ HNSCC, breast, and pancreasarcinoma cell lines, xenografts, and surgically removedesions in vitro. They also found antitumor activity bydoptive immunotherapy with ALDH1A1-specific CTLs inivo. ALDH1A1-specific CD8+ T cells could eliminateLDHbright cells, inhibit tumor growth and metastases, or

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

rolong survival of xenograft-bearing immunodeficient mice.f note, normal hepatocytes expressing ALDH1A1 were

hown to express little to no MHC class I Ag that madehem unlikely to be recognized by MHC class I-restricted,

ARTICLE IN PRESSONCH-1961; No. of Pages 9

X. Qian et al. / Critical Reviews in Oncology/Hematology xxx (2015) xxx–xxx 7

Table 2Studies of CSC-based vaccination.

Approach Model Source Ref.

Recognition of ALDH1A1-CSCs by ALDH188–96 peptide-specific CD8+ T cells HNSCC Human [42]Elimination of ALDHbright cancer cells by ALDH1A1-specific CD8+ T cells HNSCC, breast and

pancreatic carcinomaHuman [67]

SOX6-peptide-specific-CTLs could lyse glioma stem cells Glioma Human [70]CD8 defined prostate stem cell antigen specific T cell epitope could activate CD8+ effector T cells Prostate cancer Human [71]Dendritic cells (DC) loaded with neurospheres enriched with glioma CSCs could eliminate glioma

tumorsMurine glioma Murine [72]

Vaccination by DC loaded with CSCs could induce higher IFN-� production and prolong survival Glioblastoma Rat [50]Vaccination by DC loaded with CSCs could eliminate CSCs in vitro Murine melanoma

and squamous cellcarcinoma

Murine [68]

Murine prostate stem cell antigen encoding cDNA vaccination can induce long-term protectionagainst prostate cancer development

Murine prostatecancer

Murine [73]

Vaccine containing lysates of CSCs-enriched tumor cells could reduce tumor volume and occurrence Rat colon carcinoma Rat [69]V ively im

AptiCciwtmsoCdfvCgCosric

otv

5

cmiOi

obsciCe

adadoeCeptsmItpet

C

R

accination with defined human embryonic stem cells (hESCs) could effectmouse and rat ovarian cancer

LDH1A1-specific CD8+ T cells. Later, different to theserevious studies, an in vivo study on the host immune sys-em was used to selectively target CSCs. Ning et al. [68]nvestigated immunogenicity induced by murine ALDHhigh

SC used as source of antigen to prime DCs as a vac-ine for malignant melanoma and squamous cell carcinoman immunocompetent mice used as hosts. ALDHhigh CSCsere immunogenic and more effective as an antigen source

han unselected tumor cells in inducing protective antitu-or immunity. A high level of IgG produced by splenocytes

ubjected to CSC-tumor-lysate-pulsed DCs and the bindingf the antibody from CSC-vaccinated murine hosts to theSCs which resulted in the CSCs lysis via complement-ependent cytotoxicity have been observed. CTLs generatedrom peripheral blood mononuclear cells or splenocytes har-ested from CSC-vaccinated hosts were capable of killingSCs in vitro. Consistent with the findings of the aboveroups, Duarte et al. [69] first demonstrated an ALDHhigh

SC-based vaccine could reduce both tumor volume andccurrence in a rat colon carcinoma syngeneic model. In thistudy, 50% of the CSC-based vaccinated animals becameesistant to tumor development and CSC-based vaccinationnduced a 99.5% reduction in tumor volume compared to theontrol group.

These studies provided a greater view of the immune biol-gy of CSCs. Vaccination with CSCs (Table 2) has showno be effective in killing CSCs specifically, reducing tumorolume and preventing tumor recurrence.

