REVIEWS

Microbiota: a key orchestrator

of cancer therapy
Soumen Roy and Giorgio Trinchieri
Abstract | The microbiota is composed of commensal bacteria and other microorganisms that live
on the epithelial barriers of the host. The commensal microbiota is important for the health and
survival of the organism. Microbiota influences physiological functions from the maintenance of
barrier homeostasis locally to the regulation of metabolism, haematopoiesis, inflammation,
immunity and other functions systemically. The microbiota is also involved in the initiation,
progression and dissemination of cancer both at epithelial barriers and in sterile tissues. Recently,
it has become evident that microbiota, and particularly the gut microbiota, modulates the
response to cancer therapy and susceptibility to toxic side effects. In this Review, we discuss
the evidence for the ability of the microbiota to modulate chemotherapy, radiotherapy and
immunotherapy with a focus on the microbial species involved, their mechanism of action and the
possibility of targeting the microbiota to improve anticancer efficacy while preventing toxicity.

Germ-free animals
Animals raised in strict sterile
conditions that have no
microorganisms living in or on
them.
Commensalism
A symbiotic relationship
between two species in which
one species benefits without
causing harm to the other.
Pathobionts
Resident commensal
microorganisms that under
certain conditions may acquire
pathogenic potential.
Mutualism
A symbiotic relationship
between two species that is
beneficial for both species.
Cancer and Inflammation
Program, Center for Cancer
Research, National Cancer
Institute, National Institutes
of Health, Bethesda,
Maryland 20892, USA.
Correspondence to G.T.
trinchig@mail.nih.gov
doi:10.1038/nrc.2017.13
Published online 17 Mar 2017
The human microbiota is the ensemble of bacteria and
other microorganisms (archaea, fungi, protozoa, as well
as human, fungal, bacterial and protozoan viruses) that
inhabit the epithelial barrier surfaces of our body 1. The
microbiota affects physiological functions, particularly
metabolism, neurological and cognitive functions, hae-
matopoiesis, inflammation and immunity 2,3. Germ-free
animals are functionally immature in many physiolog-
ical systems, including innate immunity, and they are
highly susceptible to infection by pathogens4. However,
the germ-free condition in itself is not lethal. Germ-free
rodents maintained in a sheltered sterile environment
and fed with a controlled diet containing factors, such
as vitamins that are normally supplied by the intesti-
nal microbiota, can survive significantly longer than
conventionally raised animals4–6. The local microbiota
affects the functions and regulates the immunity of the
epithelial barrier on which it resides7–10. In addition,
the microbiota also exerts systemic effects10,11 (FIG. 1)
through mechanisms that are less well understood than
those mediating the local effects.
The gut microbiota comprises approximately 3 × 1013
bacterial cells that mostly exhibit commensalism with the
host 12. However, when the intestinal ecology is altered,
commensal bacteria that are referred to as pathobionts
(for example, Clostridium difficile or vancomycin-
resistant Enterococcus) may expand and acquire
pathogenic characteristics13. The gut microbiota also
exhibits mutualism with the host that modulates immu-
nity, promotes bone marrow haematopoiesis and also
regulates maturation and function of tissue-resident
haematopoietic cells of yolk sac origin such as the
microglia in the central nervous system14–16. The gut
microbiota interacting with epithelial and stromal intes-
tinal cells regulates barrier functions, mucosal immune
homeostasis7–9, host–microbiota symbiosis, prevention
of infestation by pathogens, control of overgrowth by
pathobionts, metabolism of indigestible dietary fibre,
synthesis of vitamins and regulation of metabolism,
including the prevention of obesity 7–9,17–19. An intact
and functional gut epithelium maintains a healthy
body, and gut epithelial homeostasis is maintained
by continuous crosstalk between the gut microbiota,
immune cells and the mucosal barrier 20.
The composition of the microbiota is shaped by host
genetics, colonization at the time of birth, type of birth
delivery, an individual’s lifestyle, incidence of diseases
and exposure to antibiotics21–23. Microbiota composition
evolves during the first few years of human life before
maturation into an adult-like microbiota22. After that,
the composition of the microbiota in the gut and other
epithelial barriers remains relatively constant through-
out adult life, although it could still be affected by diet,
changes in lifestyle, disease and disease treatment 24,25.
The microbiota present at the epithelial barrier and
particularly in the gut influences local and systemic
metabolic functions, inflammation and adaptive immu-
nity, which modulate cancer initiation, progression
and response to anticancer treatment. Host genetics and
environmental factors, including nutrition, modify the
composition of the microbiota both in humans and
in the mouse26–29. In the mouse, the pro-carcinogenic

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 1

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Local effects of Systemic effects of lifestyle in part may indirectly affect carcinogenesis and
gastrointestinal microbiota gastrointestinal microbiota
response to cancer therapy through modification of
microbiota composition.
• Translocation of bacteria or Owing to its ability to modulate host metabolism,
bacterial products and toxins
• Metabolites (e.g. small and inflammation and immunity, the microbiota is involved
medium chain fatty acids, in the initiation and/or progression of various types of
choline derivatives, secondary
bile acids, vitamins, hormones cancer both at the epithelial barriers and in sterile tis-
and nutrients) sues3 (BOX 1). Although much progress has been made
• Innate and adaptive immune in cancer therapy — and in particular recent advances
cell migration
• Cytokines in immunotherapy are very promising — a significant
• Endocrine (cortisol) and neural proportion of cancer patients still do not respond to
(vagus and enteric nervous therapy. In this Review we discuss the body of data in
system) pathways
experimental models and increasingly in clinical stud-
ies that suggests that the composition of the microbiota
regulates the efficacy of anticancer therapy and that tar-
• Nutrient absorption • Metabolism geting the microbiota may improve drug efficacy and/or
• Synthesis of vitamins • Neurological, behavioural and cognitive reduce adverse effects.
Physiological

• Metabolism of bile and functions

functions

hormones • Cardiovascular and musculoskeletal functions

• Fermentation of carbohydrates • Haematopoiesis and myeloid cell functions Chemotherapy and microbiota
• Morphogenesis • Circadian rhythm After the discovery of the cytotoxic effects of nitro-
• Barrier strengthening • Ageing gen mustards during the Second World War, cytotoxic
• Mucosal immunity • Inflammation and immunity
chemotherapeutic agents were developed that still today
Non-neoplastic

• Inflammatory bowel disease • Obesity, insulin resistance and type II diabetes remain a major staple of cancer therapy 34. Cytotoxic
• Gastritis and gastric ulcers • Cardiovascular diseases
pathology

drugs are classified according to their mechanisms of

• Autoimmunity: CNS, eye and joints
• Non-alcoholic steatohepatitis action, as alkylating agents, heavy metals including plati­
• Gout num, antimetabolites, cytotoxic antibiotics and spindle
• Stomach cancer (Helicobacter • MALT, ocular and skin lymphoma poisons. Most forms of chemotherapy display antitu-
pylori) • Thymic lymphoma mour activity by targeting DNA integrity and cell divi-
• Colorectal carcinoma (Escherichia • Hepatocellular carcinoma
Cancer

coli, Fusobacterium spp. • Mammary carcinoma

and enterotoxigenic Bacteroides
fragilis)
• Gallbladder carcinoma
(Salmonella enterica Typhi)
• Cancer therapy gastrointestinal
therapy
sion in the dividing cancer cells. Cytotoxicity may also
• Pancreatic cancer be induced by the effect of drugs on other cellular com-
• Prostate cancer partments such as mitochondria and cell membranes35.
• Sarcoma
• Ovarian cancer Chemotherapy is not specific and its use is always asso-
ciated with significant toxicity for tissues that have a high
• Cancer chemotherapy, immunotherapy

rate of cell replacement and division36.

Cancer

toxicity and radiotherapy efficacy

• Cancer therapy toxicity
• Graft-versus-host disease Drug metabolism. The gut microbiota affects drug
Nature Reviews
Figure 1 | Local and systemic effects of the gastrointestinal microbiota. The | Cancer pharmacokinetics, anticancer activity and toxicity at
abundant microbiota present on the gastrointestinal mucosa affects local mucosal various levels3,37. A schematic representation of the role
homeostasis, functions and immunity7–9. Many of the mechanisms by which various of the microbiota on the metabolism of drugs following
bacterial species and their products and metabolites affect mucosal physiology and enteral (for example, oral) or parenteral (for example,
pathology have been described7–10. However, the presence and composition of the gut intravenous) administration is presented in FIG. 2. The
microbiota also systemically affects the functions of most physiological systems, the rate of absorption and bioavailability of many oral drugs
pathology and the response to therapy in distant organs10. Some of the demonstrated or depends on their exposure in the gut to both host and
proposed mechanisms by which the gastrointestinal microbiota can achieve this are
bacterial enzymes before entering the circulation38,39.
listed in the box at the top of the figure205. The microbiota and specific microbial species
Xenobiotics induce changes in the composition and
also affect neoplastic pathology both at the local level in the gastrointestinal
tract145,188,206,207 and systemically in organs that are not normally associated with the gut shape the physiology and gene expression of the gut
microbiota192–194,208–212. Although these mechanisms have been studied primarily for the microbiota40,41, further modulating its effect on drug
most abundant intestinal microbiota organisms, microorganisms colonizing other metabolism. Biotransformation of drugs mediated by the
epithelial barriers, for example, the mouth and the skin, are also expected to mediate gut microbiota includes many chemical reactions, with
both local and systemic effects195,196. In addition to its association with cancer reduction and hydrolysis affecting the largest number
development, the microbiota also has both a local and a systemic role in modulating the of drugs42. Other reactions include functional group
efficacy and toxicity of cancer therapy3. CNS, central nervous system; MALT, removal, N‑oxide cleavage, proteolysis, denitration,
mucosa-associated lymphoid tissue. deconjugation, amine formation and/or hydrolysis,
thiazole ring opening, acetylation and isoxazole scis-
phenotype expressed by some genetically mutated mice sion43. Commensal microorganisms also decrease the
has been shown to be transferred to wild-type mice by absorption of certain drugs by physical binding and
microbiota transfer 30–32, and the transfer of the faecal segregation44. More than 40 drugs have been shown
Xenobiotics microbiota of patients who are responsive to cancer ther- to be metabolized by the gut microbiota43. However,
Foreign chemical substances,
including drugs, that are not
apy into germ-free mice has been shown to endow those among anticancer drugs, only the nitroreduction of the
naturally produced by the animals with an ability to respond efficiently to the ther- radiation sensitizer misonidazole, the hydrolysis of
organism. apy 33. Thus, it can be speculated that host genetics and the antimetabolite methotrexate and the deconjugation

