Ovarian Cancer Screening: An ASCO Presscast

The American Society of Clinical Oncology held a presscast of studies to presented this coming June, and one has caught the attention of news organizations worldwide. It is a study of a two-step ovarian cancer screening program, conducted by MD Anderson Cancer Center. The study used an algorithm known as ROCA (Risk of Ovarian Cancer Algorithm) to assign a risk to each individual patient. This risk then set further guidelines in place: if low risk, annual CA-125 blood tests were performed; if intermediate risk, CA-125 was performed every three months; if high risk, the patient underwent transvaginal ultrasound and gynecologic oncology referral.
 
Over three-thousand women participated in this multi-institutional study and 85% continued on annual CA-125 evaluations over the 9-year course of the study. Approximately 500 women were triaged to CA-125 draws every 3 months at an annual rate of screening every 3 months of 7%. However, triage to ultrasound and gynecologic oncology referral was less than 1% on an annual basis. Even more significant, with the use of two-step screening, only 8 women underwent study-directed surgery and of those 8, 3 were found to have invasive ovarian cancer, stages I or II, and 2 were found to have tumors of borderline potentital. The positive predictive value of the risk algorithm (ROCA) was 37.5%; specificity of the two-step screening process was 99.7%. These far exceed the expectations of a screening strategy, particularly for a disease whose prevalence in normal-risk women is low.
 
These preliminary results are quite interesting, and if proven true in the larger ongoing trial (with a target enrollment of over 200,000) there may indeed be a mechanism to screen women for ovarian cancer. But, as Dr. Martee Hensley, a medical oncologist at Memorial Sloan-Kettering Cancer Center, there are multiple parameters that must be asked of a particular screening strategy: (1) it is highly sensitive to detect early-stage ovarian cancers? (2) it finds only ovarian cancer, and not other lesions of the ovary? (3) it performs well in normal-risk women? (4) it improves survival among women who are screened?
 
Thus far, the presscast indicates that the ROCA-based two-step screening strategy may indeed meet parameters (1) and (3). It appears to detect ovarian lesions, but only 3/8 women undergoing surgery had invasive ovarian cancer--I excluded borderline ovarian tumors because their natural history is drastically different from invasive ovarian cancer. So, we need to await a larger study to evaluate this parameter further. However, it is the fourth parameter that makes these findings, while promising, not ready for primetime. We must know if screening will improve the survival of those women who undergo screening. To answer this question will take more than 9 years to answer, but fortunately, larger trials are underway, both here and Europe aimed at answering this quesiton, the holy grail in early stage diagnosis of ovarian carcinoma.
 
For further information:
 
Hensley ML. A step forward for two-step screening for ovarian cancer. J Clin Oncol 2010; 28:2128-30.
 
Lu KH, Skates S, Bevers TB, et al. A prospective US ovarian cancer screening study using the risk of ovarian cancer algorithm (ROCA). J Clin Oncol 2010; 28:7s (suppl, abstr 5003).