Oophorectomy and risk of primary liver cancer and fatty liver disease among women in the UK Clinical Practice Research Datalink

Application Number
Lay Summary

Liver cancer rates have been rising rapidly in the United Kingdom. Non-alcoholic fatty liver disease (NAFLD) is a common liver disease where fat accumulates in the liver of individuals who drink very little alcohol. NAFLD has also been increasing in recent years and can potentially progress to liver cancer. Notably, in almost all areas worldwide, liver cancer and NAFLD rates are higher in men than women. While this may be partially explained by higher rates of known risk factors among men, such as viral infections or excessive alcohol consumption, these factors do not fully explain the sex differences. One possibility is that hormones could be related to the sex differences.
In animal models, oestrogen administration/removal of testicles reduces liver cancer development in male rodents, while testosterone administration/removal of ovaries increases liver cancer development in female rodents. In human studies, an increased risk of liver cancer and NAFLD was seen among women who had surgical removal of their ovaries. However, these previous population studies were based on small numbers or self-report. Thus, we propose to conduct a study of liver cancer and NAFLD in relation to ovary removal. This investigation could provide further knowledge about the benefits and harms of oophorectomy and hormone use in women.

Technical Summary

We aim to conduct two case-control studies to assess the association between oophorectomy and the subsequent development of liver cancer or non-alcoholic fatty liver disease (NAFLD) among individuals in the CPRD. We also propose to include alcoholic fatty liver disease (ALFD) as a comparator group for NAFLD, as the aetiologies underlying these two types of liver disease differ. We will identify all eligible primary liver cancer and fatty liver disease cases in this population, and controls will be matched to cases at a 4:1 ratio on age, sex, general practice, and length of time in the CPRD. Oophorectomy will be classified as ever oophorectomy and bilateral oophorectomy. We will additionally classify oophorectomy by duration of time between oophorectomy and outcome. We will use conditional logistic regression to assess crude and adjusted risk estimates (odds ratios [ORs] and 95% confidence intervals [CIs]) for oophorectomy and the risk of liver cancer or fatty liver disease, subclassified as NAFLD or AFLD. In addition to variables controlled through matching, a comprehensive list of potential confounders will be evaluated.

Health Outcomes to be Measured

Primary liver cancer or fatty liver disease, including the subtypes of NAFLD and AFLD


Katherine McGlynn - Chief Investigator - National Institutes of Health - USA
Baiyu Yang - Collaborator - Roche
Barry I Graubard - Collaborator - National Cancer Institute ( NCI )
Jessica Petrick - Collaborator - National Cancer Institute ( NCI )