Ovarian cancer is the 5th most common cause of cancer related death in UK women. The majority of women are diagnosed late and only 46 out of every 100 UK women survive for 5 years after diagnosis. Early diagnosis is likely to result in better patient outcomes including survival.
However, early diagnosis is challenging. The symptoms of ovarian cancer are vague and the same symptoms occur in non-worrying medical conditions, so it is can be difficult for GPs to decide which patients need to be sent to hospital urgently for more tests and which can be reassured. Simple blood tests, such as CA125, can be used to help GPs make these decisions. However, we don't know how good CA125 is when used in primary care or what 'cut-off point' to use for an abnormal result.
In this study, we aim to determine how effective CA125 is at picking up cancer in women visiting their GP with symptoms which could be caused to ovarian cancer, and identify the most appropriate abnormal CA125 cut-off. This work will help GPs to make decisions regarding investigation and referral of symptomatic women.
Ovarian cancer has the worst prognosis of any gynaecological cancer. Early diagnosis is likely to improve survival, and, while symptoms occur in all stages, they are also common in benign conditions. Tests are needed to help distinguish malignant from benign disease in symptomatic patients.
The serum biomarker CA125 is frequently elevated in women with ovarian cancer. It is used as a first line investigation in primary care, in the UK and internationally, in patients presenting with symptoms that might be caused by ovarian cancer. Despite widespread use, the diagnostic accuracy of CA125 in the primary care population has not been established and the current 'abnormal threshold' (35u/ml) is not based on primary care data.
In this prospective cohort study, we will determine the diagnostic accuracy (sensitivity, specificity, PPV, NPV) of CA125 in a symptomatic primary care population and identify CA125 thresholds that equate to a range of risk thresholds (PPVs). As CA125 levels and ovarian cancer risk are influenced by patient variables, we will produce stratified thresholds based on key variables e.g. age. This work will allow GPs to make decisions about further investigation and referral based on patient risk.
Primary outcome:
- Ovarian cancer diagnosis
Secondary outcomes:
- Diagnosis of a cancer other than ovarian
- Histological type and morphology, stage, size, grade of ovarian cancer at diagnosis
- Death from ovarian cancer or death from another cancer
Fiona Walter - Chief Investigator - University of Cambridge
Garth Funston - Corresponding Applicant - University of Cambridge
Emma Crosbie - Collaborator - University of Manchester
Gary Abel - Collaborator - University of Exeter
Michael Roberts - Collaborator - University of Cambridge
Owain Jones - Collaborator - University of Cambridge
William Hamilton - Collaborator - University of Exeter
HES Admitted Patient Care;NCRAS Cancer Registration Data;ONS Death Registration Data;Patient Level Townsend Score