Update on Ovarian Cancer Screening and Diagnostic Tests

By Krishnansu S. Tewari, MD

Ovarian cancer continues to thwart our efforts to detect it in the earliest stages, so it remains the deadliest of the gynecologic cancers. It is estimated that 22,000 new cases will occur in US women in 2012 and that 80 percent of those women will not survive for longer than five years following diagnosis. Physician scientists and other researchers are racing against the clock to discover a reliable ovarian cancer screening test to allow for early detection, when it is most amenable to treatment. This article summarizes some of the exciting advances in ovarian cancer screening by looking carefully at what is considered the gold standard in evidence-based medicine: the clinical trial.

 

Biomarkers: CA-125 and Beyond

CA-125 is a tumor marker that was discovered more than 30 years ago and is the only tumor marker for ovarian cancer detection approved by the US Food and Drug Administration (FDA).1 Also known as biomarkers, tumor markers are proteins that can be detected by a blood test. Elevated CA-125 levels can indicate ovarian cancer.

CA-125 is found in only approximately 75 percent of cases, however, and it is primarily elevated in serous carcinoma, a subtype of ovarian cancer. It is not expressed by other ovarian cancers, such as mucinous tumors. Additional shortcomings of CA-125 include the fact that it is not very sensitive at catching early-stage ovarian cancer and that it can be present in abnormally high levels in many different benign (non-cancerous) gynecologic and non-gynecologic conditions. For these reasons CA-125 is not a suitable screening test for ovarian cancer in the general population, and the search continues for more-sensitive and more-informative biomarkers.

More than 200 potential biomarkers have been found at either abnormally higher or lower levels in women with ovarian cancer when compared with healthy volunteers. Some of these proteins are involved in growth and angiogenesis (formation of new blood vessels that provide cancer cells with the nutrients they need to survive), whereas others are involved in the ability of a cancer to invade nearby tissues and ultimately metastasize (spread throughout the body). Some biomarkers are involved in critical processes for cancer cell survival, including the ability to conserve energy or hide from a patient’s immune system, which is always trying to remove cancerous growths. Several candidate biomarkers are involved in blood coagulation, which has been recognized to occur in patients with cancer. Finally, there are those biomarkers that have as yet undefined functions (HE4, for example).2

Unfortunately, when 35 of the most promising biomarkers were tested in blood samples taken from women up to two years before they were diagnosed with ovarian cancer, none of the biomarkers (including HE4) outperformed CA-125 in detecting what could be considered a “fingerprint” or molecular sign of the earliest stage of ovarian cancer.3,4 The search for meaningful biomarkers continues.

 

The Ovarian Component of the PLCO Study

The objective of the ovarian component of the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial was to estimate whether screening reduces mortality from ovarian cancer in healthy women ages 55 to 74 who still have their ovaries. A total of 34,261 women were enrolled in the trial and were randomly assigned to no screening interventions or to yearly vaginal ultrasounds plus CA-125. Although there were cases of ovarian cancer detected among participants, they were often advanced cases, so the trial was unable to catch the disease early. Also, for each case of ovarian cancer discovered, 20 women underwent surgery, meaning that 19 patients underwent unnecessary surgery for benign conditions for every one case of ovarian cancer diagnosed. These results were initially reported in 2009.5

Two important updates from the PLCO study have been recently published. In 2011 we learned that there was no difference in the mortality rates due to ovarian cancer between the women who did not undergo screening and those who were screened.6 This means that even though more ovarian cancers were found in women assigned to the group that received annual ultrasounds plus CA-125, because the majority of these screen-detected cases were advanced-stage cases, the screening did not result in a significantly diminished death rate from ovarian cancer. Additionally, screening resulted in more than 3,000 false-positive results and a total of 1,080 surgeries—the majority of which were for benign conditions—and fully 15 percent of patients who underwent surgery suffered serious surgical complications. Clearly, a more sensitive screening tool is needed that can detect ovarian cancer in its earliest stages and better discriminate between benign and cancerous conditions.

We recently had another update involving blood samples taken from patients on the PLCO trial.7 Scientists probed these blood samples for seven other promising biomarkers, but, even when combined with CA-125, this panel of markers was not found to be more sensitive than CA-125 alone in detecting ovarian cancer.

