Comparing endometriotic lesions with eutopic endometrium: time to shift focus?

Artificial intelligence-based tissue segmentation and cell identification in multiplex-stained histological endometriosis sections

STUDY QUESTION

How can we best achieve tissue segmentation and cell counting of multichannel-stained endometriosis sections to understand tissue composition?

SUMMARY ANSWER

A combination of a machine learning-based tissue analysis software for tissue segmentation and a deep learning-based algorithm for segmentation-independent cell identification shows strong performance on the automated histological analysis of endometriosis sections.

WHAT IS KNOWN ALREADY

Endometriosis is characterized by the complex interplay of various cell types and exhibits great variation between patients and endometriosis subtypes.

STUDY DESIGN, SIZE, DURATION

Endometriosis tissue samples of eight patients of different subtypes were obtained during surgery.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Endometriosis tissue was formalin-fixed and paraffin-embedded before sectioning and staining by (multiplex) immunohistochemistry. A 6-plex immunofluorescence panel in combination with a nuclear stain was established following a standardized protocol. This panel enabled the distinction of different tissue structures and dividing cells. Artificial intelligence-based tissue and cell phenotyping were employed to automatically segment the various tissue structures and extract quantitative features.

MAIN RESULTS AND THE ROLE OF CHANCE

An endometriosis-specific multiplex panel comprised of PanCK, CD10, α-SMA, calretinin, CD45, Ki67, and DAPI enabled the distinction of tissue structures in endometriosis. Whereas a machine learning approach enabled a reliable segmentation of tissue substructure, for cell identification, the segmentation-free deep learning-based algorithm was superior.

LIMITATIONS, REASONS FOR CAUTION

The present analysis was conducted on a limited number of samples for method establishment. For further refinement, quantification of collagen-rich cell-free areas should be included which could further enhance the assessment of the extent of fibrotic changes. Moreover, the method should be applied to a larger number of samples to delineate subtype-specific differences.

WIDER IMPLICATIONS OF THE FINDINGS

We demonstrate the great potential of combining multiplex staining and cell phenotyping for endometriosis research. The optimization procedure of the multiplex panel was transferred from a cancer-related project, demonstrating the robustness of the procedure beyond the cancer context. This panel can be employed for larger batch analyses. Furthermore, we demonstrate that the deep learning-based approach is capable of performing cell phenotyping on tissue types that were not part of the training set underlining the potential of the method for heterogenous endometriosis samples.

STUDY FUNDING/COMPETING INTEREST(S)

All funding was provided through departmental funds. The authors declare no competing interests.

TRIAL REGISTRATION NUMBER

N/A.

PARP-1, EpCAM, and FRα as potential targets for intraoperative detection and delineation of endometriosis: a quantitative tissue expression analysis

BACKGROUND

Endometriosis is a gynecological disease characterized by the presence of endometrial tissue in abnormal locations, leading to severe symptoms, inflammation, pain, organ dysfunction, and infertility. Surgical removal of endometriosis lesions is crucial for improving pain and fertility outcomes, with the goal of complete lesion removal. This study aimed to analyze the location and expression patterns of poly (ADP-ribose) polymerase 1 (PARP-1), epithelial cell adhesion molecule (EpCAM), and folate receptor alpha (FRα) in endometriosis lesions and evaluate their potential for targeted imaging.

METHODS

Gene expression analysis was performed using the Turku endometriosis database (EndometDB). By immunohistochemistry, we investigated the presence and distribution of PARP-1, EpCAM, and FRα in endometriosis foci and adjacent tissue. We also applied an ad hoc platform for the analysis of images to perform a quantitative immunolocalization analysis. Double immunofluorescence analysis was carried out for PARP-1 and EpCAM, as well as for PARP-1 and FRα, to explore the expression of these combined markers within endometriosis foci and their potential simultaneous utilization in surgical treatment.

