Abstract

There has been growing interest in the potential of the altered metabolic state typical of cancer cells as a drug target. The antidiabetes drug, metformin, is now under intense investigation as a safe method to modify cancer metabolism. Several studies have used window of opportunity in breast cancer patients before neoadjuvant chemotherapy to correlate gene expression analysis, metabolomics, immunohistochemical markers, and metabolic serum markers with those likely to benefit. We review the role metabolite measurement, functional imaging and gene sequencing analysis play in elucidating the effects of metabolically targeted drugs in cancer treatment and determining patient selection.

Metabolism as a Target for Cancer Therapy

Since Warburg (1) noted that cancer cells have a higher rate of aerobic glucose consumption it has been understood that cancer cells have a distinct metabolic phenotype. Functional imaging techniques, such as fluorodeoxyglucose (18F-FDG) Positron emission tomography–computed tomography (PET-CT), already make use of altered metabolism in cancer cells for clinical imaging, but there is a growing realization that altered metabolism also offers an exciting drug target. Many in vitro and in vivo studies have shown that the targeting of enzymes involved in metabolism can halt the proliferation of cancer cells or lead to apoptotic cell death. For example, recently it has become clear that glutamine is an especially important substrate for fatty acid synthesis and the replenishment of Krebs cycle intermediates. The targeting of key enzymes involved in glutamine metabolism, such as isocitrate dehydrogenase or glutaminase, with small molecule inhibitors may be an effective therapeutic strategy (2). It is also possible to exploit the Warburg effect itself. Dichloroacetate stimulates the activity of the enzyme pyruvate dehydrogenase leading to an increase in pyruvate uptake into the mitochondria and enhanced glucose oxidation. “Mitochondrial reactivation” in this way is associated with increased apoptosis and reduced tumor xenograft growth (3). The dysregulated lipid metabolism in cancer cells is another potential therapeutic target. The inhibition of fatty acid synthase a key enzyme involved in fatty acid synthesis, a process typically upregulated in cancer cells, leads to selective cancer cell death. Similarly xenograft cancer models when treated with the fatty acid synthase inhibitor C75 showed evidence of fatty acid synthesis inhibition, apoptosis, and inhibition of tumor growth (4).

Metformin—an Example of a “Window of Opportunity” Biomarker Study for a Metabolic Drug

There is currently great interest in the potential of the antidiabetes drug, metformin, as a cancer therapy. This was initially driven by a series of epidemiological studies which suggested that diabetic patients taking metformin had a reduced cancer incidence compared with their counterparts with diabetes using other therapies. A meta-analysis suggested a 31% reduction in overall summary relative risk (SRR = 0.69; 95% confidence interval [CI] = 0.61 to 0.71) for all cancer types. Some studies have suggested an especially large risk reduction for hepatocellular and pancreatic cancer (60%–80%) while a meta-analysis showed there may a lesser effect in colon (SRR = 0.64; 95% CI = 0.38 to 1.08) and breast cancer (SRR = 0.70; 95% CI = 0.28 to 1.77) (5–9). A subsequent retrospective study of diabetic patients receiving neoadjuvant chemotherapy for breast cancer suggested that the pathological complete response rate for patients taking metformin was greater (24% in the metformin group vs 8% in the nonmetformin group, P = .007) when compared with patients on other diabetes drugs (10). A series of in vitro and in vivo studies have reinforced the potential of metformin as an anticancer drug. For example, in vitro and in vivo work using breast cancer cell line models has shown synergy between chemotherapy and metformin and evidence of metformin selectively targeting cancer stem cells (11). Anticancer effects have also been seen in a variety of other cancer cell types, including prostate, glioma, and endometrial cancer cell lines (12–15).