 

Development of the Risk of Ovarian Cancer Algorithm and the ROCA Screening Trials in Low-Risk Women

In a strategy to improve the sensitivity of CA-125 in detecting ovarian cancer, the Risk of Ovarian Cancer Algorithm (ROCA) was designed. The basic concept is to use a woman’s particular CA-125 level as the yardstick (or baseline) against which any fluctuations or changes in CA-125 can be measured over time. Risk estimates, or a ROCA score of developing ovarian cancer, can then be provided by plugging these CA-125 changes over time into a mathematical model that includes the woman’s age.8 Even though CA-125 can be abnormally elevated in non-cancerous conditions, the hypothesis is that CA-125 levels should steadily increase over time in a woman who is ultimately going to develop ovarian cancer, whereas the CA-125 levels would be expected to remain typically stable or even decrease in those with non-cancerous conditions (endometriosis, for example). Theoretically, by monitoring the ROCA score carefully, doctors may be able to intercept the disease before it starts to spread, leading to higher cure rates.

Two important ROCA studies are ongoing. The first involves 3,238 women ages 50 to 74 who have no significant family history of breast or ovarian cancer.9 This clinical trial is being performed in the United States by the National Cancer Institute’s Cancer Genetics Network, the Early Detection Research Network, and the Specialized Program of Research Excellence in Ovarian Cancer. Each participant in the trial had her CA-125 tested on a yearly basis, and the results were plugged into the ROCA software. Based on the ROCA result, women were triaged to the next annual CA-125 (low risk), a repeat CA-125 in three months (intermediate risk), or a transvaginal ultrasound study with referral to a gynecologic oncologist (high risk). Based on the results of the clinical findings and the ultrasound result, the gynecologic oncologist then made the decision whether to proceed with surgery.

The second important ROCA study is being performed in the United Kingdom: the UK Collaborate Trial of Ovarian Cancer Screening.10 In this study more than 200,000 postmenopausal women ages 50 to 74 have been randomly assigned to one of the following: no screening; annual CA-125 with ROCA followed by transvaginal ultrasound if the ROCA is worrisome; or screening with transvaginal ultrasound only on a yearly basis.

One of the exciting things about both of these ROCA studies is that they appear to be very consistent in that the specificity for both the US and UK studies is 99.8 percent. Specificity is used in medical statistics to describe how well an intervention (in this case, the ROCA score) correctly identifies the percentage of healthy people as not having a certain condition (in this case, ovarian cancer). The positive predictive value for the US ROCA study is 37.5 percent, which is very similar to that reported for the UK ROCA study of 35.1 percent. The positive predictive value is also a medical statistics tool that allows us to determine the percentage of those with a positive test result (a concerning ROCA score, for example) who actually have a condition (ovarian cancer, in this case). We are currently awaiting survival data from both of these studies to determine if the mortality rates from ovarian cancer are significantly lower using the ROCA calculations.

 

BRCA1 and BRCA2 Mutations and a New Theory for the Origin of Ovarian Cancer

Dangerous mutations in the breast/ovarian cancer susceptibility genes—BRCA1 and BRCA2—significantly increase a woman’s lifetime risk of breast cancer to 60 to 85 percent and her risk of ovarian cancer to 10 (BRCA2) and 26 (BRCA2) percent.11-16 Women who have one of these mutations are counseled to undergo more-frequent screening with mammography, CA-125, transvaginal ultrasound, and breast/pelvic examination. With respect to the ovarian cancer risk, women with BRCA1/2 mutations who have completed their childbearing or have reached the age of 35 are often counseled to seriously consider precautionary removal of the ovaries. This is referred to as risk-reducing prophylactic bilateral salping-oophorectomy.

A surprising finding over the years has been the disproportionately large number of fallopian tube cancers that have been found in these prophylactic surgery specimens compared with the number of ovarian cancers. This finding has led to our current view of hereditary ovarian cancer as being a disease that actually may originate in the fallopian tube.17 Many researchers are studying this hypothesis at the molecular level, and there is some evidence that the fallopian tube may be the origin of nonhereditary ovarian cancers as well.

