{"paper_id":"8e389de3-e1c4-455d-8fe2-e760cf09b12e","body_text":"A growing number of studies revealed that 10% of women have been affected by\nendometriosis, a chronic gynecological disorder. It is widely considered a major\ncause of dysmenorrhea, pelvic pain, and infertility ( Barnhart  et al. , 2002 ;  Omland  et al. , 2005 ;  Gupta  et al. , 2008 ). Currently, it is not only a\nhormone-dependent disease but also a persistent inflammatory condition characterized\nby immune dysregulation and an altered peritoneal microenvironment ( Harada  et al. , 1999 ;  Gupta  et al. , 2008 ).\nThe disorder negatively impacts fertility, especially in women undergoing assisted\nreproductive technologies. Clinical and experimental studies demonstrate that\nendometriosis disrupts ovarian, tubal, and uterine physiology, resulting in poor\noocyte quality, impaired fertilization, and decreased implantation rates ( Barnhart  et al. , 2002 ;  Omland  et al. , 2005 ).\nInflammatory activation within the peritoneal cavity promotes recruitment of\nmacrophages and excessive secretion of cytokines such as IL-1β, IL-8,\nTNF-α, and VEGF ( Iwabe  et\nal. , 1998 ;  von Wolff\n et al. , 1999 ;  Mahnke\n et al. , 2000 ;  Witz,\n2000 ;  Ueda  et al. ,\n2002 ). These proinflammatory mediators enhance angiogenesis, fibrotic\nremodeling, granulosa cell apoptosis, and oxidative stress, ultimately compromising\nreproductive potential ( Khan  et\nal. , 2014 ;  Da Broi  et\nal. , 2016 ;  Sanchez\n et al. , 2016 ; 2017;  Giacomini  et al. , 2017 ;  Wu  et al. , 2017 ).\nElevated levels of IL-1β, IL-8, and TNF-α in follicular and peritoneal\nfluid have been strongly associated with impaired oocyte competence, reduced\nfertilization, and suboptimal embryo development ( Khan  et al. , 2014 ;  Da\nBroi  et al. , 2016 ;  Giacomini  et al. , 2017 ;  Sanchez  et al. , 2017 ;  Wu  et al. , 2017 ). Persistent inflammation drives lesion\nproliferation and recurrence, even after conventional therapy ( Santamaria  et al. , 2012 ;  Becker  et al. , 2017 ;  Viganò  et al. , 2018 ).\nCurrent management relies mainly on hormonal therapy. Dienogest, a fourth-generation\nprogestin, provides symptom relief by suppressing ovulation and inducing endometrial\natrophy ( Andres  et al. , 2015 ;\n Bedaiwy  et al. , 2017 ;\n Wu  et al. , 2022 ;  Muzii  et al. , 2023 ). However,\nit does not fully resolve immune imbalance, is associated with side effects, and\nrecurrence is common upon discontinuation ( Andres\n et al. , 2015 ;  Becker\n et al. , 2017 ;  Bedaiwy\n et al. , 2017 ;  Szukiewicz, 2022 ;  Wu  et\nal. , 2022 ;  Muzii  et\nal. , 2023 ;  Piriyev\n et al. , 2025 ). Meta-analyses confirm that while\nDienogest alleviates symptoms, it does not prevent disease progression ( Andres  et al. , 2015 ;  Wu  et al. , 2022 ;  Muzii  et al. , 2023 ).\nScholars have paid close attention to Mesenchymal stem cells (MSCs) because they\ncontain immunomodulatory and regenerative characteristics ( Uccelli  et al. , 2008 ). Bone marrow-derived\nMSCs (BMMSCs) are well characterized and fulfill ISCT minimal criteria: adherence to\nplastic, fibroblast-like morphology, positive for CD73, CD90, and CD105, and\nnegative for CD45 ( Dominici  et al. ,\n2006 ;  Ghaneialvar  et\nal. , 2018 ;  Schmelzer  et\nal. , 2019 ). These cells exert their effects via paracrine\nmechanisms, including secretion of cytokines, growth factors, and extracellular\nvesicles ( Dominici  et al. ,\n2006 ;  Uccelli  et al. ,\n2008 ;  Chen  et al. ,\n2018 ;  Ghaneialvar  et\nal. , 2018 ;  Schmelzer  et\nal. , 2019 ;  Bian  et\nal. , 2022 ;  Műzes &\nSipos, 2022 ;  Kulesza  et\nal. , 2023 ).