Multimodal Screening Strategy for Ovarian Cancer

HOUSTON—A multimodal strategy for screening asymptomatic postmenopausal women for ovarian cancer shows promise of being able to find the disease early and improve survival. Researchers for the Ovarian Cancer Screening Project (OCSP) at St. Bartholomew’s Hospital, London, are testing a strategy combining the tumor marker CA 125 with transvaginal ultrasound and a mathematical instrument called the Risk for Ovarian Cancer Algorithm (ROCA).

Speaking at an ovarian cancer conference sponsored by M.D. Anderson and Memorial Sloan-Kettering, Ian Jacobs, MD, a gynecologic oncologist at St. Bartholomew’s and an OCSP collaborator, said that the studies have produced compelling evidence in support of this new approach.

“Our results to date demonstrate that this multimodal strategy is not only feasible and cost-effective but also highly specific and sensitive, with a positive predictive value of over 20%,” he said.

 Ovarian cancer occurs predominantly in postmenopausal women. It presents at a late stage, and fewer than 30% of its victims survive for 5 years. If the disease can be detected early, in stage I, it is possible that overall survival will be dramatically increased. Dr. Jacobs feels these factors present a convincing argument for more aggressive screening research.

The inaccessibility of the ovaries and the absence of a confirmed premalignant condition have complicated attempts to develop preclinical screening strategies for ovarian cancer. However, recent advances in tumor marker interpretation and ultrasound technology have made such screening viable, he said.

“CA 125 elevation can be detected several years before women present with clinical signs of ovarian cancer,” Dr. Jacobs said. “Transvaginal ultrasound is an effective follow-up to CA 125, and the ROCA enhances the clinician’s ability to interpret CA 125 results.”

In phase I of the OCSP study, 22,000 postmenopausal women underwent initial prevalence screening involving CA 125, ultrasound if the CA 125 level was 30 U/mL or higher, and surgery if the ultrasound scan showed an abnormality. The results showed sensitivity of 79% at 1-year follow-up, specificity of 99.9%, and positive predictive value of 27%. CA 125 was not elevated in the preclinical phase of nonovarian malignancy but was a powerful indicator of relative risk for ovarian cancer (RR = 36 for CA 125 greater than 30 U/mL).

In the second phase of the study, the same 22,000 women were randomized to receive no further screening or annual screening for 3 years. The 16 women who continued screening and developed ovarian cancer had a median survival of 72.9 months, compared with 41.8 months for the 20 women in the control group who developed ovarian cancer (P = .011).

“Among women who developed ovarian cancer, we noted a significant difference in duration of survival between the women who continued to be screened for 3 years and those who discontinued screening,” Dr. Jacobs said. “There was an 85.5% rate of compliance with follow-up screening, which we consider satisfactory. And the positive predictive value was high at 21%.”

In the third phase of the study, ROCA was used to interpret serial CA 125 levels. “Retrospective analysis with ROCA suggested that mathematical-based interpretation of the presence of this tumor marker could increase the sensitivity of screening to greater than 85% at 1-year follow-up,” Dr. Jacobs said.

Phase IV of the research program is a randomized controlled trial of 120,000 women using the ROCA to interpret CA 125 levels as a primary test and ultrasound as a secondary test. Recruitment is continuing, but preliminary results reveal a specificity of 99.9% and positive predictive value of 25%, he said.

Although some issues must be addressed before widespread ovarian cancer screening can be advocated, Dr. Jacobs feels this clinical series presents encouraging evidence for the positive effect of screening on survival for women with ovarian cancer. “Our data justify further investigation to provide definitive information about the impact of multimodal ovarian cancer screening on survival and on the health economic and psychosocial implications of screening in general.”