Tumor Markers and Screening - Ovarian and Fallopian Tube Cancer

Usha Menon and Ian J. Jacobs , Practical Gynecologic Oncology 

One of the established strategies for combating cancer is screening the asymptomatic population for premalignant conditions and early-stage disease. These screening strategies are based on criteria laid down by the World Health Organization . Mass screening for cervical cancer fulfills most of these tenets, and organized screening programs in numerous countries have led to a significant reduction in cervical cancer mortality . Ovarian cancer is the other gynecologic malignancy that may meet the criteria of a disease for which population screening is justified. The disease is usually diagnosed in advanced stages when chances for long-term survival are poor. Effective treatment is available for early-stage disease and there is preliminary evidence that early detection may be beneficial. Mass screening for endometrial cancer is unlikely to be of benefit because women present in early stages with symptomatic disease. However, screening of “high-risk” populations may be useful. Vaginal and vulvar cancers are too rare to justify screening, although it is important to raise the awareness of these conditions among the elderly population.

Screening of cancers is based on detection of tumor markers. The term “tumor marker” is poorly defined and can be used to denote any change that indicates the presence of cancer. Markers may be biochemical substances produced by or in response to the tumor; cytologic, molecular or cytogenetic events detected in exfoliated cells; architectural abnormalities detected by ultrasound and other imaging modalities; or vascular changes detected by colposcopy or color-flow Doppler. The rapid growth of imaging technology, molecular biology, and cytogenetics continually adds to this ever-expanding armamentarium. Ideally, tumor markers should be tumor specific, allow detection of minimal disease, and quantitatively reflect tumor burden.

For screening protocols, the value of the marker depends heavily on its sensitivity (proportion of cancers detected by a positive test) and specificity (proportion of those without cancer identified by a negative test), which must be well established before the protocol is adopted into routine practice.

Biochemical tumor markers can be broadly classified into tumor-specific and tumor-associated antigens. 

An important aspect of screening is defining the risk groups to target for screening. Even for cervical cancer, where mass screening is the norm, age may be used to define the population. For example, in the United Kingdom, screening is limited to women between the ages of 20 and 64 years. Risk groups for sporadic ovarian cancer are defined by postmenopausal status and age (50 years or older), whereas for hereditary ovarian malignancy, they are defined by family history criteria and presence of BRCA1 and BRCA2 mutations. Increased risk based on family history is also the basis of defining a target population for endometrial cancer screening. 

Ovarian and Fallopian Tube Cancer

Deficiencies in our knowledge of the molecular and biologic events in ovarian carcinogenesis have hampered our ability to screen for this disease. A true precursor lesion for ovarian cancer has not been identified, limiting the goal of screening to detection of asymptomatic, early-stage disease. Biochemical, morphologic, vascular, and cytologic tumor markers have all been explored with varying success. There is as yet only preliminary evidence that ovarian cancer screening might reduce mortality (6), and the most appropriate option for women in the general population is participation in ongoing trials.

Biochemical Markers

Circulating antigens released by the tumor predominate in this group, the best known being CA125. CA125 is an antigen expressed by fetal amniotic and celomic epithelium. In the adult, it is found in tissue derived from celomic epithelium (mesothelial cells of the pleura, pericardium, and peritoneum) and mullerian epithelium (tubal, endometrial, and endocervical). The surface epithelium of normal fetal and adult ovaries does not express the determinant, except in inclusion cysts, areas of metaplasia, and papillary excrescences.

A serum CA125 of 35 U/mL, initially measured using the homologous assay and representing 1% of healthy female blood donors, is usually accepted as the upper limit of normal. This cutoff value is fully retained by the CA125 II assay (12), which is now preferred because of reduced interassay variation. An upper limit of normal of 35 U/mL is an arbitrary cutoff and may not be ideal for certain applications of CA125. For example, in postmenopausal women or in patients after hysterectomy, CA125 levels tend to be lower than in the general population, and lower cutoffs may be more appropriate; 20 and 26 U/mL, respectively, have been suggested.

Interest in CA125 as a screening test was initiated by the fact that approximately 83% of patients with epithelial ovarian cancer had CA125 levels ³35 U/mL. Elevated levels were found in 50% of patients with stage I disease and in more than 90% of women with more advanced. In addition, it became apparent that CA125 could be elevated in the preclinical asymptomatic phase of the disease because elevated levels were found in 25% of 59 stored serum samples collected 5 years before the diagnosis of ovarian cancer. In a prospective ovarian cancer screening study of Swedish women, a specificity of 97% and positive predictive value of 4.6% were achieved using CA125 (³30 U/mL) in 4,290 volunteers aged 50 years and older. Similar specificity (96.6%) and positive predictive value (4.2%) were obtained more recently on screening 2,550 volunteers in New Zealand. The low specificity and positive predictive value of CA125 used as the sole screening test for ovarian cancer is in part due to the marker being elevated in other cancers (pancreatic, breast, bladder, liver, lung), as well as in benign disease (diverticulitis, uterine fibroids, endometriosis) and physiologic conditions (pregnancy and menstruation). 

Improving Specificity of Ovarian Cancer Screening

Specificity of screening with CA125 was improved by limiting its use to postmenopausal women who were at increased risk of ovarian cancer and in whom many of the nonmalignant conditions mentioned previously do not occur. A CA125 of 30 U/mL or more in an asymptomatic postmenopausal woman is associated with a 36-fold increased risk of ovarian cancer in the subsequent year.

Specificity for screening can be further improved by the addition of pelvic ultrasound as a second-line test to assess ovarian volume and morphology. Using multimodal screening tests that incorporated sequential CA125 and pelvic ultrasound, a specificity of 99.9% and positive predictive value of 26.8% (approximately four operations for each cancer detected) for detection of ovarian and fallopian tube cancer was achieved in 22,000 postmenopausal women. Other studies, using a similar multimodal approach, have reported lower positive predictive values.

Risk of Ovarian Cancer Algorithm

Further improvements to the strategy have been made by a more sophisticated approach to interpretation of CA125 results. It has been observed that elevated CA125 levels in women without ovarian cancer are static or decrease with time, whereas levels associated with malignancy tend to rise. This finding has been incorporated into an algorithm that uses age, rate of change of CA125, and absolute levels of CA125 to calculate an individual's risk of ovarian cancer (ROC). Transvaginal ultrasound and refined interpretation of scan findings in women with elevated CA125 levels may further improve the specificity of multimodal screening and decrease the number of women without ovarian cancer who have a positive screening result. The sensitivity of CA125 using multimodal screening has been reported as 78.6% to 100% at 1 year. The ROC algorithm increases the sensitivity of CA125 compared with a single cutoff value, because women with normal but rising levels are identified as being at increased risk.