\nIn reproductive models, BMMSCs have been shown to reduce granulosa cell apoptosis,\nenhance ovarian function, and modulate immune responses ( Chen  et al. , 2018 ;  Rajabzadeh  et al. , 2019 ;  Bozorgmehr  et al. , 2020 ;  Na & Kim, 2020 ;  Liao\n et al. , 2021 ;  Cui\n& Jing, 2024 ;  Kavaldzhieva  et\nal. , 2025 ). However, direct comparisons between BMMSC\ntherapy and standard hormonal treatments for endometriosis remain scarce ( Vernon & Wilson, 1985 ;  Grümmer, 2006 ;  Bedaiwy  et al. , 2017 ). Considering the central\nroles of IL-1β and IL-8 in lesion maintenance and infertility ( Iwabe  et al. , 1998 ;  von Wolff  et al. , 1999 ;  Mahnke  et al. , 2000 ;  Witz, 2000 ;  Ueda  et al. , 2002 ;  Khan\n et al. , 2014 ;  Da Broi\n et al. , 2016 ;  Giacomini  et al. , 2017 ;  Sanchez  et al. , 2017 ;  Wu  et al. , 2017 ), evaluating BMMSCs against Dienogest\nis crucial.\nThe novelty of this study lies in its direct head-to-head comparison of BMMSCs,\nDienogest, and their combination in a murine endometriosis model. Previous studies\nhave demonstrated that MSCs possess immunomodulatory and regenerative potential in\nendometriosis by reducing inflammatory cytokines, inhibiting fibrosis, and improving\novarian function ( Dominici  et al. ,\n2006 ;  Uccelli  et al. ,\n2008 ;  Chen  et al. ,\n2018 ;  Rajabzadeh  et al. ,\n2019 ;  Bozorgmehr  et al. ,\n2020 ;  Na & Kim, 2020 ;  Liao  et al. , 2021 ;  Cui & Jing, 2024 ;  Kavaldzhieva  et al. , 2025 ). However, these\ninvestigations primarily focused on MSC therapy alone, without systematic evaluation\nagainst current pharmacological standards. To our knowledge, no prior studies have\ncompared BMMSC therapy directly with Dienogest, a fourth-generation progestin widely\nused in clinical practice for endometriosis management ( Andres  et al. , 2015 ;  Bedaiwy  et al. , 2017 ;  Wu  et al. , 2022 ;  Muzii  et al. , 2023 ). By demonstrating that\nBMMSC monotherapy not only suppressed IL-1β and IL-8 expression but also\nsignificantly reduced lesion size, while Dienogest failed to do so, this study\nprovides unique preclinical evidence of BMMSCs’ superior disease-modifying and\nfertility-preserving potential ( Iwabe  et\nal. , 1998 ;  von Wolff\n et al. , 1999 ;  Mahnke\n et al. , 2000 ;  Witz,\n2000 ;  Ueda  et al. ,\n2002 ;  Dominici  et al. ,\n2006 ;  Uccelli  et al. ,\n2008 ;  Khan  et al. ,\n2014 ;  Da Broi  et al. ,\n2016 ;  Giacomini  et al. ,\n2017 ;  Sanchez  et al. ,\n2017 ;  Wu  et al. ,\n2017 ;  Chen  et al. ,\n2018 ;  Ghaneialvar  et\nal. , 2018 ;  Rajabzadeh  et\nal. , 2019 ;  Schmelzer\n et al. , 2019 ;  Bozorgmehr  et al. , 2020 ;  Liao  et al. , 2021 ;  Bian  et al. , 2022 ;  Műzes & Sipos, 2022 ;  Kulesza  et al. , 2023 ;  Cui & Jing, 2024 ;  Kavaldzhieva\n et al. , 2025 ).\nThis study aimed to compare the effects of BMMSCs, Dienogest, and their combination\non IL-1β and IL-8 expression and lesion size in a murine endometriosis model.\nWe hypothesized that BMMSC monotherapy would provide superior disease-modifying\neffects compared to Dienogest or combined therapy.\n\nThis experimental study was approved by the Animal Care and Use Committee of\nUniversitas Airlangga, Surabaya, Indonesia (Approval No. 3.KEH.145.10.2024). All\nprocedures adhered to international standards for animal research. Humane\nendpoints were defined, and animals were monitored daily for weight, activity,\nand distress. Mice were euthanized using isoflurane overdose followed by\ncervical dislocation, in accordance with AVMA Guidelines (2020).\nThe experiment was conducted between August and December 2024 at the Stem Cell\nLaboratory, Institute of Tropical Disease, and the Faculty of Veterinary\nMedicine, Universitas Airlangga, Surabaya.\nBone marrow aspirates were obtained from 3-month-old female mice (25-30 g). Under\nlocal anesthesia, tibial marrow was collected from 5-10 donor mice, yielding ~3\nmL of aspirate, which was transferred into heparinized tubes containing an equal\nvolume of minimum essential medium-alpha (Invitrogen, USA) and stored at 4°C\nuntil processing.