Ovarian Cancer Screening Fails to Detect Early-Stage Disease

Roxanne Nelson

April 06, 2009

April 6, 2009 — Screening for ovarian cancer fails to detect early disease and leads to a high rate of unnecessary surgeries, according to new data from the large Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. In addition, the data so far have failed to provide clear evidence of an association between screening and ovarian-cancer-related mortality.

Screening for ovarian cancer was carried out with both transvaginal ultrasonography and the serum biomarker CA 125, and the results are reported in the April issue of Obstetrics & Gynecology.

According to an accompanying editorial, on the basis of these and other data, there does not appear to be a benefit in screening the general population with the combination of these 2 tests.

The low prevalence of ovarian cancer is the "real enemy of screening," writes the editorialist, David G. Mutch, MD, from the department of obstetrics and gynecology at Washington School of Medicine, in St. Louis, Missouri.

Any test that is designed or that is being used to detect ovarian cancer "must have an unusually high sensitivity and specificity to have a positive-predictive value that is sufficient to make it worthwhile to perform a relatively risky procedure (surgical removal of the ovaries) on a patient with a positive test," Dr. Mutch comments.

Definitive Test is Reduction in Mortality

"The definitive measure of whether any test is a good clinical screening test is a reduction in the mortality rate as a result of the screening," said lead author Edward Partridge, MD, director of the University of Alabama Comprehensive Cancer Center, in Birmingham.

But "a determination on whether screening with these 2 modalities will reduce ovarian cancer mortality must await the final results of the PLCO trial," the researchers report.

The majority of cancers identified were late stage (III or IV), when effective treatment options are limited. This is in seeming contrast to preliminary screening results from the ongoing United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS), which found that 48.3% of cancers detected by transvaginal ultrasound alone or by a combination of ultrasound and CA 125 testing were stage I or II. As reported by Medscape Oncology , the effect of screening on mortality from that British trial has also not yet been determined; final data will not be available for several more years.

The PLCO authors also found that the ratio of surgeries to screen-detected cancers was high through 4 rounds of screening. Annual screening with transvaginal ultrasonography and CA 125 failed to identify early-stage disease and led to a high rate of unnecessary removal of the ovaries.

"By definition, all screening tests cause harm, because if you have a screening test and it's positive, that leads to something else being done," said Dr. Partridge in a video talk. "All screening tests lead to something, so if they have a high false-positive rate, they lead to more things being done unnecessarily."

All screening tests lead to something, so if they have a high false-positive rate, they lead to more things being done unnecessarily.

In the accompanying editorial, Dr. Mutch points out that an effective screening test should lead to outcomes that are better after presymptom diagnosis and lead to better treatment, and the reduced morbidity/mortality should outweigh harms from false-positive tests. In addition, benefits from the test should be achieved at an acceptable level of risk.

This has not been the case for ovarian screening, however. "The prevalence of ovarian cancer is very low, no test has been shown to alter the natural course of the disease, and the intervention used to evaluate a positive screen — oophorectomy — carries significant morbidity and even mortality," he writes.

Low Positive-Predictive Values

The objective of the ovarian cancer component of the PLCO trial was to determine whether screening with CA 125 and transvaginal ultrasonography reduced mortality from ovarian cancer in healthy women who were 55 to 74 years at study enrollment. A total of 34,261 women underwent annual screening with CA 125 tests and transvaginal ultrasound examinations for 4 years, and then underwent 2 additional rounds of screening with CA 125 only.

Study enrollment began in 1993 and was completed in 2001, and participants are being followed for at least 13 years. All of the women in the screening group had intact ovaries, and compliance ranged from 83.1% to 77.6%.

Among the women who received at least 1 screening test (n = 30,630), 11.1 % had at least 1 positive result — 8.1% had at least 1 positive transvaginal ultrasound examination and 3.4% had at least 1 positive CA 125 test — and screening positivity rates showed little variability across all screening centers.

A total of 89 ovarian and peritoneal cancers were diagnosed, of which 60 were identified by screening. The authors reported that the positive-predictive value for both screening tests combined ranged from 1.0% to 1.3%, and the yield per 10,000 women screened was 4.7 to 6.2.

The biopsy rate among women with positive screens declined from 33.8% during the first year of screening to 13.8% by the fourth year. Conversely, among the women who did undergo a biopsy, the percentage diagnosed with invasive cancer increased significantly over the course of the study, rising from 2% during the first year to 9.5% during the fourth.

Majority of Cancers Detected in Late Stages

After the first round of screening, positive screen results led to 566 surgeries, resulting in a diagnosis of 18 invasive cancers, of which 83% were stage III or IV. The ratio of surgeries to invasive cancers was 31 to 1. Subsequent annual screening led to an additional 604 surgeries, for a total of 42 more screen-detected invasive cancers, resulting in a surgery-to-cancer ratio of 14 to 1. Of this group, 67% of these cancers were stage III or above.

The authors note that even though the number of surgeries required to detect a cancer was subsequently reduced from the baseline round of screening, the ratio remained somewhat elevated and the stage distribution of the detected cancers was only minimally improved. Taken together, the surgery-to-detected-cancer ratio was 19.5 to 1 over 4 rounds of screening, and 72% of the cancers detected from screening were late stage.

Similarities Noted Between Studies

Even though the findings from the UKCTOCS study showed a higher rate of detection of early cancers, the authors of the PLCO trial point out that the 2 studies actually demonstrated similar findings otherwise. In the UKCTOCS study, there was a cancer yield of 7.7 per 10,000 women screened and a biopsy-to-detected-cancer ratio of 3.2 to 1, they write. To progress to a diagnostic follow-up, women in the UKCTOCS had to have an elevated CA 125, and "these figures are perhaps most appropriately compared with the corresponding PLCO results from women with a positive CA 125, which were a yield (at baseline) of 5.2 per 10,000 and a biopsy-to-detected-case ratio of approximately 4.5 to 1."

In his commentary, Dr. Mutch reiterates that in the PLCO study, screening failed to detect ovarian cancer at an early stage. Instead, "greater than 70% of the patients in this data set had advanced-stage disease, as is expected in published data on unscreened populations," he writes.

"Clearly, more research with the development of new and better technologies is needed if we are going to help identify patients with ovarian cancer at an earlier stage," he concludes.

The study was funded by National Cancer Institute intramural and extramural funds and by individual contracts from the National Cancer Institute to each of the 10 screening centers and to the coordinating center. Study coauthor Bruce Kessel, MD, from the Pacific Health Research Institute, in Honolulu, Hawaii, has received research funding from Vivus, Wyeth, and Proctor & Gamble; has been a consultant to Novartis, Merck, and Eli-Lilly & Co; and has served on the speakers bureau for Bayer and Merck. The other authors have disclosed no relevant financial relationships. Dr. Mutch has disclosed no relevant financial relationships.

Obstet Gynecol. 2009:113:772-774, 775-782. Abstract, Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.