Prostate Cancer, Diabetes, and Metformin…What Are the Relationships?
The following editorial was written by Dr. Michael Pollak, an expert in the topic area of tumor metabolism, and has been a PCF-funded investigator for nearly 11 years. More recently in his career, Dr. Pollak has been interested in the clinical application of Metformin for the prevention of prostate cancer progression in men with metastatic disease.
By Michael Pollack, MD
McGill University, Montreal, Canada
Metformin is a drug used in the treatment of type 2 diabetes. It is safe, inexpensive, and widely prescribed around the world – more than 25 million people take metformin pills daily. There is increasing interest in the possibility that this agent or derivatives may have relevance to prostate cancer. At this stage, there are no firm conclusions, but rather intriguing clues that are under investigation.
The first set of clues came from pharmacoepidemiology (the science concerned with the benefit and risk of drugs used in populations and analysis of the outcomes of drug therapies): surveys of large diabetic populations showed that those who were prescribed metformin appeared to have reduced risk of many cancers, including prostate cancer. Evidence from this kind of retrospective analysis is not as definitive as those from randomized clinical trials, but the protective effects reported in some studies were large enough that laboratory scientists began to investigate the compound – with considerable initial scepticism. However, instead of debunking the population studies, several studies demonstrated an antiproliferative effect of metformin on prostate cancer cells (antiproliferative pertains to a substance used to prevent the spread of cells).
How could an anti-diabetic drug slow cancer growth or decrease cancer risk? The laboratory work suggests several possible mechanisms. One is an indirect – it involves no interaction of the drug with the prostate cancer cells, but rather an effect on the hormonal balance of the patient that creates a less favorable environment for cancer progression. Many hormone levels may be altered in a favorable direction by metformin, but one of interest is insulin – insulin levels are too high in type 2 diabetes, and are reduced by metformin. Prior studies have shown that higher insulin levels are associated with worse prognosis of prostate cancer, because some prostate cancer cells are stimulated to proliferate by insulin. Furthermore, obesity, which is associated with elevations in insulin, worsens prognosis of prostate cancer. So metformin, acting indirectly, may be of particular value in obese prostate cancer patients.
A point of special interest concerns androgen deprivation therapies, widely used in prostate cancer treatment. These treatments are of proven value, but their benefit is limited, as resistance develops to both medical and surgical castration, and also to newer androgen targeting agents including abiraterone and MDV-3100. It has been shown that while androgen deprivation reduces androgen levels, it tends over time to raise insulin levels, so that some forms of castration resistance may in the long run lead to insulin –stimulated cancer growth, despite continued inhibition of androgen-stimulated growth. If metformin prevents the hyperinsulinaemia associated with androgen deprivation, it may extend the duration of benefit of androgen deprivation therapies (hyperinsulinaemia is a condition in which there are excess levels of circulating insulin in the blood). This possibility is being examined in both laboratory studies and clinical trials.
Metformin has also been shown to have separate direct mechanisms of action. Here, rather than acting to change the patient’s hormone levels, the drug acts directly on prostate cancer cells to cause death and/or inhibit proliferation. This involves an action of the drug that interferes with cellular energy metabolism: metformin interferes with the process by which nutrients are converted to energy : under some circumstances, cells react to the energy deficiency by adjusting their functions to use less energy – but a cancer cell that is minimizing energy consumption cannot exhibit aggressive metastatic or proliferative behaviour. In other cases, the energy stress caused by metformin is sufficient to cause cell death. Novel therapies that target cellular energy supply are of considerable interest to the research community – the surprise here is that a candidate agent turns out to be an ‘old’ molecule used in diabetes treatment.
With all this evidence, one might conclude that men at risk for prostate cancer and men with prostate cancer (that is, all men) should start taking metformin. This would be going beyond the evidence: there are important gaps in knowledge that must be addressed. With respect to the direct actions, it is not certain that sufficient tissue drug levels are achieved with conventional doses of metformin used in diabetes. It is possible that metformin derivatives have better pharmacokinetic properties than metformin, which after all was never designed for applications in oncology. With respect to the indirect mechanisms of action, it appears that the reductions in insulin and other hormonal changes that may be beneficial do not occur in all men, but rather in subsets of men who have abnormally high baseline levels – so there may be a need to select specific subgroups of men who will benefit. Finally, there are complexities in the interactions between diabetes and prostate cancer and also in cross-talk between insulin and androgen signalling systems. Men with type II diabetes actually have slightly lower prostate cancer risk than the general population, an observation which may relate to lower androgen activity in diabetes. Nevertheless, in men with prostate cancer, high insulin levels are associated with worse prognosis, and a recent study provides evidence that insulin actually can stimulate the local production of androgens by prostate cancer cells – a process blocked in laboratory studies by metformin.
While this area has been of increasing interest to academic investigators in recent years, the pharmaceutical industry has been reluctant to invest in this area of research as metformin is a generic drug. However, this may change with the possibility that metformin derivatives may not only have superior antineoplastic activity as compared to the metformin, but also be patentable. There also is interest in combining metformin with approved agents for prostate cancer treatment, to increase response duration. So while metformin use for prostate cancer treatment or prevention cannot be recommended at this time, there is a consensus that research in this area should be prioritized.