 

Inherited Risk of Ovarian Cancer and Implications for Screening in High-Risk Women

A trial by the Gynecologic Oncology Group, called GOG 199, enrolled women with either BRCA mutations or a strong family history of ovarian cancer. Patients had a baseline CA-125 and transvaginal ultrasound performed and then chose to either have risk-reducing surgery (prophylactic removal of the ovaries and the fallopian tubes, for example) or continue to be screened at three-month intervals with ROCA evaluation.18

Although the data on the ROCA part of this study are not ready for analysis, the results of the nearly 1,000 high-risk patients who chose prophylactic surgery were presented at this year’s Annual Meeting of the American Society of Clinical Oncology (ASCO 2012).19 The surgical results revealed that in patients with these mutations who do not as yet have any cancer-related symptoms, there were actually some cancers that had arisen from the ovaries, fallopian tubes, or inner peritoneal lining of the body. The fallopian tube cancers fortunately had not started to invade the surrounding tissue. Many of the cancers were detected using only a microscope, and all the cancers were of the serous variety, which is the most common type of ovarian cancer.

The take-home message for patients from these important results was that 3 percent of women with BRCA mutations were found to have the early stages of cancer before any symptoms developed, and women in this group should therefore discuss current recommendations regarding prophylactic removal of the ovaries and the fallopian tubes. If prophylactic surgery is performed, the pathologist needs to look carefully for any hidden early cancers.

An additional finding from this trial was that endometrial cancers were also discovered in patients whose uterus was removed at the time of prophylactic removal of the ovaries and the tubes. This is important because the origin of the fallopian tubes lies within the uterus, and many oncologists currently recommend prophylactic hysterectomy when patients with BRCA mutations or a strong family history of breast or ovarian cancer undergo prophylactic removal of the ovaries and the tubes.

 

OVA and the Risk of Ovarian Malignancy Algorithm (HE4 plus CA-125)

Though we are still awaiting data on the ROCA score and are still lacking an adequate screening test, we have recently seen encouraging progress in diagnostic testing. A major breakthrough has been realized for women who are found to have a seemingly isolated pelvic mass—a cystic or solid mass arising from one ovary without any other suspicious findings (ascites fluid, nodules, or masses) in the abdomen or pelvis—and have been recommended surgery (see sidebar “The Importance of Proper Surgery and Staging”).

The OVA1 is a diagnostic blood test that looks at five markers (CA-125, transthyretin, apolipoprotein A1, beta-2 microglobulin, and transferrin) and has been shown to outperform CA-125 and a gynecologist’s clinical assessment of a pelvic mass.20 This means that OVA1 more accurately predicted when a pelvic mass was malignant compared with CA-125 testing and clinical evaluation by a doctor.

In a clinical trial of more than 500 women with isolated pelvic masses, the combination of OVA1 plus physician assessment correctly identified cancers missed by physician assessment alone and also detected more than 75 percent of cancers missed by CA-125 alone.21 Important attributes of OVA1 include its ability to identify different types of ovarian cancers, its ability to detect both early- and advanced-stage ovarian cancers, the possibility that it may identify nonovarian cancers, and that it provides a composite score that can be interpreted depending on the patient’s menopausal status. (On a scale of 1 to 10, an abnormal score for a premenopausal woman is > 5, whereas an abnormal score for postmenopausal woman is > 4.4.) OVA1 was approved by the FDA in 2009 and should be used only for the triage of patients who have a pelvic mass and need to undergo surgery. OVA1 is not to be used as a screening test, although now that the test has been cleared by the FDA to help evaluate pelvic masses, plans are under way to study it as a screening test.

The second major breakthrough in the preoperative triage of pelvic masses comes in the form of the Risk of Ovarian Malignancy Algorithm (ROMA), which makes use of two serum biomarkers (CA-125 and human epididymis protein 4, or HE4).22,23 In a clinical trial of 472 patients with pelvic masses, ROMA had a sensitivity of 93.8 percent in detecting ovarian cancer.24 (Sensitivity is a medical statistical tool that is the counterpart of specificity, discussed earlier.) Using ROMA as the test and ovarian cancer as the disease we are interested in understanding, the sensitivity gives us the proportion of ovarian cancer patients who are correctly identified by ROMA as having this disease. ROMA (CA-125 plus HE4) was approved by the FDA in 2011 and, like OVA1, should be used for the preoperative triage of a woman with a pelvic mass and not for ovarian cancer screening. There have been no head-to-head comparisons of OVA1 with ROMA, so it is not known which test is better.

