Metabolic Syndrome and Prostate Cancer Risk in a Population-Based Case–Control Study in Montreal, Canada

Audrey Blanc-Lapierre; Andrea Spence; Pierre I. Karakiewicz; Armen Aprikian; Fred Saad; Marie-Élise Parent


BMC Public Health. 2015;15(913) 

In This Article


Study Population

The Prostate Cancer & Environment Study (PROtEuS), described previously,[24–27] is a population-based case–control study conducted in Montreal, Canada, to assess the role of environmental factors in PCa risk. Eligible subjects were men, younger than 76 years of age at the time of diagnosis or selection, residents of the greater Montreal area, registered on Quebec's permanent electoral list (continually updated) and Canadian citizens.

Cases were all patients newly diagnosed with primary histologically confirmed PCa, actively ascertained through pathology departments across seven French hospitals in the Montreal area between 2005 and 2009. This covered over 80 % of all PCa cases diagnosed in the region of Montreal during the study period according to registry information. Concurrent to case recruitment, controls were randomly selected from the electoral list of French-speaking men residing in the same districts as cases and frequency-matched to cases in 5-year age groups.

Study participants represented 79.4 % of eligible cases and 55.5 % of eligible controls. This study was approved by the Ethics Committees of the following institutions: Institut national de la recherche scientifique, Centre de Recherche du Centre Hospitalier de l'Université de Montréal, Hôpital Maisonneuve-Rosemont, Hôpital Jean-Talon, Hôpital Fleury, and Hôpital Charles-LeMoyne. All participants provided written informed consent.

Data Collection

MetS diagnosis was assigned in a similar fashion among cases and controls, based on the information from the questionnaire. During face-to-face interviews, subjects provided the following information: socio-demographic and anthropometric characteristics (including height and weight at different ages), family history of cancer, and PCa screening history. They were also asked to report any medical condition lasting at least 6 months, including diabetes, hypertension and benign prostate hyperplasia (BPH), and medications taken to treat them, with ages at beginning and end, and duration. Information was collected about lifestyle factors such as physical activity at home, work and leisure, smoking habits, alcohol consumption and dietary habits. Hip and waist circumferences (WC) were measured by the interviewer. The degree of aggressiveness of PCa, defined by the Gleason score, was extracted from prostate biopsy pathology reports and the pre-biopsy prostate specific antigen (PSA) level at diagnosis from patient files.

MetS Definition

We assessed the presence of a MetS history according to definitions from three organizations: the Adult Treatment Panel III from the National Cholesterol Education Program (NCEP-ATPIII) revised by the American Heart Association/National Heart, Lung and Blood Institute, the World Health Organization (WHO) and the International Diabetes Federation (IDF)[6] (Table 1). The prevalence of MetS components (diabetes, hypertension, dyslipidemia and abdominal obesity) was estimated at 2 years before the index date (diagnosis for cases/interview for controls) or at interview for WC-based obesity, and at different ages (20, 40, 50 and 60 years). The timing of MetS onset was based on the first concomitant presence of three individual components. Overall, 1.1 % of cases and 0.8 % of controls whose information about metabolic disorders was not sufficient to conclude about MetS presence were excluded from analyses.

Blood pressure, fasting glucose, TG and HDL-C blood levels were not available from the study. We used therefore medical histories of hypertension, diabetes and dyslipidemia (hypercholesterolemia, hypertriglyceridemia, or intake of the following lipid lowering drugs: statins, niacin, fibrates, resins or ezetimib). Diabetic subjects receiving insulin therapy since diabetes onset, who never took oral anti-diabetic drugs, were considered not to have T2D. Central obesity was defined on the basis of BMI or WC according to the MetS definitions considered (Table 1). As the appropriate WC cutoff for abdominal obesity may vary by ethnic origin, we also used the waist-to-height ratio (i.e., the Index of Central Obesity: ICO) which has been shown to be more effective in assessing abdominal obesity across ethnicities with a simple cutoff.[28] In sensitivity analyses, we used the NCEP-ATPIII MetS definition which corresponds to the most recent harmonized definition and includes the WC cutoff of 102 cm for abdominal obesity recommended by Health Canada.[7]

Statistical Analyses

Unconditional logistic regression was used to determine the risk of PCa associated with MetS. We also assessed the risk of PCa according to age at MetS first onset (≤40, 41–50, 51–60 and > 60 yrs), and according to the number of MetS components.

Risks of low-grade (Gleason scores <7 or 3 + 4) and high-grade (Gleason scores >7 or 4 + 3) PCa[29] were estimated in polynomial logistic models, and their respective regression coefficients were compared using a Wald test. All regression models were systematically adjusted for age (continuous), ancestry (European/Sub-Saharan African/Asian/Greater Middle East/Other/Don't know), first-degree family history of PCa (Yes/No/Do not know) and recent PSA or Digital Rectal Examination (DRE) screening (No/≤2 years/>2 years/Do not know). The other covariates tested were family income (<$C30 000/$C30 000–79 999/$C80 000 and more/Preferred not to respond/Do not know), education (Primary or secondary/College or university), BPH (Yes/No), ever use of aspirin or 5-alpha-reductase inhibitors (Yes/No), smoking (cigarette pack-years), alcohol consumption (drink-years), physical activity (Very/Moderately/Not very active), dietary habits (annual frequency of fruits and vegetables intake), and changes in fat or sweets intake in the last 20 years (More/Less/No change). Variables retained in the final regression model were those which, when excluded, increased the Akaike Information Criterion by at least 5.

Sensitivity Analyses

The association between MetS and PCa was examined separately in two age groups (<65 vs. ≥65 year-old at index date) to investigate the potential competing risk represented by cardiovascular causes of death, which would be expected to be less common in the younger age group.

An analysis was performed restricting subjects to those screened for PCa (PSA or DRE) within two years of the index date, thereby limiting the inclusion of controls with a potentially undiagnosed PCa. We also ran an analysis restricted to subjects screened with DRE in the last five years, to evaluate the impact on our results of a potentially lower sensitivity of PSA screening due to decreased PSA levels among MetS subjects.[30,31]

We investigated the contribution of T2D in the PCa risk associated with MetS. We examined whether the association between PCa and MetS was different among subjects of Sub-Saharan ancestry.

The risk associated with each component was estimated in a multivariate model including other components. Finally, we assessed whether changes in ORs associated with diabetes and dyslipidemia occurred after adding metformin and statin use (yes/no) in the model.

All analyses were performed using SAS software (9.3; SAS Institute Inc., Cary, NC, USA). A two-sided P value less than 0.05 was considered statistically significant.