The efcacy and safety of probiotics in people with

cancer: a systematic review

M. G. Redman
1,2
*
, E. J. Ward
3,4
& R. S. Phillips
2,3
1
Hull York Medical School, York;
2
Centre for Reviews and Dissemination, York;
3
Department of Paediatric Oncology and Haematology, Leeds General Inrmary, Leeds;
4
Paediatric Dietetic Department, Leeds General Inrmary, Leeds, UK
Received 18 November 2013; revised 15 February 2014; accepted 20 February 2014
Background: Probiotics are living microorganisms that are generally thought of as being benecial to the recipient. They
have been shown to be effective in people with acute infectious diarrhoea, and cost-effective in antibiotic-associated diar-
rhoea. Probiotics may have a role in people with cancer, as various cancer treatments often lead to diarrhoea. However,
as people with cancer are often immunocompromised, it is important to assess for adverse events (AEs) such as infec-
tion, which could potentially be a consequence of deliberate ingestion of living microorganisms.
Design: A systematic review was carried out to collect, analyse and synthesise all available data on the efcacy and
safety of probiotics in people with cancer (PROSPERO registration: CRD42012003454). Randomised, controlled trials,
identied through screening multiple databases and grey literature, were included for analysing efcacy, while all studies
were included for the analysis of safety of probiotics. Primary outcomes were the reduction in duration, severity and inci-
dence of antibiotic-associated diarrhoea and chemotherapy-associated diarrhoea, and AEs, especially probiotic-asso-
ciated infection. Where possible, data were combined for meta-analysis by a random-effects model, assessing causes of
heterogeneity, including differences in strains, dosage and patient characteristics.
Results: Eleven studies (N= 1557 participants) were included for assessing efcacy. Results show that probiotics may
reduce the severity and frequency of diarrhoea in patients with cancer and may reduce the requirement for anti-diarrhoeal
medication, but more studies are needed to assess the true effect. For example comparing probiotic use to control 25
groups on effect on Common Toxicity Criteria 2 grade diarrhoea, odds ratio (OR) = 0.32 [95% condence interval (CI)
of 0.130.79; P = 0.01]. Seventeen studies (N= 1530) were included in the safety analysis. Five case reports showed
probiotic-related bacteraemia/fungaemia/positive blood cultures.
Conclusion(s): Probiotics may be a rare cause of sepsis. Further evidence needs to be collated to determine whether
probiotics provide a signicant overall benet for people with cancer.
Key words: probiotics, systematic review, cancer, diarrhoea, health care-associated infection
introduction
Probiotics are dened by the World Health Organisation and
Food and Agricultural Organization of the United Nations as:
Live microorganisms which when administered in adequate
amounts confer a health benet on the host [1]. Lactobacillus
and Bidobacterium are commonly used strains, though Sac-
charomyces cerevisiaea yeastis also used as a probiotic [2].
A previous Cochrane Review [3] in immunocompetent
patients demonstrated probiotics reduce episodes and duration
of presumed/proven acute infectious diarrhoea. Diarrhoea
related to cancer therapy incurs additional costs largely due to
more admissions to hospital and time spent there [4]. Probiotics
have been argued to be cost-effective in the context of antibiotic-
associated diarrhoea [5]. Therefore, it is worth considering if
probiotics are effective in people with cancer from both a
patients perspective and a nancial perspective.
Diarrhoea induced by chemotherapy use is the most common
associated toxicity which leads to the chemotherapy regime
being stopped or reduced; one factor contributing to this is that
chemotherapeutic agents can alter the recipients normal pro-
tective gut microora [6]. Diarrhoea is unpleasant for the
patient and may reduce their tolerance for undergoing radio-
therapy and chemotherapy; they may also require further treat-
ment to prevent associated morbidity and mortality [7]. As
infections are common and the gut microora plays a role in
immunity [6], probiotics should be evaluated both for efcacy in
preventing infection and for safety, particularly to investigate
*Correspondence to: M. G. Redman, Hull York Medical School, John Hughlings Jackson
Building, University of York, York YO10 5DD, UK; Tel: +44-870-124-5500; E-mail:
[email protected]
r
e
v
i
e
w
review
Annals of Oncology 00: 111, 2014
doi:10.1093/annonc/mdu106
The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: [email protected].
Annals of Oncology Advance Access published May 17, 2014

