Differential Effect of Hyperglycemia on the Odds of Cancer Among the Adult Population of the United States.

ABSTRACT

Objective Accumulating evidence indicates a relationship between diabetes and cancer risk, with obesity, insulin resistance, and hyperglycemia being implicated as the major underlying pathogenetic mechanisms of increased cancer risk among people with diabetes. We aim to assess the differential effect of dysglycemia (prediabetes and diabetes) on the strength of association (odds) of cancer amongst the adult US diabetic population.  Material and methods We analyzed data from the 1997-2013 National Health Interview Survey (NHIS) dataset, which applies a multistage area probability sampling design. We used descriptive statistics and logistic regression analyses to test the strengths of the association between diabetes, prediabetes, and cancer before and after adjusting for major risk factors for cancer, including age and body mass index (BMI). Results A total of 722,532 individuals were surveyed, with a mean age of 47.18 ±0.3 years (±SEM) and a BMI of 26.9 ±0.01 kg/m2. Between 1997 and 2013, BMI increased from 26.0 to 27.4 kg/m2, the diabetes rate increased from 4.1% to 7.6%, and associated cancer rates increased from 6.6% to 9.0%. Body mass index was 27.1 vs. 26.8 kg/m2, P < 0.01, for those with and without cancer, respectively. The unadjusted odds ratio for cancer was 1.92 (1.78-2.08) (95% CI) and 2.20 (2.13-2.27) for prediabetes and diabetes, respectively. After adjusting for age, BMI, race, and cigarette smoking, the odds ratio for cancer was 1.12 (1.03-1.22), P < 0.01, and 1.15 (1.11-1.18), P <0.01, for prediabetes and diabetes, respectively.  Conclusion Among US adults, the increasing rate of diabetes over the years was associated with an increased rate of cancer. Diabetes and prediabetes have a graduated effect on cancer risk. While obesity is generally implicated as an underlying pathophysiologic link between diabetes and cancer, our study showed a modest difference in BMI between those with and without cancer. In addition, the effect of diabetes and prediabetes on the odds of cancer persisted after adjusting for BMI. These data collectively suggest that hyperglycemia is an attractive pathophysiologic mechanism that may play a role in increasing the odds of cancer among diabetic and prediabetic populations. Our study is consistent with the accumulating evidence implicating hyperglycemia in the pathogenesis of cancer, where glucose is used in PET scanning to detect cancer (the Warburg effect), and the ketogenic diet appears to be useful in cancer management, enhancing the effect of chemotherapeutic agents.