Everyone knows about cancer. According to the World Health Organization eight million people died of one of the many forms of cancer 2007 and this number is expected to grow to more than 12 million by 2030. However, unlike many other significant diseases, cancer is not confined to a continent or socioeconomic cohort. Also unlike other entrants on the WHO’s top 10 there is no vaccine or wonder drug. This insidious disease requires surgery, chemotherapy or radiotherapy all of which wreak havoc on the patient during and often long after treatment. But recently novel research looking at using certain bacteria as a therapy is gaining traction that may result in new treatment options that are cheap, easy to produce, noninvasive and if the current research is any indication capable of complete remission in some cases.

Figure 1. Clostridium difficile by micrograph

Recent advances in bacterial cancer therapies suggest this idea is a new one but in fact using bacteria as a potential therapy for cancers dates back to the late 1800s and very early 1900s where gangrenous patients were observed to cure themselves of tumours as did patients suffering from what was once called acute onset cellulitis (now known as erysipelas or St Anthony’s Fire). During this time some very informative observations were made:

“I had found one case of malignant round-celled sarcoma of the neck. During the operation to remove the tumour it was identified that it involved the deeper tissues of the neck such that no attempt was made to remove it. The patient suffered from two attacks of erysipelas. During these attacks of erysipelas, the tumour of the neck entirely disappeared and the patient left the hospital in good health. I tracked down the patient and found him alive and well seven years later”—Dr. William Coley, 1920.

The usefulness of bacteria is limited to certain types of cancer as the requirement for this therapy to be useful is tumours large enough to be dead in the middle. As the cancerous cells grow they have to outcompete the cells around them for resources. To do this the cancerous cells unlock various cellular growth checkpoints through mutation such as checking the DNA is in a good condition before division or could start to release factors that accelerate their own growth. This activity eventually results in the formation of a tumourous mass.

This presents a problem for the tumour as nutrients and in particular oxygen can only diffuse so far from the capillaries and if extra tissue is being laid down between capillaries, essentially pushing them apart, then tissue death will occur beyond where the nutrients can diffuse to.

This puts pressure in the growing tumour, as cancerous cells are amongst those dying, a common solution must be developed to supply nutrients. This process is called angiogenesis. Angiogenesis is the direct growth of blood vessels and requires the presence of hypoxic cells and angiogenic signals. The ability of cancer cells to promote angiogenesis is seen in almost all tumours that exceed 1 mm in diameter or become located 100 mm. from the nearest blood vessel.


Figure 2. Most tumours begin growing within normal tissues (dormant) (a) until they out grow they nutrient availability of the local environment and cell death occurs alongside rapid cell division. The cell death, largely due to hypoxia induces angiogenesis which restores a supply of nutrients and facilitates further tumour growth. Angiogenesis begins with perivascular detachment and vessel dilation (b), followed by angiogenic sprouting (c), new vessel formation and maturation, and the recruitment of perivascular cells (d). Blood-vessel formation will continue as long as the tumour grows, and the blood vessels specifically feed hypoxic and necrotic areas of the tumour to provide it with essential nutrients and oxygen (e).

With a newly established nutrient supply the cancerous tumour can continue to grow but eventually the tumour will become so large that the centre of the tumour is too far from a blood vessel to be provided with nutrients and then the cells at the centre of these tumours must die.

Large tumours with dead or necrotic nodes (necrosis can develop as one large deposit or multiple small foci in the centre of the tumourous tissue) are very common and in many cases act as a marker of the primary tumour where metastases are observed. This makes them very interesting target locations for therapeutics even though direct treatment of the necrosis itself has not been shown to aid recovery.

The current limitations with traditional treatments are reasonable well known and this stems largely from the nature if these therapies. Chemotherapy and radiotherapy are designed to kill all fast growing cells including cancerous cells but other cells grow quickly too leading to hair loss, depletion of the immune system, fatigue and fertility problems. It’s the inability to target the therapy that results in much tissue damage associated with treatment. So naturally its been suggested that if a there were a way to target chemo- or radio- therapy these treatments would sho significantly less toxicity. But how do you target tumours alone?