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

whether probiotics cause infection themselves. There is current-
ly uncertainty as to the occurrence of adverse events (AEs) after
probiotic consumption. In generally healthy people, no AE of a
serious nature have been reported [3]. However, it is essential to
investigate the safety of probiotic use in immunocompromised
cancer patients, as case reports have identied a Lactobacillus
strain used in probiotic therapy to be involved with sepsis [8].
Current UK dietary advice is for neutropenic cancer patients to
avoid products containing probiotics [9]; however, Gibson et al.
recommend that in patients with pelvic malignancies, consump-
tion of probiotics containing Lactobacillus species may help pre-
vent diarrhoea secondary to chemotherapy or radiotherapy [10].
A systematic review and meta-analyses were carried out to
assess the safety of probiotics in patients with malignancy and to
determine whether probiotics are benecial through assessing
quantitative markers such as grade of diarrhoea.
methods
A protocol was registered on PROSPERO (the international
register of systematic reviews, registration: CRD42012003454) [11].
eligibility
Randomised, controlled trials (RCTs) were considered for asses-
sing the efcacy of probiotics. Both RCT and non-RCTs were
also considered for assessing the safety of probiotics.
Studies were deemed eligible if they:
Included people with a diagnosis of cancer who have received
probiotics.
Reported health outcomes such as antibiotic-associated diar-
rhoea, gastrointestinal infection, mucositis, AEs.
For efcacy assessment, probiotics had to be randomised in
comparison to not receiving probiotics.
The primary outcomes to assess were:
The proportion of people who suffered any AEs, especially
probiotic-associated infection.
The duration, severity and incidence of antibiotic-associated
diarrhoea and chemotherapy-associated diarrhoea.
Secondary outcomes were:
Faecal organic acid concentration.
Faecal bacteriological examination.
NK cell number.
search strategy
Databases and sources searched included: the Cochrane Central
Register of Controlled Trials, Medline, EMBASE, Literatura
Latino-Americana e do Caribe em Cincias da Sade, Allied and
Complementary Medicine (AMED), Database of Abstracts of
Reviews of Effects, American Society of Clinical Oncology,
International Society of Paediatric Oncology, Multinational
Association of Supportive Care in Cancer, International Cancer
Research Portfolio, National Cancer Institute Clinical Trials,
National Cancer Research Institute, Current Controlled Trials
and Centerwatch.
A 40-step search strategy was produced and used for
Medline, EMBASE and AMED (see protocol) without language
limitation. For the other databases, a simpler strategy was used.
The search strategies were run from database inception until
December 2012. Both published and unpublished studies were
included. Grey literature was sought, including on-going clinic-
al trials, conference proceedings and abstracts. Authors and
experts in the eld were contacted to request additional unpub-
lished trials and data, where possible. Reference lists of each
included study were screened, and forward citations searched
using Google Scholar.
selection of studies
For studies found through Medline, EMBASE and AMED,
titles and abstracts were screened by two independent assessors.
Non-English studies were screened by uent medical academics.
For other databases, a second reviewer double-checked a nar-
rowed down list of potential studies and a nal list of studies to
include was agreed upon. Where there was uncertainty about
the relevance of the studies, the full text was obtained to further
evaluate.
data collection
Data about the efcacy of probiotic treatment were extracted
using a tailored form and checked by the second reviewer. The
form included study demographics, trial design, probiotic regi-
mens and outcomes (see supplementary File S1, available at
Annals of Oncology online). A similar form was used for the
safety of probiotic treatment (see supplementary File S2, avail-
able at Annals of Oncology online).
Where data were unclear, the primary author was contacted
requesting further information. Further information was suc-
cessfully obtained regarding three studies [1214].
Each RCT was scrutinised for quality using the Cochrane
Collaborations Risk of bias tool [15] and non-RCTs reviewed
using guidance from Loke et al. [16].
Data were input into RevMan 5.2 [17] software for analysis.
statistical analysis
Where outcome measures were comparable, datasets were
pooled in meta-analyses. I
2
was used to evaluate between-study
heterogeneity. I
2
50% was deemed to represent signicant het-
erogeneity [15, 18] warranting further investigation.
Data were pooled using random-effects models; the Mantel
Haenszel method for analysis of dichotomous data and inverse
variance models for continuous data, as suggested by the
Cochrane Handbook for Systematic Reviews of Interventions [15]
and Systematic Reviews: CRDs Guidance for Undertaking
Reviews in Health Care [18].
Random effects meta-analyses provide condence intervals
for their average estimates of effect, and prediction intervals
(PIs), indicating the potential effect of treatment when it is
applied within an individual study setting, as this may be differ-
ent from the average effect [19]. These were calculated accord-
ing to the method described by Higgins et al. [20] (as cited by
Riley et al. [19]), when datasets contained at least three studies.
Sensitivity analyses were carried out.
i | Redman et al.
review
Annals of Oncology

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

results
included studies
Adapted PRISMA ow diagrams [21] display the process (see
Figure 1A and B) for including studies, resulting in 11 RCTs
assessing the efcacy of probiotics and 17 studies assessing
safety. Further details about the eligible studies are provided in
Tables 1 and 2. Ten ongoing studies were also found (see
Supplementary File S3, available at Annals of Oncology online).
Three studies in Chinese could not be translated due to resource
limitations. Other included studies were all in English.
quality assessment
The individual breakdown of risk of bias for each RCT is dis-
played in Figure 2, while Figure 3 displays risk of bias across all
RCTs. These show that performance and detection bias were
the items that scored the overall highest risk of bias. The quality
assessment of studies for the safety analysis (using the Loke
335 records after de-duplication
437 records identified from Ovid
MEDLINE (R) In-Process & Other
Non-Indexed Citations and Ovid
MEDLINE (R); Ovid EMBASE 1974 to
2012 December 13
319 records excluded 335 records screened
30 full-text articles assessed for eligibility
13 full-text articles excluded
(did not fit inclusion criteria)
3 excluded as unable to
translate from Chinese
2 articles not held anywhere
in UK
1 article had waiting list
11 RCTs included in data synthesis
Other databases,
conference proceedings,
references of potential
includes, forward
referencing and grey
literature screened
5941 records after de-duplication
6751 records identified from Ovid
MEDLINE (R) In-Process & Other
Non-Indexed Citations and Ovid
MEDLINE (R); Ovid EMBASE 1974
to 2012 December 13
5907 records excluded 5941 records screened
41 full-text articles assessed for eligibility 21 full-text articles
excluded as did not fit
inclusion criteria
3 excluded as unable to
translate from Chinese(x2)
and German.
17 studies included in safety analysis
Other databases,
conference proceedings,
references of potential
includes, forward
referencing and grey
literature screened
A
B
Figure 1. (A) An adapted PRISMA ow diagram [21] showing the implementation of the search strategy for the efcacy of probiotics in people with cancer.
(B) An adapted PRISMA ow diagram [21] showing the implementation of the search strategy for the safety of probiotics in people with cancer.
doi:10.1093/annonc/mdu106 |
Annals of Oncology
review

