Cell Research (2017) 27:109-118.

© 2017 IBCB, SIBS, CAS All rights reserved 1001-0602/17 $ 32.00

REVIEW www.nature.com/cr

Regulatory T cells in cancer immunotherapy

Atsushi Tanaka1, 2, Shimon Sakaguchi1
1
Laboratory of Experimental Immunology, WPI Immunology Frontier Research Center, Osaka University, Suita 565-0871, Japan;
2
Department of Frontier Research in Tumor Immunology, Graduate School of Medicine, Osaka University, Suita 565-0871, Japan

FOXP3-expressing regulatory T (Treg) cells, which suppress aberrant immune response against self-antigens, also
suppress anti-tumor immune response. Infiltration of a large number of Treg cells into tumor tissues is often associ-
ated with poor prognosis. There is accumulating evidence that the removal of Treg cells is able to evoke and enhance
anti-tumor immune response. However, systemic depletion of Treg cells may concurrently elicit deleterious autoim-
munity. One strategy for evoking effective tumor immunity without autoimmunity is to specifically target terminally
differentiated effector Treg cells rather than all FOXP3+ T cells, because effector Treg cells are the predominant cell
type in tumor tissues. Various cell surface molecules, including chemokine receptors such as CCR4, that are specifi-
cally expressed by effector Treg cells can be the candidates for depleting effector Treg cells by specific cell-depleting
monoclonal antibodies. In addition, other immunological characteristics of effector Treg cells, such as their high ex-
pression of CTLA-4, active proliferation, and apoptosis-prone tendency, can be exploited to control specifically their
functions. For example, anti-CTLA-4 antibody may kill effector Treg cells or attenuate their suppressive activity. It
is hoped that combination of Treg-cell targeting (e.g., by reducing Treg cells or attenuating their suppressive activity
in tumor tissues) with the activation of tumor-specific effector T cells (e.g., by cancer vaccine or immune checkpoint
blockade) will make the current cancer immunotherapy more effective.
Keywords: Treg; CTLA-4; cancer; immunotherapy
Cell Research (2017) 27:109-118. doi:10.1038/cr.2016.151; published online 20 December 2016

Introduction Among the various mechanisms of immunological

self-tolerance, immune suppression by endogenous Fox-
A number of studies have shown that self-antigen or p3+CD25+CD4+ Treg cells is essential and indispensable
tumor antigen-specific CD4+ and CD8+ T cells are present as illustrated by spontaneous autoimmune disease devel-
in healthy individuals [1-4]. How such self- or tumor-re- opment when Treg cells are rendered deficient. For ex-
active T cells are controlled in healthy or tumor-bearing ample, mutations of the gene encoding the Treg-specific
individuals remains to be determined. Mechanisms for transcription factor Foxp3 impair Treg cell development
the maintenance of immunological self-tolerance (i.e., and cause a fatal multi-organ autoimmune disease called
unresponsiveness to self-antigens) not only prevent au- immune dysregulation, polyendocrinopathy, enteropathy,
toimmunity but also hamper effective tumor immunity and X-linked (IPEX) syndrome [6]. Depletion of Fox-
because many tumor antigens recognized by autologous p3+CD25+CD4+ Treg cells by a variety of methods is also
lymphocytes are normal self-antigens or quasi-self-anti- able to cause similar autoimmune diseases in otherwise
gens with genetic mutations. This is one reason why it is normal rodents [7].
difficult to elicit strong tumor immunity in cancer-bear- On the other hand, it is now well substantiated that a
ing patients by cancer vaccine alone [5]. It also suggests large number of Treg cells infiltrate into tumor tissues
that effective tumor immunity can be evoked by breach- of various cancers and their abundant presence is often
ing a certain mechanism(s) of immunological self-toler- associated with poor clinical prognosis. Experimentally,
ance systemically or locally in tumor tissues. the role of Treg cells in tumor immunity was first demon-
strated by an attempt to determine a common basis be-
tween tumor immunity and autoimmunity [8]. Removal
Correspondence: Shimon Sakaguchi of Treg cells using cell-depleting anti-CD25 antibody,
E-mail: shimon@ifrec.osaka-u.ac.jp either by in vivo antibody administration to mice or
Regulatory T cells in cancer immunotherapy
110