. Conclusion

Understanding how the immune system affects CSCs inancer development and progression has been one of the

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

ost challenging questions in immunology. Recent researchndicates a dual role of CSCs with the immune system.n one hand, CSCs can be recognized and destroyed or

nhibited in their outgrowth by the immune system. On the

GNJ

munize against Ovarian cancer Human [74]

ther hand, CSCs can promote tumor progression eithery immunoediting for CSCs that are more suitable tourvive in an immunocompetent host or by establishingonditions that facilitate tumor outgrowth within the tumormmune-microenvironment. Therefore, specific targeting ofSCs by immunotherapeutic approaches may lead to morefficacious and lasting therapeutic results in the future.

Up till now, the discovery of specific tumor-associatedntigens expressed by CSCs as well as the progress ineveloping CSC-based vaccination, provides the potentialpplicability of targeting CSCs and may enable more rapidevelopment of combinational therapies that act effectivelyn CSCs. Nonetheless, it seems necessary to address sev-ral points before immunotherapeutic approaches targetingSCs can be brought into clinical trials. These include theffective isolation of CSCs from bulk tumor mass to measureotential immunotherapeutic effects on CSC, to determinehe antigen-profile presented on CSCs specifically to identifypecific CSC targets as well as the induction and enhance-ent of antigen processing and presentation of CSC epitopes.

n addition, it is also necessary to combine other aspects ofumor-mediated interference with the host immune system, inarticular in HPV+ and HPV− HNSCC (reviewed by Alberst al. [12]) with the effect of CSCs, and to develop novelherapy and improve therapeutic outcome.

onflict of interest

The authors have no conflict of interest to declare.

eviewers

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

Kazuaki Chikamatsu, M.D., Ph.D., Gunma Universityraduate School of Medicine, Otolaryngology-Head andeck Surgery, 3-39-22, Showa-machi, Maebashi 371-8511,

apan.

ARTICLE IN PRESSONCH-1961; No. of Pages 9

8 Oncolog

I

A

oPvM

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

X. Qian et al. / Critical Reviews in

Grzegorz Dworacki, M.D., Ph.D., Dept. of Clinicalmmunology, Rokietnicka 5D, Poznan, 60-806, Poland.

cknowledgments

This study was supported by the Key Science and Technol-gy Innovation Team of Zhejiang (2010R50048), Zhejiangrovincial Program for the Cultivation of High-level Inno-ative Health Talents and Key Laboratory of Laboratoryedicine, Ministry of Education, China.

eferences

[1] Albers AE, Chen C, Koberle B, et al. Stem cells in squamous head andneck cancer. Crit Rev Oncol Hematol 2012;81:224–40.

[2] Prince ME, Sivanandan R, Kaczorowski A, et al. Identification of asubpopulation of cells with cancer stem cell properties in head and necksquamous cell carcinoma. Proc Natl Acad Sci U S A 2007;104:973–8.

[3] Pries R, Witrkopf N, Trenkle T, Nitsch SM, Wollenberg B. Potentialstem cell marker CD44 is constitutively expressed in permanent celllines of head and neck cancer. In Vivo 2008;22:89–92.

[4] Chen C, Wei Y, Hummel M, et al. Evidence for epithelial-mesenchymaltransition in cancer stem cells of head and neck squamous cell carci-noma. PLoS ONE 2011;6:e16466.

[5] Clay MR, Tabor M, Owen JH, et al. Single-marker identification of headand neck squamous cell carcinoma cancer stem cells with aldehydedehydrogenase. Head Neck 2010;32:1195–201.

[6] Biddle A, Liang X, Gammon L, et al. Cancer stem cells in squamouscell carcinoma switch between two distinct phenotypes that are prefer-entially migratory or proliferative. Cancer Res 2011;71:5317–26.

[7] Krishnamurthy S, Dong Z, Vodopyanov D, et al. Endothelial cell-initiated signaling promotes the survival and self-renewal of cancerstem cells. Cancer Res 2010;70:9969–78.

[8] Mitchem JB, Brennan DJ, Knolhoff BL, et al. Targeting tumor-infiltrating macrophages decreases tumor-initiating cells, relievesimmunosuppression, and improves chemotherapeutic responses. Can-cer Res 2013;73:1128–41.

