Reference
Bo S, Ciccone G, Rosato R, et al. Cancer mortality reduction and metformin. A retrospective cohort study in type 2 diabetic patients. Diabetes Obes Metab. 2012 Jan;14(1):23-29.
Design
Health data were compared looking for association between metformin use and risk of cancer-related or overall risk of death.
Participants
Starting in 1995, the University Hospital Clinic in Turin, Italy collected data on 3,685 patients treated for type-2 diabetes for comparison with the general population of the city. The patients were followed for 4.5 years.
Study Medication and Dosage
Data from the patients were analyzed by type of medication the patients received to treat diabetes. Cox regression analysis was performed, dividing the patients into subgroups based on whether they were treated with metformin, insulin, or sulfonylureas.
Outcome Measures
All-cause mortality rates and cancer-related mortality rates.
Key Findings
Drug Used
| Metformin | Sulfonylureas | Insulin | |
| All-cause death risk | 9.2% | 13.1% | 26.8% |
| Cancer-related death | 1.6% | 3.0% | 4.8% |
Patients using metformin had the lowest risk of dying from all causes, including from cancer.
Metformin users showed a significantly lower cancer mortality risk (HR=0.56; 95% CI: 0.34–0.94). Each 5 years of metformin exposure was associated with a reduction in cancer death by 0.73. Every 5 years of insulin exposure was associated with 1.25-fold increase in other-than-cancer death.
Practice Implications
Obviously if there is a choice in treatment for type-2 diabetics, metformin seems to be preferable. Patients treated with metformin are less likely to die for any reason, including cancer. Of course, one could easily argue that the choice to use metformin in this population was simply a measure of disease severity and that those who were prescribed sulfonylureas or insulin had more advanced diabetes.
The data on metformin's lowering risk of dying from cancer-related causes is of particular interest. Over the last few years a series of papers, including a recent one on metformin and ovarian cancer survival, have suggested that metformin might be of benefit in cancer. Epidemiologic data and clinical evidence suggest that metformin, because it lowers circulating insulin, could be important in treating cancer associated with hyperinsulinemia such as breast cancer and colon cancer. Independent of this effect, metformin may inhibit cancer cell growth by inhibiting mammalian target of rapamycin (mTOR) signaling and protein synthesis.
This paper seen in context with a series of closely related studies bolsters the idea that we should consider using metformin with cancer patients—certainly those with diabetes, even if it is mild.
Why are we discussing prescription diabetes drug in Natural Medicine Journal?
Though we think of metformin as a prescription drug, we might also think of it as an herbal derivative. Metformin, technically a biguanide, was originally derived from the herb Galega officinalis (usually known as French lilac or Goat’s Rue). Use of this herb to treat diabetic symptoms, in particular polyuria and halitosis, dates back to ancient Egypt and the Middle Ages.1
Biguanides were identified as the active compound in French lilac in the 1920s and several drugs were developed for use by diabetics. Two, phrenformin and buformin, are no longer sold because of toxicity. Metformin, the third biguanide, remains a common prescription, without significant toxicity; 120 million prescriptions are written for it each year.2
While metformin has been used in Britain since 1958 and in Canada since 1972, it has been used in the United States only since 1995. Metformin is also prescribed to treat polycystic ovary syndrome (PCOS) and may have an impact on ovarian cancer.3
That metformin might mitigate cancer risk was first noticed in data from retrospective epidemiologic studies of diabetic patients. Evans et al in 2005, writing in the British Medical Journal, were among the first to hypothesize an anticancer action. Metformin works by targeting the enzyme AMPK (AMP activated protein kinase), which induces muscles to take up glucose from the blood. At the time of their writing, a protein kinase known as LKB1 had recently been identified as an upstream regulator of AMPK. LKB1 is required to activate AMPK. Either metformin or exercise can activate it. LKB1 is a tumor suppressor. Thus Evans et al thought metformin might also reduce cancer risk. They examined British data from 314,127 people collected from 1993 to 2001; 11,876 were newly diagnosed with type-2 diabetes during that period.