Despite intense investigation it is still not clear as to the exact mechanism of metformin’s anticancer effect and whether the drug acts directly on cancer cells (Figure 1). Direct cellular effects may be mediated by AMPK, a protein kinase central to energy homeostasis in the cell. Metformin is a mitochondrial toxin and inhibits complex 1 of the mitochondrial respiratory chain leading to a cellular energy stress (16). Activation of AMPK, after sensing increases in the ratios of adenosine monophosphate (AMP)/adenosine triphosphate (ATP) and adenosine diphosphate (ADP)/adenosine triphosphate (ATP), downregulates anabolic pathways, such as mammalian target of rapamycin (mTOR) and fatty acid synthesis, and upregulates catabolic pathways such as glycolysis and fatty acid oxidation, leading to a halt in cancer cell proliferation. Alternatively, metformin may exert an indirect anticancer effect via its influence on the insulin axis. It is known that obesity and type 2 diabetes are associated with cancer risk and upregulated insulin and insulin-like growth factor 1 signaling may promote tumor growth (17). AMPK activation by metformin in the liver inhibits hepatic gluconeogenesis reducing serum glucose levels and subsequently circulating insulin. This effect is likely to be especially prominent in obese or diabetic individuals (18).

Figure 1.

There are two hypotheses as to the mechanism behind metformin’s anticancer effect. Upper panel, the direct effect in which metformin generates an energy stress in tumor cells via its inhibition of complex 1 and subsequent activation of AMPK leads to catabolic downstream effects and inhibition of cancer cell proliferation. Lower panel, the indirect effect in which AMPK activation in the liver results in reduced circulating insulin levels and subsequent inhibition of the IGF1/phosphoinositide 3-kinase (PI3K) axis in tumor cells. ACC1 = acetyl CoA carboxylase 1; FAS = fatty acid synthase; IGF1 = insulin growth factor 1; IR = insulin receptor.

Figure 1.

There are two hypotheses as to the mechanism behind metformin’s anticancer effect. Upper panel, the direct effect in which metformin generates an energy stress in tumor cells via its inhibition of complex 1 and subsequent activation of AMPK leads to catabolic downstream effects and inhibition of cancer cell proliferation. Lower panel, the indirect effect in which AMPK activation in the liver results in reduced circulating insulin levels and subsequent inhibition of the IGF1/phosphoinositide 3-kinase (PI3K) axis in tumor cells. ACC1 = acetyl CoA carboxylase 1; FAS = fatty acid synthase; IGF1 = insulin growth factor 1; IR = insulin receptor.

Breast cancer is a heterogenous disease and this is reflected in the varying response to targeted anticancer drugs between patients, for example, for estrogen receptor and HER2 expression. There are now approximately 50 clinical trials worldwide investigating the clinical effects of metformin in the cancer setting (19). However, at an early stage of metformin’s development as a cancer drug it is important that biomarkers are developed to select patients who may be suitable for metformin treatment and inform the design and interpretation of phase III trials. The history of cancer drug development is littered with examples of promising treatments tested in unselected patient populations in which clinical benefit could not be demonstrated. Already one phase III adjuvant study of metformin has completed recruitment in an unselected breast cancer population before satisfactory biomarker development (19). Our group’s recent detailed pharmacodynamic study of the early response of primary breast cancers to bevacizumab, in which dynamic magnetic resonance imaging was correlated with gene expression profiles has attempted to address this for anti–vascular endothelial growth factor therapy (20) and demonstrated the marked short-term heterogeneity of response.

