BMI, long-term changes in BMI, and risk of cancer mortality in a large cohort study

Reporter: Gita Suneja, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 4, 2012

Presenter: Niloofar Taghizadeh
Presenter's Affiliation: Department of Epidemiology, University Medical Center Groningen, Groningen, Netherlands


  • Many studies have found an association between body mass index (BMI) and the risk of development of cancer.
  • The proposed mechanisms include hormonal systems involving insulin-like growth factor, leptin, sex steroids, and adipokines.
  • Existing studies:
    • Dispute whether this association differs between males and females.
    • Have inconsistent results for mortality risks associated with BMI and different types of cancers
    • Do not address the effect of changes in BMI on risk
  • The objective of this study was to measure associations of BMI, and changes in BMI, with the most common types of cancer, in a large, population-based cohort.

Materials and Methods

  • The authors used a large, general population-based database in Vlagtwedde-Vlaardingen to extract 8,465 cases from 1965-1990.
  • BMI was assessed at the first survey and last survey, as well as annual time points in the study period. Each subject had 1 to 8 surveys.
  • BMI was associated with the risk of mortality from the most common types of cancer (lung, colorectal (CRC), breast and prostate cancer) with follow-up on mortality status until 2009.
  • The association was assessed based on tertiles of the annual change in BMI (defined as the difference between BMI at last survey and first survey divided by the time between last and first survey).
  • BMI was categorized into three groups: < 25 kg/m2, 25-30 kg/m2, and ≥ 30 kg/m2.
  • Changes in BMI were also categorized: < 0.02 kg/m2/yr, 0.02-0.2 kg/m2/yr, and > 0.2 kg/m2/yr for the analyses.
  • The multivariate Cox regression model was adjusted for age, smoking, and gender.
  • The primary outcomes was cancer mortality.


  • Among all 8,645 subjects, 1194 died due to cancer (275 from lung cancer, 134 from colorectal cancer, 117 from breast cancer, and 83 from prostate cancer).
  • Mortality from all types of cancer was significantly increased in subjects with BMI > 30 kg/m2 (HR 1.22, 95% CI 1.00-1.48). This effect was more pronounced in females (HR 1.38, 95% CI 1.06-1.81) and in never smokers (HR 1.39, 95% CI 1.02-1.90).
  • Prostate cancer mortality was significantly increased in males with BMI 25-30 kg/m2 (HR 2.04, 95% CI 1.90-3.83) and > 30 kg/m2 (HR 2.61, 95% CI 1.02-6.67).
  • This association between prostate cancer mortality and BMI was higher in smokers.
  • Lung cancer mortality risk was decreased in subjects with BMI 25-30 kg/m2 (HR 0.71, 95% CI 0.54-0.93) and > 30 kg/m2 (HR 0.82, 95% CI 0.50-1.32), and was particularly pronounced in males, in smokers, and in smoking males.
  • There were no significant associations between BMI and colorectal or breast cancer mortality.
  • There were no significant associations between long-term change in BMI and mortality from any types of cancer analyzed. However, short-term change in BMI decreased mortality from all types of cancer, particularly in females.
  • Finally, "chronic obesity" was found to have an association with higher mortality from all types of cancer.

Author's Conclusions

  • Increase in BMI is associated with an increased risk of mortality from all types of cancer in females and an increased mortality risk from prostate cancer in males
  • Increase in BMI was associated with a decreased lung cancer mortality risk, especially in males.
  • More research is needed into the biological mechanisms that link BMI to cancer.

Clinical Implications

  • This study is highly relevant and timely given the large obesity epidemic facing most developed countries.
  • Rates of obesity has increased dramatically in the United States with further projected increases in the near future.
  • Prior studies have provided compelling evidence that obesity is linked to cancer incidence and mortality for cancers of the endometrium, kidney, gallbladder, breast, colon, rectum, pancreas, thyroid, and esophagus. Data on other cancers have been mixed. (Calle, NEJM, 2003).
  • Proposed mechanisms include chronic hyperinsulinemia and insulin resistance, estrogen excess, direct or indirect effects on tumor growth regulators (mTOR), and chronic inflammation, however the underlying biologic mechanisms are not well understood.
  • Most importantly, like tobacco smoking, obesity may represents a potentially avoidable cause of cancer incidence and death, and the attributable fraction of death from obesity is high.
  • A 2007 study using NCI Surveillance, Epidemiology, and End Results (SEER) data, estimated that in the United States, about 34,000 new cases of cancer in men (4%) and 50,500 in women (7%) were due to obesity.
  • A projection of the future health and economic burden of obesity in 2030 estimated that continuation of existing trends in obesity will lead to about 500,000 additional cases of cancer in the United States by 2030.
  • The contributions of this study include investigation of changes in BMI, as well as chronic obesity as a risk factor.
  • Further effort needs to be focused on understanding the processes linking obesity with cancer to develop new public health approaches to prevention and early detection of cancer in the at-risk overweight/obese population.