The Use of Ketogenic Diets in Cancer Patients: A Systematic Review

Clinical and Experimental Medicine (2021) 21:501–536
https://doi.org/10.1007/s10238-021-00710-2
REVIEW ARTICLE
The use of ketogenic diets in cancer patients: a systematic review

Maximilian Römer1 · Jennifer Dörfler1 · Jutta Huebner1
Received: 9 February 2021 / Accepted: 26 March 2021 / Published online: 3 April 2021
© The Author(s) 2021
Abstract
Ketogenic diets are a widely known, yet controversial treatment for cancer patients. In this review, we summarize the clini-
cal evidence for anti-tumor effects, as well as the effects on anthropometry, quality of life, adverse events and adherence in
cancer patients. In April 2019, a systematic search was conducted searching five electronic databases (EMBASE, Cochrane,
PsychInfo, CINAHL and Medline) to find studies analyzing the use, effectiveness and potential harm of a ketogenic diet in
cancer patients of any age as sole or complementary therapy. From all 19.211 search results, 46 publications concerning
39 studies with 770 patients were included in this systematic review. The therapy concepts included all forms of diets with
reduced carbohydrate intake, that aimed to achieve ketosis for patients with different types of cancer. Most studies had a low
quality, high risk of bias and were highly heterogeneous. There was no conclusive evidence for anti-tumor effects or improved
OS. The majority of patients had significant weight loss and mild to moderate side effects. Adherence to the diet was rather
low in most studies. Due to the very heterogeneous results and methodological limitations of the included studies, clinical
evidence for the effectiveness of ketogenic diets in cancer patients is still lacking.
Keywords Humans · Metabolism · Ketogenic diet · Low-carbohydrate diet · Carbohydrate-restricted · Cancer
Abbreviations Introduction
KD Ketogenic diet
RCT Randomized controlled trial Current cancer treatment is largely based on surgery, radia-
CT Controlled trial tion and chemotherapy. Despite the advances in these fields
PSADT Prostate-specific antigen doubling time and the implementation of targeted therapies and immune
PFS Progression-free survival checkpoint inhibitors, many cancer patients still suffer from
PCS Physical component summary a poor prognosis and search for alternative or complemen-
MCS Mental component summary tary treatments. Since there is a growing recognition of the
SF-12 Short Form (12) Health Survey impact of dietary interventions on human health [1], many
QoL Quality of life cancer patients try to optimize their diet to improve their
HNC Head and neck cancer prognosis and reduce treatment-associated side effects [2].
FM Fat mass For these patients and professionals alike, the ketogenic
FFM Fat-free mass diet (KD) is compelling due to its success in treating epi-
AE Adverse events lepsy [3] and its theoretical foundation. The proposed anti-
SoC Standard of care tumor effect relies on Warburg’s observation, that cancer
DLT Dose-limiting toxicity cells prefer anaerobic glycolysis, even in the presence of
CTCAE NCI Common Terminology Criteria for oxygen [4]. Further, cancer cells use glycolysis for rapid
Adverse Events cell proliferation [5] and the formation of metastases [6].
OS Overall survival Hence, KDs, which are high in fat and low in carbohydrates
[7], try to reduce the amount of glucose in the body, that
the cancer cells can utilize [8, 9]. The exact ratio of macro-
* Maximilian Römer nutrients differs between the specific variations of this diet
maximilian.roemer@uni-jena.de [10]. Probably, the most renown adaption of this diet is a
1
Klinik Für Innere Medizin II, Hämatologie Und
4:1 fat-to-carbohydrate + protein ratio [7]. Such an approach
Internistische Onkologie, Universitätsklinikum Jena, Am was used successfully in cellular and animal studies [11,
Klinikum 1, 07747 Jena, Germany
13
Vol.:(0123456789)
502 Clinical and Experimental Medicine (2021) 21:501–536
12]. Nonetheless, there were also contradicting studies that tumor progression of prostate cancer). Language restrictions
showed that there are cancer cell lines, which can utilize were made to English and German.
fatty acids and ketone bodies [13–16].
Our aim in this review was to systematically assess Study selection
whether the results from in vitro studies translated to clini-
cal evidence of anti-tumor efficiency and further analyze the A systematic research was conducted using five databases
impact that a KD has on the quality of life and anthropom- (Medline (Ovid), CINAHL (EBSCO), EMBASE (Ovid),
etry of the patients. Cochrane CENTRAL and PsycINFO (EBSCO)) in April
2019. For each of these databases, a complex search strategy
was developed, consisting of a combination of MeshTerms,
Method keywords and text words in different spellings connected to
cancer and ketogenic diets. The detailed search string is pro-
Criteria for including and excluding studies vided in online resource 1. The search string was highly sen-
in the review sitive, since it was largely unrestricted by filters for study or
publication type. After importing the search results into End-
Inclusion and exclusion criteria are listed in Table 1 based Note X9, all duplicates were removed and a title–abstract
on a PICO model. According to the recommendations of screening was carried out by two independent reviewers
the Cochrane Effective Practice and Organization of Care (MR, JD). In case of disagreement, consensus was made
(EPOC) systematic reviews, review and meta-analyses, ran- by discussion. After that, all full texts were retrieved and
domized controlled studies (RCT), non-randomized con- screened again independently by both reviewers. When title
trolled studies (CT), uncontrolled studies (process moni- and abstract did not have sufficient information for screening
toring, uncontrolled before–after studies and time series purposes, a full-text copy was retrieved as well. Addition-
analyses) and observational studies were included [17]. We ally, bibliography lists of all retrieved articles were manually
additionally included case series and case studies, due to the screened for relevant studies. Such studies were included if
low number of publications on this topic. Criteria for reject- they provided a comprehensive description of the study. The
ing studies were primary prevention, gray literature, other study flow during this process is presented in Fig. 1.
publication types than primary investigation/report (e.g.,
comments, letters, abstracts) or precancerous conditions if Assessment of risk of bias and methodological
the results of the patients with cancer were not reported sep- quality
arately. Additionally, studies were excluded if they reported
no patient-centered outcomes (laboratory parameters, except All characteristics were assessed by two independent review-
PSA which was considered as a surrogate parameter for ers (MR, JD). In case of disagreement, a third reviewer was
consulted (JH) and consensus was made by discussion.
Table 1 Inclusion and exclusion criteria based on a PICO model

PICO Inclusion criteria Exclusion criteria
Patient Cancer patients (all entities and stages) Patients with precancerous conditions or carcinoma in situ
Primary prevention
Preclinical studies
Intervention Every intervention based on a ketogenic diet
No restrictions regarding the type of KD, dose, mode of application
KD applied as sole or supplementary treatment
Comparison All possible control groups (active control, placebo, standard/guide-
line/usual care)
Outcome Mortality (overall survival)
Morbidity (progression-/disease-free interval, tumor response)
Patient-reported outcomes (i.e., quality of life or other important
psychological outcomes like psychological well-being, fatigue, as
well as physical and mental adverse effects)
Weight and body composition
Toxicity and adverse events (CTCAE)
Others Language: German and English Gray literature (conference articles, abstracts, letters, ongo-
Full publication ing studies, unpublished literature, etc.)
Full text not available in German or English
13
Clinical and Experimental Medicine (2021) 21:501–536 503
Fig. 1 Preferred reporting

items for systematic reviews
Records identified through Additional records identified
Identification
and meta-analyses flow chart
database searching through other sources
displaying the study selection
(n = 19.205) (n = 6 )
process
Records after duplicates removed

(n = 16.373 )
Screening
Records screened Records excluded
(n = 16.373 ) (n = 16.325 )
Full-text articles assessed Full-text articles excluded,

for eligibility with reasons
Eligibility
(n = 48) (n =3 )
Studies included in
qualitative synthesis
(n =45)
Included
Studies included in
quantitative synthesis
(meta-analysis)
(n =0)
Risk of bias and methodological quality Data extraction
The risk of bias of the included RCTs and CTs was analyzed Data extraction was performed by one reviewer (MR) and
with the SIGN-Checklist [18] for controlled trials version controlled by two independent reviewers (JD, JH). As a
2.0. The AMSTAR-2 instrument for systematic reviews template for data extraction, the evidence tables from the
was used for reviews and systematic reviews. Other study national Guideline on Complementary and Alternative
types were analyzed based on the Cochrane Risk of Bias Medicine in Oncological Patients of the German Guideline
tool [19]. Further, these studies were rated with the Oxford Program in Oncology (https://www.leitlinienprogramm-
criteria. Additional criteria concerning methodology were onkologie.de/english-language/) were used. Due to a large
size of population, application of power analysis, adequacy overlap in studies included in the systematic reviews, only
of statistical tests (e.g., control of premises or multiple test- the data from the individual studies, which were included
ing) and selective outcome reporting (report of all assessed in the systematic reviews, were extracted. Extraction was
outcomes with specification of statistical data as the p-value) limited to data from primary literature and other sources,
as well as possible conflicts of interest. which provided a comprehensive description of the study,
meeting the inclusion criteria.
Data synthesis
No studies were suitable for a pooled analysis; hence, only

a narrative analysis can be presented here.
13
Results based on the Oxford criteria. These results and additional

