Combined magnetic resonance imaging with serum CA125 for dysmenorrhea in adenomyosis

The baseline characteristics, which included age, body mass index (BMI), serum CA125, uterine position and volumes, lesion position and volumes, distribution of dysmenorrhea and menorrhagia and symptom recurrence were summarized in Table 1 . Representative imaging characteristics of C4 classification were shown in Fig.  1 . Table 1 Clinicopathological features in total cohort. Variable Total cohort Age (years) 43.8 ± 5.2 BMI (kg/m 2 ) 23.6 ± 2.8 Parity 1 (1, 2) Uterus position (n(%))  Anteverted 292 (58.2%)  Retroverted 192 (38.2%)  Neutral 18 (3.6%)  Uterus volume(cm 3 ) 231.3 ± 124.5 Adenomyosis location (n(%))  Anterior 148 (29.5%)  Posterior 263 (52.4%)  Fundus 38(7.6%)  Lateral (left or right) 27(5.4%)  Anterior/posterior/fundus/lateral 26(4.2%)  Lesion volume (cm 3 ) 54.6 ± 56.8  Subcutaneous fat thickness (mm) 19.1 ± 8.2  Distance (mm) 22.0 ± 14.8  Serum CA125 (U/ml) 55.9 (30.0, 106.4) Dysmenorrhea (n(%)  Yes 381(75.9%)  No 121(24.1%) Menorrhagia (n(%))  Yes 272 (54.2%)  No 230 (45.8%) Symptom recurrence (n(%))  Yes 133 (26.5%)  No 369 (73.5%) BMI, body mass index; CA125, cancer antigen 125; distance, distance from the posterior surface of the adenomyosis lesion to the sacrum. Fig. 1 Representative imaging characteristics of C4 classification were shown in Fig. 1. Clinicopathological features in total cohort. BMI, body mass index; CA125, cancer antigen 125; distance, distance from the posterior surface of the adenomyosis lesion to the sacrum. Representative imaging characteristics of C4 classification were shown in Fig. 1. Table 2 showed χ2 value in C4, considered as the suitable one for CA125, was higher than those in other classification criteria (χ2 = 22.003, P  < 0.001). Besides, no matter intrinsic or external subtype, serum CA125 of Group A was higher than that of Group B (Z = 22.003, P  < 0.001). In C1, serum CA125 levels in subtype-2 (60.0 U/ml (30.2 U/ml, 124.8U/ml)) were higher than those in subtype-1 (52.2 U/ml (31.1 U/ml, 81.3 U/ml), P  = 0.007). In C2, serum CA125 levels in subtype-4 (106 U/ml (47 U/ml, 182.5 U/ml) were higher than those in other subtypes ( P  = 0.006). In C5, serum CA125 in subtype-6 was higher when compared with that in other subtypes ( P  = 0.006). There existed no significant differences for CA125 in C3 ( P  > 0.05) . Table 2 The identification of different classifications based on MRI for serum CA125. Variable Subtype-1 Subtype-2 Subtype-3 Subtype-4 Subtype-5 Subtype-6 χ 2 P -value Classification 1 17 52.2 (31.1, 81.3) 60.0 (30.2, 124.8) 7.380 0.007 Classification 2 13 55 (27.5,98.5) 57(30.3,104.8) 45(24,67.5) 106(47,182.5) 12.395 0.006 Classification 3 14 55 (28.8,108.5) 45(28.0,69.0) 60(33,123) 4.444 0.108 Classification 4 15 20 (12.0,42.0) 33.5(13.7,61.8) 55(30,108) 93(51-) 35(26.3,67.8) 59(34,123) 22.003  < 0.001 Classification 5 16 50.5 (27.0,93.5) 72.5(23.3,121.8 58(31,108) 45(29.5,67.5) 94(41.8, 173) 123(51,213.5) 16.265 0.006 * P  < 0.05. The identification of different classifications based on MRI for serum CA125. * P  < 0.05. According to the results of priority study, dysmenorrhea and menorrhagia rates were different upon the size of uterine wall based on C4. Patients were included into the lesion size of uterine wall ≥ 2/3 group (Group A, N = 446, 88.8%, and 68% failure with GnRH-a or levonorgestrel within 2 years) and the lesion size of uterine wall < 2/3 group (Group B, N = 56,11.2%, and 35.7% failure with GnRH-a or levonorgestrel within 2 years). The uterine volumes (236.9 ± 125.6 cm3), lesion volumes (56.1 ± 55.7 cm3), symptom recurrence(28%) and CA125 57.9 U/ml (31.1 U/ml, 111.1 U/ml) in Group