{"paper_id":"38f38814-40b1-4068-9619-25b122e988f2","body_text":"1 \n \n   \nAging increases ovarian cancer growth, metastasis, and immunosuppression that can be \nalleviated by inhibiting hedgehog signaling \n \nAsha Kumari1, Mohamed Halaby Elbahoty1, Resha Rajkarnikar1, Khushi Sureja1, Mehri \nMonavarian1, Liz Macias Quintero1, Katherine Fuh3, Daniel J Tyrrell 1, Lalita A Shevde1,2, \nCamilla Margoli1, Karthikeyan Mythreye1,2#.  \n \n1 Department of Pathology, University of Alabama at Birmingham, Birmingham, Heersink \nSchool of Medicine, Birmingham, Alabama, USA. \n2 O’Neal Cancer Center, University of Alabama, Heersink School of Medicine, Birmingham, \nAlabama, USA. \n3 Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Helen Diller \nFamily Comprehensive Cancer Center, University of California San Francisco, San Francisco, \nCA. \n \n# Correspondence: \nKarthikeyan Mythreye, Ph.D.   \nDepartment of Pathology \nO'Neal Cancer Center  \nUAB, The University of Alabama at Birmingham, Birmingham, AL, USA \nWTI 320B | 1824 Sixth Avenue South | Birmingham, AL  35294 \nE-mail: mythreye@uab.edu \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n2 \n \n \nAbstract \nOvarian cancer incidence and mortality increase with age, yet the impact of physiological aging \non tumor progression and the tumor immune microenvironment remains poorly defined. Here we \nshow using orthotopic implantation of two syngeneic models representing distinct cellular \norigins (ovarian surface epithelial and fallopian tube -derived) in young versus aged mice, that \naged hosts exhibit markedly higher tumor burden, ascites accumulation, and proliferation. \nSelective follicle depletion in young mice using VCD did not recapitulate these effects, \nindicating that age -associated microenvironmental changes beyond hormonal decline drive \ntumor growth. Spatial transcriptomics revealed distinct intratumoral heterogeneity in both age \ngroups. Comparison of CD45\n+ cells between aged and young tumors showed Hedgehog \nsignaling enrichment and immunosuppressive signatures in aged hosts, with elevated M2 \nmacrophages and Foxp3\n+ regulatory T cells . Notably, pharmacologic inhibition of Hedgehog \nsignaling with vismodegib in aged mice suppressed tumor growth, reduced metastatic spread, \nand decreased infiltration of CD206\n+ macrophages and Foxp3 + T cells while sparing CD8 + T \ncells. Our findings provide proof -of-concept that vismodegib can reduce tumor growth and \nspecific immunosuppressive populations in aged hosts, suggesting Hedgehog inhibition as a \npotential immunomodulatory strategy for older ovarian cancer patients or those tumors that \nexhibit Hedgehog pathway activation. \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n3 \n \nIntroduction  \nAge represents one of the strongest risk factors across most cancer types, with both cancer \nincidence and mortality rates increasing dramatically in patients over 65 years. However, the \nmechanisms by which physiological aging contributes to tumor progression remain incompletely \nunderstood. Aging is associated with broad alterations in the tissue microenvironment  that \ninclude but are not limited to  changes in extracellular matrix composition, accumulation of \nsenescent cells, and shifts in immune cell populations, all of which may create a pro -tumorigenic \nenvironment. Ovarian cancer represents a particularly compelling model, as incidence and \nmortality increase sharply after menopause, with a 26% higher mortality among aged patients \ncompared to other cancer types  [1, 2]. In this context, hormonal decline has historically been \nconsidered a driver of increased susceptibility, yet whether broader age -associated \nmicroenvironmental changes contribute independently remains unclear. High- grade serous \novarian carcinoma (HGSOC), the most lethal gynecologic malignancy, often originates in the \nfallopian tube with preferential metastasis to the ovary, much like other ovarian cancer subtypes  \n[3-5], making the aging ovarian microenvironment a potential key determinant of disease \nprogression for this group of cancers. Despite this clinical reality, most preclinical ovarian cancer \nstudies employ young animal mouse models,  that bypass ovarian age -specific tumor -host \ninteractions [11], limiting  our understanding of how physiological aging influences tumor \nbehavior.  \nOlder ovarian cancer patients also experience poorer outcomes, suggesting that age -\nassociated changes in the ovarian and peritoneal cavity may contribute to tumor progression and \ntreatment response to therapies [6]. Ovarian aging also extends beyond follicle decline, including \nextracellular matrix composition  changes [7] accumulation of senescent and multinucleated \novarian stromal cells and macrophages [8] and significant shifts in immune cell populations. \nThese immune shifts  in mouse models , include higher levels of  monocyte recruitment followed \nby increased alternatively activated macrophage (M2) populations [9] and a shift towards \nadaptive immunity [10]. T hese changes in the immune milieu may predispose aged ovaries to \nincreased tumor burden. Whether hormonal decline alone accounts for age -associated tumor \nprogression, or whether broader aging -related changes are required, remains unclear . Prior work \nin a 85 week old mouse model using intraperitoneal injections, showed increased tumor burden \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n4 \n \nand tumor infiltrating lymphocytes (TIL) with  altered B cell related pathways in the peritoneal \nadipose tissues [11].  