2 | ADVANCE ONLINE PUBLICATION www.nature.com/nrc

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Biotransformation of the liver-detoxified form of the topoisomerase I
The chemical alteration of a inhibitor irinotecan (also known as CPT‑11) (discussed
xenobiotic, such as a drug, in more detail below) have clearly been shown to be
within the body. affected by the gut microbiota43.
Probiotic
In addition to the direct effect of microorgan-
Live microorganisms that are isms and microbial enzymes on drug absorption and
consumed by humans and metabo­lism39–41, the gut microbiota indirectly affects
animals as food supplements the metabolism of both orally and systemically delivered
for their potential
drugs through modulation of gene expression and the
health-promoting qualities.
physiology of the local mucosal barrier and of distant
organs such as the liver 45–47. In the livers of germ-free
mice compared with specific pathogen-free mice, some
members of the large cytochrome P450 (Cyp450) gene
family are increased (for example, the Cyp2a, Cyp2b and
Cyp3a families involved in xenobiotic steroid metabo-
lism) whereas others are decreased (for example, the
Cyp4a family involved in fatty acid or arachidonic acid
metabolism)45–49. Germ-free mice also consistently
Box 1 | Microbiota effects on cancer initiation and progression
The health of the organism requires maintenance of epithelial barriers on which the
microbiota resides. The barriers maintain a peaceful relationship with commensal
microorganisms while protecting the host from pathogens and pathobionts.
Dysbiosis, which is an alteration in the composition of the microbiota associated with
pathology, disrupts the physiological interaction between epithelial cells and the
microbiota, results in breach of the barriers, induces inflammatory pathologies and
may contribute to cancer initiation and progression182–184.
The cytokine interleukin‑18 (IL‑18) mediates IL‑22‑dependent intestinal mucosal
protection. Mice unable to produce, process or respond to IL‑18 (for example, mice
genetically deficient in the genes encoding IL‑18, IL‑18 receptor (IL‑18R), myeloid
differentiation primary response 88 (MYD88), inflammasomes or inflammasome
signalling molecules) display a dysbiosis that results in enhanced susceptibility to
chemically induced colitis and colon carcinogenesis32,170,185,186. This susceptibility can
be transferred to wild-type mice by cohousing or faecal transfer32,170. Deficiency for
other immunologically relevant genes may also enhance the susceptibility to colon
carcinogenesis owing to a dysbiosis that can be transferred to wild-type mice187. Some
of the mechanisms by which microbial species either in humans or in experimental
animals induce colorectal carcinogenesis have been characterized187–189. For example,
Fusobacterium nucleatum, associated with human colorectal cancer, expands
tumour-promoting myeloid cell populations188,189, activates host β‑catenin–WNT
signalling by the binding of its FadA adhesin to E‑cadherin188,189 and modulates
anticancer immune response by triggering the inhibitory receptor T cell
immunoreceptor with immunoglobulin (Ig) and ITIM domains (TIGIT) on T and natural
killer cells via the microbial Fap2 protein190. However, among the various bacteria
described to be associated with human cancer, only Helicobacter pylori has been
proved to be a human carcinogen causing gastric cancer, as its elimination from the
host prevents disease187,191.
The gut microbiota also affects cancer at distant sites. Colonic infection with
Helicobacter hepaticus not only enhances colon carcinogenesis but also favours the
generation of mammary and prostate carcinoma in the ApcMin/+ mouse model of
colorectal cancer192 and augments chemical and viral transgenic liver
carcinogenesis184,193. In a mouse model of ataxia telangiectasia, the gut microbiota
composition observed in different mouse colonies or upon experimental
perturbation determines thymic lymphoma incidence and survival194. In humans, a
history of periodontal disease and antibodies against oral bacteria have been found
to be associated with increased risk of pancreatic cancer195. Specifically, the
presence in the oral microbiota of Porphyromonas gingivalis and Aggregatibacter
actinomycetemcomitans was observed to be significantly associated with increased
risk of pancreatic cancer, and the presence of the phylum Fusobacterium was
associated with decreased risk196. The effects of the microbiota on carcinogenesis in
distant organs are mediated in part by the modulation of the systemic inflammatory
tone, oxidative stress and inflammation-induced genotoxicity towards leukocytes
and epithelial cells192–194,197.
overexpress the genes encoding xenobiotic-sensing
receptors and transcription factors such as constitutive
androstane receptor (Car; also known as Nr1i3), CAR-
regulated gene P450 oxidoreductase (Por), aryl hydro-
carbon receptor (Ahr), peroxisome proliferator-activated
receptor‑α (Ppara) and nuclear factor erythroid 2‑related
factor 2 (Nrf2; also known as Nfe2l2)45–49. This altered
gene expression results in the faster metabolism of many
xenobiotics in germ-free mice, supporting the role of the
microbiota in regulating drug metabolism and detoxi-
fication45. Interestingly, although colonization of germ-
free mice with microbiota taken from conventionally
raised mice re-establishes normal gene expression, colo-
nization of germ-free mice with the probiotic preparation
VSL#3 normalizes only a proportion of the genes altered
in germ-free mice48. Thus, the heterogeneity of response
to drug therapy and susceptibility to toxic effects in can-
cer patients may be due at least in part to differences
in gut microbiota composition and activity between
individual patients50.
In addition to oral drugs, injected drugs are also
in part firstly metabolized in the liver, before biliary
excretion into the gut, where they are exposed to the
gut microbiota and can be further metabolized and
reabsorbed (FIG. 2). Irinotecan is an intravenous chemo­
therapeutic drug used for colorectal cancer treat-
ment 51. Irinotecan is transformed into its active form,
SN‑38, by liver and small intestine tissue carboxyl­
esterase and then detoxified in the liver by host UDP-
glucuronosyltransferases into inactive SN‑38‑G before
being secreted into the gut 51. In the gut, SN‑38‑G can
be reconverted by bacterial β‑glucuronidases into active
SN‑38, which induces significant intestinal toxicity and
diarrhoea52. In tumour-bearing rats, irinotecan chemo-
therapy increased the abundance of Clostridium cluster XI
and Enterobacteriaceae in the proximal colon, indicat-
ing induction of an altered microbiota composition that
could directly affect inflammation or change the propor-
tion of bacterial species expressing β‑glucuronidase53.
In the human gut, β‑glucuronidase activity was found
mostly in the Firmicutes phylum, particularly within
Clostridium clusters XIVa and IV54,55. Antibiotic treat-
ments that decrease the number of β‑glucuronidase-
positive bacterial species or the use of an inhibitor
specific for bacterial β‑glucuronidase effectively treat
intestinal inflammation induced by irinotecan therapy
in experimental animals56,57. Use of probiotics in clini-
cal trials has shown a modest decrease in irinotecan‑
induced diarrhoea58. As the intestinal toxicity of irino-
tecan in humans is severe and dose limiting, there is
much interest in the development of selective bacterial
β‑glucuronidase inhibitors to be used in clinical trials
to reduce toxicity 59.
Response to chemotherapy. Direct interaction with
bacteria can affect the efficacy of chemotherapeutic
drugs60. Among 30 drugs tested in vitro in the presence
of non-pathogenic Gram-negative Escherichia coli or
Gram-positive Listeria welshimeri, the activity of 10
of them was found to be inhibited by one or both spe-
cies whereas the efficacy of 6 others was enhanced60.

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 3

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

a Enteral drug metabolism b Parenteral drug metabolism

E5 P2
E1
Approximately 10% of the drug Shortly after administration
Enteral drug enters the circulation via the >90% of the drug is
administration hepatic veins, and reaches available in the circulation
E2 for the target tumour
target tumours and other tissues
100% of the drug reaches the
GI tract and is absorbed into P3
the liver via the portal circulation
Systemic toxicity
Heart
First pass metabolism Lung
P4
E3 in the liver Hepatic vein
Phase I Each minute, 29% of
Altered gene Liver the circulating blood
expression Gall bladder P1 and drugs reach the
(CYP450 enzymes) Portal vein liver via the splanchnic
Parenteral drug circulation
Activation Bile duct administration
Stomach
Detoxification
Small intestine
Phase II Large intestine
Deconjugation
E4
~90% of the drug metabolites Gut microbiota Gut microbiota
from the liver are excreted • Biotransformation of xenobiotics Reactivation of liver detoxified xenobiotics
with bile into the intestinal affecting absorption and bioavailability
tract where they are excreted • Reactivation of liver-detoxified xenobiotics
in stools or reabsorbed
through the portal circulation Drug elimination by kidney Drug elimination by kidney
excretion and via the biliary tract excretion and via the biliary tract

Figure 2 | Major pathways of drug metabolism and the role of microbiota following enteral Nature
(forReviews
example,| Cancer
oral)
or parenteral (for example, intravenous) administration. a | Enteral drug metabolism. Orally administered drugs
(E1) sit in the stomach for 30–45 minutes before reaching the intestine and being absorbed into the liver by the
portal circulation (E2). In the intestine, host and microbial enzymes induce metabolic alterations to the drug that
together with direct binding to bacterial products and segregation control intestinal absorption43. In the liver,
following phase I and phase II processing (first pass metabolism; E3), approximately 90% of the oral drug is
metabolized and destroyed or eliminated through biliary secretion (E4). The drugs secreted into the intestine via the
biliary duct can be reabsorbed via the portal circulation or excreted in stools. As a consequence, only 10% of the oral
drug enters the circulation through the hepatic veins and is available to reach the target tumours and other tissues
(E5). Phase I and phase II processing are also affected by the gut microbiota through the regulation of the level of
host enzymes involved in drug processing. b | Parenteral drug metabolism. Following intravenous administration (P1)
close to 100% of the drug enters the circulation and is available to reach the target tumours (P2); however, the drug
is also distributed systemically, inducing adverse toxic reactions (P3). Any remaining drug not retained in tissues can
be rapidly excreted by the kidney. Each minute 29% of the circulating drug is transported via the splanchnic
circulation (hepatic, mesenteric and splenic arteries) to the liver (P4), where the drug is processed similarly to
enterally administered drugs. The detoxified drugs that are secreted from the liver to the intestine through the
biliary excretion route can be reactivated by bacterial enzymes, inducing intestinal toxicity. CYP450, cytochrome
P450; GI, gastrointestinal.

High-performance liquid chromatography (HPLC) and Platinum-based antineoplastic drugs such as

mass spectrometry analysis showed that contact with
the bacteria resulted in biotransformation of the drugs.
The ability of bacteria to inhibit the antitumour effect of
gemcitabine and to increase that of the prodrug CB1954
was confirmed in vivo in mice by intratumour inocu-
lation of E. coli 60. In vivo, doxorubicin maintains its
efficacy against subcutaneous tumours in microbiota-
depleted mice; however, in one study, overgrowth
of Parabacteroides distasonis in mice treated with
broad-spectrum antibiotics was observed to abrogate
its antitumour effect 47. Thus, the results in experimen-
tal animals suggest a complex and variable interaction
between several chemotherapeutic agents and micro-
biota. However, detailed mechanistic studies in vivo
have been reported only for platinum compounds and
cyclophosphamide (CTX)61,62.
oxaliplatin and cisplatin kill tumour cells by inhibiting
DNA replication and by targeting cellular membranes
and mitochondria63. They form intra-strand platinum–
DNA adducts that lead to DNA double-strand breaks
(DSBs). In addition to killing tumour cells, plati-
num drugs cause severe intestinal toxicity, nephro-
toxicity, disruption of blood–brain barrier integrity,
oto­toxicity leading to hearing loss and peripheral neu-
ropathy 64–69. Like other chemotherapeutic agents, the
genotoxic platinum compounds are toxic for the rap-
idly regenerating intestinal mucosal cells, leading to
disruption of barrier functions70,71. Barrier breaching
enables commensal microorganisms and pathogens
to access the mesenteric lymph nodes and also the
blood circulation, leading to septicaemia and systemic
inflammation70.

4 | ADVANCE ONLINE PUBLICATION www.nature.com/nrc

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Pathogenic T helper 17 cells
(pTH17 cells). A CD4+ T cell
subset that simultaneously
expresses markers of TH1 cells
(T‑bet transcription factor,
interferon–γ (IFN–γ) and CXC
chemokine receptor 3 (CXCR3))
and of TH17 cells (RORγT
transcription factor,
interleukin‑17 (IL‑17) and C-C
chemokine receptor 6 (CCR6)).
The antitumour effect of oxaliplatin or cisplatin
treatment on subcutaneous transplantable tumours
is dramatically decreased in germ-free mice or in
mice in which gut commensals have been depleted by
broad-spectrum antibiotics62 (FIG. 3). In microbiota-
depleted mice, oxaliplatin induces a lower level of
inflammatory gene expression in tumours than
observed in conventionally raised mice62. The pres-
ence of gut microbiota is not necessary for the drug
to penetrate the tumour and to form platinum–DNA
adducts. However, oxaliplatin-induced DNA damage
in the tumour cells is decreased in microbiota-depleted
mice62. Known components of the cytotoxic effects of
platinum drugs that mediate the DNA and other cellu-
lar damage are reactive oxygen species (ROS)72,73. The
source of the ROS, required for induction of DNA
damage and apoptosis in tumour and tissue cells was
expected to be the cancer cells themselves upon inter-
action with the drug 73. However, the absence of gut
microbiota in mice was shown to prevent the paracrine
production of ROS by tumour-infiltrating myeloid cells
via NADPH oxidase 2 (NOX2)62. Consistently, oxalip-
latin is not effective in mice that are genetically defi-
cient for Cybb (which encodes the gp91phox chain of
NOX2) or in mice in which myeloid cells are depleted
by antibody treatment 62. Similarly, re‑association
of antibiotic-treated mice with the probiotic bac-
terial species Lactobacillus acidophilus restores the
cisplatin antitumour effect and recovers some of
the cisplatin-induced inflammatory gene expression
that is observed in conventionally raised mice74. The
mechanism by which the intact gut microbiota or
L. acidophilus prime intratumour myeloid cells for
ROS production in response to platinum drugs is
dependent on signalling through myeloid differentia-
tion primary response 88 (MYD88)-associated innate
immune receptors (also known as pattern recognition
receptors (PRRs)), although the precise receptors
involved remain to be determined62 (FIG. 4).
Oxaliplatin induces immunogenic cancer cell death,
a cell death modality that is preceded by autophagy and
that exposes cell surface-associated and soluble immuno­
stimulatory signals, thus stimulating an antitumour
immune response through the activation of antigen-
presenting cells 75. The adaptive T  cell-mediated
cytotoxic response against oxaliplatin-treated tumours
results in long-lasting tumour regression and cure in
most mouse tumour models62,75,76. The activation of
T cells following immunogenic cell death is dependent
on an inflammatory response involving the induction of
interleukin‑1β (IL‑1β). This is mediated by the dual acti-
vation of the innate immune receptor Toll-like recep-
tor 4 (TLR4) by nuclear high mobility group protein
B1 (HMGB1), a damage-associated molecular pattern
(DAMP) released by dying cells, and of the NOD-like
receptor, pyrin domain-containing 3 (NLRP3) inflam-
masome77. DAMPs react with germline-encoded PRRs
and activate inflammation, innate resistance and adap-
tive immunity 78 (FIG. 4). The ability of another chemo-
therapeutic agent, anthracycline, to elicit antitumour
T cell immunity requires the activation of another
innate immune receptor, formyl peptide receptor 1
(FPR1) on dendritic cells, not by microbial products
but through interaction with the endogenous ligand
annexin A1 from dying cancer cells, which promotes
stable contacts between the dying cancer cells and the
dendritic cells79 (FIG. 4). Unlike oxaliplatin, cisplatin
does not induce immunogenic cell death and in mouse
tumour models mediates an initial effective and rapid
antitumour effect with almost complete destruction
of the tumours; however, this effect is short lived and
tumours grow back within about a week, killing most
of the animals62. In microbiota-depleted mice, the early
genotoxic and cytotoxic effects of both platinum com-
pounds as well as the oxaliplatin-induced long-term
survival are decreased62. These findings suggest that
the microbiota composition modulates both the early
cytotoxic effect and the adaptive immune response,
possibly by independent mechanisms.
Similarly to oxaliplatin, the alkylating agent CTX
also induces immunological cell death80. The anti­
tumour activity of CTX in tumour-bearing mice
induces alteration of the gut mucosa within 1–2 days,
with increased mucosal permeability and transloca-
tion of gut bacteria into the mesenteric lymph nodes61.
Furthermore, CTX treatment alters the composition of
commensals in the small intestine of tumour-bearing
mice within a week, similar to what has been observed
in cancer patients61,81. More specifically, this chemo-
therapeutic agent selectively reduces the abundance of
several bacterial phyla in the small intestinal mucosa
while it enhances that of Gram-positive bacteria, such
as Enterococcus hirae, which translocate into mesenteric
lymph nodes61,82. Following CTX-induced immuno­
genic tumour cell death, these translocated bacteria
increase the intratumoural CD8+ T cell/T regulatory
(Treg) cell ratio and activate pathogenic T helper 17 cells
(pTH17 cells) and memory TH1 cell immune responses,
resulting in the induction of an antitumour adaptive
immune response that can be detected in the spleen of
the treated animals61,82,83. The pTH17 cell responses and
the antitumour effect of CTX treatment are reduced in
germ-free mice and in mice depleted of Gram-positive
bacteria by treatment with antibiotics61. Moreover, the
antitumour effect of CTX can be restored in microbiota-
depleted mice by adoptive transfer of pTH17 cells61. In
addition, another bacterial species, the Gram-negative
Barnesiella intestinihominis accumulates in the colon
and promotes the infiltration of interferon‑γ (IFNγ)-
producing γδ T cells in cancer lesions of CTX-treated
mice82. The immune sensor nucleotide-binding oli-
gomerization domain-containing 2 (NOD2) limits the
ability of these microorganisms to translocate and to
modulate CTX anticancer activity. It is noteworthy that
the presence of E. hirae- and B. intestinihominis-specific
TH1 cells correlated with a more favourable prognosis in
patients with advanced lung or ovarian cancer treated
with chemo-immunotherapy 82.
Drug-induced toxicity. There is little information cur-
rently available regarding the ability of probiotics to pre-
vent platinum-compound toxicity. Oral treatment with a