transfer of cell suspension depleted in vitro of CD25+ nosine, and cAMP [6] (Figure 1 and Table 1). Given that
Treg cells into histocompatible T-cell-deficient mice, ectopic Foxp3 expression in Tconv cells is able to confer
effectively eradicated a variety of inoculated syngeneic Treg-like suppressive activity, the molecule(s) mediating
tumors [8, 9]. The mice showed an increase of tumor-in- a core suppressive mechanism may well be controlled
filtrating CD8+ T cells with strong tumor-specific killing by Foxp3. In addition, among various mechanisms of
activity, and upon re-challenge with the same tumor Treg-dependent suppression, those important for main-
cells, exhibited more rapid rejection than the primary taining self-tolerance (i.e., the suppression mechanisms
rejection, indicating the establishment of tumor-specific whose impairment causes autoimmune disease) have the
immunity [8, 10]. These studies have thus demonstrated most impact on tumor immunity. On these assumptions,
that the removal of Treg cells is able to evoke effective there are only a few molecules whose expression is con-
anti-tumor immunity by abrogating immunological un- trolled by Foxp3 directly or indirectly and whose defi-
responsiveness to syngeneic tumors, albeit it may also ciency abrogates Treg-suppressive function and causes
cause autoimmunity, especially if Treg cells are depleted severe autoimmune diseases. The candidates include IL-
systemically. 2, IL-2 receptor subunits, and CTLA-4. Foxp3 indeed
In this review, we discuss molecular basis of Treg controls the expression of these molecules and deficien-
functions and their behavior in tumor tissues, and strat- cies of IL-2, CD25 (IL-2 receptor α-chain), CD122 (IL-
egies to target Treg cells, in particular their subsets, in 2 receptor β-chain), or CTLA-4 produce similar autoim-
order to evoke effective anti-tumor immunity in humans, mune diseases as observed in Foxp3 deficiency [6].
without eliciting deleterious autoimmunity. A cardinal feature of Treg cells is that they consti-
tutively express high-affinity IL-2 receptor composed of
Treg cell function in relation to tumor immunity CD25, CD122, and CD132 (common γ-chain) at a high lev-
el, but scarcely produce IL-2. Treg cells therefore highly de-
T-cell receptor repertoire of Treg cells pend on exogenous IL-2, which is mainly produced by acti-
The T-cell receptor (TCR) repertoire of Treg cells vated Tconv cells, for their survival and proliferation. Foxp3
is broad and skewed to a certain extent to recognizing binds to and attenuate AML1 and NFAT, two transcription
self-antigens. That is, in the course of T-cell selection in factors necessary for IL-2 production, thus repressesing
the thymus, a developing Treg cell exhibits a higher TCR IL-2 production [15, 16]. In fact, neutralization of circu-
affinity than a conventional T (Tconv) cell for the MHC/ lating IL-2 by administration of anti-IL-2 antibody can
self-peptide ligand that selects both [11]. Assuming that compromise Treg function and survival, causing severe
TCR recognition of peptides is cross-reactive (and de- autoimmunity as produced by Treg deficiency [6]. On
generate) and a particular TCR is able to recognize a mil- the other hand, the constitutive expression of high-af-
lion different peptides of 10 amino acid length [12, 13], finity IL-2 receptor combined with self-insufficiency in
the TCR repertoire of Treg cells as well as Tconv cells IL-2 production determines that Treg cells absorb IL-2
is broad and able to recognize a wide spectrum of self from their surroundings, thus limiting the amount of IL-2
and non-self antigens including quasi-self-tumor anti- available for Tconv cells and consequently suppressing
gens. Given the antigen-primed state of endogenous Treg the activation and proliferation of the latter [17]. Accord-
cells (as illustrated by higher level expression of T-cell ingly, addition of exogenous IL-2 abrogates Treg-sup-
accessory molecules such as LFA-1), it is reasonable to pressive function in vitro [14, 18]. Therefore, IL-2 and
assume that Treg cells recognizing a particular self- or IL-2 receptor can be key targets for controlling Treg cell
tumor antigen are more easily activated than naive Tconv survival and suppressive function [19].
cells recognizing the same antigen, ensuring Treg-medi- CTLA-4 is a highly potent co-inhibitory molecule
ated dominant tolerance [14]. expressed constitutively by Treg cells and by Tconv cells
after activation. In mice, Treg-specific deletion of CTLA-
Treg-mediated suppression mechanisms 4 elicits systemic hyper-proliferation of Tconv cells and
Treg cells are able to control not only T cells but also fatal autoimmune diseases affecting multiple organs, in-
B cells, NK cells, dendritic cells (DCs), and macro- cluding severe myocarditis [20]. Recently, heterozygous
phages via humoral and cell-cell contact mechanisms [6]. CTLA-4 mutations in humans were identified in patients
A variety of molecules are involved in Treg-mediated with multiple autoimmune symptoms accompanied by
suppression mechanisms, including CTLA-4 (cytotoxic impaired suppressive function of Treg cells [21, 22]. As
T-lymphocyte-associated protein 4), IL-2, IL-10, TGF-β, CTLA-4 has much higher affinity than CD28 for their
IL-35, GITR (glucocorticoid-induced TNF receptor), common ligands CD80 and CD86, CTLA-4 expressed by
LAG3 (lymphocyte-activation gene 3), granzyme B, ade- Treg cells may physically outcompete CD28 on Tconv

SPRINGER NATURE | Cell Research | Vol 27 No 1 | January 2017

 Atsushi Tanaka and Shimon Sakaguchi
111

Figure 1 Treg suppression mechanisms. Treg cells, which scarcely produce IL-2, deprive IL-2 from the surrounding via
their high affinity IL-2 receptor, making it unavailable for responder T cells. They also constitutively express CTLA-4, which
down-modulates CD80/CD86 expression by antigen-presenting cells (APCs), thus depriving co-stimulatory signal to respond-
er T cells. Treg cells also produce immune-suppressive cytokines such as IL-10, which also down-modulates APC functions.
Under this deprivation of co-stimulatory signal, responder T cells with high-affinity TCRs for the presented antigen die by
apoptosis, those with intermediate affinity TCRs are rendered anergic, and those with low-affinity TCRs stay dormant. This
IL-2/IL-2 receptor-dependent and CTLA-4-dependent mechanism forms a core basis of Treg-mediated suppression in various
tissues including cancer.

Table 1 Key mechanisms of suppression by Treg cells

Molecule(s) Mechanism of suppression
IL-2 receptor/IL-2 Constitutive expression of high affinity IL-2 receptor α chain (CD25) and dependency on
exogenous IL-2 by Treg cells together limit the amount of IL-2 available to Tconv cells,
thereby hindering the activation and proliferation of the latter
CTLA-4 Constitutively expressed CTLA-4 on Treg cells preferentially binds to and downregulates
CD80/CD86 co-stimulatory molecules on antigen-presenting cells, depriving Tconv cells of
the co-stimulatory signal
IL-10 and other immune-suppressive Treg cells produce immune-suppressive cytokines, such as IL-10 and TGF-β; form extracellular
cytokines and substances adenosine from ATP by CD39 and CD73; and can also mediate direct killing of Tconv or antigen-
presenting cells by secreting granzymes

cells for binding to CD80/CD86 on antigen-presenting 25, 26]. Thus, Treg expression of CTLA-4 is essential for
cells, thereby inhibiting co-stimulation of Tconv cells Treg-mediated immune suppression.
[23, 24]. Furthermore, CTLA-4 on Treg cells down-mod- Treg cells require TCR stimulation to exert suppres-
ulates expression of CD80 and CD86 on DCs, thereby sive activity; without antigen stimulation, they fail to
hindering the activation of Tconv cells at this level [20, suppress immune response [14]. TCR stimulation of Treg

