High ER-alpha predicts poor prognosis in women with deep endometriosis: a tissue micro-array study

Abstract

Endometriosis is regulated by sex steroid hormones through their receptors. Progressive lesion development and resistance to treatment may be mediated by aberrant steroid receptor regulation. Reliable markers of aggressive disease are lacking and few studies have investigated the relationship between steroid receptor expression and clinical symptoms. We analyzed receptor expression using aTissue Micro-Array (TMA) to determine if levels of estrogen receptor-α, progesterone receptor, androgen receptor, aromatase or CXCL12 in deep endometriosis that can predict symptom recurrence after surgery and response to the progestin treatment. Subjects who had undergone a complete surgical excision of symptomatic deep endometriosis (N=92) were prospectively enrolled in the study between 2013 and 2015. Exclusion criteria were: incomplete surgical excision, concomitant hysterectomy and/or bilateral oophorectomy and the use of GnRH analogs 6 month before surgery. The presence and severity of dysmenorrhea, dyspareunia, dyschezia and non-menstrual pain (NMP) were assessed using a Visual Analogue Scale. Recurrence of painful symptoms at one-year follow-up evaluation was recorded. A minimum of two tissue cylinders of 3 mm diameter were punched from each endometriosis tissue block. Before inclusion in the TMA, a representative slide from each paraffin block was analyzed by two pathologists, confirming endometriosis (glands and stroma). Endometriosis was identified in 83 subjects who had >2 samples from each block. Five-micrometer transverse sections were obtained, placed on a glass slide; immunochemical staining was carried out on all samples simultaneously and scored by two blinded observers. The intensity of expression of ER-α, PR, AR, aromatase and CXCL12 were correlated with progestin use and clinical outcome using chi square and linear regression. All subjects had deep lesions and a frozen pelvis was found in 41. Remarkably, 55% of women had moderate-to-severe NMP despite 46% using chronic oral progestin (MPA 2.5 mg/day or DNG 2 mg/day) treatment at the time of surgery. Pain Recurrence rate at one-year was 19%. Higher ER-α expression increased the likelihood of experiencing moderate to severe dysmenorrhea (p<0.04) and deep dyspareunia (p<0.03) at the time of surgery in women not receiving hormonal treatment. Progestin treatment was associated with diminished ER expression. In women receiving progestin therapy, persistently high ER-α expression increased the likelihood of deep dyspareunia, severe dischezia as well as pain recurrence at one-year (P<0.02). ER expression correlated with symptom severity in deep endometriosis. Failure of progestin therapy to lower ER predicted pain recurrence within one year. ER expression may serve as a prognostic biomarker of aggressive endometriosis. Estrogen modifying rather than progestin based therapies may be targeted to patients with high ER-α, allowing for precision medicine in endometriosis care.