Negative Association Between the Use of TZD's and the Risk of Cancer

Previous | Next

This article originally posted 29 October, 2009 and appeared in Issue 493

The ability of thiazolidinediones such as rosiglitazone to lower blood sugar levels in Type 2 diabetes is an indisputable benefit. But, the benefit has been tempered by the inconsistent evidence regarding the use of thiazolidinediones and the risk of cancer.

While a reduced risk of breast, lung, prostate, and colon cancer in diabetes patients receiving thiazolidinediones has been reported^1,2 so have conflicting results.³

Now, a study that utilized the Hong Kong Diabetes Registry containing information on over 6,000 Chinese patients with Type 2 diabetes has reported a negative association between the use of rosiglitazone, metformin, and gliclazide, and the risk of cancer.

Alice Kong, The Chinese University of Hong Kong, Hong Kong, China, presented the study at the 20th World Diabetes Congress of the International Diabetes Federation (IDF).

The Hong Kong database has been collecting patient information since 1995. Specifically, patient data from the Prince of Wales Hospital in Hong Kong was assessed based on the European DiabCare protocol. A cutoff date of July 30, 2005, was imposed and data concerning all diabetic patients from 1995 to 2005 in the central database of the Hong Kong Hospital Authority and Hong Kong Death Registry was obtained.

Of the 7,920 identified patients, 1,817 patients were excluded due to Type 1 diabetes, non-Chinese ethnicity, history of cancer, or missing data. The resulting 6,103 patients (median age, 57 years; 54% male) comprised the study population. The patients had been diagnosed with diabetes for a median of 6 years (range, 2-11 years).

At a median follow-up of 4.91 years (range, 2.81-6.98 years), 271 participants (4.44%) had developed cancer, representing an incidence rate of 9.21 (95% confidence interval [CI], 8.12-10.30) per 1,000 person-years.

Comparison of the data from these patients with those who had not developed cancer revealed that cancer development tended to occur in older patients, in those who smoked and drank alcohol, and in those who had been diagnosed with diabetes for a longer time.

Biochemically, cancer development was more evident in those with higher systolic blood pressure, higher urine albumin/creatinine ratio, lower estimated glomerular filtration rate and low-density lipoprotein cholesterol (<2.8 mmol/L), and albuminuria and low-density lipoprotein cholesterol exceeding 3.8 mmol/L.

Cancer development was also linked to lessened use of gliclazide, metformin, rosiglitazone, statins, and angiotensin blockers.

After adjusting the data for age, gender, body mass index, tobacco, and alcohol consumption, use of gliclazide, metformin, and rosiglitazone was found to be associated with a reduced risk of cancer (hazard ratio of 0.73, 0.58, and 0.16, respectively; P < .0168, P < .0001, and P = .0091, respectively).

The data will help clarify the controversy regarding the association of thiazolidinedione use and cancer risk in Type 2 diabetics, adding weight to the view that thiazolidinediones do not increase the risk of cancer in diabetics, said the researchers.

Oct. 20^th, at the IDF conference[Presentation title: Associations Between the Use of Oral Anti-Diabetic Drugs and the Risk of Cancer in Type 2 Diabetes Patients. Abstract D-0771

Koro C et al. Pharmacoepidemiol Drug Saf. 2007;16:485-492.
Govindarajan R et al. J Clin Oncol. 2007;25:1476-1481.
Ramos-Nino ME et al. BMC Med. 2007;5:17.