\nSamples were diluted with phosphate-buffered saline (PBS; Sigma, USA) and\ncentrifuged twice at 1600 rpm for 15 min. The mononuclear cell fraction (buffy\ncoat) was isolated by Ficoll density gradient (GE Healthcare, UK), washed with\nPBS, resuspended in 6 mL of complete culture medium (CCM; Invitrogen, USA), and\nseeded into 5 cm 2  culture dishes. Cultures were maintained at 37°C in\n5% CO₂. After 24 h, non-adherent cells were removed by washing with PBS and\nreplaced with fresh CCM. Medium was changed every 3 days until cultures reached\n60-80% confluence.\nCells were subcultured every 5 days until passage 4. Phenotypic confirmation was\nperformed by immunofluorescence staining, requiring negativity for CD45 and\npositivity for CD73, CD90, and CD105, consistent with ISCT minimal criteria\n( Dominici  et al. ,\n2006 ;  Uccelli  et\nal. , 2008 ;  Ghaneialvar\n et al. , 2018 ;  Schmelzer  et al. , 2019 ). These markers have also\nbeen validated in previous studies characterizing MSCs in various species ( Chen  et al. , 2018 ;  Na & Kim, 2020 ).\nThirty-two female BALB/c mice (3 months old, 25-30 g; Charles River, USA) were\nacclimatized for one week before randomization into four groups (n=8 each).\nAnimals were housed under controlled conditions (22±2°C, 55±5%\nhumidity, 12/12 h light-dark cycle) with free access to standard chow and water.\nRandomization was computer-generated, and outcome assessors were blinded.\nExclusion criteria included perioperative death or failed lesion induction.\nK+:  Endometriosis control group\nP1:  BMMSC therapy (1 × 10 6  cells in 0.2 mL PBS,\ni.p., day 15)\nP2:  BMMSC + Dienogest therapy (same BMMSC dose + dienogest 1\nmg/kg/day p.o. from day 15-29)\nP3:  Dienogest therapy (1 mg/kg/day p.o. from day 15-29)\nThroughout the experiment, animals were monitored daily for body weight,\nactivity, grooming, and signs of distress ( Harada  et al. , 1999 ). No mortality occurred.\nNecropsy of major non-reproductive organs (liver, kidney, spleen) revealed no\ngross abnormalities, indicating that BMMSC administration did not induce\nsystemic toxicity ( Dominici  et\nal. , 2006 ;  Uccelli\n et al. , 2008 ).\nEndometriosis was induced following previously described protocols ( Gupta  et al. , 2008 ;  Bedaiwy  et al. , 2017 ). This\nmurine model was selected because surgical induction by transplantation of\nendometrial tissue reliably reproduces the pathophysiological features of human\ndisease, including peritoneal lesion formation, estrogen dependency, and chronic\ninflammatory cytokine responses ( Vernon &\nWilson, 1985 ;  Bedaiwy  et\nal. , 2017 ).\nMice received intramuscular cyclosporin A (10 mg/kg; Sandimmune, Novartis,\nSwitzerland) and estrogen priming. A 0.1 mL suspension of human endometrial\ntissue was injected intraperitoneally to establish lesions. Estrogen\nsupplementation (5.4 mg/mouse; equivalent to 10IU per 1 mg) was administered\ndaily from days 1-5. By day 14, visible lesions had formed. BMMSCs\n(1×10 6  cells/mouse) were injected intraperitoneally on day\n15. All animals were sacrificed on day 29, and tissues collected for analysis.\nEuthanasia was performed in accordance with the protocol approved by the\nInstitutional Animal Care and Use Committee of Universitas Airlangga (Approval\nNo. 3.KEH.145.10.2024).\nFormalin-fixed, paraffin-embedded ovarian tissues were sectioned at 5 µm\nand deparaffinized in xylene (3 × 3 min), followed by graded ethanol\nrehydration (100%, 95%, 70%) and rinsing in distilled water. Endogenous\nperoxidase activity was blocked using peroxidase solution (27°C, 10 min).\nSections were incubated in blocking serum (25°C, 10 min), then with polyclonal\nanti-IL-1β and anti-IL-8 primary antibodies (Bioss Antibodies) for 10\nmin.