Serum HSF1 is upregulated in endometriosis patients and serves as a potential diagnostic biomarker

This prospective study included 18–50 years‐old females who underwent laparoscopic surgery due to suspected EMS, pelvic pain of unknown origin, adnexal cysts, infertility, or uterine myomas. The participants were divided into two groups based on laparoscopic visual diagnosis and histological confirmation: a control group and an EMS group. Patients who received hormone therapy within 3 months before surgery, had a history of malignant diseases, current infections or inflammation, or systemic autoimmune diseases were excluded from the study. A detailed questionnaire was administered to all patients to assess EMS‐related pain symptoms and obtain a detailed patient disease history, resulting in a well‐characterized patient cohort. The presence of EMS was confirmed by laparoscopic and histological analyses and was further classified into stages according to the revised guidelines of the American Society for Reproductive Medicine (ASRM). 18 Patients who showed no EMS lesion detected by laparoscopic assessment were included in the control group. The menstrual cycle stage was assessed on the day of surgery by histological analysis of endometrial biopsies from women undergoing diagnostic curettage or, in cases where diagnostic curettage was not available, based on the patients' reported menstrual cycle duration and the day after the last menstruation. Between April 2020 and April 2022, a total of 144 eligible females participated in the study, of whom 92 were diagnosed with EMS and the remaining 52 females comprised the control group. Our study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of our hospital. Prior to the initiation of general anesthesia, a 5 mL sample of peripheral venous blood (PB) was collected from each participant and immediately transferred to a sterile silicone tube. The PB samples were kept on ice during transportation to the laboratory where they were centrifuged at 450 × g for 7 min. The resulting serum was isolated from the PB samples and stored at 40°C until thawed for subsequent assays. The time from blood collection to laboratory transfer for qRT–PCR detection was no more than 12 h. Total RNA was isolated from serum using TRIzol reagent (Takara, Tokyo, Japan), followed by reverse transcription into complementary DNA using a PrimeScript RT kit (Takara). Expression of HSF1 was determined by qRT–PCR using an SYBR Premix Ex Taq II kit (Takara), with GAPDH as the control gene. The 2 −ΔΔCt method was used to calculate gene expression. The qPCR primers used in this study are listed in Table  1 . All testing results are shown in Table  S1 . qPCR primers. Serum CA125 levels were determined using an ELISA kit (MBS2020370, MyBiosource, San Diego, CA, USA) in strict accordance with the provided instructions. The assays were performed by researchers who were blinded to the study's purpose. To identify potential downstream target genes of the human transcription factor HSF1, we utilized two databases, AnimalRFDB3.0 ( http://bioinfo.life.hust.edu.cn/AnimalTFDB/#!/ ) 19 and hTFtarget ( http://bioinfo.life.hust.edu.cn/hTFtarget#!/ ). 20 We downloaded the EMS dataset GSE25628 , which includes eight cases of ectopic endometria from EMS patients and 6 cases of endometria from healthy women, from the Gene Expression Omnibus (GEO) database. Using the R package limma, we conducted differential analysis on the ectopic endometria and normal endometria samples of GSE25628 . Genes with |log2‐fold change (FC)| > 1.5 and a p value 1.5 were defined as upregulated genes, and those with log2 FC≤ − 1.5 were classified as downregulated genes. Volcano plots and heatmaps were generated using the R packages “ggplot2” and “pheatmap.” SPSS version 18.0 software (SPSS Inc., Armonk, NY, USA) and GraphPad Prism 8.0 software (GraphPad Software Inc., San Diego, CA, USA) were utilized for data analysis and graphing. The normal distribution was assessed using the Shapiro–Wilk test. Measurement data are presented as the mean ± standard deviation (SD) and were compared using the t test. Enumeration data are presented as case numbers and were compared using the chi‐square test. Receiver operating characteristic (ROC) analysis was conducted to evaluate the diagnostic accuracy of serum HSF1 for EMS patients. Logistic regression analysis was employed to assess the associations between HSF1 and clinical characteristics of EMS patients, which are presented as adjusted odds ratios (ORs). First, univariate analysis was performed for each variable, and variables with p  < 0.1 underwent multivariate logistic regression analysis, using the Enter method for variable selection. p values were attained from the two‐sided test. In all statistical references, a value of p  < 0.05 was suggestive of statistically significant differences.

