Doxorubicin: an update on anticancer molecular action, toxicity and novel drug delivery systems


  • Oktay Tacar,

    1. School of Biomedical and Health Sciences, Victoria University, St Albans, Australia
    Search for more papers by this author
  • Pornsak Sriamornsak,

    1. Department of Pharmaceutical Technology
    2. Pharmaceutical Biopolymer Group (PBiG), Faculty of Pharmacy, Silpakorn University, Nakhon Pathom, Thailand
    Search for more papers by this author
  • Crispin R. Dass

    Corresponding author
    1. School of Biomedical and Health Sciences, Victoria University, St Albans, Australia
      Crispin R. Dass, School of Biomedical and Health Sciences, Bldg 6, Victoria University, St Albans 3021, Australia. E-mail:
    Search for more papers by this author

Crispin R. Dass, School of Biomedical and Health Sciences, Bldg 6, Victoria University, St Albans 3021, Australia. E-mail:


Objectives  The frontline drug doxorubicin has been used for treating cancer for over 30 years. While providing a cure in select cases, doxorubicin causes toxicity to most major organs, especially life-threatening cardiotoxicity, which forces the treatment to become dose-limiting.

Key findings  Doxorubicin is known to bind to DNA-associated enzymes, intercalate with DNA base pairs, and target multiple molecular targets to produce a range of cytotoxic effects. For instance, it causes the activation of various molecular signals from AMPK (AMP-activated protein kinase inducing apoptosis) to influence the Bcl-2/Bax apoptosis pathway. By altering the Bcl-2/Bax ratio, downstream activation of different caspases can occur resulting in apoptosis. Doxorubicin also induces apoptosis and necrosis in healthy tissue causing toxicity in the brain, liver, kidney and heart. Over the years, many studies have been conducted to devise a drug delivery system that would eliminate these adverse affects including liposomes, hydrogel and nanoparticulate systems, and we highlight the pros and cons of these drug delivery systems.

Summary  Overall the future for the continued use of doxorubicin clinically against cancer looks set to be prolonged, provided certain enhancements as listed above are made to its chemistry, delivery and toxicity. Increased efficacy depends on these three aims being met satisfactorily as discussed in turn in this review.


The global epidemic of cancer is constantly increasing due to a steady increase in the population growth rate and ageing. Lifestyle choices across the developing world currently attribute to 64% of deaths. In 2012, an estimated 1 638 910 new cases of cancer were diagnosed.[1] Cancer is characterised as a disease associated with unsuppressed growth and spread of anomalistic cells. Cancers are caused by both internal and external factors from insufficient diets, inherited mutations to radiation and tobacco.[1] Decades often pass from the initial exposure before the cancer is detectable. By this time the cancer is usually at a late stage making treatment difficult if not impossible. A tumour is described as being neoplasm, uncoordinated clusters of cells that have proliferated aggressively, and show unnatural growth. Tumours are comprised of two components; the proliferating cells and the stroma, which comprises blood vessels and connective tissue. For the tumour to progress to metastasis, enzymes such as those of the matrix metalloproteinase class must break down the local host tissue. Once capillaries are damaged or leaking, the broken clusters of tumour cells can travel to distant regions throughout the body.[2] Once malignant, the tumour can induce angiogenesis allowing the tumour to grow its own blood supply to provide itself with the essential nutrients and oxygen for further growth at the secondary sites.

The development of cancer involves a series of complicated events, often taking decades to occur. In most cases the final result is altered genes that regulate differentiation and cell growth. In terms of gene alteration to form a cancer cell, a normal cell's genome acquires mutations at tumour suppressor genes, proto-oncogenes and other genes involved in cell growth regulation. Molecular markers are constantly being studied to determine how their presence or absence influences cancer development. An example of a tumour suppressor is p53. The p53 tumour suppressor protein is one of the most important cell cycle controls in normal cells and in human cancer cells too, and is altered in almost every case of the latter. Under normal cellular conditions, p53 is responsible for terminating cell growth when physiological stress and damage is encountered. This occurs by initiating various apoptosis pathways to eradicate the damaged cells.[3] Since preneoplastic cells have a strong tendency to eliminate functioning p53 protein, it suggests that tumour cells use this to reduce the risk of apoptotic death.[4] Another class of genes that are highly affected that results in cancer are oncogenes, which are responsible for stimulating cellular growth.[5] These alterations at a genetic level can be major, to the point where entire chromosomes are either lost or gained as a result of errors acquired during mitosis. Yet the most common of all are the mutated changes to the genomic DNA's nucleotide sequence. It is important to note that a single genetic change is not enough to promote malignant tumour development. Several genetic mutations are responsible for altering the multistep processes involved in cancer cell production. As these cancerous cells begin to clone different cytogenetical versions of themselves, they become quite different from the original cancer cell.[2] As a result these genetic alterations cause the tumour cells to respond differently to treatments and behave unpredictably. This makes treatment sensitivity differ within the same cancer type, even in the same patient, where one tumour growth may respond in a different way to other tumours.

Cellular sensitivity has a major role in chemotherapy, a primary treatment technique responsible for treating millions of sufferers worldwide. This form of treatment is often referred to as antineoplastic therapy and is used to eradicate all neoplastic cells via pharmacological drug administration, surgery or radiotherapy. If complete elimination fails, the drug is used to reduce the number of neoplastic cells to improve the patient's chance of survival and reduce their symptoms. A major problem is that the tumour cells will usually be in different phases of the cell cycle, the majority of them in a phase of decelerated growth. Cells suffer decelerated growth due to the lack of physical space and nutrients as a result of competition, and this can also occur in larger tumours. The drugs work best when the cells are in the dividing (exponential) phase, thus the vast majority of the cells will not be affected and show resistance to the chemotherapy drug. One of the problems with antineoplastic drugs is that they are not specifically selective for neoplastic cells and will eliminate normal healthy cells as well, especially those that divide often. As multiple cell lines are destroyed, organ toxicity occurs. Cells responsible for triggering many of the body's crucial immune responses are also affected, depressing the patient's immune system and increasing their risk of developing viral or bacterial infections.[6]

Chemotherapy drugs can be divided into groups based on several factors including their chemical composition and function. Alkylating agents directly damage DNA to prohibit the tumour cells from dividing further. Antimetabolites interfere with the synthesis of DNA and RNA by substituting for the normal building blocks required for normal DNA replication and transcription. Antitumour antibiotics (anthracyclines) interfere with the enzymes involved in DNA replication and are capable of inflicting their action regardless of what cell cycle phase the cell is in, though the preference would be in mitotic cells.[6] Topoisomerase inhibitors interfere with topoisomerase, which is an enzyme responsible for separating the double strands of DNA, whilst mitotic inhibitors are natural products that stop mitosis, usually interacting with and perturbing the microtubule spindle machinery facilitating mitosis. This review will focus on the anthracycline doxorubicin, one of the most commonly used chemotherapeutic drugs to date. Its primary action is to inhibit topoisomerase I and II and intercalating into DNA to interfere with its uncoiling, ultimately inducing programmed cell death.