The results of several clinical biomarker studies of metformin in the breast cancer setting have recently been published or were reported at the 2012 San Antonio Breast Cancer Symposium (Table 1) (21–25). Without exception these studies took place in a 2–4-week window between breast cancer diagnosis and surgery with the collection of tumor samples comprising of a premetformin biopsy sample and postmetformin surgical sample after an intervening short course of metformin. Diabetic patients were excluded from these trials and one study was enriched for obese patients, but otherwise the breast cancer population was unselected. All of these trials have focused on metabolic serum markers and changes in intratumoral proliferation (Ki67) or apoptosis (terminal deoxynucleotidyl transferase dUTP nick end labeling [TUNEL]) using immunohistochemistry. Results to date from these studies suggest that metformin may inhibit both cancer cell proliferation, measured by Ki67, and reduce insulin resistance in this nondiabetic population, as measured by the homeostasis model assessment index. The largest study to date did not demonstrate a clear relationship between estrogen or progesterone receptor status and Ki67 (21). In addition, one limited study examined for changes in gene expression in a small cohort of patients (see below). Of note, no published study has used functional imaging to determine metabolic response to metformin or assessed for intratumoral changes in metabolite concentrations.

Table 1.

Neoadjuvant studies of potential metformin biomarkers of effect in breast cancer patients*

Study Enrollment and design Population Effect of metformin on serum metabolic markers Effect of metformin on immunohistochemical markers 
Bonanni et al. (21) and Cazzaniga et al. (22) 200
Metformin (n = 100), placebo (n = 100) 
Untreated breast cancer
Nondiabetic 
Decrease in CRP (−0.8mg/L, P = .029) and total cholesterol (−9.4mg/dL, P = .001)
Decreased glucose (−2.8mg/dL, P = .15) and trend toward reduced insulin (−0.2 mU/L, P = .92) if BMI > 27 
10.5% decrease in Ki67 if HOMA > 2.8 (P = .045)
No effect on TUNEL 
Hadad et al. (23) 55
Metformin (n = 47), placebo (n = 8) 
Untreated breast cancer
Nondiabetic 
No significant change in serum insulin—note patients received intravenous glucose before postmetformin blood sampling 5.1% decrease in Ki67 (P = .041), decrease in pAKT (P = .04), and increase in pAMPK (P = .04) 
Kalinsky et al. (24) 35
No control arm 
Untreated breast cancer
Nondiabetic
Enriched for obese women 
Significant decrease in total cholesterol (−16.4mg/dL, P < .01) and leptin (−2ng/mL, P = .03)
Trend toward decrease in insulin (−4.1 uIU/mL, P = .06), HOMA (−1.0, P = .06), and adiponectin (−388ng/ dL, P = .06)
No significant change in glucose or IGF-BP3 
 
Niraula et al. (25) 39
No control arm 
Untreated breast cancer
Nondiabetic 
Decrease in glucose (−0.14 mmol/L, P = .045) and HOMA (−0.21, P = .047)
No significant change in insulin, leptin, or CRP 
3.0% decrease in Ki67 (P = .016) and 0.5% increase in TUNEL (P = .004) 
Study Enrollment and design Population Effect of metformin on serum metabolic markers Effect of metformin on immunohistochemical markers 
Bonanni et al. (21) and Cazzaniga et al. (22) 200
Metformin (n = 100), placebo (n = 100) 
Untreated breast cancer
Nondiabetic 
Decrease in CRP (−0.8mg/L, P = .029) and total cholesterol (−9.4mg/dL, P = .001)
Decreased glucose (−2.8mg/dL, P = .15) and trend toward reduced insulin (−0.2 mU/L, P = .92) if BMI > 27 
10.5% decrease in Ki67 if HOMA > 2.8 (P = .045)
No effect on TUNEL 
Hadad et al. (23) 55
Metformin (n = 47), placebo (n = 8) 
Untreated breast cancer
Nondiabetic 
No significant change in serum insulin—note patients received intravenous glucose before postmetformin blood sampling 5.1% decrease in Ki67 (P = .041), decrease in pAKT (P = .04), and increase in pAMPK (P = .04) 
Kalinsky et al. (24) 35
No control arm 
Untreated breast cancer
Nondiabetic
Enriched for obese women 
Significant decrease in total cholesterol (−16.4mg/dL, P < .01) and leptin (−2ng/mL, P = .03)
Trend toward decrease in insulin (−4.1 uIU/mL, P = .06), HOMA (−1.0, P = .06), and adiponectin (−388ng/ dL, P = .06)
No significant change in glucose or IGF-BP3 
 
