The review, “Diagnostic accuracy of cancer antigen 125 for endometriosis: a systematic review and meta-analysis,” was published in BJOG.
Endometriosis, a painful, chronic condition that affects approximately 6.3 million females in the U.S. and millions more worldwide, happens when tissue which normally lines the uterus (the endometrium) grows outside the uterus, usually in growths around the ovaries, fallopian tubes, and ligaments that support the uterus. Diagnosis is considered uncertain until proven by a laparoscopy, a surgical procedure that shows the location, size, and extent of the growths.
CA 125 is a glycoprotein antigen normally expressed in tissues derived from coelomic epithelia (ovaries, fallopian tubes, peritoneum, pleura, pericardium, colon, kidney, and stomach). Serum CA 125 is elevated in about 80 percent of women with advanced epithelial ovarian cancer, but elevated levels have also been associated with endometriosis. Measuring CA 125 in women with suspected endometriosis could help to confirm the condition and eliminate the need for surgical exploration.
In the review, Martin Hirsch at the Women’s Health Research Unit, Barts and The London School of Medicine and Dentistry, and his colleagues searched databases for studies evaluating the diagnostic accuracy of serum CA 125 for confirmed endometriosis. Twenty-two studies involving 3,626 patients whose endometriosis was diagnosed surgically were identified.
The stage of endometriosis was defined based on the American Fertility Society classification. Subgroup analyses were performed based on mild (stage 1 or 2) or advanced (stage 3 or 4) endometriosis.
Fourteen studies were included in the meta-analysis. Using data from those studies, the researchers tested the predictive accuracy using a cut-off of 30 IU/mL in serum CA 125 for the diagnosis of endometriosis. They found that sensitivity was 52.4%, while the specificity was 92.7%. In mild endometriosis, the sensitivity was 24.8%. In stage 3 and 4 cases, sensititivity was 63.1%.
Dr. Peter Kovacs, MD, PhD, recently presented his opinion of the study. His comment was published as a Viewpoint attached to the previously published study report in BJOG.
Kovacs maintained that the positive, noninvasive screening could be used merely to select patients for additional diagnostic steps that could include laparoscopy to confirm endometriosis, assess its stage, and decide on the best therapeutic route.
But, referring to the overall 52.4% sensitivity with a cut-off of 30 IU/mL, he wrote: “… this cut-off falsely identifies a good proportion of patients as having endometriosis. In addition, a negative test does not rule out endometriosis; therefore one is still left with the dilemma of whether to proceed with surgical exploration or to try empirical treatment for the given symptom.”