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

method [16]) highlighted that many studies had an unclear
denition of AEs and specically raised concerns about report-
ing bias, given the lack of clarity about how AEs were measured.
efcacy of probiotics
Four RCTs looked at the frequency of grade 3 diarrhoea,
according to the National Cancer Institute Common Toxicity
Criteria (now called the Common Terminology Criteria for
Adverse Events) [39] (CTC) (see Table 3). As displayed in
Figure 4, meta-analysis comparing probiotic to control group
showed an OR of 0.72 with a 95% CI of 0.421.25, a 95% PI of
0.411.27.
Four RCTs looked at the frequency of CTC grade 2 diar-
rhoea. Figure 5 shows the meta-analysis, which, comparing pro-
biotic to control group, results in OR = 0.32 (95% CI 0.130.79;
PI 0.110.97; P = 0.01) suggesting probiotics are benecial in re-
ducing the frequency of CTC grade 2 diarrhoea.
Stool consistency was comparable across three RCTs, as
shown in Figure 6. Urbancsek et al. [28] only clearly dened
formed stools, which they labelled as normal. Giralt et al. [12]
used the Bristol Stool Chart to compare stool consistencies. A
rating of 7 on the Bristol Stool Chart was equated to liquid
stools, whereas a rating of either 5 or 6 was equated to soft/
semi-solid stools. When comparing probiotic groups to control
groups, for liquid stools OR = 0.46 (95% CI 0.045.64; P = 0.55),
whereas for soft/semi-solid stools OR = 1.91 (95% CI 0.18
20.78; P = 0.60). For formed/solid stools, OR = 1.18 (95% CI
0.692.04; P = 0.54).
Two RCTs looked at average daily bowel movements, but
insufcient data were provided for meta-analysis. The mean dif-
ference between probiotics and control in the single study was
9.60 stools per day (95% CI 10.45 to 8.75; P < 0.00001).
The use of anti-diarrhoeal (rescue) medication can be consid-
ered a surrogate marker for severity of diarrhoea. Three studies
evaluated the use of anti-diarrhoeal medication with an
OR = 0.63 (95% CI 0.271.45; PI 0.201.99; P = 0.28) of taking
anti-diarrhoeal medication in the probiotics group (Figure 7).
A secondary outcome measure was faecal bacteriological
comparison. Three RCTs [24, 25, 29] looked at faecal bacterio-
logical components; the change in bacteriological counts were
combined where possible. The evidence was very limited and
uncertain with regards to total anaerobe, bacillus and en-
terococci counts, but showed a signicant mean reduction in
enterobacteriaceae count of 1.98 [log10 colony-forming units
(CFU)/g] of faeces (95% CI 2.56 to 1.39; P < 0.00001)
(Figure 8).
The two remaining secondary outcomes were NK cell
number and faecal organic acid concentration. Both of these
outcomes were only investigated by Wada et al. [29], who did
not nd an increase in the amount of NK cells in the blood of
those consuming probiotics. They found that faecal organic acid
concentrations remained normal until week 5, from which point
pH became constantly <7.0 in those consuming probiotics [29].
safety of probiotics
Seventeen studies were found that met the inclusion criteria.
Three other studies [4042] were unable to be evaluated. The 17
studies found included 1530 people (756 consuming probiotics,
774 not consuming probiotics).
It is unclear how many individuals suffered AEs, as, in some
studies, individual events rather than people suffering from
events were reported. There were 105 AE in those consuming
probiotics, and 145 AE in those not consuming probiotics.
There were no deaths attributed by the authors to probiotic-
associated infection in the probiotic group. However, there was
Table 1. Characteristics of included RCTs for efcacy analysis
Study rst
author
Country of
study
Therapy (RT, CHT,
surgery)
Probiotic administered
Castro [14] Brazil RT Lactobacillus casei Shirota and Bidobacterium breve
Chitapanarux
[22]
Thailand RT Lactobacillus acidophilus, Bidobacterium bidum (Inoran)
Delia [23] Italy RT VSL#3 (Lactobacillus casei, Lactobacillus plantarum, Lactobacillus acidophilus,
Lactobacillus delbrueckii subsp. Bulgaricus, Bidobacterium longum, Bidobacterium
breve, Bidobacterium infantis, Streptococcus salivarius subsp. thermophilus)
Germain [13] Canada RT CHT surgery Bilact (Lactobacillus acidophilus LAC-361 and Bidobacterium longum BB-536)
Gianotti [24] Italy Surgery Lactobacillus johnsonii, Bidobacterium longum (with maltodextrin)
Giralt [12] Spain RT CHT Lactobacillus casei DN-114 001, Streptococcus thermophilus, Lactobacillus delbrueckii
subsp. Bulgaricus
Liu [25] China Surgery Lactobacillus plantarum, Lactobacillus acidophilus, Bidobacterium longum
Osterlund [26] Finland Adjuvant CHT
following surgery
Lactobacillus rhamnosus
Sharma [27] India RT + CHT Lactobacillus brevis
Urbancsek [28] Hungary RT Lactobacillus rhamnosus
Wada [29] Japan CHT Bidobacterium breve strain Yakult (BBG-01)
RT, radiotherapy; CHT, chemotherapy.
| Redman et al.
review
Annals of Oncology

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Table 2. Characteristics of included studies for safety analysis
Study rst
author
Type of study Total people (subgroups) Probiotic administered Summary of potential adverse events (AEs)
Abd El-Atti
[30]
Case report 1 Multispecies 0 AE
Bellette [31] Case report 1 Colotium (ADVITEC)Culture
showed growth of Candida
pelliculosa, Candida krusei,
A. corymbifera and Aspergillus
avus.
Appendicitis and liver abscesses
Cesaro [32] Case report 1 Saccharomyces boulardii Saccharomyces cerevisiae fungaemia
Chitapanarux
[22]
Randomised
control trial
(RCT)
63 (placebo = 31;
probiotics = 32)
Lactobacillus acidophilus and
Bidobacterium bidum
(Inoran)
0 AE
Delia [23] RCT 482 analysed
(placebo = 239;
probiotics = 243)
VSL#3 (multispecies) 0 AE
Giralt [12] RCT 85 (placebo = 41;
probiotics = 44)
Lactobacillus casei DN-114 001,
Streptococcus thermophilus and
Lactobacillus delbrueckii subsp.
bulgaricus
0 AE
Henry [33] Case report 1 Saccharomyces boulardii
(Perenterol)
Saccharomyces cerevisiae found on blood
cultures
LeDoux [34] Case report 1 Lactobacillus acidophilus but not
clear if additional organisms
Persistent Lactobacillus acidophilus bacteraemia
on serial blood cultures for 3 days
Liu [25] RCT 100 analysed
(placebo = 50;
probiotics = 50)
Lactobacillus plantarum,
Lactobacillus and Bido-
bacterium longum
0 AE
Malkov [35] Case series 10 Bacillus oligonitrophilus KU-1 5 potential AE- Sicchasia (patient withdrew),
blood pressure rise 3 (patients probiotics
paused), ICP gain
Mehta [36] Case report 1 Unclear but did contain
Lactobacillus acidophilus
Lactobacillus acidophilus on blood cultures
though not clear to tell if symptomatic
Naito [37] RCT 202 analysed (group
without
probiotics = 102; group
with probiotics = 100)
Lactobacillus casei 126 AE in group without probiotics; 80 AE in
group with probiotics unclear how many
individuals these were distributed over. Wide
range of gastrointestinal and urinary
symptoms - unable to differentiate from
malignancy (transitional cell carcinomas) or
chemotherapy
Oggioni [38] Case report 1 Bacillus subtilis spores
(Enterogermina)
Blood cultures positive for B. subtilis
Osterlund [26] RCT 148 (group without
probiotics = 97, group
with probiotics = 51)
Lactobacillus rhamnosus GG No probiotic = 2 of 51; probiotic = 9 of 97all
cases of neutropenic infection (but no
growth of Lactobacillus in blood cultures)
Sharma [27] RCT 188 analysed
(placebo = 95,
probiotic = 93)
Lactobacillus brevis CD2 Placebo group = (7 grade II dysphagia,
6 grade II nausea and vomiting) + 1 died
after developing grade IV neutropenia and
sepsis; probiotic group = 1 grade II
dysphagia; 1 developed acute myocardial
infarction after 4 weeks of anticancer therapy
- all attributed to chemotherapy by authors
Urbancsek
[28]
RCT 205 (placebo = 103,
probiotic = 102)
Antibiophilus sachets (containing
Lactobacillus rhamnosus)
Placebo = 2 GI problems (mild to moderate),
1 labial oedema; probiotic = 3 GI
problems (mild to moderate)
Wada [29] RCT 40 (placebo = 22;
probiotic = 18)
Bidobacterium breve strain Yakult
(BBG-01)
0 AE
doi:10.1093/annonc/mdu106 | ,
Annals of Oncology
review