Doxorubicin – brief background

In its unaltered form, doxorubicin has shown great treatment potential, being regarded as one of the most potent of the Food and Drug Administration-approved chemotherapeutic drugs.[7] The ability to combat rapidly dividing cells and slow disease progression has been widely acknowledged for several decades, limited only by its toxicity on noncancerous cells in the human body. The drug is a nonselective class I anthracycline, possessing aglyconic and sugar moieties. The aglycone is comprised of a tetracyclic ring with quinine-hydroquinone adjacent groups, methoxy substituent short side chain followed by the carbonyl group. The sugar component (also known as daunosamine) is attached to one of the rings by a glycosidic bond. This is comprised of a 3-amino-2,3,4–trideoxy-L-fucosyl moiety.[8]

Doxorubicin metabolism

Many studies on the pharmacokinetics of doxorubicin have been conducted assessing the treatment range from single or multiagent therapy against a range of tumour types. Most of these studies have shown doxorubicin disposition to be multiphasic after intravenous injection. When intravenously infused it is often followed by a triphasic plasma clearance. This gives the distribution half-life of doxorubicin of 3–5 min, pointing to the drug's rapid uptake by cells. Doxorubicin's terminal half-life of 24–36 h suggests doxorubicin takes far longer to be eliminated from the tissue than its uptake.[9] Steady-state distribution of the drug is imperative to reduce the risk of toxicity. The range of steady distribution ranges from 500–800 l/m2, and this allows bodily tissues to take up a potent amount of doxorubicin.[9] Doxorubicin and its major metabolite doxorubicinol bind to plasma proteins. Like most drugs, doxorubicin enters the cell via passive diffusion, generally accumulating to intracellular concentrations that exceed the extracellular compartments by 10- to 500-fold. The doxorubicin within the nuclear compartments are higher than located in the cell cytoplasm by 50-fold. More specifically, the amount of doxorubicin within the nucleus can reach a saturation level of 340 µm, this can be represented as one molecule of doxorubicin being intercalated at every fifth base pair on the DNA strand.[10,11] The free intracellular doxorubicin left (2% of total intracellular drug) is equally distributed indiscriminately among the other organelles (Golgi apparatus, lysosomes and mitochondria).[12,13] Doxorubicin seems to accumulate mostly in the liver, most likely due to the organ's role in metabolism. In conjunction, the concentration of doxorubicin in bone marrow and white blood cells is 200–500 times higher than in the plasma. As doxorubicin is distributed into tissues rapidly, this speed of distribution is matched by the rapid drop of doxorubicin levels in the blood. Doxorubicin has the ability to penetrate tissues highly effectively, whilst its ability to remain inside nucleated cells is due to its lipophilic characteristics and DNA intercalating/binding properties. Intriguingly, despite the high penetrating power of doxorubicin, it cannot pass through the blood–brain barrier.

The chemical modification of doxorubicin in the body (biotransformation) occurs in the liver. The ketone of the C-13 yielding doxorubicinol molecule undergoes stereo-specific reduction. For this process to occur, two metabolising enzymes are needed; aldo-ketoreductase and ubiquitous cytoplasmic carbonyl reductase. For both doxorubicin and doxorubicinol to be metabolised, both must undergo a range of reactions such as hydrolytic glycosidic and reductive cleavage, O-sulfation, O-demethylation and O-glucuronidation. The sugar components of both doxorubicin and doxorubicinol are eliminated when the glycosidic bonds of the sugar undergo acid-catalysed hydrolysis. The hydrolysis of the glycosidic bond causes doxorubicinone to be derived from doxorubicin and doxorubicinolone from doxorubicinol.[9]

A semi-quinone intermediate and subsequent protonation of the C7-aglycone radical removes the C7-linked daunosamine sugar group to produce 7-deoxydoxorubicinone from doxorubicin and 7-deoxydoxorubicinolone from doxorubicinol. In order for the 7-deoxyaglycones to be excreted, they must first undergo demethylation to conjugate with sulfonic or glucuronic acid. It is important to note that doxorubicin aglycones have only been detected in the biological fluids of some patients, and in minute concentrations in comparison with doxorubicin and doxorubicinol.[14–16]

Plasma clearance of doxorubicin is normally in the range of 324–809 ml/min/m2 ( 2011) and is mediated by the hepatobiliary pathway. Half of the excreted drug is in the bile, usually being excreted within 5–7 days of treatment, whilst only 5–12% of the drug appears in the patient's urine during the same time period, with 3% of the drug found in urine in the form of doxorubicinol. After 24 h, 10–20% of the drug is excreted in faeces, and 50% after 150 h. In obese women, the rate of systemic clearance is notably decreased. Following a bolus injection of doxorubicin the plasma concentration shows levels of doxorubicinol rapidly increasing and depleting parallel to doxorubicin levels. If the level of drug infusion continues for an extended period of time, the concentration of doxorubicinol will eventually transcend over doxorubicin.[9]

Drug action (biological effects) – apoptosis, autophagy and necrosis

As mentioned above, doxorubicin acts by binding to DNA-associated enzymes, it can intercalate the base pairs of the DNA's double helix.[8] By binding to multiple molecular targets such as topoisomerase enzymes I and II, a range of cytotoxic effects occur in conjunction with antiproliferation, thus resulting in DNA damage.[17] The apoptosis pathway is triggered when the attempt to repair the breaks in DNA fail and cellular growth is inhibited at phases G1 and G2. Doxorubicin is also known to intercalate itself into the DNA, with the inhibition of both DNA and RNA polymerase, ultimately ceasing DNA replication and RNA transcription.

