Dichloroacetate (DCA) as a potential metabolic-targeting therapy for cancer

Dichloroacetate (DCA) as a potential metabolic-targeting therapy for cancer

E D Michelakis1, L Webster1 and J R Mackey2

  1. 1Department of Medicine, University of Alberta, Edmonton, Canada
  2. 2Department of Oncology, University of Alberta, Edmonton, Canada

Write to Dr. ED Michelakis at the University of Alberta Hospital’s Department of Medicine, 8440-112 Street, Edmonton, Alberta, Canada. Email:

Submitted on December 18, 2007, revised on April 28, 2008, approved on July 4, 2008, and posted online on September 2, 2008.

10.1038/sj.bjc.6604554 British Journal of Cancer (2008) 99, 989–994.

Published online 2 September 2008

Top of page


Not only is the Warburg effect, or aerobic glycolysis, the basis for identifying cancer with metabolic imaging, but it may also be linked to the resistance to apoptosis that distinguishes cancer. The glycolytic phenotype in cancer may be connected to a (perhaps reversible) reduction of mitochondrial function and appears to be the common denominator of various molecular aberrations in cancer. The generic medication dichloroacetate is an orally ingestible small chemical that enhances the flux of pyruvate into the mitochondria by inhibiting the pyruvate dehydrogenase kinase, favouring glucose oxidation over glycolysis. This inhibits the growth of tumors both in vitro and in vivo by reversing the decreased mitochondrial apoptosis seen in cancer. Here, we examine the clinical and scientific justification for the quick implementation of this promising metabolic modulator in early-stage cancer clinical studies.


mitochondria, metabolism, apoptosis, potassium channels, positron emission tomography, glycolysis

Top of page

The approach to treating cancer must change.

Despite some successes since it was launched in the United States in 1971, the “war against cancer” is currently fighting. Despite large investments from industry and the general public, oncology has an incredibly low success rate in the clinical development of successful experimental treatments; it is less than a third of that in cardiovascular or infectious diseases. (Kamb et al, 2007). Drug development in cancer frequently focuses on targets required for the survival of all dividing cells, leading to narrow treatment windows. Non-essential targets give higher selectivity but are less effective. The potency and specificity of imatinib for CML make it an exception in cancer therapy since the dependence of CML cells on Ableson kinase is only brought on by a chromosomal translocation in the malignant clone. Finding a crucial target that is only effective against cancer cells is incredibly rare. (Kamb et al, 2007). TheThe amazing heterogeneity and adaptability of cancer cells is the main cause of the subpar effectiveness of cancer therapies. Histologically identical malignancies with different molecular characteristics commonly differ from one another, and molecular heterogeneity frequently exists within a single tumor. Scientists and clinical oncologists are increasingly of the opinion that there are a variety of malignancies. The realization that specialized medications must be designed and evaluated for molecularly defined tumors and that effects in one malignancy may not necessarily apply to another has crucial consequences.

The most difficult problem still involves selectively inducing cell death (mostly apoptosis) in cancer cells but not in normal cells. A practical anticancer treatment would be inexpensive (perhaps an orally accessible tiny chemical) and simple to deliver. Not only are the majority of brand-new anti-cancer medications out of reach for millions of patients from developing nations, but they are also prohibitively expensive for many people in rich nations without robust health insurance.

Targeting more “distal” pathways that integrate many proximal signals is one strategy to address the issue of heterogeneity of “proximal” molecular pathways in cancer, provided that the shared distal pathways continue to be crucial and unique to cancer cells. Most solid tumors have a distinct metabolism that combines a number of proximal pathways and leads to modification of the mitochondria, which results in a glycolytic phenotype and a potent resistance to apoptosis. The mitochondria may be the main target of cancer therapies rather than merely incidental players in the genesis of cancer, according to mounting data. Dichloroacetate (DCA), a mitochondria-targeting small chemical that permeates most tissues following oral treatment, can correct this cancer-specific metabolic reprogramming. (Bonnet et al, 2007; Pan and Mak, 2007). By non-invasively and prospectively measuring glucose uptake in tumors using positron emission tomography (PET) imaging, it is also possible to monitor the molecular and direct metabolic response to DCA. Such metabolic tactics may be able to alter the cancer field’s experimental therapy paradigm.