[9] Kindt N, Preillon J, Kaltner H, et al. Macrophage migration inhibitoryfactor in head and neck squamous cell carcinoma: clinical and experi-mental studies. J Cancer Res Clin Oncol 2013;139:727–37.

10] Borovski T, De Sousa EMF, Vermeulen L, Medema JP. Cancer stemcell niche: the place to be. Cancer Res 2011;71:634–9.

11] Rosenberg SA, Yang JC, Restifo NP. Cancer immunotherapy: movingbeyond current vaccines. Nat Med 2004;10:909–15.

12] Albers AE, Strauss L, Liao T, Hoffmann TK, Kaufmann AM. T cell-tumor interaction directs the development of immunotherapies in headand neck cancer. Clin Dev Immunol 2010;2010:236378.

13] Liao T, Kaufmann AM, Qian X, et al. Susceptibility to cytotoxic Tcell lysis of cancer stem cells derived from cervical and head and necktumor cell lines. J Cancer Res Clin Oncol 2013;139:159–70.

14] Wei J, Barr J, Kong LY, et al. Glioma-associated cancer-initiating cellsinduce immunosuppression. Clin Cancer Res 2010;16:461–73.

15] Diehn M, Cho RW, Lobo NA, et al. Association of reactive oxy-gen species levels and radioresistance in cancer stem cells. Nature2009;458:780–3.

16] Xu J, Muller S, Nannapaneni S, et al. Comparison of quantum dottechnology with conventional immunohistochemistry in examiningaldehyde dehydrogenase 1A1 as a potential biomarker for lymph nodemetastasis of head and neck cancer. Eur J Cancer 2012;48:1682–91.

17] Singh A, Settleman J. EMT, cancer stem cells and drug resistance: an

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

emerging axis of evil in the war on cancer. Oncogene 2010;29:4741–51.18] Floor S, van Staveren WC, Larsimont D, Dumont JE, Maen-

haut C. Cancer cells in epithelial-to-mesenchymal transition and

[

y/Hematology xxx (2015) xxx–xxx

tumor-propagating-cancer stem cells: distinct, overlapping or samepopulations. Oncogene 2011;30:4609–21.

19] Mani SA, Guo W, Liao MJ, et al. The epithelial–mesenchymal transitiongenerates cells with properties of stem cells. Cell 2008;133:704–15.

20] Sarrio D, Franklin CK, Mackay A, Reis-Filho JS, Isacke CM. Epithe-lial and mesenchymal subpopulations within normal basal breastcell lines exhibit distinct stem cell/progenitor properties. Stem Cells2012;30:292–303.

21] Shigeishi H, Biddle A, Gammon L, et al. Maintenance of stemcell self-renewal in head and neck cancers requires actions ofGSK3beta influenced by CD44 and RHAMM. Stem Cells 2013;31:2073–83.

22] Yang WH, Lan HY, Huang CH, et al. RAC1 activation mediates Twist1-induced cancer cell migration. Nat Cell Biol 2012;14:366–74.

23] Kong D, Li Y, Wang Z, et al. miR-200 regulates PDGF-D-mediatedepithelial–mesenchymal transition, adhesion, and invasion of prostatecancer cells. Stem Cells 2009;27:1712–21.

24] Qian X, Wagner S, Ma C, et al. ALDH1-positive cancer stem-likecells are enriched in nodal metastases of oropharyngeal squamous cellcarcinoma independent of HPV status. Oncol Rep 2013;29:1777–84.

25] Pang R, Law WL, Chu AC, et al. A subpopulation of CD26+ cancerstem cells with metastatic capacity in human colorectal cancer. CellStem Cell 2010;6:603–15.

26] Currie MJ, Beardsley BE, Harris GC, et al. Immunohistochemical anal-ysis of cancer stem cell markers in invasive breast carcinoma andassociated ductal carcinoma in situ: relationships with markers of tumorhypoxia and microvascularity. Hum Pathol 2013;44:402–11.

