EDITORIALThe rising tide of cardiotoxicity …The rising tide of cardiotoxicity associated with cancer chemotherapy Gary D. Lopaschuk, Mazankowski Alberta Heart Institute, University

CONTENTS

Conte

nts

EDITORIAL .................................................................................................................................................................................... 3The rising tide of cardiotoxicity associated with cancer chemotherapyG. D. Lopaschuk

BASIC ARTICLE........................................................................................................................................................................ 5Cardiotoxicity in cancer therapeuticsC. Zuppinger

MAIN CLINICAL ARTICLE.............................................................................................................................................. 9Assessing and monitoring cardiac function in cancer patientsK. von Klemperer and K. Fox

METABOLIC IMAGING ....................................................................................................................................................15Florine-18 labeled thia fatty acids for PET imaging of fatty acid oxidation in heart and cancerM. K. Panday, A. Bansal and T. R. DeGrado

NEW THERAPEUTIC APPROACHES.................................................................................................................20Metabolic modulation as a novel cancer treatmentP. Dromparis, G. Sutendra and D. Michelakis

FOCUS ON TRIMETAZIDINE (Vastarel® MR) ............................................................................................27Cancer and inhibition of fatty acid oxidationA. Huqi

CASE REPORT ........................................................................................................................................................................31A case of multifactorial late left ventricular dysfunction after cancer treatmentC. Lestuzzi

REFRESHER CORNER....................................................................................................................................................34The Warburg effect - aerobic glycolysis in proliferating cellsJ. S. Jaswal

HOT TOPICS..............................................................................................................................................................................39Warfarin and dabigatran: an historical handoverA. Huqi

GLOSSARY..................................................................................................................................................................................41G. D. Lopaschuk

Heart Metab. (2011) 51:1

The rising tide of cardiotoxicityassociated with cancer

chemotherapyGary D. Lopaschuk, Mazankowski Alberta Heart Institute, University of Alberta,

Edmonton, Canada

Correspondence: Professor Gary D. Lopaschuk, 423 Heritage Medical Research Center,University of Alberta, Edmonton, Alberta T6G 2S2, Canada.

Tel: +1 780 492 2170; fax: +1 780 492 9753e-mail: [email protected]

Advances in cancer treatment over the last few decades have resulted in marked improve-ments in patient survival. Unfortunately, many of these new cancer therapies are also asso-

ciated with toxic side effects on the heart. Consequently, the use of newer potentially cardio-toxic cancer drugs, combined with an improvement in patient survival from the cancer, hasresulted in a sharp rise in the incidence of heart failure in cancer patients. In fact, in many can-cer patients the risk of developing heart disease may be higher than the risk of cancer recur-rence [1]. Indeed, the American College of Cardiology/American Heart Association guidelinesdefine patients receiving chemotherapy as Stage A heart failure patients, since they have anincreased risk of developing heart failure [2]. The cardiotoxicity associated with cancer chemo-therapy can range from mild and temporary alterations in left ventricular ejection fraction (LVEF)to severe and irreversible life-threatening heart failure. Because of the importance of cardiotoxi-city associated with cancer therapy, this edition of Heart & Metabolism addresses the impor-tant issue of cancer and the heart.

Many of the pharmacological approaches to treat cancer involve a mechanism of action thatis actually toxic to the cardiomyocyte. This includes approaches to promote free radical pro-duction, promote apoptosis, prevent angiogenesis, inhibit metabolism, and promote ischemia.It is not surprising, therefore, that a number of these cancer cytotoxic agents have the potentialfor cardiotoxicity. This includes the use of agents such as the anthracylines, alkylating agents,microtubule targeting agents, monoclonal antibodies, tyrosine kinase inhibitors, and cytokinesto treat cancer. The potential for cardiotoxicity during cancer treatment is not confined to acuteuse of cancer chemotherapy, but can also manifest in the chronic setting. Chiara Lestuzzinicely highlights this in the Case Report in this issue of Heart & Metabolism, which describesa case of late left ventricular dysfunction in a long-term survivor of cancer. This case highlightsthe need to closely monitor survivors of cancer and aggressively treat any risk factors for thedevelopment of heart disease. This type of patient should benefit from the many recentadvances made in treating heart disease. One of these involves advances made inanti-coagulant therapy. The Hot Topics article by Alda Huqi describes how newer-generationcompounds, which includes dabigatran, may replace the use of warfarin as standardanti-coagulant therapy.

Editoria

lEDITORIAL - GARY D. LOPASCHUK

Heart Metab. (2011) 51:3–4 3

The mechanisms by which cancer cytotoxic agentsresult in cardiotoxicity are multifactorial. The BasicArticle by Christian Zuppinger describes potentialmechanisms as to how some of these cytotoxic che-motherapies for cancer may exert their cardiotoxiceffects. The Main Clinical Article by von Klempererand Fox also nicely highlights what cardiac manifesta-tions these cancer chemotherapies can exert on theheart. This includes both the acute and chronic cardiacdysfunction that can result from the use of these can-cer therapies. In addition, this article also describeshow to identify and detect this cardiac dysfunction.

Detecting and monitoring cardiac dysfunction incancer patients can involve a number of diagnosticmethods including the use of echocardiography,radionucleotide ventriculography, cardiac magneticresonance imaging, and the use of biomarkers. Posi-tron emission tomography (PET) has long been animportant tool in detecting cancers, particularly bymonitoring glucose uptake by the cancer cell. Indeed,as described in the Refresher Corner article by JagdipJaswal, cancer cells display a phenomenon called theWarburg effect, in which rapidly proliferating cells haveenhanced glycolytic rates, which is uncoupled from thesubsequent oxidation of the glucose. This increaseduptake of glucose by cancer cells has been success-fully exploited using PET to detect tumors. Use of PETto image glucose uptake (using 18F-deoxyglucose) isan important tool for detecting tumors. In addition,however, PET can also be used to image the heart,and potentially to diagnose cardiac abnormalities.The Metabolic Imaging article by Mukesh Pandey, Adi-tya Bansal, and Timothy DeGrado describes somenovel fatty acid oxidation imaging agents that can beused to assess fatty acid oxidation in the failing heart.These authors also describe how these PET imagingagents may also be useful in detecting tumorigeniccells.

As mentioned above, rapidly proliferating cancercells have a distinct switch in metabolism, away frommitochondrial oxidative metabolism and towardsglycolysis. The decrease in mitochondrial oxidativemetabolism may actually decrease apoptosis andpromote survival of the tumorigenic tissue. The New

Therapeutic Approaches article by Peter Dromparis,Gopinath Sutendra, and Evangelos Michelakisdescribes a potentially exciting new approach to treat-ing cancer, which involves promoting mitochondrialoxidation of the pyruvate derived from the acceleratedglycolysis seen in tumorigenic tissue. These authorshave pioneered a new metabolic strategy for treatingcancer that involves the stimulation of pyruvate dehy-drogenase (PDH), using the metabolic agent dichloro-cetate [3]. By stimulating PDH, glycolysis is bettercoupled to glucose oxidation in the tumor cell, result-ing in an effective decrease in tumorigenesis. Anotherapproach to stimulating PDH is to inhibit mitochondrialfatty acid oxidation, which indirectly results in anincrease in mitochondrial PDH activity. The Focus onTrimetazidine (Vastarel® MR) article by Alda Huqidescribes the potential use of the fatty acid oxidationinhibitor trimetazidine to treat cancer. The potentialadvantage of this therapeutic approach is that trimeta-zidine may also decrease the severity of cardiac toxic-ity in cancer, as a number of studies have shown thattrimetazidine has efficacy in treating heart failure.

In summary, cardiac dysfunction is an importantconsideration when deciding on the best therapeuticapproach to treat cancer. This calls for new relation-ships to be developed between the oncologist and thecardiologist in detecting, treating, and preventingcardiac dysfunction associated with the new cancertreatments. The development of novel therapies totreat cancer that involve metabolic modulation of thetumor cell may be one approach effectively treatingcancer, while lessening the possibility of cardiacdysfunction. ●

References

1. Yeh ET (2006) Cardiotoxicity induced by chemotherapy andantibody therapy. Annu Rev Med 112:1853–1887

2. Yeh ET et al (2004) Cardiovascular complications of cancertherapy: diagnosis, pathogenesis, and management. Circulation109:3122–3131

3. Bonnet S et al (2007) A mitochondrial-K+ channel axis is sup-pressed in cancer and its normalization promotes apoptosisand inhibits cancer growth. Cancer Cell 11:37–51

EDITORIAL - GARY D. LOPASCHUK

4 Heart Metab. (2011) 51:3–4

http://www.ncbi.nlm.nih.gov/pubmed/


http://www.ncbi.nlm.nih.gov/pubmed/15226229






Cardiotoxicity in cancer therapeuticsChristian Zuppinger, Cardiology, Bern University Hospital, Bern, Switzerland

Correspondence: Christian Zuppinger PhD, Cardiology, Bern University Hospital, MEM E808,Murtenstrasse 35, CH-3010 Bern, Switzerland, Tel. +41 31 6329143,

Fax. +41 31 6328837, e-mail: [email protected]

AbstractIt has been increasingly recognized that cardiac disease is a comorbid condition for cancer patientsthat deserves special attention. Although the primary goal of cancer therapy remains to cure the life-threatening disease, it is also important to maintain the highest possible quality of life, especially sincemore cancer patients survive either without cancer or with cancer as a manageable, chronic disease.Cardiovascular complications of anticancer therapies have been known for quite some time, particularlywith cytotoxic therapies such as anthracyclines used at higher cumulative doses and after mediastinalradiotherapy. Frequently, cancer cells are, or become by mutation, dependent on the activity of specifictyrosine kinases, which promote proliferation and survival. In addition, when tumor size reaches a criti-cal extent, the cancer cells depend on vascularization for further growth and metastasis. The advent oftargeted monoclonal antibodies and tyrosine kinase inhibitors has revolutionized the treatment of sev-eral types of malignancies. However, many of these targeted growth and survival pathways are alsoimportant for the homeostasis of non-cancerous tissue including the myocardium, and modulatorsmay affect protein synthesis and degradation, energy production, calcium handling and tissue integrity.

Keywords: cancer therapy, cardiotoxicity, anthracyclines, cell death, mitochondria

■ Heart Metab. (2011) 51:5–8

Introduction

Cardiac toxicity associated with cancer therapies can range from asymptomatic subclinicalabnormalities, including electrocardiographic changes and temporary left ventricular ejectionfraction decline, to life-threatening congestive heart failure or acute coronary syndrome.Delayed toxicities may be less relevant for the acute cancer treatment, but much more so inthe neo- and adjuvant setting where the overall duration of treatment can measure severalyears. Progress has been made in reducing the side effects of oncological treatment, forexample by giving an iron chelator for the inhibition of free radicals, by improvements in imag-ing technologies for radiotherapy, or with different treatment schedules and drug formulationssuch as liposomes in order to target the therapy more precisely to the tumor [1]. Meanwhile,a large number of molecular targets for new anticancer-therapies have been developed [2].

Cardiotoxicity of cytotoxic chemotherapies

Cytotoxic chemotherapeutic drugs can be divided in anthracyclines, alkylating agents, anti-metabolites, microtubule targeting agents, and others. Some of these drugs were discoveredas naturally occurring compounds in plants or microorganisms, and they generally lead togrowth inhibition and eradication of rapidly dividing cells. Anthracyclines are still widely usedfor many malignancies, despite their well-described cumulative cardiotoxicity, which is often

Basic

Article

BASIC ARTICLE - CHRISTIAN ZUPPINGER

Heart Metab. (2011) 51:5–8 5

irreversible, and can lead to congestive heart failure[1, 3]. While there are a number of discrepancies instudies on anthracycline effects in cardiomyocytesregarding the primary mechanism—which could beexplained by differences in drug dosage, animalmodel, and duration of treatment [4]—the formationof reactive oxygen species (ROS), changes in mito-chondrial function [5], and finally mismanagement ofcellular calcium handling appears to be a commonendpoint [1, 6]. The irreversible loss of cardiomyo-cytes and hypothetical stem cells might also contrib-ute to the late effects of anthracyclines, which canoccur many years after treatment and is especiallyimportant for the treatment of pediatric patients [7].Alkylating agents like cyclophosphamide and cisplatinmay cause acute symptoms of chest pain, arrhyth-mias, and cardiac ischemia with elevated enzymesindicative of myocardial infarction after a single dose[8]. As with the anthracyclines, the formation of ROSis proposed as a cellular mechanism for the cardiotoxi-city of alkylating agents. For the class of the antimeta-bolites, it is mainly 5-fluorouracil (5-FU) that is knownfor the risk of cancer therapy-associated cardiotoxicity.Coexisting coronary heart disease and hypertensionfrequently worsens the problem [9]. Microtubule tar-geting agents, like the taxanes, with paclitaxel as thebest-studied example, can lead to bradycardia andother electrical disorders and to thrombosis [8]. Thecombined use of paclitaxel with anthracyclinesenhances their cardiotoxicity [10]. In contrast to othercytotoxic drugs such as the anthracyclines, the effectof paclitaxel on cardiomyocytes does not compriseimmediate cell death [11].

Cardiotoxicity of targeted therapies

So-called targeted therapies are usually designed witha molecular target in mind that has already been iden-tified as a modulator of cell division and growth. Ideally,such a target, or family of target proteins, is morecommonly found in those cells that have become, forexample by genetic mutation, continuously dividingcancer cells. One of the first therapies of this kind, tras-tuzumab, is based on humanized antibodies againstthe receptor tyrosine kinase HER2. This receptor isfound overexpressed in some tumors, frequently inbreast cancer, making the HER2 receptors (corre-sponding to ErbB2 in rodents) a seemingly ideal targetfor cancer therapy. However, the use of trastuzumab

in combination with anthracyclines was associatedwith cardiac dysfunction in up to 28% of patients in apivotal trial [12]. The cardiac dysfunction observedduring adjuvant treatment with trastuzumab for up to2 years seems to be due to an impairment of contrac-tility, rather than loss of cardiomyocytes, and is revers-ible in the majority of cases [13]. A mechanism withoutcardiomyocyte loss is also suggested by results fromin vitro studies using adult rodent cardiomyocytes[11, 14]. The experience with trastuzumab and theexperiments using neuregulin-1 beta in basic cardio-vascular research has led to the hypothesis that survivalfactors such as the endothelium-derived neuregulin-1beta, which binds to the heterodimer ErbB2/ErbB4 incardiomyocytes, play an important role for myofibrillarintegrity, contractile function and metabolic homeosta-sis in the myocardium challenged by cytotoxic cancertherapies and in general in stress conditions [14, 15].

The other and larger group of targeted therapies istyrosine kinase inhibitors (TKIs), which bind to theATP pocket of tyrosine kinases, either to receptors orintracellular kinases. Many new TKIs are currently indevelopment and in early clinical trials. Among themare inhibitors of the mitogen-activated protein kinase(MAPK) and phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signalingpathways that are also important in the cardiovascular

Fig. 1 Primary cardiac microvascular endothelial cells were isolatedfrom adult rats, cultured on glass-bottom dishes, stained alive withJC-1 and examined in the fluorescence microscope after treatmentwith sunitinib for 18 hours (a: untreated, b: 7.5 nM sunitinib, 15 nMsunitinib, 30 nM sunitinib). High mitochondrial membrane potentialpromotes the formation of dye aggregates, which fluoresce in red,while low potential results in the appearance of monomeric JC-1and diffuse green flourescence.


6 Heart Metab. (2011) 51:5–8

system. Careful cardiac monitoring seems warrantedas these trials proceed [2]. Some TKIs are multi-targeted, i.e., less specific for a single protein kinase.The premise of less selective TKIs is that one or moreof these additional targets may also play a role in dis-ease progression and its inhibition will lead to betteranticancer efficacy [16].

Antiangiogenic targeted therapies are currentlyused in clinical practice in the form of antibodies,such as the anti-vascular endothelial growth factor(VEGF)-A therapy bevacizumab, or as TKIs such as

sorafenib and sunitinib. Sunitinib inhibits, amongothers, VEGF receptor 1–3, c-Kit, platelet-derivedgrowth factor receptor (PDGFR)-A/B, rearranged dur-ing transfection (RET), FMS related tyrosine kinase3 (FLT3), and colony stimulating factor 1 receptor(CSF1R) [16]. The mechanism for the observed cardi-otoxicity of sunitinib is probably related to the inhibitionof 5’ adenosine mono phosphate-activated proteinkinase (AMPK) in cardiomyocytes and can leadto cell death at micromolar concentrations in vitro[17, 18]. We and others have observed changes in

ErbB2ErbB2 ErbB4ErbB4

Neuregulin

Ras/Raf

MEK

PI-3-kinase

Akt/PKBERK1/2

Stressors

Anthracyclines

Taxanes

Antimetabolites

Radiation therapy

Ischemia/reperfusion

OxidativeStress

[Ca2+]i

Proteases

Sorafenib

EverolimusSunitin

ib

Sunitinib

Gene Expression

GATA-4

myofibrillar degradationautophagy, apoptosis

necrosis Cardiomyocytes

Cardiac microvascular endothelium

VEGF-A

VEGF-A

HIF1

Bevacizumab

Bor

tezo

mib

Trastuzumab

VEGFR-2

Cytochrome Crelease

VEGFR-1

Lapatinib

Proteasome Bcl-2

Bcl-XL

mitochondriamTOR

Doxorubicin

Hypoxia

Perifosine

TAS

106

IMC-A12

IGF-IR

Fig. 2 Some myocardial targets of cancer therapies and potentially cardioprotective signaling pathways in cardiomycoytes and as partof the paracrine signaling of microvascular endothelial cells and cardiomyocytes. Green arrows show therapeutics that potentially interferewith signaling components in the heart, although mentioning a certain therapy does not imply clinically relevant cardiotoxicity. Top left:Cytotoxic therapies and other stressors often lead to oxidative stress in cardiomyocytes and can cause protein degradation and cell death.Top right: Microvascular endothelial cells, among others, release factors that enhance protein synthesis and survival under stress condi-tions. Neuregulin and VEGF-A are examples, activating the MAPK and PI3K signaling pathways in cardiomyocytes. Bottom: Gene expres-sion depends on transcription factors, which are also targets. GATA-4 drives the expression of antiapoptotic mitochondrial proteins Bcl-2and Bcl-XL, but is inhibited by doxorubicin, and HIF-1 alpha is important for VEGF-A production in hypoxia, but is inhibited by new targetedand antiangiogenic therapies.


Heart Metab. (2011) 51:5–8 7

mitochondrial function by sunitinib in different celltypes, as visually demonstrated with the cationic dyeJC-1, that selectively enters into mitochondria andreversibly changes fluorescence emission color fromred to green as the membrane potential decreases(Fig. 1). Antiangiogenic therapies could also have animpact on the heart by the induction of microvasculardysfunction and/or the reduction of microvessel den-sity. However, pathologic changes seen in culturedcells and small rodents do not necessarily translateinto clinically significant cardiac toxicity, and thedescribed mitochondrial effect of sunitinib is not acommon feature in TKIs with a similar inhibition spec-trum [19]. It is important to note, that cardiotoxicity isnot a class effect of TKIs, and it is reasonable toassume, that only those therapies targeted at essentialkinases in the myocardium and in the vasculature willshow cardiotoxicity. Nevertheless, the combined useof therapies, which have not shown significant cardio-toxic potential in single use, could create a potentiallyharmful and unexpected double-hit on themyocardium.

Conclusions

Cardiovascular side effects of the new signaling inhibi-tors appear to be particularly relevant in patients withpreexisting cardiovascular conditions and risk factors,since they depend on survival pathways that nowbecome targets in modern oncology. Sometimes, itmight even appear as oncologists and cardiologisthave opposing interests, since most of the cellular fac-tors discussed in recent years for a role in cardiopro-tection are on the list of new targets under study fortheir use in anticancer therapy (Fig. 2). On the contrary,given the unique problems presented by the cancerpatient with cardiovascular disease, multidisciplinaryworking among cardiologists, oncologists and cellbiologists should be encouraged [20]. Cardiovascularside effects of cancer therapeutics may become amanageable problem in future if we are able to learnmore about the role of signaling pathways in the myo-cardium. Still, this topic will continue to be topical in theyears to come. ●

Acknowledgements This work was supported by a researchgrant from the Swiss National Science Foundation (3100A0-120664) to C. Zuppinger.