Niraula et al. (25) 39
No control arm 
Untreated breast cancer
Nondiabetic 
Decrease in glucose (−0.14 mmol/L, P = .045) and HOMA (−0.21, P = .047)
No significant change in insulin, leptin, or CRP 
3.0% decrease in Ki67 (P = .016) and 0.5% increase in TUNEL (P = .004) 

*BMI = body mass index; CRP = C-reactive protein; HOMA = homeostasis model assessment index (an estimate of insulin resistance, HOMA formula: fasting blood glucose [mmol] × insulin [mU/L]/22.5); IGF-BP3 = insulin-like growth factor binding protein-3; TUNEL = terminal deoxynucleotidyl transferase dUTP nick end labeling.

Study Design

In our study 40 patients with primary breast cancer have received a 2-week course of metformin before commencing neoadjuvant chemotherapy. Patients undergo dynamic 18F-FDG PET-CT scans, breast core biopsies (for gene profiling, metabolomics, and immunohistochemistry), and fasting blood samples for pharmacodynamic assessment pre- and postmetformin (Figure 2). In contrast to the studies discussed above both the pre- and postmetformin tumor samples were collected via ultrasound-guided core biopsy (rather than the second being from a surgical specimen). Care has been taken to ensure that the same area is sampled on each occasion and secondly that the biopsy is taken from the (less necrotic) rim of the tumor.

Figure 2.

Schema of window of opportunity study to assess for biomarkers of metformin’s metabolic effect in breast cancer. Patients will have a baseline and postmetformin 18F-FDG PET-CT scan, three breast tumor biopsies for mRNA sequencing, metabolomics profiling and immunohistochemistry, and serum blood samples for serum metabolic markers. 18F-FDG = fluorodeoxyglucose; mRNA = messenger RNA; PET-CT = positron emission tomography–computed tomography; XR = extended release formulation.

Figure 2.

Schema of window of opportunity study to assess for biomarkers of metformin’s metabolic effect in breast cancer. Patients will have a baseline and postmetformin 18F-FDG PET-CT scan, three breast tumor biopsies for mRNA sequencing, metabolomics profiling and immunohistochemistry, and serum blood samples for serum metabolic markers. 18F-FDG = fluorodeoxyglucose; mRNA = messenger RNA; PET-CT = positron emission tomography–computed tomography; XR = extended release formulation.

Dynamic 18-FDG PET-CT Scans

Functional imaging is already used in the determination of early drug response for cancer agents both within the clinical trial setting and routine clinical care (26). With the advent of an era of cancer treatments that specifically target metabolism, it can be expected that these techniques will become more critical for this purpose. In vitro investigation using cancer cell lines has demonstrated that a metformin-induced energy stress leads to upregulated glycolysis and increased glucose uptake (2). A recent xenograft model, using colon cancer cell lines, demonstrated increased intratumoral 18F-FDG uptake 24 hours after intraperitoneal injection of metformin (27). Hence, 18F-FDG was chosen as the most suitable tracer to define a metabolic response for metformin, the advantage also being that it is already widely available in the clinic.

Our initial analysis will focus on changes in the mean at a 50% threshold and maximum standardized uptake values (SUVmean50% and SUVmax, respectively) as measures of 18F-FDG uptake between 50 and 60 minutes postinjection. The SUV is a ratio of the local tracer accumulation to the injected activity per body volume. It is equal to one when there is homogenous distribution of tracer in tissue and is greater than one if there is accumulation in tissue (28). However, SUV at a single late timepoint cannot differentiate between unmetabolized 18F-FDG in the vessels, tumor interstium and cells, and intracellular phosphorylated 18F-FDG (ie, different tracer compartments). Hence we also plan to collect dynamic PET-CT data continuously from 0 to 60 minutes postinjection, from which the exchange of tracer between blood and tissue over time will be evaluated using two and three compartment models. An input function is required to determine the kinetics rate constants of these models, and in this study the 18F-FDG left ventricular blood time-activity-curve will be used (29). The assessment of tracer kinetics in this way aims to separate the impact of tumor perfusion, cell membrane transport, and intracellular metabolism on 18F-FDG tumor uptake—potentially allowing the individual effects of metformin on these distinct processes to be studied.