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 5

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

L. acidophilus Gram-negative L. johnsonii A. shahii Burkholderiales B. breve

bacteria E. hirae B. thetaiotaomicron B. longum
B. intestinihominis B. fragilis B. adolescentis
Commensals

Intestinal
epithelium

MYD88

Mechanism
of anticancer CTLA4
therapy

ROS
PD1 or
PDL1
DNA damage TNF

T cell-mediated
• Pt-DNA adduct Promotion of Licensing of
Tumours T cell-mediated killing following Haemorrhagic
• Genotoxicity antitumour T cell-mediated
tumour killing immunogenic necrosis
and cell death immunity killing
cell death

Cisplatin or TBI and adoptive CpG-ODN and

Treatments Untreated Cyclophosphamide Anti-CTLA4 Anti-PDL1
oxaliplatin T cell transfer anti-IL-10R

Apoptotic cell Bacteria Blood vessels CTL Dendritic Fibroblast

cell

Macrophage Monocyte Neutrophil pTH17 or TH1 cell Tumour cells TLR4 TLR9

Nature Reviews | Cancer

probiotic combination of lyophilized live L. acidophilus
and Bifidobacterium bifidum was shown to prevent intes-
tinal toxicity in cancer patients treated with both radio­
therapy and cisplatin84. Combined with experimental
data that L. acidophilus supports the anticancer effect of
cisplatin, these clinical data suggest the possibility that
probiotic bacterial species such as L. acidophilus may
favour the antitumour effect while preventing some of
the toxic side effects of the drug 74,84. Inhibition of NOX
proteins by acetovanillone protects mice from cisplatin
nephrotoxicity by blocking early ROS production in
drug-damaged proximal tubular cells and subsequent
ROS production by tissue myeloid cells85. Thus, the gut
microbiota may regulate both the antitumour effect
and the toxicity of platinum compounds by similarly
modulating tissue and tumour ROS production62,85.
The intestinal chemotoxicity of methotrexate is medi-
ated in part by activation of TLR4 by either microbial
products or endogenous DAMPs86,87 (FIG. 4). Activation
of TLR2 protects the mucosa against methotrexate-
induced damage by increasing the expression of the ABC
transporter multidrug resistance protein 1 (MDR1; also
known as P‑glycoprotein and ABCB1), which regulates
the efflux of xenobiotics from intestinal epithelial cells86–88.

6 | ADVANCE ONLINE PUBLICATION www.nature.com/nrc

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

◀ Figure 3 | The gut microbiota regulates anticancer therapies. In physiological
conditions, the commensals in the intestinal lumen are prevented from translocating
through the intestinal mucosa by an intact epithelial barrier covered by a mucus layer
that is poorly permeable to microorganisms. Treatment with platinum (Pt) drugs, total
body irradiation (TBI), cyclophosphamide (CTX) and anti-cytotoxic T lymphocyte-
associated antigen 4 (CTLA4) all cause damage to the mucus layer, which disrupts
barrier integrity and enables bacteria to penetrate the lamina propria, which lies
beneath the epithelium. Translocated bacteria activate innate immune cells and
initiate local and systemic inflammation. Mechanisms of gut-associated toxicity and
tumour clearance vary based on treatment type, and the microbial species that have
been demonstrated to affect these mechanisms are listed at the top of the figure.
Gut commensals, through myeloid differentiation primary response 88 (MYD88)-
associated receptors, prime myeloid cells for the generation of reactive oxygen
species (ROS), which, in the presence of Pt–DNA adducts formed in response to
cisplatin or oxaliplatin, cause DNA damage62,74. TBI used to condition patients before
they receive adoptive T cell therapy induces mucosal damage and translocation of
commensals, which through Toll-like receptor 4 (TLR4) signalling activate dendritic
cells to sustain proliferation and cytotoxic functions of the transferred T cells153.
CTX induces immunological cell death of tumour cells, which elicits the generation
of antitumour pathogenic T helper 17 (pTH17) cells, TH1 cells and cytotoxic
T lymphocytes (CTLs); CTX also induces damage to the mucosa and translocation
of commensal bacteria that activate tumour antigen-presenting dendritic cells,
enhancing the antitumour immune response61,82. During CpG-oligodeoxynucleotide
(ODN)–anti-interleukin‑10 receptor (IL‑10R) therapy, the gut microbiota, through
TLR4 signalling, primes tumour-infiltrating myeloid cells to respond to the TLR9
agonist CpG-ODN, producing tumour necrosis factor (TNF) and other inflammatory
cytokines that induce haemorrhagic necrosis of the tumour and an antitumour
immune response62. Anti‑CTLA4 immunotherapy promotes both antitumour and
anti-commensal immunity; the anti-commensal immunity against specific genera,
such as Burkholderiales and Bacteroidales (Bacteroides thetaiotaomicron and
Bacteroides fragilis), results in mucosal damage and bacterial translocation but
also serves as an adjuvant for the antitumour response33. The efficacy of anti-
programmed cell death protein 1 ligand 1 (PDL1) therapy in generating antitumour
immunity by preventing programmed cell death protein 1 (PD1) interaction with
PDL1 is enhanced by the presence in the gut microbiota of Bifidobacterium spp.
(Bifidobacterium breve, Bifidobacterium longum and Bifidobacteri‑um adolescentis)152.
A. shahii, Alistipes shahii; B. intestinihominis, Barnesiella intestinihominis; E. hirae,
Enterococcus hirae; L. acidophilus, Lactobacillus acidophilus; L. johnsonii, Lactobacillus
johnsonii.
Doxorubicin induces severe side effects such as cardio­
myopathy and intestinal mucositis associated with
major changes in both the oral and the intestinal
microbiota89–91. Although these results suggest that the
alterations in microbiota composition contribute to tox-
icity, the innate immune receptors responsible for the
microbiota-induced toxicity have not been fully char-
acterized89–91. However, NOD2 stimulation by bacterial
Cachexia
A wasting syndrome with muramyl dipeptide has a protective effect and prevents
muscle atrophy and loss of mucosal damage by doxorubicin92 (FIG. 4).
weight and adipose tissue, The gut microbiota regulates pancreatic beta-cell mass,
often associated with cancer adipose tissue inflammation and uptake of lipids. It ulti-
and cancer therapy.
mately promotes diet-induced obesity through the bile
Bystander effect acid nuclear receptor farnesoid X receptor (FXR) path-
In radiobiology, collateral way 93,94. Fat metabolism and adipose tissue are affected
damage exhibited by in many patients with tumours as well as in platinum
unirradiated cells in response
drug-treated individuals, resulting in cachexia95–98. Not
to signals received from nearby
irradiated cells. only can cancer therapy aggravate the devastating effects
of cancer-associated cachexia99,100, but also many chemo-
Abscopal effect therapeutic agents, including cisplatin, can directly induce
In radiotherapy, a multi-organ failure and muscle wasting that resembles the
phenomenon whereby local
radiotherapy induces tumour
properties of cancer cachexia100–102. In addition, the over-
regression at sites distant from all toxicity of cancer treatment is augmented in patients
the irradiated site. with cachexia in part as a consequence of altered drug
pharmacokinetics in these patients103,104. Although the
mechanism underlying cachexia in these conditions
remains poorly understood, the close relationship between
the gut microbiota and energy metabolism raises the pos-
sibility that the composition of the microbiota could be
involved in the pathogenesis of this condition and there-
fore, could be targeted for therapy98,105,106. Use of probiotics
alone or together with a nutritional supplement-enriched
diet was shown to improve body weight in patients and
mice with cancer-associated cachexia107,108. Recent stud-
ies in mice have shown that the colonization of the gut
by the E. coli strain O21:H+ prevents muscle wasting
induced by infection or intestinal damage by engaging
the NLR family CARD domain-containing protein 4
(NLRC4) inflammasome and the insulin-like growth fac-
tor 1 (IGF1) signalling pathway109. Additional mechanistic
studies and rigorous clinical trials are needed to estab-
lish whether modulation of gut microbiota homeo­stasis
could become an effective clinical approach to treat cancer
co‑morbidities such as anorexia and cachexia as well as
cancer therapy adverse events.
Radiotherapy and microbiota
Many cancer patients receive ionizing radiation therapy
(RTX) that is genotoxic for tumour cells and may be cura-
tive for localized cancers. The classic paradigm in radi-
ation biology assumed that the cellular nucleus was the
only target of radiation and DNA damage was induced by
direct deposition of energy or production of ROS through
radiation-induced dissociation of intracellular water mol-
ecules. However, ionizing radiation also induces non-
targeted effects on non-irradiated cells such as the
bystander effect on nearby cells; systemic radio-adaptive
responses including inflammatory and immune reactiv-
ity; and genomic instability 110,111. Bystander and systemic
effects are secondary to DNA damage and are mediated
by the disruption of gap junction proteins involved in
cell–cell interactions112 and by the release of extracellular
mediators, including ROS, nitric oxide (NO), cytokines
and exosomes78,113–116. Thus, radiation, similarly to the
tissue damage associated with infection by pathogens,
induces the release of DAMP stress signals.
Response to RTX. Relatively little is known about
whether and how the microbiota regulates the response
to RTX. Local irradiation can induce immunogenic
tumour cell death and promote systemic inflammation
and immunity 80,117. However, the effects of radiation
are complex, activating both immunostimulating and
immunosuppressive responses and may be insufficient
to activate a protective anticancer immune response80,117.
Ionizing radiation exerts antitumour responses outside
the field of radiation — known as the abscopal effect
— that are immune mediated and require the activa-
tion of antigen-presenting dendritic cells and immune
T cells118. As the gut microbiota has been shown to affect
the immune response induced by immunogenic cell
death in both chemotherapy and immunotherapy (dis-
cussed below)61,62,119, it can be hypothesized that the gut
microbiota also fulfils a role in the immunostimulatory
effects of RTX.

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 7

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

TLR4 TLR4 TLR2 TLR1 or FPR1 or

? ? TLR6 FPR2

NLRP3
TRAM MAL
+ Gα
MYD88 TRIF Gγ
Gβ

Oxaliplatin or Methotrexate Oxaliplatin or CTX Radiation Mucosal homeostasis

cisplatin • Hyperexpression • Induction of IL-1β • Migration of COX2+ mesenchymal
Priming of tumour- • Exacerbated mucosal injury • Immunogenic stem cells to the crypt base Ligand: bacterial and
infiltrating cell death • Mucosal protection cellular formyl peptides
myeloid cells for Ligand: bacterial LPS • NOX1-dependent ROS production and other DAMPs
NOX2-dependent inducing the NRF2 system
ROS production Ligand: cellular
CpG-ODN HMGB1 Anthracycline
Priming of tumour infiltrating Ligand: MAMPs from L. rhamnosus Induction of T cell
Ligand: MAMP myeloid cells for TNF production immunity
from L. acidophilus
Ligand: LPS (A. shaii) Methotrexate Ligand: endogenous
Increased detoxification by MDR1 annexin-A1
TBI and adoptive T cell transfer expressed on dying
Intestinal mucosal bacterial Ligand: MAMPs from L. reuteri cancer cells
translocation
Ligand: LPS NOD2
Radiation
Induction of caspase
Endosome 1-mediated intestinal and
haematopoietic cell death
Oxaliplatin or
TLR3 TLR3 Double-strand doxorubicin
Radiation DNA breaks • Protection against
Induction of intestinal damage
massive cell • Bacterial translocation
death AIM2
Ligand: bacterial
Leakage of muramyl dipeptide
cellular RNA Nucleus

Figure 4 | Microbiota-triggered innate immune receptors. Microbial
products (microorganism-associated molecular patterns; MAMPs) and
endogenous ligands, often released following tissue damage
(damage-associated molecular patterns; DAMPs) interact with
membrane-bound and cytoplasmic innate immune receptors regulating
nutrition, metabolism, tissue homeostasis, inflammation, innate and
adaptive immunity and, to a lesser extent, morphogenesis213–217. The gut
microbiota promotes platinum cancer therapy by signalling through
myeloid differentiation primary response 88 (MYD88), an adaptor for
both the Toll-like receptor (TLR) and the interleukin‑1 receptor (IL‑1R)
families61,62. ? denotes that the identity of the receptors are yet to be
determined. Activation of TLR4 by MAMPs primes tumour myeloid cells
to respond to the TLR9 agonist CpG-oligodeoxynucleotide (ODN),
induces mucosal bacterial translocation following total body irradiation
(TBI), which favours optimal anticancer activity of adoptive T cell transfer
and mediates mucosal toxicity by methotrexate62,86,87. In response to
immunogenic cell death mediated by oxaliplatin or cyclophosphamide
(CTX), activation of TLR4 by DAMPs in combination with NOD-like
receptor, pyrin domain-containing 3 (NLRP3) inflammasome activation
induces antitumour T cell activation77. Activation of TLR2 by MAMPs
protects against mucosal damage induced by chemotherapy or
radiation86–88,130–132. Radiation activates TLR3 and the inflammasome
Nature Reviews | Cancer
absent in melanoma 2 (AIM2) by inducing leakage of cellular RNA and
double-strand DNA breaks, respectively, resulting in massive cell death
and tissue damage 127–129 . The cytoplasmic nucleotide-binding
oligomerization domain-containing 2 (NOD2) receptor recognizing
bacterial muramyl dipeptides participates in the regulation of intestinal
mucosa homeostasis and protects against chemotherapy-induced
mucosal damage and bacterial translocation82,92. The N‑formyl peptide
receptors (FPRs) recognize bacterial peptides as well as endogenous
ligands 218. FPR2 expressed on apical and lateral membranes of the
colonic crypt has a crucial role in regulating intestinal homeostasis and
inflammation219. The ability of anthracycline to elicit antitumour T cell
immunity requires the interaction of FPR1 on dendritic cells with the
endogenous ligand annexin‑A1, which promotes stable contacts
between dying cancer cells and dendritic cells79. A. shahii, Alistipes shahii;
COX2, cyclooxygenase 2; HMGB1, high mobility group protein B1;
L. acidophilus, Lactobacillus acidophilus; L. reuteri, Lactobacillus reuteri;
L. rhamnosus, Lactobacillus rhamnosus; LPS, lipopolysaccharide; MDR1,
multi-drug resistance protein 1; NOX, NADPH oxidase; NRF2, nuclear
factor erythroid 2‑related factor 2; ROS, reactive oxygen species;
TNF, tumour necrosis factor.