Random controls were selected from the general population, people without cancer but matched for age, sex, and year of diagnosis. Odds ratios were adjusted for confounders such as smoking and obesity. People with type-2 diabetes who had been prescribed metformin had a lower risk of getting cancer, which decreased with the number of prescriptions they had received. For example, those who had received 1–11 prescriptions had an adjusted odds ratio of being diagnosed with cancer of 0.82, while those who had received 12–31 prescriptions had an even lower odds ratio of 0.75.4
These researchers wrote a follow up study that compared 4,085 diabetics who used metformin with and an equal number of diabetics who did not. Of the metformin users, 7.3% were diagnosed with cancer during the period of the study, compared to 11.6% of the non-metformin users. Cancer was diagnosed sooner in the non-users—2.6 years into the study, compared to 3.5 years in the metformin users. After adjusting for confounders such as sex, age, smoking, etc., using metformin reduced risk of a cancer diagnosis by 37% (HR=0.63; [0.53–0.75]).5
A comprehensive meta-analysis by Decensi et al was published in November 2010 and concluded that taking metformin decreased risk for a number of cancers. Data from 11 studies were combined to yield information on 4,043 cancer events and 529 cancer deaths. Those taking metformin rather than other antidiabetic drugs had a 31% reduction in risk (0.69; 95% CI: 0.61–0.79).6
In July 2009, Jiralerspong et al from MD Anderson reported that diabetic women taking metformin responded better to cancer treatment than diabetic women who were not. They studied 2,529 women with breast cancer and compared their response rates to treatment. They tracked pathologic complete response rates (pCRs) that are considered predictive of overall survival. Those diabetic women receiving metformin while receiving adjuvant treatment for breast cancer had significantly higher pCRs (24%) compared to diabetic patients not receiving metformin (pCR 8%) and non-diabetic patients not receiving metformin (pCR 16%).
Despite this increase in pCR, metformin did not significantly improve the estimated 3-year relapse-free survival rate in this study.7
A 2010 study by Patel et al of men with prostate cancer did not find that metformin use brought significant benefit. They tracked 612 men with prostate cancer, of whom 112 were diabetic metformin users, 98 were diabetic non-metformin users, and the rest did not have diabetes. All the men were treated with radical prostatectomy. Biochemical recurrences (BCR) were tracked after surgery. Five-year BCR-free survival was 75.0% for non-diabetic men, compared with 66.1% for metformin users and 59.3% for diabetic non-metformin users (P=0.004). Metformin use brought a modest but not statistically different reduction in risk of recurrence (HR=0.94; 95% CI: 0.6–1.5; P=0.817).8
Cell culture studies suggest that metformin can inhibit growth through a variety of different mechanisms and in a variety of different cancer types, including breast, prostate, colon, endometrial, ovarian, and glioma.9–16
A paper presented at the June 2010 ASCO meeting suggests that the changes from taking metformin in breast cancer patients are almost immediate.
A modest number of clinical trials have been completed using metformin in non-diabetics evaluating its effect on cancer.
In September 2010, Hosono et al reported that after just a month of taking small 250 mg/day doses of metformin, a dozen patients with “aberrant crypt foci (ACF), … an endoscopic surrogate marker of colorectal cancer … had a significant decrease in the mean number of ACF” compared to a control group. These early results suggest that “metformin suppresses colonic epithelial proliferation and rectal ACF formation in humans, suggesting its promise for the chemoprevention of colorectal cancer.”
A paper presented at the June 2010 ASCO meeting indicates that the changes from taking metformin in breast cancer patients are almost immediate. Hadad et al enrolled 47 non-diabetic women with breast tumors >1 cm. The women were randomized after their initial biopsies had been taken. Some of the women received metformin and some did not. Second tissue biopsies were taken a few weeks later at the time of surgery and the first and second samples compared. Taking metformin down regulated 4 specific genes in these women: PDE3B, SSR3, TP53, and CCDC14. PDE3B is a critical regulator of cAMP that affects the activation/phosphorylation of AMPK.17
Seen in the light of these other recent studies, the recent paper by Bo et al featured in this discussion is just one more piece of evidence suggesting a potential use for metformin in treating cancer.
Which patients might best respond to metformin treatment? Though considerable research still needs to be done, Ryan Dowling and colleagues from Toronto in an April 2011 review state “that patients exhibiting hyperinsulinemia and tumors expressing the insulin receptor, LKB1, and TSC2 would benefit most from metformin therapy, while patients with normal circulating insulin levels and tumors lacking expression of the insulin receptor, LKB1, and TSC2 would likely be unresponsive to the drug.” They point out that there are “a number of clinical trials examining the use of metformin as a cancer therapy … underway including studies in prostate, breast, endometrial and pancreatic cancer patients.”18
Perhaps one of the most ambitious of the trials is one being conducted by The National Cancer Institute of Canada Clinical Trials Group (NCIC CTG). They have started a large phase III clinical trial looking at the effect of metformin versus placebo in over 3,500 women with early stage breast cancer.19
Taken together, this information should encourage us when evaluating cancer patients to assess their blood sugar levels with great care and if there is any hint of elevation, to encourage their use of metformin. We should follow the research on this topic closely, watching for new developments in the near future.
For more research involving integrative oncology, click here.