\nAfter washing in PBS, slides were incubated with horseradish\nperoxidase-conjugated secondary antibody (25°C, 10 min), developed using\ndiaminobenzidine (DAB, 10min), and counterstained with hematoxylin and eosin (3\nmin). Sections were dehydrated, mounted, and examined under light microscopy\n(Nikon H600L, DS-Fi2 digital camera, Nikon Image System). Brown cytoplasmic\nstaining was interpreted as positive expression in glandular epithelial and\nstromal cells ( Iwabe  et al. ,\n1998 ;  Mahnke  et al. ,\n2000 ;  Muzii  et al. ,\n2023 ).\nEndometriotic lesion dimensions were evaluated using ImageJ v1.54 (NIH, USA).\nLesions were photographed, calibrated against a scale bar, outlined, and\nmeasured automatically to calculate the surface area (mm 2 ). Such\nimage-based quantitative analysis is widely applied in regenerative and\nMSC-related studies, including wound healing models ( Andres  et al. , 2015 ;  Wu  et al. , 2022 ;  Muzii  et al. , 2023 ). Results were\nexpressed as mean lesion size per group. In addition to calculating lesion size,\nthe sample was analyzed semi-quantitatively using the IRS semi-quantitative\nscale, as expressed in  Table 1 .\nIRS semi-quantitative scale.\nThis study applied different analysis techniques depending on the data\ndistribution pattern. For example, One-way analysis of variance (ANOVA) was\napplied to normally distributed variables. Meanwhile, if the data were not\nnormally distributed, the Kruskal-Wallis test was performed. Following that, the\nstudy conducted a post hoc Bonferroni correction for multiple comparisons.\nIL-1β expression followed normal distribution and was analyzed using\nANOVA, whereas IL-8 expression and lesion size were analyzed using the\nKruskal-Wallis test ( Harada  et\nal. , 1999 ). The work has been reported in line with the\nARRIVE guidelines 2.0.\n\nThis study consisted of three stages: stem cell preparation, induction of the\nendometriosis model, and evaluation of BMMSC effects on IL-1β, IL-8, and\nlesion area. Stem cells were confirmed as mesenchymal by immunophenotyping\n(CD73 + , CD90 + , CD105 + , and CD45 - ),\nand their homing capacity was verified by PKH26 luminescence ( Iwabe  et al. , 1998 ;  Uccelli  et al. , 2008 ;  Andres  et al. , 2015 ;  Becker  et al. , 2017 ;  Viganò  et al. , 2018 ). To validate the\nmodel, mice from all groups (K + -P3) were euthanized on day 14, showing\nestablished peritoneal lesions.\nBone marrow aspirates from mouse tibiae were processed according to standard\nprotocols in the Stem Cell Laboratory, Institute of Tropical Disease,\nUniversitas Airlangga. Cultured cells exhibited fibroblast-like morphology,\nappearing elongated, flattened, and spindle-shaped with large nuclei.\nSubculturing was performed every 5 days until the fourth passage. By passage 4,\ncultures displayed the characteristic swirling growth pattern of mesenchymal\nstem cells ( Figure 1 ).\nFigure 1 Mus musculus bone marrow mesenchymal stem cell culture. (a)\nMesenchymal stem cell morphology. Cells appear to be small cell\nbodies (fibroblast-shaped), which are long and flattened with a\nlarge nucleus; (b) swirling pattern in passage four [inverted\nmicroscope, 40×.\nMus musculus bone marrow mesenchymal stem cell culture. (a)\nMesenchymal stem cell morphology. Cells appear to be small cell\nbodies (fibroblast-shaped), which are long and flattened with a\nlarge nucleus; (b) swirling pattern in passage four [inverted\nmicroscope, 40×.\nStem cells developed in vitro can be characterized using both genotypic and\nphenotypic approaches. In this study, phenotypic identification was performed by\nimmunocytochemistry using monoclonal antibodies conjugated with fluorescein\nisothiocyanate (FITC) (F3651; Sigma, St. Louis, MO). BMMSCs were confirmed by\nthe expression of CD73, CD90, and CD105, and the absence of CD45 to exclude\nhematopoietic contamination ( Figure 2 ).