References

1. Anderson B, Sawyer DB (2008) Predicting and preventing thecardiotoxicity of cancer therapy. Expert Rev Cardiovasc Ther6(7):1023–1033

2. Peng X, Pentassuglia L, Sawyer DB (2010) Emerging antican-cer therapeutic targets and the cardiovascular system: is therecause for concern? Circ Res 106(6):1022–1034

3. Singal P, Iliskovic N (1998) Doxorubicin-induced cardiomyo-pathy. N Engl J Med 339(13):900–905

4. Gewirtz DA (1999) A critical evaluation of the mechanisms ofaction proposed for the antitumor effects of the anthracyclineantibiotics adriamycin and daunorubicin. Biochem Pharmacol57(7):727–741

5. Tokarska-Schlattner M, Zaugg M et al (2006) New insights intodoxorubicin-induced cardiotoxicity: the critical role of cellularenergetics. J Mol Cell Cardiol 41(3):389–405

6. Lim CC, Zuppinger C, Guo X et al (2004) Anthracyclinesinduce calpain-dependent titin proteolysis and necrosis in car-diomyocytes. J Biol Chem 279(9):8290–8299

7. Alvarez JA, Scully RE et al (2007) Long-term effects of treat-ments for childhood cancers. Curr Opin Pediatr 19(1):23–31

8. Yeh ET, Tong AT et al (2004) Cardiovascular complications ofcancer therapy: diagnosis, pathogenesis, andmanagement. Cir-culation 109(25):3122–3131

9. Senkus E, Jassem J (2011) Cardiovascular effects of systemiccancer treatment. Cancer Treat Rev 37(4):300–311

10. Gianni L, Salvatorelli E, Minotti G (2007) Anthracycline cardio-toxicity in breast cancer patients: synergism with trastuzumaband taxanes. Cardiovascular Toxicol 7(2):67–71

11. Pentassuglia L, Timolati F et al (2007) Inhibition of ErbB2/neur-egulin signaling augments paclitaxel-induced cardiotoxicity inadult ventricular myocytes. Exp Cell Res 313(8):1588–1601

12. Slamon D, Leyland-Jones B, Shak S et al (2001) Use of che-motherapy plus a monoclonal antibody against HER2 for met-astatic breast cancer that overexpresses HER2. N Engl J Med344(11):783–792

13. Suter TM, Procter M, van Veldhuisen DJ et al (2007)Trastuzumab-associated cardiac adverse effects in the her-ceptin adjuvant trial. J Clin Oncol 25(25):3859–3865

14. Sawyer DB, Zuppinger C, Miller TA, Eppenberger HM, SuterTM (2002) Modulation of anthracycline-induced myofibrillardisarray in rat ventricular myocytes by neuregulin-1beta andanti-erbB2: potential mechanism for trastuzumab-inducedcardiotoxicity. Circulation 105(13):1551–1554

15. De Keulenaer GW, Doggen K, Lemmens K (2010) The vulner-ability of the heart as a pluricellular paracrine organ: lessonsfrom unexpected triggers of heart failure in targeted ErbB2anticancer therapy. Circ Res 106(1):35–46

16. Cheng H, Force T (2010) Molecular mechanisms of cardio-vascular toxicity of targeted cancer therapeutics. Circulationresearch 106(1):21–34

17. Kerkela R, Woulfe KC et al (2009) Sunitinib-induced cardio-toxicity is mediated by off-target inhibition of AMP-activatedprotein kinase. Clin Transl Sci 2(1):15–25

18. Hasinoff BB, Patel D, O’Hara KA (2008) Mechanisms of myo-cyte cytotoxicity induced by the multiple receptor tyrosinekinase inhibitor sunitinib. Mol Pharmacol 74(6):1722–1728

19. French KJ, Coatney RWet al (2010) Differences in effects onmyo-cardium and mitochondria by angiogenic inhibitors suggest sepa-rate mechanisms of cardiotoxicity. Toxicol Pathol 38(5):691–702

20. Eschenhagen T, Force T, Ewer MS, de Keulenaer GW et al(2011) Cardiovascular side effects of cancer therapies: a positionstatement from the Heart Failure Association of the EuropeanSociety of Cardiology. Eur J Heart Fail 13(1):1–10


8 Heart Metab. (2011) 51:5–8
































































Assessing and monitoring cardiac functionin cancer patients

K. von Klemperer and K. Fox, Cardiology, Imperial College Hospital’s Trust, London, United Kingdom

Correspondence: Dr. Kevin Fox, Honorary Clinical Senior Lecturer, Imperial College London, Cardiology DepartmentCharing Cross Hospital, Fulham Palace Road, London W6 8RF, United Kingdom. e-mail: [email protected]

AbstractThe use of chemotherapy is associated with improved survival rates for patients with neoplastic dis-ease. Cardiac toxicity is a feature of several therapeutic agents (e.g., anthracyclines) and patientsrequire careful monitoring during treatment. The adverse effects of chemo- and radiotherapy may notbe apparent for some time post completion of treatment and patients may require long-term follow-upby oncologists and cardiologists. In this invited article we review the therapies associated with signifi-cant cardiac toxicity and provide guidance on the assessment of patients during chemotherapy andduring chronic surveillance.

Keywords: cancer, chemotherapy, heart failure, cardiotoxicity

■ Heart Metab. (2011) 51:9–14

Introduction

Progress made in screening and therapies have greatly reduced morbidity and mortality due tocancer over the past few decades. In 2006, there were 1.13 million cancer survivors in theUnited Kingdom who were up to 10 years from diagnosis [1]. Some of these therapies dohowever carry a significant risk of cardiotoxicity and after-cancer recurrence; cardiac eventsare the most common cause of death in these patients. While coronary ischemia, arrhythmia,thromboembolism, and QT prolongation have all been noted as acute complications of specificchemotherapies (Table 1) [2], in this review we focus on cardiomyopathy due to cancertherapy.

The mechanisms of left ventricular (LV) dysfunction have been extensively studied, yet fewprospective clinical studies exist and there is a lack of consensus and guidelines regardingbest practice in terms of assessing, monitoring, and managing these patients. The risk-benefitcalculations are therefore challenging for the oncologists and cardiologists involved in theircare.

Chemotherapy agents

Anthracyclines (doxorubicin, epirubicin, idarubicin)

Anthracyclines are recognized as highly efficacious and are still one of the most commonlyused chemotherapies in treating adult solid organ and hematological tumors. They are alsorenowned for causing a dose-dependent and progressive toxic cardiomyopathy [3].MainCli

nicalArticle

MAIN CLINICAL ARTICLE - K. VON KLEMPERER

Heart Metab. (2011) 51:9–14 9

There are three distinct types of cardiotoxicity asso-ciated with anthracyclines:

• Acutely (<1%) or subacutely, they can cause apericarditis-myocarditis syndrome, or acute left ven-tricular failure (usually reversible).

• A chronic, progressive cardiomyopathy can presentwithin the first year following the start of treatment.

• Late-onset cardiomyopathy can manifest years todecades after anthracycline treatment. In the pediat-ric population, the effects are often not seen until upto 25 years post-chemotherapy.

We know from a large experience of endomyocardialbiopsies that anthracyclines destroy myocytes. Therehave been considerable efforts into finding ways toidentify those at risk and protect the myocardiumfrom anthracyclines. Identified “risk factors” are: under-lying cardiovascular disease, pregnancy, hypertension,pre-existing cardiac disease, age >65 or <18 at time oftherapy initiation, combination chemotherapy, radiationtherapy, intravenous bolus administration, high cumu-lative dose [4, 5]. Retrospective analyses suggest thatdoxorubicin cardiotoxicity occurs at dosages consider-ably lower than first appreciated: the cumulative dos-age that correlated with a 5% incidence of heart failureranged between 400 and 450 mg/m2 [6].

Attempts to minimize the cardiotoxicity of anthracy-clines include dose limitation, schedule modification,

use of less cardiotoxic analogues, and use of cardio-protective agents. There has been some evidence thatadding dexrazoxane (a free radical scavenger) can becardioprotective [7]. There are, however, concernsthat it may interfere with the antitumor effect of theanthracyclines and the American Society of ClinicalOncology currently advises against its routine use [8].

Alkylating agents (cyclophosphamide, ifosfamibe)

Acutely, cyclophosphamide has been associated witha hemorrhagic myocarditis. It has also been associ-ated with heart failure in up to 28% of patients. Cardi-otoxicity has been shown to be dose related, irrever-sible and typically manifests within 2 weeks of therapy[9]. Risk factors for cardiotoxicity include previousanthracycline use and previous radiotherapy.

Humanized monoclonal antibody based tyrosinekinase inhibitors (bevacizumab, trastuzumab)

Cancer patients receiving trastuzumab (Herceptin®) area relatively new population (it was first approved by theUnited States Food and Drug Administration in 1998). Itis indicated in combination with paclitaxel for first-linetreatment and as a single agent for second- or third-line treatment of metastatic breast cancer. Approxi-mately 25% of women with breast cancer will have atumor overexpressing human epidermal growth factor2 (HER2) and addition of trastuzumab to adjuvant

Therapy LV dysfunction Ischemia Prolonged QT Arrythmia

AnthracyclinesDoxorubicin, epirubicin, idarubicin >5% incidenceAlkylating agentsCyclophosphamide, ifosfamide >5% incidenceHumanized monoclonal antibodiesTrastuzumab, bevacizumab >5% incidence <2% incidenceSmall molecule tyrosine kinase inhibitorsSunitinib, dasatinib, lapatinib 2-5% incidence >5% incidenceErlotinib, sorafenib 2-5% incidence >5% incidenceRadiotherapy >5% incidenceAntimetabolites5Fluorouracil, capecitabine >5% incidenceAntimicrotubulePaclitaxel 2-5% incidence >5% incidenceProteosome InhibitorsBotezomib 2-5% incidence

Table 1 Incidence of cardiac complications of commonly used chemotherapies. A summary of the major chemotherapeutic classesand their estimated incidence of cardiac complication based on meta-analysis (modified from Yeh and Bickford [2]).


10 Heart Metab. (2011) 51:9–14

therapy has been shown to reduce their risk of recur-rence at 3 years by half and improve survival by approx-imately a third [10]. It has also recently been approvedfor use in HER2-positive metastatic stomach cancer.

The pivotal trials of trastuzumab observed an unex-pected risk of cardiac dysfunction. Patients with breastcancer are commonly treated with both trastuzumaband an anthracycline, which poses the problem of iso-lating effects of either agent from a potential synergisticinteraction. Cardiotoxicity has however been shown tobe rare in patients on trastuzumab alone.

Endomyocardial biopsy specimens evaluated post-trastuzumab exposure have not shown any changesthat suggest significant myocyte destruction [11].HER2 receptors exist in the heart and it is thoughtthe HER2/erbB2 system may modulate the effects ofoxidative stress in the cardiac myocyte. Some postu-late that these drugs interfere with myocyte meta-bolism and, in the patient with impaired contractilereserve due to previous or concomitant anthracyclineor alkylating agents, this may manifest as deteriorationin cardiac function.

Unlike anthracyclines trastuzumab rarely causesdeath, is not cumulatively dose related, and the major-ity of patients who developed trastuzumab-relatedcongestive heart failure and discontinued trastuzumabhad recovery of left ventricular ejection fraction (LVEF)with subsequent follow-up [12]. The long-term effectsof these drugs remain to be seen.

Small molecule tyrosine kinase inhibitors(lapatinib, imatinib, sumatanib)

The cardiac safety of lapatinib was recently evaluated inthe 3,689 patients enrolled in phase I to III lapatinib clin-ical trials. Of these patients, 1.6% experienced a car-diac event. In patients treated with prior anthracyclines,trastuzumab, or neither, the incidence of cardiac eventswas 2.2%, 1.7%, and 1.5%, respectively. The meantime to onset of cardiac events was 13 weeks [13].

Radiotherapy

Mediastinal radiation potentially leads to inflammationand progressive fibrosis of all of the heart structures. Itis known to cause pericardial disease, acceleratedcoronary artery disease and has been implicated inconduction disease although causality is difficult toestablish. In comparison to the systolic dysfunctionseen with chemotherapy these patients have beenshown to develop a restrictive cardiomyopathy. The

occurrence and manifestation of radiation-relatedcardiac disease depends mostly on radiation dose,volume of the heart exposed, and specific radiationdelivery techniques used. A study of 4,122 patientsfound the relative risk of cardiac death was 12.5 atradiation doses of 5-15 Gy, and 25.1 at >15 Gy [14].The Oxford meta-analysis of randomized trials alsodemonstrated an increased risk of ischemic heart dis-ease [15], however the majority of the patientsincluded in these studies would have been exposedto large volumes of radiotherapy and modern radiationtherapy techniques (used since the mid 1980s) withthree-dimensional planning, computed tomographyguided therapy and lower radiation doses seem tohave significantly reduced this risk [16].

Defining and detecting cardiac dysfunction

In order to assess and monitor these patients we haveto identify how we recognize and define “cardiac dys-function”. When the pivotal trials of trastuzumab estab-lished that there was associated cardiotoxicity [17], acardiac review committee established the followingcriteria to identify cardiac dysfunction:

(1) A decrease in cardiac LVEF that was eitherglobal or more severe in the septum.(2) Symptoms of congestive heart failure.(3) Associated signs of heart failure.(4) Decline in LVEF of at least 5% to less than 55%with accompanying signs or symptoms of conges-tive heart failure, or a decline in LVEF of at least10% to below 55% without accompanying signsor symptoms.

Initial assessmentCurrently in most institutions it is the oncologists whoare assessing these patients. Ideally prior to antineo-plastic therapy, they will conduct a detailed baselineassessment of functional capacity, known cardiachistory as well as risk factors. Notably many of thesymptoms of cancer can be difficult to distinguishfrom cardiac symptoms (chest pain, breathlessness,leg swelling). Oncologists should identify patients atparticular risk of cardiotoxicity so that the choice of ther-apy and risk factors may be optimized and they can bemonitored more vigilantly.

Prior to starting therapy, one needs a baseline ECGand study of cardiac function. The most common non-invasive method of monitoring myocardial toxicity hasbeen the assessment of LV systolic function, with either


Heart Metab. (2011) 51:9–14 11

radionuclide ventriculography or echocardiography.The two measurements are not interchangeable; thesame technique should be used for serial measure-ments. Where possible, this should also include thesame operator, machine, and calculation algorithm.

After baseline assessment, patients with LV dys-function or modifiable cardiac risk factors should bereferred for cardiology review. There is some trial evi-dence for the efficacy of ACE inhibitors in preventingthe decrease in LVEF observed in patients after high-dose chemotherapy [18]. The regime of chemotherapymight be modified accordingly.

In the United Kingdom, the National Institute forHealth and Clinical Excellence recommends cardiacfunctional assessments should be repeated every3 months during trastuzumab treatment. If the LVEFdrops by 10% or more from baseline and to below50%, trastuzumab treatment should be suspended.A decision to resume trastuzumab therapy should bebased on a further cardiac assessment and a fullyinformed discussion of the risks and benefits betweenthe individual patient and his or her clinician [19]. Thereare consensus guidelines for anthracyclines thatadhere to a similar protocol, however they advocatelong-term follow-up on a yearly or two-yearly basisdepending on the cumulative dose of chemotherapygiven and the risk profile of the patient.

Methods of monitoring

MUGA

Radionuclide ventriculography or multiple gated acqui-sition scan (MUGA) is frequently used. In its favor, itis less subject to observer variability [20] than echo-cardiography, however it does involves exposure toradiation, is associated with higher cost and does notgive us any information other than ejection fraction incomparison to echocardiography, which may identifypericardial or valvular disease as well. Choice betweenechocardiography and MUGA is based on providerpreference or regional practice.

Echocardiography

Echocardiography is widely available and relativelyinexpensive and hence is often the initial investigationof choice for evaluating systolic function. While dis-puted by some, ejection fraction is currently thestandard method of screening for cardiotoxicity.

Poor quality echo images due to surgery and inter-observer variability may confound these measure-ments. Contrast or 3D echocardiography mightimprove reproducibility, and the laboratories perform-ing such studies should have evidence that they canreliably and reproducibly identify a 10% change in EFas a true change [21].

Many studies have aimed to identify more sensitiveindicators of early cardiotoxicity, including Tei Index,tissue Doppler, and myocardial strain [22, 23]. Severalhave suggested that diastolic dysfunction is an earlysign of anthracycline-induced cardiac dysfunction.These are generally small studies and further validationis necessary [24–26].

Other methods currently being evaluated

While most of the methods currently used do monitorfor deterioration in systolic dysfunction, there is anargument that cardiac dysfunction is not best appre-ciated by measurements of systolic function. It may bepreceded by asymptomatic or sub clinical cardiacdysfunction undetectable at bedside evaluation and/or defined by abnormalities measured by standardnon-invasive imaging techniques

Dobutamine stress echo

It is known that patients with mild and sub-clinicaldeterioration of systolic LV function can compensatefor a decreased cardiac output with a number of adap-tive mechanisms, i.e., increasing preload, heart rate,and contractility during stress. Dobutamine stressecho is an established method of assessing contractilereserve in ischemic and valvular heart disease. It is agenerally well-tolerated test, which can be repeated. Itdoes however require a less widely available skill setand its role in assessing patients on chemotherapy isless well established [27]. No studies have evaluatedthe benefits of non-invasive evaluation of asymptom-atic cancer survivors for cardiac dysfunction. Yeung etal tested 29 children, 19 of whom had received anthra-cyclines up to 6 years earlier, with exercise echocardi-ography. They found an average increase in fractionalshortening of 3% in the anthracycline group comparedwith an average increase of 23% in the control group,although fractional shortening at rest was normal in allparticipants [28]. This raises the question of how wewould interpret and act on such findings.


12 Heart Metab. (2011) 51:9–14

CMR

Cardiac magnetic resonance imaging (CMR) has thebenefits of low intra/inter-observer variability and hightest-retest reproducibility. It also has the ability to detectearly myocardial damage. Wassmuth et al have demon-strated that CMR may be useful at detecting thesub-clinical effects of chemotherapy. Increased contrastenhancement on day 3 predicted significant decline at28 days [29]. It could be useful in patients with poorecho pictures and in fact as more CMR servicesbecome available, it may serve as a simple non-invasive tool in following up patients post chemotherapy.

Biomarkers

It is well recognized that cardiac dysfunction maybecome evident weeks, months or years after high-dose chemotherapy. The possibility of developing anearly marker of myocardial injury would allow us toappropriately monitor those at higher risk. Troponinand BNP have both been explored in detecting cardiacdamage and risk stratifying patients. Cardinale et aldemonstrated that troponin I is a sensitive and specificmarker for predicting the development and severity ofventricular dysfunction. In one study of 703 patients,elevation of troponin I levels 0–72 hours after chemo-therapy administration and at 1 month, predicted a latedecline in LVEF and a composite of cardiac events[30]. Cardinale also demonstrated a strong relation-ship was also found between NT-proBNP value at72 hours, and LVEF changes at 12 months. Whilethese initial studies of biomarkers as a screening toolare promising the numbers are small and more work inthis area is needed for further validation.

Conclusion

Patients with cancer have a substantial risk of devel-oping heart disease as a result of chemo- and radiationtherapy. Cardiotoxicity, not only negatively affects thecardiac outcome of patients with cancer, but alsolargely reduces the range of suitable therapies. Theoptimal interval and duration as well the cost effective-ness of cardiac monitoring remain undefined. Thatbeing said, from the evidence and consensus guide-lines we do have it is possible to design sensible moni-toring protocols. These need to be clearly defined andregular quality assurance is key. The process of asses-sing and following up these patients needs to be sys-tematic not only to avoid missing pathology, but also

for future research purposes. Most importantly wherecardiac dysfunction is identified, there needs to be aclear avenue for communication and collaborationbetween oncologists and cardiologists within a multi-disciplinary cancer team. ●

References

1. http://info.cancerresearchuk.org/cancerstats/types/breast/mortality.