A were higher than those in Group B ( P  < 0.001, Table 3 ). Table 3 Patient characteristics in subtype groups. Variable MRI-based severity group (n = 446, 88.8%) MRI-based non-severity group (n = 56, 11.2%) Test value P -value Age (years) 43.9 ± 5.2 42.6 ± 5.6 1.777 0.076 BMI (kg/m 2 ) 23.6 ± 2.8 24.1 ± 2.8 − 1.337 0.182 Parity 1 (1, 2) 1 (1, 1) − 0.901 0.368 Uterus position 1.115 0.572  Anteverted 263(59.0%) 29 (51.8%)  Retroverted 167 (37.4%) 25 (44.6%)  Neutral 16 (3.6%) 2 (3.6%)  Uterine volume (cm 3 ) 236.9 ± 125.6 185.0 ± 105.0 2.889 0.004  Adenomyosis location 2.117 0.714  Anterior 135 (30.3%) 13(23.2%)  Posterior 231 (51.8%) 32(12.2%)  Fundus 33 (7.4%) 5 (8.9%)  Lateral (left or right) 25(5.6%) 2 (3.6%)  Anterior/posterior/undus/lateral 22(4.9%) 4 (7.2%)  Adenomyotic lesion (cm 3 ) 56.1 ± 55.7 32.3 ± 51.6 3.067 0.003 Subcutaneous fat thickness (mm) 19.2 ± 8.0 20.5 ± 15.3 − 0.637 0.527  Distance (mm) 23.9 ± 14.7 25.1 ± 15.7 − 0.504 0.614  Hemoglobin (g/dl) 97.7 ± 19.3 100.2 ± 24.7 − 0.397 0.694  CA125 (U/ml) 57.9(31.1, 111.1) 35.1(16.8, 67.0) − 3.236 0.001 Dysmenorrhea (n(%)) 2.227 0.136  Yes 343(76.9%) 38(67.9%)  No 103 (23.1%) 18(32.1%) Menorrhagia (n(%)) 2.333 0.127  Yes 247(55.4%) 23(44.2%)  No 199 (44.6%) 29(55.8%) Symptom recurrence (n(%)) 9.883 0.002  Yes 125 (28.0%) 8 (15.4%)  No 321 (72.0%) 66 (84.6%) Group A, the lesion size of uterine wall ≥ 2/3; Group B, the lesion size of uterine wall < 2/3; BMI, body mass index; CA125, cancer antigen 125; distance from the posterior surface of the adenomyosis lesion to the sacrum. Patient characteristics in subtype groups. Group A, the lesion size of uterine wall ≥ 2/3; Group B, the lesion size of uterine wall < 2/3; BMI, body mass index; CA125, cancer antigen 125; distance from the posterior surface of the adenomyosis lesion to the sacrum. We further explored the relationship between CA125 and dysmenorrhea in different groups, as shown in Fig.  2 . CA125 levels in patients suffering from dysmenorrhea were higher when compared with those in patients without dysmenorrhea in total cohort(Z = -3.597, P  < 0.001, Fig.  2 A) and Group A (Z = -3.150, P  = 0.002, Fig.  2 B), while no significant difference was found in Group B (Z = -1.350, P  = 0.177, Fig.  2 C). In terms of those with dysmenorrhea, CA125 levels of Group A were higher than those in Group B (Z = -2.503, P  = 0.012). Fig. 2 CA125 level in dysmenorrhea and non-dysmenorrhea group in different subgroups. CA125 levels in patients with dysmenorrhea were higher than those in patients without dysmenorrhea in cohort(Z = − 3.597, P  < 0.001, Fig ( A ) and the volume of uterine wall ≥ 2/3 subgroup (Z = − 3.150, P  = 0.002, Fig ( B ), while no significant difference the volume of uterine wall < 2/3 subgroup(Z = − 1.350, P  = 0.177, Fig ( C ). For those with dysmenorrhea, CA125 levels of the volume of uterine wall ≥ 2/3 subgroup were higher than those the volume of uterine wall < 2/3 subgroup (***Z = − 2.503, P  = 0.012). CA125 level in dysmenorrhea and non-dysmenorrhea group in different subgroups. CA125 levels in patients with dysmenorrhea were higher than those in patients without dysmenorrhea in cohort(Z = − 3.597, P  < 0.001, Fig ( A ) and the volume of uterine wall ≥ 2/3 subgroup (Z = − 3.150, P  = 0.002, Fig ( B ), while no significant difference the volume of uterine wall < 2/3 subgroup(Z = − 1.350, P  = 0.177, Fig ( C ). For those with dysmenorrhea, CA125 levels of the volume of uterine wall ≥ 2/3 subgroup were higher than those the volume of uterine wall < 2/3 subgroup (***Z = − 2.503, P  = 0.012). In total cohort, univariate logistic regression demonstrated that age (OR = 0.909, 95% CI 0.869 ~ 0.951, P  < 0.001), CA125 (OR = 1.004, 95% CI 1.001 ~ 1.017, P  = 0.012) and subcutaneous fat thickness (OR = 0.97, 95% CI 0.94–0.99, P  = 0.007) were related to dysmenorrhea of patients with adenomyosis. However, there were no associations between BMI, the lesion size of uterine wall (≥ 2/3 vs.  0.05). According to multivariate logistic regression, age (OR = 0.900, 95% CI 0.855 ~ 0.947, P  = 0.007), and CA125 (OR = 1.004, 95% CI 1.001 ~ 1.007, P  = 0.018) were correlated with dysmenorrhea of patients with adenomyosis, while subcutaneous fat thickness showed no significant association with dysmenorrhea ( P  = 0.073) (Table 4 ). Table 4 Binary logistic regression analysis of factors impacting dysmenorrhea of adenomyosis in total cohort. Variable Univariate Multivariate OR 95%CI P OR 95%CI P Age (years) 0.909 0.869 ~ 0.951  < 0.001 0.900 0.855 ~ 0.947  < 0.001 BMI (kg/m 2 ) 0.940 0.873 ~ 1.012 0.098 1.015 0.927 ~ 1.111 0.753 Menorrhagia 1.162 0.774 ~ 1.748 0.468 1.144 0.727 ~ 1.801 0.561 (Yes vs. No) The volume of uterine wall 1.577 0.864 ~ 2.881 0.138 1.299 0.644 ~ 2.622 0.465 (≥ 2/3 vs. < 2/3) Serum CA125 (U/ml) 1.004 1.001 ~ 1.017 0.012 1.004 1.001 ~ 1.007 0.018 Subcutaneous fat thickness (mm) 0.967 0.944 ~ 0.991 0.007 0.975 0.948 ~ 1.002 0.073 CA125, cancer antigen 125; BMI, body mass index. Binary logistic regression analysis of factors impacting dysmenorrhea of adenomyosis in total cohort. CA125, cancer antigen 125; BMI, body mass index. In severity group (Group A) based on MRI, multivariate logistic regression showed that age (OR = 0.897, 95% CI:0.849 ~ 0.948, P  < 0.001) and CA125 (OR = 1.004, 95% CI 1.001 ~ 1.008, P  = 0.021) were related to dysmenorrhea of patients with adenomyosis, while did not significantly associated with dysmenorrhea in Group B ( P  = 0.073) (Table 5 ). Table 5 Binary logistic regression analysis of factors for dysmenorrhea of adenomyosis in subgroups. Variable The volume of uterine wall ≥ 2/3(Group A) The volume of uterine wall < 2/3(Group B) OR 95%CI P OR 95%CI P Age (years) 0.897 0.849 ~ 0.948  < 0.001 0.887 0.763 ~ 1.031 0.119 BMI (kg/m 2 ) 1.019 0.924 ~ 1.124 0.701 0.983 0.750 ~ 1.289 0.903 Menorrhagia 1.272 0.783 ~ 2.066 0.331 0.494 0.123 ~ 1.985 0.320 (Yes vs. No) Serum CA125 (U/ml) 1.004 1.001 ~ 1.008 0.021 0.999 0.987 ~ 1.012 0.908 Subcutaneous fat thickness (mm) 0.976 0.943 ~ 1.011 0.173 0.967 0.914 ~ 1.023 0.246 CA125, cancer antigen 125; Group A, the lesion size of uterine wall ≥ 2/3; Group B, the lesion size of uterine wall < 2/3; BMI, body mass index. Binary logistic regression analysis of factors for dysmenorrhea of adenomyosis in subgroups. CA125, cancer antigen 125; Group A, the lesion size of uterine wall ≥ 2/3; Group B, the lesion size of uterine wall < 2/3; BMI, body mass index. ROC curves were employed to assess the performance of CA125 in predicting dysmenorrhea in the total cohort and Group A. In the total cohort, the optimal cutoff value for CA125 was 44.8 U/mL, with a sensitivity of 58.3%, specificity of 62.0%, and an area under the curve (AUC) of 0.601 (95% CI: 0.556–0.644; P  = 0.006). In Group A, the optimal cutoff remained 44.8 U/mL, yielding a sensitivity of 59.2%, specificity of 59.2%, and an AUC of 0.594 (95% CI: 0.547–0.640; P  = 0.002). While no significant difference was observed in the non-severity subgroup. Then, we investigated the factors causing the increased CA125 levels in MRI-based severity group by multivariate logistic regression. Based on the results, the lesion volume positively related to elevated CA125 levels (OR = 1.016, 95%: 1.008–1.025, P  < 0.001), while the ages exhibited a negative correlation with elevated CA125 levels (OR = 0.943, 95%: 0.898–0.991, P  = 0.021) (Table 6 ). Table 6 Multivariate binary logistic regression analysis of factors for CA125 in MRI-based severity group. Variable B S.E Wald OR 95%CI P Age (years) − 0.058 0.025 5.342 0.943 0.898–0.991 0.021 BMI (kg/m2) − 0.037 0.046 0.643 0.964 0.880–1.055 0.423 Menorrhagia − 0.137 0.236 0.336 0.872 0.549–1.386 0.562 (Yes vs. No) Uterine volume (cm 3 ) 0.000 0.001 0.058 1.000 0.998–1.003 0.810 Lesion volume (cm 3 ) 0.016 0.004 15.408 1.016 1.008–1.025  < 0.001 Subcutaneous fat thickness (mm) 0.007 0.017 0.195 1.007 0.975–1.041 0.659 CA125, cancer antigen 125; BMI, body mass index. Multivariate binary logistic regression analysis of factors for CA125 in MRI-based severity group. CA125, cancer antigen 125; BMI, body mass index.

This retrospective study involved 502 patients, who were diagnosed as adenomyosis by MRI scan and treated by FUAS for dysmenorrhea or menorrhagia. Patients who had complete demographic and disease characteristics data were extracted from medical records in the Affiliated Nanchong Central Hospital of North Sichuan Medical College from June 2017 to March 2021. The study was conducted in accordance with the Declaration of Helsinki and performed in accordance with relevant guidelines/regulations. Approved by ethics committee of the Affiliated Nanchong Central Hospital of North Sichuan Medical College (permit number 2021/104), the study waived the need of obtaining informed consent due to the retrospective nature. The inclusion criteria for patients were patients with dysmenorrhea or menorrhagia secondary to adenomyosis, diagnosed by MRI before FUAS, received CA125 test and MRI examination before FUAS, and performed by FUAS 7 , 8 . The exclusion criteria for patients were those receiving no pre-FUAS CA125 test or MRI examination, with uterine fibroids or endometriosis or suspected malignant tumor in MRI imaging, which might lead to dysmenorrhea or menorrhagia, with inflammatory disease within a week, as well as a history of medicine for dysmenorrhea within 6 months, influencing the symptoms investigation before FUAS 7 , 8 . Adenomyosis was diagnosed by imaging physicians based on MRI examination, which was defined as the maximal junctional zone thickness (JZmax) max ≥ 12 mm) 18 , 19 , or the ratio of JZmax to the entire myometrium thickness > 40% 14 , or the difference between the maximal and minimal JZ thickness > 5 mm 18 , 20 . The lesion size of uterine wall was defined as the size of lesion invasive into uterine wall. The uterine volume and the adenomyotic lesion volume were calculated using the following formula: 4/3πABC, where A, B, and C represent the long diameter, wide diameter, and thickness diameter of the lesion, respectively 2 . As patient-reported pain during menstrual cycles, dysmenorrhea secondary to adenomyosis was acquired from medical records, and recorded in the clinical encounter before FUAS and follow-up 2 , 21 . Menorrhagia secondary to adenomyosis was set as the clinically recorded increasing patient-reported menstrual blood volume, and recorded in the clinical encounter before FUAS and follow-up 22 . Symptom recurrence was set as patient-reported dysmenorrhea and/or menorrhagia 12 months after a period when symptomatic relief lasted for at least 3 months after FUAS 2 . CA125 test samples of collecting and measurement followed the standard procedure, which were transported to the laboratory center at room temperature, measured by two-step immunoassay for the quantitative determination with flexible assay protocols within 24 h before FUAS, and assayed in a central laboratory 7 , 8 . Imaging system by a 1.5 T MRI system informed the images information of pelvic-enhanced MRI patients. Based on the professional