DCA’s preclinical research (showing effectiveness in a variety of tumours and relatively low toxicity) (Bonnet et al, 2007), Rapid clinical translation is strongly justified by the drug’s structure—a very small molecule—low cost (it is a generic medication), and the fact that DCA has been used by humans for more than 30 years. Here, we elaborate on the scientific justification and go over a few useful points that are crucial to the clinical assessment of DCA as an anticancer treatment.

Top of page

The metabolism of cancer cells

Aerobic glycolysis (GLY), or the use of glycolysis for energy production despite the presence of oxygen, is a characteristic of the majority of cancers. The Warburg effect was first noticed by Warburg in 1929, and he postulated that it was caused by mitochondrial dysfunction that prevented the oxidation of glucose based on mitochondria. (Warburg, 1930). Cancer cells upregulate glucose receptors and dramatically enhance glucose uptake in an effort to “catch up” with GO since it is significantly more efficient than GLY at creating ATP (producing 36 vs. 2 ATP per glucose molecule). The majority of solid tumors currently have much higher glucose uptake and metabolism than non-cancerous cells, according to positron emission tomography imaging. (Figure 1). Given that GLY is not normally seen in normal tissues outside of skeletal muscle during intense exercise, this bio-energetic differential between cancer and normal cells may provide a very specific therapeutic target. The glycolytic profile has typically been considered as a consequence of cancer growth rather than a cause, therefore interest in targeting tumour metabolism has been modest. Nevertheless, this area of experimental oncology has remained controversial. Furthermore, employing an evolutionary model of carcinogenesis makes it challenging to comprehend the glycolytic profile of cancer at first glance. First, why would these extremely energy-hungry, rapidly multiplying cells choose to use GLY rather than the much more effective GO? Second, GLY causes severe lactic acidosis, which may be hazardous to both the tissues around the cancer cells and the cancer cells themselves. The metabolic hypothesis of cancer has recently come back into favor, which suggests that these facts are not as contradictory as they initially seem to be. (Gatenby and Gillies, 2004):

Figure 1.


An extensive glioblastoma tumor can be seen on a brain MRI along with considerable cerebral oedema and areas of necrosis inside the tumor. A similar FDG-Glucose PET from the same patient is shown on the right. The tumour has substantially higher glucose uptake than the nearby brain tissue.

Full figure and legend (49K)


Top of page

Glycolysis offers an early adaptation to the hypoxic microenvironment in carcinogenesis

Gatenby and Gillies (2004) recently proposed that as early carcinogenesis often occurs in a hypoxic microenvironment, the transformed cells have to rely on anaerobic GLY for energy production. Hypoxia-inducible factor (HIF) is activated in hypoxic conditions and it has been shown to induce the expression of several glucose transporters and most of the enzymes required for GLY (Semenza et al, 1994). For instance, HIF stimulates pyruvate dehydrogenase kinase expression (PDK) (Kim et al, 2006), an enzyme that acts as a gatekeeper and controls the flow of carbohydrates (pyruvate) into the mitochondria. Pyruvate dehydrogenase (PDH) is inhibited in the presence of activated PDK, limiting the amount of pyruvate that can enter the mitochondria, where GO can occur. In other words, activated PDK encourages GLY completion in the cytoplasm through the conversion of pyruvate to lactate, while inhibited PDK guarantees effective coupling of GLY and GO.

Tumors first seek to compensate by increasing the intake of glucose by their cells. Furthermore, Gatenby and Gillies (2004) List a few ways that lactic acidosis promotes tumor growth, including enhanced cell mobility and propensity for metastasis, breakdown of the extracellular matrix, and stimulation of angiogenesis coupled with HIF. The aerobic glycolytic profile endures even when tumors eventually get vascularized and are no longer notably hypoxic (although some tumors stay hypoxic at the core due to poor neo-vessel quality). This implies that cancer cells have a survival advantage due to the (originally adaptive) metabolic modification. In fact, new research indicates that switching to a glycolytic phenotype confers resistance to apoptosis. (Plas and Thompson, 2002) (Kim and Dang, 2005, 2006).

Top of page

Resistance to apoptosis is linked to glycolysis.