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

one death in a person consuming probiotics [38], as discussed
later, which was attributed to progression of malignancy.
There was also one death in the placebo group due to neutro-
penia and sepsis [27]. One patient was withdrawn from probiot-
ic treatment due to sicchasia, while three paused then
recommenced their probiotic treatment due to a rise in blood
pressure [35].
In the report by Cesaro et al. [32], an 8-month-old baby with
acute myeloid leukaemia had been receiving Saccharomyces bou-
lardii capsules (and uconazole prophylaxis), developed a fever
after a completing a course of chemotherapy, and S. cerevisiae
was isolated from the blood culture. The baby was treated with
Amphotericin B until they recovered from neutropenia, had
their occluded central venous catheter removed and no other
cause of infection was found [32]. The authors noted that
routine laboratory methods lead to difculties distinguishing
S. boulardii and S. cerevisiae [32].
Saccharomyces cerevisiae was reportedly found on blood cul-
tures by Henry et al. [33]. A 65-year-old male was treated with
S. boulardii for 2 days, then developed a fever, inammatory
syndrome and neutrophilic leucocytosis [33]. Results from
six consecutive blood cultures showed S. cerevisiae (no other
infection was identied) [33]. The patient was treated with
Amphotericin B and improved [33].
+
+ + +
+ + + + +
+ +
+
+
+
+
+ + + + +
+ +
+
+ + +
+
+
+ + + +
+
+
+
= Low risk of bias
Key
Castro
R
a
n
d
o
m

s
e
q
u
e
n
c
e

g
e
n
e
r
a
t
i
o
n

(
s
e
l
e
c
t
i
o
n

b
i
a
s
)
A
l
l
o
c
a
t
i
o
n

c
o
n
c
e
a
l
m
e
n
t

(
s
e
l
e
c
t
i
o
n

b
i
a
s
)
B
l
i
n
d
i
n
g

o
f

p
a
r
t
i
c
i
p
a
n
t
s

a
n
d

p
e
r
s
o
n
n
e
l

(
p
e
r
f
o
r
m
a
n
c
e

b
i
a
s
)
B
l
i
n
d
i
n
g

o
f

o
u
t
c
o
m
e

a
s
s
e
s
s
m
e
n
t

(
d
e
t
e
c
t
i
o
n

b
i
a
s
)
I
n
c
o
m
p
l
e
t
e

o
u
t
c
o
m
e

d
a
t
a

(
a
t
t
r
i
t
i
o
n

b
i
a
s
)
S
e
l
e
c
t
i
v
e

r
e
p
o
r
t
i
n
g

(
r
e
p
o
r
t
i
n
g

b
i
a
s
)
O
t
h
e
r

b
i
a
s
Chitapanarux
Delia
Germain (high dose)
Gianotti (high dose)
Giralt
Liu
Osterlund
Sharma
Urbancsek
Wada
= high risk of bias
= unclear risk of bias

?
?
?
?
?
?
?
? ?
?
?
?
?
?
?
? ?
? ?
? ?
? ? ?
? ? ?
? ?
?
Figure 2. Risk of bias for each included randomised, controlled trial for efcacy analysis, judged according to Cochrane Risk of bias assessment tool. Data
from [15].
o | Redman et al.
review
Annals of Oncology

