Ovarian cancer remains the most deadly gynecologic malignancy in the United States with more than 14,000 deaths in 2016. Yet, the prevalence remains low with approximately 22,000 cases in 2016. Stage at diagnosis is one of the strongest predictors of overall survival. The 5-year overall survival is more than 90% with stage I disease; this drops to 25% for those with distant metastases. Unfortunately, three-quarters of patients have disease spread beyond the ovary at the time ovarian cancer is clinically identified.
In this update, we will review:
• The fundamentals of ovarian cancer screening.
• How to identify patients who would benefit from surveillance.
• The usefulness of tumor markers.
• The results from recent large ovarian cancer screening trials.
Screening is a critical part of secondary prevention through early disease detection, when patients are asymptomatic and treatment can stop progression. Core principles of a good screening test are that the test is noninvasive, tolerable to the patient, and not costly. The disease should pose a major health threat and be detected at a stage at which intervention can impart a survival advantage. Most critically, the test should be sensitive and specific (i.e., detect disease when it is truly present and rarely be positive in the absence of disease).
The likelihood that a screening test accurately diagnoses the disease is referred to as the positive or negative predictive value. For example, the positive predictive value refers to the probability that a patient has ovarian cancer when the test is positive. These predictive values rely strongly on the prevalence of the disease in the test population. Herein lies the major challenge with ovarian cancer screening: The average lifetime risk of developing ovarian cancer is low, approximately 1 in 70. Even a clinical test with 100% sensitivity and 99% specificity would have a positive predictive value of just 4.8%.1.
A further challenge for ovarian cancer screening is that the confirmatory test requires a major surgical procedure – oophorectomy or ovarian cystectomy – with its own potential risk and harm. Currently, all North American expert groups, including the U.S. Preventive Services Task Force and the Society of Gynecologic Oncology, recommend against screening patients who are at average risk for ovarian cancer.
Screening Vs. Case Finding
A significant distinction should be made between average-risk patients and high-risk patients. Ob.gyns. frequently encounter high-risk patients who would benefit from regular surveillance or case finding (for example, patients with BRCA deleterious mutations or with Lynch syndrome). There are multiple risk factors for ovarian cancer, but the strongest known is family history, which is present in 15% of ovarian cancer patients. Having one relative with ovarian cancer increases the lifetime risk of ovarian cancer up to 5%. When a patient reports having one or more family members with ovarian cancer, it is important to differentiate between a common sporadic presentation and a rare familial cancer syndrome. ACOG Practice Bulletin 103 provides excellent guidance on which patients warrant formal genetic risk assessments by a genetic counselor.2
During the last 25 years, screening for ovarian cancer in an average-risk population has been evaluated in multiple large prospective studies using serum tumor markers (i.e., CA 125) and ultrasound results.
CA 125 and HE4 tumor markers are frequently elevated in ovarian cancer and have been studied in ovarian cancer screening. However, while having a high sensitivity for detecting disease, they are nonspecific because they are also elevated in numerous benign conditions and therefore have not proven to be a useful screening tool in the average-risk population. There are clinically available tumor marker panels that are not intended for screening. Rather, they clarify the uncertainty of the presurgical adnexal mass evaluation by providing a risk score. High risk scores are generally managed in conjunction with a gynecologic oncology referral.
Combined assessment of both ultrasound findings and tumor marker levels shows more promise with respect to prediction of ovarian cancer. However, a systematic review of 25 ovarian cancer screening studies concluded that screening low-risk populations should not be included in clinical practice until randomized trials assessed the effect on mortality and the risk of adverse events. Three large randomized controlled trials have been completed to date.3,4,5
The U.K. Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) results appear promising. However, the revealing analysis was post hoc since the original study design did not take into account the inherent delayed effect in screening studies. While these results may provide a basis for future successful screening for ovarian cancer, confirmatory further analysis is pending, using additional data over a period of the next 3 years.
Ultimately, we are all excited about the possibility of effective screening protocols for ovarian cancer and await completed analyses of UKCTOCS. Until their benefits are confirmed, screening and preventive measures should be limited to those at high risk for ovarian cancer.
To read this entire article on MDEdge.com, please click here.