comments on methodology are provided in Table 4.
The systematic search revealed 19.205 results. Six stud-
ies were added by hand search. At first, duplicates were Efficacy of the ketogenic diet
removed leaving 16.373 studies. After screening title and
abstract, 48 studies remained to complete review. Finally, 45 The study characteristics and all relevant results reported in
publications were analyzed in this review, including 5 SR, the included RCTs and CTs are presented in Table 5. Similar
1 review and additionally 5 publications on 3 RCTs, 2 con- information concerning the included single-arm studies and
trolled studies and 33 single-arm studies and case reports, case reports is presented in Table 6.
presented in 32 publications. All in all, 765 patients were
described in 39 publications. Survival and disease progression
Characteristics of included studies Results from RCTs and CTs
Concerning the RCTs and CTs, 322 patients were included Overall survival was only analyzed in one RCT [23]. In this
and 250 of them were analyzed, due to 72 drop-outs. The study, the overall survival (OS) for a subgroup of patients
mean age of patients (only reported in 3 studies) ranged from with neoadjuvant treatment for breast cancer was signifi-
44.8 to 66.3 years and the range of age from 38 to 76 years cantly higher in the intervention group (p = 0.04). However,
(reported in 1 study). One publication only reported the no data for the entire study population are presented, which
median age, which was 72 years. A total of 156 (62%) par- also consisted of patients with metastatic disease.
ticipants were female and 94 (38%) were male. Concerning One RCT assessed the effects of the diet on prostate-
the studies with a fixed duration of intervention, the extent specific antigen doubling time (PSADT) as a surrogate
of the diet ranged from 3 to 6 months. parameter for progression of disease [24]. Per protocol,
In the single-arm studies and case series, a total of 443 there was no between-group difference concerning the
patients were included and analyzed. The age of the included PSADT (p = 0.446). Only in post hoc exploratory analysis
patients ranged from 3 to 92 years. Information about the with adjusting for multiple baseline covariates and proposed
gender of the included patients could be obtained for 370 of hemoconcentration, a significantly increased PSADT could
the 443 patients. Out of these 370 participants, 184 (50%) be found.
were female and 186 (50%) were male. Duration of the die-
tary intervention in these studies reached from a single, 3 h Results from single‑arm studies and case reports
long, application of parenteral nutrition [20] to a single case
with more than 31 months of oral KD [21]. Only five of these studies compared reported and expected
The KDs prescribed in the included studies varied exten- survival, which was derived from historical controls [25–29].
sively between studies (see Table 6 for details concerning In one study [25], two of the patients were analyzed and their
the prescribed diets) were in most articles not described in survival was comparable with the expected survival, simi-
adequate detail, and the majority did not utilize standardized lar to another study where all of the different subgroups of
dietary protocols. Furthermore, the methods used for assess- patients had an OS in line with the historical controls [28].
ing ketone body levels and diet compliance varied widely, Two other studies [26, 27] found a numerically better than
with some studies not measuring them at all. expected survival. However, no statistical analysis was per-
formed. One study, however, reported a lower-than-expected
Excluded studies survival for the patients receiving a KD [29].
Another study compared the subgroup of patients, who
A list of the studies excluded after full-text screening and received bevacizumab salvage treatment while on a KD with
the reasons for exclusion are presented in online resource 2. other patients treated with bevacizumab in the same hospi-
tal, who did not receive a KD. There was no difference in
Risk of bias in included studies median progression-free survival (PFS) (p = 0.38) [30].
Even though most studies reported on tumor stability
The methodical quality of the included RCTs and CTs was and progression, the results were highly heterogeneous and
rated according to the SIGN checklists [18], and the results the tools and methods used for this assessment were only
are presented in Table 2. Other study types were analyzed reported in a minority of them in adequate detail. Further-
based on the Cochrane Risk of Bias tool [22], with the more, there was no analysis for statistical significance of
results presented in Table 3. These studies were further rated the findings.
13
Table 2 Risk of bias in the included RCTs and CTs according to the SIGN checklist
Reference Study type Standardized rating of risk of bias Additional comments on methodology Evidence
level
(Oxford)
Freedland et al. [24] RCT Rating according to SIGN PRO: in accordance with the ethical guidelines of 1b–
Positive: 4 points the US Common Rule; randomization stratified by
Uncertain: 2 points center and BMI; compliance surveilled in arm A with
Negative: 3 points weekly urine ketone measurement; comprehensive
Overall quality: acceptable and adequate analysis including the most important
factors; power analysis
Contra: small sample size; no possibility for separation
of the effects of weight loss and carbohydrate deficit;
no information about approval by the ethics commit-
tee; no intention to treat analysis
Khodabakhshi et al. [23] RCT Rating according to SIGN PRO: study protocol approved by responsible research 2b–
Positive: 3 points institute; power analysis; groups are comparable
Uncertain: 3 points Contra: small sample size, especially concerning the
Negative: 2 points subgroup of neoadjuvant-treated patients; no survival
Overall quality: acceptable analysis for the whole study population, only for
subgroup of neoadjuvant-treated patients; unclear if
intention to tread analysis was actually performed,
since only information about patients that completed
the study is given; duration of follow-up for survival
analysis is longer than timeframe from start of patient
enrollment to submission of the article
Cohen et al. [32, 42, 56] RCT Rating according to SIGN PRO: compliance surveilled in arm A with weekly urine 2b
Positive: 4 points ketone measurement; inclusion of the demographic
Uncertain: 2 points characteristics as covariates during analysis; power
Negative: 3 points analysis; groups are comparable; study approved by
Overall quality: acceptable the local institutional review board
Contra: no adjustment for multiple testing; no intention
to treat analysis; high drop-out; possible side effects
are not mentioned
Klement et al. [33] CT Rating according to SIGN PRO: compliance surveilled in arm A with weekly 3b
Positive: 3 points beta-hydroxybutyrate blood measurement and patient
Uncertain: 1 point questioning; analysis including the most important
Negative: 0 points factors using linear mixed effects model; groups are
Overall quality: acceptable comparable; study approved by the ethics committee
Contra: small sample size, especially in arm A; no
possibility for separation of the effects of ketogenic
diet and amino acid supplementation; low objectively
measured diet adherence using blood beta-hydroxybu-
tyrate levels (69%)
Ok et al. [34] CT Rating according to SIGN PRO: study protocol approved by responsible institu- 3b
Positive: 3 points tional review board; groups are comparable
Uncertain: 0 points Contra: small sample size; no possibility for separation
Negative: 1 point of the effects of carbohydrate deficit and smaller but
Overall quality: acceptable more frequent servings per day in intervention group;
no power analysis; high drop-out; short follow-up
RCT: randomized controlled trial; CT: non-randomized controlled trial; SIGN checklist: Scottish Intercollegiate Guidelines Network Methodol-
ogy: Checklist 2: Randomized Controlled Trials
An exception is the study of Fine et al. [31], which Feasibility and adherence
reported that patients with stable disease or partial remis-
sion on PET scan after the diet exhibited significantly higher Results from RCTs and CTs
dietary ketosis than those with progressive disease (n = 4,
p = 0.018). Out of the included 322 patients, which were included in
13
Table 3 Risk of bias in the Study RSQ AC BPP BOA IOD SR OSB
included single-arm studies and
case reports according to the Champ et al. [45] − − − − + + +
Cochrane risk of bias tool
Fearon et al. [44] − − − − + + +
Fine et al. [31] − − − − + + +
Jansen and Walach [54] − − − − + + −
Klement and Sweeney [41] − − − − + + +
Martin-McGill et al. [35] − − − − − + −
Martin-McGill et al. [49] − − − − + + −
Rieger et al. [30] − − − − + + −
Schmidt et al. [36] − − − − − + +
Tan-Shalaby et al. [37] − − − − − + +
van der Louw et al. [29] − − − − − + +
Woodhouse et al. [28] − − − ? + + +
Zahra et al. KETOLUNG [38] − − − − − + +
Zahra et al. KETOPAN [38] − − − − − + +
Bozzetti et al. [48] − − − − + + +
Schwartz et al. [39] − − − − + + +
Zuccoli et al. [46] − − − − + + +
Tóth and Clemens [47] − − − − + + +
Bozzetti et al. [20] − − − − + + +
Branca et al. [57] − − − − + + +
Nebeling et al. [40] − − − − + + +
Rossi-Fanelli et al. [43] − − − − + + +
Schroeder et al. [58] − − − − + + +
Artzi et al. [21] − − − − + + +
Iyikesici [26] − − − − + + +
Iyikesici [27] − − − − + + +
Strowd et al. [25] − − − − + + +
Moore [50] − − − − + + −
Elsakka et al. [59] − − − − + + +
Schwalb et al. [55] − − − − + + −
Brünings [60] − − − − ? ? ?
Brünings [61] − − − − ? ? ?
Schütz [62] − − − − ? ? ?
the 5 studies 72 drop-outs occurred (24.7%). From the 72 the quality of life (QoL) and functioning of the patients.
drop-outs, 38 (53%) were part of the intervention group and After adjusting for baseline values and chemotherapy score,
34 (47%) of the control group [23, 24, 32–34]. the PCS score was significantly better in the KD group.
There were no significant between-group differences con-
Results from single‑arm studies and case reports cerning the MCS score [32].
QoL was measured in 4 studies using the EORTC QLQ-
Feasibility and diet adherence was analyzed in 13 studies. C30 questionnaire [29, 36, 37, 41]. The results were overall
In total, 84 out of 139 patients (60%) were able to continue inconsistent, but most often reporting stable or decreasing
the diet for the duration of the intervention [21, 25, 28–31, QoL [29, 36, 41].
35–40].
Changes in body weight
Quality of life
Results from RCTs and CTs
Cohen et al. used the physical component summary (PCS)
and mental component summary (MCS) out of the Short All 3 RCTs reported a significant higher weight loss in the
Form (12) Health Survey (SF12) questionnaire to measure KD group than in the control group [23, 24, 42]. Freedland
13
Table 4 Risk of bias in the included single-arm studies and case reports rated with the Oxford criteria
Reference Study type Standardized Additional comments on methodology Evidence
rating of risk of level
bias (Oxford)
Champ et al. [45] Retrospective single-arm clinical study – PRO: study approved by responsible 4
institutional review board, adherence
checked with urine and blood ketone
bodies measurements
Contra: small sample size, no standard-
ized KD, no possibility for separation
of the side effects caused by KD and
concurring radio-chemotherapy
Fearon et al. [44] Crossover study – PRO: study approved by local hospital 4
ethical committee, crossover design
to minimize confounding by covari-
ates
Contra: small sample size, no wash-out
period resulting in possible carryover
effects, extremely short duration of
intervention
Fine et al. [31] Prospective single-arm pilot study – PRO: study approved by responsible 4
committee on clinical investigations,
adherence checked with written
food-recall records and blood ketone
bodies measurements
ized KD, no possibility for separa-
tion of the effects caused by KD and
weight loss
Jansen and Walach [54] Systematic, prospective cohort study – PRO: 4
Contra: small sample size, number
of observations for the majority of
the variables reported insufficient to
perform a reliable statistical analysis;
no standardized KD, no information
about an approval by the responsible
ethics committee; no information
about the occurrence of side effects;
potential conflict of interest: the
first author is a shareholder of the
company, that patients were specifi-
cally informed about as a source for
ketogenic food
Klement and Sweeney [41] Prospective – PRO: study approved by institutional 4
Case reports ethics review board, adherence
checked with food diaries written by
the patients and monitoring of ketone
levels in urine and blood
Contra: very small sample size, no
standardized KD, no possibility for
separation of the effects caused by
KD and radio(chemo)therapy
13
Table 4 (continued)
bias (Oxford)
Martin-McGill et al. [35] Randomized, mixed methods, feasibil- – PRO: study approved by local research 4
ity study ethics committee, adherence checked
with food diaries written by the
patients and monitoring of ketone
levels in urine and blood; randomized
Contra: small sample size, potential
conflict of interest: the first author
received a PhD studentship of the
company, that provided the medium-
chain triglyceride nutritional products
used in Arm A; two co-authors
received salary costs from the same
company
Martin-McGill et al. [49] Prospective single-arm pilot study – PRO: study approved by local 4
Research, Development and Innova-
tion committee; adherence checked
with food diaries written by the
patients and monitoring of ketone
levels in urine
Contra: very small sample size,
potential conflict of interest: the first
author received a PhD studentship
from a company, that produces KD
foods and supplements
Rieger et al. [30] Prospective single-arm pilot study – PRO: study approved by local institu- 4
tional review boards of the partici-
pating hospitals; adherence checked
with nutritional questionnaires and
monitoring of ketone levels in urine
Contra: relatively small sample size,
potential conflict of interest: one of
the co-authors is the founder of a
company, that produces KD foods
and supplements and provided the
nutritional packages used in the
study; data not stratified by center
Schmidt et al. [36] Prospective, single-arm pilot study – PRO: study approved by local ethics 4
committee; adherence checked with
patient documenting food intake and
monitoring of ketone levels in urine
Contra: small sample size, no stand-
ardization of KD despite carbohy-
drate intake
Tan-Shalaby et al. [37] Single-arm prospective feasibility trial – PRO: study approved by local Inde- 4
pendent Review Board
ized KD; no possibility for separa-
weight loss
van der Louw et al. [29] Prospective single-arm feasibility study – PRO: study approved by local 4
medical ethical committee; adher-
ence checked with monitoring of the
ketone body levels in the blood
Contra: small sample size, no pos-
sibility for separation of the effects
caused by KD and radio-chemother-
apy
13
Table 4 (continued)
bias (Oxford)
Woodhouse et al. [28] Retrospective single-arm feasibility – PRO: study approved by local insti- 4
study tutional review board; adherence
checked with monitoring of the
caused by KD and radio-chemo-
therapy; retrospective study that
only includes patients who achieved
ketosis
Zahra et al. KETOLUNG [38] Prospective single-arm phase 1 clinical – PRO: study approved by local insti- 4
trial tutional review board; adherence
the patients and monitoring of the
apy
Zahra et al. KETOPAN [38] Prospective single-arm phase 1 clinical – PRO: study approved by local insti- 4
trial tutional review board; adherence
the patients and monitoring of the
apy
Bozzetti et al. [48] Single case report – PRO: adherence secured, due to paren- 4
teral feeding
CONTRA: only a single patient
analyzed
Schwartz et al. [39] Case Report – PRO: study approved by local insti- 4
tutional review board; adherence
CONTRA: extremely small sample
size
Zuccoli et al. [46] Case Report – PRO: 4
CONTRA: only a single patient
analyzed; no possibility for separa-
radio-chemotherapy; no systematic
assessment of adverse effects
Tóth and Clemens [47] Case report – PRO: adherence checked with monitor- 4
ing of ketone levels in urine
CONTRA: only a single patient ana-
lyzed; no possibility for separation of
the effects caused by KD and radio-
therapy; no systematic assessment of
adverse effects; no standardized KD
Bozzetti et al. [20] Single-arm prospective Study – PRO: power analysis 4
Contra: small sample size; due to die-
tary intervention of only 3-h results
can hardly be translated to the effects
of a long-term dietary intervention
13
Table 4 (continued)
bias (Oxford)
Branca et al. [57] Single case report – PRO: 4