RTX toxicity. The biggest limitations to the effectiveness
and safety of RTX in cancer therapy remain the hetero-
geneity in radiation sensitivity of different cancer types
as well as local and systemic toxicity and stress responses
that cause significant morbidity and may also impair anti-
tumour immunity120,121. RTX is associated with damage to
healthy tissues and, in particular, actively proliferating tis-
sues such as bone marrow and epithelia (for example, the
skin and the digestive tract mucosa) are highly affected by
radiation exposure122. The observed changes in micro-
biota composition at epithelial surfaces in patients and
mice treated with RTX have been proposed to contrib-
ute to the pathogenesis of oral mucositis, diarrhoea,
enteritis, colitis and bone marrow failure123–125. RTX
induces apoptosis in the intestinal crypts, breach of
the intestinal barrier and alterations in the microbiota
composition122. These alterations enable pathobionts to
gain access to the intestinal immune system, leading

8 | ADVANCE ONLINE PUBLICATION www.nature.com/nrc

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

to inflammation in the gut as well as changes to the
microbiota-mediated homeostatic control of mucosal
and systemic innate and adaptive immunity 11,123,126
(FIG. 1).
The severity of RTX-induced oral mucositis and
enteropathy may impose a limitation on therapy
completion. The TLR3 for double-stranded RNA
(dsRNA) regulates irradiation-mediated intestinal
toxicity. Tlr3−/− mice are sensitive to p53‑dependent
radiation-induced apoptosis but are protected against
TLR3‑dependent massive cell death following radiation-
induced leakage of cellular RNA 127,128. Tlr3 −/− mice
survive longer when exposed to ionizing radiation
and show less intestinal toxicity than wild-type mice,
suggesting that blocking TLR3 signalling may reduce
radiation-mediated gastrointestinal damage127,128 (FIG. 4).
Radiation-induced DNA DSBs also activate the DNA
sensor absent in melanoma 2 (AIM2) inflammasome,
resulting in cell death and tissue damage129 (FIG. 4). In
contrast, TLR2‑activating microorganisms such as
the probiotic Lactobacillus rhamnosus GG have been
shown in mice to protect the intestinal mucosa against
chemotherapy- or radiotherapy-induced toxicity by
relocating cyclooxygenase 2 (COX2)-expressing cells
from the villi to the base of the intestinal crypts130
and inducing ROS, which activate the cytoprotective
NRF2 system131,132 (FIG. 4). Indeed, probiotics have been
proved in some clinical studies to be beneficial in pre-
venting radiation-induced enteropathy. Preparations
Box 2 | Role of intestinal microbiota in haematopoietic stem cell transplantation
Allogeneic haematopoietic stem cell transplantation (allo-HSCT) and bone marrow
transplantation are routine procedures for the treatment of haematopoietic
neoplasias198. Before infusing haematopoietic cells, a conditioning regimen using
high-dose chemoradiotherapy or chemotherapy alone is administered to cancer
patients in order to eliminate the malignant cells followed by the graft to restore
haematopoiesis. Although allo-HSCT represents a curative procedure for some
patients, frequent complications can arise, such as graft-versus-host disease (GVHD),
systemic infections and bacteraemia (the presence of bacteria in the blood)199.
Following allo-HSCT, changes in gut microbiota composition and reduced diversity
were observed in both mice and patients199–202, which are probably due to
therapy-associated mucosal damage, inflammation and immune response against
the commensal bacteria, as well as the use of antibiotics. Depending on the type
of antibiotic used during the treatment, the risk of dominance in the gut of
Enterococcus, Streptococcus and various proteobacteria can increase over time after
allo-HSCT203. Dominance of Enterococcus could evolve into vancomycin-resistant
Enterococcus bacteraemia whereas dominance of proteobacteria could evolve into
Gram-negative bacteraemia203. Mortality is significantly higher in patients with lower
diversity in the gut microbiota after HSCT and the diversity of the microbiota at
engraftment is an independent predictor of mortality in allo-HSCT200. In addition, loss
of the genus Blautia as a consequence of the use of antibiotics with specificity for
anaerobic bacteria was associated with increased GVHD and reduced survival204.
Thus, the use of antibiotics, although required to prevent infections during the
engraftment phase, might affect treatment-related toxicities and favour domination
by pathogens and pathobionts. It has been proposed that reconstitution of the
microbiota lost during allo-HSCT through transplantation after haematopoietic
engraftment with autologous faecal preparations collected and stored before the
initiation of conditioning, or restoration of specific bacteria, may improve the clinical
outcome200. Mouse data showing that administration of Lactobacillus spp. reduces the
severity of GVHD201 suggest that the restoration of specific bacterial species through
clinical use of probiotic preparations could be effective in ameliorating HSCT-related
toxicity.
containing L. acidophilus, B. bifidum, Lactobacillus casei
and particularly the VSL#3 formulation contain-
ing Bifidobacterium, Lactobacillus and Streptococcus
spp. have been found to be protective against pelvic
radiation-induced gut toxicity, significantly reducing
the incidence of severe diarrhoea and delaying the
necessity for rescue treatment with loperamide123,133.
Administration of Lactobacillus brevis CD2 lozenges
during radiation and chemotherapy treatment of
patients with head and neck cancer also decreased the
incidence of therapy-induced mucositis and increased
the treatment completion rate134.
Total body irradiation (TBI) is part of the preparative
regimen for haematopoietic stem cell (or bone marrow)
transplantation (BOX 2) and for adoptive T cell transfer
immunotherapy. Although probiotic treatment amelio-
rates local RTX-induced toxicity, germ-free mice survive
longer than conventionally raised mice after TBI and
require a higher radiation dose to induce enteropathy
and 50% mortality 135. After irradiation, germ-free mice
have fewer apoptotic endothelial cells in the intestinal
mucosa and less lymphocyte infiltration than conven-
tionally raised mice135. The resistance of germ-free mice
to the enteric toxicity of TBI could be due to the absence
of gut commensals and pathobionts that in conven-
tionally raised animals may contribute to the mucosal
inflammation and damage that follow the initial insult
mediated by RTX. However, one of the major mecha-
nisms responsible for the resistance of germ-free mice to
TBI is the production of angiopoietin-like 4 (ANGPTL4;
also known as FIAF), a protein inhibitor of lipoprotein
lipase that is suppressed by the presence of the micro-
biota in conventionally raised mice135. ANGPTL4 has
pleiotropic effects on lipid metabolism, angiogenesis and
tissue repair 136. Similarly to conventionally raised mice,
Angptl4 genetically deficient germ-free mice are suscep-
tible to intestinal damage. The transcription of Angptl4 is
regulated by the PPAR family in response to bacteria that
induce the production of small chain fatty acids from
indigestible carbohydrates137–139. Conversely, the bacte-
ria that induce Angptl4 expression include those such as
the Bifidobacterium, Lactobacillus and Streptococcus spp.
that mediate protection against RTX-induced mucosi-
tis and colitis in patients, providing an explanation for
the paradoxical observation that both germ-free mice
and mice treated with probiotic bacteria are resistant to
radiotherapy toxicity.
Mice are more susceptible to irradiation toxicity
during daytime than at night-time140–142. The circadian
rhythm affects radiation-induced apoptosis and activa-
tion of p53‑responsive genes in the intestine, bone mar-
row and peripheral blood140–142. As the circadian rhythm
is associated both with diurnal variation of gut microbial
composition and short chain fatty acid production, and
with cyclic expression in the gut of microbiota-detecting
innate immune receptors resulting in oscillation in the
number of circulating inflammatory cells that may mod-
ulate radiation sensitivity 143–147, it is possible to hypoth-
esize that the observed changes in radiation sensitivity
may be linked to the circadian variation in microbiota
composition.

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 9

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

The data presently available suggest that the micro­
biota composition modulates the response and repair fol-
lowing irradiation-induced damage. However, the ability
of the microbiota to license the genotoxicity of chemo-
therapeutic agents by priming for ROS production and
inflammatory responses62 suggests that the microbiota
may also control the direct nuclear effect of ionizing
radiation and/or its bystander effect. Recently, it has
been reported that conventionally raised mice exposed
to high-energy protons are more susceptible than mice
with a restricted gut microbiota to single-strand and per-
sistent dsDNA damage in peripheral blood leukocytes148.
A better understanding of the mechanism of radiation
non-targeted effects and their regulation by the commen-
sal microbiota or its therapeutic manipulation will be
invaluable in increasing therapeutic effectiveness, reduc-
ing collateral toxicity of radiotherapy and managing
the health effects of accidental exposure to radiation.
Immunotherapy and microbiota
In most patients treated with traditional cancer ther-
apies, tumours become resistant to therapy and the
chances of tumour recurrence are high149. Recently,
immunotherapy approaches have shown potential in
treating haematopoietic and solid cancers150 and have for
the first time achieved long-lasting complete responses
in a proportion of patients with metastatic melanoma
and lung cancer who were unresponsive to conventional
therapies. However, the efficacy of immunotherapy is
still limited by the variability of the immune response
in different patients and the different susceptibility of
tumour types151. The emerging knowledge of the abil-
ity of the gut microbiota to modulate the response to
immuno­therapy offers new possibilities to improve its
efficacy by targeting the microbiota33,62,152.
Adoptive T cell transfer. The first report of the ability
of the gut microbiota to sustain the anticancer effect of
immune therapy was the observation that the efficacy
of adoptive transfer of tumour-specific cytotoxic T cells
following TBI was reduced in mice treated with anti-
biotics153. The proliferation of the transferred T cells in
the tumour and their antitumour cytotoxic activity is
enhanced by the TBI-induced translocation of the gut
microbiota into the mesenteric lymph nodes153. This
effect of the translocated microbiota requires TLR4
signalling, as TLR4‑deficient mice respond poorly to
the treatment whereas the administration of a TLR4
ligand, lipopolysaccharide (LPS), can enhance the anti-
tumour response induced by adoptive T cell transfer in
non-irradiated lymphodepleted mice153 (FIG. 4). These
observations may help to explain why patients with
metastatic melanoma treated with adoptive cell therapy
using tumour-infiltrating lymphocytes respond better
if a conditioning regimen including myelo­ablative
radiotherapy is used154. Similarly, morbidity and sur-
vival after haematopoietic stem cell transplantation
preceded by a conditioning regimen using TBI and/or
chemotherapy have been shown to be modulated by
the composition of the patient’s gut microbiota after
therapy (BOX 2).
CpG-oligodeoxynucleotide intratumour therapy.
Intratumoural treatment with the TLR9 agonist
CpG-oligodeoxynucleotide (CpG-ODN) induces an
antitumour effect that is potentiated when the immuno­
suppressive effect of the cytokine IL‑10, produced
by tumour-infiltrating Treg cells and myeloid cells, is
prevented by using IL‑10 receptor (IL‑10R) antibod-
ies155–157. CpG-ODNs act by inducing the secretion from
myeloid cells of pro-inflammatory cytokines such as
tumour necrosis factor (TNF) and IL‑12 that provoke
a rapid haemorrhagic necrosis and redirect tumour-
infiltrating macrophages and dendritic cells from an anti-
inflammatory to a pro-inflammatory state. In this
pro-inflammatory microenvironment, an antigen-
specific adaptive T cell antitumour immunity is elicited
that can clear tumours in most conventionally raised
mice155. In contrast, in germ-free mice or mice treated orally
with a cocktail of non-absorbable antibiotics (vanco­mycin,
neomycin and imipenem), treatment of sub­cutaneous
tumours with CpG-ODNs and anti‑IL‑10R is largely
inefficient and the tumours progress62. In microbiota-
depleted mice, tumour-infiltrating myeloid cells fail
to produce pro-inflammatory cytokines in response to
CpG-ODNs and neither the TNF-dependent haemor-
rhagic necrosis nor the antitumour adaptive immunity
are efficiently induced by the treatment62.
In tumours of microbiota-depleted mice, the total
number of infiltrating inflammatory monocytes before
treatment is unaltered while the number of monocyte-
derived Ly6C+ major histocompatibility complex (MHC)
class II+ cells is reduced62. These data suggest impaired
differentiation of bone marrow-derived infiltrating
inflammatory monocytes into cells with the charac-
teristics and functions of macrophages and dendritic
cells in the tumour microenvironment of microbiota-
depleted mice. However, only after CpG-ODN treatment
are major differences in gene expression, particularly in
genes encoding inflammatory products and markers, for
example, TNF and IL‑12, observed in tumour-infiltrating
myeloid cell subsets between microbiota-depleted and
conventionally raised mice. A partial reduction in
the response of the myeloid cell to CpG-ODNs is also
observed in TLR4‑deficient mice, and oral administra-
tion of the TLR4 agonist LPS to microbiota-depleted
mice reconstitutes the responsiveness of the myeloid
cells62 (FIG. 4). These results suggest that the gut micro­
biota, in part through TLR4 signalling, systemically
primes myeloid cells, including tumour-infiltrating cells,
for responsiveness to TLR9, a phenomenon that resem-
bles the ‘trained’ innate resistance described with various
innate effector cell types158.
TNF production in response to CpG-ODNs corre-
lates both positively and negatively with the frequency
of various bacterial genera in the faecal microbiota of
mice at the time of the treatment (FIG. 3). For example,
the frequencies of the Gram-negative Alistipes and the
Gram-positive Ruminococcus genera are positively cor-
related with TNF production whereas the frequencies of
the Lactobacillus genus, including Lactobacillus murinum,
Lactobacillus intestinalis and Lactobacillus fermentum
are negatively correlated59. After mice were exposed to

