Evaluation of CA125 in relation to pain symptoms among adolescents and young adult women with and without surgically-confirmed endometriosis

Endometriosis is a painful gynecologic disease affecting one in ten reproductive aged women worldwide [ 1 ]. Although over 50% of adults with endometriosis report onset of severe pelvic pain during adolescence, there is an average seven-year delay in diagnosis due to the current gold standard of surgical visualization for definitive diagnosis [ 2 – 4 ]. Presently, no reliable biomarkers exist for non-invasive diagnosis of endometriosis [ 5 , 6 ]. Though many attempts have been made to identify and validate specific biomarkers, studies have been conducted mostly in adult populations limited by poor methodological quality including small sample sizes, lack of consideration of disease heterogeneity, and inappropriate control groups [ 6 , 7 ]. Cancer Antigen125 (CA125) is a high molecular-weight glycoprotein normally expressed on tissues derived from the coelomic and mullerian epithelia including the uterus endometrium [ 8 ]. CA125 has been reported to be elevated in endometriosis patients and is the most commonly described, extensively studied endometriosis biomarker to date [ 6 , 9 – 13 ]. A recent report suggested that CA125 may be predictive of endometriosis in symptomatic women with gynecological pain and/or subfertility [ 10 ]. However, this study was based on only 30 endometriosis cases and 28 surgical controls with heterogeneous endometriosis and control subtypes. Patients with endometriosis have been shown to experience different levels of dysmenorrhea, non-cyclic/general pelvic pain, and dyspareunia, which are the most frequently reported pain outcomes among these women [ 9 ]. However, few studies have correlated this symptomatology with biomarker levels, and no studies correlating pain with biomarker levels have been performed in young patient populations. Endometriosis diagnosed in adolescence often presents with pain symptoms and superficial peritoneal lesions, which is often different from endometriosis diagnosed in adults [ 14 ]. Thus, it is important to understand biomarker performance in this population when aiming for earlier diagnosis of endometriosis. The objective of this study was to examine whether CA125 correlates with different types and severity of pain among adolescents and young women with and without endometriosis using data collected at enrollment from the longitudinal cohort of the Women’s Health Study: From Adolescence to Adulthood (A2A).