RESULTS

Gene expression analysis revealed that PARP-1, EpCAM, and FOLR1 (FRα gene) are more highly expressed in endometriotic lesions than in the peritoneum, which served as the control tissue. The results of the immunohistochemical study revealed a significant increase in the expression levels of all three biomarkers inside the endometriosis foci compared to the adjacent tissues. Additionally, the double immunofluorescence analysis consistently demonstrated the presence of PARP-1 in the nucleus and the expression of EpCAM and FRα in the cell membrane and cytoplasm.

CONCLUSION

Overall, these three markers demonstrate significant potential for effective imaging of endometriosis. In particular, the results emphasize the importance of PARP-1 expression as a possible indicator for distinguishing endometriotic lesions from adjacent tissue. PARP-1, as a potential biomarker for endometriosis, offers promising avenues for further investigation in terms of both pathophysiology and diagnostic-therapeutic approaches.

Distribution of smooth muscle actin and collagen in superficial peritoneal endometriotic lesions varies from the surrounding microenvironment

RESEARCH QUESTION

Do different subtypes of superficial peritoneal endometriotic lesions exist, based on the presence and morphology of smooth muscle, collagen fibres and immune cell populations?

DESIGN

A retrospective cohort study of 24 patients, from across the menstrual cycle, with surgically and histologically confirmed endometriosis. Immunofluorescence was used to delineate the CD10 stromal area of lesions (n = 271 lesions from 67 endometriotic biopsies), and then smooth muscle actin (SMA) positive tissue and immune cell populations (CD45+ and CD68+) were quantified within and adjacent to these lesions. Second harmonic generation microscopy was used to evaluate the presence and morphology of type-1 collagen fibres within and surrounding lesions.

RESULTS

Overall, immune cell numbers and the area of SMA and collagen within endometriotic lesions tended to be low, but a spectrum of presentations significantly varied, particularly in the adjacent tissue microenvironment, based on lesion locations, the morphology of endometriotic gland profiles, or both. Lesions in which collagen fibres formed well aligned capsules around the CD10+ stromal border were identified compared with lesions in which collagen fibre distribution was random. Considerable inter- and intra-patient variability in the morphology of SMA and collagen was observed within and surrounding lesions.

CONCLUSION

These data demonstrate considerable diversity in the presence of immune cells and morphology of SMA and collagen within, but even more so, surrounding endometriotic lesions, even within individual patients. This heterogeneity, especially within individual patients, presents a challenge to incorporating these cell and tissue types into any new endometriosis classification systems or prognostic approaches.

Extensive heterogeneity in the expression of steroid receptors in superficial peritoneal endometriotic lesions

RESEARCH QUESTION

Is the expression of steroid hormone receptors (oestrogen receptor-α and progesterone receptor A/B) and proliferative markers (Bcl-2 and Ki67) uniform among superficial peritoneal endometriotic lesions?

DESIGN

A retrospective cohort study of 24 patients with surgically and histologically confirmed endometriosis. Immunofluorescence was used to determine the proportion of oestrogen receptor-α (ERα), progesterone receptor A/B, Bcl-2 and Ki67 positive cells in 271 endometriotic lesions (defined as endometriotic gland profile/s within an individual region of CD10 stromal immunostaining from a single biopsy) from 67 endometriotic biopsies from 24 patients. Data were analysed to examine associations related to menstrual cycle stage, lesion location and gland morphology.

RESULTS

Oestrogen receptor-α and progesterone receptor A/B expression in superficial peritoneal endometriotic lesions was extremely heterogeneous. Bcl-2 immunostaining in endometriotic lesions was also variable, whereas Ki67 immunostaining was minimal. Menstrual cycle stage associations were limited in steroid hormone receptor and Bcl-2 expression in lesions. Patterns in progesterone receptor A/B and Bcl-2 immunostaining were associated with lesion location. Bcl-2 was differentially expressed, based on lesion gland morphology.

