Current issues of ACP Journal Club are published in Annals of Internal Medicine


Diagnosis

Screening asymptomatic women for ovarian cancer is not recommended

ACP J Club. 1995 Jan-Feb;122:22. doi:10.7326/ACPJC-1995-122-1-022


Source Citations

Carlson KJ, Skates SJ, Singer DE. Screening for ovarian cancer. Ann Intern Med. 1994 Jul 15;121;124-32.

American College of Physicians. Screening for ovarian cancer: recommendations and rationale. Ann Intern Med. 1994 Jul 15;121:141-2.


Abstract

Objective

To review the evidence for screening asymptomatic women for ovarian cancer with ultrasonography (US), CA 125 radioimmunoassay (CA 125), or both.

Data sources

Studies were identified using MEDLINE (1982 to 1993), bibliographies of relevant papers, and personal files. English-language studies and those with English-language abstracts were selected.

Study selection

Studies were selected if ovarian cancer risk factor assessment was limited to formal epidemiologic studies of defined populations. Studies on the effectiveness of treatment and diagnostic evaluations were limited to randomized controlled trials and studies of case series. For calculating sensitivities and specificities of US and CA 125, only studies in which ovarian cancer was confirmed by surgery were included.

Data extraction

Risk factors and relative risk for ovarian cancer, sensitivity and specificity for US and CA 125, and the number of patients with and without ovarian cancer. Sensitivity, specificity, and the predictive values of tests were calculated after pooling the data.

Main results

The lifetime probability of a woman in the United States developing ovarian cancer is 1 in 70 and the mean age at presentation is 59 years. The risk for ovarian cancer increases with age and family history of ovarian cancer and decreases with oral contraceptive use and pregnancy. Combining 14 studies (7 of women with known or suspected ovarian cancer and 7 screening studies) of US, the summary estimate of sensitivity was 85% (95% CI 80% to 90%) and specificity was 93.7% (CI 93.3% vs 94.3%). Combining 20 studies (17 of women with known or suspected ovarian cancer and 3 screening studies) evaluating CA 125, the summary estimate of sensitivity was 78% (CI 73% to 83%) and specificity was 98.9% (CI 98.6% to 99.2%). US had a positive predictive value < 1% for women with no risk factors and 2% for women with a family history of ovarian cancer. For CA 125, the positive predictive value increased to 3% for women with no risk factors and 10% for women with a family history of ovarian cancer.

Conclusions

For women with no family history of ovarian cancer, screening with ultrasonography or CA 125 radioimmunoassay is not recommended. For women with a positive family history, other risk factors, such as oral contraceptive use and parity, should be considered when making individualized decisions about screening.

Source of funding: American College of Physicians.

For article reprint: Dr. K.J. Carlson, Medical Practices Evaluation Center, Massachusetts General Hospital, Boston, MA 02114, USA. FAX 617-724-6725


Commentary

This study by Carlson and colleagues presents a thorough review of the risk factors for ovarian cancer and the accuracy of US and CA 125 as modalities for ovarian cancer screening.

The review by Carlson and colleagues pools data from screening studies to produce summary estimates of the sensitivity, specificity, and positive predictive value of US and CA 125 in screening for ovarian cancer. The authors conclude that the predictive values of these screening modalities are too low to justify recommendation of their use for women who do not have a family history of ovarian cancer. This conclusion is consistent with those of other professional societies including the American College of Obstetricians and Gynecologists and the U.S. Preventive Services Task Force.

Less clear, however, is the best approach for ovarian cancer screening in women with a family history of ovarian cancer in a first- or second-degree relative or those with evidence of 1 of the hereditary ovarian cancer syndromes. Although the higher incidence of disease in these groups may warrant screening, no specific protocols have been recommended by the professional societies. The recommendations provided in this article and used in the American College of Physicians' guidelines are to consider other risk factors, inform women of the risks and benefits of screening, and make decisions about screening after discussing options with the patient. In the case of the hereditary ovarian cancer syndrome, consultation with a specialist in the field of gynecology is recommended.

These recommendations are appropriate given our experience with ovarian cancer screening. Future research efforts should focus on how to identify and effectively screen women with a family history of ovarian cancer or evidence of a hereditary ovarian cancer syndrome. Such an approach offers the best hope of reducing morbidity and mortality from ovarian cancer.

Marilyn M. Schapira, MD
Zablocki Veterans Affairs Medical CenterMilwaukee, Wisconsin, USA