Correlation of Functional Imaging With Gene Expression Profiling

The determination of a metabolic response to metformin using functional imaging can then be correlated with changes in the other pharmacodynamic assays in this study to screen for further potential biomarkers. We plan to use messenger RNA sequencing of breast cancer biopsy samples to identify transcriptional changes in response to metformin. However, the transcriptional response to any treatment is likely to be highly variable between tumor samples. We and other groups have previously used pathway analysis of genes associated with particular processes or tumor environments (20,30,31). A previous clinical study examining gene expression changes of 14 paired pre- and postmetformin primary breast cancer samples using RNA microarray analysis found 34 gene sets to be overexpressed and 60 to be underexpressed after metformin treatment. In particular this study identified both upregulation in the tumor necrosis factor receptor 1 pathway and underexpression of p53 signaling, suggestive of an effect of metformin on key pathways known to regulate tumor cell proliferation, invasion, and metastasis (23).

Other biomarkers may reflect changes in the expression of specific genes associated with resistance and sensitivity to metformin. For example, in the above RNA microarray study there was prominent downregulation of phosphodiesterase 3B, a known regulator of the lipolytic and glycogenlytic effects of insulin (23).

AMPK activation has previously been shown to upregulate glucose transporter expression and GLUT1 expression correlates with primary breast cancer uptake of 18F-FDG on PET-CT (32,33). The glutamine reductive carboxylation pathway that allows metformin-treated cancer cells to utilize glutamine as a carbon source at a time of mitochondrial toxicity and changes in expression of its regulators such as isocitrate dehydrogenase will be of interest (34). Nonsynonymous polymorphisms of OCT1, a cell surface transporter associated with metformin uptake, have been associated with metformin resistance (35). Additionally, mutations in the mitochondrial genome and in particular complex 1 genes, may be also be key markers of resistance.

Mass Spectrometry

Mass spectrometry has long been used as a tool for biomarker discovery but mainly in carrying out proteomic screens. The use of stable isotope labeling for in vitro studies in conjunction with nuclear magnetic resonance and mass spectrometry has been vital to a growing understanding of the intricacies of cancer cell substrate utilization for key processes vital to proliferation, such as fatty acid and amino acid synthesis. Similarly, clinical studies using these techniques have also proved invaluable in investigating the altered metabolism associated with diabetes and obesity (36–38). There is an important place for this methodology to study drugs targeting cancer metabolism. Study participants can be given a stable isotope, for example, deuterated water or (U-13C) palmitic acid, the evening before each biopsy to allow tracing through the fatty acid synthesis pathway.

Immunohistochemistry

Most studies will use a formalin-fixed sample for immunohistochemical analysis. Initial molecular markers of interest will aim to assess metformin’s effect on AMPK and mTOR signaling (pAMPK, pS6, and pEIF4-BP1) and proliferation using the validated proliferation marker Ki67. Given that these are predominantly phosphoproteins, rigorous sample handling is vital with immediate fixation in formalin and rapid processing built into the study design (39).