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

LeDoux et al. [34] reported Lactobacillus acidophilus bacter-
aemia during 3 days of blood cultures in a 38-year-old male with
AIDS and stage IV Hodgkins disease being treated with probiotics
containing L. acidophilus. He had nished chemotherapy 3 weeks
prior and had had methicillin-susceptible Staphylococcus aureus
and Prevotella loescheii Hickman catheter bacteraemia [34],
treated with ceftriaxone. After 4 days of antibiotics, his Hickman
line was removed, and after 6 days of antibiotics, his cultures from
blood and previous catheter site showed L. acidophilus bacter-
aemia [34]. He was hospitalised on day 10 of antibiotics. By day 3
of hospitalisation, his blood cultures were sterile [34].
Mehta et al. [36] reported a 69-year-old gentleman with
mantle cell lymphoma who had been consuming probiotics for
severe mucositis developed during conditioning before an au-
tologous haematopoietic stem cell transplant. Lactobacillus acid-
ophilus was grown on his blood cultures after the transplant,
and the report describes the resolution of his fever, symptoms
and blood count after the yogurt was stopped [36].
A 73-year-old male with chronic lymphocytic leukaemia
(CLL) was reported by Oggioni et al. [38] as consuming Bacillus
subtilis spores (Enterogermina). Bacillus subtilis was then found
on blood cultures, and remained despite multiple antibiotic
treatment [38]. The patient died within a few days, which was
attributed to CLL with central nervous system involvement,
rather than the B. subtilis positive blood cultures [38].
Bellette et al. [31] reported a 10-year-old girl with an isolated
medullary relapse of acute lymphatic leukaemia, who had been
consuming a probiotic mixture containing Absidia corymbifera
[31]. The girl developed appendicitis followed by sub-hepatic
abscesses, which were found to contain A. corymbifera [31]. She
was treated with Amphotericin B and developed no further
abscesses [31].
The other potential AEs, as shown in Table 2, are of similar
frequencies between groups consuming probiotics and groups
not consuming probiotics.
discussion
This review found 11 RCTs of probiotics in cancer and iden-
tied 17 studies reporting on AEs. The studies were heteroge-
neous in treatments used; strain, dose and duration of probiotic
(s); the patients ages, comorbidities, cancers and therapies
received; and in outcomes assessed, potentially explaining some
of the between-study heterogeneity of the results.
The risk of bias in the efcacy RCTs mainly concerned detec-
tion bias and performance bias. However, the impact of this
may not be substantial given the objective nature of most
outcome measures, such as number of stools per day and use of
anti-diarrhoeal medication.
As a qualitative measurement tool, the Loke method [16] for
quality assessing the safety of probiotics (Supplementary File S4,
available at Annals of Oncology online) highlighted that many
studies were unclear on their denition of an AE, and how they
were measured.
Table 3. National Cancer Institute Common Terminology Criteria
for adverse events. Data from [39]
Toxicity
grade
Criteria
1 Stools: Increase of <4 per day; mild increase
colostomy output
Ostomy output: Mild increase
2 Stools: Increase of 46 per day
Ostomy output: Moderate increase
3 Stools: Increase of 7 or more per day
Ostomy output: Severe increase
Other: loss of continence, hospitalization, limiting
activities of daily living
4 Life-threatening consequences
5 Death
Random sequence generation (selection bias)
Allocation concealment (selection bias)
Blinding of participants and personnel (performance bias)
Blinding of outcome assessment (detection bias)
Incomplete outcome data (attrition bias)
Selective reporting (reporting bias)
Other bias
0% 25% 50% 75% 100%
High risk of bias Unclear risk of bias Low risk of bias
Figure 3. Bar chart comparing percentage risk of bias for each item as judged according to Cochrane Risk of bias assessment tool. Data from [15].
doi:10.1093/annonc/mdu106 | ,
Annals of Oncology
review

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Sensitivity analyses showed no qualitative change in conclu-
sions where meta-analyses were still possible when changes
between studies were assessed. Subgroup analysis could not be
carried out due to the small number of heterogeneous studies.
efcacy
CTC grade 2 and 3 diarrhoea were useful indicators for fre-
quency and severity of diarrhoea. Meta-analysis found that
those in the probiotic group had a signicantly reduced inci-
dence of CTC grade 2 diarrhoea, (OR = 0.32; 95% CI 0.13
0.79; PI 0.110.97; P = 0.01), but was unclear if CTC grade 3
diarrhoea was also reduced (OR = 0.72; 95% CI 0.421.25;
PI 0.411.27; P = 0.24).
Probiotics were also not clearly associated with a decreased
use of rescue (anti-diarrhoeal) medication (OR = 0.63; 95% CI
0.271.45; PI 0.201.99; P = 0.28), but the CI are wide, so rm
conclusions cannot be drawn.
Stool consistency reects the severity/incidence of diarrhoea.
This pattern of results suggests that there may be a shift from
liquid stools to soft/semi-solid stools when trial participants
consumed probiotics. However, the results of these analyses
were all statistically non-signicant, so while point estimates
suggest liquid stools tended to be less common in the probiotic
group (OR = 0.46; 95% CI 0.045.64; P = 0.55), and soft/semi-
solid stools possibly occurred more commonly (OR = 1.91; 95%
CI 0.1820.78; P = 0.60), such assertions are speculative.
In the one study reporting mean number of average daily
bowel movements, it showed a reduction with probiotics of
9.60 stools per day (95% CI 10.45 to 8.75; P < 0000.1).
Given the lack of studies contributing to these data, it would be
unwise to draw rm conclusions based on this result.
The nal quantitative result is with regards to a secondary
outcome of faecal bacteriological count, which was not studied
extensively. Considering the faecal bacteriological composition
is important in understanding a scientic basis for any effects of
probiotics. Alongside faecal organic acid concentration, it can
be used as a surrogate measure to compare changes to the gut
ora, which is an important mucosal barrier to infection. These
are still important aspects to pursue as they may give some indi-
cation into the viability of probiotics and their effectiveness at
altering the guts ora. There is scope for further investigation
into this area.
Study or subgroup
Castro
Chitapanarux
Germain/Demers
Giralt
Total (95% CI)
Total events
Events
7
3
44
44
98
Total
20
32
59
44
155
Events
13
14
74
41
142
Total
20
31
89
41
181
Weight
28.7%
26.6%
44.8%
100.0%
M-H, Random, 95% CI
0.29 [0.08, 1.06]
0.13 [0.03, 0.50]
0.59 [0.27, 1.33]
Not estimable
0.32 [0.13, 0.79]
Probiotic Control Odds ratio Odds ratio
M-H, Random, 95% CI
0.01 0.1 1 10 100
Favours probiotic Favours control
Heterogeneity: t
2
= 0.31; c
2
= 3.82, df = 2 (P = 0.15); I
2
= 48%
Test for overall effect: Z = 2.46 (P = 0.01)
Figure 5. Forest plot of grade 2 diarrhoea (CTC). PI was calculated as 0.110.97.
Study or subgroup
Chitapanarux
Germain/Demers
Giralt
Osterlund
Total (95% CI)
Total events
Heterogeneity: t
2
= 0.08; c
2
= 4.06, df = 3 (P = 0.26); I
2
= 26%
Test for overall effect: Z = 1.17 (P = 0.24)
Events
0
14
20
21
55
Total
32
59
44
97
232
Events
1
28
15
19
63
Total
31
89
41
51
212
Weight
2.8%
34.5%
28.1%
34.6%
100.0%
M-H, Random, 95% CI
0.31 [0.01, 7.98]
0.68 [0.32, 1.43]
1.44 [0.61, 3.45]
0.47 [0.22, 0.98]
0.72 [0.42, 1.25]
Probiotic Control Odds ratio Odds ratio
M-H, Random, 95% CI
0.01 0.1 1 10 100
Favours probiotic Favours control
Figure 4. Forest plot of grade 3 diarrhoea (Common Toxicity Criteria). Prediction interval (PI) was calculated as 0.411.27. M-H, MantelHaenszel meta-
analysis method; random, random-effects model; CI, condence interval, upper and lower ends are displayed within the bracketed region; I
2
represents hetero-
geneity; the P value in overall effect represents the probability of the overall effect being due to chance; the diamond on the plot represents the pooled odds
ratio with its width representing the CIs; the odds ratio of each study, as labelled by its main author, is represented by a square, with a horizontal line demarcat-
ing the 95% CIs.
8 | Redman et al.
review
Annals of Oncology