Contra: only a single patient analyzed;
no assessment of adverse effects
Nebeling et al. [40] Case reports – PRO: study approved by local 4
institutional review board; adher-
ence checked with food diaries and
monitoring of the ketone body levels
in blood and urine
Contra: small sample size; no system-
atic assessment of adverse effects;
no possibility for separation of the
effects caused by KD and radio-
chemotherapy
Rossi-Fanelli et al. [43] 3-Arm prospective Study – PRO: adherence secured, due to paren- 3b
teral feeding
Contra: no assessment of adverse
effects
Schroeder et al. [58] Prospective quantitative study – PRO: study approved by local research 4
ethics committee; prospective study
effects; no standardized diet; due
to dietary intervention lasting only
4 days at most, results can hardly be
translated to the effects of a long-
term dietary intervention
Artzi et al. [21] Prospective,2 arm pilot study – PRO: study approved by local insti- 4
tutional review board; adherence
ketone body levels in the urine
effects; small sample size; control
group added retrospectively
Iyikesici [26] Single-arm retrospective study – PRO: due to the retrospective nature no 4
institutional review board approval
required
Contra: no standardized diet; no pos-
caused by the KD and the additional
treatments, including: polychemo-
therapy and hyperthermia
Iyikesici [27] Single-arm retrospective study – PRO: due to the retrospective nature no 4
institutional review board approval
required
Contra: no standardized diet; no pos-
caused by the KD and the additional
treatments, including: polychemo-
therapy, hyperbaric oxygen therapy
and hyperthermia
Strowd et al. [25] Single-arm study – PRO: study approved by institutional 4
review board; adherence checked
with monitoring of the ketone body
levels in blood and urine
Contra: no structured assessment of
adverse effects; small sample size
13
Table 4 (continued)
bias (Oxford)
Moore [50] Single case report – PRO: 4

adverse effects; no possibility for
KD and chemotherapy
Elsakka et al. [59] Single case report – PRO: study approved by institutional 4
review board
adverse effects; no possibility for
KD and other treatments including
surgery, radiation, chemotherapy and
other novel treatments
Schwalb et al. [55] Case reports – PRO: 4
Contra: small sample size; no struc-
tured assessment of adverse effects;
no possibility for separation of the
effects caused by the KD and the
additional novel treatments, including
high dose vitamin D, colostrum and
multiple food supplements; two of
the authors own companies, which
produced most of the food supple-
ments used in this trial
Brünings [60] Case reports – PRO: 4
Contra: historic study, from a cur-
rent standpoint outdated and often
subjective methods used to assess the
effects of the diet
Brünings [61] Case reports – PRO: 4
effects of the diet
Schulte and Schütz [62] Case reports – PRO: 4
effects of the diet
KD ketogenic diet
et al. [24] found a weight loss of 12.1 kg in the interven- by Klement et al. [33], only regression coefficients for the
tion group, compared to a weight loss of 0.5 kg in the con- changes in body weight were provided. Here, a significantly
trol group (p < 0.001) during the 6 months of the diet. The higher reduction of body weight was reported for the sub-
study of Khodabakhshi et al. [23] reported a significantly group of breast cancer patients (p = 0.00014) and rectal can-
larger weight loss in the intervention group than in the con- cer patients (p = 0.01). However, in the subgroup of HNC
trol group over the course of a 3 month diet with 6.3 kg (head and neck cancer) patients the regression coefficient for
compared to 1.3 kg, respectively (p < 0.001). Over the same “Time × KD” implied a significant positive effect of the KD
3-month duration Cohen et al. [42] detected a weight loss on the body weight of the patients (p = 0.008) [33].
of 6.1 kg in the intervention group and 3 kg in the control
group (p < 0.05). Results from single‑arm studies and case reports
In one of the controlled trials by Ok et al. [34], there
were no significant differences in the reduction of body Changes in bodyweight were analyzed in 15 studies. A sta-
weight between both groups (p = 0.475). In the other trial tistical analysis to check for significance was performed
13
Table 5 Study characteristics and outcomes reported in the included RCTs and CTs
512
References Study type N Cancer site Age Intervention/duration Endpoints Outcomes
13
Freedland et al. [24] RCT Included patients N = 57 Prostate cancer Median: 72y Arm A: A low-carbohydrate diet, goal: 1. PSADT 1. Per protocol, no difference was found
Analyzed patients N = 45 (≤ 20 g per day), estimated actual 2. Weight loss in log-transformed PSADT over the
Arm A: N = 27 carbohydrate intake: 37 g/day; supervi- 3. BMI 6-months between arms using a T-test
Arm B: N = 18 sion by dietitians by telephone weekly 4. Waist circumfer- (mean values in LCD vs. control: 21 vs.
Drop-out Arm A: 4 for the first 3 months and then every ence 15 months, p = 0.446)
Arm B:8 2 weeks for the last 3 months 5. Adverse events Post hoc exploratory analyses of
Arm B: Control group (no dietary PSADT: after adjusting for key baseline
intervention) covariates including
Duration: 6 months baseline PSA, pre-study PSADT, treat-
ment received (surgery vs. radiation)
and accounting for hemoconcentration
during the study, LCD significantly
lowered log-transformed PSADT (28 vs
13 months, p = 0.021)
2. Significantly higher weight loss in
Arm A, than in Arm B; Arm A pretest:
197.5 kg, Δ from baseline − 12.1 kg,
Arm B pretest: 196.2 kg, Δ from base-
line − 0.5 kg; between-group compari-
son at the end of the study p < 0.001
3. Significantly higher BMI reduction in
29.0 kg/m2, Δ from baseline − 3.9 kg/
m2 Arm B pretest: 29.7 kg/m2, Δ from
baseline-0.2 kg/m2; between-group
comparison at the end of the study
p < 0.001
4. Significantly higher waist circumfer-
ence reduction in Arm A, than in Arm
B; Arm A pretest: 107.0 cm, Δ from
baseline − 11.8 cm Arm B pretest:
110.7 cm, Δ from baseline-0.5 cm;
between-group comparison at the end
of the study p < 0.001
5. Similar number of AEs at baseline
in both groups; numerically more
AEs in Arm A at 3 months (30 vs 19)
and slightly more AEs in Arm A at
6 months (19 vs 15); only mild and one
moderate AE (nausea) reported
Table 5 (continued)
Khodabakhshi et al. [23] RCT Included patients N = 77 Breast cancer Intervention Arm A: Medium-chain triglycerides 1. Overall survival 1. No data for the whole study population
Analyzed patients N = 60 group: (MCT) based ketogenic diet (6% calo- 2. Weight given; significantly prolonged survival
Arm A: N = 30 Mean: ries from Carbohydrates [CHO], 19% 3. BMI in a subgroup of only neoadjuvant
Arm B: N = 30 44.8 years protein, 20% MCT, 55% fat); Patients 4. Body fat patients; log rank test for Kaplan–Meier
Drop-out Arm A: 10 Control group: received 500 ml of MCT oil from the p = 0.04
Arm B:7 Mean:45.2 years Nutricia Company every 2 weeks 2.Significantly higher weight loss in
Arm B: Standard Diet (55% Arm A, than in Arm B; Arm A pretest:
CHO, 15% protein, and 30% fat) 71.7 kg, Δ from baseline − 6.3 kg,
Duration: 3 months Arm B pretest: 70.5 kg, Δ from
baseline—1.3 kg; between-group
comparison at the end of
the study p < 0.001
3. Significantly higher BMI reduction in
28.47 kg/m2, Δ from
baseline 2.57 kg/m2 Arm B pretest:
28.44 kg/m2, Δ from baseline-0.64 kg/

m2; between-group comparison at the
end of the study p < 0.001
4. Significantly higher reduction of body
fat, adjusted for baseline value, in Arm
A, than in Arm B;
A pretest: 35.8%, Δ from baseline − 6.7%,
Arm B pretest: 34.5%, Δ from base-
line—3.7%; between-group comparison
at the end of the study p = 0.03
13
513
Table 5 (continued)
514
13
Cohen et al. [32, 42, 56] RCT Included patients N = 73 Ovarian cancer, Endo- Mean: Arm A: Ketogenic diet (70% [≥ 125 g]: 1. Physical and 1. Significant between-group difference in
Analyzed patients N = 45 metrial cancer 60.2 years 25% [≤ 100 g]: 5% [< 20 g] energy per mental health PCS after adjusting for baseline values
Arm A: N = 25 day from fat, protein, and carbohy- status and chemotherapy status (p = 0.04),
Arm B: N = 20 drates) 2. Energy level with fat loss added to the model,
Drop-out Arm A: 12 Arm B: American Cancer Society diet 3. Hunger and the effect was no longer significant
Arm B:16 (ACS: high in fiber, low in fat) satiety, and food (p = 0.064)
Individual diet advice from certified dieti- cravings 2. no significant between-group difference
tians. Weekly e-mails or phone calls. 4. Body composi- in MCS, only a subgroup of the partici-
One face-to-face meeting after baseline tion pants in the intervention group without
assessment concurrent chemotherapy reported a
Duration: 3 months statistically significant improvement of
23% in energy level from baseline to
12 weeks (p = 0.02)
3. significant less cravings for starchy
foods and fast-food fats after adjusting
for baseline values and chemotherapy
status in the intervention group
(p = 0.03 and p = 0.04, respectively)
measured with FCI
4.significantly higher reduction of total
body mass in Arm A, than in Arm
B; Arm A pretest: 81.2 kg, Δ from
baseline − 6.1 kg, Arm B pretest: 89 kg,
Δ from baseline − 3 kg; between-group
p < 0.05 significantly higher reduction
of total fat mass in Arm A, than in
Arm B; Arm A pretest: 37.9 kg, Δ
from baseline − 5.2 kg, p < 0.05, Arm
B pretest: 44.1 kg, Δ from baseline
− 2.9 kg, p > 0.05; between-group
p < 0.05 no significant differences in
lean body mass between Arm A and
Arm B; Arm A pretest: 43.2 kg, Δ from
baseline − 0.9 kg, p > 0.05, Arm B pre-
test: 44.9 kg, Δ from baseline − 0.1 kg,
p > 0.05; between-group comparison at
the end of the study p > 0.05
Table 5 (continued)
Klement et al. [33] Controlled Included patients N = 85 Rectal cancer, head From 38 to Arm A: ketogenic diet with additional 1. Diet adherence 1. subjectively reported by patients: 100%
study Analyzed patients N = 81 and neck cancer 76 years consumption of non-glucogenic amino 2. Body composi- objectively measured using blood BHB
Arm A: N = 20 Breast cancer acids, patients are provided with tion changes levels: 69%
Arm B: N = 61 literature regarding ketogenic diet; 2. Regression coefficients for body
Drop-out Arm A: 2 opportunity to speak with a dietician composition changes, according to the
Arm B:2 Arm B: control (no dietary intervention); linear mixed-effects model. Effects
in case of dietary counseling: official of the KD over time were described
recommendations of the German Soci- with the coefficient “KD x Time” in
ety for nutrition provided to the patient the study:
Duration: as long as the patients received Rectal cancer patients:
RT (median duration: 35-40 days) Regression coefficient for body weight
change in Arm A compared to Arm B:
− 0.4 kg/week, p = 0.011
Regression coefficient for fat-free mass
0.0 kg/week, p = 0.9467
Regression coefficient for fat mass

− 0.5 kg/week, p = 0.000889
HNC patients:
Regression coefficient for body weight
change in Arm A compared to Arm
B: + 0.6 kg/week, p = 0.00823
Regression coefficient for fat free mass
B: + 0.4 kg/week, p = 0.03423
Regression coefficient for fat mass change
in Arm A compared to Arm B: + 0.2 kg/
week, p = 0.3296 breast cancer patients:
Regression coefficient for body weight
− 0.3 kg/week, p = 0.00124
Regression coefficient for fat free mass
B: + 0.1 kg/week, p = 0.1655
Regression coefficient for fat mass
− 0.4 kg/week, p = 8.49 × 10–5
13
515
Table 5 (continued)
516

Ok et al. [34] Controlled Included patients N = 30 Pancreato-biliary intervention Arm A: Ketogenic diet (3–6%, 14–27%; 1. Average energy 1. Arm A:61.3%; Arm B: 38.5%; p < 0.05,
13
study Analyzed patients N = 19 cancer group: 70–80% energy per day from carbo- intake rate significant higher in Arm A
Arm A: N = 10 Mean: hydrates, protein, and fat) served as 3 2. Average protein 2. Arm A:63.5%; Arm B:37.7%; p > 0.05,
Arm B: N = 9 57.8 years meals and 3 snacks per day intake rate no significant difference
Drop-out Arm A: 10 Control group: Arm B: usual Korean diet (55–65%, 3. Frequency of 3. Arm A: average number of Problems
Arm B:1 Mean: 7–20%, 15–30% energy per day from meal intake- per person 1.3
66.3 years carbohydrates, protein and fat) served related problems Arm B: average number of problems
as 3 meals per day 4. Body composi- per person 2; p > 0.05, no significant
Duration: Measurement of meal compli- tion difference
ance, energy and protein intake: 4. No significant differences in the
10 days reduction of body weight in Arm A,
Measurement of body composition compared to Arm B;
and frequency of meal intake-related Arm A pretest: 64.6 kg, Δ from baseline
problems: till 1 st outpatient visit after − 4 kg, Arm B pretest: 56.2 kg, Δ from
surgery (mean hospital stay for Arm baseline − 3.5 kg; between-group com-
A = 12 days) parison at the end of the study p = 0.475
significantly less reduction of body cell
mass in Arm A, than in Arm B; Arm
A pretest: 28.9 kg, Δ from baseline
− 1.9 kg; Arm B pretest: 27.4 kg, Δ
from baseline − 2.9 kg; between-group
p = 0.049; no significant differences
in body fat mass between Arm A and
Arm B; Arm A pretest: 18.2 kg, Δ from
baseline − 1.1 kg
Arm B pretest: 13.7 kg, Δ from base-
line + 0.5 kg; between-group compari-
son at the end of the study p = 0.086
PSADT prostate-specific antigen doubling time, AE adverse event, BMI body mass index, PCS physical component summary, MCS mental component summary, FCI food craving inventory, RT
radiotherapy, BHB beta-hydroxybutyrate, HNC head and neck cancer
in 8 studies, of which 4 found a significant reduction in one moderate AE (nausea) were reported. The number of
bodyweight [30, 36, 37, 41]. Three of these studies found AEs was similar at baseline but increased drastically in the
a non-significant decrease in body weight [25, 31, 43] and KD group (30 vs 19 reported AEs) at 3 months. At 6 months,
only Fearon et al. [44] showed a significant increase in body the number of AEs had subsided back to baseline in the KD
weight. group and was again close to the number in the control group
Out of the remaining seven studies, where no statistical at the same time.
analysis was performed, weight loss during the diet occurred Ok et al. [34] assessed the number of meal intake-related
in six studies [38, 39, 45–47], while only one study showed problems and postoperative complications. No significant
an increase in body weight [48]. differences between both groups in either of the two cat-
One study assessed the change in BMI and reported a egories occurred.
median decline of 1.04 kg/m2, without checking for signifi-
cance [28]. Results from single‑arm studies and case reports
Changes in body composition Adverse events were monitored in 19 studies. A validated

tool was used in 11 of the 19 studies [26–29, 31, 35, 37, 38,
Results from RCTs and CTs 45, 46]. Since many studies combined KD with standard of
care (SoC) chemotherapy and/or radiation therapy, it was
Changes in body composition were analyzed in one RCT often not possible to determine the cause of the reported
and both CTs. The RCT by Cohen et al. found a significant AEs. Most of the AEs were mild to moderate. The most
higher reduction of total fat mass in the KD group (− 5.2 kg) common AEs include: fatigue [31, 45], constipation [29,
than in the control group (− 2.9 kg), while no significant 31], diarrhea [29, 35] as well as nausea and vomiting [29,
differences concerning the lean body mass occurred [42]. 35]. Further reported AEs were: deep venous thrombosis,
Klement et al. provided primarily the regression coef- asymptomatic hypoglycemia, nephrolithiasis, leg cramps,
ficients for the fat mass (FM) and fat-free mass (FFM). In dyspepsia, dry mouth, hyperuricemia, hyperlipidemia,
the subgroup of rectal cancer patients, a significantly greater pedal edema, anemia, neutropenia and febrile neutropenia,
loss of FM occurred in the KD group, without significant thrombocytopenia, halitosis, pruritus, hypoglycemia, hyper-
differences in FFM. A comparable result was reported in kalemia, hypokalemia, hypomagnesemia, flu-like symptoms,
the subgroup of breast cancer patients, who experienced a low carnitine, hallucinations, allergic reaction, wound infec-
significant reduction in FM, while the FFM reduction was tion, headaches and neuropathy [26–31, 35–39, 45–50].
not significant. However, the 50 kHz phase angle, an indica- Even though most AEs were mild to moderate, there
tor for changes in cell mass, also significantly declined in the were also DLTs (dose-limiting toxicity) like CTCAE (NCI
KD group. In the subgroup of HNC patients, the regression Common Terminology Criteria for Adverse Events) grade 3
coefficients implied a significant increase in FFM in patients dehydration, grade 4 hyperuricemia [38] and a case of grade
receiving a KD [33]. 5 neutropenia, resulting in the death of the patient [26].
Ok et al. found a significantly lower reduction in body cell
mass in the KD group (− 1.9 kg) than in the control group
(− 2.9 kg), while no significant differences in body fat mass Discussion
occurred [34].
Summary of main results
Results from single‑arm studies and case reports
The basic idea of using a KD to prohibit cancer growth relies
Two studies analyzed changes in body composition [35, on the Warburg hypothesis and successful animal and cell
41]. One study showed a significant FM reduction, without culture studies. However, clinical evidence demonstrating
significant reduction in FFM [41], whereas the other study a beneficent effect on survival and anti-tumor efficiency is
showed no significant effects on body composition [35]. still lacking.
The RCT conducted by Freedland et al. [24] failed to
Adverse events detect a significant anti-tumor effect in per-protocol analysis
and an effect was only visible in a strongly adjusted explora-
Results from RCTs and CTs tory analysis. Only Khodabakhshi et al. [23] found a sig-
nificantly longer OS of the neoadjuvant treated subgroup of
Only one RCT [24] and one CT [34] monitored adverse breast cancer patients. But said data are only presented as a
events and only the CT by Ok et al. used a validated tool Kaplan–Meier plot, without any further information, despite
[34]. In the RCT by Freedland et al. [24], only mild AEs and a p value of 0.04 and the claim of a higher survival rate in
13
Table 6 study characteristics and outcomes reported in the included single-arm studies and case reports
518
Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes
13
Champ et al. [45] Retrospective single- Analyzed patients Glioblastoma multi- From 34 to 62 years Arm A: self-adminis- 1. Adverse events 1. 2 patients with grade
arm clinical study N = 53 forme tered KD 2. Bodyweight 1 constipation, 4
Arm A: N = 6 Arm B: unspecified patients with grade
Arm B: N = 47 standard American 1 fatigue, 1 patient
diet with grade 2 fatigue,
Duration: 1 patient with deep
3–12 months venous thrombosis
during treatment, 1
patient with asympto-
matic hypoglycemia,
1 patient with neph-
rolithiasis no grade 3
and higher toxicities
or symptomatic hypo-
glycemia
2. weight loss on non-
calorie-restricted KD:
1 to 27Ibs
Weight loss on calorie-
restricted KD: 46Ibs
Fearon et al. [44] Crossover study Analyzed patients Ovarian, Lung, Mean: 61 years Crossover study: 1. Protein synthesis, 1. No significant differ-
N=5 Gastric Nasogastric tube turnover and nitro- ences, mean daily N
feeding: normal, gen balance balance non-signifi-
balanced regimen on 2. Bodyweight cantly more positive
days 1–6 3. Performance status on normal, balanced
KD containing same diet, p > 0.1
total calorie and pro- 2. No significant
tein on days 7–13 change in body
Duration: 13 days weight during nor-
mal balanced diet,
p > 0.05
Significant increase in
body weight during
KD (average + 2 kg),
p < 0.05
3. Performance status
did not change during
normal balanced diet,
but increased by one
point during KD,
but no testing for the
statistical significance
was applied
Table 6 (continued)
Fine et al. [31] Prospective single- Recruited patients Diverse Mean: 62.9 years KD with targeted 1. Toxicity 1. 5 patients with grade
arm pilot study N = 12 CHO intake below 2. Metabolic effects 2 fatigue, 5 patients
Analyzed patients 5% of total energy 3. Dietary adherence with grade 1 consti-
N = 10 intake, written pation, 1 patient with
menus and samples grade 1 leg cramps
of CHO-restriction 2. Mean weight loss
products were 4% compared to
provided baseline, p = 0.08; all
Duration: 28 days patients spontane-
ously decreased their
caloric intake, mean
energy deficit: 35%,
p < 0.01 compared
with baseline
3. 5 out of 12 patients
completed all 28d of
the diet
Jansen and Walach Systematic, prospec- Analyzed patients Diverse Mean: 68.3 years Arm A: full adoption 1. TKTL 1 level 1. Reduction in TKTL
[54] tive cohort study N = 78 of a non-specified 2. Improvement in 1 was associated with
Arm A: N = 7 KD, patients cancer status adopting a KD, no
Arm B: N = 6 informed about a test for significance
Arm C N = 65 single company pro- due to insufficient
ducing KD related number of cases
food 2. Correlation between
Arm B: partial adop- improvement in
tion of a non-spec- cancer status category
ified KD, patients and full adoption of
informed about a a KD (χ2 = 33.26;
single company pro- df = 4; p = 0.00001),
ducing KD related no information
food provided about the
Arm C: patients who definitions and the
did not adopt a KD exact methods used
Duration: non-spec- to define the cancer
ified, study began status categories
11/2010, follow-up
until end of 2011
13
519
Table 6 (continued)
520
13
Klement and Sweeney Prospective Case Analyzed patients Diverse From 40 to 74 years Self-administered KD 1. QoL 1. Only measured in
[41] reports N=6 (recommended CHO 2. Bodyweight 5 out of 6 patients,
intake < 50 g/day) 3. Body composition QoL at the end of RT
during the course of decreased in 3 out of
RT/RCT; patients 5 patients and stayed
received basic consistent in 2 out
information on KD; of 5
counseling at least 2. Significant decrease
once per week in 2 patients, only
Duration: Patient analyzed individually,
dependent from 32 no analysis for the
to 73 days whole study popula-
tion performed
3. Only 4 patients ana-
lyzed; FM decreased
significantly in 3
patients, FFM did not
change significantly
Table 6 (continued)
Martin-McGill et al. Randomized, mixed Assessed for eligibil- Glioblastoma From 44 to 66 years Arm A: MCTKD 1. Long-term reten- 1. Arm A: 3 patients
[35] methods, feasibility ity: N = 57 (75%; 15%; 10% tion retained for 3 months
study Randomized: N = 12 of energy per day 2. Quality of life (drop-out = 50%)
Arm A: N = 6 from fat, protein 3. Adverse events Arm B: 1 patient
Arm B: N = 6 and carbohydrates, retained for 3 months
Retention at with 30% of fat from (drop-out = 83%)
12 weeks. N = 4 MCT nutritional 2. GHS at baseline:
Arm A: N = 3 products) Arm A: patients
Arm B: N = 1 Arm B: MKD (80%; who later withdrew:
15%; 5% of energy 72.2 ± 20.7; patients
per day from fat, who retained: 75 ± 6.8
protein and carbohy- Arm B: patients who
drates) later withdrew:
Duration: 12 weeks 70 ± 13.8; patients
who retained: 80 ± 0
GHS: at week 6: Arm
A: patients who
withdrew at week 6:
41.7 ± 0; patients who
retained: 66.7 ± 0
Arm B: patients who
withdrew at week 6:
50 ± 0; patients who
retained: 100 ± 0
3. Adverse events dur-
ing the first 6 weeks:
Arm A: diarrhea
(n = 1, CTCAE grade
1), nausea (n = 1,
CTCAE grade 1),
vomiting (n = 1,
CTCAE grade 2),
dyspepsia (n = 1,
CTCAE grade 1)
Arm B: vomiting
(n = 1, CTCAE grade
1), dry mouth (n = 1
MKD, CTCAE grade
1)
13
521
Table 6 (continued)
522
13
Martin-McGill et al. Prospective single- Enrolled: N = 6 Glioblastoma From 34 to 66 years MKD (70%: 3–5% 1. Adverse events 1. Constipation in 2
[49] arm pilot study Completed interven- [≤ 20 g] energy per 2. Body composition patients, resolved
tion: N = 4 day from fat and car- with dietary modifi-
bohydrates; protein cation
consumption was 2. No significant differ-
not restricted ences in body compo-
Duration: 12 weeks sition occurred
Rieger et al. [30] Prospective single- Included patients Glioblastoma Median: 57 years KD with CO 1. Feasibility 1. 3 out of 20 patients
arm pilot study N = 20 intake < 60 g/day, 2. Bodyweight discontinued the diet
Evaluable for effi- additionally highly 3. Tolerability after 2–3 weeks with-
ciency N = 17 fermented yoghurt 4. Efficacy out progression, due
drinks and two to reduced QoL
different plant oils 2. Significant body
were provided to be weight reduction;
consumed at will mean weight at base-
No calorie restric- line: 78.3 kg, mean
tion, patients were weight at the end
instructed to always of the diet: 76.5 kg
eat to satiety (p < 0.05)
Duration: till progres- 3. Diarrhea, constipa-
sion of the disease tion, hunger and/or
demand for glucose
were present in a
minority of patients
during the diet
4. Median PFS on
the KD alone was
5 weeks
No significant dif-
ference between
median PFS on the
KD with additional
bevacizumab treat-
ment (20.1 weeks)
and median PFS of
patients on normal
diet treated with
bevacizumab in the
same hospital dur-
ing the same period
(16.1 weeks) p = 0.38
Table 6 (continued)
Schmidt et al. [36] Prospective, single- Enrolled: N = 16 Diverse From 33 to 64 years KD with CHO limited 1. Feasibility 1. 11 out of 16 Patients
arm pilot study Completed interven- to 70 g per day and 2. Bodyweight discontinued the
tion: N = 5 20 g per meal 3. Adverse events diet, 3 out of 11 were
Two oil–protein 4. QoL unable to adhere to
shakes consumed in the diet, 6 out of 11
the morning and in discontinued due to
the afternoon progressive disease
Duration: 12 weeks and 2 out of 11 died
from progressive
disease
2. Only analyzed in 7
patients; significant
weight loss of 2 kg
from mean 68.5 kg
at baseline to 66.5 kg
at the end of the diet,
p < 0.05
3. Statistical evalua-
tion of the adverse
events and the influ-
ence on QOL is not
statistically feasible;
reported side effects
included increase in
appetite loss, consti-
pation, diarrhea and
fatigue during the diet
4. QoL was low at
baseline and stayed
relatively stable
during the interven-
tion; worsening of
fatigue, pain, dyspnea
and role function but
emotional function-
ing and insomnia
improved slightly
13
523
Table 6 (continued)
524
13
Tan-Shalaby et al. Single-arm prospec- Enrolled: N = 17 Diverse From 42 to 87 years Modified Atkins Diet 1. Feasibility 1. 13 out of 17 patients
[37] tive feasibility trial Drop-out before first with 20 to 40 g of 2. Bodyweight discontinued the diet
analysis: N = 6 CHO and restricted 3. Adverse effects before 16 weeks
Completed interven- consumption of 4. QoL 2. Significant mean
tion: N = 4 high CHO foods no weight loss of all
restrictions for calo- subjects: 7.5 kg,
ries, protein or fats p < 0.05; significant
Duration 16 weeks mean weight loss
of the patients, who
completed the diet:
12.3 kg, p < 0.05
3. Reported adverse
effects included:
hyperuricemia
(N = 7), hyperlipi-
demia (N = 2), pedal
edema (N = 2),
anemia (N = 2), hali-
tosis (N = 2), pruritus
(N = 2), hypoglycemia
(N = 2), hyperkalemia
(N = 2), hypokalemia
(N = 2), hypomagne-
semia (N = 2), flu-like
symptoms/fatigue
(N = 2)
4. Patients, who
completed at least
4 weeks of the diet
(N = 6) showed no
significant deteriora-
tion in QoL
Table 6 (continued)
van der Louw et al. Prospective single- Eligible patients: Glioblastoma multi- Median: 53.8 years Phase A: Fluid KD 1. Feasibility 1. 6 out of 9 patients
[29] arm feasibility study N = 11 forme with a 4:1 ratio (4 g 2. Adverse effects (67%) included in
Included in phase A: fat versus 1 g protein 3. QoL phase A completed
N=9 plus carbohydrates, 4. Overall survival the 14 weeks KD
Included in phase B: 90% energy from 2. Reported adverse
N=8 fat) Patients were effects included:
Completed interven- allowed a snack with CTCAE grade 1:
tion N = 6 the same 4:1 diet constipation (n = 7),
ratio once a day nausea/vomiting
Phase B: Solid-food (n = 2), hypercho-
KD (diet ratio lesterolemia (n = 1),
1.5–2.0:1) with hypoglycemia (n = 1),
MCT; (70% energy low carnitine (n = 1)
from fat with the and diarrhea (n = 1).
consistency of an CTCAE grade 2:

emulsion) hallucinations (n = 1),
Duration: 14 weeks allergic reaction
(6 weeks phase A, (n = 1) and wound
8 weeks phase B) infection (n = 1)
3. Global quality of
live at baseline: 83%,
global quality of live
at end of study: 58%;
reference value: 78%
4. The median overall
survival of the
nine patients was
12.8 months; median
survival duration
reference value is
15 months
Woodhouse et al. [28] Retrospective single- Analyzed patients: Glioma From 30 to 76 y MAD with a 0.8–1:1 1. Feasibility 1. 28 out of 29 patients
arm feasibility study N = 29 ratio (0.8-1 g fat to 2. Adverse events (96.6%) completed
1 g carbohydrate 3. Changes in BMI the 6-week diet
plus protein 2. Grade 2 constipa-
Duration: 6 weeks tion (n = 1), grade 1
fatigue and nausea
were present in the
patients
3. Median change of
BMI for all patients
was –1.04 kg/m2, not
analyzed for signifi-
cance
13
525
Table 6 (continued)
526
13
Zahra et al. KETOL- Prospective single- Screened patients: Lung Median: completed KD with 90%; 8%; 1. Feasibility 1. 2 out of 7 patients
UNG [38] arm phase 1 clinical N = 11 KD: 66 years 2% of energy per 2. Adverse events (29%) completed the
trial Enrolled patients: Did not complete: day from fat, protein 3. Bodyweight intervention
N=7 67 years and carbohydrates. 2. Reported adverse
Completed interven- All meals readily events included:
tion: N = 2 prepared for the CTCAE Grade
patients 1–2: constipation,
Duration: 42 days diarrhea, nausea,
vomiting and fatigue;
1 patient experienced
DLT (hyperuricemia
Grade 4)
3. Average weight loss:
5.6 kg
Zahra et al. KETO- Prospective single- Screened patients: Pancreas Completed KD: KD with 90%; 8%; 1. feasibility 1. 1 out of 2 patients
PAN [38] arm phase 1 clinical N=5 69 years 2% of energy per 2. adverse events (50%) completed the
trial Enrolled patients: Did not complete KD: day from fat, protein 3. bodyweight intervention
N=2 67 years and carbohydrates. 2. Reported adverse
Completed interven- All meals readily events included:
tion: N = 1 prepared for the CTCAE grade 1–2:
patients Constipation, diar-
Duration: 34 days rhea, nausea and
vomiting 1 patient
experienced DLT
(dehydration grade 3)
3. Average weight loss:
8.2 kg
Bozzetti et al. [48] Single case report N=1 Desmoid tumor 28y TPN consisting of 1. bodyweight 1. Body weight
28 kcal fat/kg body 2. adverse events increased by 1 kg
weight/day, 1.5 g (from 61 to 62 kg)
protein/kg body 2. No adverse events
weight/day; 40 g reported, no signs of
glucose/day hepatic steatosis or
Duration: 5 months liver damage
Schwartz et al. [39] Case report Included patients: Glioma From 3 to 65 years ERKD: with a 3:1 1. feasibility 1. 1 out of 2 patients
N=2 ratio of ingested 2. adverse events (50%) completed the
Completed interven- nutrients (3 g fat 3. bodyweight intervention
tion: N = 1 versus 1 g protein 2. Besides headaches
plus carbohydrates) no adverse events
20% restriction of 3. Body weight initially
calories per day decreased in both
Duration: 12 months patients and remained
stable afterward
Table 6 (continued)
Zuccoli et al. [46] Case Report N=1 Glioblastoma multi- 65 years ERKD delivering 1. bodyweight 1. bodyweight
forme 600 kcal per day, 2. adverse events decreased 3 kg (from
consisting of 42 g 58 to 55 kg) in the
fat, 32 g protein and first 14 days of the
10 g CHO per day diet
Duration: 56 days 2. No adverse events
despite grade 4
hyperuricemia
reported, resulted
in diet change to
calorie restricted non-
ketogenic diet
Tóth and Clemens Case report N=1 Rectal 62 years Paleolithic KD, 1. adverse events 1. No adverse events
[47] nutrients consumed 2. bodyweight were reported

in a fat: protein ratio 3. tumor volume 2. Bodyweight
of 2:1 animal fat, decreased 13 kg
red meats and organ (from 78 to 65 kg)
meats were encour- during the diet
aged, root vegeta- 3. Initial decrease in
bles were allowed, volume after con-
all other foods were comitant radiother-
prohibited apy; tumor volume
Duration: 24 months remained stable in the
following months, but
four hepatic metasta-
ses were detected at
the end of the diet
Bozzetti et al. [20] Single-arm prospec- N = 12 Diverse From 31 to 75 years single 3 h infusion 1. Glucose uptake 1. No statistically
tive Study of glucose-based analysis of the liver significant stimula-
(GTPN) or a lipid- metastases using tion or suppression of
based TPN (LTPN) FDG-PET FDG uptake due to
containing 4 mg the administration of
glucose/kg/min or GTPN or LTPN
2 mg lipid/kg/min,
respectively
Duration: 3 h
13
527
Table 6 (continued)
528
13
Branca et al. [57] Single Case Report N=1 Breast 66 years Self-administered 1. changes in tumor 1. Progesterone recep-
high doses of biomarkers tor status positivity
oral vitamin D3 increased from < 1%
(10,000 IU/day), and at baseline to 20%
KD rich in Oleic after the 3-week
acid intervention; HER2
Duration: 3 weeks positivity decreased
from > 10% (score
2 +) to 0% (score
0) after the 3-week
intervention
Nebeling et al. [40] Case reports N=2 Astrocytoma From 3 to 8.5 years KD with 60%; 20%; 1. Glucose uptake 1. Dose uptake ratio
10%, 10% of energy analysis of the tumor: normal cortex
per day from MCT tumor using FDG- decreased by approxi-
oil, protein, carbo- PET mately 22% in both
hydrates and dietary 2. feasibility patients
fat plus additional 2. 2 out of 2 (100%)
supplements patients were able to
Duration: 8 weeks complete the dietary
intervention
Rossi-Fanelli et al. 3-Arm prospective Enrolled: N = 27 Esophagus Median: Arm A: glucose- 1. tumor cell kinetics 1. Assessed as the
[43] Study Arm A: N = 9 Stomach Arm A: 61 years based TPN (100% 2. bodyweight fraction of cells in
Arm B: N = 9 Colon–rectum Arm B: 70 years of the calorie from S-phase; none of the
Arm C: N = 9 Arm C: 67 years dextrose) changes within and
Arm B: lipid-based between the three
TPN (80% of the arms reached statisti-
calorie from fat, cal significance
20% from dextrose) 2. None of the changes
Arm C: oral diet within and between
All diets were iso- the three arms
caloric and isoni- reached statistical
trogenous significance
Duration: 2 weeks
Table 6 (continued)
Schroeder et al. [58] Prospective quantita- N = 12 Head and neck From 50 to 86 y Unspecified western 1. metabolic changes 1. Decline of mean
tive study diet followed by in the tumor tissue lactate concentration
unspecified KD in the tumor tissue
Duration: variable, up during the KD, no
to 4 days analysis for statistical
significance per-
formed glucose and
pyruvate concentra-
tion in the tumor
tissue were stable or
even increased, no
analysis for statisti-
cal significance
performed
Artzi et al. [21] Prospective, two-arm Included: N = 9 Brain From 27 to 69 years KD based on ready- 1. feasibility 1. Diet tolerated by
pilot study intervention: N = 5 made formula, 2. ketone body levels 4/5 patients, strict
retrospectively added with a 4:1 ratio of in the brain adherence only in 2
control N = 4 ingested nutrients patients
(4 g fat versus 1 g 2. 4 out of 50 MRI
protein plus carbo- spectroscopy scans
hydrates) detected ketone bod-
Duration: variable ies in the brains of
from 2 to 31 months the patients following
the KD
None of the scans
detected ketone
bodies in the control
group
13
529
Table 6 (continued)
530
13
Iyikesici [26] Single-arm retrospec- N = 44 Lung (NSCLC) Median: 65 years Mild KD (patients 1. survival 1. After 24 weeks
tive study were encouraged 2. adverse events 42 patients (95%)
to avoid high CHO and at the termina-
food) in combination of follow-up 29
tion with HBO, patients (66%) were
hyperthermia and alive mean OS was
polychemotherapy 43 months (numeri-
administered during cally better than
induced hypogly- historical controls
cemia from other studies)
Duration: 24 weeks 2. Adverse events
Follow-up: 1–6 years reported during treat-
ment period: grade 5
neutropenia (N = 1),
grade 3 neutrope-
nia (N = 3), grade
3 anemia (N = 10),
grade 4 thrombocyto-
penia (N = 3), grade 3
fatigue (N = 5), grade
3 diarrhea (N = 8),
grade 3 neuropathy
(N = 1), all of which
were attributed to
chemotherapy
Iyikesici [27] Single-arm retrospec- N = 25 Pancreas Median: 61 years Mild KD (patients 1. survival 1. During follow-
tive study were encouraged 2. adverse events up mean OS was
to avoid high CHO 15.8 months (numeri-
food) in combina- cally better than
tion with HBO, historical controls
hyperthermia and from other studies)
polychemotherapy 2. Adverse events
administered during reported during
induced hypogly- treatment period:
cemia grade 3/4 neutrope-
Duration: mean fol- nia (N = 9), febrile
low-up: 25 months neutropenia (N = 1),
grade 3 anemia
(N = 7), grade 4
thrombocytopenia
(N = 4), grade 3 diar-
rhea (N = 2), all of
which were attributed
to chemotherapy
Table 6 (continued)
Strowd et al. [25] Single-arm study Included N = 8 Brain From 28 to 54 years MAD with20g CHO/ 1. Bodyweight 1. Non-significant
Completed interven- day restriction 2. Seizure frequency body weight decrease
tion N = 7 Duration: by a mean 3.4 kg
2–24 months (mean (p = 0.48)
13.17 months) 2. Non-significant
reduction in mean
seizure frequency
per week from 0.54
at baseline to 0.1 at
6 months (p = 0.27)
Moore [50] Single case report N=1 Glioblastoma multi- 40 years Energy-restricted KD 1. Anti-tumor effect 1. PET-CT at the end
forme with a 4:1 ratio of 2. Adverse events of the diet detected no
calorie intake (fat metabolically active
versus protein plus tumor, despite a new