However, how tumor growth, metastasis, and the immune landscape differ \nin aged versus young ovarian microenvironments has not been systematically defined . Similarly, \nsurvival analysis in a genetically engineered mouse model indicate s that nulliparous aged mice \nhad shorter survival compared to young and multiparous mice  [12] indicating that parity, and the \nlikely associated hormonal changes influence tumor outcomes.  \nHere, w e compared tumor progression and immune composition differences and \noutcomes in 60-65 week-old female mice, which represent reproductive decline corresponding to \nmenopause in women (~51 years) [13-15], versus young mice (controlling for parity) and follicle \ndepleted mice. We find that menopausal hosts support tumor progression significantly more than \nyounger hosts. Using spatial transcriptomics, we analyzed  age-associated differences in the \ntumor immune microenvironment and identified Hedgehog signaling as a pathway enriched in \nimmune cells from aged tumors. Aged tumors showed a significant increase in a spectrum of \nimmunosuppressive cell populations. Notably, pharmacologic inhibition of Hedgehog signaling \nreduced tumor progression and the presence of the same immunosuppressive cell populations . \nThese findings demonstrate that physiological aging, rather than follicular  depletion alone, is \nassociated with accelerated tumor growth and an immunosuppressive microenvironment and \nimplicate Hedgehog signaling as a potential target for immune remodeling in post -menopausal \nhosts.  \n \n \n \nResults \n \nAge-associated changes, rather than follicular depletion  alone, promote ovarian cancer \nprogression. \n \nTo investigate how physiological ovarian aging affects tumor growth and metastasis  in ovarian \nand fallopian tube- derived cancer , we compared tumor growth and metastasis in young (6- 8 \nweeks) versus aged (60-65 weeks)  nulliparous C57BL/6 mice  to eliminate physiological and \nhormonal changes  associated with pregnancy  [16]. Ovaries from a ged host mice showed \npredicted changes, including a 10.5- fold reduction in follicle count and decreased expression of \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n5 \n \novarian function markers (1.6- fold lower INHA  and 3.4-fold lower AMH  mRNA) compared to \novaries from young mice (Supplementary Figure 1A,  i- iii). To quantify differences in tumor \ngrowth in the hosts w e orthotopically implanted two distinct tumor cell models, ID8Trp53−/− \n(ovarian surface epithelial -derived) and PPNM ( fallopian tube -derived cells) into the mice \novarian bursa. Tumor growth kinetics were monitored by bioluminescence imaging in \nID8Trp53−/−implanted mice (Supplementary Fig ure 1B, i –ii) and by palpation in PPNM -\nimplanted mice. Because mice were euthanized at a uniform endpoint, we quantified progr ession \nusing a time -to-threshold analysis. The event was defined as the first attainment of ≥2.77 cm² \nperitoneal spread and yielded Kaplan Meier style estimates of the probability of remaining below \nthreshold over time. Aged hosts reached this threshold significantly earlier, with a 3.7-fold \nhigher hazard of progression. ( Supplementary Figure  1B, iii).  At endpoint (day 55 for \nID8Trp53−/−, day 42 for PPNM), aged mice displayed significantly greater tumor burden across \nall metastatic sites (Supplementary Fig ure 1B, v–vi). In the ID8Trp53−/− model, aged hosts \nshowed increased ovarian  and omental tumor mass (1.8- fold and 1.2-fold respectively), ascites \nvolume (1.6- fold), and total peritoneal/mesenteric burden (1.8- fold) relative to young hosts \n(Figure  1A; Supplementary Figure 1C, i–ii). In the PPNM model, age -related differences were \neven more pronounced, with 2.3-fold greater ovarian tumor mass, 6.2-fold greater omental tumor \nmass, 9.1-fold higher ascites volume, and 2.7-fold higher total tumor burden in aged compared to \nyoung mice (Figure 1D; Supplementary Figure 1C, iii –iv). Histological analysis revealed that \ntumors from aged hosts had higher proliferation rates  as determined by PCNA staining . \nSpecifically, ovarian tumors were 1.6-fold more proliferative  in both ID8Trp53−/−and PPNM, \nand omental tumors were 1.4-fold for ID8Trp53−/− and 1.2-fold for PPNM as compared to young \nhosts (Figure 1B-C, E-F). \n \nTo determine whether follicular depletion and associated hormonal changes alone could \naccount for increased tum or growth, we treated young mice with 4- vinylcyclohexene diepoxide \n(VCD), which selectively depletes ovarian follicles without other age -associated changes. VCD \ntreatment successfully reduced follicle number by 3.7- fold and decreased AMH  and INHA \nexpression by 14- fold and 10- fold, respectively (Supplementary Figure 1D, i- iii). ID8Trp53−/− \ncells implanted into the ovarian bursa  of follicle depleted ovaries (VCD pretreated) and follicle \nreplete ovaries (vehicle- treated) showed no significant difference at  endpoint (day 67 post -\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n6 \n \nimplantation) (Supplementary Figure 1D, iv). We confirmed no significant differences in ovarian \nor omental tumor weights, ascites volume, or total tumor burden between groups (Supplementary \nFigure 1D, v-xi). These findings demonstrate that follicle depletion in the ovaries alone do es not \nsupport tumor growth; rather, physiological age in mice strongly accelerates tumor growth in the \novaries, and metastasis to the omentum and peritoneal cavity, thereby reducing overall survival.  \n \n \n \nSpatial transcriptomics reveal  distinct intratumoral heterogeneity patterns in ovarian \ntumors from young versus aged hosts \n \nTo understand the spatial organization of the tumor microenvironment in young and aged hosts, \nand define intra and intertumoral differences, w e performed spatial transcriptomics using digital \nspatial profiling (DSP) . Regions from ovarian  tumors from young and aged mice were \nsegmented into tumor cells (CK8\n+), immune cells (CD45 +), and endothelial cells (CD31 +) by \nmultiplex immunofluorescence (Figure 2 A -B). Given the well -documented immune alterations \nassociated with ovarian cancer and ovarian aging, and the critical role of immune cells in ovarian \ncancer tumor progression, we focused our analysis for this study on comparing CD45 + enriched \nregions versus CD45+ depleted regions  (n=6 ROIs from each category ). Within young host \ntumors, CD45 + depleted regions showed enrichment  in hallmark pathways  associated with \nmetabolic reprogramming, including adipogenesis, oxidative phosphorylation, and fatty acid \nmetabolism (Figure 2C , i ). Top upregulated genes in these regions included Fxyd3  (FXYD \nDomain Containing Ion Transport Regulator 3), Prr15l  (Proline Rich 15 Like), and Flrt1  \n(Fibronectin Leucine Rich Transmembrane Protein 1) (Supplementary Table A). KEGG pathway \nanalysis revealed enrichment in glycan biosynthesis pathways (Supplementary Figure 2A , i). In \ncontrast, CD45 + enriched regions in the same young tumors displayed allograft rejection, \nepithelial-mesenchymal transition (EMT), E2F targets, and inflammatory response hallmark \npathways (Figure 2C , i). Top upregulated genes included Thbs1 (thrombospondin-1), Col8a1 \n(collagen type VIII alpha -1 chain), and Adgre1 (Adhesion G Protein- Coupled Receptor E1) \n(Supplementary Table A). KEGG pathways related to tuberculosis, malaria, and immune \ndeficiency were enriched in these regions (Supplementary Figure 2A,ii). \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n7 \n \nIn contrast to young tumors, aged host ovarian tumors displayed a distinct pattern of intratumoral \nheterogeneity. CD45+ cells depleted regions showed enrichment for MYC targets V1, cholesterol \nhomeostasis, and mTORC1 signaling hallmark pathways (Figure 2C , ii). Top upregulated genes \nincluded Clu (Clusterin), GPC1 (Glypican-1), and TGFBI  (Transforming Growth Factor Beta \nInduced) (Supplementary Table B). KEGG pathways related to protein processing were enriched \nin these regions (Supplementary Figure 2A , iii). CD45+ cells enriched regions in the same aged \ntumors were characterized by inflammatory response, allograft rejection, and KRAS signaling \nhallmark pathways (Figure 2C , ii). Top upregulated genes included HMGCS2  (3-\nHydroxymethylglutaryl-CoA Synthase 2), Jchain (Joining Chain), and Lyz2  (Lysozyme 2) \n(Supplementary Table B). KEGG pathway analysis showed enrichment in chemokine signaling, \ncytokine-cytokine receptor interaction, leukocyte transendothelial migration, and natural killer \ncell cytotoxicity pathways (Supplementary Figure 2A, iv). \nComparison of CD45+ depleted regions between the young and old ovarian tumor groups \nrevealed three common hallmark pathways: fatty acid metabolism, androgen response, and heme \nmetabolism (Supplementary Figure 2A, v) implicating these pathways as host independent . \nCD45\n+ enriched regions from the young and old ovarian groups shared the inflammatory \nhallmark (Supplementary Figure 2A, vi ). However, even in the shared inflammatory signature, \nthe genes driving this hallmark were non- overlapping, with 19 unique genes in each age group  \n(Figure 2 D, i -ii). For example, young tumors expressed Cxcl10 , associated with anti -tumor \nimmunity, whereas aged tumors expressed chemokines linked to immunosuppression, including \nCcl5, Ccl22, and Ccl17 . These findings indicate that while both young and aged tumors exhibit \nintratumoral heterogeneity, the underlying molecular programs, particularly those governing \ninflammation are fundamentally distinct, suggesting age -associated differences in immune cell \ncomposition and function. \n \nImmune cells in aged tumors are immunosuppressive with Hedgehog pathway signatures. \nTo determine whether these pathway differences reflect altered immune cell composition, we \nperformed deconvolution using CIBERSORTX  [17] . Young host tumors contained higher \nproportions of memory B cells and CD8\n+ T cells. In contrast, aged host tumors were dominated \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n8 \n \nby regulatory T cells (18% vs 4% in young) and CD4 + follicular T helper cells (Figure 2 E). \nNotably, tumor associated M2 macrophages  (TAM), that are known drivers of \nimmunosuppression in ovarian cancer  [18, 19] were markedly elevated in aged tumors (5% vs \n1% in young).  To investigate the functional programs active in these immune populations, we \nperformed GSEA on CD45 + cells themselves . In a ged host tumors CD45 + cells were \ncharacterized by IFN -α, IFN -γ, inflammatory response, and Hedgehog signaling pathways \n(Figure 2F, i-iv). In contrast, CD45 + cells from young host tumors showed enrichment in EMT, \nNotch signaling, and estrogen response early, though these did not reach statistical significance.  \nConsistent with these compositional differences, the top upregulated genes in CD45 + cells from \nyoung tumors  include GSK3A, TXNDC5, and ITGA5 ; genes that are associated with B cell \ninfiltration and prolonged survival in ovarian cancer patients (Supplementary Figure 2B , Figure \n2G, i-iii). Conversely, top upregulated genes in CD45 + cells from aged tumors  include PRM1, \nIHH (Indian Hedgehog), and TUBG1 , are associated with M2 macrophage infiltration as shown \nin other models previously [20, 21]  and poor survival outcomes in ovarian cancer patients \n(Supplementary Figure 2B, Figure 2G, iv-vi). Together, these analyses reveal that aged host \ntumors harbor immunosuppressive cell populations with Hedgehog ligand and pathways , \nidentifying this as a potential modifiable target. \n \nAged tumors are enriched for immunosuppressive macrophages and regulatory T cells.  \nTo phenotypically validate CIBERSORTX predictions, we quantified immunosuppressive cell \npopulations in ovarian and omental tumors , the later being a preferred metastatic site in ovarian \ncancer [22]. We first assessed CD45 +Arg1+ double- positive cells as a marker of \nimmunosuppressive immune populations [23, 24]. In aged hosts, ovarian tumors showed 2.8-fold \nhigher CD45 +Arg1+ cells in the ID8 Trp53−/− model (Figure 3A) and 2.42- fold higher in the \nPPNM model (Figure 3B) compared to ovarian tumors in young hosts. Similarly, omental tumors \nfrom aged hosts showed 3.52- fold ( ID8Trp53−/−, Figure 3I) and 2.08- fold (PPNM, Figure 3J ) \nhigher CD45 +Arg1+ cells. Total CD45 + infiltration followed similar trends (Supplementary \nFigure 3A -B). Since Arg1 + expressing monocytes/macrophages (CD68 +) promote tumor \nprogression and immune evasion in ovarian and other cancers [25, 26], we specifically quantified \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n9 \n \nCD68+Arg1+ cells. Ovarian tumors in aged hosts showed 3.7-fold and 2-fold higher CD68+Arg1+ \ncells compared to young hosts  in ID8Trp53−/− and PPNM models , respectively (ID8Trp53−/−, \nFigure 3C, PPNM, Figure 3D). CD68+Arg1+ cells were also numerically higher in aged omental \ntumors, reaching statistical significance in the PPNM model ( Figure 3L). The balance between \nM1 (CD80+) and M2 (CD206 +) macrophages is crucial in determining local immune milieu and \nprognosis in ovarian cancer [27, 28]. Aged host tumors showed 2.69-fold higher CD206+ cells in \novarian tumors (ID8Trp53−/−, Figure 3E) and 2.26-fold higher in the PPNM model (Figure 3F ). \nOmental tumors showed similar elevations in CD206 + cells in aged hosts (Figures 3M, 3N ). \nSupporting these findings, ascites fluid from aged mice (ID8Trp53 −/− model) showed a 2.9- fold \nhigher ratio of CD206+ to CD80+ macrophages (Supplementary Figure 3C). \nWe next evaluated regulatory T cells, a critical population known to inhibit effective anti-\ntumor immunity [29] and predicted to be elevated from our deconvolution analysis  (Figure 3H). \nFoxp3+ cells were 3 to 3.4-fold higher in aged host ovarian tumors across both models (Figures \n3G, 3H) and 2.28 -fold higher in omental tumors (Figures 3O , 3P). Consistent with tumor tissue \nfindings, ascites fluid from aged mice showed 2.5- fold higher CD25 +Foxp3+ cells \n(Supplementary Figure 3D).  Together, these findings demonstrate that the aged tumor \nmicroenvironment is dominated by immunosuppressive cell populations  including M2 \nmacrophages and regulatory T cells , across primary  (ovary) and metastatic  (omentum) sites, \nproviding a cellular basis for the enhanced tumor growth observed in aged hosts. \n \nHedgehog inhibition suppresses ovarian tumor growth and metastasis in aged models.  \nBased on the immune suppressive cell populations, accelerated tumor growth and predominance \nof Hedgehog signaling in immune cells from aged host tumors, we aimed to test the therapeutic \nbenefit of inhibiting Hedgehog (Hh) signaling in older hosts using Vismodegib (Figures 4A,4B). \nAs mice were euthanized at a uniform endpoint, disease progression was assessed using a time-\nto-threshold probability analysis of disease progression using abdominal girth changes. This \nyielded Kaplan-Meier style estimates of the p robability of remaining below the abdominal girth \nthreshold. Vismodegib- treated mice showed delayed disease progression, remaining below \nthreshold until day 41 compared to day 16 in vehicle -treated mice (Figure 4C).  Vismodegib \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n10 \n \ntreatment reduced metastatic burden , the number of mice that  developed measurable ascites \nvolume (33.3% of vismodegib vs 88.8% of vehicle-treated, Figure 4D) , diaphragmatic \nmetastases (66.6 % of vismodegib vs 87.5% of vehicle- treated mice Figure 4E), and peritoneal \nwall metastases ( 22.2 % of vismodegib mice vs 50 % of vehicle -treated mice Figure \n4F). Ovarian and omental tumor weights trended lower in vismodegib treated mice, ovarian  \nweights did not reach statistical significance (Figure 4G -H). However, tumor proliferation was \nmarkedly reduced as PCNA-positive cells were 1.47- fold lower in ovarian tumors and 1.86- fold \nlower in omental tumors, with corresponding reductions in staining intensity (1.5- fold and 2.15-\nfold, respectively, Figure 4G ,4H). These findings demonstrate that Hh inhibition delays tumor \nprogression and reduces metastatic spread in aged hosts.  \n \nHedgehog inhibition reduces immunosuppressive macrophages and regulatory T cells  in \naged tumors. \nGiven the enrichment of Hedgehog signaling in immune cells from aged tumors, we next asked \nwhether the therapeutic effects  (Figure 4) were accompanied by changes in immunosuppressive \ncell populations in ovarian and omental tumors. In ovarian tumors, vismodegib treatment \nreduced total CD45\n+ cells by 1.85- fold (Supplementary Figure 5A) and CD45 +Arg1+ cells by \n1.28-fold (Figure 5A), although the latter did not reach statistical significance. Omental tumors \nshowed only marginal changes in these populations (Supplementary Figure 5B, Figure 5B ). Hh \ninhibition has been shown to modulate macrophage polarization in other cancer types  [30, 31]. \nConsistent with this, CD206 + M2 macrophages were 2.4- fold lower in vismodegib- treated \novarian tumors (Figure 5E) but did not reach statistical significance in vismodegib -treated \nomental tumors (Figure 5 F).  Flow cytometry analysis of mice ovaries supported these findings, \nshowing trends toward higher M1 (F4/80 +CD80+) and lower M2 (F4/80 +CD206+) macrophages \nin vismodegib- treated mice (Supplementary Figure 4C, i -ii). However, the CD68 +Arg1+ \nmonocytic population was not significantly altered in either ovarian or omental tumors (Figures \n5C, 5D). Hh signaling can also modulate T cell responses in breast cancer [32] . We found that \nFoxp3\n+ regulatory T cells were markedly reduced in vismodegib- treated mice leading to 2.89-\nfold reduction in ovarian tumors (Figure 5G) and 1.91-fold lower in omental tumors (Figure 5H)  \nas compared to vehicle treated mice. Together, these findings demonstrate that pharmacologic \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n11 \n \ninhibition of Hh signaling in aged hosts suppresses ovarian tumor growth and metastatic spread, \naccompanied by reductions in CD206+ M2 macrophages and Foxp3+ regulatory T cells. Notably, \nHh inhibition effectively reduced these key immunosuppressive populations, suggesting that Hh \ninhibition selectively targets immunosuppressive rather than effector populations  in these \nmodels. \n \nDiscussion \n \nAge is among the strongest risk factors for ovarian cancer, with incidence and mortality rising \nsharply after menopause, and older patients experiencing poorer outcomes compared to younger \npatients [33-35]. While follicle depletion is a hallmark of ovarian aging, whether this change \nalone accounts for increased tumor susceptibility or whether additional age-associated alterations \nare required has remained unclear. Using orthotopic implantation of two distinct tumor models in \nyoung and aged mice , the latter corresponding approximately to human perimenopausal age, we \nfound that aged hosts exhibited significantly higher tumor burden, increased ascites \naccumulation, and elevated tumor proliferation (Figure 1 A-F), though direct clinical translation \nrequires validation in patient samples . Our data further identified enrichment of Hedgehog \nsignaling in immune cells from  aged tumors and demonstrated that pharmacologic inhibition of \nthis pathway reduces tumor burden and immunosuppressive cell populations, suggesting a \npotential immunomodulatory strategy for older patients with ovarian cancer (Figure 2F). \n \nOvarian aging encompasses far more than follicular decline, including extracellular \nmatrix remodeling, accumulation of senescent stromal cells, significant shifts in immune \ncomposition, and elevated pro- inflammatory cytokines contributing to 'inflammaging'[36]. W e \nused 4-vinylcyclohexene diepoxide (VCD), which selectively destroys primordial and primary \nfollicles, reducing estrogen, progesterone, AMH, and Inhibin A while leaving the young \nmicroenvironment otherwise intact [37].VCD- treated young mice showed no increase in tumor \nburden compared to controls, indicating that tumor -promoting effects in aged hosts arise from \nmicroenvironmental changes beyond hormonal decline (Supplementary Figure 1D). Prior studies \nusing VCD or ovariectomy in ovarian cancer contexts have yielded variable results. VCD -\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n12 \n \ninduced follicle depletion in TgCAG -LS-TAg mice or ovariectomy before SKOV3/OVCAR3 \nintraperitoneal implantation resulted in slower tumor growth, effects attributed to elevated \ngonadotropins [38] . In contrast, VCD treatment followed by carcinogen exposure induced \novarian neoplasms [39]. These studies used intraperitoneal injections or carcinogen -based \napproaches rather than orthotopic implantation into the ovarian microenvironment, which \nprovides insight into how the aged ovarian niche itself shapes tumor growth. Our orthotopic \nstudies establish that follicular depletion is insufficient but do not identify which specific age -\nassociated changes are necessary or sufficient. The relative contributions of the local ovarian \nmicroenvironment versus systemic aging effects, including alterations in circulating immune \ncells, remain to be determined. \n \nSpatial transcriptomics of the tumors revealed that b oth age groups displayed \nheterogeneous tumor landscapes