\nImmunofluorescence analysis demonstrated strong CD73 and CD90 expression, while\nno green luminescence was detected for CD45, consistent with the ISCT minimal\ncriteria ( Vernon & Wilson, 1985 ;\n Iwabe  et al. , 1998 ;\n Grümmer, 2006 ).\nFigure 2 Immunohistocytochemical examination of CD 73, CD90, CD105, and CD 45.\n(a) Observation of bone marrow mesenchymal stem cells without\nfluorescence; (b) fluorescent observation of bone marrow mesenchymal\nstem cells (fluorescent microscope, 100×).\nImmunohistocytochemical examination of CD 73, CD90, CD105, and CD 45.\n(a) Observation of bone marrow mesenchymal stem cells without\nfluorescence; (b) fluorescent observation of bone marrow mesenchymal\nstem cells (fluorescent microscope, 100×).\nPKH26 fluorescence was observed in the membranes of transplanted BMMSCs within\novarian tissue ( Figure 3 ), confirming their\nability to home to the target site ( Viganò  et al. , 2018 ;  Piriyev  et al. , 2025 ).\nFigure 3 PKH26 luminescence in ovarian preparations of mice with endometriosis\nlabeled PKH26. (a) Green filter, (b) red filter, (c) red-green\nfilter (fluorescent microscope, 4.2×).\nPKH26 luminescence in ovarian preparations of mice with endometriosis\nlabeled PKH26. (a) Green filter, (b) red filter, (c) red-green\nfilter (fluorescent microscope, 4.2×).\nFigure 4  indicates that immunohistochemical\nstaining patterns of IL-1β were observed in K+, P1, P2, and P3 groups (A,\nB, C, and D, respectively). Strong positive reactions were evident in the K+\ngroup, whereas weak positive reactions were noted in the P1 group (100×).\nMost positively stained cells were macrophages (red arrows) (E) and mesothelial\ncells (black arrows) (F), both showing cytoplasmic expression of IL-1β\n(400×) ( Iwabe  et al. ,\n1998 ;  Mahnke  et al. ,\n2000 ;  Piriyev  et\nal. , 2025 ).\nFigure 4 IL-1β protein expression in peritoneal endometriosis.\nIL-1β protein expression in peritoneal endometriosis.\nFigure 5  shows that the control group\n(K + ) exhibited the highest IL-1β and IL-8 expression,\nwhereas BMMSC-treated animals (P1) showed the lowest levels. Dienogest alone\n(P3) or in combination with BMMSCs (P2) produced some reduction but remained\nmarkedly less effective than BMMSC monotherapy ( von Wolff  et al. , 1999 ;  Ueda  et al. , 2002 ;  Sanchez  et al. , 2016 ; 2017;  Muzii  et al. , 2023 ).\nFigure 5 IL-8 protein expression in peritoneal endometriosis.\nIL-8 protein expression in peritoneal endometriosis.\nTable 2  shows that the control group (K+)\ndemonstrated the highest IL-1β and IL-8 expression, whereas BMMSC-treated\nanimals (P1) exhibited the lowest. Dienogest alone (P3) or in combination with\nBMMSCs (P2) provided some reduction but were markedly less effective than BMMSC\nmonotherapy ( von Wolff  et al. ,\n1999 ;  Ueda  et al. ,\n2002 ;  Sanchez  et\nal. , 2016 ; 2017;  Muzii\n et al. , 2023 ).\nIL-1β and IL-8 expression groups.\nKruskal-Wallis analysis showed significant differences in IL-8 expression\n(H=17.236,  p <0.001). As summarized in  Table 3 , the post-hoc Bonferroni confirmed that BMMSC\nmonotherapy significantly reduced IL-8 compared with all groups\n( p =0.001), while no differences were observed among\ncontrol, Dienogest, and BMMSC + Dienogest ( p =1.000). Similarly,\nANOVA revealed significant group differences in IL-1β expression\n(F=54.808,  p <0.001). Both BMMSC and BMMSC + Dienogest were\nlower than control ( p =0.001), with BMMSC also suppressing\nIL-1β more than Dienogest and the combination\n( p ≤0.007). No differences were found between Dienogest\nand either control or BMMSC + Dienogest. Although error bars in  Figures 2  and  3  appear visually similar among groups, statistical analyses\nconfirmed significant differences. Only BMMSC monotherapy demonstrated\nsignificant reduction in IL-1β, IL-8, and lesion size compared with\ncontrols ( p <0.05). This emphasizes that statistical\noutcomes, not only visual inspection, are essential for interpretation. Overall,\nBMMSC monotherapy provided the strongest suppression of both IL-8 and\nIL-1β ( Mahnke  et al. ,\n2000 ;  Szukiewicz, 2022 ;  Muzii  et al. , 2023 ).