Screening for ovarian cancer may lower mortality in some women, but more research is needed

A large randomized clinical trial has shown that annual screening for ovarian cancer may reduce mortality in some women, but the results were not statistically significant compared to no screening. So more research must be done on the efficacy and cost-effectiveness of screening before implementing it on a wide scale, according to researchers behind the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS).        

“This is the first ever evidence to suggest—not conclusively prove, but to suggest with a reasonable degree of confidence—that ovarian cancer screening can intervene early enough in the natural history of this disease to lead to an improvement [in mortality],” said co-lead investigator Ian Jacobs, MD, MBBS, President and Vice-Chancellor of the University of New South Wales, in Sydney, Australia.

Results of the study were presented online December 17, 2015 and simultaneously published in The Lancet.

“These results don't necessarily signal the introduction of a national screening program, but they are an exciting step forward for early detection of ovarian cancer,” commented Athena Lamnisos, CEO of The Eve Appeal, a gynecological cancer research fund, based in London, UK, which supported the study. “Medical research takes time to shift from the lab bench to the hospital bedside. Today, the UKCTOCS results move early detection one step closer.”

Previous research of ovarian cancer screening involved a blood test for CA125 protein (a tumor marker known to be elevated in women with ovarian cancer) as well as transvaginal ultrasound of the ovaries. Because these studies had not shown a clear benefit to screening, the researchers behind the current trial developed a custom Risk of Ovarian Cancer Algorithm (ROCA) designed to provide a more accurate calculation of ovarian cancer risk.

In the UKCTOCS study, investigators in 13 centers in England, Wales, and Ireland enrolled more than 200,000 postmenopausal women (ages 50 to 74) with no strong family history of ovarian cancer. They randomly assigned participants to one of three groups:

• No screening (control group).
• Annual transvaginal ultrasound screening.
• Annual multimodal screening with ROCA, and ultrasound as a second-line test.

The algorithm analyzed the results of annual blood tests for CA125 to assign a woman’s individual level of risk for ovarian cancer. If screening identified a significant CA125 increase, ROCA classified the patient’s risk as either intermediate (which prompted a repeat blood test in 3 months) or elevated (which necessitated transvaginal ultrasound and a repeat blood test within the next 6 weeks).

If abnormal results persisted, participants in both the multimodal and ultrasound-only screening groups were referred to a clinician for definitive diagnosis and treatment.

At a median follow-up of 11.1 years, 1,282 of the women had been diagnosed with ovarian cancer and 649 had died from it. Primary analysis showed that annual multimodal screening provided a mortality reduction of 15% and ultrasound-only screening provided a reduction of 11%—although these rates were not statistically significant compared with no screening.

However, when researchers compared the multimodal screening group against the control group and also excluded women found to have evidence of ovarian cancer upon entering the study, they showed that screening did have a significant 20% reduction in overall average mortality.

“If you are a pessimist, you will point out that the primary analysis in UKCTOCS did not reach statistical significance,” Dr. Jacobs said. “If you’re on the optimistic side, you would say that even though the weighted log-rank test was a post-hoc analysis…the estimate [of mortality reduction] is over 20%.”

The authors noted: “Findings from this trial suggest that for 641 women screened annually using the multimodal strategy for 14 years, one ovarian cancer death is prevented.”

The results also showed that multimodal screening offered greater mortality reduction during the latter half of the trial—28% reduction during the last 7 years of the trial compared with 8% during the first 7 years. This result suggests that an even greater difference in mortality will emerge over time, so the investigators will continue to follow these participants for an additional 3 years to learn the magnitude of this reduction.

In the meantime, the investigators are not yet recommending that screening efforts for ovarian cancer should increase.

“It’s uncertain whether or not screening can reduce ovarian cancer deaths overall,” said Fiona Reddington, PhD, Head of Population Research Funding at Cancer Research UK, a London-based cancer research charity that helped fund the study. “While this is an important step in ovarian cancer research, we would not recommend a national screening program at this point.”

Screening for ovarian cancer is not currently recommended by the American Cancer Society or the United States Preventive Services Task Force, and results from this trial are not likely to change that.

One reason for withholding such a recommendation is that screenings run the risk of false positive results and subsequent surgery. This study found that for each case of ovarian cancer detected by screening, 2 additional women in the multimodal group and 10 in the ultrasound-only group had unnecessary surgery in which the ovaries had benign or no pathology. Plus, more than 3% of patients who had false-positive surgery had a major complication. These and other harms must be fully considered in any decision making, the investigators indicated.