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

safety
A previous systematic review [3], which included 43 trials on
the use of probiotics in acute infectious diarrhoea, reported no
AEs attributable to probiotics, but one trial reporting a poten-
tially related mild hypersensitivity reaction. However, people
with cancer are more likely to be immunocompromised, so it
Study or subgroup
Chitapanarux
Giralt
Urbancsek
Total (95% CI)
Total events
Events
3
16
36
55
Total
32
44
102
178
Events
10
12
50
72
Total
31
41
103
175
Weight
21.7%
33.6%
44.6%
100.0%
M-H, Random, 95% CI
0.22 [0.05, 0.89]
1.38 [0.56, 3.43]
0.58 [0.33, 1.01]
0.63 [0.27, 1.45]
Probiotic Control Odds ratio Odds ratio
M-H, Random, 95% CI
0.05 0.2 1 5 20
Favours probiotic Favours control
Heterogeneity: t
2
= 0.33; c
2
= 5.13, df = 2 (P = 0.08); I
2
= 61%
Test for overall effect: Z = 1.09 (P = 0.28)
Figure 7. Forest plot for use of anti-diarrhoeal (rescue) medication PI was calculated as 0.201.99.
Study or subgroup
4.2.1 FORMED/SOLID
Chitapanarux
Urbancsek
Subtotal (95% CI)
Total events
4.2.2 SOFT/SEMI-SOLID
Chitapanarux
Giralt
Subtotal (95% CI)
Total events
4.2.3 LIQUID
Chitapanarux
Giralt
Subtotal (95% CI)
Total events
Total (95% CI)
Total events
Events
1
59
60
25
6
31
6
43
49
140
Total
32
102
134
32
50
82
32
50
82
298
Events
0
56
56
11
12
23
20
49
69
148
Total
31
103
134
31
62
93
31
62
93
320
Weight
6.7%
21.4%
28.0%
17.8%
18.2%
35.9%
17.5%
18.5%
36.0%
100.0%
M-H, Random, 95% CI
3.00 [0.12, 76.49]
1.15 [0.66, 2.00]
1.18 [0.69, 2.04]
6.49 [2.13, 19.81]
0.57 [0.20, 1.64]
1.91 [0.18, 20.78]
0.13 [0.04, 0.40]
1.63 [0.60, 4.46]
0.46 [0.04, 5.64]
1.06 [0.39, 2.88]
Probiotic Control Odds ratio Odds ratio
M-H, Random, 95% CI
0.01 0.1 1 10 100
Favours probiotic Favours control
Heterogeneity: t
2
= 0.00; c
2
= 0.33, df = 1 (P = 0.57); I
2
= 0%
Test for overall effect: Z = 0.61 (P = 0.54)
Heterogeneity: t
2
= 2.66; c
2
= 9.63, df = 1 (P = 0.002); I
2
= 90%
Test for overall effect: Z = 0.53 (P = 0.60)
Heterogeneity: t
2
= 2.95; c
2
= 10.69, df = 1 (P = 0.001); I
2
= 91%
Test for overall effect: Z = 0.60 (P = 0.55)
Heterogeneity: t
2
= 1.12; c
2
= 25.68, df = 5 (P = 0.0001); I
2
= 81%
Test for subgroup differences: c
2
= 0.69, df = 2 (P = 0.71); I
2
= 0%
Test for overall effect: Z = 0.12 (P = 0.90)
Figure 6. Forest plot of stool consistency.
doi:10.1093/annonc/mdu106 |
Annals of Oncology
review