CONCLUSIONS

These data demonstrate considerable diversity in the expression of steroid hormone receptors and Bcl-2 between lesions, even within an individual patient.

Elevated phosphorylation of estrogen receptor α at serine-118 in ovarian endometrioma

OBJECTIVE

To evaluate the phosphorylation of estrogen receptor α at serine-118 (phospho-ERα S118) in the endometrium, ovarian endometrioma, and deep infiltrating endometriosis (DIE).

DESIGN

Experimental study.

SETTING

University-affiliated hospital and academic research laboratory.

PATIENT(S)

Twenty-five patients underwent a hysterectomy, 18 patients underwent surgical removal of ovarian endometrioma, and 6 patients underwent DIE.

INTERVENTION(S)

Tissue samples were obtained from patients who underwent surgical procedures.

MAIN OUTCOME MEASURE(S)

Immunostaining for phospho-ERα S118, ERα, or phosphorylated p44/42 mitogen-activated protein kinase (phospho-p44/42 MAPK) was performed to evaluate the endometrium with or without endometriosis, ovarian endometrioma, and DIE. For in vitro analysis, endometrial epithelial cells (Ishikawa cells) were stimulated with estradiol (E2) or tumor necrosis factor alpha (TNFα), and the expression levels of phospho-ERα S118 and phospho-p44/42 MAPK were evaluated via Western blotting.

RESULT(S)

First, phospho-ERα S118 level was significantly higher in the glands and stroma of ovarian endometriosis samples than in those of endometrial and DIE samples. Second, colocalization of phospho-p44/42 MAPK and phospho-ERα S118 was observed in the glands of ovarian endometrioma. The proportions of cells strongly expressing phospho-p44/42 and phospho-ERα were 87% in phosphor-p44/42 MAPK-positive cells and 79% in phosphor-ERα-positive cells. Third, E2 stimulation significantly enhanced phospho-ERα S118 after 15 and 30 minutes in in vitro analysis using endometrial epithelial cells. Fourth, TNFα stimulation modestly but significantly enhanced phospho-ERα S118 after 15 and 30 minutes. Fifth, in Ishikawa cells, treatment with a p44/42 inhibitor (PD98059) significantly reduced phospho-ERα S118 by TNFα but not by E2.

CONCLUSION(S)

ERα-S118 phosphorylation was increased in ovarian endometriosis. Our findings may provide a new perspective for understanding the mechanism of increased ERα action in the pathophysiology of endometriosis.

Computer-Aided Histopathological Characterisation of Endometriosis Lesions

Endometriosis is a common gynaecological condition characterised by the growth of endometrial tissue outside the uterus and is associated with pain and infertility. Currently, the gold standard for endometriosis diagnosis is laparoscopic excision and histological identification of endometrial epithelial and stromal cells. There is, however, currently no known association between the histological appearance, size, morphology, or subtype of endometriosis and disease prognosis. In this study, we used histopathological software to identify and quantify the number of endometrial epithelial and stromal cells within excised endometriotic lesions and assess the relationship between the cell contents and lesion subtypes. Prior to surgery for suspected endometriosis, patients provided menstrual and abdominal pain and dyspareunia scores. Endometriotic lesions removed during laparoscopic surgery were collected and prepared for immunohistochemistry from 26 patients. Endometrial epithelial and stromal cells were identified with Cytokeratin and CD10 antibodies, respectively. Whole slide sections were digitised and the QuPath software was trained to automatically detect and count epithelial and stromal cells across the whole section. Using this classifier, we identified a significantly larger number of strongly labelled CD10 stromal cells (p = 0.0477) in deeply infiltrating lesions (99,970 ± 2962) compared to superficial lesions (2456 ± 859). We found the ratio of epithelial to stromal cells was inverted in deeply infiltrating endometriosis lesions compared to superficial peritoneal and endometrioma lesions and we subsequently identified a correlation between total endometrial cells and abdominal pain (p = 0.0005) when counted via the automated software. Incorporating histological software into current standard diagnostic pipelines may improve endometriosis diagnosis and provide prognostic information in regards to severity and symptoms and eventually provide the potential to personalise adjuvant treatment decisions.

