CA125 Cancer Antigen Test
Dear OncoLink "Ask the Experts,"
My insurance company has denied payment for the CA125 cancer antigen test for my wife who died from pancreatic cancer. Her oncologist frequently used this test to provide an indicator of the effectiveness of the cancer treatment. My insurance company says that its value as a diagnostic tool for malignant neoplasm of the pancreas has NOT been established. They said that I need a "current published, peer reviewed, scientific literature supporting the use of CA125 in the treatment and/or diagnosis of malignant neoplasm of pancreas." The closest article is your FAQ (CA125 Cancer Antigen 125 for ovarian cancer). However it states that cancer of the pancreas also "may cause" abnormal CA125 levels. It also does not seem to meet their peer reviewed and scientific literature requirement. Can you please send me an article that will satisfy my insurance company's requirements for malignant neoplasm of the pancreas? I believe this is a valid test for the "treatment and/or diagnosis" of all cancers including cancer of the pancreas.
Thank you in advance for your assistance.
Peter Argenta, MD and Ivor Benjamin, MD (Former co-Editor-in-Chief of OncoLink) from the Division of Gynecologic Oncology in the Department of Obstetrics and Gynecology at the University of Pennsylvania Health System, respond:
Tumor markers are antigens (usually proteins) expressed by cancer cells. Tumor markers may be located within an affected cell, at the cells' surface, or may be excreted by the cell. Excretion may occur locally, through the urinary system, through the gastro-intestinal system, or into the blood stream. With many cancers the production of tumor markers is proportional to the tumor burden. Following the levels of these markers, therefore, allows the physician to assess the course of treatment and regressions.
The development of monoclonal antibodies in the 1940's and 50's paved the way for accurate detection and quantification of individual molecules. These tests can be used to detect marker protein directly on tissue samples or to evaluate their concentrations in the blood stream. In the 1960's the detection of alpha-fetoprotein, a serum protein usually restricted to the normally developing fetus, was demonstrated to be an accurate tumor marker for gastrointestinal cancer. Since then, multiple tumor markers for many different anatomic systems have been discovered. These include not only tumor products, but elements inherent to the tumors themselves (oncogenes). (1) CA 125 is a tumor antigen initially noted to be elevated in patients with non-mucinous ovarian cancer (most ovarian cancers). (2) This protein could be identified in patients with ovarian cancer and seen to decrease after surgical removal of the tumor or successful chemotherapy. As importantly, early recurrence of ovarian cancer could be detected by rising levels of CA 125 - often before clinical evidence of recurrence was present. Unfortunately, people without cancer demonstrate a large variation in levels of CA 125 making the determination of what is normal difficult. In general women under 50 years of age have higher levels than women over 50 or men.(3) The finding of elevated CA 125 was noted to be elevated in number of conditions unrelated to ovarian disease, including gastro-intestinal cancer (specifically 63% of patients with pancreatic cancer) and in a number of benign conditions including infections, and endometriosis. Because of this, CA 125 levels are not used as a screening test to find unknown cancer but rather to follow the progress of documented cancer.
The utility of following CA 125 levels in pancreatic cancer has undergone significant scrutiny in the last decade. Most notably it has been compared with CA19-9, a tumor marker with proven prognostic value in the treatment of pancreatic cancer. In head to head comparisons CA 125 demonstrated both a lower ability to detect a problem (low sensitivity) and a higher likelihood of being abnormal in a healthy patient (low specificity).
If CA 125 has a present role in the treatment of pancreatic disease it appears to be in the differentiation of pancreatic cancer from pancreatitis. Haga showed that CA 125 level were apt to be elevated in pancreatic cancer but not in pancreatitis. When combined with the relative reliability of CA19-9 levels one may conclude both that a process is of pancreatic origin and is, or is not, malignant. (3) Once malignancy is proven there appears no additional benefit in following the CA 125 level. Even this use is presently disputed in patients with chronic pancreatitis.(4)
Preliminary evidence exists for using combinations of multiple tumor markers simultaneously to both detect and follow pancreatic cancer. Nakae at al report an algorithm using 9 tumor markers (including CA 125 and CA19-9) which appears to have both improved sensitivity and specificity for diagnosing pancreatic cancer when compared to CA19-9 alone. This has yet to be proven in a definitive fashion and whether it is better than following CA 19-9 for determining prognosis or detecting recurrence has not been examined. (4)
Based upon this information, it appears that your insurance carrier would have considered this data when deciding to decline payment for the CA 125 tests.
- Zinser JW. Tumor markers in cancer of the digestive system. Revista de Gastroenterologica 1997;62:145-8.
- Bast RC, Klug TL, St. John RN, Jenison E, Niloff JM, Lazerus H, Berkowitz RS, Leavit T, Griffiths CT, Parker L, Zurawski VR, Knapp RC. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. New England Journal of Medicine 1983;309:883-887.
- Haga Y. Sakamoto K. Egami H. Yoshimura R. Akagi M. Evaluation of serum CA-125 values in healthy individuals and pregnant women. American Journal of the Medical Sciences 1986;292(1):25-9.
- Hamori J, Arkosy P, Lenkey A, Sapy P. The role of different tumor markers in the early diagnosis and prognosis of pancreatic carcinoma and chronic pancreatitis. Acta Chirurgica Hungarica 1997;36:125-7.
- Nakae Y, Naruse S, Shibata T, Kitagawa M, Kondo T, Hayakawa T, Kumo N, Kurimoto K. Early detection of pancreatic cancer by serum markers. Japanese Journal of Clinical Pathology 1994;42:139-142.