It has been suggested that linkages between metabolic sensors, cell death, and gene transcription are directly created through the enzymes that control metabolism because several of the glycolysis-related enzymes are also significant regulators of apoptosis and gene transcription. (Kim and Dang, 2005). Hexokinase activation, for instance, results in a significant suppression of apoptosis. Activated hexokinase translocates to the mitochondrial membranes from the cytoplasm where it interacts with and inhibits several essential elements of mitochondria-dependent apoptosis. (Pastorino et al, 2005). The fact that hexokinase is upregulated and activated in many cancers is therefore not surprising. (Kim and Dang, 2006). Why does this happen? Both p53 and HIF response elements are present in the hexokinase promoter, and both mutated p53 and activated HIF increase the expression of hexokinase. (Mathupala et al, 1997). Additionally, the oncogenic protein Akt induces a glycolytic metabolic profile through a variety of mechanisms and is upregulated in many cancers. (Elstrom et al, 2004). Hexokinase’s expression and activity are both increased by Akt. (Gottlob et al, 2001; Elstrom et al, 2004). Many malignancies have mutations (loss of function mutations) in PTEN, the gene that typically inhibits Akt. Even more connections between p53 and metabolism were recently discovered: p53 directly controls the expression of a subunit of cytochrome c oxidase, a vital component of complex IV of the electron transport chain in mitochondria, as well as the expression of a crucial GLY enzyme through the production of TIGAR. (Pan and Mak, 2007)). In other words, the loss of p53 function, the most prevalent molecular abnormality in cancer, causes metabolic and mitochondrial changes that are consistent with the glycolytic phenotype. Similarly, the c-myc transcription factor can produce this same metabolic phenotype by increasing the expression of numerous GLY enzymes. (Kim and Dang, 2005, 2006).

In summary, an evolutionary theory of carcinogenesis identifies metabolism and GLY as a crucial and early mechanism of adaptation of cancer cells against hypoxia, which persists because it provides resistance to apoptosis in cancer cells. (Gatenby and Gillies, 2004). The genetic hypothesis of cancer also links GLY and metabolism to the activation of numerous other oncogenes, such as c-myc, Akt/PTEN, and p53. (Pan and Mak, 2007). It is therefore possible that this metabolic phenotype is not just a “by-product” of carcinogenesis but rather plays a key role in the pathogenesis of cancer. The ‘facilitation’ of carcinogenesis by this metabolic phenotype is undeniable, even though it is unclear whether it directly causes malignancy. (Kim and Dang, 2006). Additionally, because this metabolic signature provides the link between many different pathways, if it were to be therapeutically addressed, it might provide selectivity for malignant cells with a variety of cellular and molecular origins.

Top of page

Apoptosis and mitochondria

Apoptosis, a kind of cell death that depends on mitochondrial energy production, may be suppressed by moving metabolism from the mitochondria (GO) to the cytoplasm (GLY). (Figure 2). Apoptosis-inducing factor and cytochrome c are two pro-apoptotic mediators that are shielded inside the mitochondria. Although it is conceivable for this to happen without the voltage- and redox-sensitive mitochondrial transition pore (MTP) opening, they are released in the cytoplasm and cause apoptosis. (Halestrap, 2005). MTP opening is related to mitochondrial depolarization and increased ROS. (Zamzami and Kroemer, 2001). ROS production and mitochondrial membrane potential depend on the flow of electrons through the electron transport chain (ETC), which in turn depends on the Krebs cycle’s synthesis of electron donors (NADH, FADH2). By preventing pyruvate from entering the mitochondria and acetyl-CoA from being produced, you can prevent the Krebs cycle, the ETC, MTP opening, and apoptosis.

Figure 2.


The majority of solid tumors have a glycolytic environment, which is linked to an antiapoptotic and pro-proliferative condition. By increasing pyruvate entry into the mitochondria either by DCA or by inhibiting LDH, one can improve glucose oxidation, induce apoptosis, and decrease tumor growth and proliferation (see text for discussion).

Full figure and legend (148K)


Additionally, mitochondria can influence downstream processes connected to apoptosis and proliferation. For instance, mitochondrial uptake can control intracellular Ca++ directly, and a rise in this level is linked to enhanced proliferation and the activation of a number of transcription factors. Additionally, by activating plasma membrane K+ channels, the superoxide generated by the mitochondria can be dismutated to H2O2 and diffuse freely, controlling the input of Ca++ and the activation of caspases. K+ ions can pass through the plasma membrane thanks to the transmembrane proteins known as K+ channels. [K+]i rises as a result of K+ channel closure or reduced K+ channel expression, increasing the tonic inhibition of caspases that cytosolic K+ exerts. (Remillard and Yuan, 2004). Due to its redox sensitivity, the voltage-gated K+ channel family (Kv) can be controlled by the mitochondria. For instance, certain Kv channels, such as Kv1.5, can be activated by H2O2 derived from mitochondria. (Bonnet et al, 2007).