2. Yeh ETH, Bickford CL (2009) Cardiovascular complications ofcancer therapy. Incidence, pathogenesis, diagnosis, and man-agement. J Am Coll Cardiol. 53:2231–2247

3. Von Hoff DD, Layard M (1981) Risk factors for development ofdaunorubicin cardiotoxicity. Cancer Treat Rep 65:19–23

4. Legha SS, Benjamin RS, Mackay B, Ewer M, Wallace S, Val-divieso M et al (1982) Reduction of doxorubicin cardiotoxicityby prolonged continuous intravenous infusion. Ann Intern Med96:133–139

5. Torti FM, Bristow MR, Howes AE, Aston D, Stockdale FE,Carter SK et al (1983) Reduced cardiotoxicity of doxorubicindelivered on a weekly schedule. Assessment by endomyocar-dial biopsy. Ann Intern Med 99:745–749

6. Swain SM,Whaley FS,Ewer MS (2003) Congestive heart failurein patients treated with doxorubicin: a retrospective analysis ofthree trials. Cancer 97:2869–2287

7. Speyer JL, Green MD, Zeleniuch-Jacquotte A, Wernz JC, ReyM, Sanger J et al (1992) ICRF-187 permits longer treatmentwith doxorubicin in women with breast cancer. J Clin Oncol10:117–127

8. Hensley ML, Hagerty KL, Kewalramani T, et al (2009) Ameri-can Society of Clinical Oncology 2008 Clinical Practice Guide-line Update: use of chemotherapy and radiotherapy protec-tants. J Clin Oncol 27:127–145

9. Dow, E., Schulman, H. & Agura, E (1993) Cyclophosphamidecardiac injury mimicking acute myocardial infarction. BoneMarrow Transplant 12:169–172

10. Harbeck N,Ewer M, De Laurentiis et al (2010) Cardiovascularcomplications of conventional and targeted adjuvant breastcancer therapy. Ann Oncol doi: 10.1093/annonc/mdq543

11. Heidman A, Hudis C, Pierri MK et al (2005) Cardiac dysfunc-tion in the trastuzumab clinical trials experience. J Clin Oncol20:1215–1221

12. Ewer MS, Vooletich MT, Durand J-B et al (2005) Reversibilityof trastuzumab-related cardiotoxicity: New insights based onclinical course and response to medical treatment. J ClinOncol 23:7820–7826

13. Perez EA, Koehler M, Byrne J, Preston AJ, Rappold E, EwerMS (2008) Cardiac safety of lapatinib: pooled analysis of3689 patients enrolled in clinical trials, Mayo Clin Proc83:679–686

14. Tukenova M et al (2010) Role of cancer treatment in long-termoverall and cardiovascular mortality after childhood cancer.J Clin Oncol 28:1308–1315

15. Early Breast Cancer Trailist’s Collaborative Group (2005)Effects of radiotherapy and of differences in the extent of sur-gery for early breast cancer on local recurrence and 15 yearsurvival: An overview of randomised trials. Lancet 366; 2087–2106

16. Patt DA, Goodwin JS, Kuo YF et al (2005) Cardiac morbidityof adjuvant radiotherapy for breast cancer. J Clin Oncol.23:7475–7482


Heart Metab. (2011) 51:9–14 13























































17. Seidman A, Hudis Pierri MK et al (2002) Cardiac dysfunction inthe trastuzumab clinical trials experience. J Clin Oncol20:1215–1221

18. Cardinale D, Colombo A, Sandri MT, Lamantia G, Colombo N,Civelli M, Martinelli G, Veglia F, Fiorentini C, Cipolla CM (2006)Prevention of high-dose chemotherapy-induced cardiotoxicityin high-risk patients by angiotensin-converting enzyme inhibi-tion. Circulation 114:2474–2481

19. www.nice.org.uk/nicemedia/pdf/TA107guidance.pdf20. Van Royen N, Jaffe CC, Krumholz HM et al (1996) Compari-

son and reproducibility of visual echocardiographic and quan-titative radionucleotide left ventricular ejection fractions. AmJ Cardiol 77:845–850

21. Fox KF (2006) The evaluation of left ventricular functionfor patients being considered for, or receiving trastuzumab(Herceptin) therapy [letter]. Br J Cancer 95:1454

22. Kapusta L, Thijssen JM, Groot-Loonen J et al (2000) TissueDoppler imaging in detection of myocardial dysfunction insurvivors of childhood cancer treated with anthracyclines.Ultrasound Med Biol 26:1099–1108

23. Ganz WI, Sridhar KS, Ganz SS et al (1996) Review of tests formonitoring doxorubicin-induced cardiomyopathy. Oncology53:461–470

24. Marchandise B, Schroeder E, Bosly A, Doyen C, Weynants P,Kremer R, et al (1989) Early detection of doxorubicin cardio-

toxicity: interest of Doppler echocardiographic analysis of leftventricular filling dynamics. Am Heart J 118:92–98

25. Ganz WI, Sridhar KS, Forness TJ (1993) Detection of earlyanthracycline cardiotoxicity by monitoring the peak filling rate.Am J Clin Oncol. 16:109–112

26. Tassan-Mangina S, Codorean D, Metivier M et al (2006)Tissue Doppler imaging and conventional echocardiographyafter anthracycline treatment in adults: early and late altera-tions of left ventricular function during a prospective study.Eur J Echocardiogr 7:141–146

27. Klewer SE, Goldberg SJ, Donnerstein RL, Berg RA, Hutter JJJr (1992) Dobutamine stress echocardiography: a sensitiveindicator of diminished myocardial function in asymptomaticdoxorubicin-treated long-term survivors of childhood cancer.J Am Coll Cardiol 19:394–401

28. Yeung ST, Yoong C, Spink J, Galbraith A, Smith P (1991)Functional myocardial impairment in children treated withanthracyclines for cancer. Lancet 337:816–818

29. Wassmuth R, Lentzsch S, Erdbruegger U, Schulz-Menger J,Doerken B, Dietz R, Friedrich MG (2001) Sub clinical cardio-toxic effects of anthracyclines as assessed by magnetic reso-nance imaging-a pilot study. Am Heart J 141(6):1007–1013

30. Cardinale D, Sandri MT, Martinoni A et al (2000) Left ventricu-lar dysfunction predicted by early troponin I release after high-dose chemotherapy, J Am Coll Cardiol 36:517–522


14 Heart Metab. (2011) 51:9–14



















































Fluorine-18 labeled thia fatty acidsfor PET imaging of fatty acid oxidationin heart and cancerMukesh K. Pandey, Aditya Bansal and Timothy R. DeGrado, Department of Radiology, Brigham and Women’s

Hospital, Harvard Medical School, Boston, Massachusetts, USA

Correspondence: Timothy R. DeGrado, 75 Francis Street, Boston, Massachusettes 02115, USA.Tel.: +1 617-525-8939, Fax: +1 617-264-5291, e-mail: [email protected]

AbstractMyocardial fatty acid oxidation (FAO) imaging is a noninvasive technique that can measure FAO rates intissues for research applications in animals and humans, as well as clinical applications in managingpatients with metabolic disorders. FAO imaging has great potential in diagnosis and monitoring ofpatients with ischemic heart disease, cardiomyopathies, myocarditis, acute coronary syndrome, andheart failure. Applications of FAO imaging in oncology and endocrinology are also highly anticipated. Forover 20 years, our laboratory has investigated fluorine-18 labeled thia-substituted fatty acid analogsas positron emission tomography (PET) probes of myocardial FAO. These FAO probes share a commondesign motif of metabolic trapping in the myocardium subsequent to their commitment to the mitochon-drial FAO pathway, in analogy to the design of 2-[18F]fluoro-2-deoxy-D-glucose (FDG) as a metabolicallytrapped probe of glucose transport and phosphorylation. This mini-review describes the development ofthese FAO probes, from the seminal 6-thia substituted analog, 14-[18F]fluoro-6-thia-heptadecanoate(FTHA), to the most recently developed oleate-based FAO analog, 18-[18F]fluoro-4-thia-oleate (FTO). Itis shown that small changes in thia fatty acid analog structure can exert profound differences in thebiodisposition and specificity of these probes to indicate myocardial FAO, particularly in conditions ofoxygen deprivation. The potential of these probes for imaging of FAO in cancer is supported by initialuptake studies in cultured cancer cells. Thus, 18F-labeled thia fatty acid analogs have significant potentialto play an important role as clinical PET probes of FAO in cardiovascular diseases, oncology, and futureanticipated applications in endocrinology and neurology.

Keywords: PET; fatty acid oxidation; thia fatty acid; FTHA; FTP; FTO

■ Heart Metab. (2011) 51:15–19

Introduction

There remains a great demand for advancements of molecular imaging techniques to allownoninvasive assessment of biochemical processes for the metabolic characterization ofhuman diseases. Positron emission tomography (PET), and to a lesser extent, single photonemission computed tomography (SPECT) and magnetic resonance imaging (MRI) can provideimportant information on ion and metabolic fluxes in human tissues in a noninvasive and quan-titative manner. Energy metabolic pathways in heart and tumors are sensitively regulated andimmediately reflect any dietary and hormonal changes or abnormalities. Thus, monitoring ofMe

tabolicIma

ging

METABOLIC IMAGING - MUKESH K. PANDEY

Heart Metab. (2011) 51:15–19 15

substrate metabolism provides substantial informationabout the disease state and therapeutic progress. Themajor energy producing substrates are fatty acids(derived from plasma nonesterified fatty acids or locallipolysis from circulating lipoproteins), glucose, lactate,and ketone bodies. The glucose-based PET probe,FDG, has already set a benchmark in clinical care byindication of glucose utilization in tissues exhibitinghigh glycolytic rates, including a broad spectrum ofcancer types [1] and ischemic but viable myocardium[2]. Fatty acid-based SPECT probes are currently inroutine practice in Japan, but not in the United States.11C-labeled fatty acid PET radiotracers have been inuse for over 35 years, but have been mainly confinedto research studies due to their short physical half-life(20.4 m) and the complex nature of the quantitation ofthe dynamic PET images [3–7]. Quantitation of fattyacid oxidation (FAO) rates is not possible in ischemicmyocardium because the diffusion rates of the oxida-tion product 11CO2 and unoxidized 11C-palmitate areboth rapid, thereby confounding the model fit [6, 7].However, 11C-labeled fatty acid probes have value toindicate accumulation of exogenous fatty acids in themyocardial triglyceride pool, because this componentis characterized by a slow turnover. Over the last20 years, our group has been involved in the develop-ment of 18F-labeled thia fatty acids as metabolicallytrapped probes of FAO. The half-life of 18F (110 min-utes) allows for regional distribution of probes, whilethe presence of the sulfur heteroatom blocks theβ-oxidation of the fatty acid and also renders themolecule as a poor substrate for incorporation intocomplex lipids [8, 9]. In this mini-review, we intend tohighlight the incremental development and promisesoffered by thia fatty acids as potential probe moleculesfor FAO imaging in heart and cancer.

14-(R,S)-[18F]fluoro-6-thia-heptadecanoic acid(FTHA)

FTHA (Fig. 1) was the first-generation thia fatty acidprobe synthesized in 1990 and evaluated as a PETprobe of fatty acid metabolism in mice [8–9]. Afterintravenous administration of FTHA, the probe wasrapidly taken up by tissues as evident from its rapidblood pool clearance. The heart showed highestuptake of 39.8±3.0%ID/g at 5 minutes and subse-quently cleared with a biological half-life of about

2 hours. The heart:blood ratios of uptake were4.3±0.4, 20±6, 41±6, and 82±16 at 0.25, 1, 5and 60 minutes respectively. The myocardial trappingof 18F-radioactivity was drastically reduced bypretreatment of mice with the carnitine palmitoyl-transferase-1 (CPT-1) inhibitor POCA, showing CPT-1 dependent uptake of FTHA. FTHA has shown highmyocardial uptake, longer retention and rapid clear-ance from the bloodstream in humans also, making ita useful tracer of fatty acid metabolism for PET [9–10].In a mechanistic study, the net retention of FTHA wasdepressed in ischemic porcine myocardium, butremained unchanged in hypoxic myocardium [11].The lower retention in ischemia confirmed the applica-bility of FTHA for imaging ischemic myocardium [11],but the lack of sensitivity to lower FAO rates in hypoxicmyocardium motivated further tracer development toimprove specificity to monitor FAO rates. Neverthe-less, FTHA continues to be the most-investigated thiafatty acid PET probe, and has been used as a fatty aciduptake probe in human studies in heart [12–15], liver[16], skeletal muscle [12], and brain [17].

16-[18F] fluoro-4-thia-palmitate (FTP)

The lack of correspondence of FTHA trapping to FOArates in hypoxic myocardium was the driving force todevelop a second-generation thia fatty acid analog. In2000, the palmitate-based analog, 16-[18F] fluoro-4-thia-palmitic acid (FTP, Fig. 1), was identified as aFAO probe [18]. FTP was evaluated in a rat modelwith varying dietary conditions: fed (30minutes), fasted(30 minutes), fasted (120 minutes), and fasted (30 min-utes with CPT-I inhibitor “etomoxir”). The highest uptakewas observed in heart, liver and kidneys, regardless of

14(R,S)-[18F]fluoro-6-thia-heptadecanoic acid (FTHA)

16-[18F]fluoro-4-thia-palmitic acid (FTP)

18-[18F]fluoro-4-thia-oleic acid (FTO)

18F S OH

O

S OH

O

18F

S OH

O18F

S

Fig. 1 Structures of 18F thia fatty acids [18F]FTHA, [18F]FTP and[18F]FTO.


16 Heart Metab. (2011) 51:15–19

dietary status [18]. The new tracer FTP was also eval-uated in Langendorff perfused rat heart to study thekinetics and relationship of tracer retention to FAOrate in normoxic and hypoxic myocardium. FTP trap-ping in the rat myocardium correlated well with [9,10-3H]palmitate oxidation rates in both normoxic and hyp-oxic conditions. In the same model, the myocardialaccumulation of the 6-thia fatty acid analog, 17-[18

F]fluoro-6-thia-heptadecanoate, was insensitive tothe decrease in palmitate oxidation rate in hypoxichearts. Thus, the placement of the thia-substituent atthe fourth position of the fatty acid analog significantlyimproved the specificity of the probe for indication ofFAO. It was speculated that 4-thia fatty acid analogshave a more rapid turnover in pools of FAO intermedi-ates (i.e., acyl-CoAs and acyl-carnitines) than 6 thiafatty acid analogs that allow these pools to clear themyocardium if they are not further metabolized bymitochondrial β-oxidation. FTP showed high myocar-dial uptake and retention in porcine heart, consistentwith metabolic trapping seen in rats [18]. Detailedmechanistic studies of FTP uptake in isolated rathearts demonstrated FAO dependent metabolic accu-mulation that was in proportion to FAO rates as mea-sured by [9,10-3H]palmitate oxidation [18]. The ratio ofproportionality, which we denoted as the “lumpedconstant (LC)” in good analogy with FDG modelingmethodology, was decreased in hypoxic conditions,suggesting somewhat different affinities of FTP andpalmitate at transport and/or metabolic control pointsof fatty acid disposition within the myocyte [19]. How-ever, changes in the fatty acid composition of the per-fusion medium did not influence the LC, which mayallow FTP to be a prototypical fatty acid analog forindication of overall FAO [19]. The only major short-coming observed with FTP during these studies wasits lower retention over time in rat myocardium: myo-cardial clearance was about 70% at 2 hours relative to30 minutes [18].

18-[18F] fluoro-4-thia-oleate (FTO)

The suboptimal myocardial retention of FTP in rat myo-cardium prompted the development of the oleate-based probe 18-[18F]fluoro-4-thia-oleic acid (FTO,Fig. 1) [20]. FTO was recently synthesized and evalu-ated in rats with and without CPT-1 inhibition [20]. FTOwas motivated from oleate’s relative abundance inplasma, and a clinical study showing that dietary oleate

had a somewhat higher disposition relative to palmi-tate toward whole-body FAO [21]. Synthesis of thelabeling precursor for FTO required an 11-step syn-thetic process due to the inclusion of 18-bromo leav-ing group, 9-cis-double bond and the 4-thia substitu-ent [20]. FTO showed excellent myocardial imagingcharacteristics and superior myocardial retentionthan the previously developed tracers FTHA, andFTP. FTO showed three- to four-fold higher heart:background tissue radioactivity ratios than FTP. FTOuptake by heart was approximately reduced to 80%by pretreatment of CPT-1 inhibitor etomoxir indicatinghigh dependence on CPT-1 mediated mitochondrialtransport. The microPET images of FTO accumulationin the rat myocardium were clearly superior to thosewith FTP [20]. The folch-type extraction analysisshowed 70–90% of the 18F-radioactivity to beprotein-bound in heart, liver and skeletal muscle.These values were significantly higher than those forFTP and FTHA, suggesting higher specificity of mito-chondrial trapping. The preliminary data with FTO indi-cate this probe to be the most specific myocardial FAOprobe to date based on the thia fatty acid concept.

Mechanism of metabolic trapping of thia fatty acids

Based on our findings with FTHA, FTP and FTO,we propose a plausible mechanism for the uptakeand metabolic trapping in myocardium of terminally18F-labeled 4-thia fatty acid analogs, using FTO as anexample (Fig. 2). FTO enters into the cardiomyocyte viafatty acid transporters (CD36 /FATP (fatty acid trans-porter protein) [22]. In the cytosol, FTO is esterified toFTO-CoA by fatty acyl-CoA synthase (FACS). FTO-CoA is then transferred to carnitine via CPT-1. Theacyl carnitine is then shuttled across the inner

FTO Lipids

FTO

hydrolase

ACS

FTO-carnCAT

CPT-ICPT-II

MTPVLAD

β-ox.

Mitochondrion

Myocyte

FAT/CD36 FTO-CoA

FTO-carnFTO-CoA

LC-THIOL

PROTEIN-BOUND

(slow)

?

?

(slow)

(slow)

Fig. 2 Proposedmetabolism scheme for FTO in the cardiomyocyte.


Heart Metab. (2011) 51:15–19 17

mitochondrial membrane where it gets convertedback to FTO-CoA by CPT-2. In the mitochondrialmatrix, FTO-CoA may undergo two subsequentsteps of β-oxidation, forming the 3-hydroxy acyl-CoAmoiety, and then spontaneously decomposes to along-chain thiol, 14-[18F]fluoro-tetradecane-1-thiol,which in turn covalently or noncovalently binds to vari-ous mitochondrial proteins. Differential centrifugationidentified the mitochondrial fraction as containing thepredominant amount of retained 18F-radioactivity, whilenative-gel electrophoresis of heart extracts showedthat the 18F-radioactivity was associated with a broadspectrum of molecular weights, evidencing the non-specific nature of the protein binding (unpublisheddata). Thus, the accumulation of protein-bound18F-radioactivity in tissue is a direct readout for FAOof exogenous fatty acids. Slow clearance processesremain unclarified, particularly for FTHA and FTP. Twolikely mechanisms of clearance are slow releases ofcarnitine esters and/or β-oxidation metabolites (Fig. 2,dotted arrows).