This process occurs as doxorubicin enters the cell through diffusion using its higher affinity to bind to the cytoplasm's proteasome. A doxorubicin proteasome complex is formed when doxorubicin binds to the proteosome's 20S subunit, where it is then translocated through the nuclear pore complexes into the nucleus. Doxorubicin has a higher affinity for nuclear DNA over the proteasome it is attached to, allowing it to dissociate itself from the proteasome and bind to the DNA.[8] Other doxorubicin actions include free radical generation which causes further DNA damage, inhibition of macromolecule production, DNA unwinding/separation and increase in alkylation.[18,19] A recent study reported doxorubicin's ability to intercalate with not only nuclear DNA, but also mitochondrial DNA.[20] Furthermore, doxorubicin can affect the cell membrane directly by binding to plasma proteins causing enzymatic electron reduction of doxorubicin. This can cause the formation of highly reactive species of hydroxyl free radicals. Free radicals are responsible for the dangerous side effects of toxicity elicited by the drug's use, though these same mechanisms make doxorubicin a potent anticancer drug, allowing it to be efficacious against various forms of cancer.

Results have suggested also that doxorubicin results in autophagy, being cytoprotective as a response to DNA damage. Nuclear enzyme (poly (ADP-ribose) polymerase-1) (PARP-1) activation is a vital event that decides whether the cell will undergo autophagy. Genotoxic stress can cause PARP-1 to hyperactivate, which in turn depletes both NAD+ and ATP. The cell will then experience energy failure that would be irreversible, resulting in cell death.[19] PARP-1's role in doxorubicin-induced autophagy is made evident in the testing of PARP-1-deficient cells, where moderately high concentrations of doxorubicin were unable to induce cell death. Normally, when doxorubicin hyperactivates PARP-1 the cell experiences cell death that shows characteristics of both autophagy and necrosis. This event shows evidence that the key event that triggers the cell to enter autophagy is the collapse of cellular energy as a result of PARP-1 hyperactivation. PARP-1 inhibition prevents cellular energy collapse and thus protects the cell from necrosis and delays doxorubicin-induced autophagy, often resulting in apoptosis. This study shows PARP-1's activation to be an important regulator of a cell's fate. This is simply because PARP-1 can repair lesions caused by doxorubicin treatment, whilst during its inhibition PARP-1 cannot repair the DNA damage, resulting in cell death. Ultimately, PARP-1 requires a balance between energy crisis and DNA damage, which depends on the dose of doxorubicin administered and PARP-1 activation.

The specific form of cell death resulting from doxorubicin treatment varies depending on the concentration of the drug, treatment duration and specific form of cancer. By determining the optimum concentration that would effectively kill cancer cells whilst minimising toxicity. Also, by evaluating further molecular pathways and proteins involved in drug resistance whilst furthering our current knowledge on ceramide and the ABC transport family, future drug therapies could be optimised to effectively treat cancer.

Examples of molecular signals activated by/involved in doxorubicin action in normal and cancer cells

Doxorubicin activation of AMP-activated protein kinase and induction of apoptosis

Cellular mechanisms of chemotherapeutic agents are constantly studied to optimise cancer treatment. In regards to myocardial toxicity, studies have shown that doxorubicin induces AMP-activated protein kinase (AMPK) activation in embryonic ventricular myocardial H9c2 rat cells. When reactive oxygen species (ROS)-dependent liver kinase B1 (LKB1) is activated it provides the upstream signal necessary for AMPK activation. Once activated, doxorubicin-induced H9c2 apoptosis occurs. Recently discovered, AMPK is a protein kinase that acts as an intracellular energy status sensor, activated to reserve cellular energy and a fundamental regulatory role of both cell survival and death under pathological stress (for example osmotic, hypoxia or oxidative stress).[21–34] To mediate its apoptotic effects, the modulation of multiple downstream targets must occur such as c-Jun N-terminal kinase (JNK), p53 and mTORC1regulation.[30,35–38] An example of this process is seen in apoptosis of insulin-producing MIN6 cells and ϐ cells, where activation of AMPK leads to JNK activation, ultimately inducing cell death in cultured cells.

AMPK activation induces p53 activation, which initiates apoptosis in B16 melanoma cells.[39] It is thought that doxorubicin induces the activation of AMPK via ROS-dependent LKB1 activation, allowing it to act as the upstream signal. Doxorubicin-induced ROS production can occur in multiple ways. The first involves triggering a Fenton-type reaction by chelating intracellular iron, producing hydroxyl radicals that are highly reactive. Other reactions involve ‘redox cyclers’ that react with flavoprotein reductase to generate superoxide when molecular oxygen is present. In detail AMPK modulates cell death by activating JNK kinase in liver cells and insulin-producing β cells.[31,35] This stimulates AM251, a cannabinoid antagonist that depresses hepatoma HepG2 cell viability via the JNK pathway. The detailed mechanisms behind JNK stimulation by AMPK are relatively unknown, warranting future investigation. A recent study highlighted AMPK as being the most important upstream signal for the phosphorylation of p53 at serine 15, as well as its activation in the presence of doxorubicin, ultimately causing programmed cell death and arresting the cell cycle.[25,26,39]

The effects of doxorubicin on the Bcl-2/Bax apoptosis pathway

Other studies have explored the effects of doxorubicin on the Bcl-2/Bax apoptosis pathway.[40] By treating the breast cancer cell line MCF-7 with doxorubicin, a decrease in time- and concentration-dependent Bcl-2 occurred whilst Bax increased. Chromosomal rearrangement can cause the overproduction of Bcl-2 proteins, leading to an increased survival rate of lymphocytes. Bcl-2 is known for its antiapoptotic properties, which are usually opposed by Bax, and this balance between the two is critical as to whether a cell enters apoptosis or survives.[41,42]

