FROM JOURNAL OF CLINICAL ONCOLOGY
A cancer risk algorithm using serial measurements of serum cancer antigen 125 (CA-125) detected more cases of ovarian cancer than did a single-threshold rule, according to a large, prospective screening study published in the Journal of Clinical Oncology.
The findings “highlight the need to examine serial change in biomarker levels in the context of screening and early detection of cancer. Reliance on predefined single-threshold rules may result in biomarkers of value being discarded,” wrote Dr. Usha Menon, professor of gynecological cancer at University College London (J. Clin. Oncol. 2015 June 20 [doi: 10.1200/JCO.2014.59.4945 ]).
The risk of ovarian cancer algorithm (ROCA) compares an individual’s CA-125 serial profile with that of cancer cases and healthy controls to estimate a risk for ovarian cancer. The United Kingdom Collaborative Trial of Ovarian Cancer Screening enrolled 202,638 women and randomly allocated about 50,000 to multimodal screening using CA-125 velocity measurements and ROCA. From 2002 to 2011, 296,911 incident screens were performed, which was a median 7 screens per participant.
Out of 155 women with invasive epithelial ovarian or tubal cancers (iEOC), ROCA detected 85.8%, whereas single measurements of CA-125 as thresholds 35, 30, and 22 U/mL would have identified 41.3%, 48.4%, and 66.5%, respectively. The ROCA had significantly larger area under the receiver operating curve (ROC) 0.915) than did single CA-125 measurements (0.915 vs. 0.869; P = .0027). Most of the screen-detected iEOCs (82.0%) were aggressive type II, contrary to concerns that screening identifies more indolent cancers.
Surgery as a result of positive screen results occurred in 640 women (0.2% of all screens) and of these, 154 (24.1%) had primary ovarian and/or tubal malignancies detected, representing a lower false-positive rate than other screening strategies.
The median time from annual screen to surgery was 20 weeks, but for women who had CA-125 in the normal range (and intermediate ROCA), the median time to surgery was 30 weeks, because of a requirement for repeat tests. Modifications to the screening strategy to decrease this time may be warranted given that this group had a higher proportion of stage I and II iEOCs compared with those who had intermediate ROCA coupled with above-normal CA-125, did not require more testing, and had a median time to surgery of 12 weeks.