Serum Metabolic Markers

Serum metabolic markers have been assayed, in particular those relevant to the insulin axis (glucose, insulin, c-peptide, insulin-like growth factor 1, and insulin-like growth factor binding protein-3) and lipid metabolism (high-density lipoprotein and low-density lipoprotein cholesterol, and triglycerides). The baseline metabolic parameters in conjunction with the assessment of the body mass index of participants may also help to define whether the metabolic/anticancer effect of metformin is likely to be most relevant to cancer patients with features of the metabolic syndrome. Other studies have also assessed for changes in these and other serum metabolic markers after metformin treatment in cancer patients (see Table 1) and correlation with immunohistochemical markers, in particular Ki67, has suggested a relationship between the metabolic syndrome and metformin’s effect in breast cancer (21). The clinical effects of metformin in the cancer setting have thus far largely been reported in a diabetic and often obese population and this study should further define any correlation of intratumoral metabolic effects (18F-FDG uptake and metabolomic profiling with mass spectrometry) with metformin’s systemic effects, in particular on the insulin axis. This would go some way toward answering the question as to whether metformin generates its anticancer effects directly or indirectly (Figure 1) and whether it is likely to have utility in a nondiabetic and/or nonobese cancer population.

Conclusions

Biomarker discovery is vital to cancer drug development and this also applies to a new era of treatments targeting metabolism. However, the pharmacodynamic techniques used and markers assayed for may be very different compared with that for traditional cancer drug discovery. Metformin is the first such drug of this class to undergo extensive clinical evaluation and in vitro work and animal models have already suggested several possible markers of resistance and sensitivity. We expect the results of studies incorporating the above features will be vital in informing future large-scale clinical trial design, the potential applications of metformin in the cancer setting, and cancer metabolism drug development in general.

Funding

This work is supported by funding from the Oxford Cancer Imaging Centre, the Oxford National Institute for Health Biomedical Research Centre, and the Breast Cancer Research Foundation.