27] Jiang F, Qiu Q, Khanna A, et al. Aldehyde dehydrogenase 1 is atumor stem cell-associated marker in lung cancer. Mol Cancer Res2009;7:330–8.

28] Leemans CR, Braakhuis BJ, Brakenhoff RH. The molecular biology ofhead and neck cancer. Nat Rev Cancer 2011;11:9–22.

29] Jung YS, Kato I, Kim HR. A novel function of HPV16-E6/E7 inepithelial–mesenchymal transition. Biochem Biophys Res Commun2013.

30] Tang AL, Owen JH, Hauff SJ, et al. Head and neck cancer stem cells:the effect of HPV – an in vitro and mouse study. Otolaryngol HeadNeck Surg 2013;49:252–60.

31] Rietbergen MM, Martens-de Kemp SR, Bloemena E, et al. Cancerstem cell enrichment marker CD98: a prognostic factor for survivalin patients with human papillomavirus-positive oropharyngeal cancer.Eur J Cancer 2014;50:765–73.

32] Morrison R, Schleicher SM, Sun Y, et al. Targeting the mechanisms ofresistance to chemotherapy and radiotherapy with the cancer stem cellhypothesis. J Oncol 2011;2011:941876.

33] Ogawa K, Yoshioka Y, Isohashi F, et al. Radiotherapy targeting can-cer stem cells: current views and future perspectives. Anticancer Res2013;33:747–54.

34] Chen YW, Chen KH, Huang PI, et al. Cucurbitacin I suppressedstem-like property and enhanced radiation-induced apoptosis in headand neck squamous carcinoma-derived CD44(+)ALDH1(+) cells. MolCancer Ther 2010;9:2879–92.

35] Chen YC, Chang CJ, Hsu HS, et al. Inhibition of tumorigenicity andenhancement of radiochemosensitivity in head and neck squamous cellcancer-derived ALDH1-positive cells by knockdown of Bmi-1. OralOncol 2010;46:158–65.

36] Bourguignon LY, Wong G, Earle C, Chen L. Hyaluronan-CD44v3 inter-action with Oct4-Sox2-Nanog promotes miR-302 expression leadingto self-renewal, clonal formation, and cisplatin resistance in cancerstem cells from head and neck squamous cell carcinoma. J Biol Chem2012;287:32800–24.

37] Boiko AD, Razorenova OV, van de Rijn M, et al. Human melanoma-initiating cells express neural crest nerve growth factor receptor CD271.Nature 2010;466:133–7.

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

38] Busse A, Letsch A, Fusi A, et al. Characterization of small spheresderived from various solid tumor cell lines: are they suitable targets forT cells? Clin Exp Metastasis 2013;30:781–91.

ARTICLE IN PRESSONCH-1961; No. of Pages 9

Oncolog

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

B

li

Andreas E. Albers MD, PhD is a specialist in otorhinolar-yngology, head and neck surgery. His basic scientific interests

X. Qian et al. / Critical Reviews in

39] Chen YC, Chen YW, Hsu HS, et al. Aldehyde dehydrogenase 1 isa putative marker for cancer stem cells in head and neck squamouscancer. Biochem Biophys Res Commun 2009;385:307–13.

40] Isfoss BL, Holmqvist B, Alm P, Olsson H. Distribution of alde-hyde dehydrogenase 1-positive stem cells in benign mammarytissue from women with and without breast cancer. Histopathology2012;60:617–33.

41] Okamoto A, Chikamatsu K, Sakakura K, et al. Expansion and charac-terization of cancer stem-like cells in squamous cell carcinoma of thehead and neck. Oral Oncol 2009;45:633–9.

42] Visus C, Ito D, Amoscato A, et al. Identification of human aldehydedehydrogenase 1 family member A1 as a novel CD8+ T-cell-definedtumor antigen in squamous cell carcinoma of the head and neck. CancerRes 2007;67:10538–45.