FAO imaging in cancer

It is well recognized that not all tumor cell types utilizeglucose as primary energy substrate. FAO can providefor a predominant fraction of ATP production in tumorsthat have sufficient oxygen supply, such as prostatecancer [23]. Indeed, the low rate of glycolysis in early-stage prostate cancer severely limits the applicability ofthe FDG-PET method to staging of patients with newlydiagnosed disease. We have performed a preliminarystudy of the uptake of [18F]FTP in cultured 9L ratglioma, LNCaP human prostate and PC-3 humanprostate cancer cell lines [24] (Fig. 3). FTP was takenup avidly by the cancer cells. Hypoxic incubationresulted in an increase in overall uptake, consistentwith AMP-kinase activation of fatty acid transport[22]. Folch-type extractions were performed to indi-cate the FAO-dependent metabolic incorporation of 18

F-radioactivity into protein. The LNCaP cell line,which is derived from well-differentiated, androgen-dependent prostate cancer, showed the highest frac-tionation of FTP into the protein-bound phase, corrob-orating the findings of high levels of FAO expressed inearly-stage prostate cancer [23]. 18F-labeled straight-chain and β-methylated fatty acids have been shownto be taken up in rat tumor models [25, 26], howeverthese agents do not metabolically trap in a FAO-

dependent manner. To our knowledge, PET imagingstudies in cancer patients with fatty acid analogs haveyet to be done, but we believe there exists consider-able potential. ●

8

7

6

5

4

3

2

1

020 m 120 m 120 m20 m

normoxia hypoxia

A

Inco

rpo

rati

on

in e

ach

frac

tio

n (%

)

ProteinAqOrg

8

7

6

5

4

3

2

1

020 m 120 m 120 m20 m

B

Inco

rpo

rati

on

in e

ach

frac

tio

n (%

)

89

101112

76543210

20 m 120 m 120 m20 m

C

Inco

rpo

rati

on

in e

ach

frac

tio

n (%

)

Fig. 3 Uptake and disposition of [18F]FTP in 9L rat glioma (A),LNCaP human prostate (B) and PC-3 human prostate (C) cancercells. Uptake is expressed as percentage of total radioactivityadded to the culture medium prior to incubation. Cells were incu-bated for either 20 m or 120 m in normoxic (21% O2) or hypoxic(1% O2) conditions, then washed thrice with PBS to remove extra-cellular radiotracer. Folch-type extractions were performed in chloro-form/methanol (2:1) and 50% urea / 5% sulfuric acid, and radioac-tivity counted in the aqueous, organic and pellet (protein-bound)fractions. In general, hypoxic incubation resulted in increaseduptake of FTP and lower fractionation to the protein-bound frac-tions, indicating a relatively lower partitioning of the probe to FAO.High protein-bound fractions were seen in LNCaP cells, indicatinga high relative rate of FAO. Data from Bansal et al [24].


18 Heart Metab. (2011) 51:15–19

Acknowledgement This work was supported by NIH(R01 HL-63371, R01 CA108620).

References:

1. Poeppel TD, Krause BJ, Heusner TA, Boy C, Bockisch A,Antoch G (2009) PET/CT for the staging and follow-up ofpatients with malignancies. Eur J Radiol 70:382–392

2. Knuuti J, Tuunanen H (2010) Metabolic imaging in myocardialischemia and heart failure. Q J Nucl Med Mol Imaging 54:168–176

3. Shoghi KI, Finck BN, Schechtman KB, Sharp T, Herrero P,Gropler RJ, Welch MJ (2009) In vivo metabolic phenotypingof myocardial substrate metabolism in rodents: differentialefficacy of metformin and rosiglitazone monotherapy. CircCardiovasc Imaging 2(5):373–381

4. Schelbert HR, Schwaiger M (1986) PET studies of the heart. In:Phelps M, Mazziotta JC, Schelbert HR. eds. Positron emissiontomography and autoradiography. Principles and applicationsfor the brain and heart. New York: Raven Press, p. 581–661

5. Fox KAA, Abendschein DR, Ambos HD, Sobel BE, BergmanSR (1985) Effects of metabolized and nonmetabolized fattyacid from canine myocardium: implications for quantifyingmyocardial metabolism tomographically. Cire Res 57:232–243

6. Rosamond TL Abendschein DR. Sobel BE, Bergmann SR.Fox KAA (1987) Metabolic fate of radiolabeled palmitate inischemic canine myocardium: implications for positron emis-sion tomography. J Nucl Med 28:1322–1329

7. Wyns W, Schwaiger M, Huang SC, Buxton DB, Hansen H,Selin C, Keen R, Phelps ME, Schelbert HR (1989) Effectsof inhibition of fatty acid oxidation on myocardial kinetics ofC-labeled palmitate. Circ Res 65:l787–1797

8. DeGrado TR (1991) Synthesis of 14 (R, S)-[18F] fluoro-6-thia-heptadecanoic acid (FTHA). Journal of Labeled Compoundsand Radiopharmaceuticals 29:989–995

9. DeGrado TR, Coenen HH, Stocklin G (1991) 14(R, S)-[18F]fluoro-6-thia-heptadecanoic acid (FTHA): evaluation in mouse of a newprobe of myocardial utilization of long chain fatty acids. J NuclMed 32:1888–1896

10. Ebert A, Herzog H, Stain GL, Henrich MM, DeGrado TR,Coenen HH, Feinendegen LE (1994) Kinetics of 14(R, S)-fluorine-18-fluoro-6-thia-heptadecanoic acid in normal humanhearts at rest, during exercise and after dipyridamole injection.J Nucl Med 35:51–56

11. Renstrom B, Rommelfanger S, Stone CK, DeGrado TR,Carlson KJ, Scarbrough E, Nickles RJ, Liedtke AJ, HoldenJE (1998) Comparison of fatty acid tracers FTHA and BMIPPduring myocardial ischemia and hypoxia. J Nucl Med39:1684–1689

12. Takala TO, Nuutila P, Pulkki K, Oikonen V, Grönroos T, SavunenT, Vähäsilta T, Luotolahti M, Kallajoki M, Bergman J, Forsback S,Knuuti J (2002) 14(R,S)-[18F]Fluoro-6-thia-heptadecanoic acidas a tracer of free fatty acid uptake and oxidation in myocardiumand skeletal muscle. Eur J Nucl Med Mol Imaging 29:1617–1622

13. Wallhaus TR, Taylor M, DeGrado TR, Russell DC, Stanko P,Nickles RJ, Stone CK (2001) Myocardial free fatty acid andglucose use after carvedilol treatment in patients with conges-tive heart failure. Circulation 103:2441–2446

14. Taylor M, Wallhaus TR, DeGrado TR, Russell DC, Stanko P,Nickles RJ, Stone CK (2001) An evaluation of myocardialfatty acid and glucose uptake using PET with [18F]fluoro-6-thia-heptadecanoic acid and [18F]FDG in patients withcongestive heart failure. J Nucl Med 42:55–62

15. Turpeinen AK, Takala TO, Nuutila P, Axelin T, Luotolahti M,Haaparanta M, Bergman J, Hämäläinen H, Iida H, Mäki M,Uusitupa MI, Knuuti J (1999) Impaired free fatty acid uptakein skeletal muscle but not in myocardium in patients withimpaired glucose tolerance: studies with PET and 14(R,S)-[18F]fluoro-6-thia-heptadecanoic acid. Diabetes 48:1245–1250

16. Iozzo P, Turpeinen AK, Takala T, Oikonen V, Solin O, Ferran-nini E, Nuutila P, Knuuti J (2003) Liver uptake of free fattyacids in vivo in humans as determined with 14(R, S)-[18F]fluoro-6-thia-heptadecanoic acid and PET. Eur J Nucl MedMol Imaging 30:1160–1164

17. Karmi A, Iozzo P, Viljanen A, Hirvonen J, Fielding BA, VirtanenK, Oikonen V, Kemppainen J, Viljanen T, Guiducci L,Haaparanta-Solin M, Någren K, Solin O, Nuutila P (2010)Increased brain fatty acid uptake in metabolic syndrome.Diabetes 59:2171–2177

18. DeGrado TR, Wang S, Holden JE, Nickles RJ, Taylor M, StoneCK (2000) Synthesis and preliminary evaluation of (18)F-labeled 4-thia palmitate as a PET tracer of myocardial fattyacid oxidation. Nucl Med Biol 27:221–231

19. DeGrado TR, Kitapci MT, Wang S, Ying J, Lopaschuk GD(2006) Validation of 18F-fluoro-4-thia-palmitate as a PETprobe for myocardial fatty acid oxidation: effects of hypoxiaand composition of exogenous fatty acids. J Nucl Med47:173–181

20. DeGrado TR, Bhattacharyya F, Pandey MK, Belanger AP,Wang S (2010) Synthesis and preliminary evaluation of 18-18F-fluoro-4-thia-oleic Acid (FTO) as a PET probe of fatty acidoxidation. J Nucl Med 51:1310–1317

21. DeLany JP, Windhauser MM, Champagne CM, Bray GA(2000) Differential oxidation of individual dietary fatty acids inhumans. Am J Clin Nutr 72:905–911

22. Lopaschuk GD, Ussher JR, Folmes CDL, Jaswal JS, StanleyWC (2010) Myocardial fatty acid metabolism in health anddisease. Physiol Rev 90:207–258

23. Liu Y (2006) Fatty acid oxidation is a dominant bioenergeticpathway in prostate cancer. Prostate Cancer Prostatic Dis9:230–234

24. Bansal A, DeGrado TR (2010) Comparison of [14C]choline and[18F]fluoro-4-thia-palmitate (FTP) uptake in hypoxic 9L gliomacells. J Nucl Med 51(Suppl 1):198P

25. Kubota K, Takahashi T, Fujiwara T, Yamada S, Sato T, KubotaR, Iwata R, Ishiwata K, Ido T, Matsuzawa T (1991) Possibilityfor tumor detection with fatty acid analogs. Int J Rad ApplInstrum B 18:191–195

26. Greig NH, Nariai T, Noronha JG, Schmall B, Larson DM,Soncrant TT, Rapoport SI (1991) Brain tumor imaging in ratsusing the positron emitting fatty acid dl-erythro-9,10-[18F]difluoropalmitate. Clin Exp Metastasis 9:67–73


Heart Metab. (2011) 51:15–19 19
























































































































Metabolic modulation as a novelcancer treatment

Peter Dromparis, Gopinath Sutendra and Evangelos D. Michelakis, Department of Medicine,University of Alberta, Edmonton, Canada

Correspondence: Evangelos D. Michelakis, University of Alberta, Edmonton, T6G 2B7 Canada.Tel: +1 780-407-1576, fax: +1 780-407-6452, e-mail: [email protected]

AbstractThe Warburg effect refers to the cancer’s metabolic shift from mitochondrial oxidation to anaerobic gly-colysis, even when oxygen is present. Although this metabolic anomaly was long believed to be due toirreversible mitochondrial damage, it now appears mitochondria in cancer are actively suppressed.Since mitochondria are extensively integrated into the metabolic, signaling and apoptotic biology of thecell, mitochondrial suppression in combination with glycolysis-driven metabolism allows cancer to mas-terfully inhibit intrinsic cell death mechanisms and promote rapid proliferation even in sub-optimal condi-tions. In this review, we analyze cancer’s metabolic strategy and describe how active mitochondrialsuppression is compatible with both the genetic and evolutionary models of cancer. Furthermore, wediscuss several strategies for mitochondrial activation and review current pre-clinical and clinical studiesof agents showing promise in effectively and selectively targeting cancer.

Keywords: cancer, metabolic, apoptosis, dichloroacetate, glycolysis

■ Heart Metab. (2011) 51:20–26

Introduction

We currently perceive cancer as an accumulation of numerous chromosomal, genetic and bio-chemical abnormalities that result in dysfunctional cells. However, if so extensively damaged,how does cancer escape death, even flourish, in hypoxic and acidic conditions? Perhaps can-cer should be viewed as the “healthiest“ of cells, developing exquisite mechanisms to sup-press cell death (which is mostly regulated by mitochondria) and survive in otherwise uninhab-itable microenvironments, essentially achieving what medicine strives for—immortality. Oneway that cancer could achieve this is by developing strategies to effectively suppress mito-chondrial function. Here, we summarize cancer’s metabolic strategies to evade death and dis-cuss potentially effective and selective mitochondria-targeting therapies that induce apoptosisin cancerous but not non-cancerous tissues.

Mitochondria: a central role in cellular biology

Mitochondria produce energy through the oxidation of carbohydrates and lipids. In glucoseoxidation (GO), pyruvate is de-carboxylated to acetyl-CoA by the gate-keeping mitochondrialenzyme pyruvate dehydrogenase (PDH). In fatty acid oxidation (FAO), fatty acids enter themitochondria through carnitine palmitoyl-transferase-1 (CPT-1) and are also converted toacetyl-CoA (Fig. 1). The Krebs’ cycle extracts electrons derived from acetyl-CoA and feedsNe

wTh

erape

uticA

pproa

ches

NEW THERAPEUTIC APPROACHES - PETER DROMPARIS

20 Heart Metab. (2011) 51:20–26

them into the electron transport chain (ETC). Aselectrons transfer, H+ are extruded across the innermitochondrial membrane, creating an electrochemicalgradient, the mitochondrial membrane potential (ΔΨm).ΔΨm drives H+ re-entry through adenosine-5’-tri-phosphate (ATP) synthase, where they combine withoxygen to produce ATP, completing oxidative phos-phorylation. Thus, the flux of fuel into the mitochon-dria and respiration are closely linked with ΔΨm,which can be imaged with voltage-sensitive dyes andthus be a marker of mitochondrial activity. In addition,because ΔΨm is critical in setting the threshold forapoptosis as described below, fuel supply is alsointimately linked to cell death.

Mitochondria are major regulators of apoptosis.Upon opening of the mitochondrial transition pore

(MTP), pro-apoptotic mediators are released into thecytoplasm and ignite the apoptotic process. MTP is amega-channel that spans the mitochondrial mem-brane and is both redox- and voltage-sensitive, thusregulated in part by ΔΨm (Fig. 1) [1]. While ΔΨmdepolarization promotes MTP opening, hyperpolariza-tion increases its opening threshold, establishingan apoptosis-resistant state. Thus, ΔΨm could alsobe a surrogate for apoptosis resistance. Furthermore,since ΔΨm depends on metabolism, fuel supplyand apoptosis are intrinsically linked. As discussedbelow, most solid cancers are characterized by hyper-polarized mitochondria compared to non-canceroustissues, pointing to mitochondria’s critical role in thewell-known apoptosis resistance in cancer.

Mitochondria are extensively integrated into cellsignaling. In addition to H+ extrusion, the ETC alsogenerates superoxide (mitochondria-derived reactiveoxygen species –mROS). Unstable mROS, like super-oxide, can be dismutated into more stable mROS andleave the mitochondria to regulate redox-sensitive tar-gets like the voltage-gated potassium channels (Kv) inthe plasma membrane. As mitochondria are importantO2 sensors, this axis (ETC-mROS-Kv channels) is thebasis of hypoxic pulmonary vasoconstriction. mROSalso regulate other redox-sensitive targets like the tran-scription factors p53 [2] and hypoxia inducible factor-1α(HIF1α) [3], important in both vascular diseases andcancer (Fig. 1).

Mitochondria may also signal by releasingmetabolicsubstrates. Alpha-ketoglutarate (αKG) is a Krebs’cycle intermediate that, once in the cytosol, acts asa co-factor for prolyl-hydroxylases which degradeHIF1α [3]. In other words, this critical transcription fac-tor that drives angiogenesis in both cancer and vascu-lar diseases, is regulated by at least two mitochondria“signals”, mROS and αKG, both of which are linked tofuel processing and mitochondrial respiration.

Mitochondria can also sequester calcium. Being themost negatively charged organelles, they function asCa++ sinks, in a ΔΨm-dependent manner. Amongthe myriad of Ca++-dependent signaling processesare transcription factors that are integral in both vascu-lar disease and cancer, like the nuclear factor of acti-vated T-cells (NFAT) [3].

Therefore, suppression of mitochondrial functionhas the potential to influence a myriad of cellularprocesses critical in cancer biology. Suppressed

Glucose

DNAsynthesis

Fatty acids

Fatty acids

Citrate

mROS

∆Ψm

Acetyl-CoA

Randle Cycle

Acetyl-CoA

Membranesynthesis

PPP

GLY

Pyruvate

1

3

5

1 2

3

5

6

4

64

2

Glycoslysis Inhibitors Mitochondria Activators3-bromopyruvate, lonidamine, 2-deoxyglucose

PKM2 inhibitors

LDH inhibitors

PDK inhibitors (DCA)

FAS Inhibitors

FAO Inhibitors

PDH

PDK

DiffusibleMetabolitesETC

MTP

HIF1α

αKG

CTP-1

Krebs’Cycle

Lactate + HNAD +

+

+H

Fig. 1 Metabolic targets in cancer therapy. A metabolic shiftaway from mitochondrial oxidation to alternative pathways supportsrapid proliferation and apoptosis resistance. Mitochondrial suppres-sion is achieved in part by inhibition of PDH, which resultsin decreased mitochondrial signaling molecules (mROS and αKG)and mitochondrial membrane potential hyperpolarization, closingthe MTP. The decrease in signaling molecules can activate numer-ous transcription factors including HIF1α, which sustains this phe-notype. The glycolytic pathway (GLY) provides sufficient ATP andsupports DNA synthesis (via PPP) and other biosynthetic pathways.Numerous mutations in the Krebs’ cycle (see text) can result in anaccumulation of citrate which is critical for the de novo synthesisof fatty acids required for proliferating cells. Increasing mitochondrialfunction by promoting flux into the mitochondria has shown promiseas an anticancer strategy (see text). This can be done directly withPDK inhibitors like DCA (4), or indirectly with LDH inhibitors (shunt-ing pyruvate into the mitochondria) (3), PKM2 inhibitors (increasingpyruvate synthesis) (2), FAS inhibitors (5), and FAO inhibitors (6)(stimulate GO via the Randle cycle). Strategies that inhibit glycolysis(1) result in rapid depletion of cellular ATP and exhibit a lack of selec-tivity to cancer cells.


Heart Metab. (2011) 51:20–26 21

mitochondria would increase MTP opening thresh-olds, suppress mROS, inhibit Kv-channels, and acti-vate HIF1α and NFAT, all promoting a proliferative andanti-apoptotic state.

Mitochondria as integrators of early geneticand environmental signals in cancer

A number of common molecular abnormalitiesdescribed in cancer have direct metabolic and mito-chondrial effects that result in suppressed oxidativephosphorylation and mitochondrial function. Forexample, p53 loss-of-function, cancer’s most com-mon genetic anomaly, strongly promotes a glycolyticphenotype by upregulating glycolytic enzymes includ-ing phosphoglycerate mutase and the rate-limitingenzyme hexokinase (HK) [4]. p53 also regulates Tp53-induced glycolysis and apoptosis regulator (TIGAR),an enzyme with fructose-2,6-bisphosphatase activity[2]. When suppressed, fructose-2,6-bisphosphateaccumulates and potently simulates the rate-limitingglycolytic enzyme phosphofructokinase-1. Moreoverp53 regulates the expression of the ETC proteincytochrome-c oxidase and suppresses the expressionof glucose transporters (GLUTs) [2]. Thus, as a result ofp53 loss-of-function, ATP supply is maintained byelevated cytoplasmic glycolysis and glucose uptake,compatible with the increased 18-fluorodeoxyglucosepositron emission tomography (PET) signal in mosttumors compared to non-cancerous tissues.

Similarly, activation of Akt (or PTEN loss-of-function)not only enhances the activity of rate limiting glycolyticenzymes but also translocates GLUTs to the cell mem-brane. c-myc upregulates nearly all glycolytic enzymesincluding Hk [4, 5]. Intriguingly, in addition to regulatingcarbohydrate metabolism, several cytoplasmic glyco-lytic enzymes directly suppress apoptosis [5] or evenmodulate mitochondrial function. For example, acti-vated Hk translocates to the outer mitochondrial mem-brane and binds the voltage-dependent anion channel(VDAC) [6]. This prevents anion efflux contributing tothe ΔΨm hyperpolarization that characterizes mostcancers [7] and prevents MTP opening [6].

Mitochondrial enzyme mutations also impair mito-chondrial function and promote a glycolytic pheno-type. The Krebs’ cycle enzymes succinate dehydro-genase (SDH) [8], fumarate hydratase (FH) [9] andisocitrate dehydrogenase (IDH) [10, 11] are all associ-

ated with cancers like renal cell carcinoma, paragan-gliomas, or glioblastomas.

Thus, mitochondria appear to integrate a largenumber of diverse genetic signals that all result in acommon phenotype (i.e., hyperpolarized mitochondriawith suppressed function, upregulated glycolysis andsecondary signaling consequences) promoting prolif-eration and suppressing apoptosis. The evolving met-abolic theory of cancer that suggests that the mito-chondrial and metabolic remodeling promotes a pro-proliferative and apoptosis-resistant state is compati-ble with the genetic theory of cancer that has domi-nated the field for forty years but has failed to delivereffective and selective cancer therapies (except fewand isolated examples, such as imatinib mesylate).

