Patients with low levels of prediagnostic adiponectin were approximately 40% more likely to develop pancreatic cancer than those with higher adiponectin levels, according to results of a nested case-control study.
Researchers conducted the investigation to further understand how the mechanics of insulin resistance and chronic inflammation affect the development of pancreatic cancer.
The researchers collected prediagnostic levels of adiponectin from 468 patients with pancreatic cancer and 1,080 matched controls from five prospective cohorts in the United States.
Median plasma adiponectin was 6.2 mcg/mL in case patients and 6.8 mcg/mL in controls (P=.009).
The researchers observed an inverse association between plasma adiponectin and pancreatic cancer risk. The association was consistent across the five cohorts (P=.49) and independent of other markers of insulin resistance such as diabetes, BMI, physical activity or plasma C-peptide.
Multivariate analysis results indicated that, compared with those in the lowest quintile of adiponectin, the OR for pancreatic cancer risk was 0.61 (95% CI, 0.43-0.86) for quintile two, 0.58 (95% CI, 0.41-0.84) for quintile three, 0.59 (95% CI, 0.4-0.87) for quintile four and 0.66 (95% CI, 0.44-0.97) for quintile five (P=.04).
A restricted cubic spline regression analysis validated the nonlinear link (P<.01).
Factors such as sex, smoking, BMI, physical activity or C-peptide did not modify the link (all P>.10).
“Early detection by the assessment of adiponectin has the potential to improve the survival rates of pancreatic tumor patients,” Jianliang Zhang, PhD, associate professor of oncology at Roswell Park Cancer Institute, and Steven N. Hochwald, MD, FACS, of the department of surgical oncology at Roswell Park, wrote in an accompanying editorial. “It is also inviting to speculate that the therapeutic interventions to increase the levels of circulating adiponectin may prevent the development of pancreatic cancer and/or improve the survival of patients with malignancy.”
For more information:
Bao Y. J Natl Canc Inst. 2012;doi:10.1093/jnci/djs474.
Zhang J. J Natl Canc Inst. 2012;doi:10.1093/jnci/djs522.
Disclosure: Hochwald and Zhang report no disclosures.