The first human trial of metformin for precancerous colorectal lesions hints that the diabetes drug may be poised for a second incarnation – this time as a chemopreventive agent for several cancers, including lung, colon, and prostate cancers, and even breast malignancies.
The small study, published in the September issue of Cancer Prevention Research, found that after 1 month of oral metformin, nondiabetic patients who already had at least one colorectal adenoma removed experienced a significant reduction in their number of colorectal aberrant crypt foci – an endoscopic finding that some consider a precursor of adenomatous colon polyps.
A second study, published in the same issue of the journal, found that mice who received the drug after exposure to a potent lung carcinogen developed significantly fewer lung tumors than did those who received a placebo.
In light of these data, physicians may want to look closely at metformin when choosing an oral antidiabetic agent for their patients, Dr. Phillip Dennis of the National Cancer Institute said during a press briefing Sept. 1.
“It’s true that we don’t have an indication for metformin as a cancer chemopreventive,” said Dr. Dennis, lead author of the mouse study. “But a clinician who has four different oral agents for diabetes has to make a choice for his patients, and in that process, this early evidence that metformin may have an oncologic benefit will certainly play a role.”
A growing body of data suggests that metformin reduces the risk of several types of cancer, but thus far the only human evidence had come from observational studies of patients with diabetes. Three epidemiologic studies suggest that metformin decreases the incidence of cancer and cancer-related deaths in diabetic patients by 15%-77% (Euro. J. Cancer 2010;46:2369-2380). In vitro and animal studies suggest similar reductions.
The study released today shows the drug’s potential not only to reduce the number of precancerous lesions, but also to decrease the level of proliferating cell nuclear antigen (PCNA), a marker of tumor virulence, wrote Dr. Kunihiro Hosono of the Yokohama City University School of Medicine, Japan, and colleagues: “This first reported trial of metformin for inhibiting colorectal carcinogenesis in humans provides preliminary evidence that metformin suppresses colonic epithelial proliferation and rectal aberrant crypt foci, suggesting its promise for the chemoprevention of colorectal cancer” (Cancer Prev. Res. 2010;3:1077-83).
The murine study examined metformin’s potential to prevent lung cancers. Mice that had been exposed to tobacco carcinogens and consumed or were injected with metformin had up to 72% fewer lung tumors than did placebo-treated mice, reported Dr. Dennis and his colleagues (Cancer Prev. Res. 2010;3:1066-76).
Although the new data are exciting, metformin is not ready to grab the spotlight, Dr. Ernest Hawk cautioned in an interview.
“The story is quite compelling, suggesting that metformin may have a role in therapy or in maintenance treatment for cancer survivors,” said Dr. Hawk, division head of Cancer Prevention and Population Sciences at the University of Texas M.D. Anderson Cancer Center, Houston. “Before we make any assumptions, though, we need many more trials to figure out if it will live up to the potential it seems to have.”
The Japanese study included 23 nondiabetic patients who had an average of five adenomas removed during a screening colonoscopy. Nine of the patients were treated with 250 mg/day of metformin for 1 month; the remaining 14 did not receive any treatment.
At baseline, the 23 patients had a mean of eight aberrant cryptal foci (ACF). “ACF are tiny lesions that develop in the earliest stage of colorectal carcinogenesis and consist of large, thick crypts,” Dr. Hosono and his team wrote. “ACF have been shown to be precursor lesions of the adenoma-carcinoma sequence in humans, and have been suggested as a suitable surrogate end point for colorectal chemoprevention.”
After the treatment period, patients were examined again; ACF were detected by methylene blue staining of rectal mucosa. The number of lesions decreased significantly in the active group, from a mean of nine per patient to six (P = .007). There was no change in the untreated group (mean of seven at each exam). The mean number of ACF considered dysplastic also decreased significantly in the treated group, while again, there was no change in the untreated group. The drug also proved safe, with no patient experiencing hypoglycemia.
The researchers also examined rectal epithelial proliferation by proliferative cell nuclear antigen (PCNA) immunoassay. The PCNA index decreased significantly in the treated group but remained steady in the untreated group. However, the drug did not affect the percentage of cells undergoing apoptosis.
Dr. Hawk, who is also the T. Boone Pickens Distinguished Chair for Early Prevention of Cancer at M.D. Anderson, cautioned against overinterpreting the results. The short duration and small study size should temper enthusiasm somewhat – as should the debate about whether ACF are indeed precursors of colorectal cancer and whether affecting them in any way eventually affects cancer risk.
“Frankly, the Japanese are very good at these ACF studies. And while they have been very reproducible in animal models, human results have been much more difficult to confirm. A number of American investigators have tried to replicate earlier Japanese ACF studies using aspirin, a known colorectal cancer chemopreventive agent, and have not been able to get the same results.”
It’s unclear whether the difference lies in the treatment time, study design, or even the basic differences between a homogeneous Japanese population and the typical heterogeneous American study group, Dr. Hawk said. “Whatever it is, whether the results can be reproduced is the most important question.”
While larger studies are necessary to discover the unknowns of metformin, he said, the drug already has a lot going for it terms of the known. “We have so much data on metformin, and it has such a strong track record of safety, that I believe it’s an agent worth exploring.”
A review co-released with the studies suggested that metformin may be a more effective chemopreventive agent in patients with hyperinsulinemia. “There is evidence indicating that the growth of untransformed epithelial cells and a subset of cancers is stimulated by insulin, and that the systemic insulin-lowering action of metformin might inhibit the proliferation of these cancers,” wrote Dr. Michael Pollak of McGill University and Jewish General Hospital, Montreal. Cancers stimulated by insulin include endometrial, cervical, breast, and prostate cancers, as well as lung and colorectal malignancies. “This action of metformin shares mechanistic features with the manner by which caloric restriction inhibits the growth of certain cancers,” he said (Cancer Prev. Res. 2010;3:1060-5).
If Dr. Pollak’s theory is correct, Dr. Hawk said, metformin or other biguanides may help stem the flood of obesity- and diabetes-related cancers that researchers have predicted for several years. A recent study concluded that obesity could soon become the leading cause of cancer in women – possibly overtaking smoking as the leading cause (Int. J. Cancer 2010;126:692-702).
“The rise of diabetes-related cancer speaks to the importance of pursuing this investigation vigorously,” Dr. Hawk said. “We know that this is a problem of enormous impact for our future.”