43] Ochsenreither S, Majeti R, Schmitt T, et al. Cyclin-A1 representsa new immunogenic targetable antigen expressed in acute myeloidleukemia stem cells with characteristics of a cancer-testis antigen.Blood 2012;119:5492–501.

44] Nishizawa S, Hirohashi Y, Torigoe T, et al. HSP DNAJB8 controlstumor-initiating ability in renal cancer stem-like cells. Cancer Res2012;72:2844–54.

45] Yamada R, Takahashi A, Torigoe T, et al. Preferential expression ofcancer/testis genes in cancer stem-like cells: proposal of a novel sub-category, cancer/testis/stem gene. Tissue Antigens 2013;81:428–34.

46] Ferris RL, Whiteside TL, Ferrone S. Immune escape associated withfunctional defects in antigen-processing machinery in head and neckcancer. Clin Cancer Res 2006;12:3890–5.

47] Lopez-Albaitero A, Nayak JV, Ogino T, et al. Role of antigen-processing machinery in the in vitro resistance of squamous cellcarcinoma of the head and neck cells to recognition by CTL. J Immunol2006;176:3402–9.

48] Leibowitz MS, Srivastava RM, Andrade Filho PA, et al. SHP2 is over-expressed and inhibits pSTAT1-mediated APM component expression,T-cell attracting chemokine secretion, and CTL recognition in head andneck cancer cells. Clin Cancer Res 2013;19:798–808.

49] Ogino T, Shigyo H, Ishii H, et al. HLA class I antigen down-regulationin primary laryngeal squamous cell carcinoma lesions as a poor pro-gnostic marker. Cancer Res 2006;66:9281–9.

50] Xu Q, Liu G, Yuan X, et al. Antigen-specific T-cell response from den-dritic cell vaccination using cancer stem-like cell-associated antigens.Stem Cells 2009;27:1734–40.

51] Knutson KL, Lu H, Stone B, et al. Immunoediting of cancers may leadto epithelial to mesenchymal transition. J Immunol 2006;177:1526–33.

52] Gjerdrum C, Tiron C, Hoiby T, et al. Axl is an essential epithelial-to-mesenchymal transition-induced regulator of breast cancer metastasisand patient survival. Proc Natl Acad Sci U S A 2010;107:1124–9.

53] Kudo-Saito C, Shirako H, Takeuchi T, Kawakami Y. Cancer metastasisis accelerated through immunosuppression during Snail-induced EMTof cancer cells. Cancer Cell 2009;15:195–206.

54] Akalay I, Janji B, Hasmim M, et al. Epithelial-to-mesenchymal transi-tion and autophagy induction in breast carcinoma promote escape fromT-cell-mediated lysis. Cancer Res 2013;73:2418–27.

55] Albers A, Abe K, Hunt J, et al. Antitumor activity of human papillo-mavirus type 16 E7-specific T cells against virally infected squamouscell carcinoma of the head and neck. Cancer Res 2005;65:11146–55.

56] Albers AE, Ferris RL, Kim GG, et al. Immune responses to p53 inpatients with cancer: enrichment in tetramer+ p53 peptide-specific Tcells and regulatory T cells at tumor sites. Cancer Immunol Immunother2005;54:1072–81.

57] Hoffmann TK, Dworacki G, Tsukihiro T, et al. Spontaneous apoptosisof circulating T lymphocytes in patients with head and neck cancer andits clinical importance. Clin Cancer Res 2002;8:2553–62.