It is here that bacteria can prove their worth. Bacterial species such as Clostridia, Listeria monocytogenes and Salmonella cannot grow well or in some cases at all in the presence of oxygen and so find it very difficult to grow in most locations of the body unless its necrotic.

L. monocytogenes was one of the most heavily researched species early on in regards to a potential therapy due to our ability to control an infection using antibiotics whilst allowing the gram positive aerobe to undergo its normal invasive lifecycle. This approach is very rudimentary but does allow for tumour regression in some cases. To improve the efficacy of the treatment recombinant Listeria expressing tumour-specific antigen were used to great effect in vaccinating against tumours but specificity in these treatments remained an issue as the Listeria were unable to target the tumours and in fact preferred the oxygenated environments instead of the hypoxic environment of the tumour and the surrounding tissue.

In order to increase the specificity of the potential therapy a shift to anaerobes was made. This concentrates the growth of the bacteria within the necrotic centre of the tumours. One of the most investigated species was the gram negative facultative anaerobe Salmonella. Although naturally invasive of all cells, particularly cells of the gut lumen, recombinant strains lacking amino acid synthesis machinery were found to grow preferentially in the necrotic centres of tumours. Further modifications, such as the removal of Lipid A synthesis machinery (a component of the highly immunogenic lipopolysaccharide), made the bacteria even safer and less likely to spread in the host while still retaining the ability to occupy the necrotic lesions of tumours.

But a significant risk remained in administering a live, viable culture of an invasive pathogen into a patient and over time a safer option was again sought after.

More recently researchers have gone back to one of the earliest observations of the interactions between bacteria and tumours to arrive at a species with a number of properties making it an ideal candidate for tumour therapy. Clostridia species are obligate anaerobes and require an oxygen depleted environment to grow and survive. This makes them perfect for targeting necrotic tissue but also makes them incapable of causing systemic disease.

Clostridia also have another trick up their sleeve for researchers to exploit. Clostridia are naturally sporulating bacteria allowing them to only be metabolically active in the locations where the necrotic tissues are found and inert everywhere else. This presents significant advantages as the administration of inert spores that appear to not bother the immune system greatly on primary exposure is much safer than the administration of viable invasive pathogens.

The deliberate use of Clostridial species to treat cancer is one of the earliest attempts to control cancer. By 1813 researchers such as Vautier had observed that gangrenous patients cleared tumours before succumbing to the gangrene but the causative agent of the gangrene was not established at this stage. Continued research over the next 100 years did not make a lot of headway and by the early 1950’s the experimental clearance of tumours in mice was possible but due to the unavailability of a non-pathogenic strain experimental animals did not often survive the treatments.

The slow progress took its toll on the field as researchers re-focused in other areas while non-pathogenic Clostridial species were being identified in the late 50s and early 60’s. In the mid-60s Möse and Möse made the next big breakthrough. After injecting spores of the environmental and non-pathogenic strain Clostridium butyricum (renamed Clostridium oncolyticum and then later renamed again to Clostridium sporogenes) directly into tumours they found regression and in some cases remission in experimental tumour models. Möse and Möse observed that following the injection of spores [that] “The tumour softened noticeably and fluctuated on palpation” and eventually the tumours “broke through … [releasing a] … discharge of brownish liquid necrotic masses which had the consistency of thin pus”.

These observations indicated that even the non-pathogenic strains were resulting in some cell death of the cells bordering the necrotic centres of these tumours and it has since been identified that this cell death occurs due to the build up of toxic bacterial metabolites.

Despite the positive activity observed over the last 20 years in particular a purely bacterial therapy for cancer treatment will not be the full answer to cancer. The real promise lies in combination therapies that place bacterial approaches alongside traditional approaches.