Top of page

According to preliminary research, DCA prevents cancer cells from becoming resistant to apoptosis via reversing mitochondrial remodeling.

In contrast to non-cancer cell lines, we recently demonstrated that various cancer cell lines (including non-small cell lung cancer, breast cancer, and glioblastoma) have hyperpolarized mitochondria. (Bonnet et al, 2007), a discovery that Dr. Chen first discussed at the Dana Farber Institute in the 1980s (Chen, 1988). Reduced levels of mitochondrial ROS and lower Kv channel expression and activity were linked to this. In the cancer cells, the Ca++-sensitive transcription factor NFAT was also active (i.e., nuclear). It has been demonstrated that the transcription factor NFAT lowers the level of the Kv channel Kv1.5 and raises the antiapoptotic protein bcl-2. The reduction of pyruvate entry would eventually lead to a reduction in the flux of electrons in the ETC and a consequent reduction in the production of ROS, the closure of the existing redox-sensitive Kv channels, and an increase in intracellular Ca++. All of these characteristics are consistent with an antiapoptotic state and could be secondary to a suppressed mitochondrial activity. The closure of the redox-sensitive MTP and mitochondrial hyperpolarization may both be influenced by the reduced ROS. There would be compensatory GLY as a result of the decreased pyruvate entrance into the mitochondria (and hence the lower GO). Increased hexokinase levels would contribute to the hyperpolarization of the mitochondria; it is known that hexokinase is translocated to the mitochondrial membrane in a glycolytic environment where it inhibits the voltage-dependent anion channel, causing the mitochondria to become hyperpolarized and suppressing apoptosis. (Pastorino et al, 2005) (Figure 2).

Pyruvate was delivered to the mitochondria at a higher rate as a result of dichloroacetate’s activation of the pyruvate dehydrogenase. As expected, DCA raised GO and depolarized the mitochondria, bringing the membrane potential back to that of non-cancerous cells while having no negative effects on the non-cancerous cells’ mitochondria. (Figure 2). Surprisingly, all the aforementioned characteristics of the cancer cells were “normalized” as a result of the rise in GO and the depolarization of the mitochondria: ROS increased, NFAT was deactivated, and Kv channel function and expression increased. Most importantly, both cytochrome c and apoptosis-inducing factor efflux from the mitochondria induced apoptosis in the cancer cells. In xenotransplant models, this led to a reduction in tumor growth both in vitro and in vivo. (Bonnet et al, 2007) (Figure 3). Along with the induction of apoptosis by DCA in breast cancer, glioblastoma cell lines, and non-small cell lung cancer reported in our initial publication (Bonnet et al, 2007), DCA was just recently proven to cause endometrial cells to undergo apoptosis. (Wong et al, 2008) and prostate (Cao et al, 2008) Independent confirmation of our findings showed that the same technique mostly killed cancer cells. Furthermore, as expected, activating mitochondria by DCA boosts O2 consumption in tumors and significantly improves the efficacy of hypoxia-specific chemotherapies in animal models. (Cairns et al, 2007).

Figure 3.


Dichloroacetate inhibits tumor development in vivo and depolarizes mitochondria. Non-small cell lung cancer cells are shown on the left with TMRM loaded both before and after DCA treatment (the higher the red fluorescence the higher the mitochondrial membrane potential; nuclei in blue). Rats that were naked had the same cells injected into their flanks. With a rodent PET-CT, these rodents are pictured on the right (GammaMedica). DCA therapy reduces the size and glucose uptake in the tumor, according to concurrent CT and FDG-Glucose PET imaging.

Full figure and legend (149K)


It is crucial to emphasize that merely blocking GLY will not encourage pyruvate entry into the mitochondria, or reactivate mitochondria. Several non-cancerous tissues that rely on GLY for energy production will also be poisoned by it. Because apoptosis is an energy-intensive process requiring active mitochondria, inhibiting GLY causes ATP depletion and necrosis instead of apoptosis, which has previously been tested as a potential cancer treatment. (Xu et al, 2005). The “trick,” rather than just inhibiting GLY, is to improve the GLY to GO coupling. One way to achieve this is by bringing pyruvate into the mitochondria, inhibiting LDH, activating PDH, and enhancing GO. (Figure 2). Recent research that inhibits LDH (by siRNA), which promotes the transfer of pyruvate into the mitochondria (in that sense mimicking DCA), also promotes cancer apoptosis and slows tumor growth in vitro and in mice xenotransplants, lends support to this hypothesis. (Fantin et al, 2006).

