New Hope for Finding Ovarian Cancers in High-Risk Women
by Sue Friedman and Dr. Janiel Cragun
Genetics involves the study of genes, the body’s master code that determines how individual cells evolve into skin, muscle, bone and nerve. By understanding genetics, we can recognize changes to genes—BRCA mutations which lead to hereditary breast and ovarian cancers, for example—and take actions to manage the risk for these cancers. Which genes are present and “switched on” in a particular cell affects which proteins the cells make. Proteins are the chemicals of life that distinguish different cells from each other, allowing a skin cell to cover our body, a muscle cell to move parts of our body, and a nerve cell to conduct electrical impulses. The science of proteomics identifies the thousands of proteins produced by the body, and how they function in health and illnesses such as cancer.
Until recently, scientists were capable of studying only very large proteins in small volumes, and identifying just one or two proteins at a time. The application of these proteins towards cancer screening is limited. Some may be present in non-cancerous conditions,such as inflammation or infection,or are specific to more than one cancer type. This is the case with CA125,a protein test commonly used with transvaginal ultrasound and rectalpelvic examination to diagnose gynecologic cancers, follow ovarian cancer progression, or screen for ovarian cancer in high-risk women. The test produces less than optimal results: CA125 is elevated in only 80% of patients with advanced disease and 50-60% of women with early stage ovarian cancer; CA125 can be elevated in some women without cancer, thus making the test difficult to interpret.
Advances in proteomics are improving our capability to identify both smaller proteins and larger numbers of proteins in a single mass analysis. These advances allow us to search for “protein patterns” that might indicate specific cancers more accurately, previously an impossible task. In an article published in 2002 in The Lancet, investigators conducting preliminary research in protein pattern recognition were able to distinguish blood samples from ovarian cancer patients from patients who didn’t have ovarian cancer. Based on this study, many researchers are optimistic that proteomics can help in ovarian cancer detection. An elusive disease to date, ovarian cancer is often discovered in later stages.
Development of a screening test for ovarian cancer faces challenges. To be considered clinically useful, a screening tool for a disease as rare in the general population as ovarian cancer must be highly sensitive (having a high true positive rate) and specific (having a high true negative rate). Considering the fear and anxiety caused by a falsely positive test, and the resulting potential for invasive surgery, some experts believe that unless a test with near-perfect specificity and sensitivity to cancer is developed, the benefits of ovarian cancer screening in the general population don’t outweigh the risks.
"There is always a balance between making something available because we need it right now and making sure it stands up to the rigors of good scientific research,” says Dr.Tim Rebbeck, researcher of epidemiology at University of Pennsylvania Abramson Cancer Center. “The medical community needs to have confidence in any screening test before it becomes a standard of care.”
But what are the consequences of not detecting ovarian cancer? The ovaries are naturally hidden, tucked away in the body cavity, obscuring tumors from study. If ovarian cancer is caught in early stage, the cure rate can reach 90%. Currently only 23% of ovarian cancers are diagnosed in stage I.
Women with BRCA mutations have a higher prevalence of ovarian cancer than the general population. How does the standard for a near-perfect test translate for this high-risk population? According to Dr.Rebbeck, the answer may depend on a woman’s individual circumstances, including factors such as her age, whether she carries a BRCA1 or BRCA2 mutation, and her tolerance for the potential for false positives. For example, a woman who is already considering prophylactic oophorectomy may be willing to trade specificity for increased sensitivity; she might worry more about missing an ovarian cancer that is already present than a receiving a false positive test. A 37-year old woman with a BRCA1 mutation who is undecided about having a child might be willing to risk a one percent chance false positive rate in exchange for the certainty that a negative test will mean she doesn’t have ovarian cancer.
Until a proteomic test is commercially available these are largely speculative concerns. Proteomic research, however, is advancing rapidly. There’s growing optimism it will lead to improved ovarian cancer detection for both the general and high-risk populations. Efforts are focusing on improving reproducibility of these tests. Recent studies in prostate cancer have shown encouraging reproducibility between institutions, giving hope of standardizing this process for screening purposes. New studies on LPA, a lipid that is elevated in ovarian cancer, are showing promise. Meanwhile research to validate protein patterns for use in detection of ovarian cancer in BRCA carriers is continuing. While near-perfect sensitivity and specificity may be the ultimate goal, any improvement over our current ability to detect ovarian cancer in high-risk patients would be embraced by this community.
To learn more about screening tests currently available for ovarian cancer, and more about research studies enrolling women for ovarian cancer detection, visit the FORCE website.
Dr. Janiel Cragun is a post-doctoral fellow studying ovarian and endometrial cancer at the H. Lee Moffitt Cancer Center and Research Institute in Tampa, Florida.
The Language of Science
Scientists use the following criteria when considering the utility of a screening test:
- Sensitivity is the probability of a positive test among those who have the disease (true positives). If two women in a group of 100 have ovarian cancer, for example, a highly sensitive test would find positive results in both women.
- Specificity is the probability of a negative test among those who do not have the disease (true negatives). If only two women in the group of 100 have ovarian cancer, but 10 falsely test positive, the test is not very specific.
The ideal test is both sensitive and specific. If someone tests positive, they likely have the disease; if they test negative they likely don’t. It is difficult to develop a test that is highly sensitive and highly specific. Usually one quality is compromised at the expense of another.
- Reproducibility refers to the key ability to achieve consistent results with the same test at different times and/or different facilities.
Petricoin EF III,Ardekani AM, Hitt BA, et al.
Use of proteomic patterns in serum to identify ovarian cancer. The Lancet, vol. 359, no. 9306, 572-577.
Sutphen R, Xu Y,Wilbanks GD, et al. (July
Lysophospholipids are potential biomarkers of ovarian cancer. Cancer Epidemiology Biomarkers & Prevention, vol. 13, no. 7, 1185-1191.
Posadas EM, Davidson B,Kohn EC (2004)
Proteomics and ovarian cancer: implications for diagnosis and treatment: a critical review of the recent literature. Current Opinion in Oncology, vol.16, 478-484.
Semmes OJ, Fen A,Adam B, et al. (2005)
Evaluation of serum protein profiling by surface enhanced laser desorption/ionization time of flight mass spectrometry for the detection of prostate cancer: I. assessment of platform reproducibility. Clinical Chemistry, vol. 51, no.1, 102-112.
Disclaimer: Health links are made available for educational purposes only. This information should not be interpreted as medical advice. All health information should be discussed with your health care provider. Please read our full disclaimer for more information.
This site has been made possible by a generous grant from Morphotek.