{"paper_id":"55c5585f-fe27-4f4b-9e30-cb33731aab98","body_text":"Feasibility study of the application of Magnetic Resonance Elastography (MRE)  to \ndiagnose adenomyosis \n \nJain, V1,2 and Hojo, E1,3; McKillop, G2; Oniscu, A2; Le, Y3, Chen, J3,4,5, Ehman, R3, Roberts, \nN1; Critchley, HOD1,2. \n1. Centre for Reproductive Health, Institute for Regeneration and Repair, University of \nEdinburgh, Edinburgh, United Kingdom \n2. Royal Infirmary Edinburgh, Edinburgh, United Kingdom \n3. Mayo Clinic, Rochester, Minnesota, USA \n4. Department of Radiology, Nanjing Drum Tower Hospital, Nanjing, China \n5.  eResoundant, Inc., Rochester, MN, USA \n \nJoint Corresponding Authorship: Professor Hilary Critchley and Professor Neil Roberts: Centre \nfor Reproductive Health, Institute for Regeneration and Repair, University of Edinburgh, 4 -5 \nLittle France Drive, Edinburgh, EH15 4UU. Email: hilary.critchley@ed.ac.uk and \nneil.roberts@ed.ac.uk  \n \nPresent address for E. Hojo - Mayo Clinic, Rochester, Minnesota, USA \nPresent address for A. Oniscu - Karolinska University Hospital, Sweden \n \n  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \nNOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.\n\nStructured Abstract: \nIntroduction: Adenomyosis is an under -recognised condition in which definitive diagnosis is \nonly possible via histology after hysterectomy, an unacceptable option for those wishing to \npreserve fertility. Recent cellular/molecular studies indicate adenomyotic lesions ma y be \nfibrotic leading to increased uterine tissue stiffness. 3D Magnetic Resonance Elastography \n(MRE) is a novel imaging technique that allows in vivo measurement of tissue stiffness (via \nelastograms). 3D MRE has not been reported to study adenomyosis. The feasibility study aimed \nto utilise a novel  3D MRE protocol to measure global uterine stiffness and to investigate its \npotential application for non-invasive in vivo diagnosis of adenomyosis. \nMaterials and Methods: 3D MRE protocol was conducted on one heal thy volunteer (control) \nand four patients with suspected adenomyosis and heavy menstrual bleeding ( HMB), \ndiagnosed via transvaginal ultrasound and clinical history (REC:20/SS/0123 and 19/SS/0102). \nTwo patients underwent hysterectomy, and representative uterine tissue samples were assessed \nfor (i) histological presence of adenomyosis via H&E staining; (ii) cellular/molecular measures \nof tissue stiffness (collagen [picrosirius red], α -smooth muscle actin, e -cadherin); (iii) \nrelationship between in vivo assessment of the uterus via MRI images and 3D MRE findings \nwith in vitro uterine tissue histology from the same individuals. \nResults: 3D MRE  was successfully used to acquire elastogra ms for  four patients with \nadenomyosis (diffuse n=3, focal n=1) and one healthy volunteer. Calculated global uterine \nstiffness was higher in women with adenomyosis (2.93kPa; range 2.34 – 3.39kPa) compared \nto a healthy volunteer (2.04kPa). Areas of stiffness on 3D  elastograms reflected adenomyotic \nchanges visualised via conventional MRI, with the added benefit of also correlating with \nhistology/immunohistochemical assessment for markers of tissue stiffness. \nDiscussion: A novel 3D MRE protocol has been applied to ob tain the global uterine stiffness \nin four women with HMB and suspected adenomyosis , and one healthy volunteer . 3D MRE \nhas the potential to provide superior non -invasive tissue characterisation in vivo when \ncompared to conventional MRI in the assessment of adenomyosis due to the correlation of \nimaging and tissue findings. Further studies are now needed to confirm the above exploratory \nfindings, prior to performing a potential clinical trial. \n \nKeywords (3-5): adenomyosis, magnetic resonance elastography (MRE), diagnosis, fibrosis, \nabnormal uterine bleeding  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nIntroduction: \nAdenomyosis is a debilitating uterine disorder that is associated with heavy menstrual bleeding \n(HMB) and painful periods. It is defined as the presence of endometrial glands and stroma in \nthe myometrium at least  2.5 mm from the endo -myometrial junction (Bird, McElin, and \nManalo-Estrella 1972; Camboni and Marbaix 2021). Definitive diagnosis of adenomyosis can \nbe obtained by a hysterectomy however this is a major, invasive gynaecological operation that \nis also fertility-ending, and not acceptable to many reproductive-aged women. Presently, there \nis a lack of an accurate, reproducible, non -invasive diagnostic modality that  can equal the \nhistological definition and as a consequence, the condition remains under-diagnosed and under-\nresearched. T ransvaginal UltraSound (US) and Magnetic Resonance Imaging (MRI), have \nbeen investigated as potential non-invasive techniques for diagnosing suspected adenomyosis. \nThe reported sensitivities  of 83% , and 88% , respectively (Meredith, Sanchez -Ramos, and \nKaunitz 2009; Novellas et al. 2011) are reasonable but not always achieved in practice.  