References

1.
Warburg
O
.
On the origin of cancer cells
.
Science
 .
1956
;
123
(
3191
):
309
314
.
2.
Fendt
SM
Bell
EL
Keibler
MA
et al
Metformin decreases glucose oxidation and increases the dependency of prostate cancer cells on reductive glutamine metabolism
.
Cancer Res
 .
2013
;
73
(
14
):
4429
4438
.
3.
Michelakis
ED
Webster
L
Mackey
JR
.
Dichloroacetate (DCA) as a potential metabolic-targeting therapy for cancer
.
Br J Cancer
 .
2008
;
99
(
7
):
989
994
.
4.
Kuhajda
FP
.
Fatty acid synthase and cancer: new application of an old pathway
.
Cancer Res
 .
2006
;
66
(
12
):
5977
5980
.
5.
Decensi
A
Puntoni
M
Goodwin
P
et al
Metformin and cancer risk in diabetic patients: a systematic review and meta-analysis
.
Cancer Prev Res (Phila)
 .
2010
;
3
(
11
):
1451
1461
.
6.
Currie
CJ
Poole
CD
Gale
EA
.
The influence of glucose-lowering therapies on cancer risk in type 2 diabetes
.
Diabetologia
 .
2009
;
52
(
9
):
1766
1777
.
7.
Li
D
Yeung
SC
Hassan
MM
Konopleva
M
Abbruzzese
JL
.
Antidiabetic therapies affect risk of pancreatic cancer
.
Gastroenterology
 .
2009
;
137
(
2
):
482
488
.
8.
Hassan
MM
Curley
SA
Li
D
et al
Association of diabetes duration and diabetes treatment with the risk of hepatocellular carcinoma
.
Cancer
 .
2010
;
116
(
8
):
1938
1946
.
9.
Donadon
V
Balbi
M
Ghersetti
M
et al
Antidiabetic therapy and increased risk of hepatocellular carcinoma in chronic liver disease
.
World J Gastroenterol
 .
2009
;
15
(
20
):
2506
2511
.
10.
Jiralerspong
S
Palla
SL
Giordano
SH
et al
Metformin and pathologic complete responses to neoadjuvant chemotherapy in diabetic patients with breast cancer
.
J Clin Oncol
 .
2009
;
27
(
20
):
3297
3302
.
11.
Hirsch
HA
Iliopoulos
D
Tsichlis
PN
Struhl
K
.
Metformin selectively targets cancer stem cells, and acts together with chemotherapy to block tumor growth and prolong remission
.
Cancer Res
 .
2009
;
69
(
19
):
7507
7511
.
12.
Ben Sahra
I
Laurent
K
Loubat
A
et al
The antidiabetic drug metformin exerts an antitumoral effect in vitro and in vivo through a decrease of cyclin D1 level
.
Oncogene
 .
2008
;
27
(
25
):
3576
3586
.
13.
Gotlieb
WH
Saumet
J
Beauchamp
MC
et al
In vitro metformin anti-neoplastic activity in epithelial ovarian cancer
.
Gynecol Oncol
 .
2008
;
110
(
2
):
246
250
.
14.
Cantrell
LA
Zhou
C
Mendivil
A
Malloy
KM
Gehrig
PA
Bae-Jump
VL
.
Metformin is a potent inhibitor of endometrial cancer cell proliferation–implications for a novel treatment strategy
.
Gynecol Oncol
 .
2010
;
116
(
1
):
92
98
.
15.
Isakovic
A
Harhaji
L
Stevanovic
D
et al
Dual antiglioma action of metformin: cell cycle arrest and mitochondria-dependent apoptosis
.
Cell Mol Life Sci
 .
2007
;
64
(
10
):
1290
1302
.
16.
El-Mir
MY
Nogueira
V
Fontaine
E
Avéret
N
Rigoulet
M
Leverve
X
.
Dimethylbiguanide inhibits cell respiration via an indirect effect targeted on the respiratory chain complex I
.
J Biol Chem
 .
2000
;
275
(
1
):
223
228
.
17.
Pollak
M
.
Insulin and insulin-like growth factor signalling in neoplasia
.
Nat Rev Cancer
 .
2008
;
8
(
12
):
915
928
.
18.
Foretz
M
Hébrard
S
Leclerc
J
et al
Metformin inhibits hepatic gluconeogenesis in mice independently of the LKB1/AMPK pathway via a decrease in hepatic energy state
.
J Clin Invest
 .
2010
;
120
(
7
):
2355
2369
.
19.
A phase III randomized trial of metformin vs placebo in early stage breast cancer. Clinicaltrials.gov Web site. https://clinicaltrials.gov/ct2/show/NCT01101438?term=metformin+breast+cancer&rank=8. Updated April 10, 2015. Accessed April 17, 2015.
20.
Mehta
S
Hughes
NP
Buffa
FM
et al
Assessing early therapeutic response to bevacizumab in primary breast cancer using magnetic resonance imaging and gene expression profiles
.
J Natl Cancer Inst Monogr
 .
2011
;
2011