Top of page

DCA: clinical experience and the mechanism of action

Having a 150 Da size, dichloroacetate is a tiny molecule (see structure in Figure 3) explaining in part the drug’s high bioavailability and its ability to enter the brain, one of the traditional chemotherapy sanctuary sites. DCA inhibits PDK at concentrations of 10-250 M or 0.15-37.5 g ml1 in vitro, activating PDH in a dose-dependent manner. (Stacpoole, 1989). To date, four different isoforms of PDK have been identified that have variable expression and sensitivity to the inhibition by DCA (Sugden and Holness, 2003). In our previously published preclinical research, we demonstrated that PDK2 inhibition with siRNA perfectly mimicked DCA effects. PDKII is the isozyme constitutively expressed in the majority of tissues and with the highest sensitivity to DCA. (Bonnet et al, 2007).

DCA has a 100% bioavailability when taken orally. To determine the ideal dose of DCA, numerous studies using both IV and oral forms were conducted. The reduction in lactate levels in the blood and cerebrospinal fluid was the end point that was measured. The immediate effect of DCA’s inhibition of PDK (and subsequent activation of PDH) is a reduction in lactate levels. PDH activity in muscle biopsies was directly measured in several studies that administered DCA to patients. Dichloroacetate reduces lactate levels by more than 60% and directly activates PDH by 3-6 fold when given at a dose of 35-50 mg kg1. (Howlett et al, 1999; Parolin et al, 2000).

The pharmacokinetics of DCA in healthy volunteers adhere to a straightforward one-compartment model, but in significantly aberrant situations like severe lactic acidosis or cirrhosis, they are more complicated. By an unidentified mechanism, dichloroacetate slows its own metabolism, and after repeated dosages, its clearance declines. (Stacpoole et al, 2003). Although the half-life is initially less than an hour with the first dose, it becomes several hours with repeated doses. This action, nevertheless, reaches a plateau, and chronic use does not result in further increases in DCA serum levels. The same is true for the metabolites of DCA (which do not have any biologic effect, at least on PDH). For instance, the serum DCA levels after five years of ongoing treatment with oral DCA at 25 mg kg1 are only marginally higher than the levels after the initial doses (and continue to be in the vicinity of 100 g ml1) (Mori et al, 2004). Due to the fact that DCA “locks” PDK in a sustained inactive state, the effects on lactate levels are sustained and continue even after the DCA levels drop.

Over the past 40 years, a large number of children and adults, including healthy volunteers and subjects with various disease states, have been exposed to DCA. Since the 1969 first description (Stacpoole, 1969), The effects of lactic acidosis worsening severe malaria, sepsis, congestive heart failure, burns, cirrhosis, liver transplantation, and congenital mitochondrial illnesses have been examined, and DCA has been shown to reduce their symptoms or haemodynamic effects. dosages of 12.5 to 100 mg kg per day, administered orally or intravenously, were utilized in single-arm and randomised studies of DCA (reviewed in (Stacpoole et al, 2003)). DCA was successful in reducing lactate levels across the board, but it had no effect on how the main disease progressed (for example sepsis).

The first two randomised control studies of chronic oral therapy with DCA for congenital mitochondrial disorders were reported in 2006, despite the fact that more than 40 nonrandomized trials of DCA in small patient cohorts have been described. In the first, oral DCA was given to 30 MELAS syndrome patients in a blinded, placebo-controlled research at a dose of 25 mg kg per day (mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes) (Kaufmann et al, 2006). In contrast to 4 out of 15 participants in the placebo arm, the majority of individuals recruited in the DCA arm experienced symptomatic peripheral neuropathy, which resulted to the study’s cancellation. By nine months after stopping the DCA, 17 out of 19 patients experienced at least some improvement in their peripheral neurological symptoms. This neurotoxicity had characteristics of an axonal, sensory polyneuropathy with length dependence but no demyelination. There were no additional toxicity noted. Due to primary or secondary effects on peripheral nerves, peripheral neuropathy frequently exacerbate MELAS. For instance, these patients also have diabetes and peripheral neuropathy caused by diabetes.

A different randomised placebo-controlled double-blinded research, in contrast, was unable to demonstrate any substantial DCA harm, including peripheral neuropathy. Only one of the 21 congenital lactic acidosis patients in this trial who received DCA orally at a dose of 25 mg kg per day for six months had minor peripheral neuropathy. Serial nerve conduction investigations were unable to show a difference between the 2 arms’ incidence of neuropathy (placebo vs DCA). One patient from each group complained hand tremor, muscle rigidity of the upper extremity, and drowsiness. (Stacpoole et al, 2006).