We enrolled a total of 144 females aged of 18–50 years. Among them, 92 were diagnosed with EMS (44 cases of ASRM classification I/II EMS, 48 cases of ASRM classification III/IV EMS); the control group consisted of 52 females. The clinical characteristics of the study participants are shown in Table  2 . There were no significant differences in parameters such as age, body mass index (BMI), age at menarche, and menstrual cycle between the two groups ( p  > 0.05). However, there were significant differences between the two groups in terms of dysmenorrhea, dyspareunia, pelvic pain, nulliparity, and CA125 expression ( p  < 0.05). Clinical characteristics of the study groups. Note : Age, BMI, age at menarche, menstrual cycle, and CA125 were analyzed by the independent t test. Dysmenorrhea, dyspareunia, pelvic pain, and nulliparity were analyzed by chi‐square test. p  < 0.05 indicated differences with statistical significance. Abbreviations: ASRM, American Society for Reproductive Medicine; BMI, body mass index; CA125, carbohydrate antigen 125. Serum HSF1 expression was measured in both the control and EMS groups, and the results showed that HSF1 was notably upregulated in the serum of EMS patients ( P  < 0.05, Figure  1A ) and its expression also increased with an increase in the ASRM classification level ( p  < 0.05, Figure  1B ). Based on ROC analysis, the cutoff value of serum HSF1 in EMS was 1.12 (area under the curve [AUC]: 0.857), with a sensitivity and specificity of 91.30% and 63.46%, respectively ( p  < 0.05, Figure  1C ). These results suggest that HSF1 may serve as a reliable biomarker for the diagnosis of EMS. Determination of serum HSF1 and its diagnostic value. (A, B) Serum HSF1 expression in different populations was determined by qRT–PCR, Control, n  = 52, Endometriosis (EMS), n  = 92, ASRM I/II: n  = 44, ASRM III/IV: n  = 48. C: ROC analysis which showing the cutoff value of serum HSF1 in EMS of 1.12 (AUC: 0.857), sensitivity of 91.30%, specificity of 63.46%. Data in Panels A and B were analyzed by the independent t test. ** p  < 0.01. To further investigate the correlation between HSF1 and EMS‐related symptoms, we conducted dichotomy analysis in both the control population and EMS patients using the mean values of age (years), BMI (kg/mm 2 ), age at menarche (years), menstrual cycle (days), and serum HSF1 expression as critical values. In addition, patients were dichotomized based on the presence or absence of dysmenorrhea, dyspareunia, pelvic pain, and nulliparity. These variables were then subjected to univariate logistic regression analysis (Figure  2A ), and those with p  < 0.1 were included in the multivariate logistic regression analysis. After adjustment, the results showed that dysmenorrhea, dyspareunia, nulliparity, and serum HSF1 were independent risk factors for EMS (Figure  2B ). Correlation analysis of serum HSF1 and clinical features of endometriosis (EMS) patients. The control population and EMS patients were divided based on the means of age (years), BMI (kg/mm 2 ), age at menarche (years), menstrual cycle (days), and serum HSF1 expression serving as critical values and conducted in patients with or without dysmenorrhea, dyspareunia, pelvic pain, and nulliparity. (A) Association between each variable and EMS was analyzed by univariate logistic regression analysis; (B) Variables with p  < 0.1 were subjected to multivariate logistic regression analysis, which revealed dysmenorrhea, dyspareunia, nulliparity, and serum HSF1 as independent risk factors for EMS. p  < 0.05 indicated statistically significant differences. We conducted further analysis to explore the relationship between serum HSF1 expression and the severity of EMS. Patients were divided into two groups according to the ASRM classification: the ASRM I/II group and the ASRM III/IV group. Univariate logistic regression analysis (Figure  3A ) was performed to investigate the association between serum HSF1 expression and EMS severity. Variables with P  < 0.1 were included in multivariate logistic regression analysis. After adjustment, the results revealed that dysmenorrhea and serum HSF1 were independent risk factors for the severity of EMS (Figure  3B ). Correlation analysis between serum HSF1 and clinical features of endometriosis (EMS) patients with different severities. The control population and EMS patients were divided based on the means of age (years), BMI (kg/mm 2 ), age at menarche (years), menstrual cycle (days), and serum HSF1 expression serving as critical values and conducted in patients with or without dysmenorrhea, dyspareunia, pelvic pain, and nulliparity. (A) The association between each variable and ASRM classification of EMS was analyzed by univariate logistic regression analysis. (B) Variables with p  < 0.1 were subjected to multivariate logistic regression analysis, which revealed dysmenorrhea and serum HSF1 as independent risk factors for ASRM classification of EMS. p  < 0.05 indicated statistically significant differences. To investigate the potential regulatory mechanism and function of HSF1, we used the AnimalRFDB3.0 and hTFtarget online databases to identify potential downstream factors of HSF1 and identified the intersection (Figure  4A ). In addition, we downloaded the EMS dataset GSE25628 , which provides gene expression data in ectopic and healthy endometria, from GEO database and identified DEGs in EMS (Figure  4B ). Through comparison between genes in the intersection of Figure  4A and DEGs in GSE25628 , prostaglandin E synthase enzyme 3 (PTGES3), heat shock protein 90 alpha family class A member 1 (HSP90AA1), and heat shock protein family B (small) member 1 (HSPB1) showed significantly differential expression in EMS (Figure  4B , Table  3 ). Among them, PTGES3 and HSP90AA1 were downregulated in EMS and HSPB1 upregulated. We hypothesized that the regulatory role of HSF1 in EMS may be associated with PTGES3, HSP90AA1, and HSPB1, and further studies should focus on these three genes. Downstream target genes of HSF1. (A) Venn diagram showing the intersection of HSF1 downstream target genes predicted by the AnimalRFDB3.0 database and hTFtarget database. (B) Volcano plot revealing differentially expressed genes (DEGs) of endometriosis (EMS) in the GSE25628 dataset. Upregulated genes are shown in red, downregulated genes are shown in green, and undifferentiated genes are shown in black. The labeled gene names are from the genes in the intersection of panel A. The downstream target genes of HSF1 that have aberrant expression in EMS. Abbreviations: adj. P .Val, adjust P .Value; HSP90AA1, heat shock protein 90 alpha family class A member 1; HSPB1, heat shock protein family B (small) member 1; logFC, log2 fold change; PTGES3, prostaglandin E synthase 3.