(
43
):
71
74
.
21.
Bonanni
B
Puntoni
M
Cazzaniga
M
et al
Dual effect of metformin on breast cancer proliferation in a randomized presurgical trial
.
J Clin Oncol
 .
2012
;
30
(
21
):
2593
2600
.
22.
Cazzaniga
M
Decensi
A
Pruneri
G
.
Effect of metformin on apoptosis in a presurgical trial in non-diabetic patients with breast cancer
.
Cancer Res
 .
2012
;
72
(
24 suppl
):
116s
.
23.
Hadad
S
Iwamoto
T
Jordan
L
et al
Evidence for biological effects of metformin in operable breast cancer: a pre-operative, window-of-opportunity, randomized trial
.
Breast Cancer Res Treat
 .
2011
;
128
(
3
):
783
794
.
24.
Kalinsky
K
Crew
KD
Refice
S
.
Pre-surgical trial of metformin in overweight and obese, multi-ethnic patients with newly diagnosed breast cancer
.
Cancer Res
 .
2012
;
72
(
24 suppl
):
117s
.
25.
Niraula
S
Dowling
RJ
Ennis
M
et al
Metformin in early breast cancer: a prospective window of opportunity neoadjuvant study
.
Breast Cancer Res Treat
 .
2012
;
135
(
3
):
821
830
.
26.
Ben-Haim
S
Ell
P
.
18F-FDG PET and PET/CT in the evaluation of cancer treatment response
.
J Nucl Med
 .
2009
;
50
(
1
):
88
99
.
27.
Habibollahi
P
van den Berg
NS
Kuruppu
D
Loda
M
Mahmood
U
.
Metformin–an adjunct antineoplastic therapy–divergently modulates tumor metabolism and proliferation, interfering with early response prediction by 18F-FDG PET imaging
.
J Nucl Med
 .
2013
;
54
(
2
):
252
258
.
28.
Strauss
LG
Conti
PS
.
The applications of PET in clinical oncology
.
J Nucl Med
 .
1991
;
32
(
4
):
623
48; discussion 649
.
29.
de Geus-Oei
LF
Visser
EP
Krabbe
PF
et al
Comparison of image-derived and arterial input functions for estimating the rate of glucose metabolism in therapy-monitoring 18F-FDG PET studies
.
J Nucl Med
 .
2006
;
47
(
6
):
945
949
.
30.
Buffa
FM
Harris
AL
West
CM
Miller
CJ
.
Large meta-analysis of multiple cancers reveals a common, compact and highly prognostic hypoxia metagene
.
Br J Cancer
 .
2010
;
102
(
2
):
428
435
.
31.
Desmedt
C
Haibe-Kains
B
Wirapati
P
et al
Biological processes associated with breast cancer clinical outcome depend on the molecular subtypes
.
Clin Cancer Res
 .
2008
;
14
(
16
):
5158
5165
.
32.
Bos
R
van Der Hoeven
JJ
van Der Wall
E
et al
Biologic correlates of (18)fluorodeoxyglucose uptake in human breast cancer measured by positron emission tomography
.
J Clin Oncol
 .
2002
;
20
(
2
):
379
387
.
33.
Hardie
DG
Ross
FA
Hawley
SA
.
AMPK: a nutrient and energy sensor that maintains energy homeostasis
.
Nat Rev Mol Cell Biol
 .
2012
;
13
(
4
):
251
262
.
34.
Mullen
AR
Wheaton
WW
Jin
ES
et al
Reductive carboxylation supports growth in tumour cells with defective mitochondria
.
Nature
 .
2012
;
481
(
7381
):
385
388
.
35.
Shu
Y
Sheardown
SA
Brown
C
et al
Effect of genetic variation in the organic cation transporter 1 (OCT1) on metformin action
.
J Clin Invest
 .
2007
;
117
(
5
):
1422
1431
.
36.
Diraison
F
Pachiaudi
C
Beylot
M
.
Measuring lipogenesis and cholesterol synthesis in humans with deuterated water: use of simple gas chromatographic/mass spectrometric techniques
.
J Mass Spectrom
 .
1997
;
32
(
1
):
81
86
.
37.
Diraison
F
Moulin
P
Beylot
M
.
Contribution of hepatic de novo lipogenesis and reesterification of plasma non esterified fatty acids to plasma triglyceride synthesis during non-alcoholic fatty liver disease
.
Diabetes Metab
 .
2003
;
29
(
5
):
478
485
.
38.
Evans
K
Burdge
GC
Wootton
SA
Clark
ML
Frayn
KN
.
Regulation of dietary fatty acid entrapment in subcutaneous adipose tissue and skeletal muscle
.
Diabetes
 .
2002
;
51
(
9
):
2684
2690
.
39.
Pinhel
IF
Macneill
FA
Hills
MJ
et al
Extreme loss of immunoreactive p-Akt and p-Erk1/2 during routine fixation of primary breast cancer
.
Breast Cancer Res
 .
2010
;
12
(
5
):
R76
.