carbohydrates) enhancement area
Total calories calcu- in MR
lated 25% below 2. No significant
BMR fatigue or reduced
Duration: 4 months mental capacity
reported, patient was
able to continue his
work and exercise
regime
Elsakka et al. [59] Single case report N=1 Glioblastoma multi- 38 years KD with a 4:1 ratio 1. Anti-tumor effects 1. Good surgical out-
forme of calorie intake 2. Body weight come and regressive
(fat versus protein 3. QoL changes in histopa-
plus carbohydrates), 4. Anti-tumor effect thology
delivered as calorie 2. Body weight
restricted diet, decreased 9.3 kg dur-
combined with ing the intervention
intermittent fasting, 3. No clinical or neu-
HBOT, other novel rological symptoms
therapies and SOC reported, despite
treatment reduced weight no
Duration: 20 months discomfort
4. After subtotal tumor
resection, radio- and
chemotherapy station-
ary disease
13
531
Table 6 (continued)
532
13
Schwalb et al. [55] Case reports N=6 Diverse From 55 to 73 years Very low CHO diet 1. anti-tumor effects 1. shrinkage of tumor
(not further speci- 2. effect on cancer or stable disease was
fied) with a multi- related symptoms reported during the
tude of supplements, intervention
including amino 2. subjective improve-
acids and Vitamin ment reported in
D3 combined with some cases
SOC therapy
Duration: variable
Brünings [60] Case reports N = 14 Head and neck KD with as little 1. Anti-tumor effects 1. visible remission
CHO as possible 2. Adverse events after 2–3 weeks, but
(estimated < 50 g per rebound effect after
day), combined with 2–3 months on the
insulin administra- diet
tion 3 × per day 2. no adverse events
were reported
Brünings [61] Case reports N = 30 Extra-cranial KD with as little 1. Anti-tumor effects 1. Tumor shrinkage in
CHO as possible 2. QoL some cases
(estimated < 50 g per 2. Improvement in
day), combined with general condition and
insulin administra- positive effects on
tion 3 × per day clinical symptoms
Schütz [62] Case reports N = 23 Extra-cranial KD with as little 1. anti-tumor effect 1. no anti-tumor effects
CHO as possible 2. QoL found
(estimated < 50 g per 2. reduced pain sever-
day), combined with ity, but also fatigue
insulin administra- and deteriorated
tion 3 × per day orientation
KD ketogenic diet, CHO carbohydrate, TKTL 1 transketolase-like-1, RT radiotherapy, RCT radio-chemotherapy, QoL quality of live, FM fat mass, FFM fat free mass, MCT medium-chain tri-
glyceride, MKD modified ketogenic diet, GHS global health status, PFS progression-free survival, MAD modified Atkins diet, DLT dose-limiting toxicity, TPN total parenteral nutrition, ERKD
restricted ketogenic diets, FDG-PET [8F]-2-fluoro-2–deoxy-d-glucose positron emission tomography, HER2 human epidermal growth factor receptor 2
the KD group. It is also noteworthy that the follow-up time Finally, it should be noted that definite conclusions are
in this study’s Kaplan–Meier plot appears to be 26 months. still difficult to ascertain from the available data, due to a
However, the recruitment started in 07/2017 and stopped in high level of bias in most studies, a small number of patients
10/2018. The finished article was received by the publish- with high level of adherence and the lack of a control group
ing journal in 02/2019. This is just 4 months after the last and randomization, further increasing especially allocation,
patient was recruited. The resulting follow-up is 19 months and performance bias. It should also be noted that in several
at most for the first patients recruited. Furthermore, the KD studies the authors had a potential conflict of interest, due to
was only administered for 3 months. These inconsistencies financial and non-financial support or owning shares from
raise serious concerns regarding the presented data. Addi- companies producing products used in a KD [30, 35, 49,
tionally, no data were provided for the subgroup of meta- 54, 55].
static patients in this publication. Even though the data from Furthermore, the studies are highly heterogenous, in
these patients were not published as an original publication, many cases not limited to one cancer type and often use the
they are reported in a systematic review by Klement et al. KD complementary to other therapies, limiting the possibil-
[51]. Here, the patients in the KD group had a numerically ity to assess whether effects and AEs were caused by the diet
shorter OS (p = 0.078). or other simultaneous interventions—this also impairs the
The studies in this review showed an overall low adher- possibility to pool the results to perform a meta-analysis.
ence to the KD, but the drop-out rates varied greatly between
studies. Important reasons for low adherence were: limita- Limitations of this work
tions in monitoring and delivery [39], patients finding the
meals unpalatable [38] and problems trying to integrate the Some limitations of this systematic review must be men-
diet into family life [36]. tioned. For once, due to the heterogeneity of the included
QoL was only assessed in a few studies. The RCT RCTs no meta-analysis could be conducted, and no mod-
by Cohen et al. [32] was only able to show a significant erators of the effects caused by a KD could be determined.
improvement in perceived physical functioning after adjust- Furthermore, only studies published in English or German
ing for several variables and without adjusting for weight were included in this review.
loss, which attenuated the effect. No beneficial effects on
mental functioning were found in this trial. This is in line
with other studies, which also failed to show a QoL benefit Conclusion
of the KD [29, 36].
Almost all controlled and non-controlled studies showed Even though a variety of studies have been conducted in
a weight loss during the KD, which was often significant, if the past on KDs for cancer patients, evidence for increased
statistical analysis was performed [23, 24, 42]. This is rather survival, anti-tumor efficacy and a reduction of side effects
concerning, since malnourishment, sarcopenia and cancer is lacking, even in the most recent controlled trials. More
cachexia have been shown to negatively impact clinical out- robust and consistent clinical evidence from larger patient
comes and greatly reduce QoL [52, 53]. For patients with groups with comparable methodology, thorough dietary pro-
an increased risk of cancer cachexia, a KD can therefore tocols and an assessment of side effects using validated tools
be detrimental and the idea of implementing a KD in these are necessary, before a KD can be recommended to most
patients should raise serious safety concerns. Nevertheless, cancer patients. Currently possible side effects including
studies analyzing body composition revealed that the loss weight loss as well as patient co-morbidities must be care-
of fat mass appears to be more pronounced than the loss in fully weighed when considering applying a KD to cancer
fat-free mass [41, 42]. patients. To form a final judgment about the efficiency of a
The studies in this review showed a variety of adverse KD in Oncology, a randomized controlled trial with a well-
events related to a KD. The most frequent were fatigue [31, designed control group and sufficient power to also detect
45], constipation [29, 31], diarrhea [29, 35] as well as nausea evidence for absence of anti-tumor effects is necessary.
and vomiting [29, 35]. Despite the fact that most of these
were only mild to moderate several serious AEs like grade Supplementary Information The online version contains supplemen-
tary material available at https://d oi.o rg/1 0.1 007/s 10238-0 21-0 0710-2.
3 dehydration and grade 4 hyperuricemia [38] and a case of
grade 5 neutropenia occurred [26]. Especially problematic Authors’ contributions MR and JH contributed to development of the
is, that many studies did not measure AEs and the ones that protocol. MR and JD collected data and performed analysis. MR and
did, often attributed those that happened entirely to the SoC JD performed data interpretation. JH critically revised the article and
anti-cancer treatments [26]. Thus, the AEs of a KD seem to approved the final version submitted for publication.
be underreported.
13
Funding Open Access funding enabled and organized by Projekt 8. Hübner J, Marienfeld S, Abbenhardt C, Ulrich CM, Löser C.
DEAL.. The work of JD was funded in parts (search of the litera- How useful are diets against cancer? Dtsch Med Wochenschr.
ture, title–abstract screening) by the German Guideline “S3 Leitlinie 2012;137(47):2417–22. https://doi.org/10.1055/s-0032-1327276.
Komplementärmedizin in der Behandlung von onkologischen Patien- 9. Huebner J, Marienfeld S, Abbenhardt C, et al. Counseling patients
tInnen (Registernummer 032-055OL)” funded by the German Cancer on cancer diets: a review of the literature and recommendations
Aid (Fördernummer 11583) within the German Guideline Program for clinical practice. Anticancer Res. 2014;34(1):39–48.
in Oncology. 10. Poorshiri B, Barzegar M, Tahmasebi S, Shiva S, Raeisi S,
Ebadi Z. The efficacy comparison of classic ketogenic diet and
Availability of data and materials The datasets generated during and/or modified Atkins diet in children with refractory epilepsy: a
analyzed during the current study are available from the corresponding clinical trial. Acta Neurol Belg. 2019. https://doi.org/10.1007/
author on reasonable request. s13760-019-01225-0.
11. Klement RJ, Champ CE, Otto C, Kämmerer U. Anti-tumor
effects of ketogenic diets in mice: a meta-analysis. PLoS ONE.
Declarations 2016;11(5):e0155050. https://d oi.o rg/1 0.1 371/j ourna l.p one.0 1550
50.
Conflict of interest The authors declare no conflicts of interest. 12. Khodadadi S, Sobhani N, Mirshekar S, et al. Tumor cells growth
and survival time with the ketogenic diet in animal models: a
Ethics approval Not applicable. systematic review. Int J Prev Med. 2017;8:35. https://doi.org/10.
4103/2008-7802.207035.
Consent to participate Not applicable. 13. Sperry J, Condro MC, Guo L, et al. Glioblastoma utilizes fatty
acids and ketone bodies for growth allowing progression during
Consent for publication All authors consent to the publication of this ketogenic diet therapy. iScience. 2020;23(9):101453. https://doi.
work. org/10.1016/j.isci.2020.101453.
14. Kallinowski F, Vaupel P, Runkel S, et al. Glucose uptake, lactate
release, ketone body turnover, metabolic micromilieu, and pH dis-
Open Access This article is licensed under a Creative Commons Attri- tributions in human breast cancer xenografts in nude rats. Cancer
bution 4.0 International License, which permits use, sharing, adapta- Res. 1988;48(24 Pt 1):7264–72.
tion, distribution and reproduction in any medium or format, as long 15. Bonuccelli G, Tsirigos A, Whitaker-Menezes D, et al. Ketones
as you give appropriate credit to the original author(s) and the source, and lactate “fuel” tumor growth and metastasis: evidence that
provide a link to the Creative Commons licence, and indicate if changes epithelial cancer cells use oxidative mitochondrial metabolism.
were made. The images or other third party material in this article are Cell Cycle. 2010;9(17):3506–14. https://doi.org/10.4161/cc.9.17.
included in the article’s Creative Commons licence, unless indicated 12731.
otherwise in a credit line to the material. If material is not included in 16. Martinez-Outschoorn UE, Prisco M, Ertel A, et al. Ketones
the article’s Creative Commons licence and your intended use is not and lactate increase cancer cell “stemness,” driving recurrence,
permitted by statutory regulation or exceeds the permitted use, you will metastasis and poor clinical outcome in breast cancer: achiev-
need to obtain permission directly from the copyright holder. To view a ing personalized medicine via metabolo-genomics. Cell Cycle.
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 2011;10(8):1271–86. https://doi.org/10.4161/cc.10.8.15330.
17. (EPOC) CEPaOoC. What study designs can be considered for
inclusion in an EPOC review and what should they be called?
EPOC resources for review authors, 2020. 2020.
References 18. (SIGN). SIGN. Methodology checklist 2: controlled trials. 2014.
19. Sterne JAC, Savović J, Page MJ, et al. RoB 2. A revised tool for
1. Hickson M, Child J, Collinson A. Future dietitian 2025: informing assessing risk of bias in randomised trials. BMJ. 2019;366:14898.
the development of a workforce strategy for dietetics. J Hum Nutr https://doi.org/10.1136/bmj.l4898.
Diet. 2018;31(1):23–32. https://doi.org/10.1111/jhn.12509. 20. Bozzetti F, Gavazzi C, Mariani L, Crippa F. Glucose-based total
2. Maschke J, Kruk U, Kastrati K, et al. Nutritional care of cancer parenteral nutrition does not stimulate glucose uptake by humans
patients: a survey on patients’ needs and medical care in real- tumours. Clin Nutr. 2004;23(3):417–21. https://d oi.o rg/1 0.1 016/j.
ity. Int J Clin Oncol. 2017;22(1):200–6. https://doi.org/10.1007/ clnu.2003.09.012.
s10147-016-1025-6. 21. Artzi M, Liberman G, Vaisman N, et al. Changes in cerebral
3. Martin-McGill KJ, Bresnahan R, Levy RG, Cooper PN. Ketogenic metabolism during ketogenic diet in patients with primary brain
diets for drug-resistant epilepsy. Cochrane Database Syst Rev. tumors: (1)H-MRS study. J Neurooncol. 2017;132(2):267–75.
2020;6(6):CD001903. https://doi.org/10.1002/14651858.CD001 https://doi.org/10.1007/s11060-016-2364-x.
903.pub5. 22. Higgins JPT TJ, Chandler J, Cumpston M, Li T, Page MJ, Welch
4. Warburg O. On the origin of cancer cells. Science. VA, editors. Cochrane handbook for systematic reviews of inter-
1956;123(3191):309–14. https://doi.org/10.1126/science.123. ventions version 6.1 (updated September 2020). Cochrane, 2020.
3191.309. 2020.
5. Hanahan D, Weinberg RA. Hallmarks of cancer: the next gen- 23. Khodabakhshi A, Akbari ME, Mirzaei HR, Mehrad-Majd H, Kal-
eration. Cell. 2011;144(5):646–74. https://doi.org/10.1016/j.cell. amian M, Davoodi SH. Feasibility, safety, and beneficial effects
2011.02.013. of MCT-based ketogenic diet for breast cancer treatment: a ran-
6. Bettum IJ, Gorad SS, Barkovskaya A, et al. Metabolic reprogram- domized controlled trial study. Nutr Cancer. 2020;72(4):627–34.
ming supports the invasive phenotype in malignant melanoma. https://doi.org/10.1080/01635581.2019.1650942.
Cancer Lett. 2015;366(1):71–83. https://doi.org/10.1016/j.canlet. 24. Freedland SJ, Allen J, Jarman A, et al. A randomized controlled
2015.06.006. trial of a 6-month low-carbohydrate intervention on disease pro-
7. Allen BG, Bhatia SK, Anderson CM, et al. Ketogenic diets as an gression in men with recurrent prostate cancer: carbohydrate and
adjuvant cancer therapy: history and potential mechanism. Redox prostate study 2 (CAPS2). Clin Cancer Res. 2020;26(12):3035–
Biol. 2014;2:963–70. https://d oi.o rg/1 0.1 016/j.r edox.2 014.0 8.0 02. 43. https://doi.org/10.1158/1078-0432.Ccr-19-3873.
13
25. Strowd RE, Cervenka MC, Henry BJ, Kossoff EH, Hartman AL, an IRB-approved energy-restricted ketogenic diet protocol and
Blakeley JO. Glycemic modulation in neuro-oncology: experience review of the literature. Cancer Metab. 2015;3:3. https://doi.org/
and future directions using a modified Atkins diet for high-grade 10.1186/s40170-015-0129-1.
brain tumors. Neurooncol Pract. 2015;2(3):127–36. https://doi. 40. Nebeling LC, Miraldi F, Shurin SB, Lerner E. Effects of a
org/10.1093/nop/npv010. ketogenic diet on tumor metabolism and nutritional status in
26. Iyikesici MS. Feasibility study of metabolically supported pediatric oncology patients: two case reports. J Am Coll Nutr.
chemotherapy with weekly carboplatin/paclitaxel combined with 1995;14(2):202–8. https://d oi.o rg/1 0.1 080/0 73157 24.1 995.1 0718
ketogenic diet, hyperthermia and hyperbaric oxygen therapy 495.
in metastatic non-small cell lung cancer. Int J Hyperthermia. 41. Klement RJ, Sweeney RA. Impact of a ketogenic diet intervention
2019;36(1):446–55. https://doi.org/10.1080/02656736.2019. during radiotherapy on body composition: I. Initial clinical expe-
1589584. rience with six prospectively studied patients. BMC Res Notes.
27. Iyikesici MS. Long-term survival outcomes of metabolically sup- 2016;9:143. https://doi.org/10.1186/s13104-016-1959-9.
ported chemotherapy with gemcitabine-based or FOLFIRINOX 42. Cohen CW, Fontaine KR, Arend RC, et al. A ketogenic diet
regimen combined with ketogenic diet, hyperthermia, and hyper- reduces central obesity and serum insulin in women with ovarian
baric oxygen therapy in metastatic pancreatic cancer. Complement or endometrial cancer. J Nutr. 2018;148(8):1253–60. https://doi.
Med Res. 2020;27(1):31–9. https://doi.org/10.1159/000502135. org/10.1093/jn/nxy119.
28. Woodhouse C, Ward T, Gaskill-Shipley M, Chaudhary R. Fea- 43. Rossi-Fanelli F, Franchi F, Mulieri M, et al. Effect of energy sub-
sibility of a modified Atkins diet in glioma patients during strate manipulation on tumour cell proliferation in parenterally fed
radiation and its effect on radiation sensitization. Curr Oncol. cancer patients. Clin Nutr. 1991;10(4):228–32. https://doi.org/10.
2019;26(4):e433–8. https://doi.org/10.3747/co.26.4889. 1016/0261-5614(91)90043-c.
29. van der Louw E, Olieman JF, van den Bemt P, et al. Ketogenic 44. Fearon KC, Borland W, Preston T, Tisdale MJ, Shenkin A, Calman
diet treatment as adjuvant to standard treatment of glioblastoma KC. Cancer cachexia: influence of systemic ketosis on substrate
multiforme: a feasibility and safety study. Ther Adv Med Oncol. levels and nitrogen metabolism. Am J Clin Nutr. 1988;47(1):42–8.
2019;11:1758835919853958. https://d oi.o rg/1 0.1 177/1 75883 5919 https://doi.org/10.1093/ajcn/47.1.42.
853958. 45. Champ CE, Palmer JD, Volek JS, et al. Targeting metabolism
30. Rieger J, Bähr O, Maurer GD, et al. ERGO: a pilot study with a ketogenic diet during the treatment of glioblastoma mul-
of ketogenic diet in recurrent glioblastoma. Int J Oncol. tiforme. J Neurooncol. 2014;117(1):125–31. https://doi.org/10.
2014;44(6):1843–52. https://doi.org/10.3892/ijo.2014.2382. 1007/s11060-014-1362-0.
31. Fine EJ, Segal-Isaacson CJ, Feinman RD, et al. Targeting insu- 46. Zuccoli G, Marcello N, Pisanello A, et al. Metabolic manage-
lin inhibition as a metabolic therapy in advanced cancer: a pilot ment of glioblastoma multiforme using standard therapy together
safety and feasibility dietary trial in 10 patients. Nutrition. with a restricted ketogenic diet: case report. Nutr Metab (Lond).
2012;28(10):1028–35. https://doi.org/10.1016/j.nut.2012.05.001. 2010;7:33. https://doi.org/10.1186/1743-7075-7-33.
32. Cohen CW, Fontaine KR, Arend RC, Soleymani T, Gower BA. 47. Tóth C, Clemens Z. Treatment of rectal cancer with the paleo-
Favorable effects of a ketogenic diet on physical function, per- lithic ketogenic diet: a 24-months follow-up. Am J Med Case Rep.
ceived energy, and food cravings in women with ovarian or endo- 2017;5(8):205–16.
metrial cancer: a randomized, controlled trial. Nutrients. 2018. 48. Bozzetti F, Cozzaglio L, Gavazzi C, et al. Total nutritional
https://doi.org/10.3390/nu10091187. manipulation in humans: report of a cancer patient. Clin Nutr.
33. Klement RJ, Schäfer G, Sweeney RA. A ketogenic diet exerts 1996;15(4):207–9. https:// d oi. o rg/ 1 0. 1 016/ s 0261- 5 614(96)
beneficial effects on body composition of cancer patients during 80243-3.
radiotherapy: an interim analysis of the KETOCOMP study. J 49. Martin-McGill KJ, Marson AG, Tudur Smith C, Jenkinson MD.
Tradit Complement Med. 2020;10(3):180–7. https://doi.org/10. The Modified ketogenic diet in adults with glioblastoma: an evalu-
1016/j.jtcme.2019.03.007. ation of feasibility and deliverability within the national health
34. Ok JH, Lee H, Chung HY, et al. The potential use of a ketogenic service. Nutr Cancer. 2018;70(4):643–9. https://doi.org/10.1080/
diet in pancreatobiliary cancer patients after pancreatectomy. 01635581.2018.1460677.
Anticancer Res. 2018;38(11):6519–27. https://doi.org/10.21873/ 50. Moore K. Using the restricted ketogenic diet for brain cancer
anticanres.13017. management: comments from neuro-oncologist. In: Seyfried TN,
35. Martin-McGill K, Cherry G, Marson A, Smith CT, Jenkinson M. editor. Cancer as a metabolic disease: on the origin, management,
Ketogenic diets as an adjuvant therapy in glioblastoma (KEAT- and prevention of cancer. Hoboken: Wiley; 2012. p. 397–400.
ING): a mixed method approach to assessing trial feasibility. 51. Klement R, Brehm N, Sweeney R. Ketogenic diets in medi-
Neuro Oncol. 2018. https://doi.org/10.1093/neuonc/noy148. cal oncology: a systematic review with focus on clinical out-
36. Schmidt M, Pfetzer N, Schwab M, Strauss I, Kammerer U. comes. Med Oncol. 2020;37:14. https:// d oi. o rg/ 1 0. 1 007/
Effects of a ketogenic diet on the quality of life in 16 patients s12032-020-1337-2.
with advanced cancer: a pilot trial. Nutr Metab. 2011;8(1):54. 52. Ryan AM, Power DG, Daly L, Cushen SJ, Ní Bhuachalla Ē, Prado
https://doi.org/10.1186/1743-7075-8-54. CM. Cancer-associated malnutrition, cachexia and sarcopenia:
37. Tan-Shalaby JL, Carrick J, Edinger K, et al. Modified Atkins diet the skeleton in the hospital closet 40 years later. Proc Nutr Soc.
in advanced malignancies—final results of a safety and feasibil- 2016;75(2):199–211. https://doi.org/10.1017/s00296651150041
ity trial within the Veterans Affairs Pittsburgh Healthcare Sys- 9x.
tem. Nutr Metab (Lond). 2016;13:52. https://doi.org/10.1186/ 53. Bruggeman AR, Kamal AH, LeBlanc TW, Ma JD, Baracos VE,
s12986-016-0113-y. Roeland EJ. Cancer cachexia: beyond weight loss. J Oncol Pract.
38. Zahra A, Fath MA, Opat E, et al. Consuming a ketogenic diet 2016;12(11):1163–71. https://doi.org/10.1200/jop.2016.016832.
while receiving radiation and chemotherapy for locally advanced 54. Jansen N, Walach H. The development of tumours under a
lung cancer and pancreatic cancer: the University of Iowa experi- ketogenic diet in association with the novel tumour marker
ence of two phase 1 clinical trials. Radiat Res. 2017;187(6):743– TKTL1: a case series in general practice. Oncol Lett.
54. https://doi.org/10.1667/r r14668.1. 2016;11(1):584–92. https://doi.org/10.3892/ol.2015.3923.
39. Schwartz K, Chang HT, Nikolai M, et al. Treatment of glioma 55. Schwalb M, Taubmann M, Hines S, Reinwald H, Ruggi-
patients with ketogenic diets: report of two cases treated with ero M. Clinical observation of a novel, complementary,
13
immunotherapeutic approach based on ketogenic diet, chondroitin metabolic therapy and modified standard of care: a 24-month
sulfate, vitamin D3, oleic acid and a fermented milk and colos- follow-up. Front Nutr. 2018;5:20. https://doi.org/10.3389/fnut.
trum product. Am J Immunol. 2016;12:91–8. https://doi.org/10. 2018.00020.
3844/ajisp.2016.91.98. 60. Brünings W. Beitraege zum Krebsproblem. 1. Mitteilung:
56. Cohen CW, Fontaine KR, Arend RC, Gower BA. A ketogenic diet Ueber eine diaetetisch-hormonale Beeinflussung des Krebses.
is acceptable in women with ovarian and endometrial cancer and Muenchener Medizinische Wochenschrift. 1941;117–23.
has no adverse effects on blood lipids: a randomized. Control Trial 61. Brünings W. Beitraege zum Krebsproblem. 2. Mitteilung: Kli-
Nutr Cancer. 2019. https://d oi.o rg/1 0.1 080/0 16355 81.2 019.1 6458 nische Anwendungen der diaetetisch-hormonalen Krebsbeein-
64. flussung (“Entzuckerungsmethode”). Muenchener Medizinische
57. Branca JJ, Pacini S, Ruggiero M. Effects of pre-surgical vitamin Wochenschrift. 1942;71–6.
D supplementation and ketogenic diet in a patient with recurrent 62. Schulte G, Schütz H. Insulin in der Krebsbehandlung. Muenchener
breast cancer. Anticancer Res. 2015;35(10):5525–32. Medizinische Wochenschrift. 1942;648–50.
58. Schroeder U, Himpe B, Pries R, Vonthein R, Nitsch S, Wollenberg
B. Decline of lactate in tumor tissue after ketogenic diet: in vivo Publisher’s Note Springer Nature remains neutral with regard to
microdialysis study in patients with head and neck cancer. Nutr jurisdictional claims in published maps and institutional affiliations.
Cancer. 2013;65(6):843–9. https://doi.org/10.1080/01635581.
2013.804579.
59. Elsakka AMA, Bary MA, Abdelzaher E, et al. Management
of glioblastoma multiforme in a patient treated with ketogenic
13