opinion, adenomyosis was also divided according to ultrasound-based classification system (C1, the morphological features of C1 were also assessable by MRI imaging 14 , 17 ) and the MRI findings (C2-C5 13,14–17 ). Based on the five classification criteria, imaging physicians (ZJ, 10 years of experience; JZ, with 10 years of experience) and a gynecologist (YT, with 11 years of experience) were individually invited to reclassify adenomyosis 2 . Different classification standard criteria are listed in Supplementary Table 1. The classification parameters were used if the results were consistent between two readers, the third reader was invited to discuss the final classification parameters in the case of ambiguous or controversial results 2 . Preparation before and during the FUAS procedure conformed to the guidelines of the Focused Ultrasound Tumor Therapeutic System (Model-JC200, Chongqing Haifu Medical Technology Co., Ltd, Chongqing, China) 2 . To consolidate the effectiveness of FUAS in the treatment of adenomyosis, gonadotropin-releasing hormone agonists (GnRH-a)/ the levonorgestrel were suggested for patients without fertility requirements 2 . While after FUAS, we followed up patients via MRI and questionnaire survey to know whether they had dysmenorrhea and/or menorrhagia relief or recurrence on site or telephones by patient-reported return of dysmenorrhea/menorrhagia after ≥ 3 months of post-FUAS relief, which could support some information for FUAS outcomes and whether combined with suspected malignant tumor. While the symptom after FUAS relapse, GnRH-a or levonorgestrel intrauterine system or retreated with FUAS might be suggested for patients. While the imaging based on MRI suspected malignant tumor, hysterectomy or resection of lesion might be suggested according to the age and whether hoping to preserve fertility. Analysis were performed in total cohort and stratified by the lesion size of uterine wall based on MRI: Group A, the lesion size of uterine wall ≥ 2/3, MRI-based severity group; Group B, the lesion size of uterine wall < 2/3, MRI-based non-severity group. Continuous variables were summarized by the mean and standard deviation or median (P25, P75). In addition, interquartile range (IQR) and categorical variables were shown to be counts and percentages. This study adopted χ2 test, t test or Mann-Whiney U test for examining the differences of demographic and clinical factors between patients with and without dysmenorrhea in total cohort and subgroups. A logistic regression mode was used to explore the correlation between clinicopathological features (age, the lesion volume, the uterine volume, subcutaneous fat thickness, serum CA125) and dysmenorrhea by odds ratios (ORs) with 95% confidence intervals (95% CIs) in total cohort and subgroups. Using a Receiver Operating Characteristic (ROC) curve, the optimal cut-off of CA125 levels for dysmenorrhea before FUAS in adenomyosis were investigated. The logistic regression model was used for establishing the associations between elevated CA125 and dysmenorrhea before FUAS in adenomyosis in total cohort and subtype groups. To investigate the factors (age, the lesion size) influencing CA125 level before FUAS, multivariate logistic regression model was employed. GraphPad Prism 7.0 (Version X; La Jolla, CA, USA) was used for image editing. SPSS 22.0 (IBM, Armonk, NY) was adopted for the statistical analysis, and P  < 0.05 was thought to be of significance. The data was reviewed by the Data Surveillance and Monitoring Committee of the State Key Laboratory of Ultrasound in Medicine and Engineering. This committee is consisted of multidisciplinary experts in the clinical research.