Peripheral neuropathy is more common in adult MELAS patients, which may indicate an inherent propensity to this complication in MELAS or its associated illnesses, such as diabetes mellitus; this toxicity may also depend on the patient’s age. In conclusion, peripheral neuropathy appears to be a potentially serious adverse effect of DCA. The risk for DCA-associated peripheral neuropathy may depend on whether cancer patients have received prior or concurrent neurotoxic medication, as peripheral neuropathy is a common side effect of taxane, platinum, and vinca-alkaloid chemotherapies.

Top of page

DCA: Cancer clinical trials?

There is strong evidence that DCA may help treat human cancer in preclinical in vitro and in vivo models. (Bonnet et al, 2007; Cairns et al, 2007; Cao et al, 2008; Wong et al, 2008). The idea is supported by the finding that LDH inhibition improved survival in mice with human cancer xenotransplants by inducing apoptosis and inhibiting growth. (Fantin et al, 2006). The mechanistic studies of DCA in human tissues following oral administration, the pharmacokinetic and toxicological data from randomised studies for 6 months, and the 5-year use case reports total 40 years of human experience. This suggests that early-phase clinical trials can be easily translated.

Numerous cancer types could be used as test subjects for dichloroacetate. Given that (i) a wide range of signaling pathways and oncogenes cause resistance to apoptosis and a glycolytic phenotype, (ii) most carcinomas have hyperpolarized/remodeled mitochondria, and (iii) the majority of solid tumors have increased glucose uptake on PET imaging, it is possible that DCA may be effective in a wide range of different tumor types. However, only non-small cell lung cancer, glioblastoma, breast, endometrial, and prostate cancer have provided direct preclinical evidence of DCA’s anticancer effects. Additionally, the absence of mitochondrial hyperpolarization in some cancers, such as sarcomas, lymphomas, neuroblastomas, and oat cell lung cancer (Chen, 1988), imply that DCA might not be successful in such circumstances. The top cancers to study should be those with little or no relevant therapy alternatives, such as advanced lung cancer or recurrent glioblastoma.

DCA has never been administered to a cancer patient during a clinical trial. It is unknown if dose ranges previously investigated will produce DCA concentrations that are lethal within tumors. Additionally, patients with advanced cancer had different nutritional and metabolic profiles overall than those in the previous DCA studies. Additionally, DCA neurotoxicity may be predisposed to by prior exposure to neurotoxic chemotherapy. To determine the maximally tolerated and physiologically active dose, meticulously executed phase I dose escalation and phase II trials with repeated tissue biopsies are needed. Clinical trials using DCA must closely assess neurotoxicity and devise precise dose-reduction plans to control side effects. Additionally, it will be necessary to define the pharmacokinetics in the cancer population.

Clinical trials using DCA as a single agent or in direct comparison with other agents are justified by the preclinical experience with DCA monotherapy. DCA, however, might be a desirable “apoptosis-sensitizer” agent because it “unlocks” cancer cells from an apoptosis resistance state. In an effort to improve their efficacy, lower the dosages needed, and reduce the toxicity of conventional therapies, DCA could be administered both before and after chemotherapy or radiation therapy. (Cairns et al, 2007).

The potential for tracking metabolic modulation through imaging and diagnostic studies is prompted by the capacity to view metabolism as an integrator of numerous diverse signaling pathways. Important queries that must be addressed in clinical trials utilizing DCA include those already mentioned, such as: (i) Can PET be used to document non-invasively a reversal of the glycolytic phenotype in response to DCA or as a predictor of clinical response? (ii) Can clinical response to DCA be predicted by mitochondrial membrane potential or the acute effects on DCA in fresh tumour biopsies, thereby facilitating patient selection?

Since DCA is a generic drug and early industry support might be limited, funding for such trials would be difficult for the academic community. It might be possible to raise money from philanthropies to support early phase I–II or modest phase III trials. The public will be further encouraged to directly fund these efforts if these trials, however, show favorable efficacy and toxicity, and national cancer organizations like the NCI may be motivated to directly contribute to the design and structure of larger trials. Even if DCA does not prove to be the “dawn of a new era,” it is significant to keep in mind (Pan and Mak, 2007), Starting and finishing clinical trials using a generic drug will be a monumental symbolic and practical task. The ‘dogma’ that systemic anticancer therapy studies cannot take place without corporate sponsorship now inhibits the potential of a number of promising therapies that might not be commercially viable for pharmaceutical companies. In that regard, the clinical assessment of DCA will represent a paradigm change unto itself, in addition to its scientific foundation.