10 | ADVANCE ONLINE PUBLICATION www.nature.com/nrc

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Prebiotics
Non-digestible food
ingredients, often containing
fibre, that promote the growth
of beneficial microorganisms in
the intestines.
Enterotypes
Classification of individuals
based on the composition of
their gut bacterial ecosystem,
each enterotype having distinct
clusters of organisms with
characteristic predominant
bacterial species.
antibiotics, recolonization of their gut microbiota with
Alistipes shahii restored the ability of myeloid cells to
produce TNF, whereas oral administration of L. fer­
mentum to conventionally raised mice impaired TNF
production62. These results indicate that, although
complete depletion of the gut microbiota results in abo-
lition of the ‘training’ of myeloid cells for response to
CpG-ODNs, different bacterial species can have oppo-
site effects. Thus, changes in gut microbiota composition
or alteration in the frequency of individual species that
can be obtained by therapeutic approaches such as the
use of antibiotics, prebiotics or probiotics, could be used
to modulate the response to immunotherapies.
Immune checkpoint inhibitors. In many cancer patients,
antitumour immunity is dormant or suppressed but can
be reactivated by releasing the immunological brakes that
tumours use to escape antitumour immunity 150,159–161. In
particular, the immune checkpoint inhibitors (antibod-
ies against cytotoxic T lymphocyte-associated antigen 4
(CTLA4) expressed on activated T effector cells and
Treg cells and programmed cell death protein 1 (PD1)
or its ligand PD1 ligand 1 (PDL1)) have shown strong
antitumour activity in experimental animal models and
substantial and long-lasting clinical efficacy in patients
with cancer (melanoma, renal cell cancer, lung cancer
and others being tested in ongoing clinical trials)162.
Interpatient variability in therapeutic responsiveness and
the failure of certain types of tumour to respond remain
of great concern154. In addition, immune checkpoint
inhibitors can induce immune-related adverse effects,
particularly colitis and inflammation of the pituitary
gland in response to CTLA4 antibodies, and thyroid
dysfunction and pneumonitis following blockade of the
PD1–PDL1 interaction163. Recently, two studies have
reported the association of the gut microbiota in regu-
lating the efficacy of anti‑CTLA4 and anti‑PDL1 cancer
therapy 33,152.
Subcutaneous tumours respond poorly to
anti‑CTLA4 in antibiotic-treated or germ-free mice33.
CTLA4 antagonism induces T cell-mediated mucosal
damage in the ileum and colon that is associated with
an alteration in the composition of the intestinal and
faecal microbiota33. In general, the relative frequency
in tumour-bearing mice of both Bacteroidales and
Burkholderiales (for example, Bacteroides thetaiotao­
micron, Bacteroides uniformis and Burkholderia cepa­
cia) decreases, whereas that of Clostridiales increases,
upon anti‑CTLA4 treatment. However, the frequency
of the immunoregulatory Bacteroides fragilis remains
constant 33. Oral feeding of either B. thetaiotaomicron or
B. fragilis to microbiota-depleted mice restores the ther-
apeutic response to anti‑CTLA4 by inducing matura-
tion of intratumoural dendritic cells and a TH1 response
detected in the tumour-draining lymph nodes33. The
feeding of mice with combined B. fragilis and B. cepacia
also restores the therapeutic response to anti‑CTLA4,
but unlike single administration with either B. theta­
iotaomicron or B. fragilis, the combined treatment also
significantly decreases the extent of intestinal damage
and colitis associated with the antitumour response33.
In patients treated with anti‑CTLA4, it was simi-
larly observed that an increased frequency of the
Bacteroidetes phylum correlated with resistance to
colitis whereas underrepresentation of bacterial spe-
cies expressing genetic pathways involved in polyamine
transport and vitamin B biosynthesis was associated
with an increased risk of colitis164. Individual antibiotics
change the composition of the gut microbiota, affect-
ing the anti‑CTLA4 response; for example, vanco­mycin
enhances the efficacy of CTLA4 blockade in mice by
decreasing the abundance of Gram-positive bacteria
while preserving the Gram-negative Bacteroidales and
Burkholderiales33.
In both mice and patients, anti‑CTLA4 induces a
TH1 response specific for either B. thetaiotaomicron or
B. fragilis 33. The response to anti‑CTLA4 in microbiota-
depleted mice can be restored by adoptive transfer of
B. fragilis-specific T cells33. These results are reminiscent
of the long-lived microbiota-specific TH1 cell response
that is generated in acute intestinal infections and which
primes for a rapid TH1 polarized immune response to
the pathogens during subsequent infections in the same
anatomical location165. How the microbiota-specific TH1
cells induced by anti‑CTLA4 can reach distant tumour
microenvironments and help to mount a tumour-specific
immune response are intriguing questions.
Analysis of the faecal microbiota from patients with
melanoma before and after treatment with anti‑CTLA4
showed a change in the relative proportions of three
dominant enterotypes observed in these patients33.
Enterotype A was dominated by Prevotella spp.
whereas enterotypes B and C were driven by different
Bacteroides spp. The number of patients with enterotype
C increased after treatment at the expense of patients
with enterotype B. Transfer of the faecal microbiota from
patients with each of the three human enterotypes into
tumour-bearing, germ-free mice showed that only the
transfer of enterotype C resulted in colonization with
B. thetaiotaomicron or B. fragilis and enhanced respon-
siveness to anti‑CTLA4. This observation indicates that
anti‑CTLA4 may in some patients alter the composi-
tion of the gut microbiota in a direction that favours its
antitumour activity 33.
Unlike anti‑CTLA4, PDL1 blockade does not induce
intestinal damage and does not seem to have an absolute
requirement for the presence of the gut microbiota152,166.
However, when subcutaneous B16.SIY melanoma
growth was compared between C57BL/6 mice obtained
from two different vendors, Jackson (JAX) and Taconic,
it was observed that tumours grew more quickly in
Taconic mice than in JAX mice152. This correlated with
the ability of the mice to induce an immune response
against the tumour as shown by higher infiltration of
CD8+ T cells in the tumour of JAX mice compared with
Taconic mice152. Although anti‑PDL1 had an antitumour
effect in both Taconic and JAX mice, the slower-growing
tumour of JAX mice was more profoundly affected,
with almost complete arrest of tumour progression152.
Taconic mice cohoused with JAX mice acquired the
same rate of tumour growth, antitumour resistance and
responsiveness to anti‑PDL1 as observed in JAX mice152.

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 11

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

The response to anti‑PDL1 in C57BL/6 mice trans-
planted with the B16.SIY tumours is significantly
associated with the relative frequency in the gut
microbiota of the Bifidobacterium genus, includ-
ing Bifidobacterium breve, Bifidobacterium longum
and Bifidobacterium  adolescentis 152. Oral adminis-
tration to mice of a commercially available probiotic
cocktail of Bifidobacterium spp. (including B. breve and
B. longum), alone or with anti‑PDL1, enhanced CD8+
T cell-induced antitumour activity 152. Furthermore,
the effect of Bifidobacterium was abolished in CD8+
T cell-depleted mice, indicating that Bifidobacterium
action is dependent on cytotoxic T cell activity 152. These
results indicate that the mechanism by which anti‑PDL1
treatment enhances the antitumour immune response
does not require microbiota-specific inflammation and
immune activation, unlike that induced by anti‑CTLA4.
However, the therapeutic effectiveness of anti‑PDL1
treatment can be facilitated in tumour-bearing animals
with a pre-existing antitumour immune response when
Bifidobacterium spp. are well represented in the gut
microbiota.
Concluding remarks
The abundant gut microbiota regulates carcinogenesis
and response to antitumour therapy both in the intes-
tinal tract and in extra-intestinal tissues, although the
role of the microbiota at other epithelial barriers remains
poorly understood11. Many of the mechanisms by which
the gut microbiota affects inflammation, immunity,
carcinogenesis and response to therapy at the local
level have been characterized7–10. However, much less is
known about the distant effects by which the micro­biota
at the epithelial barriers regulates not only immunity
and carcinogenesis, but also many of the physiological
functions of the organism11.
Most studies in the 3‑year-old scientific field
investigating the modulation of cancer therapy by the
microbiota have been in mice, and the translation of
these experimental findings to the clinic remains a
challenge. Experimental mouse tumour models using
transplantable cell lines, chemical carcinogenesis or
genetically modified mice have been instrumental in
major advances in immune cancer therapy and in the
detailed analysis of the molecular mechanisms involved
in carcinogenesis and cancer therapy, yet they fail to
fully reproduce the features of spontaneous human
tumours167. In addition, the composition of the gut
microbiota varies between mice, even in mice with
identical genetic backgrounds and housed under sim-
ilar conditions. Identical mouse strains obtained from
different vendors may also have differences in micro-
biota composition that profoundly affect their physi­
ology 152,168. The phenotype of genetically modified
mice is not always due to a direct effect of the mutated
gene but in some cases is mediated by the presence in the
mutated mouse strains of an altered micro­biota com-
position; the altered microbiota is responsible for the
observed phenotype and can be transmitted by cohousing
or faecal transplant to wild-type mice, which acquire the
same phenotype as the genetically modified mice169,170.
A microbiota imbalance that in genetically altered mice
influences their phenotype is not limited to bacteria
but can involve fungi, protozoa, viruses and reactiva-
tion of endogenous retroviruses171–176. Thus, preclini-
cal ana­lysis of cancer drugs in mice can be profoundly
affected by the composition of the microbiota in differ-
ent animal populations, necessitating great caution in
the interpretation of the results and in their translation
to the clinic.
Identification of the most favourable microbiota
composition in clinical situations will require a very
extensive human database with careful analysis of the
correlation of different bacterial species with clini-
cal response. It is possible to confirm the correlative
results of human studies by transplanting human faecal
microbial communities into germ-free mice177. These
humanized mice are stably and heritably colonized and
at least in part reproduce the bacterial diversity of the
donor microbiota177,178. They are susceptible to per-
turbations, such as changes in diet, that are known to
modify the human microbiota177. Furthermore, even if
mice are colonized with the human microbiota, their
physiological and immune responses are similar but not
identical to those of humans4. For example, the genus
Bifidobacterium, which has been shown to induce an
immune response against tumours and facilitate the
therapeutic efficacy of anti‑PDL1 in mice152, can acti-
vate immune cells using two innate immune receptors,
TLR2, which is mostly immunosuppressive, and TLR9,
which is strongly immunostimulating, in the mouse179.
However, the cellular expression of TLR9 in humans
is very different from its expression in the mouse.
Mouse TLR9 is expressed on all myeloid and dendritic
cells whereas in humans its expression is restricted to
plasmacytoid dendritic cells and B cells180. Thus, it is
not possible yet to conclude that activation of TLR9 in
humans by Bifidobacterium spp. has the same immuno­
stimulating activity as observed in the mouse, and
detailed clinical data are required to determine whether
Bifidobacterium-containing probiotics would stimulate
antitumour activity also in patients.
Once the most favourable microbiota composition
for each clinical condition has been identified, the next
challenge will be how to modify the patient’s micro­biota.
The resilience and stability of the gut microbiota and
its responsiveness to physiological, pathological
and environmental changes are characteristics that
would enable us to use the microbiota composition as a
biomarker, a diagnostic tool and possibly a therapeutic
target 24. Although the field of therapeutic intervention
through targeting the microbiota is still primitive, some
approaches are already possible. For example, the valida-
tion of the microbiota as a therapeutic target is provided
by studies showing that patients can be recolonized
with a resilient and stable modified microbiota to fight
antibiotic-resistant pathogens181. The ultimate goal is to
discover a bacterial species or a combination of species
that both reduces systemic toxicity and promotes anti-
cancer therapy. Thus, targeting the microbiota in cancer
and other diseases is likely to become one of the next
frontiers for precision and personalized medicine.