Our study showed that C4 based on MRI was the suitable classification criterion for CA125 in patients with adenomyosis before FUAS. Uterine volumes, lesion volumes, CA125 levels and symptom recurrence after FUAS in MRI-based severity group were shown to be higher than those of the non-severity group. For those with dysmenorrhea, CA125 levels in MRI-based severity group were shown to be higher than those of the non-severity group, while they showed no significant difference in those without dysmenorrhea. CA125 level of 44.8 U/ml was found to be the optimal cut-off point of dysmenorrhea in severity group based on MRI. When we extended the study to investigate the factors related to the elevated CA125 in MRI-based severity group, the lesion volume was considered as the positively one, while the ages negatively one. Consist with the results that adenomyosis is actually related to expression of uterine inflammatory mediators and cytokines 6 , 7 , including tumor necrosis factor-α, β (TNF-α, β), interleukin (IL)-1β, IL-18 and CA125, our results again demonstrated that elevated CA125 level was associated with dysmenorrhea secondary to adenomyosis in patients without uterine fibroids and endometriosis 7 . It indicated the higher level of serum CA125, the higher possibilities of dysmenorrhea secondary to adenomyosis 7 . Classification criterion based on MRI also could inform the severity of dysmenorrhea. In the previous study, we found the severity disease was different considered the lesion size of uterine wall. In this study, we found that the C4 was the suitable classification criterion for CA125. Furthermore, we explored the association between CA125 and dysmenorrhea based on C4. While the lesion size of uterine wall ≥ 2/3, CA125 was positively associated with the severity of dysmenorrhea. In the previous study, we confirmed the commonly used diagnosis value, and 35 U/ml was the optimal cut-point for dysmenorrhea 23 . In this study, we further investigated CA125 as 48.4 U/ml was the optimal cut-point for dysmenorrhea in the patients whose lesion size of uterine wall ≥ 2/3. Meanwhile, no matter what is the cut-off point of CA125 for dysmenorrhea, elevated CA125 level was positively associated with the lesion volume, and negatively associated with the age. While our cohort focused on perimenopausal women, recent evidence highlights adenomyosis in adolescents with distinct diagnostic challenges (e.g., non-invasive imaging criteria) and management priorities (symptom control and fertility preservation 24 . The coexistence of adenomyosis and endometriosis in adolescents further complicates diagnosis and management, underscoring the need for comprehensive evaluation. As the main symptom of adenomyosis, dysmenorrhea is probably the results of inflammation, neurogenesis, angiogenesis, and contractile abnormalities in the endometrial and myometrial components 6 . Dysmenorrhea can be adjusted by abnormal genetic, including CYP1A1 and A2, catechol-O-methyltransferas, Cytochrome P450, lipoxygenase-5 and Cyclooxygenase-2 6 , 25 , 26 , involving the key processes in adenomyosis development. Dysmenorrhea could be illuminated by myometrial hypercontractility, and indicated by higher expression of oxytocin receptors and increased contractile amplitude of uterine smooth muscle cells (uSMCs) in adenomyotic uteri, resulting in hyperestrogenism, progesterone resistance, and inflammatory microenvironment 6 . Then, inflammatory microenviroment promoted the ectopic endometrium secretion of significantly higher CA125 level compared with the normal endometrium 27 . The high expression of IL-1β, CRH, and UCN observed in adenomyotic lesions may mediate prostaglandins synthesis and stimulate peritoneal irritation and peritoneal stretch, and therefore CA125 reached the blood circulation through changing endothelial permeability 6 , 28 . While elevated CA125 levels primarily reflect the inflammatory processes associated with adenomyosis, potential confounding factors such as subclinical infections or lesion degeneration must be considered 29 . In the present study, we implemented strict exclusion criteria to eliminate these confounding variables by excluding patients with concurrent infections or evidence of lesion degeneration. Besides, Hiratsuka D et al. indicated that advanced and extrinsic adenomyosis were potential predictors for progesterone resistance for ademoyosis 30 . Furthermore, our study attempted to explore the combined value of MRI and CA125 for dysmenorrhea secondary to adenomyosis. The results showed that when the MRI imaging informing the invasion wall ≥ 2/3, the larger areas of lesion, the higher levels of CA125 and higher recurrence after FUAS in adenomyosis. Meanwhile, CA125 may aggravate inflammation through promoting ectopic endometrium migration and adhesion in the surrounding myometrium in adenomyosis development 9 , which might result in a relatively severity level of the disease. While the MRI imaging informing the invasion lesion of uterus wall, the patients have higher possibilities of dysmenorrhea before FUAS. These findings might offer clues for investigating the pathogenesis of dysmenorrhea based on MRI and CA125. The suitable classification criterion based on MRI for CA125 might be beneficial for exploring the possible mechanisms