 

Conclusion

Women in affluent countries such as the United States have a 1 percent lifetime risk of developing ovarian cancer, even if they do not have a family history of the disease. This review has carefully presented and summarized the clinical trials that are being performed in the United States and the United Kingdom to help develop a reliable screening strategy for early detection of ovarian cancer. The goal of cancer screening is early detection, which may allow for an expeditious cure without substantial losses in quality of life.

The most promising screening test on the horizon is the ROCA model, which uses each individual asymptomatic woman’s CA-125 as a baseline against which changes in CA-125 levels are measured on an annual basis to determine if there is a concerning rise. We also reported on two new diagnostic (nonscreening) blood tests—OVA1 and ROMA (aka CA-125 plus HE4)—both of which can help gynecologists decide which patients found to have an ovarian mass are at highest risk of having an ovarian cancer. Appropriate referrals to a gynecologic oncologist for patients with a positive OVA1 or positive ROMA can result in lifesaving surgery.  _

References

Bast RC Jr, Klug TL, St John E, et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. New England Journal of Medicine. 1983;309:883-87.

Hellström I, Raycraft J, Hayden-Ledbetter M, et al. The HE4 (WFDC2) protein is a biomarker for ovarian carcinoma. Cancer Research. 2003;63(13):3695-700.

Cramer DW, Bast RC Jr, Berg CD, et al. Ovarian cancer biomarker performance in prostate, lung, colorectal, and ovarian cancer screening trial specimens. Cancer Prevention Research. 2011;4:365-74.

Zhu CS, Pinsky PF, Cramer DW, et al. A framework for evaluating biomarkers for early detection: validation of biomarker panels for ovarian cancer. Cancer Prevention Research. 2011;4:375-83.

Partridge E, Kreimer AR, Greenlee RT, et al. Results from four rounds of ovarian cancer screening in a randomized trial. Obstetrics & Gynecology. 2009;113:775-82.

Buys SS, Partridge E, Black A, et al. Effect of screening on ovarian cancer mortality: the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Randomized Controlled Trial. Journal of the American Medical Association. 2011;305:2295-303.

Moore LE, Pfeiffer RM, Zhang Z, et al. Proteomic biomarkers in combination with CA 125 for detection of epithelial ovarian cancer using prediagnostic serum samples from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial. Cancer 2012;118:91-100.

Lu KH, Kates S, Bevers TB, et al. A prospective US ovarian cancer screening study using the Risk of Ovarian Cancer Algorithm (ROCA). Journal of Clinical Oncology. 2010;28:15s (suppl; abstr 5003).

Skates SJ, Mai P, Horick NK, et al. Large prospective study of ovarian cancer screening in high-risk women: CA125 cut-point defined by menopausal status. Cancer Prevention Research. 2011;4(9):1401-8.

Menon U, Gentry-Maharaj A, Hallett R, et al. Sensitivity and specificity of multimodal and ultrasound screening for ovarian cancer, and stage distribution of detected cancers: results of the prevalence screen of the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). Lancet Oncology. 2009;10(4):327-40.

Bowcock AM, Anderson LA, Friedman LS, et al. THRA1 and D17S183 flank an interval of < 4 cM for the breast-ovarian cancer gene (BRCA1) on chromosome 17q21. American Journal of Human Genetics. 1993;52:718-22.

Simard J, Feunteun J, Lenoir G, et al. Genetic mapping of the breast-ovarian cancer syndrome to a small interval on chromosome 17q12-21: exclusion of candidate genes EDH17B2 and RARA. Human Molecular Genetics. 1993;2:1193-99.

Tonin P, Serova O, Simard J, et al. The gene for hereditary breast-ovarian cancer, BRCA1, maps distal to EDH17B2 in chromosome region 17q12-q21. Human Molecular Genetics. 1994;3:1679-82.