Actively suppressed mitochondria are also compat-ible with the evolutionary model of cancer, which sug-gests that early carcinogenesis is hypoxia-driven [12].Here, decreased oxygen suppresses respiration andlimits proton influx through ATP synthase. This wouldresult in mitochondrial hyperpolarization and subse-quent mitochondria-dependent metabolic signalingand apoptosis suppression. HIF1α facilitates a shiftfrom mitochondrial oxidation to glycolysis, whichensures ATP production even in sustained hypoxia[12]. In addition to upregulating GLUTs, HIF1α upregu-lates pyruvate dehydrogenase kinase (PDK), theenzyme that tonically inhibits PDH, limiting pyruvateflux into the mitochondria (Fig. 1). When angiogenesisoccurs and tumor oxygen delivery is restored, theupregulated PDK maintains mitochondrial suppres-sion, and at this point, loss of secondary mitochondrialsignals (mROS, αKG) sustains HIF1α even in theabsence of hypoxia, maintaining a glycolytic state.

Mitochondrial suppression in cancer may have pro-proliferative effects in addition to the apoptosis inhibi-tion and HIF1α driven angiogenesis. Metabolitesthat are now not oxidized in the mitochondria areshifted toward biosynthetic pathways important forrapidly proliferating cells [13]. For instance, glucose-6-phosphate (the product of Hk) can be shunted intothe pentose phosphate pathway (PPP) wherebyribulose-5-phosphate is produced for nucleotide syn-thesis (Fig. 1). This pathway also produces NADPH,which protects against oxidative stress and is used inde novo lipid and amino acid synthesis [13]. Thus thebeneficial effects of an overall mitochondrial suppres-sion in cancer may extend to multiple levels.


22 Heart Metab. (2011) 51:20–26

Mitochondrial activators in cancer

Not surprisingly, therapies targeting single genetic ormolecular abnormalities are ineffective in most can-cers, which are commonly heterogeneous in nature.For example, there is significant diversity in glioblas-toma multiform (GBM), where a given tumor includesmultiple abnormalities and therefore is unlikely torespond to single target strategies [14]. A common fea-ture among all GBM (and most solid tumor) cells is met-abolic remodeling. In the 1920s, Warburg suggestedthat this feature of most cancers is caused by abnormalmitochondria. However, his theory was dismissed andmost considered mitochondria inhibition as a second-ary effect, possibly a result of oxidative damage.

DCA and preclinical studies

Recently, we showed that cancer cells treated withdichloroacetate (DCA), a PDK inhibitor (and thus aPDH activator), acutely increased GO [15]. BecauseGO occurs exclusively in the mitochondria, thesedata suggested that cancer mitochondria are perhapsfunctionally and reversibly suppressed, not perma-nently damaged, providing strong rationale for thedevelopment of similar drugs as novel cancer thera-pies. We first published the preclinical effects of DCAin cancer in 2007. We found that non-small cell lung,breast and GBM cancer cells had hyperpolarized ΔΨmcompared to non-cancerous cells and that DCA rap-idly depolarized mitochondria to normal levels, sug-gesting restoration of mitochondrial function [15]. Inkeeping with this, DCA decreased the glycolysis toGO ratio by enhancing GO and lowering lactate pro-duction. DCA induced mitochondria-dependent apo-ptosis and reduced proliferation both in vitro and invivo [15]. DCA-treated cancer cells had increasedmROS, activated Kv-channels and inhibited NFAT.Importantly, DCA did not affect mitochondria andtheir downstream targets in non-cancerous cells.DCA also decreased tumor size in a xenotransplantmodel (Fig. 2). Since then, many independent studieshave confirmed our findings (reviewed in [16]) in colon[17], prostate [18], endometrial [19] or metastaticbreast cancer [20]. In vivo, DCA was effective inattenuating aggressive metastatic breast cancer[20] (Fig. 2). In another study, DCA was chemicallylinked to cisplatin, a commonly used cancer drugthat induces DNA cross-linking [21]. This newmolecule, mitaplatin, was cancer selective, inducing

mitochondrial-dependent apoptosis and reversing theapoptosis resistance that limits cisplatin’s effectiveness.

DCA and early clinical experience

DCA has been used clinically for decades to symptom-atically treat congenital mitochondrial diseases includ-ing PDH deficiencies and their resulting lactic acidosis.This, along with promising pre-clinical findings pro-vided the rationale for early phase clinical trials in can-cer. The first such study was recently published exam-ining DCA’s effects in GBM, a highly vascular anddeadly brain cancer [22]. Examination of 49 freshlyexcised tumors showed mitochondrial hyperpolariza-tion compared to non-cancerous brain tissue, whichwas reversed with acute DCA, supporting reversiblemitochondrial inactivation in this tumor. The tumorsshowed high PDK2 expression, the most DCA-sensitive PDK isoform [23] (Fig. 3a). DCA was orallyadministered to five (previously treated) patients at adose of 6.25mg/kg bid giving plasma concentrationsof 0.266-0.626mM after at least three months,well within the range for PDK2 inhibition (Ki=0.2mM[23]). Importantly, none of the patients developedhematologic, hepatic, renal or cardiac toxicity. Periph-eral neuropathy was the only apparent toxicity, which

CT scan

Con

trol

∆Ψm PET Lung Metastasis

DC

A-t

reat

ed

Human non-smallcell lung cancer cells

Human non-small cell lungcancer xenotransplant model

Human breast cancerxenotransplant model

Fig. 2 Metabolic modulation of human cancers with dichloroace-tate depolarizes mitochondria and reduces tumor growth andmetastases in vivo. Human non-small cell lung cancer cells stainedwith the ΔΨm-sensitive dye (TMRM- red) and nuclear stain (Dapi-blue). Acute administration of DCA depolarizes ΔΨm, as measuredby a reduction in TMRM stain (left panel). Oral administration of DCAreduces tumor size (CT) and glucose uptake (18FDG-PET) in humannon-small cell lung cancer xenotransplant rat model (middle panel)(Modified with permission from [34]). Oral administration of DCAreduces lung metastases in a xenotransplant model with humanmammary carcinoma (right panel) (Modified with permissionfrom [20]).


Heart Metab. (2011) 51:20–26 23

was reversed at lower doses. This study’s majorstrength was comparing tumor tissue obtained beforeand after DCA therapy. Post-DCA, tumor tissuesexhibited higher PDH activity, confirming PDK2 inhibi-tion in vivo. Further examination revealed p53 activa-tion, HIF1α inactivation and decreased vascularity.Moreover, the tumors were less proliferative, moreapoptotic (Fig. 3a) and some, but not all, showed evi-dence of regression or stability (Fig. 3b) [22]. Evenmore exciting was the apparent effects on putativeGBM stem cells (GBM-SC; CD133+/Nestin+), whichare the most apoptosis-resistant cells and the causeof recurrence after remission. In line with this, the

GBM-SC had the highest ΔΨm within the tumor invivo. DCA induced apoptosis in these cells in vitro(Fig. 3c) and in vivo, suggesting that this traditionallyresilient population is also metabolically vulnerable [22].

Lactate dehydrogenase

LDH-A, a downstream target of HIF1α[3], convertspyruvate into lactate, restoring NAD+ to sustain glycoly-sis and preventing pyruvate oxidation. Inhibition ofLDH-A may restore mitochondrial function by promot-ingmitochondrial pyruvate metabolism, in a sensemim-icking DCA (Fig. 1). LDH-A inhibition decreases lactate,depolarizes mitochondria, increases respiration and

PDKII is higly expressed in human GBMDCA reduces proliferation

a

b cDCA stabilises a patient with advanced GBM

DCA induces apoptosis in putativepatient+dericed GBM stem cells

Pos

t-D

CA

Pre

-DC

A

Pos

t-D

CA

Pre

-DC

A

Pos

t-D

CA

Pre

-DC

A

DCA induces apoptosis in human GBM

TUNEL

CD133 NESTINPET-MRI MERGE

BASELINE

Surgery(x2)+DCA

+ 15 MONTHS

PET-MRI MERGE TUNEL MERGE

+DAPIPDKII PCNA MERGE+DAPI

100 µm 100 µm

20 µm

Fig. 3 Metabolic modulation with DCA attenuates proliferation and induces apoptosis in human GBM tissue and putative human GBM-SC, resulting in clinical stabilization in some patients. (a) Representative images of human GBM tissue taken before and after chronic oralDCA administration. DCA reduces proliferation (PCNA-red) and cell density (nuclei-blue). GBM tumors maintain high levels of PDKII that areexpressed throughout the tumor, even after DCA, suggesting maintained efficacy during chronic use (left panel). DCA induces apoptosis(TUNEL-green) in GBM tissue (right panel). (b) T1 axial MRI superimposed with 18FDG-PET scan in a GBM patient. Note the elevatedof 18FDG uptake in the tumor compared to non-cancerous brain tissue, indicating a high reliance on glucose. After two debulking surgerieswith concurrent DCA therapy for 15 months, there was a significant reduction in tumor size and FDG-PET signal intensity (arrows). (c) DCAinduces apoptosis (TUNEL-green) in human CD133+ (purple)/Nestin+(red) GBM-SC clusters, suggesting that this classically apoptosis-resistant population of cells is susceptible to metabolism-based therapies (modified with permission from [22]).


24 Heart Metab. (2011) 51:20–26

decreases proliferation in cancer cells [24, 25]. Animalsinjected with LDH-A-knockdown cancer cells had smal-ler tumors and improved survival [24], and inhibition ofLDH-A reduced tumor volumes in lymphoma and pan-creatic xenotransplant models without kidney, liver orblood toxicity [25]. These promising preclinical resultsand the relative selectivity for cancer suggest LDH-Ainhibitors may be a viable therapeutic intervention.

Pyruvate kinase M2 (PKM2)

Conversion of phosphoenol-pyruvate to pyruvateoccurs as the final rate-limiting glycolytic step. Thisreaction is catalyzed by pyruvate kinase, which existsin high (PKM1) and low activity (PKM2) isoforms. Therecent identification of PKM2 has generated excite-ment since this isoform is selective to rapidly proliferat-ing cells [13]. Although it appears paradoxical, this low-activity isoform may actually promote proliferation byreducing ATP production (preventing allosteric inhibi-tion of prior glycolytic reactions) and shunting metabo-lites into biosynthetic pathways [13, 26]. PKM2 wouldalso reduce mitochondrial pyruvate influx (Fig. 1). Thus,targeting PKM2 not only reduces biosynthetic pro-cesses, but may also promote GO, similar to DCA.Indeed, inhibition of PKM2 increases oxygen consump-tion in some cancers and limits tumor growth in xeno-transplant models [27]. The PKM2 data are compatiblewith the PDH/DCA work and the idea of an overallsuppression of mitochondrial function in cancer.

Fatty acid metabolism inhibitors

Cancer’s highly proliferative nature creates a largedemand for fatty acids for membrane synthesis. Ratherthan extracellular sources, cancer obtains fatty acids viade novo synthesis from the Krebs’ cycle intermediatecitrate [28]. Citrate is upstream of many Krebs’ cycleenzyme mutations commonly associated with canceras described above. Mutations in these enzymeswould lead to accumulation of upstream metabolites(i.e., citrate) for biosynthetic purposes (Fig. 1). Inhibitionof several fatty acid biosynthetic steps has shown effi-cacy in preclinical cancer models [29, 30]. Intriguingly,fatty acid oxidation (FAO) inhibition also inhibits cancergrowth [31], as it indirectly promotes GO through theRandle cycle (i.e. FAO-generated acetyl-CoA inhibitsPDH) [32]. In other words, this strategy is similar toDCA in terms of “refueling” mitochondria with pyruvateand reversing the mitochondrial suppression and thesecondary upregulation of glycolysis.

Glycolysis inhibitors

Cancer’s glycolytic environment provides sufficientenergy and building blocks for proliferation while simul-taneously suppressing apoptosis. Although direct gly-colysis inhibition seems logical, this strategy rapidly(and predictably) depletes ATP causing ATP-independent (necrotic) cell death, which is likely todamage non-cancerous tissues as well. Inhibiting gly-colysis may not be selective since non-cancerous tis-sues (i.e., skeletal muscle, brain, etc.) also rely on gly-colysis. In fact, several recent clinical trials of glycolysisinhibitors have not been successfully completed.Clinical trials for 2-deoxyglucose (Hk inhibitor) forprostate cancer (NCT00633087) and intracranialneoplasms (NCT00247403) have been suspended.Phase II/III trials for the Hk-inhibitor lonidamine inbenign prostate hyperplasia have also been termi-nated (NCT00435448, NCT00237536). Another Hkinhibitor, 3-bromopyruvate, which pre-clinical worksuggests induces necrosis in cancer, has also demon-strated significant toxicities in animal models at dosesonly slightly higher than therapeutic doses[33]. In otherwords, the non-selectivity of these drugs may limit theirtranslational potential. In summary, the metabolicmodulators that improve coupling between glycolysisand GO, “refueling” mitochondria and “normalizing”remodeled metabolism should not be confused withglycolysis inhibitors, which rather than normalizemetabolism, cause energy starvation (Fig. 1).

Conclusions

Revisiting Warburg’s original hypothesis that abnormalmetabolism causes cancer has provided newfoundoptimism. Once thought to be a secondary result ofupstream abnormalities, it now appears that cancer’smetabolic remodeling is integral for survival andexposes many therapeutic targets. Various strategiesthat restore mitochondrial metabolism have shownselectivity and efficacy in pre-clinical studies andneed to efficiently be moved into clinical trials. ●

References

1. Zamzami N, Kroemer G (2001) The mitochondrion in apopto-sis: how Pandora’s box opens. Nat Rev Mol Cell Biol 2:67–71

2. Vousden KH, Ryan KM (2009) p53 and metabolism. Nat RevCancer 9:691–700


Heart Metab. (2011) 51:20–26 25





3. Dromparis P, Sutendra G, Michelakis ED (2010) The role ofmitochondria in pulmonary vascular remodeling. J Mol Med88:1003–1010

4. Kim JW, Dang CV (2006) Cancer’s molecular sweet tooth andthe Warburg effect. Cancer Res 66:8927–8930

5. Kim JW, Dang CV (2005) Multifaceted roles of glycolyticenzymes. Trends Biochem Sci 30:142–150

6. Pastorino JG, Hoek JB, Shulga N (2005) Activation of glyco-gen synthase kinase 3beta disrupts the binding of hexokinaseII to mitochondria by phosphorylating voltage-dependentanion channel and potentiates chemotherapy-induced cyto-toxicity. Cancer Res 65:10545–10554

7. Chen LB (1988) Mitochondrial membrane potential in livingcells. Annu Rev Cell Biol 4:155–181

8. Douwes Dekker PB, Hogendoorn PC, Kuipers-Dijkshoorn N,Prins FA, van Duinen SG, Taschner PE, van der Mey AG, Cor-nelisse CJ (2003) SDHD mutations in head and neck paragan-gliomas result in destabilization of complex II in the mitochon-drial respiratory chain with loss of enzymatic activity andabnormal mitochondrial morphology. J Pathol 201:480–486

9. Tomlinson IP, Alam NA, Rowan AJ, Barclay E, Jaeger EE, Kel-sell D, Leigh I, Gorman P, Lamlum H, Rahman S, RoylanceRR, Olpin S, Bevan S, Barker K, Hearle N, Houlston RS,Kiuru M, Lehtonen R, Karhu A, Vilkki S, Laiho P, Eklund C,Vierimaa O, Aittomaki K, Hietala M, Sistonen P, Paetau A, Sal-ovaara R, Herva R, Launonen V, Aaltonen LA (2002) Germlinemutations in FH predispose to dominantly inherited uterinefibroids, skin leiomyomata and papillary renal cell cancer. NatGenet 30:406–410

10. Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Drig-gers EM, Fantin VR, Jang HG, Jin S, Keenan MC, Marks KM,Prins RM, Ward PS, Yen KE, Liau LM, Rabinowitz JD, CantleyLC, Thompson CB, Vander Heiden MG, Su SM (2009)Cancer-associated IDH1 mutations produce 2-hydroxy-glutarate. Nature 462:739–744

11. Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, YuanW, Kos I, Batinic-Haberle I, Jones S, Riggins GJ, Friedman H,Friedman A, Reardon D, Herndon J, Kinzler KW, VelculescuVE, Vogelstein B, Bigner DD (2009) IDH1 and IDH2 mutationsin gliomas. N Engl J Med 360:765–773

12. Gatenby RA, Gillies RJ (2004) Why do cancers have highaerobic glycolysis? Nat Rev Cancer 4:891–899

13. Vander Heiden MG, Cantley LC, Thompson CB (2009) Under-standing the Warburg effect: the metabolic requirements ofcell proliferation. Science 324:1029–1033

14. Wen PY, Kesari S (2008) Malignant gliomas in adults. N Engl JMed 359:492–507

15. Bonnet S, Archer SL, Allalunis-Turner J, Haromy A, BeaulieuC, Thompson R, Lee CT, Lopaschuk GD, Puttagunta L, HarryG, Hashimoto K, Porter CJ, Andrade MA, Thebaud B, Miche-lakis ED (2007) A mitochondria-K+ channel axis is suppressedin cancer and its normalization promotes apoptosis and inhi-bits cancer growth. Cancer Cell 11:37–51

16. Papandreou I, Goliasova T, Denko NC (2010) Anticancerdrugs that target metabolism: Is dichloroacetate the new par-adigm? Int J Cancer 128:1001–1008

17. Madhok BM, Yeluri S, Perry SL, Hughes TA, Jayne DG (2010)Dichloroacetate induces apoptosis and cell-cycle arrest incolorectal cancer cells. Br J Cancer 102:1746–1752

18. Cao W, Yacoub S, Shiverick KT, Namiki K, Sakai Y, PorvasnikS, Urbanek C, Rosser CJ (2008) Dichloroacetate (DCA) sensi-tizes both wild-type and over expressing Bcl-2 prostate can-cer cells in vitro to radiation. Prostate 68:1223–1231

19. Wong JY, Huggins GS, Debidda M, Munshi NC, De Vivo I(2008) Dichloroacetate induces apoptosis in endometrial can-cer cells. Gynecol Oncol 109:394–402

20. Sun RC, Fadia M, Dahlstrom JE, Parish CR, Board PG, Black-burn AC (2010) Reversal of the glycolytic phenotype bydichloroacetate inhibits metastatic breast cancer cell growthin vitro and in vivo. Breast Cancer Res Treat 120:253–260

21. Dhar S, Lippard SJ (2009) Mitaplatin, a potent fusion of cis-platin and the orphan drug dichloroacetate. Proc Natl AcadSci U S A 106:22199–22204

22. Michelakis ED, Sutendra G, Dromparis P, Webster L, HaromyA, Niven E, Maguire C, Gammer TL, Mackey JR, Fulton D,Abdulkarim B, McMurtry MS, Petruk KC (2010) Metabolicmodulation of glioblastoma with dichloroacetate. Sci TranslMed 2:31ra34

23. Bowker-Kinley MM, Davis WI, Wu P, Harris RA, Popov KM(1998) Evidence for existence of tissue-specific regulation ofthe mammalian pyruvate dehydrogenase complex. BiochemJ 329 (Pt 1):191–196

24. Fantin VR, St-Pierre J, Leder P (2006) Attenuation of LDH-Aexpression uncovers a link between glycolysis, mitochondrialphysiology, and tumor maintenance. Cancer Cell 9:425–434

25. Le A, Cooper CR, Gouw AM, Dinavahi R, Maitra A, Deck LM,Royer RE, Vander Jagt DL, Semenza GL, Dang CV (2010)Inhibition of lactate dehydrogenase A induces oxidative stressand inhibits tumor progression. Proc Natl Acad Sci U S A107:2037–2042

26. Vander Heiden MG, Locasale JW, Swanson KD, Sharfi H, Hef-fron GJ, Amador-Noguez D, Christofk HR, Wagner G, Rabino-witz JD, Asara JM, Cantley LC (2010) Evidence for an alterna-tive glycolytic pathway in rapidly proliferating cells. Science329:1492–1499

27. Christofk HR, Vander Heiden MG, Harris MH, Ramanathan A,Gerszten RE, Wei R, Fleming MD, Schreiber SL, Cantley LC(2008) The M2 splice isoform of pyruvate kinase is importantfor cancer metabolism and tumour growth. Nature 452:230–233

28. Deberardinis RJ, Sayed N, Ditsworth D, Thompson CB (2008)Brick by brick: metabolism and tumor cell growth. Curr OpinGenet Dev 18:54–61