12 | ADVANCE ONLINE PUBLICATION www.nature.com/nrc

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

1. Costello, E. K., Stagaman, K., Dethlefsen, L.,
Bohannan, B. J. & Relman, D. A. The application of
ecological theory toward an understanding of the
human microbiome. Science 336, 1255–1262
(2012).
2. Bosch, T. C. & McFall-Ngai, M. J. Metaorganisms as
the new frontier. Zoology (Jena) 114, 185–190
(2011).
3. Dzutsev, A., Goldszmid, R. S., Viaud, S., Zitvogel, L. &
Trinchieri, G. The role of the microbiota in
inflammation, carcinogenesis, and cancer therapy. Eur.
J. Immunol. 45, 17–31 (2015).
4. Smith, K., McCoy, K. D. & Macpherson, A. J. Use of
axenic animals in studying the adaptation of mammals
to their commensal intestinal microbiota. Semin.
Immunol. 19, 59–69 (2007).
5. Gustafsson, B. E., Daft, F. S., McDaniel, E. G.,
Smith, J. C. & Fitzgerald, R. J. Effects of
vitamin K‑active compounds and intestinal
microorganisms in vitamin K‑deficient germfree rats.
J. Nutr. 78, 461–468 (1962).
6. Gordon, H. A., Bruckner-Kardoss, E. &
Wostmann, B. S. Aging in germ-free mice: life tables
and lesions observed at natural death. J. Gerontol. 21,
380–387 (1966).
7. De Santis, S., Cavalcanti, E., Mastronardi, M.,
Jirillo, E. & Chieppa, M. Nutritional keys for intestinal
barrier modulation. Front. Immunol. 6, 612 (2015).
8. Vaishnava, S., Behrendt, C. L., Ismail, A. S.,
Eckmann, L. & Hooper, L. V. Paneth cells directly sense
gut commensals and maintain homeostasis at the
intestinal host–microbial interface. Proc. Natl Acad.
Sci. USA 105, 20858 (2008).
This paper identified the role of microbiota
signalling in maintaining the host–commensal
homeostasis through MYD88‑coupled receptors in
epithelial cells.
9. Peterson, L. W. & Artis, D. Intestinal epithelial cells:
regulators of barrier function and immune homeostasis.
Nat. Rev. Immunol. 14, 141–153 (2014).
10. Belkaid, Y. & Naik, S. Compartmentalized and
systemic control of tissue immunity by commensals.
Nat. Immunol. 14, 646–653 (2013).
This review discusses the role of the microbiota at
different epithelial barriers in regulating immunity
both locally and systemically.
11. Belkaid, Y. & Hand, T. W. Role of the microbiota in
immunity and inflammation. Cell 157, 121–141 (2014).
12. Sender, R., Fuchs, S. & Milo, R. Revised estimates for
the number of human and bacteria cells in the body.
PLoS Biol. 14, e1002533 (2016).
13. Chow, J., Tang, H. & Mazmanian, S. K. Pathobionts of
the gastrointestinal microbiota and inflammatory
disease. Curr. Opin. Immunol. 23, 473–480 (2011).
14. Erny, D. et al. Host microbiota constantly control
maturation and function of microglia in the CNS. Nat.
Neurosci. 18, 965–977 (2015).
15. Khosravi, A. et al. Gut microbiota promote
hematopoiesis to control bacterial infection. Cell Host
Microbe 15, 374–381 (2014).
16. Trompette, A. et al. Gut microbiota metabolism of
dietary fiber influences allergic airway disease and
hematopoiesis. Nat. Med. 20, 159–166 (2014).
17. Chung, W. S. F. et al. Modulation of the human gut
microbiota by dietary fibres occurs at the species level.
BMC Biol. 14, 3 (2016).
18. Dinan, T. G. & Cryan, J. F. Regulation of the stress
response by the gut microbiota: implications for
psychoneuroendocrinology.
Psychoneuroendocrinology 37, 1369–1378 (2012).
19. Sommer, F. & Bäckhed, F. The gut microbiota engages
different signaling pathways to induce Duox2
expression in the ileum and colon epithelium. Mucosal
Immunol. 8, 372–379 (2015).
20. Wells, J. M., Rossi, O., Meijerink, M. & van Baarlen, P.
Epithelial crosstalk at the microbiota–mucosal
interface. Proc. Natl Acad. Sci. USA 108 (Suppl. 1),
4607–4614 (2011).
21. Tulstrup, M. V.‑L. et al. Antibiotic treatment affects
intestinal permeability and gut microbial composition
in Wistar rats dependent on antibiotic class. PLoS
ONE 10, e0144854 (2015).
22. Backhed, F. et al. Dynamics and stabilization of the
human gut microbiome during the first year of life. Cell
Host Microbe 17, 690–703 (2015).
23. Dominguez-Bello, M. G. et al. Delivery mode shapes
the acquisition and structure of the initial microbiota
across multiple body habitats in newborns. Proc. Natl
Acad. Sci. USA 107, 11971–11975 (2010).
References 22 and 23 describe the establishment
of the human microbiota during early life.
24. Faith, J. J. et al. The long-term stability of the human
gut microbiota. Science 341, 1237439 (2013).
25. Oh, J., Byrd, A. L., Park, M., Kong, H. H. & Segre, J. A.
Temporal stability of the human skin microbiome. Cell
165, 854–866 (2016).
26. Goodrich, J. K. et al. Human genetics shape the gut
microbiome. Cell 159, 789–799 (2014).
27. Turnbaugh, P. J. et al. A core gut microbiome in obese
and lean twins. Nature 457, 480–484 (2009).
28. David, L. A. et al. Host lifestyle affects human
microbiota on daily timescales. Genome Biol. 15, R89
(2014).
29. David, L. A. et al. Diet rapidly and reproducibly alters
the human gut microbiome. Nature 505, 559–563
(2014).
30. Garrett, W. S. et al. Colitis-associated colorectal cancer
driven by T‑bet deficiency in dendritic cells. Cancer Cell
16, 208–219 (2009).
31. Couturier-Maillard, A. et al. NOD2‑mediated dysbiosis
predisposes mice to transmissible colitis and
colorectal cancer. J. Clin. Invest. 123, 700–711
(2013).
32. Hu, B. et al. Microbiota-induced activation of epithelial
IL‑6 signaling links inflammasome-driven inflammation
with transmissible cancer. Proc. Natl Acad. Sci. USA
110, 9862–9867 (2013).
References 31 and 32 demonstrated that the
carcinogenic phenotypes associated with dysbiosis of
the microbiota in genetically mutated mice can be
transmitted to wild-type mice by microbiota transfer.
33. Vetizou, M. et al. Anticancer immunotherapy by
CTLA‑4 blockade relies on the gut microbiota. Science
350, 1079–1084 (2015).
This paper presented the first demonstration of
the role of the microbiota in modulating
responsiveness to anti‑CTLA4 therapy.
34. DeVita, V. T. Jr & Chu, E. A history of cancer
chemotherapy. Cancer Res. 68, 8643–8653
(2008).
35. Sancho-Martinez, S. M., Prieto-Garcia, L., Prieto, M.,
Lopez-Novoa, J. M. & Lopez-Hernandez, F. J.
Subcellular targets of cisplatin cytotoxicity: an
integrated view. Pharmacol. Ther. 136, 35–55
(2012).
36. Mitchell, E. P. Gastrointestinal toxicity of
chemotherapeutic agents. Semin. Oncol. 33,
106–120 (2006).
37. Spanogiannopoulos, P., Bess, E. N., Carmody, R. N. &
Turnbaugh, P. J. The microbial pharmacists within us:
a metagenomic view of xenobiotic metabolism. Nat.
Rev. Microbiol. 14, 273–287 (2016).
38. Li, H. & Jia, W. Cometabolism of microbes and host:
implications for drug metabolism and drug-induced
toxicity. Clin. Pharmacol. Ther. 94, 574–581 (2013).
39. Feng, R. et al. Transforming berberine into its
intestine-absorbable form by the gut microbiota. Sci.
Rep. 5, 12155 (2015).
40. Maurice, C. F., Haiser, H. J. & Turnbaugh, P. J.
Xenobiotics shape the physiology and gene expression
of the active human gut microbiome. Cell 152, 39–50
(2013).
41. Montassier, E. et al. Chemotherapy-driven dysbiosis in
the intestinal microbiome. Aliment. Pharmacol. Ther.
42, 515–528 (2015).
42. Wilson, I. D. & Nicholson, J. K. Gut microbiome
interactions with drug metabolism, efficacy, and
toxicity. Transl Res. 179, 204–222 (2017).
43. Haiser, H. J. & Turnbaugh, P. J. Developing a
metagenomic view of xenobiotic metabolism.
Pharmacol. Res. 69, 21–31 (2013).
44. Carmody, R. N. & Turnbaugh, P. J. Host–microbial
interactions in the metabolism of therapeutic and diet-
derived xenobiotics. J. Clin. Invest. 124, 4173–4181
(2014).
45. Bjorkholm, B. et al. Intestinal microbiota regulate
xenobiotic metabolism in the liver. PLoS ONE 4,
e6958 (2009).
46. Selwyn, F. P., Cheng, S. L., Klaassen, C. D. & Cui, J. Y.
Regulation of hepatic drug-metabolizing enzymes in
germ-free mice by conventionalization and probiotics.
Drug Metab. Dispos. 44, 262–274 (2016).
47. Selwyn, F. P., Cui, J. Y. & Klaassen, C. D. RNA-Seq
quantification of hepatic drug processing genes in
germ-free mice. Drug Metab. Dispos. 43, 1572–1580
(2015).
48. Selwyn, F. P. et al. Developmental regulation of drug-
processing genes in livers of germ-free mice. Toxicol.
Sci. 147, 84–103 (2015).
49. Kang, M. J. et al. The effect of gut microbiota on drug
metabolism. Expert Opin. Drug Metab. Toxicol. 9,
1295–1308 (2013).
50. Yip, L. Y. & Chan, E. C. Investigation of host-gut
microbiota modulation of therapeutic outcome. Drug
Metab. Dispos. 43, 1619–1631 (2015).
51. Fujita, K. & Sparreboom, A. Pharmacogenetics of
irinotecan disposition and toxicity: a review. Curr. Clin.
Pharmacol. 5, 209–217 (2010).
52. Stringer, A. M. et al. Faecal microflora and beta-
glucuronidase expression are altered in an irinotecan-
induced diarrhea model in rats. Cancer Biol. Ther. 7,
1919–1925 (2008).
53. Lin, X. B. et al. Irinotecan (CPT‑11) chemotherapy
alters intestinal microbiota in tumour bearing rats.
PLoS ONE 7, e39764 (2012).
54. Dabek, M., McCrae, S. I., Stevens, V. J., Duncan, S. H.
& Louis, P. Distribution of beta-glucosidase and beta-
glucuronidase activity and of beta-glucuronidase gene
gus in human colonic bacteria. FEMS Microbiol. Ecol.
66, 487–495 (2008).
55. McIntosh, F. M. et al. Phylogenetic distribution of
genes encoding beta-glucuronidase activity in human
colonic bacteria and the impact of diet on faecal
glycosidase activities. Environ. Microbiol. 14,
1876–1887 (2012).
56. Takasuna, K. et al. Involvement of beta-glucuronidase
in intestinal microflora in the intestinal toxicity of the
antitumor camptothecin derivative irinotecan
hydrochloride (CPT‑11) in rats. Cancer Res. 56,
3752–3757 (1996).
57. Wallace, B. D. et al. Alleviating cancer drug toxicity by
inhibiting a bacterial enzyme. Science 330, 831–835
(2010).
58. Mego, M. et al. Prevention of irinotecan induced
diarrhea by probiotics: a randomized double blind,
placebo controlled pilot study. Complement. Ther.
Med. 23, 356–362 (2015).
59. Wallace, B. D. et al. Structure and inhibition of
microbiome beta-glucuronidases essential to the
alleviation of cancer drug toxicity. Chem. Biol. 22,
1238–1249 (2015).
60. Lehouritis, P. et al. Local bacteria affect the efficacy
of chemotherapeutic drugs. Sci. Rep. 5, 14554
(2015).
61. Viaud, S. et al. The intestinal microbiota modulates
the anticancer immune effects of cyclophosphamide.
Science 342, 971–976 (2013).
This paper describes the role of the microbiota in
modulating the anticancer effect of CTX.
62. Iida, N. et al. Commensal bacteria control cancer
response to therapy by modulating the tumor
microenvironment. Science 342, 967–970 (2013).
This paper describes the role of the microbiota in
modulating the efficacy of cancer therapy with
CpG-ODNs and platinum drugs.
63. Galluzzi, L. et al. Systems biology of cisplatin
resistance: past, present and future. Cell Death Dis. 5,
e1257 (2014).
64. Roy, S., Ryals, M. M., Van den Bruele, A. B.,
Fitzgerald, T. S. & Cunningham, L. L. Sound
preconditioning therapy inhibits ototoxic hearing loss
in mice. J. Clin. Invest. 123, 4945–4949 (2013).
65. Abuzeid, W. M. et al. Molecular disruption of RAD50
sensitizes human tumor cells to cisplatin-based
chemotherapy. J. Clin. Invest. 119, 1974–1985
(2009).
66. Pabla, N. & Dong, Z. Cisplatin nephrotoxicity:
mechanisms and renoprotective strategies. Kidney Int.
73, 994–1007 (2008).
67. Wagner, J. M. & Karnitz, L. M. Cisplatin-induced
DNA damage activates replication checkpoint
signaling components that differentially affect tumor
cell survival. Mol. Pharmacol. 76, 208–214 (2009).
68. Zhu, S., Pabla, N., Tang, C., He, L. & Dong, Z. DNA
damage response in cisplatin-induced nephrotoxicity.
Arch. Toxicol. 89, 2197–2205 (2015).
69. Park, S. B. et al. Chemotherapy-induced peripheral
neurotoxicity: a critical analysis. CA Cancer J. Clin. 63,
419–437 (2013).
70. Hooper, L. V. & Macpherson, A. J. Immune
adaptations that maintain homeostasis with the
intestinal microbiota. Nat. Rev. Immunol. 10,
159–169 (2010).
71. Sonis, S. T. The pathobiology of mucositis. Nat. Rev.
Cancer 4, 277–284 (2004).
72. Kim, S., Lee, T. J., Park, J. W. & Kwon, T. K.
Overexpression of cFLIPs inhibits oxaliplatin-mediated
apoptosis through enhanced XIAP stability and Akt
activation in human renal cancer cells. J. Cell.
Biochem. 105, 971–979 (2008).
73. Laurent, A. et al. Controlling tumor growth by
modulating endogenous production of reactive
oxygen species. Cancer Res. 65, 948–956 (2005).