Smith SA, DiCioccio RA, Struewing JP, et al. Localisation of the breast-ovarian cancer susceptibility gene (BRCA1) on 17q12-21 to an interval of < or 1 1 cM. Genes, Chromosomes & Cancer. 1994;10(1):71-76.

Wooster R, Neuhausen SL, Mangion J, et al. Localization of a breast cancer susceptibility gene, BRCA2, to chromosome 13q12-13. Science. 1994;265:2088-90.

Nelson HD, Huffman LH, Fu R, et al. Genetic risk assessment and BRCA mutation testing for breast and ovarian cancer susceptibility: systematic evidence review for the US Preventive Services Task Force. Annals of Internal Medicine. 2005;143;362-79.

Norquist BM, Garcia RL, Allison KH, et al. The molecular pathogenesis of hereditary ovarian carcinoma: alterations in the tubal epithelium of women with BRCA1 and BRCA2 mutations. Cancer. 2010;116:5261-71.

Green MH, Piedmonte M, Alberts D, et al. A prospective study of risk-reducing salpingo-oophorectomy and longitudinal CA-125 screening among women at increased genetic risk of ovarian cancer. Design and baseline characteristics: a Gynecologic Oncology Group study. Cancer Epidemiology, Biomarkers & Prevention. 2008;17(3):594-604.

Mai PL, Sherman ME, Piedmonte M,…Tewari KS, et al. Pathologic findings at risk-reducing salpingo-oophorectomy among women at increased ovarian cancer risk: results from GOG-199. Journal of Clinical Oncology. 2012;30 (suppl; abstr 1519).

Miller RW, Smith A, DeSimone CP, et al. Performance of the American College of Obstetricians and Gynecologists’ ovarian tumor referral guidelines with a multivariate index assay. Obstetrics & Gynecology. 2011;117:1-9.

Ueland FR, Desimone CP, Seamon LG, et al. Effectiveness of a multivariate index assay in the preoperative assessment of ovarian tumors. Obstetrics & Gynecology. 2011;117:1289-97.

Moore RG, Brown AK, Miller C, et al. The use of multiple novel tumor biomarkers for the detection of ovarian carcinoma in patients with a pelvic mass. Gynecologic Oncology. 2008;108(2);402-8.

Moore RG, McMeekin DC, Brown AK, et al. A novel multiple marker bioassay utilizing HE4 and CA 125 for the prediction of ovarian cancer in patients with a pelvic mass. Gynecologic Oncology. 2009;112:40-46.

Moore RG, Miller C, DiSilvestro P, et al. Evaluation of the diagnostic accuracy of the risk of ovarian malignancy algorithm in women with a pelvic mass. Obstetrics & Gynecology. 2011;118:280-88.

 

The Importance of Proper Surgery and Staging

Proper cancer-removing surgery (called tumor debulking) at the time a widely metastatic ovarian cancer is discovered leads to better survival rates. Similarly, the precise and accurate surgical staging of what appears to be an early-stage ovarian cancer (when there’s just a solitary ovarian mass that upon surgical removal is found to be malignant when evaluated by the pathologist) will correctly allow the mapping of a cancer’s hidden spread pattern and determine which patients need chemotherapy and, importantly, how much chemotherapy.

For these reasons, whenever a cancer is detected, even if there is no obvious evidence that it may have spread, careful surgical mapping by a gynecologic oncologist needs to be undertaken to determine if there has been any hidden (microscopic) spread. This is done through sampling lymph nodes and performing biopsies of other areas that ovarian cancer is known to spread. The gynecologist planning to operate on a woman with what appears to be a solitary pelvic mass needs the best-possible diagnostic information preoperatively so that he or she can appropriately refer the patient to a gynecologic oncologist when the suspicion of ovarian cancer is high.

 

Queen of Hearts Foundation

The Queen of Hearts Foundation (qohf.org) is a nonprofit, grassroots community dedicated to helping find a cure for ovarian cancer through fundraising for cancer research. In May 2012 the first annual Queen of Hearts Foundation Ovarian Cancer Research Awards were presented to Fong Liu, MD, for her project that involved identifying and minimizing the healthcare disparities that may exist with ovarian cancer treatment in California, and to Ramez Eskander, MD, for his project using the PARO therapeutic robot to improve quality of life for women receiving chemotherapy for ovarian cancer.