The Use of Ketogenic Diets in Cancer Patients: A Systematic Review

Uploaded by

Copyright:

Available Formats

The Use of Ketogenic Diets in Cancer Patients: A Systematic Review

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Use of Ketogenic Diets in Cancer Patients: A Systematic Review

Uploaded by

Copyright:

Available Formats

Clinical and Experimental Medicine (2021) 21:501–536

The use of ketogenic diets in cancer patients: a systematic review

Keywords Humans · Metabolism · Ketogenic diet · Low-carbohydrate diet · Carbohydrate-restricted · Cancer

Table 1 Inclusion and exclusion criteria based on a PICO model

Fig. 1 Preferred reporting

Records after duplicates removed

Full-text articles assessed Full-text articles excluded,

Risk of bias and methodological quality Data extraction

No studies were suitable for a pooled analysis; hence, only

Results based on the Oxford criteria. These results and additional

Characteristics of included studies Results from RCTs and CTs

Branca et al. [57] Single case report – PRO: 4

Moore [50] Single case report – PRO: 4

References Study type N Cancer site Age Intervention/duration Endpoints Outcomes

28.44 kg/m2, Δ from baseline-0.64 kg/

References Study type N Cancer site Age Intervention/duration Endpoints Outcomes

Regression coefficient for fat mass

References Study type N Cancer site Age Intervention/duration Endpoints Outcomes

Changes in body composition Adverse events were monitored in 19 studies. A validated

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

consistency of an CTCAE grade 2:

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

[47] nutrients consumed 2. bodyweight were reported

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

versus protein plus tumor, despite a new

Reference Study type N Cancer site Age Intervention/duration Endpoints Outcomes

You might also like