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 13

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

74. Gui, Q. F., Lu, H. F., Zhang, C. X., Xu, Z. R. &
Yang, Y. H. Well-balanced commensal microbiota
contributes to anti-cancer response in a lung cancer
mouse model. Genet. Mol. Res. 14, 5642–5651
(2015).
75. Tesniere, A. et al. Immunogenic death of colon cancer
cells treated with oxaliplatin. Oncogene 29, 482–491
(2010).
76. Michaud, M. et al. Subversion of the chemotherapy-
induced anticancer immune response by the ecto-
ATPase CD39. Oncoimmunology 1, 393–395 (2012).
77. Ghiringhelli, F. et al. Activation of the NLRP3
inflammasome in dendritic cells induces
IL‑1beta‑dependent adaptive immunity against
tumors. Nat. Med. 15, 1170–1178 (2009).
78. Pateras, I. S. et al. The DNA damage response and
immune signaling alliance: is it good or bad? Nature
decides when and where. Pharmacol. Ther. 154,
36–56 (2015).
79. Vacchelli, E. et al. Chemotherapy-induced antitumor
immunity requires formyl peptide receptor 1. Science
350, 972–978 (2015).
80. Kroemer, G., Galluzzi, L., Kepp, O. & Zitvogel, L.
Immunogenic cell death in cancer therapy. Annu. Rev.
Immunol. 31, 51–72 (2013).
This paper reviewed the mechanisms underlying
the phenomenon of immunogenic cell death in
cancer therapy.
81. Zwielehner, J. et al. Changes in human fecal
microbiota due to chemotherapy analyzed by TaqMan-
PCR, 454 sequencing and PCR-DGGE fingerprinting.
PLoS ONE 6, e28654 (2011).
82. Daillere, R. et al. Enterococcus hirae and Barnesiella
intestinihominis facilitate cyclophosphamide-induced
therapeutic immunomodulatory effects. Immunity 45,
931–943 (2016).
This paper characterized the role of different
bacterial species in regulating antitumour T cell
responses induced by CTX.
83. Ghoreschi, K. et al. Generation of pathogenic TH17
cells in the absence of TGF-beta signalling. Nature
467, 967–971 (2010).
84. Chitapanarux, I. et al. Randomized controlled trial of
live Lactobacillus acidophilus plus Bifidobacterium
bifidum in prophylaxis of diarrhea during radiotherapy
in cervical cancer patients. Radiat. Oncol. 5, 31
(2010).
85. Wang, Y. et al. Pharmacological inhibition of NADPH
oxidase protects against cisplatin induced
nephrotoxicity in mice by two step mechanism. Food
Chem. Toxicol. 83, 251–260 (2015).
86. Cario, E. Toll-like receptors in the pathogenesis of
chemotherapy-induced gastrointestinal toxicity. Curr.
Opin. Support. Palliat. Care 10, 157–164 (2016).
87. Frank, M. et al. TLR signaling modulates side effects of
anticancer therapy in the small intestine. J. Immunol.
194, 1983–1995 (2015).
88. Mercado-Lubo, R. & McCormick, B. A. The interaction
of gut microbes with host ABC transporters. Gut
Microbes 1, 301–306 (2010).
89. Napenas, J. J. et al. Molecular methodology to assess
the impact of cancer chemotherapy on the oral
bacterial flora: a pilot study. Oral Surg. Oral Med.
Oral Pathol. Oral Radiol. Endod. 109, 554–560
(2010).
90. Niu, Q. Y., Li, Z. Y., Du, G. H. & Qin, X. M. 1H NMR
based metabolomic profiling revealed doxorubicin-
induced systematic alterations in a rat model.
J. Pharm. Biomed. Anal. 118, 338–348 (2016).
91. Rigby, R. J. et al. Intestinal bacteria are necessary for
doxorubicin-induced intestinal damage but not for
doxorubicin-induced apoptosis. Gut Microbes 7,
414–423 (2016).
92. Nigro, G., Rossi, R., Commere, P. H., Jay, P. &
Sansonetti, P. J. The cytosolic bacterial peptidoglycan
sensor Nod2 affords stem cell protection and links
microbes to gut epithelial regeneration. Cell Host
Microbe 15, 792–798 (2014).
93. Jiang, C. et al. Intestine-selective farnesoid X receptor
inhibition improves obesity-related metabolic
dysfunction. Nat. Commun. 6, 10166 (2015).
94. Parseus, A. et al. Microbiota-induced obesity requires
farnesoid X receptor. Gut 66, 429–437 (2016).
95. Das, S. K. et al. Adipose triglyceride lipase contributes
to cancer-associated cachexia. Science 333,
233–238 (2011).
96. Ruud, J. et al. Inflammation- and tumor-induced
anorexia and weight loss require MyD88 in
hematopoietic/myeloid cells but not in brain
endothelial or neural cells. FASEB J. 27, 1973–1980
(2013).
97. Suárez-Zamorano, N. et al. Microbiota depletion
promotes browning of white adipose tissue and
reduces obesity. Nat. Med. 21, 1497–1501 (2015).
98. de Matos-Neto, E. M. et al. Systemic inflammation in
cachexia — is tumor cytokine expression profile the
culprit? Front. Immunol. 6, 629 (2015).
99. Antoun, S., Baracos, V. E., Birdsell, L., Escudier, B. &
Sawyer, M. B. Low body mass index and sarcopenia
associated with dose-limiting toxicity of sorafenib in
patients with renal cell carcinoma. Ann. Oncol. 21,
1594–1598 (2010).
100. Toledo, M. et al. A multifactorial anti-cachectic
approach for cancer cachexia in a rat model
undergoing chemotherapy. J. Cachexia Sarcopenia
Muscle 7, 48–59 (2016).
101. Conte, E. et al. Cisplatin-induced cachexia in rats
causes alterations in skeletal muscle calcium
homeostasis. Biophys. J. 108 (Suppl. 1), 108a (2015).
102. Garcia, J. M., Cata, J. P., Dougherty, P. M. &
Smith, R. G. Ghrelin prevents cisplatin-induced
mechanical hyperalgesia and cachexia. Endocrinology
149, 455–460 (2008).
103. Bruggeman, A. R. et al. Cancer cachexia: beyond
weight loss. J. Oncol. Pract. 12, 1163–1171 (2016).
104. Cvan Trobec, K. et al. Influence of cancer cachexia on
drug liver metabolism and renal elimination in rats.
J. Cachexia Sarcopenia Muscle 6, 45–52 (2015).
105. Bindels, L. B. & Delzenne, N. M. Muscle wasting: the
gut microbiota as a new therapeutic target? Int.
J. Biochem. Cell Biol. 45, 2186–2190 (2013).
106. Klein, G. L., Petschow, B. W., Shaw, A. L. & Weaver, E.
Gut barrier dysfunction and microbial translocation in
cancer cachexia: a new therapeutic target. Curr. Opin.
Support. Palliat. Care 7, 361–367 (2013).
107. Yeh, K. Y. et al. Omega‑3 fatty acid-, micronutrient-,
and probiotic-enriched nutrition helps body weight
stabilization in head and neck cancer cachexia. Oral
Surg. Oral Med. Oral Pathol. Oral Radiol. 116, 41–48
(2013).
108. Varian, B. J. et al. Beneficial bacteria inhibit cachexia.
Oncotarget 7, 11803–11816 (2016).
109. Schieber, A. M. P. et al. Disease tolerance mediated by
microbiome E. coli involves inflammasome and IGF‑1
signaling. Science 350, 558–563 (2015).
110. Mavragani, I. V. et al. Key mechanisms involved in
ionizing radiation-induced systemic effects. A current
review. Toxicol. Res. 5, 12–33 (2016).
111. Azzam, E. I. & Little, J. B. The radiation-induced
bystander effect: evidence and significance. Hum. Exp.
Toxicol. 23, 61–65 (2004).
112. Vacchelli, E. et al. Trial Watch: anticancer
radioimmunotherapy. Oncoimmunology 2, e25595
(2013).
113. Apetoh, L. et al. Toll-like receptor 4‑dependent
contribution of the immune system to anticancer
chemotherapy and radiotherapy. Nat. Med. 13,
1050–1059 (2007).
114. Nikitaki, Z. et al. Systemic mechanisms and effects of
ionizing radiation: a new ‘old’ paradigm of how the
bystanders and distant can become the players.
Semin. Cancer Biol. 37–38, 77–95 (2016).
115. Ermolaeva, M. A. et al. DNA damage in germ cells
induces an innate immune response that triggers
systemic stress resistance. Nature 501, 416–420
(2013).
116. Al‑Mayah, A. et al. The non-targeted effects of
radiation are perpetuated by exosomes. Mutat. Res.
772, 38–45 (2015).
117. Demaria, S. & Formenti, S. C. Radiation as an
immunological adjuvant: current evidence on dose and
fractionation. Front. Oncol. 2, 153 (2012).
118. Demaria, S. et al. Ionizing radiation inhibition of
distant untreated tumors (abscopal effect) is immune
mediated. Int. J. Radiat. Oncol. Biol. Phys. 58,
862–870 (2004).
119. Zitvogel, L., Ayyoub, M., Routy, B. & Kroemer, G.
Microbiome and anticancer immunosurveillance. Cell
165, 276–287 (2016).
120. Deng, L. et al. Irradiation and anti‑PD‑L1 treatment
synergistically promote antitumor immunity in mice.
J. Clin. Invest. 124, 687–695 (2014).
121. Baird, J. R. et al. Radiotherapy combined with novel
STING-targeting oligonucleotides results in regression
of established tumors. Cancer Res. 76, 50–61 (2016).
122. Barker, H. E., Paget, J. T. E., Khan, A. A. &
Harrington, K. J. The tumour microenvironment after
radiotherapy: mechanisms of resistance and
recurrence. Nat. Rev. Cancer 15, 409–425 (2015).
123. Touchefeu, Y. et al. Systematic review: the role of the
gut microbiota in chemotherapy- or radiation-induced
gastrointestinal mucositis — current evidence and
potential clinical applications. Aliment. Pharmacol.
Ther. 40, 409–421 (2014).
124. Vanhoecke, B. W. et al. Low-dose irradiation affects the
functional behavior of oral microbiota in the context of
mucositis. Exp. Biol. Med. (Maywood) 241, 60–70
(2016).
125. Broin, P. Ó. et al. Intestinal microbiota-derived
metabolomic blood plasma markers for prior radiation
injury. Int. J. Radiat. Oncol. Biol. Phys. 91, 360–367
(2015).
126. Wang, A. et al. Gut microbial dysbiosis may predict
diarrhea and fatigue in patients undergoing pelvic
cancer radiotherapy: a pilot study. PLoS ONE 10,
e0126312 (2015).
127. Takemura, N. et al. Blockade of TLR3 protects mice
from lethal radiation-induced gastrointestinal
syndrome. Nat. Commun. 5, 3492 (2014).
128. Vacchelli, E. et al. Trial Watch: Toll-like receptor
agonists for cancer therapy. Oncoimmunology 2,
e25238 (2013).
129. Hu, B. et al. The DNA-sensing AIM2 inflammasome
controls radiation-induced cell death and tissue injury.
Science 354, 765–768 (2016).
130. Ciorba, M. A. et al. Lactobacillus probiotic protects
intestinal epithelium from radiation injury in a TLR‑2/
cyclo-oxygenase‑2‑dependent manner. Gut 61,
829–838 (2012).
131. Jones, R. M. et al. Lactobacilli modulate epithelial
cytoprotection through the Nrf2 pathway. Cell Rep.
12, 1217–1225 (2015).
132. Jones, R. M. et al. Symbiotic lactobacilli stimulate gut
epithelial proliferation via Nox-mediated generation of
reactive oxygen species. EMBO J. 32, 3017–3028
(2013).
133. Delia, P. et al. Use of probiotics for prevention of
radiation-induced diarrhea. World J. Gastroenterol.
13, 912–915 (2007).
134. Sharma, A. et al. Lactobacillus brevis CD2 lozenges
reduce radiation- and chemotherapy-induced
mucositis in patients with head and neck cancer: a
randomized double-blind placebo-controlled study.
Eur. J. Cancer 48, 875–881 (2012).
135. Crawford, P. A. & Gordon, J. I. Microbial regulation of
intestinal radiosensitivity. Proc. Natl Acad. Sci. USA
102, 13254–13259 (2005).
This paper describes the radioresistance of
germ-free mice and characterized the underlying
molecular mechanisms.
136. Santulli, G. Angiopoietin-like proteins: a
comprehensive look. Front. Endocrinol. 5, 4
(2014).
137. Grootaert, C. et al. Bacterial monocultures,
propionate, butyrate and H2O2 modulate the
expression, secretion and structure of the fasting-
induced adipose factor in gut epithelial cell lines.
Environ. Microbiol. 13, 1778–1789 (2011).
138. Jacouton, E. et al. Lactobacillus rhamnosus
CNCMI‑4317 modulates Fiaf/Angptl4 in intestinal
epithelial cells and circulating level in mice. PLoS ONE
10, e0138880 (2015).
139. Korecka, A. et al. ANGPTL4 expression induced by
butyrate and rosiglitazone in human intestinal
epithelial cells utilizes independent pathways. Am.
J. Physiol. Gastrointest. Liver Physiol. 304,
G1025–G1037 (2013).
140. Duncan, A. M., Ronen, A. & Blakey, D. H. Diurnal
variation in the response of gamma-ray-induced
apoptosis in the mouse intestinal epithelium. Cancer
Lett. 21, 163–166 (1983).
141. Ishihara, H. et al. Circadian transitions in radiation
dose-dependent augmentation of mRNA levels for
DNA damage-induced genes elicited by accurate real-
time RT‑PCR quantification. J. Radiat. Res. 51,
265–275 (2010).
142. Ruifrok, A. C., Weil, M. M., Thames, H. D. &
Mason, K. A. Diurnal variations in the expression of
radiation-induced apoptosis. Radiat. Res. 149,
360–365 (1998).
143. Leone, V. et al. Effects of diurnal variation of gut
microbes and high-fat feeding on host circadian clock
function and metabolism. Cell Host Microbe 17,
681–689 (2015).
144. Liang, X., Bushman, F. D. & FitzGerald, G. A.
Rhythmicity of the intestinal microbiota is regulated by
gender and the host circadian clock. Proc. Natl Acad.
Sci. USA 112, 10479–10484 (2015).
145. Mukherji, A., Kobiita, A., Ye, T. & Chambon, P.
Homeostasis in intestinal epithelium is
orchestrated by the circadian clock and microbiota
cues transduced by TLRs. Cell 153, 812–827
(2013).

