1
Supplementary Materials-Methods and Results. Materials-Methods and results of the unpublished case-control study by Gialamas et al.
Materials-Methods
A total of 104 histologically confirmed, incident cases of CRC, diagnosed according to C18, C19, and C20 codes of the International Classification of Diseases, 10th Revision [1], in 6 hospitals in Athens, Greece, were recruited in the study during a 5-year period (2002 – 2007). Study subjects had received no prior cancer treatment and had never been diagnosed with any other type of cancer, familial adenomatous polyposis, or inflammatory bowel disease. Each case was assigned 2 age- (± 5years) and gender-matched controls presenting to the same hospitals for routine health screening and who had a negative self-reported medical history for cancer, hepatic disease, major hormonal or hematological disorders, asthma, autoimmune disease, HIV infection, advanced heart failure, recent myocardial infarction, stroke, chronic kidney failure, nephrotic syndrome, acute pancreatitis, bone fracture, benign neoplasms, or other disorders of the gastrointestinal tract. Four potential controls refused or were unable to collaborate and were properly substituted for a total of 208 control subjects.
Information pertaining to demographic, anthropometric, lifestyle, and medical history variables was collected by trained interviewers, during an approximately 30 min interview, with the use of a pre-coded questionnaire. Anthropometrics were measured with standard techniques and self-reported weight 2 months before diagnosis was recorded, as reported earlier [2-4].
Venous blood samples were collected after an overnight fast (no later than 9:00 AM) from all cases and controls, for hormonal measurements and determination of glucose levels. All coded blood samples were centrifuged and sera were stored in deep freezers at -70˚C, before been air-shiped with dry ice to the Beth Israel Deaconess Medical Center in Boston, MA. Serum leptin levels were measured in 1 batch by trained technicians, blinded as to case-control status, using a radioimmunoassay procedure (Lincoln Research, Inc., St Charles, MO, USA; sensitivity 0.5 ng/ml; intra-assay CV 3.4–8.3%) [5]. Average preservation time was similar for cases and controls, although leptin and adiponectin levels do not systematically change with storage time.
The study protocol was approved by the University of Athens Medical School Ethics Committee, study design and performance was in accordance with the Helsinki Declaration of 1975, and participants provided written informed consent.
Statistical analysis
Descriptive values (mean, standard deviation [SD], percentiles) of demographic, anthropometric, and lifestyle variables, diabetes mellitus status, and hormonal levels were calculated among cases and controls, and differences were observed through χ2 test or t test. The data were modeled through multiple conditional logistic regression, using case-control status as the outcome variable and leptin (in increments of 1 SD of the hormone among controls) as the main predictor variable of interest, in order to study the association of leptin levels with CRC risk. Covariates were BMI at the time of diagnosis (in 2 kg/m2 increments), WHR (in quintiles), alcohol consumption (glasses per month), smoking (cigarettes smoked daily x years of smoking), physical exercise (minutes per day), plant food and red meat consumption (portions per month in quintiles), diabetes mellitus status (self-reported history of diabetes mellitus or fasting glucose level ≥126 mg/dL), education as well as weight change in the 2 months before diagnosis (increments have been reported earlier [6]). Subsequently, mean leptin levels by disease stage and tumor characteristics were calculated, and possible associations were investigated through multiple logistic regression analysis, adjusting for age, gender, BMI at diagnosis, WHR, and weight change during the last 2 months. The SAS statistical package (Version 9.1; SAS Institute, Cary, NC) was used in all analyses, and the level of statistical significance was set at p < 0.05.
Results
Data concerning demographic, anthropometric, lifestyle variables and diabetes mellitus status for the 104 CRC cases and 208 age- and gender-matched controls have been reported earlier [6]. Mean values and SDs of serum leptin levels of 104 CRC cases by disease stage and tumor characteristics and of 208 age- and gender-matched controls are presented in Suppl. Table 2. CRC cases had lower mean circulating leptin levels than controls (9.54 vs. 12.51 ng/ml respectively, p = 0.02). Similarly, female cases had lower leptin levels compared with controls (16.34 vs. 22.19 ng/ml, p=0.03) and a trend of borderline significance was observed in males (5.46 vs. 6.70 ng/ml, p=0.08) (data not shown). No statistically significant association of circulating leptin with CRC clinicopathological characteristics was noted after multiple logistic regression analysis, controlling for age, gender, BMI at diagnosis, WHR, and weight change during the last 2 months.
Suppl. Table 3 presents the ORs for CRC by leptin quartiles derived from conditional multiple logistic regression models, in which covariates studied were successively introduced. In the crude model (Model 1), a decreased risk of CRC was observed for the 4th vs. 1st quartile of leptin levels (OR = 0.31, p = 0.01); however, this significant association was not preserved after further adjustment for potential confounders (Models 2-6). As expected, WHR and recent weight loss were positively and BMI at diagnosis negatively associated with CRC risk, while no significant association was observed for education, alcohol consumption and smoking. A protective effect of physical exercise and plant food consumption was noted, along with a detrimental role of red meat consumption. Notably, positive correlation of leptin with BMI, among case and control groups, was found (data not shown).
References:
1. International classification of diseases, 10th revision (ICD-10), (available at:
2. Spyridopoulos TN, Petridou ET, Skalkidou A, Dessypris N, Chrousos GP, Mantzoros CS (2007) Low adiponectin levels are associated with renal cell carcinoma: a case-control study. International journal of cancer Journal international du cancer 120 (7):1573-1578. doi:10.1002/ijc.22526
3. Pischon T, Lahmann PH, Boeing H, Friedenreich C, Norat T, Tjonneland A, Halkjaer J, Overvad K, Clavel-Chapelon F, Boutron-Ruault MC, Guernec G, Bergmann MM, Linseisen J, Becker N, Trichopoulou A, Trichopoulos D, Sieri S, Palli D, Tumino R, Vineis P, Panico S, Peeters PH, Bueno-de-Mesquita HB, Boshuizen HC, Van Guelpen B, Palmqvist R, Berglund G, Gonzalez CA, Dorronsoro M, Barricarte A, Navarro C, Martinez C, Quiros JR, Roddam A, Allen N, Bingham S, Khaw KT, Ferrari P, Kaaks R, Slimani N, Riboli E (2006) Body size and risk of colon and rectal cancer in the European Prospective Investigation Into Cancer and Nutrition (EPIC). Journal of the National Cancer Institute 98 (13):920-931. doi:10.1093/jnci/djj246
4. Hozo SP, Djulbegovic B, Hozo I (2005) Estimating the mean and variance from the median, range, and the size of a sample. BMC medical research methodology 5:13. doi:10.1186/1471-2288-5-13
5. Gogas H, Trakatelli M, Dessypris N, Terzidis A, Katsambas A, Chrousos GP, Petridou ET (2008) Melanoma risk in association with serum leptin levels and lifestyle parameters: a case-control study. Ann Oncol 19 (2):384-389. doi:10.1093/annonc/mdm464
6. Gialamas SP, Petridou ET, Tseleni-Balafouta S, Spyridopoulos TN, Matsoukis IL, Kondi-Pafiti A, Zografos G, Mantzoros CS (2011) Serum adiponectin levels and tissue expression of adiponectin receptors are associated with risk, stage, and grade of colorectal cancer. Metabolism: clinical and experimental 60 (11):1530-1538. doi:10.1016/j.metabol.2011.03.020