www.cell-research.com | Cell Research | SPRINGER NATURE

Regulatory T cells in cancer immunotherapy
112

cells further upregulates CTLA-4 and other accessory Treg migration to nonlymphoid tissues
molecules, particularly adhesion molecules such as LFA- Treg cells are widely distributed in both lymphoid
1, whose deficiency compromises suppressive activity. and nonlymphoid tissues, including tumors, where they
When TCR affinity for a stimulating antigen is the same affect immune responses and inflammation. Treg cells
between Treg cells and Tconv cells, Treg cells can be ac- in nonlymphoid tissues are predominantly effector Treg
tivated to exert suppression at a much lower antigen con- cells (CD44hi CD62Llo) and are highly proliferative [36].
centration than Tconv cells [14]. This indicates that con- In mice, migration of Treg cells to nonlymphoid tissues,
stitutively high expression of LFA-1 and other accessory such as the skin and the lungs, requires chemokine re-
molecules prior to TCR stimulation may contribute to ceptor CCR4 (C-C chemokine receptor type 4) [37]. Treg
setting a lower threshold for TCR-induced activation of cells deficient in CCR4 expression fail to migrate, and
Treg cells [27]. On the other hand, another key feature of this causes severe inflammatory diseases in the skin and
Treg cells is their hypoproliferation upon TCR stimula- the lungs.
tion in vitro; albeit a fraction of Treg cells can proliferate
actively in response to TCR stimulation in vivo [14, 28- Cell fate and function of Treg-suppressed Tconv
29]. An excessive amount of IL-2 or agonistic anti-CD28 cells
antibody is able to abrogate the in vitro hypo-responsive-
ness. Ectopic expression of Foxp3 in CD4+ T cells also Until recently, the fate of responder T cells after they
converts them into a state of hypo-responsiveness upon are suppressed by Treg cells was poorly understood (e.g.,
TCR stimulation [30]. It has also been recently demon- if they die by apoptosis, become anergic, or stay dormant
strated that Treg cells, especially Ki-67+ Treg cells, are during the period of suppression). A recent in vitro study
highly dependent on tonic TCR signaling for their pro- has now shown that the presence of a large number of
liferation [31, 32], and ablation of TCR signaling results Treg cells can render self-antigen-/tumor antigen-specific
in caspase-mediated apoptosis of these cells in mice. It human CD8+ T cells (e.g., recognizing Melan-A antigen)
is also of note that Foxp3-dependent repression of the anergic [38]. These anergic CD8+ T cells are hyporespon-
expression of proximal TCR signaling molecules (e.g., sive to antigen stimulation (i.e., hypoproliferative and
ZAP-70) in Treg cells may contribute to their unique sen- producing very little IL-2 and other cytokines), express
sitivity to TCR stimulation. For example, whereas TCR co-inhibitory molecules such as CTLA-4, and are pheno-
stimulation upregulates the expression of ZAP-70 in typically naive (i.e., CD45RAhigh and CCR7+). Induction
Tconv cells, it downregulates ZAP-70 expression in Treg of anergy depends on the intensity of Treg suppression
cells [33]. Foxp3 binds to the promoter region of ZAP-70 (determined by the number and suppressive activity of
gene [33-35] and retroviral expression of Foxp3 in Tconv the Treg cells) and TCR affinity of the responder T cells.
cells reduces their ZAP-70 expression [33]. It remains to Upon Treg-mediated down-modulation of CD80/CD86
be determined how TCR signaling attenuation at the lev- expression by antigen-presenting cells, as discussed
el of ZAP-70 may contribute to Treg-specific functions. above, T cells with high affinity TCR for the stimulating
It might affect Treg cell selection in the thymus to skew antigen die by apoptosis, whereas T cells with interme-
the TCR repertoire toward higher self-reactivity [11]. diate affinity TCR are rendered anergic, and T cells with
As another possibility, the TCR signal attenuation might low-affinity TCR stay dormant. Thus, Treg cells can
rescue Treg cells from activation-induced cell death upon induce long-term self-tolerance, and hinder effective tu-
antigen exposure, enabling them to survive better than mor immunity, by determining the fate of self-reactive or
Tconv cells and exert dominant control over the latter at tumor-reactive T cells (Figure 1).
tumor sites. It is of note that a sizable fraction of self-antigen-spe-
In summary, constitutively high expression of CD25 cific CD8+ T cells (e.g., those specific for Melan-A as
and CTLA-4, dependency on exogenous IL-2, and TCR detected by peptide/MHC tetramer staining) present
stimulation have essential roles in producing and shaping in the peripheral blood of normal individuals are CT-
Treg functions, especially Treg-mediated suppression. LA-4-expressing but naive in phenotype (i.e., CCR7+
Indeed, adapting this triad is able to convert Tconv cells and CD45RAhigh) and functionally anergic. In addition,
into Treg-like suppressive T cells effective in vivo and CTLA-4-negative naive CD8+ T cells positive for the
in vitro, indicating these three events are minimally re- same tetramer staining can be activated to proliferate in
quired for constructing Treg-suppressive activity [26]. response to antigen stimulation. It appears that the latter
These molecular processes are also good targets to con- population of non-anergic naive T cells give rise to effec-
trol Treg function and development in the context of tu- tor T cells to cause autoimmune disease or mediate effec-
mor immunity. tive tumor immunity when Treg-mediated suppression is

SPRINGER NATURE | Cell Research | Vol 27 No 1 | January 2017

 Atsushi Tanaka and Shimon Sakaguchi
113

breached, as discussed below. cells (Fr. III), which do not possess suppressive activity
but can secrete pro-inflammatory cytokines. Furthermore,
Functional/phenotypic heterogeneity of human adjacent populations of highly suppressive Fr. II and
FOXP3+ T cells non-suppressive Fr. III can be better differentiated by the
expression of CD15s (sialyl Lewis x), a sugar antigen, on
There is accumulating evidence that FOXP3+ T cells suppressive Treg cells, at least for those in the peripheral
in humans are heterogeneous in phenotype and func- blood [40]. This classification of FOXP3+CD4+ T cells is
tion, consisting of suppressive and non-suppressive instrumental in defining suppressive or non-suppressive
subpopulations. For example, naive CD4+ T cells tran- FOXP3+ subpopulations, delineating developmental stag-
siently express FOXP3 at a low level upon in vitro TCR es of Treg cells, and assessing their adaptive processes in
stimulation; yet they are not measurably suppressive. physiological and pathological immune responses.
Attempts to delineate suppressive and non-suppressive
FOXP3 +CD4 + T cells present in the peripheral blood Tumor-infiltrating Treg cells
of healthy individuals have shown that FOXP3+CD4+
T cells can be divided into three subpopulations based Accumulating studies have demonstrated that a large
on expression levels of FOXP3 (or CD25) and the cell number of Treg cells infiltrate into various types of tu-
surface molecule CD45RA [39] (Figure 2). They are: mors in humans and mice. In humans, tumors in the head
(i) FOXP3loCD45RA+CD25lo cells (Fraction I or Fr. I), and neck [41], breast [42], lung [43], liver [44], gastroin-
designated as naive or resting Treg cells; (ii) FOXP3hiC- testinal tract [45, 46], pancreas [47], and ovary [48, 49]
D45RA–CD25hi cells (Fr. II), designated as effector or have been shown to harbor a large number of tumor-in-
activated Treg cells, which are terminally differentiated filtrating Treg cells [50]. Importantly, decreased ratios
from Fr. I cells upon antigen stimulation and highly sup- of tumor-infiltrating CD8+ T cells to FOXP3+ Treg cells
pressive; and (iii) FOXP3loCD45RA–CD25lo non-Treg were shown to correlate with poor prognosis, especially