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

may be that probiotic-associated infections are more likely in
this group.
Current dietary advice for neutropenic cancer patients is to
avoid products containing probiotics, which is based upon bac-
teraemia case reports and manufacturers recommendations [9],
rather than robust scientic evidence.
The 17 studies assessed in our review included 1530 people
(756 people consuming probiotics, 774 people not consuming
probiotics). There were 105 AE in those consuming probiotics,
and 145 AE in those not consuming probiotics. A wide range of
AEs were noted, including: bacteraemia/fungaemia, infection,
gastrointestinal symptoms (diarrhoea, constipation, dysphagia,
nausea and vomiting), urinary symptoms (only present in patients
with transitional cell carcinomas), sicchasia, raised blood pressure
and raised intra-cranial pressure. The heterogeneity of the different
treatment regimes and malignancies mean we cannot judge which
AEs are related to probiotic consumption. A gentleman with CLL
and persistent B. subtilis on blood cultures died, though his death
was attributed to CNS malignant involvement [38]. The concerns
around bacteraemia/fungaemia/blood culture growth are signi-
cant; although only noted in ve case reports [3234, 36, 38] of
the 756 cases described consuming probiotics, this risk needs to be
considered alongside any potential benet.
Streptococcus lactis septicaemia has also been diagnosed in a
person with CLL who consumed a non-probiotic yogurt drink
[43]. Also, S. cerevisiae fungaemia was found in a 48-year-old
cancer patient after bone marrow transplantation [44] with no
known record of probiotic consumption. Therefore, similar
organisms may lead to bacteraemia/fungaemia in patients not
known to be consuming probiotics.
further work
The search strategy for the review was robust and broad and
included grey literature.
As previously mentioned, 10 ongoing trials were found
(Supplementary File 3, available at Annals of Oncology online),
which should be incorporated into future iterations of this review.
No rm conclusions can be drawn from the review currently, but
as further studies are completed and become available, they could
be incorporated and may give more clinically convincing results.
A highly relevant unanswered question is if probiotic use could
reduce rates of Clostridium difcile infection in people with cancer.
conclusion
This systematic review demonstrates that there is currently
insufcient evidence to claim that probiotics are effective and
safe in people with cancer. Meta-analyses found that probiotics
signicantly reduced the incidence of CTC grade 2 diarrhoea,
may reduce the incidence of CTC grade 3 diarrhoea, may
reduce the average frequency of daily bowel movements and
may reduce the need for anti-diarrhoeal medication, but most of
the evidence is not clinically convincing, and they may be a rare
cause of sepsis. An effect on faecal bacteriological composition
may be found, but this needs to be examined in further trials.
Further studies, which are ongoing, need to be evaluated before
there can be sufcient condence regarding these outcomes.
acknowledgements
We are grateful for additional clarication and data from the fol-
lowing individuals: R. Cairoli, Haematology Specialist, Ospedale
Valduce, Como, Italy. M. Castro, Clinical Nutrition, University
Hospital of Santo Andre, Santo Andre, Brazil. M. Demers,
Clinical Lecturer and Clinical Nurtritionist, Laval University
and Htel Dieu de Qubec, Canada. J.-M. Durand, Internal
Medicine, Hpital de la Conception, Marseille, France. J. Giralt,
Radiation Oncology, Vall dHebron University Hospital,
Barcelona, Spain. M. Mego, Head of Translational Research
Unit, National Cancer Institute, Bratislava, Slovak Republic.
A. Mehta, Hematology and Oncology Fellow, University of
Alabama at Birmingham, Alabama, USA.
disclosure
The authors have declared no conicts of interests.
references
1. Guidelines for the evaluation of probiotics in food [Internet]. London, Ontario: FAO &
WHO, 2002. http://www.who.int/foodsafety/fs_management/en/probiotic_
guidelines.pdf (18 March 2014, date last accessed).
2. Guarner F, Khan AG, Garisch J et al. World Gastroenterology Organisation Global
Guidelines: Probiotics and Prebiotics. 2011. http://www.worldgastroenterology.org/
assets/export/userles/Probiotics_FINAL_20110116.pdf (18 March 2014, date
last accessed).
3. Allen SJ, Martinez EG, Gregorio GV et al. Probiotics for treating acute infectious
diarrhoea. Cochrane Database Syst Rev 2010; (11): CD003048.
4. Carlotto A, Hogsett VL, Maiorini EM et al. The economic burden of toxicities
associated with cancer treatment: review of the literature and analysis of nausea
and vomiting, diarrhoea, oral mucositis and fatigue. Pharmacoeconomics 2013;
31(9): 753766.
5. Hickson M, DSouza AL, Muthu N et al. Use of probiotic Lactobacillus preparation
to prevent diarrhoea associated with antibiotics: randomised double blind placebo
controlled trial. BMJ 2007; 335: 80.
Study or subgroup
Liu
Wada
Total (95% CI)
Mean
1.2
0.4
SD
1.63
3.75
Total
50
8
58
Mean
0.8
1
SD
1.41
1.44
Total
50
5
55
Weight
95.9%
4.1%
100.0%
IV, Fixed, 95% CI
2.00 [2.60, 1.40]
1.40 [4.29, 1.49]
1.98 [2.56, 1.39]
Probiotic Control Mean difference Mean difference
IV, Fixed, 95% CI
4 2 0 2 4
Favours probiotic Favours control
Heterogeneity: c
2
= 016, df = 1 (P = 0.69); I
2
= 0%
Test for overall effect: Z = 6.62 (P < 0.00001)
Figure 8. Forest plot for change in enterobacteriaceae count (log10 CFU/g of faeces).
1o | Redman et al.
review
Annals of Oncology

