Sexually transmitted infections like chlamydia and gonorrhoea may lead to a condition called pelvic inflammatory disease (PID). In women, PID results in inflammation and possible scarring within the reproductive organs - especially in the fallopian tubes, which connect the womb to the ovaries. Several research studies suggest that chlamydia and PID may increase risk for ovarian cancer. However, it is our understanding that this risk for ovarian cancer has not been measured using a large data set like the United Kingdom Clinical Practice Research Datalink, which has information from medical records on PID, chlamydia, and gonorrhoea. Therefore, we are proposing to use this data to estimate risk of ovarian cancer among women with and without a diagnosis of PID, chlamydia, and/or gonorrhoea. We will also determine if any risks are stronger for certain types of ovarian cancer (for example, "serous" tumours - those that start developing in the fallopian tubes themselves).
Infections and chronic inflammation that affect the fallopian tubes are being investigated as possible ovarian cancer initiators/promoters, as the majority of serous ovarian cancers are thought to originate in the fallopian tubes. Several studies examine risk for ovarian cancer associated with pelvic inflammatory disease (PID), but do not have information on PID-associated infections, do not include fallopian tube cancers, and often rely on patient report. The United Kingdom Clinical Practice Research Datalink (CPRD) is an ideal population to prospectively examine the associations between medical record confirmed PID, several of its underlying causes, and ovarian cancer. We plan to generate an exposure-matched cohort of women with a record of PID, chlamydia infection, and/or gonorrhoea infection diagnosed in English CPRD clinics. We will link to death registration data and the National Cancer Registration and Analysis Service to identify incident, invasive ovarian, fallopian tube, and peritoneal cancers for our primary outcome and for censoring information (death certificate-confirmed cancers will be included). Cox proportional hazards models will be used to estimate time to diagnosis of these cancers associated with record of PID and its related infections. We will also estimate cancer risks by histologic subtype and stage.
Primary outcome: ovarian and fallopian tube cancers, some peritoneal cancers
- Secondary outcome: endometrial cancer
- Secondary outcome: breast cancer
Britton Trabert - Chief Investigator - National Cancer Institute ( NCI )
Kara Michels - Corresponding Applicant - National Cancer Institute ( NCI )
- Collaborator -
Barry I Graubard - Collaborator - National Cancer Institute ( NCI )
Hamish Mohammed - Collaborator - Public Health England
Megan Clarke - Collaborator - National Cancer Institute ( NCI )
NCRAS Cancer Registration Data;ONS Death Registration Data;Practice Level Index of Multiple Deprivation