Screening Does Not Reduce Ovarian Cancer Mortality

View on the News

UKCTOCS May Shed More Light

The PLCO is the first study that is adequately powered to report on the mortality impact of screening for ovarian cancer, with the clear endpoint of primary invasive cancer, minimal contamination in the control arm, extensive followup, and transparent data analysis.

Yet, screening was limited to 6 years of the study. There was a considerable period of followup where no screening occurred, and 40% of the cancers were diagnosed after screening was completed. Thus, there is a massive dilution of the effect of screening by the development of new ovarian cancer cases during the followup period that could not be influenced by screening.

Also, PLCO used standard normal cutoffs of less than 35 U/mL for CA-125. Current thinking examines trends in CA-125 values, with serial increases in CA-125 considered possibly predictive even in they are less than 35 U/mL. Other limitations are that transvaginal ultrasound results are dependent on physician performance. Further, evaluations and followup of patients with positive screens were performed at the discretion of managing physicians.

In the UK Collaborative Trial of Ovarian Cancer Screening [UKCTOCS], an ongoing British study of ovarian cancer screening, 100,000 postmenopausal women will receive 7 years of screening through the end of 2011, with 3 additional years of followup. Their outcomes will be compared with 100,000 control women. Study results are expected in 2015. The screening strategy is controlled and a management algorithm is defined.

The utility of screening may be related to how the screening is performed or perhaps the natural history of ovarian cancer. If treatment is not making an impact on survival, then how early the disease is found is not consequential.

The true issue may be that the right screening approach has not been defined, not that screening does not work.

Usha Menon, Ph.D., RN, is with the UCL Elizabeth Garrett Anderson Institute of Women's Health, London, and is the principal investigator and trial coordinator of UKCTOCS which is funded jointly by the MRC, Cancer Research UK and NHS.



CHICAGO - Screening for ovarian cancer did not reduce the risk of dying from the disease for women with average risk, according to results of a multicenter screening study of nearly 80,000 women.

Screening nearly 40,000 post-menopausal women with transvaginal ultrasound for 4 years and CA-124 for 6 years did not decrease ovarian cancer mortality compared to nearly 40,000 control women. Further, there was evidence of harm in the 3,285 women with false positive results; 1080 women had surgery and 163 of them had 222 major complications, according to Dr. Saundra Buys, of the Huntsman Cancer Institute, University of Utah, Salt Lake City, whose presentation of the data was given on June 4 at the annual meeting of the American Society of Clinical Oncology. The study appears in the June 8 issue of JAMA (2011;305:2295-303).

Among women who underwent screening, 118 died of ovarian cancer, whereas 100 women who received usual care died as a result of the disease. The difference was not significant.

"It is possible that even an optimized program of annual screening may be insufficient to detect cancers early enough to reduce mortality. Evidence from modeling suggests that aggressive cancers progress rapidly through the early stages, limiting the ability to detect these cancers with yearly screening," Dr. Buys said.

More ovarian cancers were diagnosed in the screened women (212 vs. 176), although this difference was not significant. This suggests "that some of the additional cancers detected by screening were not clinically important and, if left undetected, may never have caused any symptoms or affected the women during their lifetimes," for example, with overdiagnosis, the researchers observed. Stage III and IV cancers were the most common in each group (77% of the screening group and 78% of the usual care group).

The women in this study were part of the PLCO (Prostate, Lung, Colorectal and Ovarian) cancer screening trial. In all, 68,557 women aged 55-74 years were assigned to either annual screening (34,253 women) or usual care (34,304 women) in 1993-2001 and remained in the analysis. Women in the screening arm were offered annual CA 125 testing for 6 years and transvaginal ultrasound for 4 years. Those in the usual care arm were not offered the screening tests. The median follow-up was 12.4 years.

Women with a CA 125 level of 35 U/mL or greater were classified as abnormal. Transvaginal ultrasound was conducted using a 5-7.5 MHz transvaginal probe. Results were considered abnormal if the ovarian volume was greater than 10 cm3, if the cyst volume was greater than 10 cm3, if any solid area or papillary projection extending into the cavity of a cystic ovarian tumor of any size was present, or if any mixed (solid and cystic) component within a cystic ovarian tumor was present.

Both the women and their physicians were notified in writing about suspicious abnormalities found through screening. The women’s primary care physicians were responsible for managing the diagnostic process to assess abnormalities.

Among women in the screening arm, there were a greater number of false positives (3,285 false positives vs. 212 true positives). Of women who had a false positive test, one-third (1,080 women) underwent surgery for biopsy to evaluate positive test results. Of these, 163 (15%) women had 222 distinct major complications.

One of the authors, Jonathan D. Clapp, reported having a financial interest in Human Genome Sciences Inc. The other authors reported that they had no relevant financial relationships. The National Cancer Institute funded the PLCO cancer screening trial.

Kerri Wachter contributed to this article.

Next Article: