MicroRNA profiling reveals potential biomarkers for the early transformation of endometriosis towards endometriosis-correlated ovarian cancer

Highlights

• •Peculiar miRNA signatures were associated with ovarian endometriosis (EMS-b) and endometriosis-correlated ovarian cancer (ECOC).
• •hsa-miR-10a-5p, hsa-miR-141–3p, hsa-miR-183–5p and hsa-miR-200a-3p were able to accurately differentiate EMS-b, endometriosis collected from patients with ECOC (EMS-k) and ECOC.
• •MiRNA profiling could improve early detection of malignant transformation of endometriosis.

Abstract

Background

Endometriosis (EMS) is a chronic, gynecological condition affecting 6–10 % of reproductive-age women. While these lesions are benign, ovarian EMS presents cancer-like features, and can progress to endometriosis-correlated ovarian cancer (ECOC) through a multistep process. Given the regulatory role of miRNAs in gene expression and biological pathways, we aimed to identify miRNAs associated with the malignant transformation of ovarian EMS, which could serve as a potential diagnostic tool for the early identification of such patients.

Methods

Global miRNA profiling was performed in 8 patients with benign ovarian EMS (EMS-b) and 29 patients with ECOC. Differential expression analysis (DEA) of miRNAs between EMS-b, EMS tissues from patients with ECOC (EMS-k) and ECOC tissues was performed. Receiver Operating Characteristic (ROC) curves were built to evaluate the binary classification performance of significant miRNAs.

Results

Comparison between EMS-b and EMS-k revealed 13 significantly deregulated miRNAs. Furthermore, when comparing ECOC and EMS-b, we observed significant deregulation of 181 miRNAs. ROC analysis revealed a panel of seven upregulated miRNAs with accuracies above 0.7 in identifying EMS-k and EMS-b. Notably, four miRNAs (hsa-miR-200a-3p, hsa-miR-141–3p, hsa-miR-183–5p, hsa-miR-10a-5p) were consistently upregulated in both EMS-k and ECOC tissues, achieving accuracies above 0.77 in distinguishing between EMS-k and EMS-b. When used to distinguish between EMS-b and ECOC tissues, these miRNAs showed accuracies even higher, above 0.94. Specifically, hsa-miR-183–5p had an accuracy of 1, hsa-miR-200a-3p and hsa-miR-141–3p of 0.97, while hsa-miR-10a-5p of 0.95.

Conclusions

Our study identified a panel of miRNA biomarkers that may serve as potential candidates for the early detection of ECOC in patients previously diagnosed with ovarian EMS.

Introduction

Endometriosis (EMS) is a chronic, disabling, inflammatory, and hormone-dependent gynecological disease that affects 6–10 % of reproductive age women [1] even if the true prevalence of EMS is still unknown [2]. This disease is characterized by endometrial tissues (gland and stroma) spreading from the uterus to the ovaries or to extra-ovarian sites resulting in endometriomas, superficial EMS or deep infiltrating EMS. Despite their benign nature, EMS lesions exhibit cancer-like features such as an increased ability to spread and infiltrate other tissues [3], resistance to apoptosis, stimulation of angiogenesis and ability to grow outside of the native uterine microenvironment. Women with EMS have an increased risk (2–10 %) of developing epithelial ovarian cancer compared to the general population, namely endometriosis-correlated ovarian cancer (ECOC), defined as the coexistence of EMS and cancer within the same ovary or the evidence of EMS transitioning to cancer through a multistep process according to the Sampson and Scott criteria [4,5].

A recent study [6] has highlighted that woman with EMS, particularly those with endometriomas and deep infiltrating EMS, are considered at higher risk for developing ovarian cancer: 16-fold higher for type I and 4-fold higher for type II ovarian cancer. In the general population, as well as in high-risk patients, such as BRCA mutation carriers, screening methods for ovarian cancer have proven ineffective, thus the only preventive option is risk-reducing surgery [7]. Moreover, to date, there are no age-specific guidelines for the timing of active surveillance in asymptomatic women with ovarian endometriomas or deep infiltrating EMS [8]. Ovaries are usually removed when there is an atypical appearance, an enlarged size, or notable changes on transvaginal ultrasound over time [9]. Currently, blood tests have little or no merit in the diagnosis of EMS cancerization and no biomarker-based tests have been translated to clinical practice. Even CA-125, which has a clinical value in ovarian cancer management, is not a significant predictor of malignant transformation of EMS [8,10].

MiRNAs, which are small, non-coding, single-stranded RNAs involved in genes regulation, have been identified as one of the key players in the pathogenesis of EMS [[11], [12], [13]]. Some of the altered miRNAs in endometriotic lesions are involved in several biological processes including progesterone resistance, altered cellular behaviors, epithelial-to-mesenchymal transition, angiogenesis, cell proliferation and cell adhesion and invasion [[11], [12], [13], [14]]. Given the high stability of miRNAs in terms of resistance to degradation (i.e. much higher than that of other RNA or DNA molecules), miRNAs could potentially serve as promising diagnostic biomarkers for women with EMS and uncertain imaging [15,16].

The aim of our study was to identify significantly deregulated miRNAs between patients with ECOC and those with benign ovarian EMS (hereafter labeled EMS-b) and to evaluate their potential as diagnostic biomarkers in discriminating between these two conditions. Moreover, we sought to identify whether ovarian EMS from patients who already developed ECOC (hereafter labeled EMS-k) displayed any significant pattern of miRNAs expression which might differentiate it from EMS-b.

Materials and methods

Study design and patient population

This was a single-center study funded by the Italian Ministry of Health (ENDO-2021–12,371,926) and registered on ClinicalTrials.gov (ID: NCT05161949). Ethical approval was obtained from the Ethics Committee under reference number CE 923/2021/Oss/AOUBo.

Patients with ECOC and EMS-b were enrolled between November 1, 2021, and December 30, 2023 at the Division of Oncologic Gynecology, IRCCS, Azienda Ospedaliero-Universitaria di Bologna, Italy. Inclusion criteria were women with: i) histologically proven ECOC arising from ovarian EMS according to Scott criteria [5] (study cohort); ii) histologically proven EMS-b (control cohort); iii) surgical resection performed at our center; iv) available material for RNA extraction. Exclusion criteria were: i) a documented history of cancer within the previous 5 years; ii) previous pelvic radiotherapy, or iii) systemic chemotherapy within the last 5 years prior to the study; iv) non-epithelial ovarian cancer (Fig. 1).

Fig. 1.

Study design and workflow. n: number of cases/samples; EMS-b: benign endometriosis; EMS-k: endometriosis from patients with ECOC; ECOC: endometriosis-correlated ovarian cancer.

miRNA isolation and analysis

Formalin-fixed paraffin-embedded (FFPE) samples were obtained from tissues removed during the patient's surgery. The histological diagnoses were made according to the WHO 2020 Classification of Female Genital Tumors [17]. Multiple serial sections of 3μM and 10 μM were sliced from the FFPE tissues collected during the patient's surgical procedure. The 3 μM sections were stained with hematoxylin and eosin and used by the dedicated pathologist to inspect and mark the regions of interest to be further dissected under microscopic guidance using a sterile blade for subsequent RNA extraction. This approach allowed for the identification of relevant areas where ovarian EMS or ECOC were present, ensuring a high purity of the samples selected for the analysis.

Total RNA was isolated from the FFPE sections using the Recover All™ total Nucleic Acid Isolation Kit (Thermo Fisher Scientific, Waltham, USA) following manufacturer's instructions. RNA quantities were evaluated using Qubit (Thermo Fisher Scientific, Waltham, USA). Samples for which at least 50 ng of RNA was obtained were selected for further processing and analysis.

miRNome profiling

Global miRNA profiling was evaluated for all samples. cDNA libraries were prepared using QIAseq miRNA Library Kit (Qiagen, Hilden, Germany.). Briefly, RNA libraries were prepared from 50 ng of RNA following the manufacturer's protocol. Libraries were assessed for quality and quantified using a High Sensitivity DNA Kit on a Tapestation system (Agilent, Santa Clara, United States); barcoded cDNA libraries were pooled together for a concentration of 4 nM and the effective concentration was re-checked on the same Agilent instrument. Samples were run on NextSeq 500 high-output (Illumina, San Diego, USA) with 75-bp single-end reads. The data obtained were analyzed using the Qiagen GeneGlobe miRNA sequencing pipelines. Briefly, the pipeline performs sample demultiplexing, adapter and low quality read trimming, mapping to the human transcriptome to identify the sequenced miRNA and counting the number of Unique Molecular Identifiers (UMIs) per miRNA to obtain the raw expression values in each sample. The expression matrix with the raw counts was exported and further processed and analyzed in R. Details regarding each sample reads which were obtained from the runs were summarized in Supplementary Table 1.

Differential expression analysis and candidate miRNAs identification

All analyses were conducted in R version 4.4.0 (2024–04–24 ucrt) - "Puppy Cup". Batch correction for different runs was done using the ComBat_seq function from the Bioconductor sva package [18]. Data normalization using a median ratio and differential expression analysis (DEA) of miRNAs between ovarian EMS and ECOC samples was evaluated using DESeq2 R-package [19]. The binary classification performance of miRNAs was evaluated using Receiver Operating Characteristic curves using the pROC package [20]. The optimal cut-off expression value was obtained using the Youden index. All p-values were corrected according to the Benjamini-Hochberg procedure (false discovery rate) [21]. Differentially expressed miRNAs were presented using Vulcano plots drawn with the EnhancedVolcano package [22].

Statistical analysis

For all variables, descriptive statistics were provided. Namely, nominal variables were presented as absolute frequencies and percentages while continuous variables were reported as median and interquartile range. MiRNA expression levels were plotted using boxplots, with the whiskers’ length set to 1.5 * interquartile range. Individual data points were also presented.

Results

A total of 37 patients were enrolled in this study. In detail, we included 8 (21 %) patients with EMS-b and 29 patients (79 %) with ECOC. Collection of ovarian EMS samples from ECOC patients was possible for 21 patients. General details of the cases included in the study can be found in Table 1.

Table 1.

Characteristics of the cases included in this study. n: number of cases; sd: standard deviation; EMS: endometriosis; ECOC: endometriosis-correlated ovarian cancer; BMI: body mass index; HRT: hormonal replacement therapy.

Variable		EMS-b n = 8	ECOC n = 29	p-value
Age at diagnosis, years, mean±sd		43.38±15.46	56.99±11.51	0.009
Age at menopause, years, mean±sd		51.5 ± 3.54	50.8 ± 3.07	0.764
Menopause, n(%)		2(25)	19(67.86)	0.030
Comorbidities, n(%)		3(37.5)	13(44.83)	0.711
History of endometriosis, n(%)		4(50)	7(24.14)	0.157
BMI, mean±sd		23.41±3.74	24.16±3.98	0.636
BMI class, n(%)	underweight	0(0)	1(3.45)	0.814
	normal	6(75)	17(58.62)
	overweight	1(12.5)	7(24.14)
	obese	1(12.5)	4(13.79)
Parous, n(%)		5(62.5)	20(68.97)	0.729
HRT, n(%)		0(0)	1(3.45)	0.594
Hormonal contraception, n(%)		6(75)	4(13.79)	0.001
History of cancer, n(%)		0(0)	5(17.24)	0.207
Familiar history of cancer, n(%)		3(75)	17(58.62)	0.530
CA-125 at diagnosis, U/mL, mean±sd		85.25±70.84	271.3 ± 484.04	0.291
CA-19.9 at diagnosis, U/mL, mean±sd		50.49±49.97	626.25±1958.22	0.416

MiRNA signature distinguishes benign endometriosis from endometriosis that progressed into ECOC

Differential expression analysis between EMS-k from ECOC patients and benign EMS-b yielded 13 deregulated miRNAs out of which 9 were upregulated (hsa-miR-10a-5p, hsa-miR-126–5p, hsa-miR-141–3p, hsa-miR-144–3p, hsa-miR-144–5p, hsa-miR-183–5p, hsa-miR-200a-3p, hsa-miR-205–5p and hsa-miR-451a) and 4 downregulated (hsa-miR-345–5p, hsa-miR-361–3p, hsa-miR-483–3p, hsa-miR-675–3p) (Fig. 2A, Supplementary Table 2). ROC curves built on the upregulated miRNAs revealed an accuracy above 0.7 in discriminating EMS-k from benign EMS-b in the case of 7 miRNAs (Fig. 2B, C, Supplementary Table 3).

Fig. 2.

A. Volcano plot showing the significantly deregulated miRNAs between EMS lesions from patients with ECOC (EMS-k) and benign EMS (EMS-b). B. ROC curves showing the accuracy of these miRNAs in discriminating between benign EMS (EMS-b) and EMS from ECOC patients (EMS-k). C. Expression levels of miRNAs significantly upregulated in EMS from ECOC patients. EMS-b: benign endometriosis; EMS-k: endometriosis tissue from patients with ECOC; ECOC: endometriosis-corelated ovarian cancer.

Diagnostic power of miRNAs significantly deregulated in ECOC

Differential expression analysis between EMS-b tissues obtained from patients with ovarian EMS and tumoral tissues from patients with ECOC revealed a total of 181 significantly deregulated miRNAs. Of these, 52 were downregulated and 129 were upregulated in ECOC tissues (Fig. 3A, Supplementary Table 4). We investigated whether any of these upregulated miRNAs could serve as potential diagnostic biomarkers in a clinical setting. For this analysis, ROC curves were built, and the results showed a wide range of sensitivities (0.31–1), specificities (0.5–1) and accuracies (0.46–1). Nevertheless, by selecting sensitivities of 1 and accuracies > 0.95, nine miRNAs exhibited excellent discrimination power in distinguishing EMS-b from ECOC samples (hsa-miR-183–5p, hsa-miR-429, hsa-miR-182–5p, hsa-miR-200c-5p, hsa-miR-200a-3p, hsa-miR-141–3p, hsa-miR-200b-3p, hsa-miR-200c-3p, hsa-miR-96–5p). Of these, hsa-miR-182–5p, hsa-miR-183–5p, hsa-miR-200c-5p and hsa-miR-429 were able to differentiate the two types of samples with excellent values of specificity, sensitivity, and accuracy of 1 (Fig. 3B, C, Supplementary Table 5). Moreover, we compared the diagnostic power of these miRNAs with the ones of CA-125 and CA-19.9 and showed that they were superior in identifying ECOC cases (p < 0.017 for all miRNAs) (Fig. 3B, Supplementary Table 6, 7).

Fig. 3.

A. Volcano plot showing the significantly deregulated miRNAs between ECOC and benign EMS. B. Expression levels of miRNAs significantly upregulated ECOC patients. C. ROC curves showing the accuracy of these miRNAs in discriminating between benign EMS patients and ECOC patients compared to the performance of currently used markers (CA-125 and CA-19.9). EMS: endometriosis; ECOC: endometriosis-correlated ovarian cancer.

Upon intersecting the datasets of upregulated miRNAs resulting from the EMS-k vs. EMS-b and ECOC vs. EMS-b comparisons, four miRNAs (hsa-miR-200a-3p, hsa-miR-141–3p, hsa-miR-183–5p, hsa-miR-10a-5p) were found to be consistently upregulated in both analyses (Fig. 4), showing increasing expression levels moving from EMS-b, EMS-k to ECOC. Moreover, these miRNAs had superior discriminatory capacity for distinguishing the three conditions. In detail, hsa-miR-200a-3p, hsa-miR-141–3p, hsa-miR-183–5p had discrimination accuracies of 0.8 in distinguishing EMS-k from EMS-b, while hsa-miR-10a-5p had an accuracy of 0.77 (Fig. 4, Supplementary Table 3). At higher thresholds these miRNAs showed accuracies above 0.94 in discriminating between EMS-b and ECOC tissues. Specifically, hsa-miR-183–5p had an accuracy of 1, hsa-miR-200a-3p and hsa-miR-141–3p of 0.97, while hsa-miR-10a-5p of 0.95 (Fig. 4, Supplementary Table 5).

Fig. 4.

ROC parameters for the diagnostic power of the miRNA signature with hsa-miR-200a-3p, hsa-miR-141–3p, hsa-miR-183–5p, hsa-miR-10a-5p); th: best discriminatory threshold; EMS-b: benign endometriosis; EMS-k: endometriosis from ECOC patients; Sp: specificity; Se: sensitivity; NPV: negative predictive value; PPV: positive predictive value; ECOC: endometriosis-correlated ovarian cancer.

Discussion

Main findings

To the best of our knowledge, this is the first study to identify significantly deregulated miRNAs able to discriminate between EMS-b, EMS-k and ECOC. In our analysis, we first compared benign EMS tissue (i.e. EMS-b) with the EMS associated with ovarian cancer (EMS-k). This comparison showed 7 miRNAs upregulated in EMS-k tissues, which had an accuracy higher than 0.7 in distinguishing the two types of samples. Second, we compared the miRNA expression in EMS-b and ECOC. As expected, this analysis showed the highest number of deregulated miRNAs, revealing 181 significant miRNAs. We then evaluated the potential diagnostic power in a clinical setting by building ROC curves; 8 miRNAs exhibited excellent discrimination power in recognizing EMS-b from ECOC samples with sensitivities of 1 and accuracies > 0.95. Interestingly, the discrimination power of these miRNAs resulted higher compared the typical biomarkers used in clinical setting i.e. CA125 and CA19.9. Finally, four miRNAs (hsa-miR-200a-3p, hsa-miR-141–3p, hsa-miR-183–5p, hsa-miR-10a-5p) showed consistent upregulation in both analyses, with increasing expression levels moving from EMS-b, EMS-k to ECOC.

The findings suggest that these miRNAs, which are involved in the carcinogenetic process of ECOC, are already dysregulated in the endometriotic tissues of patients diagnosed with ECOC. This early deregulation may play a role in the progression from EMS to malignancy, highlighting the potential of these miRNAs as biomarkers for early detection of malignant transformation of these patients. Most importantly, these miRNAs showed increased discrimination power to distinguish between EMS-k and EMS-b as well as ECOC and EMS-b.

Results in the context of current knowledge

So far, most studies have compared EMS with ovarian cancer, but limited literature is available regarding a direct comparison between EMS-b, EMS-k and ECOC. Some of the miRNAs highlighted in the present study are already known to be involved in ovarian cancer pathogenesis [[23], [24], [25], [26]]. Hsa-miR-200 and hsa-miR-429, in agreement with our data, are usually upregulated in this malignant tissue. These two miRNAs belong to the same family of miRNAs and have been reported to play a role in ovarian cancer development. The miR-200 family consists of five members, hsa-miR-200a, hsa-miR-200b, hsa-miR-200c, hsa-miR-141, and hsa-miR-429, divided into two clusters based on their seed sequence, which are crucial regulators of hallmarks of cancer. In particular, several studies highlighted that they have a critical role in regulating the epithelial-to-mesenchymal (EMT), an essential component of the invasive growth program of solid tumors. In a previous report [27], the overexpression of hsa-miR-200b/200a/429 caused the transformation and tumor formation of nontumorigenic ovarian epithelial cells, suggesting a direct role in ovarian tumor formation. Regarding hsa-miR-182 and hsa-miR-183, they are part of the same cluster (miR-183/182/96) and play a critical role during the pluripotent stem cell differentiation into sensory organs. Hsa-miR-183/182/96 is one of the upregulated clusters in ovarian cancer and is involved in modulating tumor growth, invasion, apoptosis and therapy resistance [28]. All these reports, together with our results, suggest that these miRNAs – showing lower levels in EMS-b - could then undergo a progressive increase in their expression leading to deep molecular changes able to contribute to the development of ovarian cancer.

Several studies have already indicated that miRNAs are implicated in the progression of EMS, but without focusing on their clinical potential [[23], [24], [25], [26]]. Here we identified nine miRNAs showing accuracies above 0.95 in distinguishing ECOC from EMS-b. It is worth discussing that these miRNAs demonstrated a great balance between sensitivity and specificity for this analysis, being superior to commonly used markers for the investigation of patients with ovarian cancer (such as CA-125 and CA-19.9). A similar approach was employed by Kumari et al. [29] who showed that hsa-miR-99b and hsa-miR-125a could discriminate EMS from endometrioid ovarian carcinoma and endometrioid endometrial cancer. However, their approach focused on analyzing a preselected list of miRNAs, and they did not demonstrate a relationship between EMS and tumor development in these patients. As a result, it remains unclear whether the patients' cancer was incidental or part of the malignant transformation of EMS as we recently discussed [30].

The key challenge is to validate these miRNAs in liquid biopsy samples, as this would establish their value as biomarkers. In this regard, Nakamura et al. [31] analyzed miRNA expression in blood and ascites samples from patients with EAOC and EMS, identifying significant deregulation in a panel of miRNAs. Notably, among the miRNAs identified in their study, miR-4484 was also found to be significantly upregulated in ECOC tissues in our analysis. This is an encouraging finding, as the upregulation of this miRNA in tumor tissue suggests that it may be secreted into the bloodstream/ascitic fluid and serve as a potential biomarker in liquid biopsy.

Although some improvements have been obtained by combining the diagnostic value of CA-125 with other markers such as human epididymis protein 4 (HE4) [[32], [33]], its usage as an indicator of EMS malignant transformation still remains limited as described by others [[34], [35]], and further confirmed by our study. Moreover, randomized-controlled trials have shown no benefit of serum CA-125 measurements or transvaginal ultrasound for the early detection of ovarian cancer or in reducing the mortality in the general population [36]. However, active monitoring is carried out for women at a high risk of developing ovarian cancer, such as those with a family history of ovarian or breast cancer, or a known pathogenic germline mutation such as in the BRCA1/2 genes. These women may have a risk of up to 50 % of developing ovarian cancer, compared to the 1.3 % risk in the general population. In some cases, prophylactic bilateral salpingo-oophorectomy is recommended to reduce this risk [37]. A 2021 systematic review and meta-analysis reported that women with EMS have nearly two times the risk of ovarian cancer compared to those without. This association is even stronger if we consider only clear cell (summary relative risk=3.44) or endometrioid histotypes (summary relative risk=2.33) [38]. Moreover, a recent study on 78,000 women with EMS showed a much higher risk than previously reported (HR=18.96 [95 % CI, 13.78–26.08]) in patients with deep infiltrating EMS [6]. Yet, current guidelines do not have specific recommendations for these women.

Study strengths and limitations

Our work has several strengths; first the prospective enrollment of the patients ensured the homogeneity of our cohort from the clinical, pathological, surgical and an analytical point of view which allowed us to limit multiple biases. Nevertheless, while it is a significant strength, it cannot control all potential biases. Specifically, the two study groups were not perfectly matched in terms of demographic variables, such as age. Although this mismatch may introduce a small bias in miRNA profiles, we believe that the changes induced by cancer itself are substantially greater and thus likely to dominate the variance caused by demographic differences between the two study populations. Another important limitation was that in some cases the collection of biological material was unfeasible, and in other cases, even with the available material, we were not able to obtain a sufficient yield from RNA isolation, resulting in the exclusion of six patients from the analysis. Moreover, it is important to note that the molecular alterations found were identified at the tissue level and may not be detectable in the patients’ blood, limiting their immediate applicability as diagnostic biomarkers for screening and/or diagnostic purposes.

Future perspectives and clinical implications

Currently, there is no reliable method to distinguish between patients who present EMS-b and those who present EMS-k which may eventually progress into ECOC. Transvaginal ultrasound is the primary tool used for the evaluation of these lesions but in some cases, it may fail to distinguish between EMS and ECOC [39], not to mention the distinction between EMS and EMS at risk of malignant transformation. The analysis of deregulated miRNAs between EMS-b and EMS-k from patients who had progressed to ECOC revealed significantly altered miRNAs, suggesting that, although both tissues appear histologically benign, the latter was different at a molecular level (Fig. 1a). Indeed, these miRNAs, particularly those belonging to the miR-200 family, have been identified as significant regulators of EMT , a pivotal biological process for the invasive and metastatic behavior of cancer cells. From a clinical point of view, these insights also suggest potential new biomarkers for early and non-invasive detection of malignant transformation in patients with EMS, through miRNAs detection in the bloodstream. Furthermore, targeting these miRNAs could lead to the development of novel therapeutic strategies aimed at preventing or slowing down the transition to ECOC. By validating the dysregulation of these miRNAs in liquid biopsy samples, it might be possible to monitor disease progression and adapt treatment plans accordingly, potentially leading to improved patient outcomes.

Conclusions

Our study revealed specific miRNAs that may play a key role in the initiation, promotion or progression from ovarian EMS to ECOC, and this has important clinical implications. MiRNAs exhibited higher capacity of discriminating patients with EMS-b from those with ECOC than that of currently used markers such as CA-125 and CA-19.9.

Ethics approval and consent to participate

Ethical approval was obtained from the Ethics Committee of Area Vasta Emilia Centro (AVEC) under reference number CE 923/2021/Oss/AOUBo. Informed consent was obtained from the patients involved in the study.

Novelty and impact

Peculiar miRNA signatures are specifically associated with benign ovarian endometriosis, and endometriosis-correlated ovarian cancer, with relevant implications for comprehension of ovarian carcinogenesis and personalized therapies. hsa-miR-10a-5p, hsa-miR-141–3p, hsa-miR-183–5p and hsa-miR-200a-3p were able to differentiate benign ovarian endometriosis (EMS-b) vs endometriosis collected from patients diagnosed with ECOC (EMS-k) and ECOC with excellent values of specificity, sensitivity, and accuracy. MiRNA profiling could improve early detection of malignant transformation of endometriosis in patients at risk.

CRediT authorship contribution statement

Gloria Ravegnini: Writing – review & editing, Writing – original draft, Software, Methodology, Investigation, Conceptualization. Camelia Alexandra Coadă: Writing – review & editing, Writing – original draft, Visualization, Software, Resources, Formal analysis, Data curation. Giulia Mantovani: Writing – review & editing, Validation. Antonio De Leo: Investigation. Dario de Biase: Investigation. Alessia Costantino: Investigation. Francesca Gorini: Investigation, Data curation. Giulia Dondi: Investigation, Data curation. Stella Di Costanzo: Resources, Investigation, Data curation. Francesco Mezzapesa: Data curation. Federico Manuel Giorgi: Supervision, Resources, Funding acquisition, Formal analysis. Giovanni Tallini: Supervision, Project administration. Sabrina Angelini: Supervision. Annalisa Astolfi: Resources. Lidia Strigari: Project administration, Funding acquisition. Pierandrea De Iaco: Supervision, Project administration, Funding acquisition. Anna Myriam Perrone: Writing – review & editing, Supervision, Methodology, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.