Some studies suggest that doxorubicin employs its effects by stimulating components within the Fas/Fas ligand apoptosis pathways, but other studies have produced contradictory results.[43,44] Doxorubicin is shown to downregulate Bcl-2 mRNA levels, a mechanism that is thought to be p53-independent. When a chemotherapeutic drug such as doxorubicin is administered, p53 levels are often increased, as seen when tested in MCF-7 cells, hence activating the p53 pathway. This suggests that the doxorubicin effect on Bcl-2 expression is mediated by p53 pathways. Theoretically, this means the ratio of Bcl-2/Bax must be altered as seen with doxorubicin treatment, allowing downstream activation of different caspases.[45] To achieve the required threshold in the Bcl-2/Bax ratio the chemotherapy drug must be delivered at a higher concentration or incubated for a longer duration. In addition, drugs like doxorubicin may require multiple pathways to modulate apoptosis other than the Bcl-2/Bax pathway. The Bcl-2/Bax complex is well known for its anti-apoptotic and pro-apoptotic function. The shift in ratio between the two proteins is an extremely vital event that ultimately determines the fate of a cell, especially during doxorubicin treatment.

Doxorubicin resistance

It is paramount to continually study the mechanisms of anticancer drug resistance to further understand doxorubicin action and explore apoptotic and/or autophagic biomarkers. This could potentially overcome resistance for better doxorubicin formulation and efficacy.

High mobility group box 1 protein-mediated autophagy and drug resistance

Autophagy has the ability to motivate and inhibit anticancer drug resistance depending on the nature of the drug, as well as the duration of the treatment and its impact on metabolic stress. The unique properties of each tumour type can also influence drug resistance. The high mobility group box 1 protein (HMGB1) has a significant role in regulating nuclear events such as DNA replication and repair. It is also a critical regulator of both selective and nonselective autophagy.[46] Despite this, it has shown to significantly contribute to drug resistance in osteosarcoma cells. Results have shown that inhibiting both autophagy and HMGB1 increases osteosarcoma cell sensitivity to doxorubicin both in vivo and in vitro. When doxorubicin is administered, mRNA and HMGB1 levels increase in osteosarcoma cells. The mechanisms behind this event are still unknown, although HMGB1 competes with Bcl-2 in an attempt to bind to BECN1.[46] The formation of BECN1-PtdIns3KC3 complex results and stimulates autophagy by evoking autophagosome maturation. For the HMGB1 and BECN1 interaction to occur, the upstream signal activation of the complex ULK1-mATG12-FIP200 must occur. HMGB1's ability to mediate autophagy and promote drug resistance in osteosarcoma when treated with doxorubicin shows great potential as a chemotherapeutic target in osteosarcoma.

Role of ceramide in cell resistance to doxorubicin

Other studies have explored ceramide's contribution to doxorubicin resistance in cells.[47] Ceramide is a cellular lipid messenger that modulates doxorubicin-induced cell death. Results show that doxorubicin enhances ceramide that causes the over expression of glucosylceramide synthase (GCS) genes and thus ceramide levels in the breast cancer cell line MCF-7. By enhancing ceramide glycosylation, the endowing cellular stress by doxorubicin results in the cells evolution to become cellular resistant to the drug. The exact signalling pathways for doxorubicin-mediated gene expression is still unknown. It is theorised that doxorubicin increases ceramide levels by either activating the enzymes responsible for ceramide synthesis or activating sphingomyelinase. These two processes cause apoptosis to occur through downstream signalling, yet GCS expression is upregulated by ceramide. This is a positive feedback cycle that occurs through Sp1. The cycle is known to be anti-apoptotic and is the driving force behind cellular resistance to anticancer drugs that are known to generate ceramide.[47] As previously mentioned, the transduction pathways involved in switching ceramide-induced apoptosis to ceramide-mediated drug resistance is still unknown.

Doxorubicin resistance in melanoma cells have shown to be ABCB8-mediated by protecting the genome of the mitochondria. ATP-binding cassette (ABC) proteins are expressed in a range of tumour cells. By inhibiting ABC transporters the efficacy of certain anticancer drugs greatly increases. Elliott and Al-Hajj[48] have shown that ABC transporters are responsible for chemotherapeutic resistance in many types of cancer. However, the role of ABC transporters on doxorubicin-induced resistance in melanomas is still unclear. ABCB8 has shown a considerable role in doxorubicin resistance specifically in melanomas, although the exact role of these proteins is still unknown. ABCB8 knockdown with shRNA showed a decrease in doxorubicin resistance by 3–4-fold. The mechanisms behind this form of drug resistance could be that ABCB8 is protecting the mitochondrial DNA from damage. It is known that the mitochondria of cancerous cells can undergo alterations to adapt to the demanding microenvironment on a tumour. Other studies claim that tumour drug resistance could be due to the mitochondria having a low membrane potential, ultimately protecting themselves from ROS by undergoing a high rate of glycolysis.[49] It is important to note that ABCB8 drug resistance is specific to doxorubicin and not other chemotherapeutic agents that are structurally different. Whilst the ABC transport family is known to cause resistance in a wide range of tumour types, ABCB8's signalled events in melanoma cell resistance is still unknown and demands further study to potentially target resistive cell lines of melanoma successfully.

Doxorubicin toxicity

As previously mentioned, the main side effects of anthracycline drugs are the multidirectional cytotoxic effects, with cardiotoxicity being the most prominent. Early clinical assessments during the 1970s on doxorubicin during phase II and III studies showed common side effects of acute vomiting and nausea, gastrointestinal problems, baldness, and disturbances to the neurological system (often causing hallucinations and light-headedness).[7,50] Unfortunately doxorubicin is not specifically targeted to the tumour, and it can affect the growth of many other cell types in the body. This results in the immune system becoming depressed, and as the numbers of immune cells reduce, the patient becomes more susceptible to microbial infections, fatigue and healing time decreases. The severity of these effects and their occurrence depends on the dosage of doxorubicin and the regeneration capacity of the patient's bone marrow.[51] By continuously administering doxorubicin into a small vein, problems such as phlebosclerosis (thickening of vein walls) can occur, and if extravasations occur in local tissues or organs, necrosis can arise. Other side effects of extravasation include cellulitis, thrombophlebitis and joint movement limitation as a result of painful induration (pathological hardening of tissue).[7]

Toxicity in the heart

The greatest risk of doxorubicin-induced toxicity is cardiotoxicity and various mechanisms exist for this (Figure 1). For this reason, administration of doxorubicin must be dose-limited. Patients are affected differently, with conditions varying from acute to chronic. Doxorubicin is responsible for the structural alterations of the cardiomyocytes in the heart, enlargening them. The genes responsible for this are the brain natriuretic peptide (BNP) and atrial natriuretic peptide (ANP). When highly expressed, these two genes are responsible for causing cardiac hypertrophy. An alternative way for cardiotoxicity to occur is through doxorubicin's impact on the mitochondria in cardiac muscle. By altering the level of mitochondrial protein expression, redox cycling of doxorubicin and NADH dehydrogenase are intensified.