29. Hatzivassiliou G, Zhao F, Bauer DE, Andreadis C, Shaw AN,Dhanak D, Hingorani SR, Tuveson DA, Thompson CB (2005)ATP citrate lyase inhibition can suppress tumor cell growth.Cancer Cell 8:311–321

30. Pizer ES, Thupari J, Han WF, Pinn ML, Chrest FJ, FrehywotGL, Townsend CA, Kuhajda FP (2000) Malonyl-coenzyme-Ais a potential mediator of cytotoxicity induced by fatty-acidsynthase inhibition in human breast cancer cells and xeno-grafts. Cancer Res 60:213–218

31. Samudio I, Harmancey R, Fiegl M, Kantarjian H, Konopleva M,Korchin B, Kaluarachchi K, Bornmann W, Duvvuri S, Taegt-meyer H, Andreeff M (2010) Pharmacologic inhibition of fattyacid oxidation sensitizes human leukemia cells to apoptosisinduction. J Clin Invest 120:142–156

32. Randle PJ (1998) Regulatory interactions between lipids andcarbohydrates: the glucose fatty acid cycle after 35 years. Dia-betes Metab Rev 14:263–283

33. Chang JM, Chung JW, Jae HJ, Eh H, Son KR, Lee KC, ParkJH (2007) Local toxicity of hepatic arterial infusion of hexo-kinase II inhibitor, 3-bromopyruvate: In vivo investigation innormal rabbit model. Acad Radiol 14:85–92

34. Michelakis ED, Webster L, Mackey JR (2008) Dichloroacetate(DCA) as a potential metabolic-targeting therapy for cancer.Br J Cancer 99:989–994


26 Heart Metab. (2011) 51:20–26































































































































Cancer and inhibition of fatty acid oxidationAlda Huqi, Cardiovascular Medicine Division, Cardio Thoracic Department, University of Pisa,

Pisa, Italy, and University of Alberta, Edmonton, Alberta, Canada

Correspondence: Dr. Alda Huqi, Cardiovascular Medicine Division, Cardio Thoracic Department, University of Pisa,Via Paradisa, 2, 56100 Pisa, Italy. And: 423 Heritage Medical Research Center, University of Alberta,

Edmonton, Alberta T6G 2S2, Canada.Tel: +39 32972 56426; e-mail: [email protected]

AbstractRecently, a “metabolic transformation” property with increased glycolytic rates (Warburg effect) wasdescribed for cancer cells. It has been shown that there is an active interaction between mutations ofcrucial enzymes involved in metabolic pathways and of oncogenes, whose effector proteins also exertmetabolic functions. Interestingly, because many of these metabolic changes are confined to themutant clones, targeted therapy appears very promising in terms of both safety and efficacy. Pharma-cological inhibition of glycolysis by dichloroacetate (DCA) significantly inhibits proliferation of cancercells. However, because of the pharmacokinetic characteristics, the plasma concentration levels and,therefore, toxic effects (i.e., hepatotoxicity and neurotoxicity) of DCA are difficult to predict. Similar met-abolic modulation and antiproliferative effects can be obtained through inhibition of fatty acid oxidation.Such agents have been safely used in other clinical conditions such as heart disease and, as such,deserve further testing in cancer therapy.

Keywords: cancer, Warburg effect, metabolic modulation therapy, fatty acid oxidation

■ Heart Metab. (2011) 51:27–30

Cancer continues to ferociously strike humanity. This has led to a major effort to developnew therapies to treat cancers. While major advances have been made in the treatment

of this debilitating disease, it still remains a major cause of mortality in our society. As a result,new therapies to treat and prevent cancer are being sought.

Over the past decades, cancer studies have mainly focused on identifying gene and proteinnetworks involved in the regulation of cell growth, differentiation, and death. However, althoughthe genetic mutations may be specific to the tumor, the downstream effectors are also found innormal tissues, therefore resulting in a narrow range of effective and safe molecular targets.Recently, a “metabolic transformation” property, amenable to metabolic modulation therapy,was described for cancer cells. Importantly, because many of these metabolic changes areconfined to the mutant clones, targeted therapy appears very promising in terms of both safetyand efficacy.

Aerobic glycolysis (Warburg effect) [1] constitutes a common feature of cancer cells and, assuch, appears to confer resistance to cell death. Importantly, increases in glycolytic rates incancer cells indirectly reflect also a decrease in the “mitochondrial energy production state” ofcancer cells. Similar to normal growth, apoptosis is an encoded cellular phase that impliesadequate mitochondrial energy production and signaling, which ultimately causes the release

Focu

sonVa

starel

®MR

FOCUS ON VASTAREL® MR - ALDA HUQI

Heart Metab. (2011) 51:27–30 27

of pro-apoptotic factors. Alterations in cellular meta-bolic pathways appear therefore to play an importantrole in conferring cancer cells their distinctive “immor-tality status”. In accordance with these observations,genetic analyses have shown that cancer cellsdisplaying these features have mutations of cru-cial enzymes involved in metabolic pathways (i.e.,phosphoinositol-3-kinase [PI3K], Akt, AMP-activatedprotein kinase [AMPK] and its upstream kinase LKB1,etc.) [2]. In addition, oncogenes whose effector pro-teins also exert metabolic functions (i.e., p53 targetTP53-induced glycolysis and apoptosis regulator[TIGAR], anti-apoptotic protein bcl-2, etc.) also showmutations [3]. Such findings indicate that metabolicand apoptotic pathways are more than two associ-ated phenomenon and therefore represent the scien-tific basis for considering metabolic modulation as atreatment strategy for cancer.

With regard to metabolic transformation, the keyevents that cause cancer cells to display the charac-teristic altered metabolic features are the “pseudo-hypoxic” mitochondrial oxygen sensing with abnormalactivation of the hypoxia-induced factor 1-alpha (HIF1-alpha). Increased HIF1-alpha expression activates aseries of glycolytic genes, and also suppresses theglucose oxidation activity of the mitochondria by trans-activating the pyruvate dehydrogenase (PDH) kinaseenzyme (PDK), which phosphorylates and inhibits thePDH complex [4]. PDH is a key enzyme in controllingthe rate of glucose oxidative metabolism in that it cat-alyzes the irreversible oxidation of pyruvate, yieldingacetyl-CoA and CO2 and, therefore its inhibition cre-ates a glycolytic shift of glucose metabolism (Fig. 1).This phenomenon is associated with othermitochondrial-metabolic abnormalities that include:mitochondrial hyperpolarization, decreased superox-ide dismutase-2 (SOD2) with reduced production ofreactive oxygen species (ROS), and decreased Kv(voltage gated) 1.5 expression. Loss of Kv 1.5 depo-larizes the membrane and elevates cytosolic K+ andCa2+, leading to Ca2+-calcineurin dependent activationof the proliferative transcription factor (NFAT) and inhi-bition of caspases through elevation of cytosolic K+.

The relevance of these changes in the coupling ofglycolysis to glucose oxidation on cellular proliferationcapacity is confirmed by the inhibitory effects on cellproliferation observed through pharmacological inhibi-tion of PDK. Dichloroacetate (DCA) is the most fre-

quently tested pharmacological agent for this purpose(Fig. 2). Besides metabolic changes, DCA restoresmitochondrial ROS production capacity, releases

CYTOPLASM

MITOCHONDRIA

Apoptoticfactors Apoptosis

TCAcycle

CPT1

CD-36HIF-1α

ACL

PDK

PDH

PFK-1PFK-2

GLUT4/HK

FA-CoAsynthase

B-oxidatione

Acetyl-CoA/NADH FAcyl- CoA

Acetyl-CoA

FAcyl-CoA

Fatty acids

ACC MCD

Malonyl-CoA

Acetyl-CoA

Glucose

Glu-6-P

Fru-1,6-P2

Gly

coly

sis

Lactate Pyruvate

Fig. 1 Metabolic transformation of cancer cells. Hypoxia inducedfactor-1α (HIF-1α) causes a pathological activation of pyruvatedehydrogenase (PDH) kinase (PDK), which in turn inhibits PDHactivity, therefore limiting oxidation of pyruvate, which is shiftedtowards lactate production. A reduced glucose oxidation-derivedacetyl-CoA has a crucial role in the normal signaling of mitochondriafor the release of apoptotic factors. ACC acetyl-CoA carboxylase,ACL ATP-citrate lyase, CPT1 carnitine palmtoyltransferase 1,GLUT4 glucose transporter-4, Glu-6-P glucose-6-phosphate, HKhexokinase,MCDmalonyl-CoA decarboxylase, PFK phospho fructokinase, TCA tricarboxylic acid

ACC MCD

Malonyl-CoA

Acetyl-CoA

CYTOPLASM

MITOCHONDRIA


TCAcycle

CPT1

CD-36HIF-1α

ACL

PDK

PDH

PFK-1PFK-2

GLUT4/HK

FA-CoAsynthase

B-oxidatione


Acetyl-CoA

FAcyl-CoA

Fatty acidsGlucose DCA

Glu-6-P

Fru-1,6-P2

Gly

coly

sis

Lactate Pyruvate

Fig. 2 Treatment of cancer cells with dichloroacetate (DCA). DCAinhibits pyruvate dehydrogenase (PDH) kinase (PDK), thereby caus-ing a relief of PDH, which catalyzes the decarboxylation of pyruvateto acetyl-CoA. In this way there is a decrease in glycolysis and anincrease in apoptotic capacity. ACC acetyl-CoA carboxylase, ACLATP-citrate lyase, CPT1 carnitine palmtoyltransferase 1, GLUT4transporter-4, Glu-6-P glucose-6-phosphate, HK hexokinase,HIF-1α hypoxia induced factor-1α, MCD malonyl-CoA decarboxyl-ase, PFK phospho fructo kinase, TCA tricarboxylic acid


28 Heart Metab. (2011) 51:27–30

pro-apototic factors [5] and, ultimately, decreases cellsurvival and increases apoptosis rates [4]. However,despite these promising cytological effects, the poten-tial use of DCA in cancer patients is regarded withsome degree of skepticism. In fact, when administeredfor other medical conditions (such as lactic acidosis,sepsis, burns, etc.), DCA has been reported to causeimportant adverse effects such as hepatotoxicity andneurotoxicity (although to some degree, these toxiceffects are reversible with short-term administration)[6]. Such effects would be particularly relevant in can-cer patients who are frequently on concomitant neuro-toxic chemotherapy. Moreover, since multiple admin-istrations of DCA leads to inhibition of hepatic enzymesresponsible for its own metabolism, plasma concen-tration levels and therefore toxic effects of the drugare difficult to predict [7].

DCA is not the only available agent to stimulate PDHactivity. The effects of DCA to increase glucose oxida-tion rates can be achieved by inhibiting mitochondrialfatty acid oxidation rates [8]. The reciprocal relation-ship between glucose and fatty acid metabolism ispart of a phenomenon known as the “Randle cycle.”Since glucose oxidation and fatty acid oxidation bothproduce mitochondrial acetyl-CoA, the rate of eachother’s activity has a direct reciprocal effect on therates of others pathway. Therefore stimulation of glu-cose oxidation through direct inhibition of PDK activitycan similarly be obtained by inhibiting fatty acid oxida-tion. This has been the rationale for the developmentand use of fatty acid oxidation inhibitors in other con-ditions such heart disease. However, at this point onemay argue: how can this approach be transferred tocancer cells when the latter have been defined by rely-ing on glycolysis rather than oxidative metabolism?

Nevertheless, although still recognizing glycolysis asthe main source for energy production of cancer cells,increasing the coupling of glycolysis to glucose oxida-tion by switching existing mitochondrial oxidativemetabolism from fatty acids to glucose, rather than areduced overall oxidative capacity of the mitochondria,has been proposed [9]. Therefore, similar to cardiacdisease, stimulation of glucose oxidation through inhi-bition of fatty acid oxidation may be a reasonableapproach for cancer therapy (Fig. 3). In support ofthis concept, fatty acid inhibitors such as etomoxirand ranolazine have been shown to have beneficialeffects in leukemia cells [10]. Trimetazidine, another

fatty acid oxidation inhibitor, was recently shown todose dependently induce cancer cell apoptosis inexperimental conditions [11]. Similarly, Zhou et alshowed that increasing malonyl-CoA levels (a keyenzyme that inhibits the transfer of fatty acids for oxi-dation into the mitochondria) through inhibition ofmalonyl-CoA decarboxylase (MCD), can decreaseproliferation of human breast cancer cells [12]. How-ever, the pro-apoptotic effects in the latter two studieswere mainly attributed to modifications in fatty acidsynthesis (FA) that occurred in addition to trimetazidineand the MCD inhibitor. Indeed, besides increased gly-colysis, cancer cells have been also shown to displayincreased de novo FA synthesis [13] with high levels ofthe lipogenic enzymes ATP-citrate lyase (ACL), acetyl-CoA carboxylase (ACC), and fatty acid synthase (FAS).However, modulation of FA synthesis has yielded con-trasting results. Likewise, in the study by Andela et al[11], trimetazidine was tested in addition to a peroxi-some proliferator receptor gamma (PPARγ) agonist,which, among other effects, causes an increase in FAsynthesis. On the contrary, in the study by Zhou et al,

ACC MCD

Malonyl-CoA

Acetyl-CoA

CYTOPLASM

MITOCHONDRIA


TCAcycle

CPT1

CD-36HIF-1α

ACL

PDK

PDH

PFK-1PFK-2

GLUT4/HK

MCDinhibitor

FA-CoAsynthase

B-oxidatione


Acetyl-CoA

FAcyl-CoA

Fatty acidsGlucose

Glu-6-P

Fru-1,6-P2

Gly

coly

sis

Lactate Pyruvate

Trimetazidine

Fig. 3 Treatment of cancer cells with fatty acid oxidation inhibitoryagents. Malonyl-CoA (an endogenous inhibitor of CPT1) levelsdepend on acetyl-CoA carboxylase (ACC) (involved in its produc-tion) and malonyl-CoA decarboxylase (MCD) (involved in its degra-dation) activities. Inhibition of MCD causes an increase in malonyl-CoA, thereby inhibiting the entry of fatty acyl-CoAmoieties for oxida-tion into mitochondria. Trimetazidine exerts its functions more dis-tally, inhibiting one of the enzymes of the beta-oxidative process.The common effect is a reduced pyruvate dehydrogenase (PDH)kinase (PDK), PDK activity and therefore an enhanced glucoseoxidative capacity. ACL ATP-citrate lyase, CPT1 carnitine palmtoyl-transferase 1, GLUT4 glucose transporter-4, Glu-6-P glucose-6-phosphate, HK hexokinase, HIF-1α hypoxia induced factor-1α,PFK phospho fructo kinase, TCA tricarboxylic acid


Heart Metab. (2011) 51:27–30 29

the MCD inhibitor was tested with cerulin, a FAS inhib-itor [12]. These contrasting results in the modulation ofFA synthesis may, to some degree, reflect the tissue-specific (in this case cancer cell type) expression ofenzymes involved in the regulation of FA. However,both studies showed clear reduction of cellular prolif-erative capacity, in this way highlighting the potentialrelevance of fatty acid inhibition and therefore, as men-tioned, relief of glucose oxidation, as a therapeuticapproach to treat cancer. This is ultimately in line withthe beneficial results obtained from direct stimulationof glucose oxidation with DCA. Importantly, theincreased apoptotic rates are confined only to cancerand not to normal control cellular lines. Moreover, tri-metazidine has been widely tested in human clinicaltrials, and shows a high safety profile and no majoradverse events. In this particular setting, the beneficialresults were observed in experimental conditions anddo not represent the current indication of the product(Vastarel® MR), which is an established therapy for themanagement of patients with myocardial ischemia.However, trimetazidine has been shown to havecardioprotective effects on cancer patients withanthracycline-induced acute cardiotoxity [14, 15].Alternatively, MCD inhibitors, another drug class thatinhibit fatty acid oxidation, can be further studied incancer setting.

In conclusion, altering glucose and fatty acid oxida-tion in cancer cells provides a potential new approachto treat cancer. The appealing potential of controllingcell proliferation by metabolic modulation therapy hasengendered new hopes on the horizon. Increasing glu-cose oxidation rates, through either a direct inductionof glucose oxidation or, a reduction in the fatty acidoxidation rates appears to be a logical approach totreat cancer. DCA, the main direct glucose oxidationstimulator available, has shown to have adverseeffects that may be particularly relevant in patientswho receive concomitant chemotherapeutic agents.On the other hand, inhibitory agents of fatty acid oxi-dation appear to have a safer therapeutic profile andmay therefore represent a valid alternative to DCA.

Given the quickly fatal natural history of the diseaseand the availability of these agents, efforts to translatethese findings in clinical practice are urgently needed. ●

References

1. Warburg O (1956) On respiratory impairment in cancer cells.Science 124(3215):269–270

2. Kim JW, Dang CV (2006) Cancer’s molecular sweet tooth andthe Warburg effect. Cancer Res 66(18):8927–8930

3. Bensaad K et al (2006) TIGAR, a p53-inducible regulator ofglycolysis and apoptosis. Cell 126(1):107–120

4. Kim JW et al (2006) HIF-1-mediated expression of pyruvatedehydrogenase kinase: a metabolic switch required for cellularadaptation to hypoxia. Cell Metab 3(3):177–185

5. Bonnet S et al (2006) An abnormal mitochondrial-hypoxiainducible factor-1alpha-Kv channel pathway disrupts oxygensensing and triggers pulmonary arterial hypertension in fawnhooded rats: similarities to human pulmonary arterial hyperten-sion. Circulation 113(22):2630–2641

6. Kurlemann G et al (1995) Therapy of complex I deficiency:peripheral neuropathy during dichloroacetate therapy. Eur JPediatr 154(11):928–932

7. Cornett R et al (1999) Inhibition of glutathione S-transferasezeta and tyrosine metabolism by dichloroacetate: a potentialunifying mechanism for its altered biotransformation and toxic-ity. Biochem Biophys Res Commun 262(3):752–756

8. Bonnet S et al (2007) A mitochondria-K+ channel axis is sup-pressed in cancer and its normalization promotes apoptosisand inhibits cancer growth. Cancer Cell 11(1):37–51

9. Samudio I, Fiegl M, Andreeff M (2009) Mitochondrial uncou-pling and the Warburg effect: molecular basis for the repro-gramming of cancer cell metabolism. Cancer Res 69(6):2163–2166

10. Samudio I et al (2010) Pharmacologic inhibition of fatty acidoxidation sensitizes human leukemia cells to apoptosis induc-tion. J Clin Invest. 120(1):142–156

11. Andela VB et al (2005) Inhibition of beta-oxidative respiration isa therapeutic window associated with the cancer chemo-preventive activity of PPARgamma agonists. FEBS Lett579(7):1765–1769

12. Zhou W et al (2009) Malonyl-CoA decarboxylase inhibition isselectively cytotoxic to human breast cancer cells. Oncogene28(33):2979–2987

13. Kuhajda FP (2000) Fatty-acid synthase and human cancer:new perspectives on its role in tumor biology. Nutrition 16(3):202–208

14. Pascale C et al (2002) Cardioprotection of trimetazidine andanthracycline-induced acute cardiotoxic effects. Lancet359(9312):1153–1154

15. Tallarico D et al (2003) Myocardial cytoprotection by trimetazi-dine against anthracycline-induced cardiotoxicity in anticancerchemotherapy. Angiology 54(2):219–227


30 Heart Metab. (2011) 51:27–30
















































A case of multifactorial late left ventriculardysfunction after cancer treatment

Chiara Lestuzzi, Centro di Riferimento Oncologico, Aviano, Italy

Correspondence: Chiara Lestuzzi, Cardiology, CRO, Centro di Riferimento Oncologico (National Cancer Institute),Via F. Gallini 2, Aviano (PN), Italy. Tel.: +39 0434659297; fax: +39 0434659572, e-mail: [email protected]

AbstractA case of hypokinetic cardiomyopathy in a young lady previously treated with anthracyclines and medi-astinal radiotherapy (RT) is described. The dysfunction was first considered due to anthracyclinescardiomyopathy, possibly with superimposed myocardits, and treated with enalapril and betblockersobtaining an improvement in left ventricular function. After a nine-year follow-up, left ventricular dysfunc-tion worsened and inducible silent myocardial ischemia was detected. When coronary artery disease(CAD) was diagnosed and cured by angioplasty and stenting, left ventricular function improved. In thispatient, CAD was likely secondary to both RT and dyslipidemia, and was probably a co-factor of leftventricular dysfunction. Patients treated with chest radiotherapy are at increased risk of CAD, and anyother risk factors (with particular attention to metabolic factors) should be thoughtfully searched andpromptly treated. This is mostly important in long-term survivors, and in patients treated more than30 years ago, when RT techniques were not yet planned to reduce radiation-induced heart disease.