with disease development and informing clues for evaluating FUAS outcomes. The strength of the current retrospective study was that we informed a suitable classification criterion based on MRI for CA125, further showing that CA125 was associated with dysmenorrhea secondary to adenomyosis upon severity disease based on MRI with a relatively large sample size. This might be of interest to researchers to study pathogenesis of dysmenorrhea based on MRI and CA125. However, this study still had the following limitations. Firstly, patients included in this study were those who received FUAS, which had potential limitation for selection bias and might limit the interpretation of our results to the general patient population. We will include the patients with non FUAS treatment and compare the predictive value of MRI and CA125 for clinic symptom of patients between the FUAS group and non-FUAS group in the future. Secondly, we could not conclude that CA125 was a risk factor of dysmenorrhea in MRI-based severity group in adenomyosis for the retrospective design. However, we demonstrated that CA125 is associated with dysmenorrhea secondary to adenomyosis in MRI-based severity group. Thirdly, although a few patients without complete data were excluded for the retrospective design, the clinicopathologic features of our cohort showed no significant difference from epidemiology of adenomyosis 31 . Finally, CA125 levels may be influenced by variations in sample collection and laboratory techniques, and the biomarker itself demonstrates only moderate predictive performance. To minimize potential preanalytical confounders, all CA125 samples in this study were collected and measured within 24 h prior to FUAS. Furthermore, in future studies, we plan to adopt the Numerical Rating Scale to quantify dysmenorrhea severity more precisely, rather than using a binary (yes/no) classification. We also intend to conduct well-designed prospective, multicenter validation studies to more comprehensively evaluate the predictive value of CA125 in adenomyosis..

In summary, this study informed that C4 based on MRI was the suitable one for CA125. Patients with increased CA125 > 44.8 U/ml were prone to suffer from dysmenorrhea secondary to adenomyosis in MRI-based severity group. Besides, the lesion volume positively related to elevated CA125 levels, while the ages exhibited a negative correlation with elevated CA125 levels. These findings might offer clues for investigating the pathogenesis of dysmenorrhea and exploring the possible mechanisms with disease progression based on MRI and CA125.

Adenomyosis is a commonly seen disorder of the uterus in women of child-bearing age, characterizing by the lesion invasion into uterus, present with dysmenorrhea, menorrhagia and larger uterus, affecting the quality of women 1 , 2 . Numerous patients with adenomyosis are prone to young and hope to preserve fertility 3 . Hysterectomy can effectively manage symptoms of adenomyosis while not suitable for young women. Recently, focused ultrasound ablation surgery (FUAS) has become a popular option for young patients for its good efficacy, safety, harmlessness, bloodlessness and no scar 2 , 4 , 5 . However, dysmenorrhea and/or menorrhagia may relapse within 12 months after FUAS 5 . Currently, the etiology and pathogenesis of adenomyosis and mechanism of disease progression after FUAS are not well known 6 , in which inflammation might be one main factor associated with dysmenorrhea secondary to adenomyosis. Cancer antigen 125 (CA125), a high macromolecule glycoprotein, is increasingly secreting from the embryonic coelomic epithelium to serum by peritoneal irritation or peritoneal stretch in inflammatory microenviroment 7 – 9 . It is mainly used as biomarker for investigating diagnosis and survival of ovarian cancer 10 , 11 . Our previous study has found that CA125 levels were closely correlated to dysmenorrhea in adenomyosis before focused ultrasound ablation surgery (FUAS), and CA125 levels in diffuse subtype was higher in relative to those in focal subtype, informing that CA125 might be one biomarker associated with disease severity for adenomyosis 7 . Magnetic resonance imaging (MRI) is considered as a moderately accurate test to diagnose and predict the disease severity for patients before FUAS 12 – 14 . Recently, researchers have explored the classifications of adenomyosis according to MRI 13 – 17 . Our previous study has found that based on the affected area, pattern, size (volume), localization of adenomyotic lesions, and concomitant pathology, classification criterion 4 (C4) was considered to be the suitable classification criterion for dysmenorrhea secondary to adenomyosis 15 . However, it is unclear which classification criterion is the suitable one for serum CA125 and whether classification criterion combined with CA125 could inform dysmenorrhea in adenomyosis. Thus, we retrospectively explored the data from a cohort of patients before FUAS, aiming to investigate the combined value of classification criterion and serum CA125 for dysmenorrhea, and study the factors influencing dysmenorrhea and the elevated CA125 level for investigating the possible pathogenesis of adenomyosis and mechanism of disease progression after FUAS.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1