58] Bottley G, Watherston OG, Hiew YL, et al. High-risk human

Please cite this article in press as: Qian X, et al. Biology and immunoloOncol/Hematol (2015), http://dx.doi.org/10.1016/j.critrevonc.2015.03.0

papillomavirus E7 expression reduces cell-surface MHC class Imolecules and increases susceptibility to natural killer cells. Oncogene2008;27:1794–9.

ac

y/Hematology xxx (2015) xxx–xxx 9

59] King EV, Ottensmeier CH, Thomas GJ. The immune response in HPV+

oropharyngeal cancer. Oncoimmunology 2014;3:e27254.60] Yu CC, Tsai LL, Wang ML, et al. miR145 targets the SOX9/ADAM17

axis to inhibit tumor-initiating cells and IL-6-mediated paracrine effectsin head and neck cancer. Cancer Res 2013;73:3425–40.

61] Duffy SA, Taylor JM, Terrell JE, et al. Interleukin-6 predicts recur-rence and survival among head and neck cancer patients. Cancer2008;113:750–7.

62] Riedel F, Zaiss I, Herzog D, Gotte K, Naim R, Hormann K. Serum levelsof interleukin-6 in patients with primary head and neck squamous cellcarcinoma. Anticancer Res 2005;25:2761–5.

63] Duray A, Demoulin S, Hubert P, Delvenne P, Saussez S. Immunesuppression in head and neck cancers: a review. Clin Dev Immunol2010;2010:701657.

64] deLeeuw RJ, Kost SE, Kakal JA, Nelson BH. The prognostic value ofFoxP3+ tumor-infiltrating lymphocytes in cancer: a critical review ofthe literature. Clin Cancer Res 2012;18:3022–9.

65] Ye XZ, Xu SL, Xin YH, et al. Tumor-associated microglia/macrophagesenhance the invasion of glioma stem-like cells via TGF-beta1 signalingpathway. J Immunol 2012;189:444–53.

66] Jinushi M, Chiba S, Yoshiyama H, et al. Tumor-associated macrophagesregulate tumorigenicity and anticancer drug responses of can-cer stem/initiating cells. Proc Natl Acad Sci U S A 2011;108:12425–30.

67] Visus C, Wang Y, Lozano-Leon A, et al. Targeting ALDH(bright)human carcinoma-initiating cells with ALDH1A1-specific CD8(+) Tcells. Clin Cancer Res 2011;17:6174–84.

68] Ning N, Pan Q, Zheng F, et al. Cancer stem cell vaccination conferssignificant antitumor immunity. Cancer Res 2012;72:1853–64.

69] Duarte S, Momier D, Baque P, et al. Preventive cancer stem cell-basedvaccination reduces liver metastasis development in a rat colon carci-noma syngeneic model. Stem Cells 2013;31:423–32.

70] Ueda R, Ohkusu-Tsukada K, Fusaki N, et al. Identification of HLA-A2- and A24-restricted T-cell epitopes derived from SOX6 expressedin glioma stem cells for immunotherapy. Int J Cancer 2010;126:919–29.

71] Kiessling A, Schmitz M, Stevanovic S, et al. Prostate stem cell anti-gen: identification of immunogenic peptides and assessment of reactiveCD8+ T cells in prostate cancer patients. Int J Cancer 2002;102:390–7.

72] Pellegatta S, Poliani PL, Corno D, et al. Neurospheres enriched incancer stem-like cells are highly effective in eliciting a dendritic cell-mediated immune response against malignant gliomas. Cancer Res2006;66:10247–52.

73] Garcia-Hernandez Mde L, Gray A, Hubby B, Klinger OJ, Kast WM.Prostate stem cell antigen vaccination induces a long-term protectiveimmune response against prostate cancer in the absence of autoimmu-nity. Cancer Res 2008;68:861–9.

74] Zhang Z, Chen X, Chang X, Ye X, Li Y, Cui H. Vaccination withembryonic stem cells generates effective antitumor immunity againstovarian cancer. Int J Mol Med 2013;31:147–53.

iographies

Andreas M. Kaufmann PhD is a biologist in gyneco-ogic research. His interest is in HPV, cancer biology, andmmunotherapy.

gy of cancer stem(-like) cells in head and neck cancer. Crit Rev09

re HPV, the tumor- and immunobiology of head and neckancer and its cancer stem cells.