ORLANDO – Metformin may be associated with a reduced risk of colorectal cancer in patients with type 2 diabetes, according to a meta-analysis that included mostly observational studies.
Sulfonylurea and insulin were not associated with a reduced risk and showed a nonsignificant trend toward a higher risk for colorectal cancer (CRC). No association was noted between CRC and thiazolidinedione (TZD), Dr. Siddharth Singh reported at the annual Digestive Disease Week.
Dr. Siddharth Singh
CRC incidence was reduced by 10% among patients who were on metformin (n = 10 studies; adjusted odds ratio [AOR], 0.90; 95% confidence interval, 0.82-0.99; P = .02). The results were stable when the analysis was restricted to high-quality observational studies (n = 7 studies; AOR, 0.88; 95% CI, 0.79-0.98).
CRC risk was higher in patients on a sulfonylurea (n = 8 studies; AOR 1.12; 95% CI 0.99-1.26; P = .07) or on insulin (n = 10 studies; AOR, 1.25; 95% CI, 0.91-1.71; P = 0.17). Neither result was statistically significant, however.
Dr. Singh said that the findings warrant prospective cohort studies or randomized controlled trials of the chemopreventive effect of metformin in adults with type 2 diabetes.
Diabetes is an established risk factor for colorectal cancer, and preclinical and observational studies have shown that antidiabetes drugs may affect CRC risk. The relative affect of various antidiabetes drugs on CRC risk is unclear, said Dr. Singh of the Mayo Clinic in Rochester, Minn.
Dr. Singh and his colleagues conducted a systematic review and meta-analysis of 15 studies evaluating the effects of metformin, TZDs, sulfonylureas, and insulin on colorectal cancer risk. There was considerable heterogeneity across the studies, which may be explained by study design, location, and whether the study accounted for the effects of other antidiabetic drugs, Dr. Singh said.
Five case-control studies, eight cohort studies, and two randomized controlled trials with a total of nearly 14,000 CRC cases in 841,000 diabetes patients were included in the meta-analysis.
TZD was not associated with CRC risk (n = 6 studies; AOR, 0.95; 95% CI, 0.87-1.03; P = .24).
One of the main limitations of the unpublished study was that the significant results were seen only in observational studies and not in post-hoc analysis of the randomized controlled trials. Also, there are residual confounding factors, including failure to adjust for obesity, and confounding by indication and severity of diabetes, Dr. Singh said.
"Before we take part in any proactive treatment regimen, we have to have well-designed, randomized controlled trials," said Dr. N. Jewel Samadder, assistant professor of medicine at the University of Utah, Salt Lake City. "However, the observational studies, and the meta-analysis that [follow], are certainly hypothesis driving. Diabetes itself is a risk factor [for CRC], and some of the medications may alter that risk," said Dr. Samadder, who was not involved in the study.
Dr. Singh had no disclosures. Dr. Samadder has received speaking and teaching honoraria from Cook.
On Twitter @NaseemSMiller