In total, 575 participants, including 282 laparoscopically-confirmed endometriosis cases and 293 controls, were included in the analyses ( Table 1 ). Median age at blood draw was 24 years in controls and 17 years in cases, with 68% and 89% non-Hispanic white, respectively. Most (81%) of the cases were using hormonal medication at blood draw compared to only 46% of the controls. The majority of cases (95%) were rASRM stage I or II. Overall, the average CA125 was 14.4 U/mL ranging from 3.6 to 117.8 U/mL. a Number of missings for characteristic variables: BMI (n = 3), smoking status (n = 12), race (n = 3), age at menarche (n = 2), menstrual cycle phase at blood draw (n = 90), hormonal medication use at time of blood draw (n = 82), Analgesic use at time of blood draw (n = 130), ASRM stage (n = 24), age at first endometriosis symptoms (n = 29). b Geometric mean (95%CI) adjusted for age (continuous). c For women aged ≥20 years: underweight (BMI < 18.5 kg/m 2 ), normal weight (BMI 18.5–24.9 kg/m 2 ), overweight (BMI 25–29.9 kg/m 2 ), or obese (BMI ≥ 30 kg/m 2 ) according to World Health Organization criteria; For those −2 to 2). d Among participants not on hormones at the time of blood draw. The average CA125 values were 12.5 (95%CI = 11.8–13.3) U/mL in controls and 12.1 (95%CI = 11.4–12.8) U/mL in cases (p = 0.96), and the association between demographic characteristics and CA125 values were similar in general regardless of disease status ( Table 1 ). Compared to those over age 30, younger age at blood draw was suggestively associated with lower CA125 values. Current smokers had lower CA125 values compared to former or never smokers, although there were only nine current smokers. Participants whose age at menarche was greater than 14 had higher CA125 values on average compared to those whose menarche started at a younger age. CA125 values were higher when blood was drawn during the follicular phase compared to other menstrual cycle phases. Regardless of whether the participants were diagnosed of endometriosis, those who were on hormonal medication at blood draw had lower CA125 values compared to those who were not. Analgesic use at time of blood draw was not associated with CA125 values. Among endometriosis cases, those diagnosed with advanced stage had a higher CA125 compared to those diagnosed with early stage disease. When we examined levels by endometriosis subtypes, those with endometrioma had the highest average CA125 values (51.7, 95%CI = 35.7–74.9 U/mL), although the sample size was limited to five cases. Endometriosis cases with superficial peritoneal lesions only (n = 256) and those with deep infiltrating endometriosis lesions only (n = 5) had lower CA125 values (11.7, 95%CI = 11.1–12.3 U/mL; 12.2, 95%CI = 8.5–17.5 U/mL respectively). Age at first endometriosis symptom was not associated with CA125. We examined the association between symptoms of pain type, severity, duration and CA125 values in endometriosis cases and controls ( Table 2 ). Overall, average CA125 values were similar across participants’ self-reported symptoms of ever experiencing dysmenorrhea, general pelvic pain, or dyspareunia. We did not observe a correlation between severity and frequency of these pain types and CA125 values among endometriosis cases or controls (p-trend >0.05). Since CA125 is strongly influenced by hormonal medication, we examined the association between pain symptoms and CA125 values among controls who were not on hormonal medication at time of blood draw ( S1 Table ). Controls reporting severe general pelvic pain was associated with higher CA125 (16.3, 95%CI = 11.3–23.6 U/mL) compared to those with mild general pelvic pain (13.4, 95%CI = 9.0–19.9 U/mL) although the trend was not statistically significant (p-value = 0.40). Interestingly, controls who reported never experiencing dyspareunia had higher CA125 (15.3, 95%CI = 13.5–17.4 U/mL) compared to those who ever experienced dyspareunia (13.6, 95%CI = 11.4–16.3 U/mL), with a significant trend by severity within 24 hours after vaginal intercourse (p-trend = 0.04). However, since there were only five control participants who reported severe dyspareunia, this observation may be due to chance. a Number of missings for pain variables: severity of period pain (n = 12), frequency of period pain (n = 14), ever experienced general pelvic pain (n = 6), severity of general pelvic pain (n = 11), ever dyspareunia (n = 1), severity of dyspareunia (n = 7), frequency of dyspareunia (n = 9). b Geometric mean (95%CI) adjusted for age (continuous). c Among participants who experienced period pain within the last 12 months. d Severity of pain was categorized based on the VAS scale: Mild (1–3), Moderate (4–6), Severe (7–10). e Among participants who answered the WERF EPHect version of the questionnaire. f Among participants who experienced general pelvic pain within the last 3 months. g Among participants aged 18 or older and reported ever having vaginal intercourse/penetration. h Among participants who reporting having dyspareunia at their current age range (16–20 years, 21–30 years, 31–40 years, 41+ years). i Among participants who reported having dyspareunia in the last 12 months. Among participants with self-reported dysmenorrhea at baseline, we evaluated the performance of CA125 in discriminating endometriosis cases from controls overall and by level of pain severity ( Fig 1 ). Overall, CA125 did not perform well using the clinical cutpoint of 35 U/mL (AUC = 0.51, 95%CI: 0.50–0.53) or 30 U/mL (AUC = 0.51, 95%CI: 0.49–0.53). When examining the performance of CA125 by the level of pain severity (mild, moderate, severe), the performance did not differ in discriminating endometriosis cases from controls using both cutpoints of 35 U/mL and 30 U/mL with AUCs ranging from 0.50 to 0.53. (A) Receiver Operating Curves (ROC) using CA125 cutoff of 35 U/mL and (B) using CA125 cutoff of 30 U/mL among participants presenting with dysmenorrhea (overall = solid line; mild pain = dashed line; moderate pain = dotted line; severe pain = dash-dotted line).

Our analyses suggest that CA125 values do not correlate with types of pain (i.e. dysmenorrhea, acyclic pain, and dyspareunia), or its severity or frequency among adolescents and young adult women. CA125 did not efficiently discriminate endometriosis cases from controls using the clinical cutpoints of 35 U/mL or 30 U/mL even when accounting for pain symptomatology in this young population. Average blood CA125 values were low in adolescents and young women in both endometriosis cases and controls, suggesting that cautious interpretation may be needed when measuring CA125 in adolescents and young women.