\nPairwise Comparison of IL-8 and IL-1 β  Expression\nBetween Groups.\nHaving identified the effectiveness of different approaches, the research\nevaluated the lesion area by calculating the nodule surface area using Roaster\nimage analysis.  Figure 6  compares the\nmacroscopic appearance of implant lesions and hypervascularization of\nendometriosis in the peritoneal tissue of each group (K+, P1, P2, and P3).\nFigure 6 Macroscopic appearance of implant lesions.\nMacroscopic appearance of implant lesions.\nImageJ v1.54 (NIH, USA) analysis confirmed that BMMSC therapy was the only\nintervention associated with a significant reduction. The mean lesion size was\n6.91 mm 2 , with 50% of BMMSC-treated mice completely free of lesions,\ncompared to large, hypervascular lesions in the control group (mean 52.89\nmm 2 , maximum 227.33 mm 2 ). Neither Dienogest nor\ncombination therapy demonstrated significant effects ( Table 4 ).\nLesion area (mm 2 ) across groups.\nAs shown in  Table 5 , the Post-hoc\nBonferroni after the Kruskal-Wallis test (H=10.397,  p =0.015)\nshowed that only BMMSC therapy reduced lesion size versus control\n( p =0.021), while Dienogest and BMMSC + Dienogest did not\ndiffer. BMMSC was the only effective treatment.\nAnalysis of differences in endometriotic lesion size among groups.\n\nThe BMMSCs characterized in this study exhibited the expected immunophenotypic\nprofile of mesenchymal stem cells, expressing CD73, CD90, and CD105 while lacking\nCD45, in accordance with the International Society for Cellular Therapy (ISCT)\ncriteria ( Iwabe  et al. ,\n1998 ). This confirmed that the transplanted cells were genuine BMMSCs with\nimmunomodulatory and regenerative potential. Moreover, PKH26 fluorescence verified\ntheir ability to migrate and engraft into ovarian tissue, demonstrating successful\nhoming ( Piriyev  et al. ,\n2025 ).\nEndometriosis is strongly associated with chronic pelvic inflammation, leading to\nelevated peritoneal cytokine concentrations that disrupt folliculogenesis ( Mahnke  et al. , 2000 ;  Muzii  et al. , 2023 ). In the\npresent study, immunohistochemistry revealed high expression of IL-1β and\nIL-8 in glandular epithelial and stromal cells of the control group, whereas BMMSC\ntreatment markedly suppressed both cytokines. These findings are consistent with\nearlier evidence implicating IL-1β and IL-8 in lesion proliferation,\nangiogenesis, fibrotic progression, and reduced oocyte competence ( Mahnke  et al. , 2000 ;  Szukiewicz, 2022 ;  Piriyev  et al. , 2025 ).\nBMMSCs are recognized for their capacity to downregulate inflammatory mediators\nthrough paracrine mechanisms, including secretion of IL-10, prostaglandin E₂, and\nTGF-β, as well as inhibition of NF-κB signaling ( Xu  et al. , 2025 ). Such mechanisms likely\naccount for the robust cytokine suppression observed in the BMMSC group.\nEndometriosis is characterized by a chronic inflammatory peritoneal environment.\nElevated levels of IL-1β, IL-8, and TNF-α are consistently\ndetected in the peritoneal and follicular fluid of affected women, where they\ncontribute to angiogenesis, fibrotic remodeling, granulosa cell apoptosis, and\nimpaired steroidogenesis ( Iwabe  et\nal. , 1998 ;  Mahnke\n et al. , 2000 ;  Witz, 2000 ;  Ueda  et\nal. , 2002 ;  Khan\n et al. , 2014 ;  Da\nBroi  et al. , 2016 ;  Giacomini  et al. , 2017 ;  Sanchez  et al. , 2017 ;  Wu  et al. , 2017 ). These\ncytokines have also been implicated in poor oocyte competence and reduced IVF\nsuccess rates ( Barnhart  et al. ,\n2002 ;  Omland  et al. ,\n2005 ;  Gupta  et al. ,\n2008 ;  Da Broi  et al. ,\n2016 ;  Giacomini  et\nal. , 2017 ;  Sanchez\n et al. , 2017 ;  Wu\n et al. , 2017 ;  Viganò  et al. , 2018 ). Our findings confirm\nthat IL-1β and IL-8 expression were highest in untreated controls,\nreinforcing their central role in disease persistence.