The diagnostic process for EMS is prolonged and complicated, with many women experiencing delays of 8–12 years before receiving a correct diagnosis. 4 Diagnostic laparoscopy has several limitations, hence the need for noninvasive and auxiliary tests that are sensitive and specific. While several differentially expressed TFs have been found to be involved in the pathogenesis of EMS, 6 less is known about the role of TFs in the diagnosis and treatment of the condition compared to microRNAs, lncRNAs or circRNAs. As the principal transcription factor regulating HSR and facilitating the progression of EMS, HSF1 represents a novel and promising biomarker for the disease. 15 , 16 Our study is the first to analyze the clinical significance of serum HSF1 and explore its downstream targets. Dysmenorrhea, dyspareunia, pelvic pain, and infertility are common symptoms of EMS and can serve as predictors for its diagnosis. 21 CA125, a glycoprotein secreted by mesothelial cells, has been found to be elevated both in ovarian tumors and benign conditions, such as EMS. 22 Consistent with previous studies, we observed significant differences in dysmenorrhea, dyspareunia, pelvic pain, nulliparity, and CA125 levels among EMS patients and non‐EMS individuals, highlighting their potential role as diagnostic indicators for the consideration of diagnostic laparoscopy. HSF1 is a transcription factor that plays a vital role in the HSR and is activated by various cellular stresses to protect against damage to proteins. In EMS, HSF1 upregulation is associated with cellular damaging signals, such as ovarian tumor B1, 16 and is linked to malignant transformation of the endometrium, 23 which is similar to the pathogenesis of EMS. Our study included EMS patients with upregulated serum HSF1, and we found that its increase was associated with the ASRM classification level. Therefore, HSF1 may serve as a promising target for noninvasive and auxiliary tests for EMS diagnosis and treatment. In summary, documentation of symptoms such as dysmenorrhea, dyspareunia, pelvic pain, and nulliparity is vital for detailed anamnesis, and CA125 and HSF1 can be used as diagnostic markers to improve the efficiency of diagnostic laparoscopy. Although there is no widely used biomarker for EMS diagnosis, several potential biomarkers such as miRNAs, lncRNAs, inflammatory cytokines, oxidative stress markers, and glycoproteins, have been identified in previous studies. 24 , 25 , 26 However, their diagnostic accuracy has been evaluated in small sized samples. In our study, we found serum HSF1 > 1.12 to be a reliable diagnostic biomarker for EMS (AUC: 0.857, sensitivity: 91.30%, specificity: 63.46%). Overall, severity is an essential factor in determining the need for laparoscopic surgery. 3 Our study demonstrated serum HSF1, dysmenorrhea, dyspareunia, and nulliparous to be independent risk factors for EMS, with serum HSF1 and dysmenorrhea being associated with the severity of EMS. HSF1 was found to be the top‐performing biomarker, similar to other ROC validated biomarkers, such as CA125 (AUC: 0.867, sensitivity: 81.75, specificity: 84.67), 27 glycodelin A (AUC: 0.96, sensitivity: 91.7, specificity: 75.0) and IL‐6 (AUC: 0.88, sensitivity: 93.8, specificity: 80.0). 28 In general, the diagnostic efficacy of using a combination of biomarkers is superior relative to a single biomarker, for instance, the combination of HGB, CA199, CA125, and HE4 (AUC: 0.900 versus 0.747–0.867) 27 and miR‐17, IL‐4, and IL‐6 (AUC: 0.84 versus 0.75). 29 Therefore, combination of HSF1 with other effective biomarkers may be a promising approach for noninvasive diagnosis of EMS. As a transcription factor, HSF1 plays a critical role in regulating the expression of a reservoir of HSR proteins and tumorigenesis‐related signaling pathways. 11 , 30 According to our analysis, PTGES3, HSP90AA1, and HSPB1 are not only DEGs for EMS in the GSE25628 dataset but also downstream targets of HSF1. PTGES3, a member of the prostaglandin enzyme family, has been previously reported as a cochaperone of heat shock protein 90 (HSP90) 31 and shown to exert oncogenic effects in lung adenocarcinoma and prostate cancer. 32 , 33 HSP90AA1, a member of the HSP90 family, is a potential molecular target for cancer treatment. 34 The HSF1‐HSPB1 pathway is well known for its role in maintaining cancer cell survival, 35 and HSPB1 has also been shown to be upregulated in preeclampsia. 36 However, their expression patterns and roles in EMS remain unclear. Our findings suggest that PTGES3, HSP90AA1, and HSPB1 are involved in the downstream mechanism of HSF1 in EMS, but further studies are warranted to validate their roles in EMS pathogenesis.