Figure 2 Functional classification of human FOXP3+CD4+ T-cell subpopulations in tumor tissue. Human FOXP3+ T cells in the
peripheral blood and lymph nodes are composed of heterogeneous subpopulations containing suppressive Treg cells (naive
and effector Treg cells) and activated non-Treg cells without suppression function. These subpopulations are designated as
Fraction (Fr.) I, II, and III for naive Treg (nTreg), effector Treg (eTreg), and non-Treg cells, respectively. CD25 surface mark-
er can be used in the place of FOXP3 because of their correlative expression in humans. Majority of cancers are infiltrated
predominantly by effector Treg cells, whereas certain cancers are infiltrated by both effector Treg cells and non-Treg cells.
Tumor-infiltrating effector Treg cells predominantly express various cell surface molecules including CTLA-4, CCR4, and PD-1.

www.cell-research.com | Cell Research | SPRINGER NATURE

Regulatory T cells in cancer immunotherapy
114

in patients with breast [51], gastric [46], and ovarian can- fully humanized monoclonal antibodies against CTLA-
cer [48, 49]. Furthermore, a recent meta-analysis of pre- 4 (Ipilimumab and Tremelimumab) have been tested in
viously published data indicated that in majority of solid patients with various cancer types, including melanoma,
tumors in the cervix, kidney, breast, and melanomas, prostate cancer, and renal cell carcinomas. The results
high frequency of tumor-infiltrating FOXP3+ cells was from a large, randomized, multi-center phase III clinical
significantly negatively correlated with patients’ survival study for Ipilimumab showed a significant advantage in
[52]. metastatic melanoma patients [59, 60]. Yet, how Ipili-
In contrast, prognosis of patients with Hodgkin lym- mumab augments anti-tumor immunity is still unclear.
phoma or colorectal cancer (CRC) containing increased CTLA-4, as discussed above, is constitutively expressed
numbers of Treg cells has been controversial [53-57]. on Treg cells, whereas it is transiently expressed by
Some studies have indicated that tumor infiltration of conventional T cells upon activation. Although anti-CT-
FOXP3+ T cells is correlated with a better prognosis, LA-4 mAb was first suggested to augment the activity of
whereas others have shown the contrary. These contra- tumor-infiltrating CD8+ and CD4+ T cells, recent studies
dictory results may arise from an improper interpretation have suggested another possibility that anti-CTLA-4
of the heterogeneous FOXP3+ cells (i.e., functional Treg mAb predominantly affects Treg cells, thereby enhancing
cells and non-Treg cells as described above) as a sin- anti-tumor immune responses. Using Fc receptor-defi-
gle population of Treg cells. Using the function-based cient mice, the anti-tumor activity of anti-CTLA-4 mAb
scheme to classify FOXP3+ cells infiltrating the CRC was shown to be dependent on antibody-dependent cel-
tissues, our recent study has shown that CRC patients lular cytotoxicity of Treg cells in tumor tissues instead of
can be separated into two groups: one group with tumors affecting re-activation of Tconv cells [61-63]. In cancer
infiltrated predominantly by suppression-competent patients, strong correlations have been reported between
effector Treg cells and the other group with tumors infil- the clinical efficacy of Ipilimumab and reduction of Treg
trated with a sizable fraction of FOXP3+ non-Treg cells cell numbers in tumor tissues [64-65]. It is likely that the
in addition to effector Treg cells [58]. In the latter group, Treg-depleting effect of anti-CTLA-4 mAb targets effec-
FOXP3+ non-Treg cells secrete inflammatory cytokines tor Treg cells, which are abundant in tumor tissues and
and the cytokine production is correlated with expression express high levels of CTLA-4. It remains to be deter-
of TGF-β and IL-12 genes in the tumor tissue. In this mined whether anti-PD-1 antibody (Nivolumab), another
group, high FOXP3 gene expression shows significantly checkpoint blockade antibody, possesses a Treg-deplet-
better prognosis than low FOXP3 gene expression. In ing effect in tumor tissues.
contrast, in the group with tumors infiltrated predomi-
nantly by effector Treg cells, high FOXP3 gene transcrip- Depletion of effector Treg cells in tumor tissues
tion indicates poor prognosis compared with low FOXP3 As discussed above, in various cancers effector Treg
transcription. Thus, it is critically important to assess cells are the most abundant cell type among FOXP3+
heterogeneity of FOXP3+ T cells in the tumor tissues in T cells in the tumor tissues. In order to distinguish and
order to evaluate their contribution to anti-tumor immune selectively deplete tumor-infiltrating Treg cells while
response in various cancers (Figure 2). preserving other Treg cells critical for suppressing auto-
immunity, one strategy is to specifically target effector
Cancer immunotherapy targeting Treg cells Treg cells, which are highly activated, proliferative, and
prone to death by apoptosis [66]. As effector Treg cells
Recent progress in cancer immunotherapy targeting are the predominant population in tumor tissues, deplet-
Treg cells, either deliberately or inadvertently, suggests ing effector Treg cells would shift the balance in tumor
that molecules relatively specific to Treg cells are good microenvironment from immune suppression to immune
candidates for Treg depletion or functional modulation. activation against tumor cells, despite a continual supply
These molecules include CTLA-4, GITR, CCR4, PD-1, of effector Treg cells from intact naive Treg cells. For
OX-40, and LAG3, as well as aforementioned CD25 and this purpose, surface molecules expressed specifically
CD15s (Figure 2). or selectively on effector Treg cells are good targets. For
example, CCR4 is predominantly expressed by effector
Checkpoint blockade antibody with possible Treg-deplet- Treg cells, not by naive Treg cells and Th2 cells which
ing effects do not contribute significantly to tumor immunity regula-
One of the recent breakthroughs in cancer immuno- tion in humans [67], and Treg migration and infiltration
therapy is the clinical use of anti-CTLA-4 antibody, of- into various tumor tissues appear to be dependent on the
ten referred to as the checkpoint blockade therapy. Two expression of CCR4 ligands (i.e., CCL17 and CCL22)