\nHormonal therapies, including Dienogest, primarily act by suppressing ovulation\nand inducing endometrial atrophy ( Andres\n et al. , 2015 ;  Bedaiwy  et al. , 2017 ;  Wu  et al. , 2022 ;  Muzii  et al. , 2023 ). Although they provide\nsymptomatic relief, they do not directly modulate the inflammatory\nmicroenvironment, and recurrence is common once treatment is discontinued ( Andres  et al. , 2015 ;  Becker  et al. , 2017 ;  Bedaiwy  et al. , 2017 ;  Szukiewicz, 2022 ;  Wu  et al. , 2022 ;  Muzii  et al. , 2023 ;  Piriyev  et al. , 2025 ). In the present\nstudy, Dienogest did not significantly reduce IL-1β, IL-8, or lesion size\ncompared with controls, consistent with clinical evidence of incomplete efficacy\n( Wu  et al. , 2022 ;\n Muzii  et al. , 2023 ;\n Piriyev  et al. ,\n2025 ).\nBMMSCs exert their therapeutic effects primarily through paracrine signaling,\nsecreting prostaglandin E₂, IL-10, and TGF-β, and inhibiting NF-κB\nactivation ( Dominici  et al. ,\n2006 ;  Uccelli  et\nal. , 2008 ;  Chen  et\nal. , 2018 ;  Ghaneialvar\n et al. , 2018 ;  Schmelzer  et al. , 2019 ;  Bian  et al. , 2022 ;  Műzes & Sipos, 2022 ;  Kulesza  et al. , 2023 ). These mechanisms suppress\ninflammatory cascades and modulate immune cell activity. In addition,\nBMMSC-derived extracellular vesicles and secretomes protect granulosa cells from\napoptosis, reduce oxidative stress, and promote follicular repair ( Chen  et al. , 2018 ;  Rajabzadeh  et al. , 2019 ;\n Bozorgmehr  et al. ,\n2020 ;  Na & Kim, 2020 ;\n Liao  et al. , 2021 ;\n Cui & Jing, 2024 ;  Kavaldzhieva  et al. ,\n2025 ). In the present study, BMMSC monotherapy achieved the most\npronounced suppression of IL-1β and IL-8 and significantly reduced lesion\nsize, consistent with their established immunomodulatory and regenerative\nroles.\nThe absence of synergistic effects between BMMSCs and Dienogest may be attributed\nto potential interactions between hormonal suppression and MSC regenerative\npathways. Hormonal therapy could interfere with MSC homing or paracrine\nactivity, thereby diminishing their therapeutic efficacy. Similar observations\nhave been reported in studies where immunomodulatory therapies showed reduced\neffects when combined with hormonal suppression 20-22 .\nBMMSC therapy offers several potential advantages over conventional management.\nBy directly targeting inflammation and tissue remodeling, BMMSCs may provide\ndisease-modifying and fertility-preserving benefits, which are particularly\nvaluable for women seeking pregnancy ( Barnhart\n et al. , 2002 ;  Omland  et al. , 2005 ;  Da Broi  et al. , 2016 ;  Sanchez  et al. , 2016 ; 2017;  Giacomini  et al. , 2017 ;\n Wu  et al. , 2017 ;\n Rajabzadeh  et al. ,\n2019 ;  Bozorgmehr  et\nal. , 2020 ;  Liao\n et al. , 2021 ;  Cui\n& Jing, 2024 ;  Kavaldzhieva\n et al. , 2025 ).\nAlthough this study offers important findings, several limitations must be\nacknowledged. First, the observation period was relatively short (14 days\npost-treatment). Second, angiogenic and fibrotic markers such as VEGF and\nTGF-β were not evaluated, despite their central roles in endometriosis\nprogression lin ( Xu  et al. ,\n2025 ). Finally, murine models cannot fully reproduce the complexity\nof human disease ( Vernon & Wilson,\n1985 ;  Grümmer, 2006 ).\nFuture studies should address these limitations through longer observation\nperiods, broader biomarker assessment, and eventual translation into clinical\ntrials.