EVOLVING RISK FACTORS FOR CANCERS IN AFRICAN POPULATIONS
Lead Research Organisation:
South African Medical Research Council
Department Name: UNLISTED
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Technical Summary
"Between 1995 to date (2015), over 23,000 participants aged 18 and over with any cancer type have been recruited. Data have been cleaned for 18,000 out of 23,000 cases (up to 2011), and blood samples are available for 71%, at least 17,000, of these. Because Johannesburg draws people from all over the country, cancer types recruited represent the background reported incidence of cancer in the SA black population (National Cancer Registry 2009): for example 3900 cervix, 3400 breast, 1400 oesophagus, 1140 Kaposi sarcoma, 977 head and neck, 900 lung, 794 bowel and 610 non-Hodgkin lymphoma (1995-2011). Data cleaning for 2012-2014 is in progress and will be available early in 2016.
A case-control analysis approach has previously been used routinely to calculate odds ratios of developing each exposure and cancer type of interest, by selecting the appropriate number of age- matched cancer controls that are unrelated to the exposure of interest. We also recruited about 800 people with cardiovascular disease, and some who were later found to have benign disease. Selecting several cancer types as controls allows testing of internal consistency of the exposures, and minimizes recall bias (especially when interviewing sick cases and healthy controls). This ‘text book’ approach of case-control selection (Breslow and Day 1982), has led us to a series of successful investigations in the JCS on the role of smoking, alcohol (Pacella-Norman et al 2002, Stein et al 2008b), sexual and reproductive history and hormonal contraception (e.g. Urban et al 2012). JCS contributes data to International Consortia on the Epidemiology of cancer of the breast, endometrium, ovary and cervix (ICCC 2007, CGESC 2015). With sample sizes almost doubling since these first reports, some of the risk factor analyses such as smoking (lung cancer samples increased from 423 to 900), alcohol (by type), ethnic language, occupation and urban- rural differences require re-evaluation to identify contemporary risks for the leading and emerging cancer types.
The case control method is also the approach to be adopted by the proposed GWAS studies of breast, oesophageal and cervical cancer. For these studies, cases will be selected from the JCS for genotyping, and 15,000 population controls will be used from ongoing large genetic initiatives as described in detail below."
A case-control analysis approach has previously been used routinely to calculate odds ratios of developing each exposure and cancer type of interest, by selecting the appropriate number of age- matched cancer controls that are unrelated to the exposure of interest. We also recruited about 800 people with cardiovascular disease, and some who were later found to have benign disease. Selecting several cancer types as controls allows testing of internal consistency of the exposures, and minimizes recall bias (especially when interviewing sick cases and healthy controls). This ‘text book’ approach of case-control selection (Breslow and Day 1982), has led us to a series of successful investigations in the JCS on the role of smoking, alcohol (Pacella-Norman et al 2002, Stein et al 2008b), sexual and reproductive history and hormonal contraception (e.g. Urban et al 2012). JCS contributes data to International Consortia on the Epidemiology of cancer of the breast, endometrium, ovary and cervix (ICCC 2007, CGESC 2015). With sample sizes almost doubling since these first reports, some of the risk factor analyses such as smoking (lung cancer samples increased from 423 to 900), alcohol (by type), ethnic language, occupation and urban- rural differences require re-evaluation to identify contemporary risks for the leading and emerging cancer types.
The case control method is also the approach to be adopted by the proposed GWAS studies of breast, oesophageal and cervical cancer. For these studies, cases will be selected from the JCS for genotyping, and 15,000 population controls will be used from ongoing large genetic initiatives as described in detail below."
People |
ORCID iD |