Lessons from implementing the Australian National Action Plan for Endometriosis

Abstract

Endometriosis is a common yet under-recognised chronic disease with one in nine (more than 830,000) women and those assigned female at birth diagnosed with endometriosis by the age of 44 years in Australia. In 2018, Australia was the first country to develop a roadmap and blueprint to tackle endometriosis in a nationwide, coordinated manner. This blueprint is outlined in the National Action Plan for Endometriosis (NAPE), created from a partnership between government, endometriosis experts and advocacy groups. The NAPE aims to improve patient outcomes in the areas of awareness and education, clinical management and care and research. As researchers and clinicians are working to improve the lives of those with endometriosis, we discuss our experiences since the launch of the plan to highlight areas of consideration by other countries when developing research priorities and clinical plans. Historically, major barriers for those with endometriosis have been twofold; first, obtaining a diagnosis and secondly, effective symptom management post-diagnosis. In recent years, there have been calls to move away from the historically accepted 'gold-standard' surgical diagnosis and single-provider specialist care. As there are currently no reliable biomarkers for endometriosis diagnosis, specialist endometriosis scans and MRI incorporating artificial intelligence offer a novel method of visualisation and promising affordable non-invasive diagnostic tool incorporating well-established technologies. The recognised challenges of ongoing pain and symptom management, a holistic interdisciplinary care approach and access to a chronic disease management plan, could lead to improved patient outcomes while reducing healthcare costs.

Lay summary

Endometriosis is a chronic disease where tissue like the lining of the uterus is found in other locations around the body. For the 830,000 people living with endometriosis in Australia, this often results in an immense burden on all aspects of daily life. In 2018, Australia was the first country to introduce a roadmap and blueprint to tackle endometriosis in a nationwide coordinated manner with the National Action Plan for Endometriosis. This plan was created as a partnership between government, endometriosis experts and advocacy groups. There are several other countries who are now considering similar plans to address the burden of endometriosis. As researchers and clinicians are working to improve the lives of those with endometriosis, we share our experiences and discuss areas that should be considered when developing these national plans, including diagnostic pathways without the need for surgery, and building new centres of expertise in Endometriosis and Pelvic Pain.

Is there a role for small molecule metabolite biomarkers in the development of a diagnostic test for endometriosis?

Endometriosis is a disease defined by the presence of benign lesions of endometrial-like glands and stroma outside the endometrial cavity. Affecting an estimated 11.4% of Australian women, symptoms include chronic pelvic pain, dysmenorrhea and infertility. The current gold standard of diagnosis requires an expensive and invasive laparoscopic surgery, resulting in delayed time to treatment. The identification of a non-invasive endometriosis biomarker - a measurable factor correlating with disease presence or activity - has therefore become a priority in endometriosis research, although no biomarker has yet been validated. As small molecule metabolites and lipids have emerged as a potential focus, this review with systematic approach, aims to summarize studies examining metabolomic biomarkers of endometriosis in order to guide future research. EMBASE, PubMed and Web of Science were searched using keywords: lipidomics OR metabolomics OR metabolome AND diagnostic tests OR biomarkers AND endometriosis, and only studies written in English from August 2000 to August 2020 were included. Twenty-nine studies met inclusion and exclusion criteria and were included. These studies identified potential biomarkers in serum, ectopic tissue, eutopic endometrium, peritoneal fluid, follicular fluid, urine, cervical swabs and endometrial fluid. Glycerophospholipids were identified as potential biomarkers in all specimens, except urine and cervical swab specimens. However, no individual molecule or metabolite combination has reached clinical diagnostic utility. Further research using large study populations with robust patient phenotype and specimen characterisation is required if we are to make progress in identifying and validating a non-invasive diagnostic test for endometriosis.