SALT LAKE CITY – and might soon spare patients from years of misdiagnosis and the costs and burden of diagnostic surgery, according to experts at the annual meeting of the American Society for Reproductive Medicine.
Most notably, expression levels of three microRNAs, 125-b-5p, 451a, and 3613-5p, correctly distinguished patients with chronic endometriosis from healthy controls, said Hugh S. Taylor, MD, of Yale University in New Haven, Conn.
Dr. Hugh S. Taylor
Chronic endometriosis affects about 10% of women and up to half of those with infertility, Dr. Taylor noted. The disease costs the United States at least $22 billion annually and is the second most-common reason for hysterectomy (Reprod Sci. 2009 Apr;16[4]:335-46). Its complexity means that patients face many barriers to diagnosis, particularly young women, who are often told they have “routine” menstrual pain, Dr. Taylor said.
Patients may go up to 12 years and see five or more physicians before they are diagnosed. Clinicians tend to rely on surgical diagnosis, but “there is a reluctance to perform surgery unless there is severe disease,” he added. “The lack of nonsurgical biomarkers contributes significantly to delays in diagnosis and timely intervention.”
These concerns prompted Dr. Taylor and his associates to study microRNAs – the short, noncoding, functional RNAs that promote messenger RNA breakdown or repress its translation. MicroRNA expression varies by tissue type and disease status, and occurs in a variety of body fluids, giving them real potential as nonsurgical biomarkers, Dr. Taylor said. To examine their role in endometriosis, he and his associates performed microarray profiling and confirmatory quantitative real-time polymerase chain reaction testing of serum samples from 24 women with chronic endometriosis and 24 healthy women who served as controls (Fertil Steril. 2016 Aug;106[2]:402-9).
MicroRNA 125b-5p was upregulated the most in endometriosis patients and distinguished patients from controls with a “giant” area under the receiver operating characteristic curve value of 0.974, Dr. Taylor said. Remarkably, this value rose to 1 – meaning that sensitivity and specificity both were 100% – when the researchers added another upregulated microRNA (451a) and a downregulated microRNA (3613-5p) to the model. More work is underway to understand how a test for these microRNAs would perform in larger populations, Dr. Taylor said.
MicroRNAs also are likely to play functional roles in chronic endometriosis and may mediate treatment response, he noted. For example, the microRNA 125b-5p, which is upregulated in endometriosis, increases the expression of inflammatory cytokines and tumor necrosis factor alpha in macrophages, and the aromatase inhibitor letrozole, which reduces pelvic pain in the disease, increases the expression of let-7 microRNAs, with corresponding decreases in the migration of endometrial cells (Fertil Steril. 2016 Sep 1;106[3]:673-80). “Maybe these microRNAs are changing metabolism. Maybe they are changing immune cell activity,” Dr. Taylor said. “I think they are doing a lot more than sitting around waiting for us to discover them.”
None of the 24 patients with endometriosis in his study had taken oral contraceptives in the 3 months prior to serum sampling, Dr. Taylor noted. “We need to look at oral contraception as a potential confounder,” he acknowledged. “If something is independent of the menstrual cycle phase, that is much better than a marker that is dependent on cycle phase.”
Menstrual cycle phase and oral contraceptives are just two of many potential confounders of biomarkers in chronic endometriosis, according to Linda Giudice, MD, PhD, of the University of California, San Francisco. Disease severity, as well as the type, number, and location of lesions and the presence or absence of coexisting inflammatory disorders all can potentially affect the sensitivity and specificity of a marker, she said. Consequently, “there is no single biomarker for chronic endometriosis,” but there are several candidates besides microRNAs, she added. For example, studies show that menstrual blood is readily distinguishable from peripheral blood, and closely resembles the immune environment of the uterus. Another study found that urinary peptides distinguished patients with moderate to severe endometriosis from healthy controls, and mild disease from severe disease, with sensitivities and specificities ranging from 72% to 88% (Fertil Steril. 2011 Mar 15;95[4]:1261-6).
Other potential sources of diagnostic tests include the endometrial proteome, transcriptome, and methylome, as well as endometrial stem cells, Dr. Giudice said. But for now, surgical diagnosis remains the gold standard, and the World Endometriosis Research Foundation is working to homogenize recording of surgical phenotypic information and laparoscopic specimens to improve data quality, she added.
Dr. Taylor did not report funding sources. He disclosed financial ties to Pfizer, OvaScience AbbVie, Bayer, and Euroscreen. Dr. Giudice acknowledged support from the National Institutes of Health and the UCSF NIH Human Endometrial Tissue and DNA Bank. She disclosed ties to Merck, Pfizer, NextGen Jane, AbbVie, and Juniper Pharmaceuticals.