In conclusion, our study provides evidence that serum HSF1 is upregulated in EMS patients and is an independent risk factor with excellent diagnostic value. However, the study did not investigate HSF1 expression in tissues or its location. Furthermore, limitations, such as the unknown form of mRNA in serum (in free form or in any component such as RNA complex or extracellular vesicles), and mRNA predisposition to degradation in serum, may affect our interpretations. Future studies should address these limitations and expand the sample size to better understand the clinical significance of HSF1 in EMS.

Endometriosis (EMS) is a prevalent gynecological disorder characterized by the presence of endometrium‐like tissue outside the uterus, affecting 5%–10% of reproductive‐aged women worldwide. 1 The pathogenesis of EMS is estrogen‐dependent, and its clinical presentations vary, ranging from pelvic pain and infertility to bowel symptoms (blood in stool, diarrhea, or constipation) and chronic fatigue. 2 Although laparoscopy is regarded as the gold standard for EMS diagnosis, it can be challenging to perform due to negative exams or the requirement of general anesthesia. 3 Additionally, visual diagnosis can be challenging due to the heterogeneity in lesion location and appearance. Therefore, a more easily accessible and noninvasive diagnostic method is required for prompt diagnosis and treatment. However, there are currently no validated serum biomarkers for the diagnosis of EMS. 4 Transcription factors (TFs) are essential regulators of gene expression and play a critical role in the pathogenesis of several human diseases by modulating various signaling pathways. 5 Emerging evidence supports the role of altered TF activity in the pathogenesis of EMS. 6 For example, hypoxia‐inducible factor‐1α, homeobox A10, and octamer‐binding transcription factor 4 are differentially expressed in the EMS endometrium and serum, influencing cellular behaviors that regulate EMS progression. 7 , 8 , 9 The heat shock response (HSR) is a conserved cytoprotective mechanism that counteracts stress‐induced protein denaturation. 10 Heat shock transcription factor 1 (HSF1) is a major regulator of HSR, with the ability to relieve stress and refold proteins. 11 Although EMS is a benign disease, it shares many features with cancer, such as excessive proliferation and metastasis. 12 HSF1 has been implicated in tumorigenesis and predicting cancer prognosis (e.g., cervical squamous cell carcinoma, endocervical adenocarcinoma, neck squamous cell carcinoma), 13 , 14 suggesting its potential involvement in EMS pathogenesis. Moreover, HSF1 has been shown to facilitate the growth and mobility of endometrial cells, 15 , 16 further supporting its regulatory role in EMS progression. Notably, HSF1 is overexpressed in the EMS endometrium, 17 but there are currently no relevant data on serum HSF1 concentration and its clinical significance in EMS. The primary objective of our study was to investigate the feasibility of using HSF1 as a diagnostic marker and identify novel therapeutic targets. In this study, we aimed to determine serum HSF1 expression in EMS patients, evaluate its potential role in EMS diagnosis, explore its association with disease severity, and analyze its downstream targets and functions.

All authors declare no conflict of interest.

Table S1. qRT‐PCR data of HSF1 (GAPDH as internal reference).