SPRINGER NATURE | Cell Research | Vol 27 No 1 | January 2017

 Atsushi Tanaka and Shimon Sakaguchi
115

produced by tumor cells or infiltrating macrophages [48, molecules, which are differentially controlled in Treg
68]. Indeed, the use of anti-CCR4 antibody has been and Tconv cells. As discussed above, ZAP-70 is specif-
shown to be effective in depleting effector Treg cells ically repressed in Treg cells upon TCR activation [33].
selectively and augmenting the induction of tumor anti- Targeting ZAP-70 may therefore enable selective reduc-
gen-specific CD4+ and CD8+ T cells in vivo [67]. tion of TCR signaling by interfering with TCR proximal
signaling molecules, resulting in selective death of Treg
Agonistic antibody affecting Treg suppression cells, in particular effector Treg cells, due to signal depri-
GITR is another molecule that is expressed by Treg vation-induced apoptosis (Tanaka et al., unpublished).
cells and can serve as a target for functional modulation. Similarly, an inactivating mutation (D910A mutation)
Treg cells constitutively express high level of GITR com- of phosphatidylinositol-3-kinase (PI3K) p110δ or condi-
pared with Tconv cells; however, upon activation, Tconv tional knockout of PI3K in Treg cells in mice effectively
cells also express GITR, whereas the level of GITR on augmented anti-tumor immune responses without incur-
Treg cells is further upregulated. In mice, administration ring autoimmunity in the mutant mice [79]. Although
of agonistic anti-GITR antibody (non-depleting) can these findings need to be confirmed in humans, these
abrogate Treg cell-mediated suppressive function and en- studies suggest that TCR-proximal signaling molecules
hance the effector function of Tconv cells to break down can be good targets for developing small molecule com-
immunological self-tolerance [69-71]. The same antibody pounds to achieve selective depletion of tumor-infiltrat-
in tumor-bearing mice can indeed break immunological ing effector Treg cells.
self-tolerance and evoke potent anti-tumor immune re-
sponse with an increase in IFN-γ-producing CD8+ and Autoimmunity and Treg-targeting cancer immuno-
CD4+ T cells, which can eradicate established tumors [72]. therapy
Furthermore, a chimeric anti-GITR antibody generated
by replacing the Fc portion of the original rat IgG2b with Cancer immunotherapies such as CTLA-4 and PD-1
a murine IgG2a has shown effective tumoricidal activity blockade are frequently accompanied by serious auto-
by selectively depleting Treg cells in mice [61]. The ag- immunity [59, 80, 81]. In general, the effectiveness of
onistic antibody for GITR is under clinical trials to test anti-tumor responses tends to correlate with the develop-
its efficacy in melanoma and other advanced solid tumor ment of autoimmune diseases [82, 83], especially when a
patients. Agonistic antibodies specific for other TNFR systemic Treg cell depletion approach is adopted. There
super family molecules, such as OX40, are under clinical are, however, several ways to evoke effective tumor im-
investigation [66]. munity without causing serious autoimmune reaction by
targeting specific populations of Treg cells. One is to se-
Small molecules for Treg depletion or functional modu- lectively target effector Treg cells in tumor tissues, thus
lation preserving the naive Treg cells in other tissues needed
In addition to antibody-mediated Treg depletion thera- to prevent autoimmunity. The second strategy aims to
py, small molecule drugs targeting characteristic features adjust the degree and duration of Treg depletion in order
of Treg cells have been shown to be effective. One ex- to promote tumor immunity while curbing autoimmune
ample is metronomic administration of cyclophospha- responses, taking advantage of the observation that in-
mide, an anti-neoplastic used frequently in traditional duction of autoimmune disease generally needs more
chemotherapy. As cyclophosphamide is an alkylating thorough and durable Treg cell depletion than induction
agent that interferes with DNA replication, it kills highly of anti-tumor immune responses. An additional approach
proliferating cells. Administration of high-dose cyclo- to control tumor-infiltrating Treg cells is direct injection
phosphamide severely affects all T cells. However, when of Treg-depleting antibodies to tumor tissues. It is im-
used at low doses over a long term, the drug has been portant to note the efficacy of Treg-targeting anti-tumor
shown to selectively reduce highly proliferating Treg immunotherapy can be assessed by monitoring Treg cell
cells including those in the tumor tissues, and enhance numbers and function in the blood. Assessing the genetic
anti-tumor immune responses in humans and rodents makeup (e.g., HLA haplotypes) of the patients is also im-
[73-76]. Furthermore, an increasing number of studies portant to determine the susceptibility to autoimmunity
have shown the potentials and efficacy of low-dose met- due to Treg cell reduction.
ronomic cyclophosphamide in combination with other
immunotherapeutic agents [77, 78]. Conclusions
Another potential strategy to augment tumor immu-
nity via controlling Treg cells is to target TCR signaling Since the discovery of Treg cells as a key mediator of