14 | ADVANCE ONLINE PUBLICATION www.nature.com/nrc

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

146. Nguyen, K. D. et al. Circadian gene Bmal1 regulates
diurnal oscillations of Ly6Chi inflammatory monocytes.
Science 341, 1483–1488 (2013).
147. Thaiss, C. A. et al. Transkingdom control of microbiota
diurnal oscillations promotes metabolic homeostasis.
Cell 159, 514–529 (2014).
148. Maier, I., Berry, D. M. & Schiestl, R. H. Intestinal
microbiota reduces genotoxic endpoints induced by
high-energy protons. Radiat. Res. 181, 45–53
(2014).
149. Holohan, C., Van Schaeybroeck, S., Longley, D. B. &
Johnston, P. G. Cancer drug resistance: an evolving
paradigm. Nat. Rev. Cancer 13, 714–726 (2013).
150. Couzin-Frankel, J. Breakthrough of the year 2013.
Cancer immunotherapy. Science 342, 1432–1433
(2013).
151. Mellman, I., Coukos, G. & Dranoff, G. Cancer
immunotherapy comes of age. Nature 480, 480–489
(2011).
152. Sivan, A. et al. Commensal Bifidobacterium promotes
antitumor immunity and facilitates anti‑PD‑L1 efficacy.
Science 350, 1084–1089 (2015).
This paper demonstrated that the presence of
Bifidobacterium spp. in the gut microbiota
promotes antitumour immunity in mice that is
amplified by anti‑PDL1 therapy.
153. Paulos, C. M. et al. Microbial translocation augments
the function of adoptively transferred self/tumor-
specific CD8+ T cells via TLR4 signaling. J. Clin. Invest.
117, 2197–2204 (2007).
This paper used a therapy model of adoptive T cell
transfer preceded by TBI in mice, to demonstrate
for the first time that the microbiota modulates
anticancer therapy.
154. Dudley, M. E. et al. Adoptive cell therapy for patients
with metastatic melanoma: evaluation of intensive
myeloablative chemoradiation preparative regimens.
J. Clin. Oncol. 26, 5233–5239 (2008).
155. Guiducci, C., Vicari, A. P., Sangaletti, S., Trinchieri, G.
& Colombo, M. P. Redirecting in vivo elicited tumor
infiltrating macrophages and dendritic cells towards
tumor rejection. Cancer Res. 65, 3437–3446 (2005).
156. Vicari, A. P. et al. Reversal of tumor-induced dendritic
cell paralysis by CpG immunostimulatory
oligonucleotide and anti-interleukin 10 receptor
antibody. J. Exp. Med. 196, 541–549 (2002).
157. Stewart, C. A. et al. Interferon-dependent IL‑10
production by Tregs limits tumor Th17 inflammation.
J. Clin. Invest. 123, 4859–4874 (2013).
158. Netea, M. G. et al. Trained immunity: a program of
innate immune memory in health and disease. Science
352, aaf1098 (2016).
159. Rosenberg, S. A., Yang, J. C. & Restifo, N. P. Cancer
immunotherapy: moving beyond current vaccines.
Nat. Med. 10, 909–915 (2004).
160. Eggermont, A. M. Therapeutic vaccines in solid
tumours: can they be harmful? Eur. J. Cancer 45,
2087–2090 (2009).
161. Page, D. B., Postow, M. A., Callahan, M. K.,
Allison, J. P. & Wolchok, J. D. Immune modulation in
cancer with antibodies. Annu. Rev. Med. 65,
185–202 (2014).
162. Sharma, P. & Allison, J. P. The future of immune
checkpoint therapy. Science 348, 56–61 (2015).
163. Teply, B. A. & Lipson, E. J. Identification and
management of toxicities from immune checkpoint-
blocking drugs. Oncology (Williston Park) 28 (Suppl.
3), 30–38 (2014).
164. Dubin, K. et al. Intestinal microbiome analyses identify
melanoma patients at risk for checkpoint-blockade-
induced colitis. Nat. Commun. 7, 10391 (2016).
165. Hand, T. W. et al. Acute gastrointestinal infection
induces long-lived microbiota-specific T cell responses.
Science 337, 1553–1556 (2012).
166. Yang, X. et al. Targeting the tumor microenvironment
with interferon‑β bridges innate and adaptive immune
responses. Cancer Cell 25, 37–48 (2014).
167. Zitvogel, L., Pitt, J. M., Daillere, R., Smyth, M. J. &
Kroemer, G. Mouse models in oncoimmunology. Nat.
Rev. Cancer 16, 759–773 (2016).
168. Ivanov, I. I., Frutos Rde, L., Manel, N., Yoshinaga, K. &
Rifkin, D. B. Specific microbiota direct the
differentiation of IL‑17‑producing T‑helper cells in the
mucosa of the small intestine. Cell Host Microbe 4,
337 (2008).
169. Nowarski, R. et al. Epithelial IL‑18 equilibrium controls
barrier function in colitis. Cell 163, 1444–1456
(2015).
170. Elinav, E. et al. NLRP6 inflammasome regulates
colonic microbial ecology and risk for colitis. Cell 145,
745–757 (2011).
171. Wheeler, M. L. et al. Immunological consequences of
intestinal fungal dysbiosis. Cell Host Microbe 19,
865–873 (2016).
172. Howitt, M. R. et al. Tuft cells, taste-chemosensory
cells, orchestrate parasite type 2 immunity in the gut.
Science 351, 1329–1333 (2016).
173. Kernbauer, E., Ding, Y. & Cadwell, K. An enteric virus
can replace the beneficial function of commensal
bacteria. Nature 516, 94–98 (2014).
174. Cadwell, K. The virome in host health and disease.
Immunity 42, 805–813 (2015).
175. Ramanan, D. et al. Helminth infection promotes
colonization resistance via type 2 immunity. Science
352, 608–612 (2016).
176. Young, G. R. et al. Resurrection of endogenous
retroviruses in antibody-deficient mice. Nature 491,
774–778 (2012).
177. Turnbaugh, P. J. et al. The effect of diet on the human
gut microbiome: a metagenomic analysis in
humanized gnotobiotic mice. Sci. Transl Med. 1,
6ra14 (2009).
178. Baxter, N. T., Zackular, J. P., Chen, G. Y. &
Schloss, P. D. Structure of the gut microbiome
following colonization with human feces determines
colonic tumor burden. Microbiome 2, 20 (2014).
179. Plantinga, T. S. et al. Differential Toll-like receptor
recognition and induction of cytokine profile by
Bifidobacterium breve and Lactobacillus strains of
probiotics. Clin. Vaccine Immunol. 18, 621–628
(2011).
180. Kadowaki, N. et al. Subsets of human dendritic cell
precursors express different toll-like receptors and
respond to different microbial antigens. J. Exp. Med.
194, 863–869 (2001).
181. Pamer, E. G. Resurrecting the intestinal microbiota to
combat antibiotic-resistant pathogens. Science 352,
535–538 (2016).
182. Goldszmid, R. S. & Trinchieri, G. The price of immunity.
Nat. Immunol. 13, 932–938 (2012).
183. Jobin, C. Colorectal cancer: CRC — all about microbial
products and barrier function? Nat. Rev.
Gastroenterol. Hepatol. 9, 694–696 (2012).
184. Rao, V. P. et al. Innate immune inflammatory response
against enteric bacteria Helicobacter hepaticus
induces mammary adenocarcinoma in mice. Cancer
Res. 66, 7395–7400 (2006).
185. Salcedo, R. et al. MyD88‑mediated signaling prevents
development of adenocarcinomas of the colon: role of
interleukin 18. J. Exp. Med. 207, 1625–1636
(2010).
186. Allen, I. C. et al. The NLRP3 inflammasome functions
as a negative regulator of tumorigenesis during colitis-
associated cancer. J. Exp. Med. 207, 1045–1056
(2010).
187. Sears, C. L. & Garrett, W. S. Microbes, microbiota, and
colon cancer. Cell Host Microbe 15, 317–328 (2014).
188. Kostic, A. D. et al. Fusobacterium nucleatum
potentiates intestinal tumorigenesis and modulates
the tumor-immune microenvironment. Cell Host
Microbe 14, 207–215 (2013).
189. Rubinstein, M. R. et al. Fusobacterium nucleatum
promotes colorectal carcinogenesis by modulating
E‑cadherin/beta-catenin signaling via its FadA adhesin.
Cell Host Microbe 14, 195–206 (2013).
190. Gur, C. et al. Binding of the Fap2 protein of
Fusobacterium nucleatum to human inhibitory
receptor TIGIT protects tumors from immune cell
attack. Immunity 42, 344–355 (2015).
191. IARC Working Group on the Evaluation of Carcinogenic
Risks to Humans. Schistosomes, liver flukes and
Helicobacter pylori. Lyon, 7–14 June 1994. IARC
Monogr. Eval. Carcinog. Risks Hum. 61, 1–241
(1994).
192. Poutahidis, T. et al. Pathogenic intestinal bacteria
enhance prostate cancer development via systemic
activation of immune cells in mice. PLoS ONE 8,
e73933 (2013).
193. Fox, J. G. et al. Gut microbes define liver cancer risk in
mice exposed to chemical and viral transgenic
hepatocarcinogens. Gut 59, 88–97 (2010).
194. Yamamoto, M. L. et al. Intestinal bacteria modify
lymphoma incidence and latency by affecting systemic
inflammatory state, oxidative stress, and leukocyte
genotoxicity. Cancer Res. 73, 4222–4232 (2013).
195. Farrell, J. J. et al. Variations of oral microbiota are
associated with pancreatic diseases including
pancreatic cancer. Gut 61, 582–588 (2012).
196. Fan, X. et al. Human oral microbiome and prospective
risk for pancreatic cancer: a population-based nested
case-control study. Gut http://dx.doi.org/10.1136/
gutjnl-2016-312580 (2016).
197. Westbrook, A. M. et al. The role of tumour necrosis
factor-alpha and tumour necrosis factor receptor
signalling in inflammation-associated systemic
genotoxicity. Mutagenesis 27, 77–86 (2012).
198. Gyurkocza, B., Rezvani, A. & Storb, R. F. Allogeneic
hematopoietic cell transplantation: the state of the
art. Expert Rev. Hematol. 3, 285–299 (2010).
199. Taur, Y., Jenq, R. R., Ubeda, C., van den Brink, M. &
Pamer, E. G. Role of intestinal microbiota in
transplantation outcomes. Best Pract. Res. Clin.
Haematol. 28, 155–161 (2015).
200. Taur, Y. et al. The effects of intestinal tract bacterial
diversity on mortality following allogeneic
hematopoietic stem cell transplantation. Blood 124,
1174–1182 (2014).
201. Jenq, R. R. et al. Regulation of intestinal
inflammation by microbiota following allogeneic bone
marrow transplantation. J. Exp. Med. 209, 903–911
(2012).
202. Holler, E. et al. Metagenomic analysis of the stool
microbiome in patients receiving allogeneic stem
cell transplantation: loss of diversity is associated
with use of systemic antibiotics and more
pronounced in gastrointestinal graft-versus-host
disease. Biol. Blood Marrow Transplant. 20,
640–645 (2014).
203. Taur, Y. et al. Intestinal domination and the risk of
bacteremia in patients undergoing allogeneic
hematopoietic stem cell transplantation. Clin. Infect.
Dis. 55, 905–914 (2012).
204. Jenq, R. R. et al. Intestinal blautia is associated with
reduced death from graft-versus-host disease. Biol.
Blood Marrow Transplant. 21, 1373–1383 (2015).
205. Ho, J. T. K., Chan, G. C. F. & Li, J. C. B. Systemic
effects of gut microbiota and its relationship with
disease and modulation. BMC Immunol. 16, 21
(2015).
206. Boleij, A. et al. The Bacteroides fragilis toxin gene is
prevalent in the colon mucosa of colorectal cancer
patients. Clin. Infect. Dis. 60, 208–215 (2015).
207. Koshiol, J. et al. Salmonella enterica serovar Typhi and
gallbladder cancer: a case-control study and meta-
analysis. Cancer Med. 5, 3310–3235 (2016).
208. Lecuit, M. et al. Immunoproliferative small intestinal
disease associated with Campylobacter jejuni. N. Engl.
J. Med. 350, 239–248 (2004).
209. Senff, N. J. et al. European Organization for Research
and Treatment of Cancer and International Society for
Cutaneous Lymphoma consensus recommendations
for the management of cutaneous B‑cell lymphomas.
Blood 112, 1600–1609 (2008).
210. Ferreri, A. J. et al. Chlamydophila psittaci eradication
with doxycycline as first-line targeted therapy for
ocular adnexae lymphoma: final results of an
international phase II trial. J. Clin. Oncol. 30,
2988–2994 (2012).
211. Lakritz, J. R. et al. Gut bacteria require neutrophils to
promote mammary tumorigenesis. Oncotarget 6,
9387–9396 (2015).
212. Rutkowski, M. R. et al. Microbially driven
TLR5‑dependent signaling governs distal malignant
progression through tumor-promoting inflammation.
Cancer Cell 27, 27–40 (2015).
213. Chu, H. & Mazmanian, S. K. Innate immune
recognition of the microbiota promotes host–
microbial symbiosis. Nat. Immunol. 14, 668–675
(2013).
214. McFall-Ngai, M. et al. Animals in a bacterial world, a
new imperative for the life sciences. Proc. Natl Acad.
Sci. USA 110, 3229–3236 (2013).
215. Kremer, N. et al. Initial symbiont contact orchestrates
host-organ-wide transcriptional changes that prime
tissue colonization. Cell Host Microbe 14, 183–194
(2013).
216. Hansen, C. H. et al. Patterns of early gut colonization
shape future immune responses of the host. PLoS
ONE 7, e34043 (2012).
217. Stappenbeck, T. S., Hooper, L. V. & Gordon, J. I.
Developmental regulation of intestinal angiogenesis
by indigenous microbes via Paneth cells. Proc. Natl
Acad. Sci. USA 99, 15451–15455 (2002).
218. Le, Y. et al. Biologically active peptides interacting with
the G protein-coupled formylpeptide receptor. Protein
Pept. Lett. 14, 846–853 (2007).
219. Chen, K. et al. Formylpeptide receptor‑2 contributes
to colonic epithelial homeostasis, inflammation, and
tumorigenesis. J. Clin. Invest. 123, 1694–1704
(2013).
Competing interests statement
The authors declare no competing interests.

NATURE REVIEWS | CANCER ADVANCE ONLINE PUBLICATION | 15

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.