In this \nstudy, we tested the feasibility of employing an MRI modality known as Magnetic Resonance \nElastography (MRE) for diagnosing adenomyosis.  \n Due in part to the above diagnostic challenges, there is currently a lack of understanding \nregarding the aetiology and pathophysiology of adenomyosis (Guo 2022). One theory is that \nafter an area of adenomyosis (ectopic endometrial epithelial glands and stom a) is established \nwithin the myometrium, it may undergo cyclical changes in response to fluctuating levels of \ncirculating sex steroid hormones, and similar to the eutopic endometrium, the area of \nadenomyosis may bleed. (Leyendecker, Wildt, and Mall 2009; Vannuccini et al. 2017) . \nSubsequently, adenomyosis progresses to a state where it is characterise d by fibrosis which \noccurs through epithelial –mesenchymal transition (EMT), fibroblast-to-myofibroblast trans -\ndifferentiation (FMT) and smooth muscle metaplasia (SMM) (Liu et al. 2016; Huang et al. \n2022)  \nMRE allows non-invasive quantification of the mechanical properties of living tissues \n(Manduca et al. 2021)  and, given the above-mentioned significance of the occurrence of \nfibrosis, the technique is potentially well suited to supporting the diagnosis of adenomyosis. \nMRE protocols are already well established and clinically validated for staging the severity of \nliver fibrosis (Ehman 2022). There have also been reports of the application of 2D MRE in the \nstudy of the uterus (Stewart et al. 2011; Jondal et al. 2018; Obrzut et al. 2020)  and which \ndemonstrated that 2D MRE was successful in detect ing and characterising leiomyomas but \nadenomyosis was not studied. \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nThe primary objective of the present study was to apply a novel protocol utilising state-of-\nthe-art 3D MRE to measure global uterine stiffness in four patients with suspected adenomyosis. \nTo our knowledge, this is the first study where 3D MRE of the in vivo uterus with adenomyosis, \nhas been accompanied with the histological assessment of the same uterus in vitro. Therefore, \nthe secondary objective was, in two patients who went on to have a hysterectomy shortly after \nhaving 3D MRE scans , to correlate imaging findings with  histological investigations of the \nuterus, including markers of stiffness, to determine if 3D MRE is consistent with \nhistopathology. We hypothesise d that 3D MRE has potential to non-invasively diagnose \nadenomyosis, with correlation between T2 weighted MRI findings, histological findings from \nmatched stained uterine tissue sections, and an increased tissue stiffness on the 3D MRE \nelastograms. \n \nMaterials and Methods \nPatient recruitment \nThe study received respective approvals from the NHS Health Research Authority’s Research \nEthics Committees which permitted the recruitment of healthy volunteers (control) to support \nthe development of imaging protocol s testing (Ref: 20/SS/0123) , together with  patient \nrecruitment (study participant) and use of their uterine tissue samples (Refs. 19/SS/0102 and \n20/ES/0119). Fully informed w ritten consent was obtained from all  study participants, with \none control recruited from the University of Edinburgh, and four study participants recruited \nafter attending the gynaecology outpatient services at the Royal Infirmary Edinburgh, NHS \nLothian. At recruitment, the control, who was of reproductive age, did not report any clinical \nhistory of gynaecological complaints and had a normal menstrual cycle (as per International \nFederation of Obstetrics and Gynecology ( FIGO) AUB system 1) (Munro et al. 2011; Munro \net al. 2018). None of the participants were using hormonal treatments at the time the study was \nconducted. All four study participants were women of reproductive age, with the complaint of \nHMB and suspicion of adenomyosis following transvaginal US which was reviewed in \naccordance with  Morphological Uterus Sonographic Assessment ( MUSA) guidance on \nultrasound diagnosis of adenomyosis (Van den Bosch et al. 2019; Harmsen et al. 2022). There \nwas no evidence of uterine fibroids being present in any of the participants.  \nA fl ow diagram of the study  is shown in Figure 1 . The same MR investigation, \ncomprising T2-weighted MRI and 3D MRE imaging was performed for one control and four \nstudy participants, for two of whom a hysterectomy was performed for definitive treatment of \nHMB within 2 days following the MRI and 3D MRE imaging. \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\n \nFigure 1. The pathway followed by one healthy volunteer (control) and the study participants \nwho did and did not have hysterectomy. Study participants were patients with suspected \nadenomyosis (based on transvaginal ultrasound scan and clinical history) and heavy menstrual \nbleeding (HMB). Hysterectomy occurred after the T2 weighted MRI and 3D MRE imaging.  \nHysterectomy specimens (unfixed) were sectioned to mirror the planes from the 3D MRE \nimages. MRI = magnetic resonance imaging, 3D MRE = three dimensional magnetic resonance \nelastography. [to the reproduced in colour in print and on the web] \n \nSetup for Acquisition of MRI and MRE Data  \nMR investigations were performed at the Edinburgh Imaging Facility at The Queen’s Medical \nResearch Institute (EIF -QMRI), University of Edinburgh  using a 3 T Skyra Fit MRI system \n(Siemens Healthineers, Erlangen, Germany). The participant lay supine and a 32-channel spine \ncoil was positioned posteriorly and a 30-channel body matrix coil anteriorly. The MRE actuator, \nwhich has previously been used to perform  cardiac MRE (Arani et al. 2017) , was  placed \nanteriorly on the lower abdomen, above the pubis symphysis and medial to the anterior superior \niliac spine  regions to avoid close proximity to bony structures . A  foam cube was placed \nbetween the actuator  and the body to provide good coupling of the acoustic waves . Both \nactuator and foam were secured by using an elastic belt which ensured good contact between \nthe actuator and the body and maintenance of a central position . The actuator was connected \nvia a plastic tube allowing pneumatic transmission of vibrations from the active driver of the \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nResoundant MRE system (Mayo Clinic, Minnesota, USA)  to the MRE actuator ( i.e., passive \ndriver). \n \nMR Image Analysis \nFor each participant co-aligned and co-localised series of T2 weighted anatomical images and \n3D Echo Planar Imaging (EPI) MRE images were acquired in identical true axial orientation, \nwith Field of View (FOV) 24 cm and slice thickness 3 mm covering the whole uterus. For the \nT2 weighted MR images, the imaging matrix was 640 pixels by 640 pixels, TR 7570 msecs \nand TE 86 msecs and the acquisition time was 4 minutes 40 seconds. F or MRE, a vibration \nfrequency of 60 Hz was chosen, and the imaging matrix was 80 pixels by 80 pixels, TR 6400 \nmsecs and TE 79 msecs.  For each axial level a stiffness map was produced by using a 3D wave \ninversion algorithm and output directly on the MRI system. The acquisition time was 5 minutes \n33 seconds. \n \nMR Image Analysis \nThe T2-weighted images were reviewed by a Radiologist (GMcK) in consideration of whether \nadenomyosis could be confirmed. MRE data were analysed by using ITK-SNAP (Yushkevich \net al. 2006). In particular, a Region of Interest (ROI) was drawn demarcating the whole uterus \nas it appeared on the T2 -weighted image and transferred and superimposed on the relevant \nstiffness map. The mean stiffness (kPa) value for the image pixels corresponding to the selected \nROI and its variance were computed. This procedure was repeated for all the section levels \nencompassing the whole uterus.  Significance testing did not occur due to the sample size of \nthis feasibility study. \n \nTissue Preparation and Sampling \nFollowing hysterectomy, the fresh unfixed uterus was serially sliced into transverse sections \nusing as a guide the approximate locations of the imaging planes that were prescribed in the \nMR investigation. Biopsies were taken so as to sample myometrial and endometrial tiss ue in \nanterior, posterior, right lateral, left lateral and fundal regions of the uterus. A sample was also \nobtained of endometrial tissue in the uterine cavity by using a Pipelle® endometrial suction \ncurette (Pipelle de Cornier Mark II, Laboratoire CCD, France).  \n \n \n \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nTissue Processing and Staining \nUterine tissue samples were fixed in 4% neutral buffered formalin and embedded in paraffin \nusing standard procedures. Tissue sections with 4µm thickness were cut from the samples and \nstained with haematoxylin and eos in (H&E)  and picrosirius red  (0.5% Direct Red (Sigma -\nAldrich) in Picric acid (Sigma -Aldrich)). In addition, immunohistochemistry assays were \nperformed automatically for e-cadherin (Cell Signalling Technologies, a marker of EMT) and \nalpha smooth muscle actin (Sigma-Aldrich, αSMA; a marker of FMT) monoclonal antibodies \nby using the Leica Bond-Max autostainer (Leica Microsystems GmbH, Wetzlar, Germany).  \nHistopathologist (AO), who was blinded to the sampling location, reviewed the H&E  \nstained tissue sections to potentially confirm a diagnosis of ad enomyosis, and the picrosirius \nred, αSMA and e-cadherin stained slides to confirm the presence or absence of collagen fibres, \nsmooth muscle fibres and epithelial cells, respectively.  \n \nCorrelation of MR Imaging and Uterine Tissue Staining  \nOnce pathological reporting had been completed for the uterine tissue samples and MR images, \nand stiffness measurements had been obtained, a Radiologist (GMcK), Histopathologist (AO) \nand Gynaecologist (VJ) systematically reviewed the T2-weighted images for each of the four \npatients, as well as the stained tissue sections, for the two patients who underwent hysterectomy. \nThe stained uterine tissue sections and  corresponding MR investigations were reviewed \ntogether to confirm if the histologically confirmed areas of adenomyosis were detectable on \nthe T2 weighted MRI. All reviewers were blinded to the findings of the 3D MRE investigations \nat this stage. The regions of the confirmed adenomyosis were subsequently mapped to the 3D \nMRE elastograms to correlate whether they displayed increased tissue stiffness.  \n \nResults \nAll participants completed the study. No participant was menstruating at the time the study was \nperformed. No suspected adenomyosis  was observed on the MR images obtained for the \ncontrol subject. For all four study participants who had been diagnosed as having suspected \nadenomyosis on the basis of  transvaginal ultrasound performed prior to their participation in \nthe study, the MR images acquired during the study confirmed the same diagnosis. Furthermore \n(see Table 1), the systematic review of the T2 -weighted MRI images, stained uterine tissue \nsections, and 3D MRE elastograms (stiffness maps) obtained for the two study participants in \nwhom a hysterectomy was performed , provided qualitative support for  the prediction that  \nregions which the Radiologist diagnosed as adenomyosis, corresponded to regions where the \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nHistopathologist confirmed there is evidence of fibrosis, together with increased stiffness on \nthe 3D MRE elastograms.  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nParticipant Transvaginal \nultrasound \nT2 MRI Histopathology Markers of tissue stiffness  \n3D MRE \n(marker of tissue \nstiffness with \nimaging) \nPicrosirius Red \nstaining \nAlpha smooth \nmuscle actin \n(αSMA) \nstaining \nE-cadherin \nstaining \nControl No ultrasound \nperformed \nNo adenomyosis \nsuspected in the \nuterus \nN/A Elastograms \nshowed low uterine \nstiffness \n   \nStudy \nparticipant \nAd1 \nHeterogenous \nappearances of \nthe uterus with \nan ill-defined \nmass 26 x 22 x \n25 mm, ?fibroid, \nhowever the \nanterior wall of \nthe uterus is \nthicker than the \nposterior wall \nDiffuse and \nhomogenous \njunctional zone \nexpansion with \nonly trivial \nsubmucous \ncystic change \nconsistent with \ndiffuse \nadenomyosis \nConfirmatory \nregions of \nadenomyosis \ndetected throughout \nthe uterus, no \nevidence of \nfibroids. \nAreas of increased \nstiffness throughout \nuterus \nIncreased collagen \ndeposition \nthroughout uterus \nexcept in regions \nof adenomyosis at \nsites of ectopic \nepithelial glands \nand stoma \nαSMA \nstaining \nthroughout \nmyometrium – \nless in regions \nof \nadenomyosis \ne-cadherin \nstaining \nidentified in \nepithelial \nglands of the \nregions of \nadenomyosis \nconfirming \npresence \nStudy \nparticipant \nAd2 \nHeterogenous \nmyometrium \nsuggestive of \nadenomyosis \nJunctional zone \nirregularly \nthickened with \nfocal areas of \ncystic change in \nkeeping with \nsignificant and \nextensive \nadenomyosis \nConfirmatory \nregions of \nadenomyosis \ndetected throughout \nthe uterus \nAreas of increased \nstiffness throughout \nuterus \nIncreased collagen \ndeposition \nthroughout uterus \nexcept in regions \nof adenomyosis at \nsites of ectopic \nepithelial glands \nand stoma \nαSMA \nstaining \nthroughout \nmyometrium – \nless in regions \nof \nadenomyosis \ne-cadherin \nidentified in \nepithelial \nglands of \nregions of \nadenomyosis \nconfirming \npresence \nStudy \nparticipant \nAd3 \nUterus is bulky \nwith increased \nechogenicity in \nthe anterior wall \nand some cystic \nchange with \nincreased \nPossibly very \nearly focal \nthickening of \nthe junctional \nzone suggesting \nfocal \nadenomyosis \nN/A Targeted area of \nincreased stiffness \nin the uterus to \ncorrelate with \nregion of focal \nadenomyosis seen \non MRI \n   \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nvascularity \nsuggestive of \nadenomyosis \nbut not \nsufficient to call \nclinical \ndiagnosis of \nadenomyosis \nStudy \nparticipant \nAd4 \nMyometrium \ndiffusely \nheterogenous, \nasymmetrical \nwall thickening, \nwith a focal area \n38mm in the \nanterior wall \nthat is isoechoic \nto the \nmyometrium \nsuggestive of \nadenomyosis \nwith an \nadenomyoma \nIrregular \nthickening of \nthe junctional \nzone especially \nanteriorly and \ntowards the \nfundus, \njunctional zone \nshows some \ncystic changes. \nFindings \nconsistent with \nadenomyosis \n* Areas of increased \nstiffness throughout \nuterus \n   \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\n \nTable 1: Summary of the  transvaginal ultrasound, Magnetic Resonance Imaging (MRI), 3D \nMRE and uterine tissue staining observations for participants, as applicable  per study \nparticipation. * after study completion, this patient  (Ad4) underwent hysterectomy for \ntreatment of heavy menstrual bleeding, and received a histologically confirmed diagnosis of \ndiffuse adenomyosis with probable adenomyoma. \n \nThe 3D MRE data acq uired for one of the study participants  diagnosed with  adenomyosis \n(subject Ad2) are illustrated in comparison to corresponding data acquired for the control in  \nFigure 2. The wave motion in x, y, and z directions (with corresponding zoomed in interpolated \nimages) for the patient Ad2 (Figure 2 K (O), L (P), and M (Q)) has longer wavelength than for \nthe control (Figure 2 B (F), C (G) and D (H)) indicating that regions of potential adenomyosis \nidentified by the Radiologist, are potentially corresponding to higher stiffness values in the 3D \nMRE elastograms.  \n \n \n \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\n \nFigure 2. T2-weighted MR images, with 3D MRE waves images in x-, y- and z-directions and \n3D MRE elastogram (colour map of stiffness) with a field of view of 24cm, superimposed on \nthe MRE magnitude image of the uterus of a control subject and for the study participant Ad2 \n(scale bar 0 to 5 kPa). Corresponding zoomed in interpolated images with a field of view of \n7.5cm. [to the reproduced in colour in print and on the web] \n \n \nThe measures of the stiffness of the whole uterus obtained for each participant are plotted in \nFigure 3. The median value obtained for the four patients  with adenomyosis (2.93kPa; range \n2.34 – 3.39kPa) is greater than the value measured for the healthy volunteer subject (2.04kPa).  \n \n \n \nFigure 3. Average stiffness (kPa) of the whole uterus for the healthy volunteer subject and four \npatients (Ad1, Ad2, Ad3 and Ad4). \n \n0\n0.5\n1\n1.5\n2\n2.5\n3\n3.5\n4\nHealthy Volunteer Ad1 Ad2 Ad3 Ad4\nGlobal estimated uterine stiffness (kPa)\nStudy Subject\nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nFigure 4 refers to the same patient as in Figure 2 (i.e., Ad2) and shows the resected uterus \n(A), the dissection (B) and sectioning that was performed so as to match with the T21 \nweighted MR images (C). The stained tissue sections depict the region where an adenomyotic \nlesion was detected during both the histopathological investigation of the uterus and on the \nT2 weighted MR images. The tissue has been systematically stained in (D), contains collagen \nfibres (E), smooth muscle fibres (F) and endometrial gland (G), consistent with the diagnosis \nof adenomyosis.  \n \n \n \n \n \nFigure 4. The whole uterus (A) was dissected in the transverse plane (B), in the direction and \nwith section spacing that matched the series of T2 weighted MR images (C). The red rectangle \ndepicts a region for which detailed  histological analysis was performed and which confirmed \nthe presence of endometrium, myometrium and an adenomyotic lesion  (D), with  ectopic \nendometrial glands and stroma of the adenomyotic lesion clearly visible in the magnified view. \nPicrosirius red staining indicated that excess collagen deposition was present (E), alpha smooth \nmuscle actin staining confirmed the presence of smooth muscle fibres of myometrium (F), and \ne-cadherin staining provided support for the epithelial nature of the endometrial glands within \nthe myometrial tissue (G). The scale bar corresponds to 2 mm. [to the reproduced in colour in \nprint and on the web] \n \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\n \nDiscussion \nThe results of this study provide preliminary evidence of the feasibility of applying MRE to \nevaluate global uterine stiffness in patients with suspected adenomyosis. The mean value of \nthe stiffness of the whole uterus in the participants with adenomyosis was greater than that for \nthe control subject, providing motivation for future studies with larger cohort sizes that would \nallow significance of this preliminary observation to be tested.  Furthermore, for two \nparticipants in whom a hysterectomy was performed  (for HMB) , areas of suspected \nadenomyosis on T2-weighted MR images were confirmed histologically via stained uterine \ntissue secti ons, and mapped to corresponding elastograms obtained by using 3D MRE, \nsuggesting increased tissue stiffness. These findings suggest that 3D MRE has the potential to \nallow tissue characterisation that is consistent with histopathology and which can be used to \nsupport non-invasive diagnosis of adenomyosis.  \nFurther studies, which learn from this feasibility study, are needed to confirm these \npreliminary data and validate 3D MRE as a potential new non-invasive method of diagnosing \nand monitoring adenomyosis. The study of a larger cohort, which should include patients with \nand without diffuse adenomyosis, in the first instance, would allow for investigation into \nclinical utility of this diagnostic modality. \nWith regard to previous relevant research, Hobson et al (Hobson et al. 2007) performed \nultrasound strain imaging on dissected uteri with confirmed  adenomyosis and found  no \ndifference in the strain values obtained for the endometrium and myometrium, and concluded \nthis was due to the pathological processes of the disease . However, more recently, in studies \nwhich used Transvaginal ultrasound elastography, Frank et al (Frank et al. 2016) have reported \nthat adenomyotic lesions are soft er than normal myometrium , whereas there have also been \nreports that regions of adenomyosis are stiffer than normal myometrium (Liu et al. 2018; Liu \net al. 2016). Furthermore, Liu et al. (Liu et al. 2016)  reported that the stiffness of regions of \nadenomyosis were greater in patients with heavier menstrual bleeding and this finding has been \nreproduced (Huang et al. 2022; Liu et al. 2018) . Huang et al (Huang et al. 2022)  also \ndemonstrate a gradient in the density of  markers of tissue stiffness  within selected stained \nuterine tissue sections, from the region of adenomyosis (greater density) to the endometrium \n(less density). This possibly suggests the value of tissue stiffness as measured by elastography, \nmay also change in relation to the proximity to the adenomyotic lesion.  Transvaginal \nultrasound elastography has also been used to demonstrate  the differences between  tissue \nstiffness for normal myometrium, uterine fibroids and regions of adenomyosis (Liu et al. 2018). \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nThis suggests that the marker of tissue stiffness as measured by elastography, could be \nleveraged to non -invasively differentiate between uterine fibroids and adenomyosis, a  \nclinically important differentiation to guide treatment for HMB.  \nOne of the patients (Ad3) recruited to the present study was diagnosed as having focal \nadenomyosis, suggesting that some areas of the uterus appeared normal. This is consistent with \nthe value of the stiffness of the uterus in this patient being closer to the value obtained for the \ncontrol subject , rather than the other three adenomyosis patients (Ad1, 2, 4) who had \nhistologically confirmed diffuse adenomyosis (see Figure 3).  \nThe authors acknowledge there are  limitations within this feasibility study. First, \nstatistical analyses were restricted due to the small number of participants. Second, the MRE \nmeasurements obtained in th e present study refer to  the whole uterus and not to particular \nregions of adenomyosis identified by a Radiologist. The resolution of the 3D MRE is 3mm x \n3mm x 3mm and therefore some of the histologically detected regions of adenomyosis could \nnot be confirmed via T2 weighted MRI or 3D MRE as they were too small. This is the reason \nthe tissue stiff ness of the whole uteru s, as opposed to regions of adenomyosis was utilised \nwithin this feasibility study. Despite these limitations, t he present study does , however, have \none major advantage. In previous MRE studies of the uterus, data have been acquired by using \na 2D MRE technique (Stewart et al. 2011; Jondal et al. 2018; Obrzut et al. 2020) , whereas a \n3D EPI MRE sequence was used in the present study. To the best of our knowledge this is the \nfirst study where 3D MRE of the uterus has been accompanied with histological assessment \nfrom the same individual with adenomyosis.  \nCurrently, there is no clear global consensus on the criteria for diagnosing adenomyosis \nby using either ultrasound or MRI techniques. The Morphological Uterus Sonographical \nAssessment (MUSA) provides a robust consensus on the definition of adenomyosis based on \nultrasound imaging  (Harmsen et al 2022 ). No such consensus exist s for MRI diagnosis of \nadenomyosis. The data obtained from t ransvaginal ultrasound elastography  is operator-\ndependant and the technique is not yet available in routine clinical practice. MRE, on the other \nhand, is a clinically approved diagnostic technique currently approved for use for grading of \nliver fibrosis and  is widely available globally. The advantages of MRE over the other \ntechniques are summarised in Table 2. Most importantly for clinica l translation, MRE is not \noperator dependant. MRE is also truly non -invasive, compared to transvaginal ultrasound \nelastography which requires a transducer to be inserted into the vagina, and manual \ncompression of the uterine tissues by means of the ultrasound transducer held by the operator, \nand may not be appropriate for use in younger patients and not acceptable to some women. \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nGlobal multicentre studies will be important to gather essential data to support the \nvalidation of these findings  in both women with and without adenomyosis (no uterine \npathology), before clinical translation and inclusion in the ongoing clinical health management \nof women with adenomyosis. \n \n \n Hysterectomy \n(histopathological \nassessment of \nmyometrium) \nTransvaginal \nUltrasound \nMagnetic \nResonance \nImaging \n(MRI) \nTransvaginal \nUltrasound \nElastography \n3D Magnetic \nResonance \nElastography \n(3D MRE) \nDiagnosis accuracy \n(if performed by \nspecialist) \n100% ~83% ~88% To be \ndetermined \nTo be \ninvestigated \nEstimated cost of \ndiagnosis \n++++ + ++ Unknown ++ \nNon-invasive No No Yes No Yes \nFertility sparing No Yes Yes Yes Yes \nOutpatient visit \n(No hospital \nadmission required) \nNo Yes Yes Yes Yes \nQuick No Yes Yes Yes Yes \nMonitor treatment \nresponse \nNo No No Unknown Yes \nObjective Yes No Yes No Yes \nQuantitative No No No No Yes \n \nTable 2. The defining characteristics of current and possible future modalities that may be used \nfor diagnosis of adenomyosis in comparison to the current gold standard of histopathological \nassessment of the myometrium. \n \nConclusion \nThe findings of the present study provide preliminary evidence of the feasibility of employing \n3D MRE to aid in diagnosis of adenomyosis, a significant condition affecting women’s health. \n \nAcknowledgements: \nThe authors would like to acknowledge : Ms. Catherine Murray for her help and support with \npatient recruitment ; Professor Sco tt Semple, for his help and support with  the MR \ninvestigations; Ms Moira Nicol, Dr Michael Millar and the team at SURF Histology, University \nof Edinburgh  (https://surf.ed.ac.uk/facilities/histology/) for their advice and support with \nhistological investigations;  Dr. Bradley D. Bolster and Dr. Stephan Kannengiesser at Siemens \nHealthcare for their contribution to this work; all the patients and healthy volunteers who have \ngenerously donated their time, thoughts, tissue and blood samples, and much more to all of our \nstudies. Funding: University of Edinburgh’s Institute for Regeneration & Repair (IRR) Early \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nCareer Researcher’s  Innovation Award 2022, MRC Centre for Reproductive Health (MRC \nCentre grants: G1002033 and MR/N022556/1). \n \n \nReferences: \nArani A, Arunachalam SP, Chang ICY, Baffour F, Rossman PJ, Glaser KJ, Trzasko JD, McGee \nKP, Manduca A, Grogan M, et al. Cardiac MR elastography for quantitative assessment \nof elevated myocardial stiffness in cardiac amyloidosis. J Magn Reson Imaging 2017: \n46; 1361-1367. \nBird CC, McElin TW, and Manalo-Estrella P. The elusive adenomyosis of the uterus--revisited. \nAm J Obstet Gynecol 1972: 112; 583-593. \nCamboni A and Marbaix E. Ectopic Endometrium: The Pathologist's Perspective. Int J Mol Sci \n2021: 22. \nEhman RL. Magnetic resonance elastography: from invention to standard of care. Abdom \nRadiol (NY) 2022: 47; 3028-3036. \nFrank ML, Schafer SD, Mollers M, Falkenberg MK, Braun J, Mollmann U, Strube F, Fruscalzo \nA, Amler S, Klockenbusch