www.cell-research.com | Cell Research | SPRINGER NATURE

Regulatory T cells in cancer immunotherapy
116

References

1 Danke NA, Koelle DM, Yee C, Beheray S, Kwok WW. Au-

toreactive T cells in healthy individuals. J Immunol 2004;
172:5967-5972.
2 Nishikawa H, Jäger E, Ritter G, Old LJ, Gnjatic S. CD4+
CD25+ regulatory T cells control the induction of antigen-spe-
cific CD4+ helper T cell responses in cancer patients. Blood
2005; 106:1008-1011.
3 Yu W, Jiang N, Ebert PJR, et al. Clonal deletion prunes but
does not eliminate self-specific αβ CD8+ T lymphocytes. Im-
munity 2015; 42:929-941.
4 Su LF, Kidd BA, Han A, Kotzin JJ, Davis MM. Virus-specific
CD4+ memory-phenotype T cells are abundant in unexposed
adults. Immunity 2013; 38:373-383.
Figure 3 Combination and sequence are the key for effective 5 Rosenberg SA, Yang JC, Restifo NP. Cancer immunotherapy:
cancer immunotherapy. See text. moving beyond current vaccines. Nat Med 2004; 10:909-915.
6 Sakaguchi S, Yamaguchi T, Nomura T, Ono M. Regulatory T
cells and immune tolerance. Cell 2008; 133:775-787.
7 Sakaguchi S. Naturally arising CD4+ regulatory T cells for
immunologic self-tolerance and negative control of immune
immunological self-tolerance, a common immunological
responses. Annu Rev Immunol 2004; 22:531-562.
basis for Treg-mediated suppression of autoimmunity 8 Shimizu J, Yamazaki S, Sakaguchi S. Induction of tumor
and tumor immunity has been extensively explored. immunity by removing CD25+CD4+ T cells: a common basis
The role of Treg cells in tumor immunity is now widely between tumor immunity and autoimmunity. J Immunol 1999;
accepted and therapies targeting Treg cells are under ac- 163:5211-5218.
tive investigation. Combination of Treg cell attenuation 9 Onizuka S, Tawara I, Shimizu J, et al. Tumor rejection by in
(e.g., by reducing Treg cell numbers or reducing their vivo administration of anti-CD25 (interleukin-2 receptor al-
suppressive activity in tumor tissues) with the activation pha) monoclonal antibody. Cancer Res 1999; 59:3128-3133.
10 Yamaguchi T, Sakaguchi S. Regulatory T cells in immune sur-
of tumor-specific effector T cells (e.g., by cancer vaccine veillance and treatment of cancer. Semin Cancer Biol 2006;
or immune checkpoint blockades) may mutually enhance 16:115-123.
each individual treatment. For the combination to be 11 Morikawa H, Sakaguchi S. Genetic and epigenetic basis of
more effective, the sequence of the treatments is criti- Treg cell development and function: from a FoxP3-centered
cally important (Figure 3). Tumor killing by anti-cancer view to an epigenome-defined view of natural Treg cells. Im-
drugs or ionizing irradiation, or effective tumor antigen munol Rev 2014; 259:192-205.
vaccination may release self-antigens and tumor-associ- 12 Wooldridge L, Ekeruche-Makinde J, Van Den Berg HA, et al.
A single autoimmune T cell receptor recognizes more than a
ated antigens and cause local inflammation; and this may
million different peptides. J Biol Chem 2012; 287:1168-1177.
recruit and activate Treg cells in tumor tissues, conse- 13 Birnbaum ME, Mendoza JL, Sethi DK, et al. Deconstructing
quently hampering ensuing anti-tumor immune respons- the peptide-MHC specificity of T cell recognition. Cell 2014;
es. To strongly activate the effector T cells, it is therefore 157:1073-1087.
necessary to deplete Treg cells or attenuate their sup- 14 Takahashi T, Kuniyasu Y, Toda M, et al. Immunologic
pressive activity prior to other immunological treatments self-tolerance maintained by CD25+CD4+ naturally anergic
such as vaccination and checkpoint blockade. We envis- and suppressive T cells: induction of autoimmune disease by
breaking their anergic/suppressive state. Int Immunol 1998;
age that the combination of various rationally sequenced
10:1969-1980.
immunological interventions will make current cancer 15 Ono M, Yaguchi H, Ohkura N, et al. Foxp3 controls regula-
immunotherapy more effective in the clinic. tory T-cell function by interacting with AML1/Runx1. Nature
2007; 446:685-689.
Acknowledgments 16 Wu Y, Borde M, Heissmeyer V, et al. FOXP3 controls regu-
latory T cell function through cooperation with NFAT. Cell
We would like to thank Dr James B Wing for critical reading 2006; 126:375-387.
of this manuscript. This study was supported by Grants-in-Aid for 17 Pandiyan P, Zheng L, Ishihara S, Reed J, Lenardo MJ.
Scientific Research from the Ministry of Education, Sports, and CD4+CD25+Foxp3+ regulatory T cells induce cytokine depri-
Culture (MEXT), and Japan Agency for Medical Research and vation-mediated apoptosis of effector CD4+ T cells. Nat Im-
Development (AMED). The authors have no conflicts of interest to munol 2007; 8:1353-1362.
declare. 18 Thornton AM, Shevach EM. CD4+CD25+ immunoregulatory
T cells suppress polyclonal T cell activation in vitro by inhib-