Keywords: cardiotoxicity; radiotherapy; left ventricular dysfunction; Hodgkin’s disease; coronary arterydisease; dyslipidemia

■ Heart Metab. (2011) 51:31–33

Case report

A 40-year-old woman, a former smoker with family history of hypertension and diabetes, wasfirst seen in our outpatient clinic in June 2001. Her past clinical history included Hodgkin dis-ease stage III (supraclavear and mediastinal lymph nodes, spleen) at age 14 (in 1974), treatedwith splenectomy, six courses of chemotherapy with adriamycin (ADM), bleomycin, vinblastine,dacarbazine (ABVD) up to a total dose of 300 mg/sm of ADM, and mantle field radiotherapy(RT) (30 Gy). In January 2001, after traveling in Thailand, she was admitted to a general hospi-tal with fever and worsening dyspnoea. Chest x-rays revealed bilateral pneumonitis, immuno-logical tests resulted positive to mycoplasma pneumonia. At that time electrocardiography(ECG) revealed complete left bundle branch block (LBBB).

In June 2001, the patient was asymptomatic, without cardiac murmurs or other pathologicfindings. The ECG showed sinus rhythm, with LBBB. At echocardiography the left ventricle(LV) showed normal dimensions, with asynchronous contraction, septal hypokinesis andslightly reduced ejection fraction (EF): 43%. Cytomegalovirus antibodies were elevated; otherroutine blood tests were negative. Enalapril 5 mg/day was started. The patient took a treadmillstress test in August 2001 and attained a workload of 9.3 metabolic equivalent units (METS),without symptoms (ECG not evaluable for LBBB). The patient refused coronary angiography

Case

Repo

rtCASE REPORT - CHIARA LESTUZZI

Heart Metab. (2011) 51:31–33 31

and myocardial biopsy (to rule out coronary artery dis-ease and myocarditis), and nuclear imaging tests.

Up to 2008, the patient remained asymptomatic,with EF ranging from 51% to 56% and mild diastolicdysfunction (mitral E/A <1 at pulsed Doppler) on yearlyechocardiograms. Treadmill stress tests, repeatedevery 3 years, were always clinically negative, withthe patient able to perform >9 METS. Therapyincluded enalapril and beta-blockers at the highesttolerated dose (enalapril 7.5 mg, bisoprolol 2.5 mg:up-titration was difficult because of hypotension). In2004, blood tests revealed dyslipidemia (total choles-terol 350 mg/dL), and we prescribed atrovastatin20 mg, which was changed to rosuvastatin 10 mg/dayin 2006. In November 2007, the patient underwentmastectomy for left breast carcinoma, followed by hor-mone therapy (tamoxifen and triptorelin). After surgery,the patient spontaneously stopped rosuvastatin. InNovember 2009, she was still asymptomatic, withechocardiographic EF at 52%, but further impairmentof diastolic function (pseudonormal pulsed Dopplermitral flow pattern, E/E’= 22). A myocardial perfusionstress test showed reduced septal perfusion, partiallyreversible at rest. The patient finally accepted coronaryangiography in June 2010: a 95% ostial stenosis of theright coronary artery was detected and treated withangioplasty and stenting. At echocardiography, onemonth later both systolic and diastolic functions wereimproved (EF 57%, mitral E/A <1).

Comments

The cardiotoxicity of both ADM and RT may becomeclinically evident years after treatment, mostly inpatients treated at age <18 (as our patient was) andin those who received both treatments [1, 2]. Theintracellular accumulation of doxorubicinol, an ADMmetabolite, may act as a toxic reservoir potentiatingthe damage of further cardiac insults [3]. The heartwas formerly considered resistant to radiation, butfrom the late 1970s more and more reports aboutthe late sequelae of chest RT have been published[4–6]. Radiation-induced damage of normal tissuesdepends on dose and volume: according to predictivealgorithms, new RT techniques have been developedin the past 30 years in order to concentrate the radia-tion beam on the tumor, sparing the surroundingstructures [7–10]. Since therapeutic plans have chan-ged over time, the risk of cardiac damage is obviouslyhigher for patients treated earlier and lower for those

treated more recently [11]. Long-term adverse cardiaceffects of RT include dilated and/or restrictive cardio-myopathy, pericarditis, valvular disease, arrhythmiasand coronary artery disease (CAD) [12, 13]. The patho-physiological link between RT and accelerated athero-sclerosis is the radiation-induced inflammatory reactionfollowed by endothelial and fibroblast proliferation, lipiddeposition, fibrosis, and formation of inflammatory pla-que [14–16]. The clinical incidence of CAD is relevantafter 10 years, and increases thereafter. A study of1998 reports an increased prevalence (2.9% after10 years, 13.7% after 20 years, 24.7% after 25 years)in the subgroup of patients with additional classic car-diovascular risk factors only; in a study of 2003 hyper-tension and hypercholesterolemia were significant riskco-factors; in a study of 2007 stress-induced ischemiawas found in 18.4% of patients, and among them CADwas detected in 40% of those examined <10 years andin 58% of those examined >10 years after RT, withoutany difference between the patients with and thosewithout additional risk factors; in all studies the riskwas higher in males than in females [17–19].

Since the initial stress test in 2001 did not induceclinical ischemia in our patient, and she had noadditional risk factors, we considered LV dysfunctionprobably due to late ADM cardiomyopathy, possiblywith myocarditis as co-factor. On enalapril and beta-blockers the patient remained asymptomatic and hadclinically negative stress tests for years. When dyslipi-demia was first detected, we prescribed a statin,which was later discontinued by the patient. This deci-sion could have accelerated the progression of CAD.Hormonal therapy for breast cancer, on the contrary,had probably no role on coronary pathology: tamoxi-fen has a beneficial effect on lipid profiles and on car-diac risk; for triptorelin no effect is reported [20, 21].

A peculiarity of radiation-induced CAD is the fre-quency of ostial lesions and the high prevalence ofsilent ischemia, and this was also the case for ourpatient, who was asymptomatic but whose LV dia-stolic function had worsened over time, suggesting aprogression of CAD. In fact, when the ostial lesion wasdetected and cured with angioplasty, an improvementin systolic and diastolic function was obtained. Thus,cardiac dysfunction, in this particular patient, wasprobably multifactorial and included ADM myocardialtoxicity, radiation-induced myocardial damage (both asdirect insult and as increased sensitivity to other stressfactors) and coronary artery subclinical damage by RT

CASE REPORT - CHIARA LESTUZZI

32 Heart Metab. (2011) 51:31–33

leading to early development of CAD (with dyslipidemiaas co-factor), which further impaired LV function.

Conclusion

The late cardiotoxicity of anticancer therapies is rele-vant in long-term survivors, mostly in case of curablemalignancies of young patients, leading to a largecohort of subjects at increased risk of cardiac diseasein early adulthood. Antracyclines irreversibly damagethe myocites, and prevention of late cardiac diseaseimplies the prevention of other factors potentiallyaffecting cardiac function, as hypertension, diabetesand ischemic heart disease. RT may cause myocardial,valvular, pericardial and coronary artery damage. Inpatients having undergone mediastinal or left chestwall RT, prevention should also include the control ofrisk factors such as dyslipidemia. The highest riskfor radiation-induced heart disease in general and ofCAD in particular regards subjects treated before1980–1985, for several reasons: at that time radiationtechniques were not yet planned to prevent the possiblelong-term adverse cardiac effects, it was possible thatdyslipidemia was not promptly treated (its role as CADrisk factor was still under debate and powerful therapeu-tic agents like statines were not available yet), and theincidence of heart disease increases anyway over time.

All the patients who had been cured by RT for medi-astinal lymphomas (or other thoracic tumors) and forleft breast carcinomas should absolutely avoid smok-ing and should be frequently screened, and promptlytreated, for any other CAD risk factor, with particularattention to diabetes and dyslipidemia. Moreover, aregular follow-up with echocardiography and stresstest is recommended. ●

References

1. Berry GJ, Jorden M (2005) Pathology of radiation and anthra-cyclines cardiotoxicity. Pediatr Blood Cancer 44:630–637

2. Oeffinger KC, Mertens AC, Sklar CA, Kawashima T, HudsonMM, Meadows AT, Friedman DL, Marina N, Hobbie W,Kadan-Lottick NS, Schwartz CL, Leisenring W, Robison LL(2006) Childhood Cancer Survivor Study. Chronic healthconditions in adult survivors of childhood cancer. N Engl JMed12;355(15):1572-82; doi:10.1056/NEJMsa060185

3. Salvatorelli E, Menna P, Cascegna S, Liberi G, Calafiore AM,Gianni L, Minotti G (2006) Paclitaxel and docetaxel stimulationof doxorubicinol formation in the human heart: implications forcardiotoxicity of doxorubicin-taxane chemotherapies. J Pharma-col Exp Ther 318(1):424–433; doi: 10.1124/jpet.106.103846

4. McReynolds RA, Gold GL, Roberts WC (1976) Coronary heartdisease after mediastinal irradiation for Hodgkin’s disease. AmJ Med 60(1):39–45

5. Applefeld MM, Wiernik PH (1983) Cardiac disease after radia-tion therapy for Hodgkin’s disease: analysis of 48 patients. AmJ Cardiol 51:1679–681

6. McEniery PT, Dorosti K, Schiavone WA, Pedrick TJ, SheldonWC (1987) Clinical and angiographic features of coronary arterydisease after chest irradiation. Am J Cardiol 60(13):1020–1024

7. Kehwar TS (2005) Analytical approach to estimate normal tis-sue complication probability using best fit of normal tissue tol-erance doses into the NTCP equation of the linear quadraticmodel. J Cancer Res Ther 1(3):168–179

8. Chera BS, Rodriguez C, Morris CG, Louis D, Yeung D, Li Z,Mendenhall NP (2009) Dosimetric comparison of three differ-ent involved nodal irradiation techniques for stage II Hodgkin’slymphoma patients: conventional radiotherapy, intensity-modulated radiotherapy, and three-dimensional proton radio-therapy. Int J Radiat Oncol Biol Phys 75(4):1173–1180

9. Gagliardi G, Constine LS, Moiseenko V, Correa C, Pierce LJ,Allen AM, Marks LB (2010) Radiation dose-volume effects inthe heart. Int J Radiat Oncol Biol Phys 76(3 Suppl):S77–85

10. Purdie TG, Dinniwell RE, Letourneau D, Hill C, Sharpe MB (2011)Automated planning of tangential breast intensity-modulatedradiotherapy using heuristic optimization. Int J Radiat OncolBiol Phys doi: 10.1016/j.ijrobp.2010.11.016

11. Hudson MM, Poquette CA, Lee J, Greenwald CA, Shah A,Luo X, Thompson EI, Wilimas JA, Kun LE, Crist WM (1998)Increased mortality after successful treatment for Hodgkin’sdisease. J Clin Oncol 16(11):3592–600

12. Veinot JP, Edwards WD (1996) Pathology of radiation-inducedheart disease. A surgical and autopsy study of 27 cases. HumPathol 27:766–773

13. Adams MJ, Lipsitz SR, Colan SD, Tarbell NJ, Treves ST, DillerL, Grenbaum N, Mauch P, Lipshultz SE (2004) Cardiovascularstatus in long-term survivors of Hodgkin’s disease treatedwith chest radiotherapy. J Clin Oncol 22:3139–3148

14. Basavaraju SR, Easterly CE (2002) Pathophysiological effectsof radiation on atherosclerosis development and progression,and the incidence of cardiovascular complications. Med Phys29:2391–2403

15. Boerma M, Wang J, Wondergem J, Joseph J, Qiu X, KennedyRH, Hauer-Jensen M (2005) Influence of mast cells on structuraland functional manifestations of radiation-induced heart disease.Cancer Res 65(8):3100–3107

16. Stewart FA, Hoving S, Russell NS (2010) Vascular damage asan underlying mechanism of cardiac and cerebral toxicity inirradiated cancer patients. Radiat Res 174(6):865–869

17. Glanzman C, Kaufmann P, Jenni R, Hess OM, Huguenin P(1998) Cardiac risk after mediastinal irradiation for Hodgkin’sdisease. Radiother Oncol 46:51–62

18. Hull MC, Moerris CG, Pepine CJ, Mendenhall NP (2003) Valvulardysfunction and carotid, subclavian, and coronary artery diseasein survivors of Hodgkin lymphoma treated with radiation therapy.JAMA 290:2831–2837; doi: 10.1001/jama.290.21.2831

19. Heidenreich PA, Schnittger i, Strauss HW, Vagelos RH, LeeBK, Mariscal CS, Tate JT, Horning SJ, Hoppe RT, HancockSL (2007) Screening for coronary artery disese after mediasti-nal irradiation for Hodkin’s disease. J Clin Oncol 25:43–49;doi: 10.1200/JCO.2006.07.0805

20. Grey AB, Stapleton JP, Evans MC, Reid IR (1995) The effectof the anti-estrogen tamoxifen on cardiovascular risk factors innormal postmenopausal women J Clin Endocrinol Metab80(11):3191–3195

21. Cushman M, Costantino JP, Tracy RP, Song K, Buckley L,Roberts JD, Krag DN (2001) Tamoxifen and cardiac riskfactors in healthy women: Suggestion of an anti-inflammatoryeffect. Arterioscler Thromb Vasc Biol 21(2):255–261

CASE REPORT - CHIARA LESTUZZI

Heart Metab. (2011) 51:31–33 33


















































































The Warburg effect – aerobic glycolysisin proliferating cells

Jagdip S. Jaswal, Department of Pharmacology, Faculty of Medicine and Dentistry, University of Alberta,Edmonton, Alberta, T6G 2S2, Canada

Correspondence: Jagdip S. Jaswal, 423 Heritage Medical Research Centre, Department of Pharmacology,Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, T6G 2S2, Canada.

Tel: +1 780 492 8659; fax: +1 780 492 9753; e-mail: [email protected]

AbstractThe metabolic phenotype of proliferating cells is characterized by enhanced glycolysis uncoupled frompyruvate oxidation, but instead coupled to increased lactate release even in the presence of oxygen(Warburg effect/aerobic glycolysis). Alterations in the pathways regulating glycolysis and the oxidation ofglucose-derived carbon are responsible for maintaining this phenotype. This article reviews the mole-cular mechanisms inherent to the intermediary metabolism of glucose contributing to the Warburg effect.

Keywords: cellular proliferation; glycolysis; glucose oxidation; cellular biosynthesis

■ Heart Metab. (2011) 51:34–38

Introduction

In the 1920s Otto Warburg demonstrated that rapidly proliferating mouse ascites tumor cellspreferentially converted glucose to lactate even in the presence of oxygen (Warburg effect /aer-obic glycolysis) at a much greater rate than that observed in quiescent, differentiated cells [1].These observations were later extrapolated to proliferating primary lymphocytes [2], demon-strating that the metabolic phenotype of cell proliferation is attended by specific alterationsin the intermediary metabolism of glucose. Although glycolysis generates only 7–10% ofthe adenosine-5’-triphosphate (ATP) compared to the complete mitochondrial oxidation ofglucose-derived carbon, it does, nonetheless, efficiently generate substrates that meet thebiosynthetic requirements of cellular proliferation [3]. This article provides an overview of mecha-nisms intrinsic to the intermediary metabolism of glucose that contribute to the Warburg effect.

Cellular glucose metabolism in differentiated versus proliferative tissues

Differentiated tissue such as cardiac muscle derives greater than 90% of its ATP requirementsvia mitochondrial oxidative phosphorylation [4]. The reduced forms of nicotinamide adeninedinucleotide (NADH) and flavin adenine dinucleotide (FADH2) generated during the catabolicdegradation of glucose and fatty acids, as well as in the tricarboxylic acid (TCA) cycle deliverprotons and electrons to the electron transport chain, reducing molecular oxygen to water,and generating ATP. With regards to carbohydrate metabolism, the catabolism of glucose indifferentiated tissues occurs via two separate, but coupled processes—glycolysis and subse-quent glucose (i.e., pyruvate) oxidation. In sharp contrast, cellular metabolism in proliferativecells is characterized by accelerated rates of glycolysis that are uncoupled from the glucoseoxidation.

Refre

sher

Corner

REFRESHER CORNER - JAGDIP S. JASWAL

34 Heart Metab. (2011) 51:34–38

Glycolysis, localized to the cytosolic compartment,converts glucose to pyruvate, or lactate, in the pres-ence or absence of oxygen, respectively, and gener-ates 2 mol ATP/1 mol glucose. Glycolysis consistsof an ATP utilizing stage and a subsequent ATP gen-erating stage (Fig. 1). The first enzymatic reactioncommitting glucose to catabolism by glycolysis, cata-lyzed by 6-phosphofructo-1-kinase (PFK-1), is sub-ject to allosteric inhibition by ATP, citrate, and protons

[4]. The first reaction of the ATP generating stage ofglycolysis, catalyzed by glyceraldehyde-3-phosphatedehydrogenase (GAPDH), is coupled to the reductionof NAD+ to NADH. ATP itself is generated by reactionscatalyzed by phosphoglycerate kinase and pyruvatekinase. The continual regeneration of NAD+ is requiredto ensure that glycolysis is not limited. In differentiatedcells and tissues, under aerobic conditions, NAD+ isre-generated from NADH via the malate-aspartateshuttle and the mitochondrial electron transport chain;whereas, under anaerobic conditions lactate dehydro-genase (LDH) couples the reduction of pyruvate to lac-tate with the regeneration of NAD+.

The oxidation of glucose-derived carbon requires thetransport of pyruvate into the mitochondrial matrix via amonocarboxylate (MCT) transporter [5], and its subse-quent oxidative decarboxylation, catalyzed by the pyru-vate dehydrogenase (PDH) complex yielding acetyl-CoA [6]. The PDH complex consists of PDH, PDHkinase (PDK), and PDH phosphatase (PDHP). The fluxof pyruvate through PDH is subjected to regulation byboth substrate/product ratios and covalent mechan-isms. The increased generation of pyruvate, decreasedratios of NADH/NAD+ and acetyl-CoA/CoA, as wellas PDHP-mediated dephosphorylation increase fluxthrough PDH. Conversely, increased ratios of NADH/NAD+ and acetyl-CoA/CoA, as well as PDK-mediatedphosphorylation decreases flux through PDH, therebyrestricting the oxidation of glucose-derived carbonunits [6].

Underlying molecular mechanisms

The Warburg effect is a predominant metabolic phe-notype observed in proliferating cells. Detailed studiesin a variety of cancer cell types have revealed thatthis metabolic phenotype may arise from a numberof cellular alterations and occur downstream ofgrowth promoting signal transduction pathways(PI3K-Akt-mTOR), mutations in proto-oncogenes(Myc), in response to loss of function mutations oftumor suppressor genes (p53), and transcriptionalresponses to hypoxia-inducible factor-1α (HIF1α) [3].These alterations in cellular signaling elicit concertedeffects on the intermediary metabolism of glucose inboth the ATP utilizing and ATP generating stages ofglycolysis, and facilitate enhanced glycolysis charac-teristic of the Warburg effect.