\nMechanistically, the therapeutic advantage of BMMSCs is attributed primarily to\nparacrine effects rather than cellular differentiation. Their secretome has been\nshown to suppress inflammation, inhibit angiogenesis, alleviate oxidative\nstress, and prevent granulosa cell apoptosis. By reshaping the peritoneal\nmicroenvironment, BMMSCs disrupt the self-sustaining cycle of cytokine release\nand lesion survival. The lack of synergy between BMMSCs and Dienogest may be\nexplained by hormonal suppression interfering with BMMSC regenerative and\nimmunomodulatory pathways ( Dominici  et\nal. , 2006 ;  Uccelli\n et al. , 2008 ;  Chen  et al. , 2018 ;  Na & Kim, 2020 ).\nPrevious studies have reported beneficial effects of mesenchymal stem cells in\nendometriosis models ( Chen  et\nal. , 2018 ;  Rajabzadeh\n et al. , 2019 ;  Bozorgmehr  et al. , 2020 ;  Na & Kim, 2020 ;  Liao\n et al. , 2021 ;  Cui\n& Jing, 2024 ;  Kavaldzhieva\n et al. , 2025 ). However, these investigations\ngenerally focused on MSC transplantation alone without direct comparison to\nstandard pharmacological therapy. To our knowledge, this is the first study to\nperform a head-to-head evaluation of BMMSCs, Dienogest, and their combination in\na murine endometriosis model. The novelty of our findings lies in demonstrating\nthat BMMSC monotherapy not only suppressed pro-inflammatory cytokines\n(IL-1β and IL-8) but also significantly reduced lesion size, whereas\nDienogest alone did not. This provides unique preclinical evidence of the\nsuperior disease-modifying potential of BMMSCs compared with standard care.\nFrom a clinical perspective, BMMSC therapy may represent a disease-modifying and\nfertility-preserving approach for endometriosis. The strong suppression of\nIL-1β and IL-8, together with lesion regression, provides compelling\npreclinical evidence. Nonetheless, several limitations must be acknowledged. The\nobservation period was relatively short, angiogenic and fibrotic markers such as\nVEGF and TGF-β were not assessed ( Xu\n et al. , 2025 ), and murine models cannot fully\nrecapitulate the complexity of human disease ( Vernon & Wilson, 1985 ;  Grümmer, 2006 ). Another limitation is that BMMSC\ncharacterization was performed by immunofluorescence (CD73 + ,\nCD90 + , CD105 + , CD45 + ) but was not validated\nby flow cytometry due to facility limitations. Future studies should include\nlonger-term evaluation, broader biomarker panels, and flow cytometric profiling\nto comply fully with ISCT recommendations ( Dominici  et al. , 2006 ;  Chen  et al. , 2018 ;  Na & Kim, 2020 ).\nFrom a translational perspective, BMMSC therapy holds promise as a\nfertility-preserving and disease-modifying approach for endometriosis. The\nstrong suppression of IL-1β and IL-8, together with lesion regression,\nsuggests potential applications in women seeking pregnancy while minimizing\nrecurrence. Future studies should extend the observation period, evaluate\nangiogenic and fibrotic markers such as VEGF and TGF-β ( Xu  et al. , 2025 ), and\nadvance toward clinical trials to determine long-term efficacy and safety. To\nour knowledge, this is the first study to directly compare BMMSCs with standard\nhormonal therapy in endometriosis models, reinforcing their potential as a\nregenerative therapeutic strategy.\n\nBMMSC monotherapy demonstrated superior efficacy compared with Dienogest or\ncombination therapy in suppressing IL-1β and IL-8 expression and in reducing\nlesion size in murine endometriosis models. These findings underscore BMMSCs as a\npromising regenerative, disease-modifying, and fertility-preserving therapeutic\nstrategy for the management of endometriosis ( Dominici  et al. , 2006 ;  Uccelli  et al. , 2008 ;  Chen  et al. , 2018 ;  Ghaneialvar  et al. , 2018 ;  Rajabzadeh  et al. , 2019 ;  Schmelzer  et al. , 2019 ;  Bozorgmehr  et al. , 2020 ;  Liao  et al. , 2021 ;  Bian  et al. , 2022 ;  Műzes & Sipos, 2022 ;  Kulesza  et al. , 2023 ;  Cui & Jing, 2024 ;  Kavaldzhieva\n et al. , 2025 ).","source_license":"CC-BY-4.0","license_restricted":false}