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

6. Miller AC, Elamin EM. Use of probiotics for treatment of chemotherapy-
induced diarrhea: is it a myth? JPEN J Parenter Enteral Nutr 2009; 33(5):
573574.
7. Arnold RJG, Gabrail N, Raut M et al. Clinical implications of chemotherapy-induced
diarrhea in patients with cancer. J Support Oncol 2005; 3: 227232.
8. Land MH, Rouster-Stevens K, Woods CR et al. Lactobacillus sepsis associated
with probiotic therapy. Pediatrics 2005; 115(1): 178181.
9. Beckerson J, Jones N, Lodhia S et al. Dietary advice during neutropenia. An
update and consensus from the Haematology subgroup of the BDA Oncology
Group. Complete Nutr 2012; 12(6): 4042.
10. Gibson RJ, Keefe DMK, Lalla RV et al. Systematic review of agents for the
management of gastrointestinal mucositis in cancer patients. Support Care Cancer
2013; 21(1): 313326.
11. Redman M, Phillips B, Ward E. The use of probiotics in people with cancer.
PROSPERO 2012; CRD42012003454. http://www.crd.york.ac.uk/PROSPERO/
display_record.asp?ID=CRD42012003454.
12. Giralt J, Regadera JP, Verges R et al. Effects of probiotic Lactobacillus casei
DN-114 001 in prevention of radiation-induced diarrhea: results from
multicenter, randomized, placebo-controlled nutritional trial. Int J Radiat Oncol Biol
Phys 2008; 71(4): 12131219 [unpublished data provided by the main author
also included].
13. Germain I, Desjardins J, Demers M et al. Phase III study: impact of probiotics on
diarrhea in patients treated with pelvic radiation. Int J Radiat Oncol Biol Phys
2011; 81(2 Suppl): S667S668.
14. Castro MG, Sanchez PX, Glasberg J et al. Effects of probiotic in prevention of
radiation-induced diarrhea. Clin Nutr Suppl 2009; 4: 7273.
15. Higgins J, Green S (eds). Cochrane Handbook for Systematic Reviews of
Interventions. Version 5.1.0 [updated March 2011]. The Cochrane Collaboration
2011. www.cochrane-handbook.org (18 March 2014, date last accessed).
16. Loke Y, Price D, Herxheimer A, The Cochrane Adverse Effects Methods Group.
Systematic reviews of adverse effects: framework for a structured approach. BMC
Med Res Methodol 2007; 7(1): 32.
17. Review Manager (RevMan) [Computer Program]. Version 5.2. Copenhagen: The
Nordic Cochrane Centre, The Cochrane Collaboration, 2012.
18. University of York Centre for Reviews and Dissemination. Systematic Reviews:
CRDs Guidance for Undertaking Reviews in Health Care. York: CRD, University of
York, 2009.
19. Riley RD, Higgins JPT, Deeks JJ. Interpretation of random effects meta-analyses.
BMJ 2011; 342: d549.
20. Higgins JPT, Thompson SG, Spiegelhalter DJ. A re-evaluation of random-effects
meta-analysis. J R Stat Soc Ser A Stat Soc 2009; 172(1): 137159.
21. Moher D, Liberati A, Tetzlaff J et al., The PRISMA Group. Preferred reporting items
for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med
2009; 6(6): e1000097.
22. Chitapanarux I, Chitapanarux T, Traisathit P et al. Randomized controlled trial
of live lactobacillus acidophilus plus bidobacterium bidum in prophylaxis
of diarrhea during radiotherapy in cervical cancer patients. Radiat Oncol
2010; 5: 31.
23. Delia P, Sansotta G, Donato V et al. Use of probiotics for prevention of radiation-
induced diarrhea. World J Gastroenterol 2007; 13(6): 912915.
24. Gianotti L, Morelli L, Galbiati F et al. A randomized double-blind trial on
perioperative administration of probiotics in colorectal cancer patients. World J
Gastroenterol 2010; 16(2): 167175.
25. Liu Z, Qin H, Xia Y et al. Randomised clinical trial: the effects of perioperative
probiotic treatment on barrier function and post-operative infectious complications
in colorectal cancer surgerya double-blind study. Aliment Pharmacol Ther
2011; 33(1): 5063.
26. Osterlund P, Ruotsalainen T, Korpela R et al. Lactobacillus supplementation for
diarrhoea related to chemotherapy of colorectal cancer: a randomised study. Br J
Cancer 2007; 97(8): 10281034.
27. Sharma A, Rath GK, Chaudhary SP 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 2012;
48(6): 875881.
28. Urbancsek H, Kazar T, Mezes I et al. Results of a double-blind, randomized study
to evaluate the efcacy and safety of Antibiophilus in patients with radiation-
induced diarrhoea. Eur J Gastroenterol Hepatol 2001; 13(4): 391396.
29. Wada M, Nagata S, Saito M et al. Effects of the enteral administration of
Bidobacterium breve on patients undergoing chemotherapy for pediatric
malignancies. Support Care Cancer 2010; 18(6): 751759.
30. Abd El-Atti S, Wasicek K, Mark S et al. Use of probiotics in the management of
chemotherapy-induced diarrhea: a case study. JPEN J Parenter Enteral Nutr 2009;
33(5): 569570.
31. Bellette B, Rabrin H, Berger C et al. Molecular conrmation of an absidiomycosis
following treatment with a probiotic supplement in a child with leukemia. J Mycol
Med 2006; 16: 7276.
32. Cesaro S, Chinello P, Rossi L et al. Saccharomyces cerevisiae fungemia in a
neutropenic patient treated with Saccharomyces boulardii. Support Care Cancer
2000; 8(6): 504505.
33. Henry S, DHondt L, Andr M et al. Saccharomyces cerevisiae fungemia in a head
and neck cancer patient: a case report and review of the literature. Acta Clin Belg
2004; 59(4): 220222.
34. LeDoux D, LaBombardi VJ, Karter D. Lactobacillus acidophilus bacteraemia after
use of a probiotic in a patient with AIDS and Hodgkins disease. Int J STD AIDS
2006; 17: 280282.
35. Malkov S, Markelov V, Polozov G et al. Signicant delay of lethal outcome in cancer
patients due to peroral administration of Bacillus oligonitrophilus KU-1.
ScienticWorldJournal 2006; 6: 21772187.
36. Mehta A, Rangarajan S, Borate U. A cautionary tale for probiotic use in
hematopoietic SCT patientsLactobacillus acidophilus sepsis in a patient with
mantle cell lymphoma undergoing hematopoietic SCT. Bone Marrow Transplant
2012; 48: 461462.
37. Naito S, Koga H, Yamaguchi A et al. Prevention of recurrence with epirubicin and
lactobacillus casei after transurethral resection of bladder cancer. J Urol 2008;
179: 485490.
38. Oggioni MR, Pozzi G, Valensin PE et al. Recurrent septicemia in an
immunocompromised patient due to probiotic strains of Bacillus subtilis. J Clin
Microbiol 1998; 36(1): 325326.
39. National Cancer Institute: Common Terminology Criteria for Adverse Events,
Version 4.0. 2010. http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-
14_QuickReference_8.5x11.pdf (15 April 2014, date last accessed).
40. De Vrese M. Microbiology, effects, and safety of probiotics. Monatsschr
Kinderheilkd 2008; 156: 10631069.
41. Xia Y, Yang Z, Chen HQ et al. Effect of bowel preparation with probiotics on
intestinal barrier after surgery for colorectal cancer. Zhonghua Wei Chang Wai Ke
Za Zhi 2010; 13(7): 528531.
42. Zhang JW, Du P, Gao J et al. Preoperative probiotics decrease postoperative
infectious complications of colorectal cancer. Am J Med Sci 2012; 343(3):
199205.
43. Durand J, Rousseau M, Gandois J et al. Streptococcus lactis septicemia in a
patient with chronic lymphocytic leukemia. Am J Hematol 1995; 50(1): 6465.
44. Cairoli R, Marenco P, Perego R et al. Saccharomyces cerevisiae fungemia with
granulomas in the bone marrow in a patient undergoing BMT. Bone Marrow
Transplant 1995; 15: 785786.
doi:10.1093/annonc/mdu106 | 11
Annals of Oncology
review

a
t

U
n
i
v
e
r
s
i
d
a
d
e

F
e
d
e
r
a
l

d
e

S
a
n
t
a

C
a
t
a
r
i
n
a

o
n

J
u
n
e

2
8
,

2
0
1
4
h
t
t
p
:
/
/
a
n
n
o
n
c
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m