ATP

ADP

ATP

ADP

NAD+

NADH

ADP

ATP

H2O

ADP

ATP

Glucose

Hexokinase

PhosphoglucoseIsomerase

PFK-1

Aldolase

GAPDH

ATP

Utilization

ATP

Generation

PhosphoglycerateKinase

PhosphoglycerateMutase

Enolase

PyruvateKinaseLDH

G-6-P

F-6-P

F-1,6-BP

DHAP GA-3-P

1,3-BPG

3-PG

2-PG

PEP

PyruvateLactate

NAD+ NADH

Fig. 1 Catabolism of glucose via glycolysis. Following facilitatedtransport across the plasma membrane, glucose is degraded bythe pathway of glycolysis. Glycolysis itself can be separated intoan ATP utilization stage and an ATP generating stage. The first con-sumes 2 ATP molecules in course of phosphorylating glucose andcleaving it to yield 2 molecules of GA-3-P. The ATP generating stageconverts 2 molecules of GA-3-P to either pyruvate of lactate, gener-ating 4 molecules of ATP. Glycolysis therefore generates a net2 moles of ATP per mole of glucose. Several of the enzymes directlyinvolved in glycolysis are up-regulated in order to support the War-burg effect in proliferating cells. 1,3-BPG 1,3-bisphosphoglycerate,DHAP dihydroxyacetone phosphate, F-6-P fructose-6-phosphate,F-1,6-BP fructose-1,6-bisphosphate, G-6-P glucose-6-phosphate,GA-3-P glyceraldehyde-3-phosphate, GAPDH glyceraldehyde-3-phosphate, LDH lactate dehydrogenase, PEP phosphoenolpyr-uvate, 2-PG 2-phosphoglycerate, 3-PG 3-phosphoglycerate


Heart Metab. (2011) 51:34–38 35

Growth factor-mediated activation of the PI3K-Akt-mTOR pathway via transcriptional and post-transcriptional effects increases the expression andcell surface localization of GLUT1 glucose transporters[7]. Similarly, loss of function mutations of the p53tumor suppressor de-represses the transcription ofGLUT1 and GLUT4 [8]. These alterations can increasecellular glucose uptake. Glucose entering the cytosoliccompartment is efficiently trapped as glucose-6-phos-phate (G6P) by hexokinase II, a HIF1α- and Myc-targetgene up-regulated in a variety of cancer cells [9]. Lossof p53 function also enhances flux through the ATPutilizing stage of glycolysis, secondary to the loss ofTP53-induced glycolysis and apoptosis regulator(TIGAR) [10]. TIGAR is functionally similar to thefructose-2,6-bisphospahatase (FBPase) domain ofthe bifunctional enzyme, 6-phosphofructo-2-kinase(PFK-2), which converts fructose-6-phosphate tofructose-2,6-bisphosphate, a potent allosteric stimu-lator of PFK-1. Thus, p53 can restrain flux throughPFK-1 via TIGAR-mediated degradation of fructose-2,6-bisphosphate [10], whereas loss of p53 functionvia loss of TIGAR can promote flux through PFK-1 [11].

ATP is generated by the phosphoglycerate kinaseand pyruvate kinase reactions of glycolysis and bothare of central importance in mediating the Warburgeffect. HIF1α and Myc increase the expression ofphosphoglycerate kinase in a cooperative manner[12], which contributes to increasing glycolytic flux.Pyruvate kinase (PK) catalyzes the final reaction of gly-colysis, coupling the production of pyruvate with theproduction of ATP. Proliferating cells, including cancercells predominantly express the PKM2 isoform [13].Interestingly, the affinity of dimeric PKM2 for its sub-strate PEP is low, and PKM2 is therefore nearly inac-tive at physiological/pathophysiological intracellularPEP concentrations [14]. These effects are associatedwith increased cellular ADP/ATP ratios, which enhanceflux through the PFK-1 reaction of glycolysis. Theseeffects may appear paradoxical, as the generation ofthe glycolytically derived pyruvate via the PK reaction islimited, which would be expected to decrease lactaterelease and hence decrease an important componentof the Warburg effect. However, a recent report iden-tifies a novel/alternative pathway of glycolysis in prolif-erating cells that circumvents this reaction and gener-ates pyruvate from PEP by transferring the phosphatefrom PEP to the catalytic histidine residue of the glyco-

lytic enzyme phosphoglycerate mutase [15]. The gen-erated pyruvate can subsequently be reduced to lac-tate and hence maintain the Warburg effect.

Mechanisms affecting pyruvate/lactate metabolismand lactate release are also required to support theWarburg effect. The expression of the LDH-A isoformis under the transcriptional control of both HIF1α andMyc [16]. LDH-A effectively converts glycolyticallyderived pyruvate to lactate. HIF1α also upregulatesthe expression of the plasma membrane MCT4 iso-form [17], as well as PDK1 in proliferating cells [9].Taken together, these metabolic alterations provide ameans to regenerate NAD+ (LDH-A) required for gly-colysis, facilitate lactate release (MCT4), and restrictglucose oxidation (PDK1) (Fig. 2).

The Warburg effect and anabolic metabolism

TheWarburg effect is of central importance in support-ing the requirements of cell division which requirea doubling of total biomass (including cellular nucleicacid, lipid, and protein contents) [3]. The non-oxidativearm of the pentose phosphate pathway (PPP) utilizesthe glycolytic intermediate, fructose-6-phosphate forthe synthesis of ribose-5-phosphate required fornucleotide biosynthesis [18]. Proliferating cancer cells

LDH-A

PHD

PDK-1

pyruvate

pyruvate

acetyl-CoA

lactate

MCT-4

MCT-1

cytosolic compartment

mitochondrial matrix

extracellular space

Fig. 2 Pyruvate/lactate metabolism and the Warburg effect. Pyru-vate is the usual end product of glycolysis in the presence of oxygen;however, the Warburg effect is characterized by the conversionof glucose to lactate, and the subsequent release of lactateto the extracellular space. Several alterations in the enzymesinvolved in the metabolism of pyruvate and lactate facilitate thisaspect of the Warburg effect. Increased expression of pyruvatedehydrogenase kinase-1 (PDK-1) restricts the mitochondrial oxida-tion of glucose-derived pyruvate. The increased expression of lac-tate dehydrogenase-A (LDH-A) preferentially converts pyruvateto lactate, while increased expression of monocarboxylate trans-porter 4 (MCT-4) increases the efflux of lactate from the cytosolto the extracellular space.


36 Heart Metab. (2011) 51:34–38

in particular also have high requirements for de novolipid synthesis, evidenced from the elevated expres-sion of lipogenic enzymes including ATP citrate lyase(ACL), acetyl-CoA carboxylase (ACC), and fatty acidsynthase (FAS) [19]. ACL can contribute to maintainingthe Warburg effect as it cleaves cytosolic citrate (origi-nating from the mitochondrial matrix), yielding oxaloac-etate and acetyl-CoA. To ensure that the requirementsfor de novo lipid synthesis do not deplete the TCAcycle, the cycle is replenished via the anaplerotic fluxof glutamine [20]. These effects of ACL decrease thecontent of cytosolic citrate, enhancing PFK-1 activity,as well as providing substrate (i.e., acetyl-CoA) for denovo lipid synthesis via ACC and FAS. The metabolismof cytosolic oxaloacetate can also contribute to support-ing the Warburg effect as malate dehydrogenase cou-ples the conversion of oxaloacetate to malate with theformation of NAD+. Malic enzyme subsequently couplesthe conversion of malate to pyruvate (which can subse-quently be converted to lactate and exported) with theproduction of nicotinamide adenine dinucleotide phos-phate (NADPH), which serves as a cofactor in both thePPP and de novo lipid synthesis [18]. These complimen-tary activities of the Warburg effect and several path-ways of anabolic cellular metabolism (Fig. 3) meet the

requirements of cellular proliferation, and in cancercells support tumor development and progression.

Conclusions

Alterations in the intermediary metabolism of glucose,exemplified by the Warburg effect are critical in sup-porting the biosynthetic requirements of cellular prolif-eration. The Warburg effect adapts cellular metabolicphenotype, such that enhanced flux through glycolysisprovides a number of critically important substrates forthe macromolecular synthesis of cellular biomass. Assuch a greater understanding of molecular mecha-nisms regulating the intermediary metabolism of glu-cose is relevant for the identification novel pharmaco-logical targets that can influence cellular proliferation.With regards to cancer cells, this may lead to thedevelopment of promising therapeutic agents to limittumor development and progression. ●

References

1. Warburg O (1956) On the origin of cancer cells. Science123:309–314

2. Wang T, Marquardt C, Foker J (1976) Aerobic glycolysisduring lymphocyte proliferation. Nature 261:702–705

3. DeBerardinis RJ, Lum JJ, Hatzivassiliou G., Thompson CB(2008) The biology of cancer: metabolic reprogramming fuelscell growth and proliferation. Cell Metab 7:11–20

4. Jaswal JS, Keung W, Wang W, Ussher JR, Lopaschuk GD(2011) Targeting fatty acid and carbohydrate oxidation - Anovel therapeutic intervention in the ischemic and failing heart.Biochim Biophys Acta doi: 10.1016éj.bbamcr.2011.01.015.

5. Halestrap AP, Price NT (1999) The proton-linked monocarbox-ylate transporter (MCT) family: structure, function and regula-tion. Biochem J 343(Pt 2):281–299

6. Sugden MC, Holness MJ (2003) Recent advances in mechan-isms regulating glucose oxidation at the level of the pyruvatedehydrogenase complex by PDKs. Am J Physiol EndocrinolMetab 284:E855–862

7. Edinger AL, Thompson CB (2002) Akt maintains cell size andsurvival by increasing mTOR-dependent nutrient uptake. MolBiol Cell 13:2276–2288

8. Schwartzenberg-Bar-Yoseph F, Armoni M, Karnieli E (2004) Thetumor suppressor p53 down-regulates glucose transportersGLUT1 and GLUT4 gene expression. Cancer Res 64:2627–2633

9. Kim JW, Gao P, Liu YC, Semenza GL, Dang CV (2007)Hypoxia-inducible factor 1 and dysregulated c-Myc coopera-tively induce vascular endothelial growth factor and metabolicswitches hexokinase 2 and pyruvate dehydrogenase kinase 1.Mol Cell Biol 27:7381–7393

10. Bensaad K, Tsuruta A, Selak MA, Vidal MN, Nakano K, BartronsR, Gottlieb E, Vousden KH (2006) TIGAR, a p53-inducibleregulator of glycolysis and apoptosis. Cell 126:107–120

11. Green DR, Chipuk JE (2006) p53 and metabolism: Inside theTIGAR. Cell 126:30–32

12. Qing G, Skuli N, Mayes PA, Pawel B, Martinez D, Maris JM,Simon MC (2010) Combinatorial regulation of neuroblastoma

NADP+

NADPH

NADH

NAD+

NADP+

NADPH

FAS

ACC

MHD

ME

ACL

LCFAs

citrate

acetyl-CoA malonyl-CoA

oxaloacetate

malate

pyruvate

Fig. 3 Cytosolic metabolism of citrate for de novo lipid synthesis.Citrate originating in the mitochondrial matrix can gain accessto the cytosolic compartment. In the cytosol, citrate can be cleavedby the lipogenic enzyme ATP citrate lyase, generating acetyl-CoAwhich itself can contribute to de novo lipid synthesis. Oxaloacetategenerated from the ATP citrate lyase (ACL) reaction can be con-verted to malate and subsequently pyruvate by the activitiesof malate dehydrogenase (MDH) and malic enzyme (ME) respec-tively. Pyruvate itself can be converted to lactate and released intothe extracellular space. ACC acetyl-CoA carboxylase, FAS fatty acidsynthase, LCFAs long chain fatty acids.


Heart Metab. (2011) 51:34–38 37






































tumor progression by N-Myc and hypoxia inducible factor HIF-1alpha. Cancer Res 70:10351–10361

13. Sun Q, Chen X, Ma J, Peng H, Wang F, Zha X, Wang Y et al(2011) Mammalian target of rapamycin up-regulation of pyru-vate kinase isoenzyme type M2 is critical for aerobic glycolysisand tumor growth. Proc Natl Acad Sci U S A 108:4129–4134

14. Mazurek S, Boschek CB, Hugo F, Eigenbrodt E (2005) Pyru-vate kinase type M2 and its role in tumor growth and spread-ing. Semin Cancer Biol 15:300–308

15. Vander Heiden MG, Locasale JW, Swanson KD, Sharfi H, Hef-fron GJ, et al (2010) Evidence for an alternative glycolytic path-way in rapidly proliferating cells. Science 329:1492–1499

16. Lewis BC, Shim H, Li Q, Wu CS, Lee LA, Maity A, Dang CV(1997) Identification of putative c-Myc-responsive genes:characterization of rcl, a novel growth-related gene. Mol CellBiol 17:4967–4978

17. Ullah MS, Davies AJ, Halestrap AP (2006) The plasma mem-brane lactate transporter MCT4, but not MCT1, is up-regulated by hypoxia through a HIF-1alpha-dependent mech-anism. J Biol Chem 281:9030–9037

18. Deberardinis RJ, Sayed N, Ditsworth D, Thompson CB (2008)Brick by brick: metabolism and tumor cell growth, Curr OpinGenet Dev 18:54–61

19. Swinnen JV, Brusselmans K, Verhoeven G (2006) Increasedlipogenesis in cancer cells: new players, novel targets. CurrOpin Clin Nutr Metab Care 9:358–365

20. DeBerardinis RJ, Mancuso A, Daikhin E, Nissim I, Yudkoff M,Wehrli S, Thompson CB (2007) Beyond aerobic glycolysis:transformed cells can engage in glutamine metabolism thatexceeds the requirement for protein and nucleotide synthesis.Proc Natl Acad Sci U S A 104:19345–19350


38 Heart Metab. (2011) 51:34–38
































Warfarin and dabigatran:an historical handover?

Alda Huqi, Cardiovascular Medicine Division, Cardio Thoracic Department, University of Pisa, Pisa, Italy,and University of Alberta, Edmonton, Alberta, Canada

Correspondence: Dr. Alda Huqi, Cardiovascular Medicine Division, Cardio Thoracic Department, University of Pisa,Via Paradisa, 2, 56100 Pisa, Italy. And: 423 Heritage Medical Research Center, University of Alberta, Edmonton,

Alberta T6G 2S2, Canada. Tel: +39 32972 56426, e-mail: [email protected]

It was in the early 1943 that Karl Paul Link first lectured on the anticoagulant properties ofdicumarol. Since then, warfarin and other vitamin K antagonists have been used for preven-

tion and treatment of thrombosis for almost 70 years. Because atrial fibrillation (AF) is the mostcommon heart rhythm disorder [1], and its presence increases the chance of stroke by five-fold [2], AF patients constitute the vast majority of patients with indications for anticoagulationtherapy. However, anticoagulant therapy is often a source of frustration to both patients andphysicians. Although it has been shown to reduce the incidence of stroke by 61%, only half ofeligible patients receive appropriate treatment [3]. While on warfarin, patients are required tocomplete an international normalized ratio (INR) monthly check and, if they are above orbelow the recommended range, have to go back the following week. Nonetheless, despitethe frequent check up and the institution of anticoagulation centers, patients on warfarin are<60% of the time in the therapeutic INR range [4]. In addition, many elderly patients, which rep-resent the vast majority of patients with indications to anticoagulation therapy, often presentwith limited mobility and autonomy and therefore have additional difficulties for attendinghospital-located monitoring facilities. Moreover, since warfarin effects are affected by factorssuch as food and drug-drug interactions, patients are also equipped with a long list of “thingsto avoid while on warfarin”, which is commonly a life-long therapy. Therefore, fairly enough,when implementing such a treatment, patients tend to be very reluctant and regard it as akind of a “punishment”, especially the asymptomatic ones, in whom AF diagnosis was basedon an accidental electrocardiogram (ECG) finding.

Although researchers have been looking for replacing warfarin for several decades, untilrecently no drug presented as a suitable substitute. However, after a long quiescence period,we are perhaps witnessing an historical event. Indeed, as reported in the 2010 EuropeanSociety of Cardiology guidelines for the management of patients with AF, dabigatran receivedguidance for use as an alternative to warfarin [5]. Moreover, in the more recently published,2011 American College of Cardiology/American Heart Association homologue guidelines,dabigatran obtained a “Class I, Level of Evidence B” insertion as an alternative to warfarin forAF patients [1].

Dabigatran is a “classmate” of ximelagatran, the first direct thrombin inhibitor which,despite showing clinical efficacy for prevention of thromboembolic events, led to fatal hepato-toxicity and therefore was withdrawn from clinical use. The first credentials for safety and effi-cacy issues arose from phase III clinical studies that compared dabigatran to enoxaparin or

HotT

opics

HOT TOPICS - ALDA HUQI

Heart Metab. (2011) 51:39–40 39

warfarin in patients undergoing orthopedic surgery [6].However, the conclusive results are based on the RE-LY trial, which assessed the two-year outcomes ofsome 18,113 AF patients receiving dabigatran (110or 150 mg twice daily) or warfarin. In this study, therate of the primary outcome (systemic embolism) waslower with dabigatran at a dosage of 150 mg twicedaily. Importantly, although gastro intestinal bleedingin elderly patients slightly increased, a 70% reductionin intracerebral bleeding was observed in patientsreceiving both doses of dabigatran as compared towarfarin. Moreover, continuous monitoring revealedno significant increase in hepatic enzymes associatedwith dabigatran. Nonetheless, there was a very smallbut statistically significant increase in the risk of myo-cardial infarction with dabigatran (reaching statisticalsignificance in the 110 mg regime), which may tempersome of the enthusiasm for the use of this drug.Another important issue with dabigatran is the lack ofan antidote for reversal of anticoagulation, although, inextreme cases, can be controlled by fresh plasmainfusion.

Dabigatran etexilate is an oral pro-drug rapidly con-verted by serum esterases to dabigatran, a competi-tive direct thrombin inhibitor. It reaches peak plasmaconcentrations within 1–2 hours after administrationand has a half-life of 12–17 hours. It necessitatestwice-daily administration, does not interact withCYP450 system and is excreted by kidneys. Thereare, however, other “attention” labels to its use. Dabi-gatran is a substrate of p-glycoprotein, a proteininvolved in active transportation of drugs though mem-branes. Therefore, drugs that strongly inhibit (amiodar-one, verapamil, and quinidine) or induce (rifamicin,pantoprazole) p-glycoprotein activity should be usedwith caution or avoided. Nonetheless, contrary toexpectances, such interactions did not reveal relevantin the RE-LY study.

From a practical point of view, besides evaluatingthe above-mentioned medical conditions, whenconsidering a switch from warfarin to dabigatran weshould ask whether patients have plain access toanticoagulation centers, can fully comply with thetwice-daily dosing, or prefer one to another treatment.On the other hand, cost issues will also have a greatimpact in determining treatment choices. However, apreliminary cost-effective study that based the esti-mates on the United Kingdom prices of dabigatranshowed promising results [7].

In conclusion, provided the brand new, guideline-recommended dabigatran fulfill current promises, wemay be witnessing an historical warfarin-to-dabigatranhandover. ●

References

1. Wann LS et al (2011) 2011 ACCF/AHA/HRS Focused Updateon the Management of Patients With Atrial Fibrillation (Updateon Dabigatran): A Report of the American College of CardiologyFoundation/American Heart Association Task Force on PracticeGuidelines. Circulation 123:1144–1150

2. Lin HJ et al (1996) Stroke severity in atrial fibrillation. The Fra-mingham Study. Stroke 27(10):1760–1764

3. Walker AM, Bennett D (2008) Epidemiology and outcomes inpatients with atrial fibrillation in the United States. Heart Rhythm5(10):1365–1372

4. Cios DA et al (2009) Evaluating the impact of study-level factorson warfarin control in U.S.-based primary studies: a meta-analysis. Am J Health Syst Pharm 66(10):916–925

5. Camm AJ et al (2010) Guidelines for the management of atrialfibrillation: the Task Force for the Management of Atrial Fibrilla-tion of the European Society of Cardiology (ESC). Eur HeartJ 31(19):2369–2429

6. Eriksson BI et al (2007) Dabigatran etexilate versus enoxaparinfor prevention of venous thromboembolism after total hipreplacement: a randomised, double-blind, non-inferiority trial.Lancet 370(9591):949–956

7. Freeman JV et al (2010) Cost-effectiveness of dabigatran com-pared with warfarin for stroke prevention in atrial fibrillation. AnnIntern Med. 154(1):1–11

HOT TOPICS - ALDA HUQI

40 Heart Metab. (2011) 51:39–40

























Documents

EDITORIALThe rising tide of cardiotoxicity …The rising tide of cardiotoxicity associated with cancer chemotherapy Gary D. Lopaschuk, Mazankowski Alberta Heart Institute, University