SPRINGER NATURE | Cell Research | Vol 27 No 1 | January 2017

 Atsushi Tanaka and Shimon Sakaguchi
117

iting interleukin 2 production. J Exp Med 1998; 188:287-296. regulatory T cell subsets. J Exp Med 2014; 211:121-136.
19 Boyman O, Sprent J. The role of interleukin-2 during homeo- 37 Sather BD, Treuting P, Perdue N, et al. Altering the distri-
stasis and activation of the immune system. Nat Rev Immunol bution of Foxp3+ regulatory T cells results in tissue-specific
2012; 12:180-190. inflammatory disease. J Exp Med 2007; 204:1335-1347.
20 Wing K, Onishi Y, Prieto-Martin P, et al. CTLA-4 con- 38 Maeda Y, Nishikawa H, Sugiyama D, et al. Detection of
trol over Foxp3+ regulatory T cell function. Science 2008; self-reactive CD8 + T cells with an anergic phenotype in
322:271-275. healthy individuals. Science 2014; 346:1536-1540.
21 Schubert D, Bode C, Kenefeck R, et al. Autosomal dominant 39 Miyara M, Yoshioka Y, Kitoh A, et al. Functional delineation
immune dysregulation syndrome in humans with CTLA4 mu- and differentiation dynamics of human CD4+ T cells express-
tations. Nat Med 2014; 20:1410-1416. ing the FoxP3 transcription factor. Immunity 2009; 30:899-
22 Kuehn HS, Ouyang W, Lo B, et al. Immune dysregulation 911.
in human subjects with heterozygous germline mutations in 40 Miyara M, Chader D, Sage E, et al. Sialyl Lewis x (CD15s)
CTLA4. Science 2014; 345:1623-1627. identifies highly differentiated and most suppressive FOXP3
23 Takahashi T, Tagami T, Yamazaki S, et al. Immunologic high regulatory T cells in humans. Proc Natl Acad Sci USA
self-tolerance maintained by CD25+CD4+ regulatory T cells 2015; 112:7225-7230.
constitutively expressing cytotoxic T lymphocyte-associated 41 Schaefer C, Kim GG, Albers A, et al. Characteristics of
antigen 4. J Exp Med 2000; 192:303-310. CD4+CD25+ regulatory T cells in the peripheral circulation
24 Walunas TL, Lenschow DJ, Bakker CY, et al. CTLA-4 can of patients with head and neck cancer. Br J Cancer 2005;
function as a negative regulator of T cell activation. Immunity 92:913-920.
1994; 1:405-413. 42 Liyanage UK, Moore TT, Joo H-G, et al. Prevalence of reg-
25 Qureshi OS, Zheng Y, Nakamura K, et al. Trans-endocytosis ulatory T cells is increased in peripheral blood and tumor
of CD80 and CD86: a molecular basis for the cell-extrinsic microenvironment of patients with pancreas or breast adeno-
function of CTLA-4. Science 2011; 332:600-603. carcinoma. J Immunol 2002; 169:2756-2761.
26 Yamaguchi T, Kishi A, Osaki M, et al. Construction of 43 Wolf AM, Wolf D, Steurer M, et al. Increase of regulatory T
self-recognizing regulatory T cells from conventional T cells cells in the peripheral blood of cancer patients. Clin Cancer
by controlling CTLA-4 and IL-2 expression. Proc Natl Acad Res 2003; 9:606-612.
Sci USA 2013; 110:E2116-E2125. 44 Ormandy LA, Hillemann T, Wedemeyer H, Manns MP, Gre-
27 Liu Z, Gerner MY, Van Panhuys N, Levine AG, Rudensky AY, ten TF, Korangy F. Increased populations of regulatory T cells
Germain RN. Immune homeostasis enforced by co-localized in peripheral blood of patients with hepatocellular carcinoma.
effector and regulatory T cells. Nature 2015; 528:225-230. Cancer Res 2005; 65:2457-2464.
28 Itoh M, Takahashi T, Sakaguchi N, et al. Thymus and autoim- 45 Ichihara F, Kono K, Takahashi A, et al. Increased populations
munity: production of CD25+CD4+ naturally anergic and sup- of regulatory T cells in peripheral blood and tumor-infiltrating
pressive T cells as a key function of the thymus in maintain- lymphocytes in patients with gastric and esophageal cancers.
ing immunologic self-tolerance. J Immunol 1999; 162:5317- Clin Cancer Res 2003; 9:4404-4408.
5326. 46 Sasada T, Kimura M, Yoshida Y, et al. CD4+CD25+ regulatory
29 Fisson S, Darrasse-Jeze G, Litvinova E, et al. Continuous ac- T cells in patients with gastrointestinal malignancies: possible
tivation of autoreactive CD4+CD25+ regulatory T cells in the involvement of regulatory T cells in disease progression. Can-
steady state. J Exp Med 2003; 198:737-746. cer 2003; 98:1089-1099.
30 Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell 47 Hiraoka N, Onozato K, Kosuge T, Hirohashi S. Prevalence of
development by the transcription factor Foxp3. Science 2003; FOXP3+ regulatory T cells increases during the progression
299:1057-1061. of pancreatic ductal adenocarcinoma and its premalignant le-
31 Levine AG, Arvey A, Jin W, et al. Continuous requirement sions. Clin Cancer Res 2006; 12:5423-5434.
for the TCR in regulatory T cell function. Nat Immunol 2014; 48 Curiel TJ, Coukos G, Zou L, et al. Specific recruitment of
15:1-10. regulatory T cells in ovarian carcinoma fosters immune priv-
32 Vahl JC, Drees C, Heger K, et al. Continuous T cell receptor ilege and predicts reduced survival. Nat Med 2004; 10:942-
signals maintain a functional regulatory T cell pool. Immunity 949.
2014; 41:722-736. 49 Sato E, Olson SH, Ahn J, et al. Intraepithelial CD8+ tumor-in-
33 Ohkura N, Hamaguchi M, Morikawa H, et al. T cell receptor filtrating lymphocytes and a high CD8+/regulatory T cell ratio
stimulation-induced epigenetic changes and Foxp3 expression are associated with favorable prognosis in ovarian cancer.
are independent and complementary events required for Treg Proc Natl Acad Sci USA 2005; 102:18538-18543.
cell development. Immunity 2012; 37:1-15. 50 Nishikawa H, Sakaguchi S. Regulatory T cells in tumor im-
34 Marson A, Kretschmer K, Frampton GM, et al. Foxp3 occu- munity. Int J Cancer 2010; 127:759-767.
pancy and regulation of key target genes during T-cell stimu- 51 Bates GJ, Fox SB, Han C, et al. Quantification of regulatory
lation. Nature 2007; 445:931-5. T cells enables the identification of high-risk breast cancer
35 Samstein RM, Arvey A, Josefowicz SZ, et al. Foxp3 exploits patients and those at risk of late relapse. J Clin Oncol 2006;
a pre-existent enhancer landscape for regulatory T cell lineage 24:5373-5380.
specification. Cell 2012; 151:153-166. 52 Shang B, Liu Y, Jiang S, Liu Y. Prognostic value of tumor-in-
36 Smigiel KS, Richards E, Srivastava S, et al. CCR7 provides filtrating FoxP3+ regulatory T cells in cancers: a systematic
localized access to IL-2 and defines homeostatically distinct review and meta-analysis. Sci Rep 2015; 5:15179.

