Androgen deprivation therapy for prostate cancer and risk of CAD, diabetes: A causal or casual link?

San Francisco-In the past few years, several studies have pointed to an association between androgen deprivation therapy (ADT) for prostate cancer and risk of diabetes and coronary artery disease (CAD). Enough evidence has linked ADT to cardiovascular events and diabetes to warrant a recent science advisory compiled by AUA, the American Heart Association, American Cancer Society, and American Society for Radiation Oncology (CA Feb. 2, 2010; Circulation Feb. 1, 2010).

"The case for biologic plausibility of treatment-related diabetes is relatively straightforward," Dr. Smith said during a presentation at the 2010 Genitourinary Cancers Symposium.

"ADT appears causally associated with diabetes," said Dr. Smith, who was the senior author on a recent paper examining ADT's association with cardiovascular disease and diabetes (J Natl Cancer Inst 2010; 102:39-46). "The association is statistically strong, but the effect size is small, with the number needed to harm over 100."

The case for biologic plausibility of CAD is less clear. It is well known that ADT, either by orchiectomy or by gonadotropin-releasing hormone (GnRH) agonist therapy, affects serum lipids. Total cholesterol increases by about 10%, low-density lipoprotein cholesterol and high-density lipoprotein cholesterol levels increase by 5% to 10%, and triglycerides increase by about 20% to 30%. ADT does not change the ratio of LDL and HDL cholesterol and does not affect other risk factors for cardiovascular disease, including blood pressure, or C-reactive protein and other inflammatory biomarkers.

What those changes mean in the plausible risk for myocardial infarction is very little, Dr. Smith noted. Recently published data show that, after adjusting for other risk factors, triglycerides have no independent predictive value for CAD, whereas there is a consistent relationship between higher LDL and lower HDL and increased CAD risk.

Data on ADT and CAD risk

Data from the Surveillance, Epidemiology and End Results Medicare database support Dr. Smith's conclusions on both diabetes and CAD. Based on data collected on 73,196 men age 66 years or older who had locoregional prostate cancer between 1992 and 1999 and were followed until 2001, GnRH agonist use was associated with increased risk of diabetes (adjusted hazard ratio, 1.44) and myocardial infarction (adjusted HR, 1.11) (J Clin Oncol 2006; 24:4448-56). Men treated with orchiectomy were more likely to develop diabetes (adjusted HR, 1.34) but not coronary heart disease, myocardial infarction, or sudden cardiac death.

Another way to consider the magnitude of risk is to examine the number needed to harm (NNH), Dr. Smith said. In this study, the NHH for diabetes is 123 for GnRH agonist therapy and 278 for orchiectomy. For myocardial infarction, the NHH is 384 for GnRH agonists and infinite for orchiectomy.

For the sake of comparison, Dr. Smith showed 5-year needed-to-treat numbers for primary prevention of coronary artery disease. In the case of statins, the number needed to treat is between 40 and 70 patients; for antihypertensives, 80 to 160 patients; and for aspirin, more than 300 patients.

"I'll let you do the calculations for yourself to compare the relative magnitude of potential harm by either introducing ADT or failing to introduce one of these known preventative therapies," Dr. Smith said.

"Taken together, these data suggest that androgen deprivation therapy appears causally related to diabetes," Dr. Smith concluded. "It's far less clear whether the relationship between ADT and coronary artery disease is causal or casual. At this time, I do not believe that the potential risk of cardiovascular disease should impact decisions about the use of androgen deprivation therapy in evidence-based settings."

Dr. Smith has disclosed that he is a consultant/adviser to Amgen, Cougar Biotechnology, GTx Inc., and Novartis.