Figure 1.

Potential mechanisms involved in doxorubicin apoptosis (white and gray solid boxes) and cardiomyopathy (dotted lined boxes). ARC, apoptosis repressor with caspase recruitment domain; DNA, deoxyribonucleic acid; DOX, doxorubicin; Bax, Bcl-2–associated X protein; B-cell lymphoma 2; ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal kinase; MAPK, mitogen-activated protein (MAP) kinase; MuRF, muscle RING finger-1; NFAT5, NRG1, Neuregulin-1; NFAT5, nuclear factor-activated T cell-5; NFKB, nuclear factor (NF) κB; p38, protein 38; p300, protein 300; TauT, taurine uptake transporter; TLR, toll-like receptor; TLR2, toll-like receptor-2; TLR4, toll-like receptor-4; UPS, ubiquitin proteasome system.

If doxorubicin is present, the levels of ROS are attuned by toll-like receptor 4 (TLR4), which increases the level of tumour necrosis factor-α (TNF-α). When ROS levels increase, the apoptotic cascade is activated by cytochrome c being released from the mitochondria. If a lifetime accumulation of doxorubicin nears 500 mg/m2, the risk of cardiomyopathy increases often resulting in congestive heart failure in 20% of patients.[52] As previously mentioned, the molecular mechanism behind this event involves the formation of oxygen free radicals and iron oxidation. Since doxorubicin is known to affect multiple biomarkers, the assessment of troponins and specific natriuretic peptides (proBNP and DNP) is believed to help predict doxorubicin-induced cardiotoxicity in its early stages.[53] Antioxidant administration has shown no signs of nullifying toxicity levels, whilst studies involving selective iron chelators have also returned negative or inconsistent results. These tests suggest that other mechanisms independent of both ROS and iron could contribute to doxorubicin-induced cardiotoxicity.[54] Other studies have explored the downregulation of GATA binding protein 4 (GATA-4), activation of p53 and p300 degradation, as these proteins are affected as a result of doxorubicin administration, leading to the apoptosis of cardiomyocytes. [55–58]

GATA-4 is a known transcription factor that acts on post-natal and differentiated cardiomyoctes as a survival factor. It has also shown to activate Bcl-xL, an anti-apoptotic gene.[54] As many patients undergo chemotherapy they begin to experience fatigue and ultimately suppression of their immune system. Doxorubicin can stimulate the production of cytokines and stimulate cytotoxic T lymphocyte responses while simultaneously enhancing the activity of natural killer (NK) cells.[59] By stimulating these innate and adaptive immune response components, the effects on the cardiac system become astounding, resulting in aggressive cardiac damage.

Toxicity in the Brain

When exploring doxorubicin-mediated brain toxicity, it is important to note that toxicity caused to the brain is indirect because doxorubicin is incapable of crossing the blood–brain barrier.[60,61] Doxorubicin stimulates the production of TNF-α, triggering the microglial cells in the brain to produce inflammatory cytokines. When TNF-α is produced in excess, it activates the expression of inducible nitric oxide synthase (iNOS) causing the levels of reactive nitrogen species (RNS) to rise. As RNS levels continually increase, the surrounding proteins (for example manganese superoxide dismutase, MnSOD) undergo nitration. After the protein has undergone nitration it stimulates the generation of ROS, enhancing the mitochondria's permeability transition pore. When the pores of the mitochondria release cytochrome c, the process of cell death occurs via apoptosis.[7] Ultimately these actions are responsible for the various domains of the brain that undergo cognitive impairment.[62] Patients usually show signs of impairment in both cognitive recall and visuospatial skills, but the majority of cognitive functioning is regained after one year without doxorubicin treatment.[63]

Toxicity in the liver

Another common site for doxorubicin-induced cell death and tissue damage is in the liver, with 40% of patients undergoing the treatment suffering some form of liver injury. Known for its metabolism and detoxification activity, the liver treats doxorubicin just like everything else. Thus, when a patient is being treated with the drug it is the liver that receives most of the drug, letting it accumulate and then metabolising it.[64] By metabolising the high concentrations of doxorubicin, a vast number of ROS are produced. Consequently ROS causes an excessive amount of damage ranging from DNA damage, production of lipid peroxidation, decreasing vitamin E levels, decreasing reduced glutathione (GSH) and causing oxidative processes to become imbalanced. The overproduction of ROS can cause IκB kinase (IκK) activation which phosphorylates the IκK inhibitors to activate nuclear factor kappaB (NFκB). NFκB then activates pro-inflammatory cytokines to bring about apoptosis. The final major action brought on by doxorubicin-mediated toxicity is that it is responsible for decreasing levels of inorganic phosphate, both ADP and ATP as well as AMP which causes pathological conditions in hepatocytes. The ATP-binding cassette is needed to mediate doxorubicin efflux through the cells to stop the intracellular compartments and space from accumulating doxorubicin, thus acting as a form of homeostasis to keep the internal fluid levels constant. However, the problem is that for this homeostasis mechanism to work a constant supply of ATP is needed to fuel the process, and since doxorubicin causes ATP to decrease, the cell's ability to perform energy-dependent tasks also decrease. This is one of the reasons patients undergoing chemotherapy begin to weaken and feel constant muscular and mental fatigue.