Top of page

Remarks for proof

The unique anticancer effects of DCA in prostate and endometrial cancers have been validated in two significant studies since our review was accepted. Wong JY et al., Dichloroacetate causes apoptosis in endometrial cancer cells. Dichloroacetate (DCA) sensitizes both wild-type and over-expressing Bcl-2 prostate cancer cells to radiation in vitro, according to Gao et al. and Gynecol Oncol June 2008; 109(3): 394–402. August 2008; Prostate 1; 68(11): 1223–1231.

Top of page


  1. Puttagunta L, Lopaschuk GD, Bonnet S, Harry G, Hashimoto K, Porter CJ, Andrade MA, Thebaud B, Michelakis ED, Archer SL, Allalunis-Turner J, Haromy A, Beaulieu C, Thompson R, Lee CT (2007) Cancer suppresses a mitochondria-K+ channel axis, and normalizing it encourages apoptosis and slows the spread of the disease. 37–51 in Cancer Cell 11, | Article | PubMed | ChemPort |
  2. RA Cairns, I Papandreou, PD Sutphin, and NC Denko (2007) In solid tumors, metabolic targeting of hypoxia and HIF1 can improve cytotoxic treatment. 104: 9445-9450 Proc Natl Acad Sci USA | Article | PubMed | ChemPort |
  3. Those involved include Cao W, Yacoub S, Shiverick KT, Namiki K, Sakai Y, Porvasnik S, Urbanek C, and Rosser CJ (2008) Both wild-type and overexpressed Bcl-2 prostate cancer cells are radiation sensitive in vitro when exposed to dichloroacetate (DCA). 1223–1231 in Prostate 68 | Article | PubMed | ChemPort |
  4. Mitochondrial membrane potential in living cells, Chen LB 1988. 155–181 Annu Rev Cell Biol 4 | Article | PubMed | ISI | ChemPort |
  5. Bauer DE, Buzzai M, Karnauskas R, Harris MH, Plas DR, Zhuang H, Cinalli RM, Alavi A, Rudin CM, Thompson CB, and Elstrom RL are among the authors (2004) Cancer cells’ aerobic glycolysis is stimulated by Akt. Cancer Research 64(8):3892–3899 | Article | PubMed | ISI | ChemPort |
  6. St-Pierre J, Leder P, and Fantin VR LDH- Attenuation Glycolysis, mitochondrial physiology, and tumor maintenance are all connected by an expression. 425-434 in Cancer Cell 9, | Article | PubMed | ChemPort |
  7. Gillies RJ, Gatenby RA Why is aerobic glycolysis more prevalent in cancers? National Cancer Review 4: 891-899 | Article | PubMed | ISI | ChemPort |
  8. Majewski N, Kennedy S, Kandel E, Robey RB, and Gottlob K (2001) Hexokinase in the mitochondria and the first committed step of glycolysis are both required for Akt/PKB to inhibit early apoptotic processes. Gene Development 15: 1406–1418 | Article | PubMed | ISI | ChemPort |
  9. Biochemistry: A Pore Way to Die, Halestrap A. 2005. Biology 434: 578–579 | Article | PubMed | ChemPort |
  10. Hollidge-Horvat MG, Howlett RA, Heigenhauser GJ, Hultman E, and Spriet LL (1999) effects of dichloroacetate infusion on the metabolism of human skeletal muscle before exercise. Physiology Am J 277: E18-E25 | PubMed | ChemPort |
  11. Lengauer C, Wee S, and Kamb A (2007) Why is finding new cancer treatments so challenging? 115–120. Nat Rev Drug Discov, 6. | Article | PubMed | ISI | ChemPort |
  12. Hong X, Gooch CL, Mao X, Pascual JM, Hirano M, Stacpoole PW, DiMauro S, De Vivo DC, Kaufmann P, Engelstad K, Wei Y, Jhung S, Sano MC, Shungu DC, Millar WS In MELAS, dichloroacetate produces toxic neuropathy: a controlled, randomized clinical trial. Brain 66: 324–330 | Article | PubMed | ChemPort |
  13. JW Kim and CV Dang various functions of glycolytic enzymes. Biochemical Trends 30: 142-150 | Article | PubMed | ISI | ChemPort |
  14. Cancer’s molecular sweet tooth and the Warburg effect, Kim JW, Dang CV, 2006. 66: 8927–8930, Cancer Research | Article | PubMed | ChemPort |