W, and Schmitz R. Import ance of Transvaginal \nElastography in the Diagnosis of Uterine Fibroids and Adenomyosis. Ultraschall Med \n2016: 37; 373-378. \nGuo SW. Cracking the enigma of adenomyosis: an update on its pathogenesis and \npathophysiology. Reproduction 2022: 164; R101-R121. \nHarmsen MJ, Van den Bosch T, de Leeuw RA, Dueholm M, Exacoustos C, Valentin L, \nHehenkamp WJK, Groenman F, De Bruyn C, Rasmussen C, et al. Consensus on revised \ndefinitions of Morphological Uterus Sonographic Assessment (MUSA) features of \nadenomyosis: results of modified Delphi procedure. Ultrasound Obstet Gynecol 2022: \n60; 118-131. \nHobson MA, Kiss MZ, Varghese T, Sommer AM, Kliewer MA, Zagzebski JA, Hall TJ, Harter \nJ, Hartenbach EM, and Madsen EL. In vitro uterine strain imaging: preliminary results. \nJ Ultrasound Med 2007: 26; 899-908. \nHuang Q, Liu X, Critchley H, Fu Z, and Guo SW. How does the extent of fibrosis in \nadenomyosis lesions contribute to heavy menstrual bleeding? Reprod Med Biol 2022: \n21; e12442. \nJondal DE, Wang J, Chen J, Gorny KR, Felmlee J, Hesly G, Laughlin-Tommaso S, Stewart \nEA, Ehman R, and Woodrum DA. Uterine fibroids: correlations between MRI \nappearance and stiffness via magnetic resonance elastography. Abdom Radiol (NY)  \n2018: 43; 1456-1463. \nLeyendecker G, Wildt L, and Mall G. The pathophysiology of endometriosis and adenomyosis: \ntissue injury and repair. Arch Gynecol Obstet 2009: 280; 529-538. \nLiu X, Ding D, Ren Y, and Guo SW. Transvaginal Elastosonography as an Imaging Technique \nfor Diagnosing Adenomyosis. Reprod Sci 2018: 25; 498-514. \nLiu X, Shen M, Qi Q, Zhang H, and Guo SW. Corroborating evidence for platelet -induced \nepithelial-mesenchymal transition and fibroblast -to-myofibroblast transdifferentiation \nin the development of adenomyosis. Hum Reprod 2016: 31; 734-749. \nManduca A, Bayly PJ, Ehman RL, Kolipaka A, Royston TJ, Sack I, Sinkus R, and Van Beers \nBE. MR elastography: Principles, guidelines, and terminology. Magn Reson Med 2021: \n85; 2377-2390. \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint \n\nMeredith SM, Sanchez -Ramos L, and Kaunitz AM. Diagnostic accuracy of transvaginal \nsonography for the diagnosis of adenomyosis: systematic review and metaanalysis. Am \nJ Obstet Gynecol 2009: 201; 107 e101-106. \nMunro MG, Critchley HO, Broder MS, Fraser IS, and FIGO Working Group on Menstrual \nDisorders. FIGO classification system (PALM-COEIN) for causes of abnormal uterine \nbleeding in nongravid women of reproductive age. Int J Gynaecol Obstet 2011: 113; 3-\n13. \nMunro MG, Critchley HOD, Fraser IS, and Committee FMD. The two FIGO systems for \nnormal and abnormal uterine bleeding symptoms and classification of causes of \nabnormal uterine bleeding in the reproductive years: 2018 revisions. Int J Gynaecol \nObstet 2018: 143; 393-408. \nNovellas S, Chassang M, Delotte J, Toullalan O, Chevallier A, Bouaziz J, and Chevallier P. \nMRI characteristics of the uterine junctional zone: from normal to the diagnosis of \nadenomyosis. AJR Am J Roentgenol 2011: 196; 1206-1213. \nObrzut M, Obrzut B, Zmuda M, Baran J, Cholewa M, Ehman R, and Darmochwal -Kolarz D. \nUterine leiomyomas: correlation between histologic composition and stiffness via \nmagnetic resonance elastography - a Pilot Study. Ginekol Pol 2020: 91; 373-378. \nStewart EA, Taran FA , Chen J, Gostout BS, Woodrum DA, Felmlee JP, and Ehman RL. \nMagnetic resonance elastography of uterine leiomyomas: a feasibility study. Fertil \nSteril 2011: 95; 281-284. \nVan den Bosch T, de Bruijn AM, de Leeuw RA, Dueholm M, Exacoustos C, Valentin L, \nBourne T, Timmerman D, and Huirne JAF. Sonographic classification and reporting \nsystem for diagnosing adenomyosis. Ultrasound Obstet Gynecol 2019: 53; 576-582. \nVannuccini S, Tosti C, Carmona F, Huang SJ, Chapron C, Guo SW, and Petraglia F. \nPathogenesis of adenom yosis: an update on molecular mechanisms. Reprod Biomed \nOnline 2017: 35; 592-601. \nYushkevich PA, Piven J, Hazlett HC, Smith RG, Ho S, Gee JC, and Gerig G. User-guided 3D \nactive contour segmentation of anatomical structures: significantly improved efficiency \nand reliability. Neuroimage 2006: 31; 1116-1128. \nBird CC, McElin, TW, Manalo-Estrella, P et al, 1972. The elusive adenomyosis of the uterus—\nrevisited. American Journal of Obstetrics and Gynecology, Volume 112, Issue 5, 1 \nMarch 1972, Pages 583-593 https://doi.org/10.1016/0002-9378(72)90781-8 \n \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted September 4, 2024. ; https://doi.org/10.1101/2024.09.03.24313024doi: medRxiv preprint","source_license":"CC0","license_restricted":false}