Clostridial approaches have been widely considered the best approach for cancer therapies but until recently the unavailability of genetic tools have meant relying on environmental strains and naturally occurring mutants. With the acquisition of manipulation tools the next phase of bacterial therapies for cancer treatments can begin.

Under extensive research now is the possibility of altering Clostridial species the express pro-drug converting enzymes such as Cytosine Deaminase (CD) or Thymidine Kinase (TK). CD converts the non-toxic 5-Flurocytosine into the cytotoxic 5-Flurouracil and TK phosphorylates the non-toxic Ganciclovir converting it into the active toxic compound. Ordinarily chemotherapeutic agents are administered intravenously and allowed to spread throughout the entire body before eliciting their effects on the quickly reproducing cells of the body. By including the pro-drug converting enzymes within the Clostridia the non-toxic pro-drug can be administered in higher concentrations, as the toxic form will only be present where the bacteria are expressing the enzymes required for its conversion.

Combination therapy has shown significant potential for development and many groups are now working on bringing this approach to cancer treatment to the front lines.

It remains to be seen whether anaerobic bacterial treatment of cancer will prove to be as effective as is hoped due to the limitations of oxygen concentrations. In the end the development of combination therapies may result in novel approaches that result in chemotherapeutic agents that fight the tumour from the outside in while the bacteria kill the tumour from the inside out then sporulate upon inhibitory oxygen concentrations.


Bergers, G. & Benjamin, L. E. (2003). Tumorigenesis and the angiogenic switch. Nat Rev Cancer 3, 401-410.

Dang, L. H., Bettegowda, C., Huso, D. L., Kinzler, K. W. & Vogelstein, B. (2001). Combination bacteriolytic therapy for the treatment of experimental tumors. Proc Natl Acad Sci U S A 98, 15155-15160.

Hanahan, D. & Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57-70.

Liu, S. C., Minton, N. P., Giaccia, A. J. & Brown, J. M. (2002). Anticancer efficacy of systemically delivered anaerobic bacteria as gene therapy vectors targeting tumor hypoxia/necrosis. Gene Ther 9, 291-296.

Pan, Z. K., Ikonomidis, G., Pardoll, D. & Paterson, Y. (1995). Regression of established tumors in mice mediated by the oral administration of a recombinant Listeria monocytogenes vaccine. Cancer Res 55, 4776-4779.

Pawelek, J. M., Low, K. B. & Bermudes, D. (1997). Tumor-targeted Salmonella as a novel anticancer vector. Cancer Res 57, 4537-4544.

Pawelek, J. M., Low, K. B. & Bermudes, D. (2003). Bacteria as tumour-targeting vectors. Lancet Oncol 4, 548-556.

Toso, J. F., Gill, V. J., Hwu, P. & other authors (2002). Phase I study of the intravenous administration of attenuated Salmonella typhimurium to patients with metastatic melanoma. J Clin Oncol 20, 142-152.

Van Mellaert, L., Barbe, S. & Anne, J. (2006). Clostridium spores as anti-tumour agents. Trends Microbiol 14, 190-196.


Image Credits: Image of Clostridium difficile by Janice Carr, Public Health Image Library. Angiogenesis figure and legend adapted from Bergers, G. & Benjamin, L. E. (2003). Tumorigenesis and the angiogenic switch. Nat Rev Cancer 3, 401-410.


About the Author: James Byrne is in his third year as a PhD candidate in the Discipline of Microbiology and Immunology at the University of Adelaide. His research interest concerns the function of the Polysaccharide Co-Polymerase proteins of S. pneumoniae and their interactions to control the biosynthesis of the capsular polysaccharide that encapsulate this bacterium. Using a combination of mutation screens, protein interaction analysis and in vitro assays James hope to better understand the role of these proteins and in particular the role of CpsC. A better understanding of these systems may allow for the development of novel antimicrobial agents aimed at disrupting the process of exopolysaccharide polymerisation and ligation, a known virulence factor in many organisms. James also writes and maintains the loosely infectious disease themed blog Disease of the Week!