  15. Kim JW, Tchernyshyov I, Semenza GL, Dang CV (2006) HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab 3: 177–185 | Article | PubMed | ChemPort |
  16. Heese C, Mathupala SP, and Pedersen PL (1997) Cancer cells catabolize glucose. The tumor suppressor p53 has functionally active response regions in the type II hexokinase promoter. 22776-22780 J Biol Chem | Article | PubMed | ISI | ChemPort |
  17. Mori M, Yamagata T, Goto T, Saito S, Momoi MY (2004) Dichloroacetate treatment for mitochondrial cytopathy: long-term effects in MELAS. Brain Dev 26: 453–458 | Article | PubMed |
  18. Mak TW, Pan JG A new era for metabolic targeting as an anti-cancer strategy? P14 Sci STKE | Article |
  19. Hollidge-Horvat MG, Matsos MP, Parolin ML, Spriet LL, Hultman E, Jones NL, and Heigenhauser GJ Effects of dichloroacetate activating PDH during exercise in hypoxia in human skeletal muscle Am Journal of Physiology: Endocrinology 279: E752-E761 | PubMed | ISI | ChemPort |
  20. Hoek JB, Pastorino JG, and Shulga N By phosphorylating voltage-dependent anion channels, glycogen synthase kinase 3beta activation disrupts hexokinase II’s ability to bind to mitochondria and increases chemotherapy-induced cytotoxicity. 10545-10554 in Cancer Res. | Article | PubMed | ChemPort |
  21. DR Plas and CB Thompson The regulation of programmed cell death by cell metabolism. 13: 75-78 Trends Endocrinol Metab | Article | PubMed |
  22. CV Remillard and JX Yuan K+ channel activation is a crucial pathway in programmed cell death. Am J L49-L67 in Physiol Lung Cell Mol Physiol 286 | Article | PubMed | ChemPort |
  23. Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. Semenza GL, Roth PH, Fang HM, Wang GL (1994). J Biol Chem 269: 23757–23763 | PubMed | ISI | ChemPort |
  24. Review of the pharmacologic and therapeutic effects of diisopropylammonium dichloroacetate by Stacpoole PW (1969) 9: 282-291 J Clin Pharmacol J New Drugs | PubMed | ChemPort |
  25. The pharmacology of dichloroacetate, Stacpoole PW (1989). 38: 1124–1144 Metabolism | Article | PubMed | ChemPort |
  26. Controlled clinical trial of dichloroacetate for the treatment of congenital lactic acidosis in children. Stacpoole PW, Kerr DS, Barnes C, Bunch ST, Carney PR, Fennell EM, Felitsyn NM, Gilmore RL, Greer, Henderson GN, Hutson AD, Neiberger RE, O’Brien RG, Perkins LA, Quisling RG, Shroads AL, Shuster JJ, Childhood 117: 1519–1531. | Article | PubMed |
  27. Dichloroacetate’s effectiveness as a lactate-lowering medication was studied by Stacpoole PW, Nagaraja NV, and Hutson AD in 2003. 683-691 J Clin Pharmacol | PubMed | ChemPort |
  28. MC Sugden and MJ Holness Recent developments in the mechanisms controlling pyruvate dehydrogenase complex-level PDKs’ oxidation of glucose. Am Journal of Physiology: Endocrinology 284: E855-E862 | PubMed | ChemPort |
  29. Regarding the stoffwechsel of tumors, Warburg O. Police: London
  30. Huggins GS, Munshi NC, Debidda M, Wong JY, and De Vivo I Endometrial cancer cells undergo apoptosis when exposed to dichloroacetate. 109: 394–402 Gynecol Oncol | Article | PubMed | ChemPort |
  31. Huang P, Bhalla KN, Xu RH, Pelicano H, Zhou Y, Carew JS, Feng L, and (2005) A novel approach to combat drug resistance linked to mitochondrial respiratory defects and hypoxia is to inhibit glycolysis in cancer cells. 613-621. Cancer Research | Article | PubMed | ISI | ChemPort |
  32. The mitochondrion in apoptosis: how Pandora’s box opens, Zamzami N., Kroemer G. (2001). 67–71. Nat Rev Mol Cell Biol 2 | Article | PubMed | ISI | ChemPort |