The Women’s Health Study: From Adolescence to Adulthood (A2A) is a study of adolescents and women oversampled for those surgically diagnosed with endometriosis, enrolling participants from 2012 to 2018 [ 14 ]. Briefly, endometriosis cases were enrolled from Boston Children’s Hospital (BCH) and Brigham and Women’s Hospital (BWH) and were eligible if they were 1) female; 2) aged 7–55 years; and 3) had a surgical diagnosis of endometriosis. Controls at enrollment were selected through a combination of approaches that sampled the underlying population that gave rise to the cases [ 15 ]. Specifically, controls were women without a surgical diagnosis of endometriosis and were recruited from clinics at BCH and BWH and from the local Boston community through local advertisement, online postings, and/or word of mouth. At the time of enrollment, participants were emailed a link to a REDCap survey containing extensive questions assessing lifestyle and reproductive factors as well as level of pain, treatment regimen, and medication use. Participants who did not respond to the email link after three follow-up attempts were mailed a paper copy of the questionnaire. Annual questionnaires were collected via the same methods. All questionnaires collected after January 2014 were World Endometriosis Research Foundation Endometriosis Phenome and Biobanking Harmonization Project (WERF EPHect) compliant [ 16 ], which are available online ( http://endometriosisfoundation.org/ephect ). Clinical information for A2A participants undergoing surgery at BCH or BWH were collected using the WERF EPHect surgical form, including surgically visualized appearance of peritoneal lesions, endometrioma, and deep infiltrating disease and endometriosis revised American Society for Reproductive Medicine (rASRM) stage [ 17 ]. This study was approved by the Institutional Review Board of the BWH (Partners Human Research Committee; 2015P001101). All participants provided written consent for study participation, with parental consent plus participant assent for girls < 18 years. Detailed information on symptoms of pain were obtained from the questionnaire completed by all A2A participants at enrollment of the study including whether the participant ever experienced pain with periods (dysmenorrhea), non-cyclic/general pelvic pain (pain not associated with menses), and dyspareunia in their lifetime [ 16 ]. Severity of dysmenorrhea, general pelvic pain, and dyspareunia was assessed using the visual analogue scale which is an 11-point numeric pain rating scale with 0 = no pain and 10 = worst imaginable pain [ 18 ]. Additionally, information on frequency of dysmenorrhea, general pelvic pain, and dyspareunia was collected. Blood samples were collected at enrollment in compliance with the standardized tools of WERF EPHect [ 16 , 17 , 19 , 20 ]. At the time of sample collection, participants completed a biospecimen questionnaire on which they reported date of last menstrual period, timing of last foods/beverages consumed, and recent medication and hormone use. Blood samples were collected, processed into plasma, serum and buffy coats, and stored at ≤-80°C per WERF EPHect fluids standard operating protocols [ 20 ]. CA125 was measured in 100μL of plasma using a Food and Drug Administration (FDA)-approved clinical chemiluminescent immunoassay (CA125 II) at the Clinical and Epidemiologic Laboratory at Boston Children’s Hospital. The assay reproducibility for CA125 on the E170 automated instrument was high with coefficient of variation in blinded duplicate samples of 1.15% [ 21 ]. Information on age, race, body mass index (BMI; kg/m 2 ), smoking status, age at menarche, menstrual cycle phase at time of blood draw, hormonal medication and analgesic use within 30 days from blood draw were abstracted from the questionnaire assessed at enrollment and/or biospecimen questionnaire data. For women aged ≥20 years, BMI was categorized according to the World Health Organization Criteria: underweight (BMI < 18.5 kg/m 2 ), normal weight (BMI 18.5–24.9 kg/m 2 ), overweight (BMI 25–29.9 kg/m 2 ), or obese (BMI ≥ 30 kg/m 2 ). For those −2 to 2). For endometriosis cases, information on rASRM stage and endometriosis subtype at surgery closest to blood draw was abstracted from the WERF-EPHect surgical form and self-reported age at first endometriosis symptoms was abstracted from the questionnaire obtained at enrollment of the study. Among the 656 participants (347 cases and 306 controls) who had a CA125 measurement for their blood samples collected at enrollment, we excluded participants who did not complete the questionnaire at enrollment (n = 13), completed the questionnaire at enrollment more than 60 days before/after their blood draw at enrollment (n = 59), were enrolled as a control at enrollment, but were diagnosed with endometriosis within 2 years after enrollment (n = 1), and were premenarchal at enrollment (n = 5) for a final analytic sample size of 282 surgically confirmed endometriosis cases and 293 controls. CA125 values were log-transformed to improve normality. We calculated the age-adjusted geometric means and 95% confidence intervals (CIs) of CA125 for all covariates and pain variables described above. Participants were excluded from analyses where they were missing the main exposure. Trend tests were calculated by modeling the exposure variable categories as ordinal adjusted for continuous age. Logistic regression analysis was used to calculate the Area Under the Curve (AUC) and 95% CIs in discriminating endometriosis cases from controls among participants who reported dysmenorrhea by pain levels using thresholds of 30 U/mL and 35 U/mL. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

(DOCX) Click here for additional data file.