www.cell-research.com | Cell Research | SPRINGER NATURE

Regulatory T cells in cancer immunotherapy
118

53 Álvaro T, Lejeune M, Salvadó MT, Banham AH, Roncador 69 Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S.
G, Montalba C. Outcome in Hodgkin’s lymphoma can be Stimulation of CD25+CD4+ regulatory T cells through GITR
predicted from the presence of accompanying cytotoxic and breaks immunological self-tolerance. Nat Immunol 2002;
regulatory T cells. Clin Cancer Res 2005; 11:1467-1473. 3:135-142.
54 Wilke CM, Wu K, Zhao E, Wang G, Zou W. Prognostic sig- 70 McHugh RS, Whitters MJ, Piccirillo CA, et al. CD4+CD25+
nificance of regulatory T cells in tumor. Int J Cancer 2010; immunoregulatory T cells: gene expression analysis reveals a
127:748-758. functional role for the glucocorticoid-induced TNF receptor.
55 Salama P, Phillips M, Grieu F, et al. Tumor-infiltrating Immunity 2002; 16:311-323.
FOXP3+ T regulatory cells show strong prognostic signifi- 71 Stephens GL, McHugh RS, Whitters MJ, et al. Engagement
cance in colorectal cancer. J Clin Oncol 2009; 27:186-192. of glucocorticoid-induced TNFR family-related receptor on
56 Frey DM, Droeser RA, Viehl CT, et al. High frequency of tu- effector T cells by its ligand mediates resistance to suppres-
mor-infiltrating FOXP3+ regulatory T cells predicts improved sion by CD4+CD25+ T cells. J Immunol 2004; 173:5008-5020.
survival in mismatch repair-proficient colorectal cancer pa- 72 Ko K, Yamazaki S, Nakamura K, et al. Treatment of advanced
tients. Int J Cancer 2010; 126:2635-2643. tumors with agonistic anti-GITR mAb and its effects on tu-
57 Sinicrope FA, Rego RL, Ansell SM, Knutson KL, Foster NR, mor-infiltrating Foxp3+CD25+CD4+ regulatory T cells. J Exp
Sargent DJ. Intraepithelial effector (CD3+)/regulatory (FoxP3+) Med 2005; 202:885-891.
T-cell ratio predicts a clinical outcome of human colon carci- 73 Ghiringhelli F, Larmonier N, Schmitt E, et al. CD4+CD25+
noma. Gastroenterology 2009; 137:1270-1279. regulatory T cells suppress tumor immunity but are sensitive
58 Saito T, Nishikawa H, Wada H, et al. Two FOXP3 +CD4 + to cyclophosphamide which allows immunotherapy of estab-
T-cell subpopulations distinctly control the prognosis of col- lished tumors to be curative. Eur J Immunol 2004; 34:336-
orectal cancers. Nat Med 2016; 22:679-684. 344.
59 Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival 74 Motoyoshi Y, Kaminoda K, Saitoh O, et al. Different mecha-
with ipilimumab in patients with metastatic melanoma. N nisms for anti-tumor effects of low- and high-dose cyclophos-
Engl J Med 2010; 363:711-723. phamide. Oncol Rep 2006; 16:141-146.
60 Robert C, Thomas L, Bondarenko I, et al. Ipilimumab plus 75 Ghiringhelli F, Menard C, Puig PE, et al. Metronomic cy-
dacarbazine for previously untreated metastatic melanoma. N clophosphamide regimen selectively depletes CD4 +CD25+
Engl J Med 2011; 364:2517-2526. regulatory T cells and restores T and NK effector functions in
61 Bulliard Y, Jolicoeur R, Windman M, et al. Activating Fcγ end stage cancer patients. Cancer Immunol Immunother 2007;
receptors contribute to the antitumor activities of immuno- 56:641-648.
regulatory receptor-targeting antibodies. J Exp Med 2013; 76 Ge Y, Domschke C, Stoiber N, et al. Metronomic cyclophos-
210:1685-1693. phamide treatment in metastasized breast cancer patients:
62 Simpson TR, Li F, Montalvo-Ortiz W, et al. Fc-dependent immunological effects and clinical outcome. Cancer Immunol
depletion of tumor-infiltrating regulatory T cells co-defines Immunother 2012; 61:353-362.
the efficacy of anti-CTLA-4 therapy against melanoma. J Exp 77 Hirschhorn-Cymerman D, Rizzuto GA, Merghoub T, et al.
Med 2013; 210:1695-1710. OX40 engagement and chemotherapy combination provides
63 Selby MJ, Engelhardt JJ, Quigley M, et al. Anti-CTLA-4 an- potent antitumor immunity with concomitant regulatory T cell
tibodies of IgG2a isotype enhance antitumor activity through apoptosis. J Exp Med 2009; 206:1103-1116.
reduction of intratumoral regulatory T cells. Cancer Immunol 78 Le DT, Jaffee EM. Regulatory T-cell modulation using cyclo-
Res 2013; 1:32-42. phosphamide in vaccine approaches: a current perspective.
64 Liakou CI, Kamat A, Tang DN, et al. CTLA-4 blockade in- Cancer Res 2012; 72:3439-3444.
creases IFNγ-producing CD4+ICOShi cells to shift the ratio 79 Ali K, Soond DR, Piñeiro R, et al. Inactivation of PI(3)K
of effector to regulatory T cells in cancer patients. Proc Natl p110δ breaks regulatory T-cell-mediated immune tolerance to
Acad Sci USA 2008; 105:14987-14992. cancer. Nature 2014; 509:407-411.
65 Hodi FS, Butler M, Oble DA, et al. Immunologic and clinical 80 Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and
effects of antibody blockade of cytotoxic T lymphocyte-as- immune correlates of anti-PD-1 antibody in cancer. N Engl J
sociated antigen 4 in previously vaccinated cancer patients. Med 2012; 366:2443-2454.
Proc Natl Acad Sci USA 2008; 105:3005-3010. 81 Michot JM, Bigenwald C, Champiat S, et al. Immune-related
66 Nishikawa H, Sakaguchi S. Regulatory T cells in cancer im- adverse events with immune checkpoint blockade: a compre-
munotherapy. Curr Opin Immunol 2014; 27:1-7. hensive review. Eur J Cancer 2016; 54:139-148.
67 Sugiyama D, Nishikawa H, Maeda Y, et al. Anti-CCR4 mAb 82 Phan GQ, Yang JC, Sherry RM, et al. Cancer regression and
selectively depletes effector-type FoxP3+CD4+ regulatory T autoimmunity induced by cytotoxic T lymphocyte-associated
cells, evoking antitumor immune responses in humans. Proc antigen 4 blockade in patients with metastatic melanoma.
Natl Acad Sci USA 2013; 110:17945-17950. Proc Natl Acad Sci USA 2003; 100:8372-8377.
68 Faget J, Biota C, Bachelot T, et al. Early detection of tu- 83 Attia P, Phan GQ, Maker AV, et al. Autoimmunity correlates
mor cells by innate immune cells leads to Treg recruitment with tumor regression in patients with metastatic melanoma
through CCL22 production by tumor cells. Cancer Res 2011; treated with anti-cytotoxic T-lymphocyte antigen-4. J Clin
71:6143-6152. Oncol 2005; 23:6043-6053.

SPRINGER NATURE | Cell Research | Vol 27 No 1 | January 2017