Toxicity in the kidneys

When toxicity in the kidneys occurs doxorubicin is known to cause nephropathy and proteinuria by injuring glomerular podocytes.[65–70] Doxorubicin-induced nephropathy occurs when the drug interferes with the normal functioning of the mitochondria, reducing the activity of complexes I and IV of the mitochondria. This causes the levels of triglycerides, superoxides and citrate synthase to increase, whilst levels of vitamin E and antioxidant compounds are reduced with lipid peroxidation occurring. The nephron structure is altered when leaked proteins from local passages come in contact with exposed renal tissue, ultimately leading to glomerulosclerosis. The glomeruli-affecting disease is known to cause hypertension, resistance to steroids and proteinuria, eventually leading to renal failure.[7] Unlike the liver, the kidney's ability to regenerate is poor, reducing its ability to heal itself when glomeruli are damaged. This heightens the degree of susceptibility to damage in the entire body especially the kidney itself because of its major role in regulating the chemical composition of blood and maintaining fluid balance. When the glomeruli becomes damaged it is unable to perform its normal functions causing glomerular lesions, inflammation, tubular dilation, and also affecting capillary permeability.

Other adverse reactions

Some patients have shown other adverse reactions from doxorubicin treatment such as cutaneous injuries. The patients suffering from cutaneous injuries may show signs of hyperpigmentation in dermal creases and nail beds, alopecia, itching, photosensitivity and rashes. Doxorubicin is also known to cause fever, urticaria and anaphylaxis that are associated with hypersensitivity. Further incidences have shown doxorubicin affecting the gastrointestinal system causing patients to vomit and experience mucositis during the early stages of therapy (5–10 days). Most patients recover within the next 10 days of treatment, although those with severe reactions have shown signs of ulceration and necrosis, often causing severe infections in the colon and caecum which can be fatal ( 2011).

Drug delivery systems: combination delivery

The anticancer drugs currently used in chemotherapy show cytotoxicity that is nonspecific, affecting normal cells. The treatment is shown to cause serious side effects to the cancer patient due to the unpredictable cytotoxic properties. For this reason, various experiments have been performed to develop specific drug delivery systems capable of restricting the drug's toxicity to aim its effects directly at the tumour as much as is possible. The use of these drug delivery systems are constantly being tested and improved to enhance the efficacy, selectivity and total effect of the anti-neoplastic drugs. Current drug delivery systems that show great promise include the entrapment of drugs into polymeric drug carriers, such as liposomes, hydrogels, and nanoparticles.

Liposomal delivery

The initial development of the liposomal drug delivery system showed great promise, yet it also demonstrated some critical shortcomings. The liposomes showed signs of leakage, easy recognition and removal from the circulatory system via the reticuloendothelial system (RES).[71,72] Early studies determined the reason for this was due to the particles' composition and size, affecting overall circulation time. Over time, liposomes have been formulated to reduce leakage by encapsulating the drug more effectively to remain in the circulation longer. For the drug to be successful it must overcome the three main clearance pathways controlling an intravenously injected drug's biodistribution and pharmacokinetics. The pathways consist of uptake by the cells of the bone marrow, spleen and liver from the drug-infused liposomes in the circulation, followed by drug metabolism and removal, drug leakage from the liposomes as well as the rapid elimination of the drug in free form, and finally the accumulation of drug-loaded liposomes in the tissue.[73]

To prevent initial drug leakage the lipid bilayer composition itself is altered, in conjunction with protocol changes to drug loading to stabilise the liposome's ability to contain the drug effectively.[74,75] Liposomes were soon coated with polyethylene glycol (PEG) to prolong circulation time. The PEG synthetic hydrophilic polymer, consists of a large headgroup which acts as a barrier to disrupt RES recognition, by preventing plasma opsonin interaction due to highly hydrated groups that inhibit electrostatic and hydrophobic interactions of a range of blood components on the surface of the liposome. The liposomes coated with PEG eventually became known as ‘stealth’ liposomes, one of which was eventually commercialised with doxorubicin. [76]

The lamellar vesicles in the liquid suspension of the stealth liposomes marketed as doxorubicinol are approximately 80–90 nm in size, with a lipid content of 16 mg/ml and mean doxorubicin concentration of 2 mg/ml.[73] The internal aqueous compartment of liposomes is separated from the external medium by a single lipid bilayer membrane. The doxorubicin is encapsulated within the internal compartment, in a gel phase consisting of 10 000–15 000 tightly packed molecules per vesicle. The diameter of these liposomes must be small enough to allow extravasation of the drug into the gaps found in permeable tumour blood vessels to reach the affected tissue.[77,78]

Properties of doxorubicin-loaded liposomes

Doxorubicin-loaded pegylated liposomes consist of liposomes that have undergone pegylation, where methoxypolyethylene glycols are bound to its surface. The anthracycline antibiotics are then encapsulated within the liposomes. Pegylation allows the liposomes to remain in the blood circulation for longer, whilst giving it stealth properties that allow it to go undetected by the mononuclear phagocyte system.[79,80] Animal studies examining the pharmacological properties of pegylated liposomes have shown inhibited tumour growth rate in different models of cancer, improving the overall survival of the organism and quality of life in comparison with regular doxorubicin treatment. One of the key pharmacological properties was the drug's ability to suppress anti-apoptotic pathways and further amplify apoptotic activity.

This synergistic effect has proven effective in a range of tumour model systems when delivered in combination with proteasome inhibitor bortezomib. The mean volume of doxorubicin distribution is approximately 3 l, similar to plasma volume when the combination is delivered intravenously, whilst the mean half-life was approximately 79.9 h with a clearance rate of 33.6 ml/h. Traditional doxorubicin has a far greater volume distribution (700–1100 l/m2) and clearance rate of 24–73 l/h/m2.[81]

To successfully encapsulate doxorubicin into the liposomes and ensure their effective delivery, optimal formulations have to be determined to balance two requirements; high drug payload and small vesicle size allowing extravasation into tumour tissue. Reducing the vesicle size would cause issues regarding the final payload, due to the reduced drug/lipid ratio. This is circumvented with increasing liposome drug solubility and trapped volume. The drug/lipid ratio is increased by a method known as active (remote) loading, using an ammonium sulphate gradient to force doxorubicin into the formulation of liposomes.[73] In other words, the drug is loaded post-manufacture of liposomes, whereas the majority of liposomal formulations encapsulate the drug during the formulation process.

