New Screening Tool for Ovarian Cancer Shows Promise

August 29, 2013
Anna Azvolinsky

Researchers have demonstrated that a two-step screening test can identify ovarian cancer early, before the disease progresses to an advanced, poor prognosis stage.

Researchers have demonstrated that a two-step screening test can identify ovarian cancer early, before the disease progresses to an advanced, poor prognosis stage. While the prospective study enrolled more than 4,000 participants and followed them for 11 years, a large randomized clinical study is still needed to decide whether the test could become a part of routine screening for women. Such a trial is currently ongoing in the United Kingdom.

The 200,000-patient United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) is a prospective, randomized trial that will address the sensitivity of the test. Results are expected in 2015.

Karen Lu, MD, professor of gynecologic oncology and reproductive medicine at the University of Texas MD Anderson Cancer Center in Houston, and colleagues tested the screening method on postmenopausal women of average ovarian cancer risk. The test achieved both high specificity (few false positive results) and a positive predictive value. According to the authors, the results suggest that a long-term screening approach could be the most valuable way to screen women who do not have an elevated risk of ovarian cancer. The results of the single-arm study are published in Cancer.

The most aggressive form of ovarian cancer, ovarian carcinoma, is predominantly diagnosed as a stage IV cancer, where the 5-year survival rate is approximately 18%. Ovarian cancer is the fifth leading cause of cancer among women in the United States. No screening tests to identify the disease early currently exist for this cancer. A high bar for screening accuracy exists because any positive result is followed by invasive surgery to conclusively diagnose the disease.

Lu and researchers tested a two-step screening method that have each been used independently in the past. Women in the study had an annual carbohydrate antigen 125 (CA-125) blood test. CA-125 is a blood protein and tumor marker. A rising CA-125 value has been previously linked to progressive ovarian cancer, while stable CA-125 values, even when high, are associated with benign ovarian conditions. Each participant also received a Risk of Ovarian Cancer Algorithm (ROCA) score. ROCA was previously developed from prospective trials of postmenopausal women at normal risk for ovarian cancer. In this study, women were divided into three categories based on their ROCA score: low-risk women who should receive their next CA-125 test in 12 months, intermediate-risk women who should have a repeat CA-125 test in 3 months, and high-risk women who should receive a transvaginal ultrasound (TVS) and be referred to a gynecologic oncologist.

Annually, an average of 93.3% of women were categorized as low risk, 5.8% were intermediate risk, and 0.9% were high risk. During the 11 years on study, 83.4% of women remained low risk, 13.7% had to have a repeat CA-125 test.

A total of 117 participants (2.9%) were categorized as high risk, 82 of whom had a normal TVS. Ten women who had a suspicious finding from the ultrasound screening went on to surgery. Of these women, three had a benign cystadenoma, two patients had stage I ovarian serous tumors of low malignant potential, four patients had early-stage high-grade invasive ovarian cancers, and one patient was found to have endometrial cancer.

Based on these results, the two-step screening approach had a 99.9% specificity. This means that only 0.1% of patients without cancer were identified as falsely having cancer. The test had a positive predictive value (PPV) of 40% (95% confidence interval of 12.2% to 73.8%). According to the authors, a PPV of at least 10% (equal to ten operations for each case of ovarian cancer detected) would be an acceptable risk-benefit limit.

The authors credit the high sensitivity of the new approach to the detection of rising, but still in the normal range, CA-125 levels in two of the four invasive disease cases, and the short-term, 3-month follow up of CA-125 testing.

The value of other screening biomarkers-HE4, CA-72-4, and MMP-7-are also being studied either alone or in combination.