Pharmacokinetics of doxorubicin-loaded pegylated liposomes

The pharmacokinetic profile of doxorubicin-loaded pegylated liposomes is quite different to doxorubicin alone. One of the main differences is the smaller volume distribution and clearance rate.[79,80] Most liposomes are removed from the circulation by the RES, yet this clearance pathway is overcome with a hydrophilic coating on the liposome's surface. This greatly reduces the level of interaction between different plasma components and the liposome itself.[82] This allows it to remain in the bloodstream for longer after intravenous administration. It is important to note that the 95% of doxorubicin found in the plasma is still encapsulated within the liposomes, meaning it is not bioavailable.[73] Drugs encapsulated within pegylated liposomes, specifically doxorubicin, holds the potential to be delivered selectively at tumour sites due to their small size (approximately 100 nm).

The liposomes' size allows them to penetrate through defective blood vessels supplying the tumour. Some studies have shown that higher concentrations of doxorubicin given over prolonged periods accumulate at the tumour site when delivered via liposomal encapsulation in contrast to conventional doxorubicin treatment.[82] Studies also show a reduced uptake of doxorubicin in the heart than that of conventional free doxorubicin when delivered by liposomal encapsulation. While doxorubicin uptake in the heart decreases, the uptake by the liver and skin increases 4–48 times after the initial injection, hence these particles are still not ideal, and the use of doxorubicinol has not been as prevalent in the clinic as initially expected. Studies conducted in rats with brain implanted sarcoma showed that the doxorubicin-loaded pegylated liposomes thoroughly penetrated the intracerebral tumours at the same time sparing the surrounding healthy brain tissue, depicting a critical advantage over free doxorubicin.


An earlier study reported the poor absorption rate of intravenous administration of doxorubicin, highlighting the need for a targeted drug delivery system to replace intravenous delivery specifically in osteosarcoma patients.[83] The study reported 1 g marrow accumulating 0.52 µg doxorubicin from an administration of 30 mg/m2. Intravenous administration of doxorubicin makes it quite difficult to treat bone-related tumours, given that bone is usually not a well-vascularised tissue, though this may change dramatically when a tumour grows at the bony site. Drug delivery systems that do not rely on intravenous administration such as hydroxyapatite implants and magnetic liposomes are usually avoided because the surgical insertions cause patient discomfort.[84,85]

Recently, it has been reported that hydrogel drug delivery systems may hold some promise in locally treating bone tumours, as this is feasible in human patients.[86] The system developed was a noncytotoxic, biodegradable in situ chitosan/dipotassium orthophosphate (Chi/DPO) hydrogel.[86] Over the course of 19 days, the Chi/DPO hydrogel delivery system administered doxorubicin to five-week-old female nude mice injected with the human osteosarcoma SaOS-2 cell line. The hydrogels remained liquid at room temperature and rapidly turned to gel when injected (37°C) to prevent drug leakage.[86] By keeping the Chi/DPO pH at the optimum physiological range (7.4), the hydrogels provided a desired low initial burst effect over a prolonged duration. The hydrogels decreased SaOS-2 cell proliferation by enhancing apoptosis. The increase in apoptotic activity was approximately fivefold higher than when doxorubicin was incorporated inside the hydrogels.[87] When analysing orthotopic tumour tissue histology, apoptosis levels were at an average of 37% in the hydrogel-doxorubicin cohort in comparison with the free doxorubicin inducing apoptosis rate below 15%. X-ray images showed a decrease in osteolysis in the Chi/DPO doxorubicin-treated group. Furthermore, this delivery system revealed a reduction of systemic side effects from Chi/DPO doxorubicin-hydrogel release. The controlled release of the drug decreased cardiotoxicity levels in the mice when compared histologically, whilst untreated mice exhibited vacuolisation within cardiomyocytes indicating toxicity.[87]When injected with the hydrogels, necrosis at the site of injection was also absent, confirming the reduction in side effect of drug delivery systems when compared with free doxorubicin. One possible explanation for this is that much less doxorubicin travels to other organs as only small bursts of doxorubicin are continually released, thus minimising side effects. Sustained and controlled release of doxorubicin is not influenced by the strength of the hydrogels or concentration of chitosan, rather the level of orthophosphate salt, the main factor influencing the initial burst effect and drug release. The smaller the amount of salt, the higher the release of entrapped agents. Ultimately, the degree of burst release of molecules can be modified by altering the amount of salt in the large outer pores of the hydrogels. One limitation of the Chi/DPO hydrogels is the plateau of doxorubicin release within a relatively short time.[87] Despite this, it is hypothesized that faced with in-vivo conditions, the release rate would be far greater due to the body's degradative impact on the hydrogel matrix that would cause the chitosan network to degrade and release doxorubicin. Indeed, in-vitro, when lysozyme (known to hydrolyse chitosan) was added to the hydrogels, doxorubicin was readily released.[87] When this enzyme was contained inside the hydrogel's structure, the initial burst rate increased by 2%, whilst release rate increased by sixfold. In conjunction to this, the release rate did not plateau and was well-sustained over the entire 19-day period, further highlighting the efficient and effective use of Chi/DPO hydrogels in terms of sustained and controlled doxorubicin release. Conclusively, by incorporating doxorubicin inside Chi/DPO hydrogels, the cell growth of various forms of cancer are significantly reduced as well as the undesirable side effects seen in abundance with free doxorubicin treatment.


As previously mentioned, the development of new drug delivery systems to improve drug delivery efficiently in combination with low cytotoxicity is constantly being explored. A recent study demonstrated nanoparticles as a carrier for delivery of doxorubicin to osteosarcoma cells.[88] In the past, macromolecular materials such as dextran and polyethyleneimine (PEI) have been used successfully as potential chemotherapeutic delivery carriers.[89] One such example produced dextran-graft-PEI nanoparticles loaded with doxorubicin.[88] Combination therapy is now commonly seen to effectively treat cancer. This combinational design of nanoparticles is synthesised via an imine reaction between oxidised dextran and PEI of a low molecular weight.[90] Low molecular weight PEI is vital as cytotoxicity increases relative to the molecular weight.[91] It is these PEI segments and oxidised dextran that allows for doxorubicin conjugation. When administered, a higher efficiency in hydrogel drug delivery was observed in osteosarcoma cells in comparison with free doxorubicin. When MG-63 and SaOS-2 cell lines were treated with the doxorubicin nanoparticles, the dextran-doxorubicin-PEI nanoparticles reduced cell numbers by 5–10% at a concentration range of 1–10 mg/l when compared with free doxorubicin over 24–48 h.[88] A possible explanation for this is the nanoparticles are entering the cell not by ligand-receptor affinity but via passive diffusion. Also, the cytotoxicity may be reduced due to slow-coupled doxorubicin release from the nanoparticles. These doxorubicin-coupled nanoparticles also showed selective delivery to tumours. This increases the margin of safety as well as reduces side-effects associated with doxorubicin-loaded dextran-PEI nanoparticles. Another advantage of using nanoparticles as a drug delivery system is that they can be manufactured with ease, with the added benefits of low cytotoxicity and biodegradable properties.

Another recent study involved encapsulation of doxorubicin into a chitosan-based drug delivery system via the complex coacervation method with dextran sulfate.[92] A high drug loading content of more than 99% was achieved, and the study proposed that osteosarcoma cell death with these nanoparticles occurred through apoptosis, necrosis and autophagic cell death. Treatment of mice bearing orthotopic osteosarcoma with the doxorubicin nanoparticles reduced tumour growth, decreased tumour-associated osteolysis, and reduced metastasis to the lungs. Importantly also, these nanoparticles did not cause side effects in mice – specifically to the heart or skin. Thus, doxorubicin-containing nanoparticles may prove to be useful clinically provided further studies are performed to validate such formulations in clinically-relevant animal tumour models.


This review explores the frontline chemotherapy drug doxorubicin, which has been used for treating cancer for over 30 years. Although doxorubicin remains one of the most effective anticancer agents (Table 1), it does not come without a price. Doxorubicin causes life-threatening toxicity to most major organs, primarily life-threatening cardiotoxicity. Doxorubicin is known to undergo various biological actions; binding to DNA-associated enzymes and intercalating with DNA base pairs. Doxorubicin can also target multiple molecular targets to produce a range of cytotoxic effects, ultimately resulting in DNA damage. This damage can result in further complications such as altered gene expression, one of the many ways cardiotoxicity occurs. The drug action causes the activation of various molecular signals often resulting in anti-apoptotic proteins decrease in expression, whilst pro-apoptotic proteins increase in expression as seen in Figure 1 from AMPK inducing apoptosis to influencing the Bcl-2/Bax apoptosis pathway. By altering the Bcl-2/Bax ratio, downstream cleavage and activation of caspase 9 occurs followed by the cleavage and activation of caspase 3. From this point the cell enters the point of no return and activates a series of downstream clients to induce apoptosis (Figure 1). As mentioned, doxorubicin also induces apoptosis and necrosis in healthy tissues causing toxicity in the brain, liver, kidney and heart. By impairing cardiac repair mechanisms and destroying local bone marrow cells, patients become prone to cardiotoxicity. Once such pathway involves the increased expression of NFKB, as a result there is an influx of inflammation and necrosis of cardiac tissue, resulting in cardiomyopathy (Figure 1).

Table 1.  Types of cancers that are treated with doxorubicin
Types of cancers treatable with doxorubicin
Cervical cancer[13]
Endometrial cancer[14]
Pancreatic cancer[15]
Prostate cancer[7]
Head and neck cancer[16]
Adrenal cortex cancer[17]
Skin cancer and mucous membranes (usually in patients suffering from AIDS – Kaposi's sarcoma)[7]
Bone cancer – osteosarcoma, Ewing's sarcoma[18]
Womb and uterine cancers – gestational trophoblastic tumours[7]
Lung cancer[19]
Breast cancer[20]
Stomach cancer[21]
Soft tissue sarcoma[7]
Multiple myeloma[7]
Pancreatic cancer[22]
Thyroid cancer[23]

Over the years, many studies have been conducted to devise a drug delivery system to eliminate these adverse affects. Liposomal doxorubicin drug delivery systems have shown promising results, penetrating intracerebral tumours and sparing the surrounding healthy brain tissue. Studies have shown that liposomes can be altered in multiple ways to optimise delivery. Recent locally-administrable hydrogel drug delivery systems provide an alternative to increase efficacy of doxorubicin. Furthermore, nanoparticles have been modified to act as carriers for doxorubicin, having higher efficiency in drug delivery to the tumour cells in comparison with free doxorubicin. Overall, the future for the continued use of doxorubicin clinically against cancer looks set to be prolonged, provided certain enhancements as listed above are made to its chemistry, delivery and toxicity. Increased efficacy depends on these three aims being met satisfactorily.

Future Directions

The Bcl-2/Bax apoptosis pathway is constantly being explored with various anticancer drugs to further realise the pathway's potential in cancer therapy. Whilst it is known to modulate apoptosis, more emphasis should placed on the Bcl-2/Bax pathway modulators. Doxorubicin is one such chemotherapeutic agent that affects the Bcl-2/Bax pathway. Furthering our understanding of how doxorubicin attenuates levels of Bcl-2/Bax could help realise the pathway's complete potential in chemotherapeutic treatment. It may also lead to other previously unknown signalling pathways that doxorubicin uses besides or in conjunction with Bcl-2 for causing tumour cell death.

Many patients who are successfully treated with doxorubicin develop cardiotoxicity in their later years. The toxicity can also affect other major organs, often being life-threatening. Many mechanisms have been proposed, yet the ‘ROS and iron’ theory is most common. The first step should be to investigate how ROS and iron accumulation produces toxicity, and how doxorubicin increases ROS levels. Doxorubicin is known to produce other side effects such as activation of the innate immune system, changes in gene expression and impairment of the mechanisms responsible for cardiac repair. These other side effects all contribute to overall toxicity. Studies should focus on altering doxorubicin to be more target-selective to prevent the drug from inhibiting bone marrow cell renewal and mobilisation. Recent studies show some potential for nanoparticle and hydrogel doxorubicin drug delivery systems. They are easily prepared and cost-effective compared with liposomes, though they need validation in tumour models.


Conflict of interest

The Author(s) declare(s) that they have no conflicts of interest to disclose.


This review received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.