with distinct CD45 + enriched and depleted regions; however, \nthe pathway signatures defining these regions differed substantially between age groups. In \nyoung host tumors, CD45 + depleted regions showed hallmarks of metabolic reprogramming \nwhile CD45 + enriched regions exhibited epithelial -mesenchymal transition (EMT) signatures. \nEmerging evidence indicates that EMT can serve as a tumor escape mechanism  in response to \nanti-tumor immune pressure, enabling tumor cells to evade T cell-mediated killing through MHC \nclass I downregulation and immune checkpoint upregulation (Figure 2C, i) [40-42]. The presence \nof EMT signatures in immune -enriched regions of young tumors, alongside anti -tumor \nchemokines such as Cxcl10 , suggests that tumors in young hosts may face greater immune \npressure and utilize EMT as an adaptive response. In contrast, aged host tumors showed MYC \ntargets and cholesterol homeostasis enrichment in CD45 + depleted regions, and inflammatory \nresponse signatures in CD45 + enriched regions (Figure 2C, ii). Notably, while both young and \naged tumors exhibited inflammatory hallmarks  in immune -enriched regions, the genes driving \nthese signatures were entirely non-overlapping (Figure 2D). This distinction has functional \nimplications such as with Cxcl10, enriched in young tumors, which is associated with anti-tumor \nimmunity and a favorable prognosis in ovarian cancer  [43], whereas chemokines enriched in \naged tumors, including Ccl5, Ccl22, and Ccl17 promote immunosuppression through regulatory \nT cell recruitment and are associated with chemoresistance and poor response to immunotherapy  \n[44-47]. These findings reveal that similarly labeled inflammatory signatures obscure \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n13 \n \nfundamentally different immune programs  indicating that  young tumors appear to face active \nanti-tumor immunity necessitating escape mechanisms, whereas  aged tumors exhibit signatures \nconsistent with an already established immunosuppressive microenvironment. Whether EMT -\ntargeted approaches could enhance anti -tumor immunity specifically in younger patients, or \nwhether reversing the immunosuppressive chemokine milieu could benefit older patients, \nwarrants further investigation. \n \nConsistent with our transcriptomic findings, aged host tumors exhibited significantly \nhigher infiltration of immunosuppressive populations across both tumor models, including \nCD45+Arg1+ cells, which were significantly higher in aged ovarian and omental tumors \ncompared to young hosts  (Figures 3A,3B,3I,3J) . This enrichment was driven by CD68 +Arg1+ \nmonocytes and CD206 + tumor- associated macrophages (TAMs), both of which were \nsignificantly elevated as well in aged tumors. TAMs, particularly those with M2 polarization, are \nestablished drivers of immunosuppression in ovarian cancer and are associated with poor \nprognosis, therapy resistance, and tumor progression [18, 48, 49]. The balance between M1 (anti-\ntumor) and M2 (pro-tumor) macrophages is critical in shaping the local immune milieu , and our \nfinding of a higher CD206+/CD80+ ratio in ascites from aged mice further supports a shift toward \nimmunosuppressive macrophage polarization in aged hosts  (Supplementary Figure 3C) . In \nparallel, Foxp3+ regulatory T cells were more abundant in aged tumors across both ovarian and \nomental sites (Figures 3G,3H,3O,3P). Regulatory T cells inhibit effective anti- tumor immunity \nand have been associated with poor survival , specifically in ovarian cancer , compared to other \nmalignancies [50]. The concurrent enrichment of M2 macrophages and regulatory T cells in aged \ntumors suggests that these populations may cooperate to establish an immunosuppressive \nmicroenvironment that permits accelerated tumor growth. Whether this immunosuppressive state \nreflects intrinsic properties of aged immune cells, tumor  or stroma mediated reprogramming of \ninfiltrating immune cells, or altered recruitment patterns driven by the aged host remains to be \ndetermined. \n \nThe Hedgehog pathway, while best known for its roles in embryonic development and \ntissue homeostasis, has emerged as an important modulator of immune function in the tumor \nmicroenvironment [51, 52] . Recent studies in breast cancer have begun to elucidate how \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n14 \n \nHedgehog signaling shapes immunosuppressive cell populations. Hedgehog signaling regulates \nmacrophage metabolism and polarization, promoting the M2 phenotype through altered O -\nGlcNAcylation and mitochondrial dynamics [20] . Additionally, Hedgehog signaling promotes \nTreg differentiation and activity, and its inhibition can drive Treg -to-Th17 conversion through \nmetabolic rewiring [32]. These mechanistic insights from breast cancer models suggest potential \npathways through which Hedgehog activation in aged ovarian tumors could sustain \nimmunosuppressive M2 macrophages and Tregs. O ur data demonstrate correlation , and whether \nHedgehog activation directly drives immunosuppressive cell polarization in aged ovarian tumors, \nor whether it represents a consequence of the altered aged microenvironment, remains to be \ndetermined.  \n \nInhibition of Hedgehog signaling using Vismodegib delayed disease progression in older \nmice and decreased infiltration of CD206 + M2 macrophages and Foxp3 + regulatory T cells , the \ntwo immunosuppressive populations most enriched in aged tumors. CD68 +Arg1+ monocytes \nhowever were not significantly affected, suggesting Hedgehog inhibition may preferentially \ntarget specific immunosuppressive subsets rather than broadly depleting myeloid cells (Figure 5). \nThese findings contrast with a phase II clinical trial in which vismodegib failed to improve \nprogression-free survival as maintenance therapy in ovarian cancer patients in second or third \ncomplete remission [53]. Several factors may account for this discrepancy. First, patients in that \ntrial were not selected based on age or profiled for Hedgehog pathway activation; our data \nsuggest Hedgehog signaling is specifically enriched in the aged tumor microenvironment, and \npatients without pathway activation may derive little benefit. Second, the clinical trial employed \nvismodegib as maintenance therapy after achieving remission, whereas our intervention began \nearly after tumor implantation when the immunosuppressive microenvironment was actively \ndeveloping. Perhaps most importantly, the immunomodulatory effects we observe suggest that \nHedgehog inhibition may be best suited not as monotherapy but in combination with immune \ncheckpoint inhibitors. By reducing M2 macrophages and Tregs, Hedgehog inhibition could \npotentially render tumors more responsive to checkpoint blockade , a combinatorial strategy that \nwarrants future investigation, particularly in older patients whose tumors may be most dependent \non this pathway. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n15 \n \nIn summary, we demonstrate that physiological aging creates an immunosuppressive ovarian \ntumor microenvironment characterized by M2 macrophage and Treg infiltration, with Hedgehog \nsignaling emerging as a potential mediator. These findings have several implications for future \ninvestigation. Parabiosis or bone marrow transplantation experiments could distinguish local \nfrom systemic contributions to age-associated tumor promotion. The distinct immune landscapes \nin young versus aged tumors , active immune engagement and potential for  EMT-mediated \nescape in young hosts versus pre -established immunosuppression in aged hosts  suggest that age-\nstratified therapeutic approaches may be warranted. Most immediately, our data provides \nrationale for clinical investigation of Hedgehog inhibitors in combination with immune \ncheckpoint blockade, with patient selection based on age and tumor Hedgehog pathw ay \nactivation status. \n \n \nMaterials and Methods \n \nCell lines, reagents and media:  PPNM (p53 −/−l R172H Pten−/−Nf1−/−MycOE) cells were kindly \nprovided by Dr. Robert A Weinberg, Whitehead Institute for Biomedical Research, Cambridge, \nMA, USA, through an MTA and were cultured as  described here [54]. For ID8 Trp53−/−- Luc, \nID8Trp53−/−cells [55] were transduced with Luc FUW -firefly luciferase -EGFP (FUW -FFLuc-\neGFP) lentiviral construct  gene, generously gifted by Dr.  Stewart, Department of Cell Biology \nand Physiology, Washington University School of Medicine, St Louis, to generate  ID8Trp53−/−  \nLuc cells as described  [56]. ID8Trp53−/− cells were cultured in Dulbecco’s modified Eagle’s \nmedium supplemented with 4% fetal bovine albumin, 100U penicillin , and streptomycin, \n5 μg/mL of insulin, 5 μg/mL of transferrin, and 5 ng/mL of sodium selenite . Both cell lines were \nmaintained at 37 °C with 5% CO2 in a humidified incubator, routinely checked for mycoplasma \nand experiments were conducted within 3–6 passages, depending on the cell line.  \n \nAnimals: All animal procedures were conducted in accordance with Institutional Animal Care \nand Use Committee (IACUC) guidelines at the University of Alabama at Birmingham (UAB), \nUSA. C57BL/6J  (Cat no. 000664; The Jackson Laboratory, Farmington, CT, USA,)  or \nC57BL/6N (National Institute of Aging, Bethesda, MD, USA)) nulliparous females aged 6 or 65 \nweeks, were used in this study. All animals were housed in the Wallace Tumor Institute animal \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n16 \n \nfacility at UAB with food and water ad libitum in groups of 5 mice per cage at 24°C under a 12-\nhour light/dark cycle.  \n \nEstablishment of follicle depleted ovaries : Young female mice (6 weeks old) were randomly \ndivided into two groups (n = 5 per group): 4- vinylcyclohexene diepoxide (VCD, Cat no. 94956, \nSigma-Aldrich)) treated and vehicle control. The VCD group received intraperitoneal injections \nat 160 mg/kg in sesame oil for 15 consecutive days  [57] while control mice received sesame oil \nalone, following VCD treatment, mice were aged until postnatal day 60, at which point ovaries \nwere collected for histological analysis and qRTPCR. \n \nIntrabursal implantation of ovarian cancer cells: 2x10 6 ID8Trp53−/− or PPNM  cells suspended \nin 8µl phosphate-buffered saline were injected into the bursa of the left mouse ovary. M ice were \nanesthetized prior to implantation using 2.5% isoflurane mixed with 2L/min oxygen and \nadministered buprenorphine 0.07 mg/kg as ana lgesic pre- and post -surgery for up to 48 hours \nfollowing institutional guidelines.  \n \nVismodegib treatment: Aged female mice (65 weeks old) were gavaged with 100µl (3mg/mouse) \nof Vismodegib (GDC-0449) (Selleckchem, Cat no. S1082 ) or DMSO as vehicle, three times a \nweek as described previously [20] after 48 hours of ID8 Trp53−/− cell i njection in the ovaries. \nMice were closely monitored for post -surgical recovery over 5 days, and tumor growth was \ntracked every 10 days using whole -body bioluminescence imaging (BLI) on a Perkin Xenogen \nIVIS Imaging System following intraperitoneal injection of luciferin (50 mg/kg). \nTime-to-Threshold Disease Progression Analysis (Fixed Biological Threshold) : Disease \nprogression in comparing young versus old mice was quantified using a time -to-threshold \nframework in which the event was defined as the first attainment of a pre -specified biological \ndisease burden. For peritoneal dissemination studies, the threshold was defined a priori as ≥2.77 \ncm² peritoneal spread or a corresponding abdominal girth criterion reflecting maximum tolerable \ntumor burden. For each mouse, time to reach this threshold was recorded as the event. Kaplan –\nMeier curves were generated and compared using the log -rank test, with hazard ratios estimated \nby Cox proportional hazards modeling. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n17 \n \nTime-to-Threshold Disease Progression Analysis (Control -Derived Reference Threshold) : In \nstudies using abdominal girth as the progression metric, a control -derived reference threshold \nwas employed. The threshold was defined exclusively from the control cohort as the mean \nchange in abdominal girth from baseline to endpoint. This control -derived benchmark was then \napplied uniformly to both control and treatment groups. For each mouse, the time to reach this \nbenchmark was scored as the event . Kaplan Meier curves were generated and compared using \nthe log-rank test, with hazard ratios estimated using Cox proportional hazards regression. \n \nSpatial Nanostring GeomMx analysis of Ovarian tumors : RNA profiling of ovarian tumors from \nyoung and aged mice was performed using the mouse Whole Transcriptome Atlas probes using \nGeoMx™ DSP Nanostring technology per manufacturer’s guidelines. In brief, paraffin-\nembedded 5µm ovarian sections were baked at 60°C for 1 hour. They were stained with labeled \nGoat anti-mouse CD45 (Cat no. AF114 1:100, Novus biologicals) primary antibody using FITC \nConjugation Kit (Cat no. ab188285, Abcam) and Labeled Rat anti -mouse CK8 primary antibody \n(Cat no. Sc 8010, 1:100, Santa Cruz Biotechnology) using APC Conjugation Kit  (Cat no. \nab201807, Abcam) for immune and tumor cells detection respectively. Following primary \nantibodies incubation for 2 hours at room temperature, nuclei were stained with Syto TM 82, \nOrange fluorescent Nucleic acid stain (Cat no. S11363, Sigma). Regions of interest (ROIs) were \nselected using a combination of immunofluorescent staining and H&E staining. ROIs were \nclassified as general areas to obtain an un biased overview of the microenvironment such as \ntumor cells (CK8+) or immune cells (CD45+). Upon ROI collection on the GeoMx Digital Spatial \nProfiler (DSP), libraries were prepared and sequenced on the  Illumina NovaSeq instrument \n(UAB, Genomics core facility). FASTQ files were uploaded to  the Base Space Illumina hub and \nconverted to digital count conversion (dcc) files using the GeoMx® NGS Pipeline (v2.0.21) on \nIllumina DRAGEN. The analysis was performed in R following the manufacturer’s analysis \ncode \n(https://www.bioconductor.org/packages/release/workflows/vignettes/GeoMxWorkflows/inst/do\nc/GeomxTools_RNA-NGS_Analysis.html). ROIs with less than 80% sequencing alignment or \nless than 50% sequencing saturation were removed from further analysis. The Grubbs outlier test \nwas used to identify outlier probes. The limit of quantification (LOQ) was defined as two \nstandard deviations above the geometric mean of the negative probes. ROIs were then divided \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n18 \n \ninto general areas, and CD45 + immune cell genes below LOQ in at least 10% of ROIs were \nremoved from further analysis. Normalization was performed using a signal -based quartile \nnormalization method, where individual counts are normalized against the 75th percentile of \nsignal from their ROI. Group comparisons were done using a mixed linear model as previously \ndescribed. Normalized counts can be accessed here:  doi: 10.17632/j4f8r6z3b3.1.. Normalized \ngene counts were used for Gene Set Enrichment  Analysis (GSEA) using GSEA 4.3.3 (Broad \nInstitute, UCSD, San Diego) [58], and immune cell abundances analysis CIBERSORTX   was \nused after converting gene IDs from mouse to human on SynGO  (Synaptic Gene Ontologies and \nannotations) [17, 59]. Immune infiltration association with ovarian cancer patient survival was \nperformed using Timer 2.0 [60].  Heat maps were generated using  Heatmapper2 [61] and KEGG \npathway analysis was performed using the SRPlot platform [62]. \n \nImmunohistochemistry: Ovarian and omental tumor sections were stained as previously \ndescribed [55]. Antigen retrieval was performed by boiling the section in sodium citrate buffer \n(pH 6.0) for 30 minutes, followed by incubation in 3% hydrogen peroxide at room temperature \nfor 15 minutes to block endogenous peroxide activity; sections were then washed twice in PBS \nfor 5 minutes each and blocked with Background Punisher (BioCare, Cat. no. BP974) for 15 \nminutes at room temperature. Sections were then incubated at 4°C overnight in primary \nantibodies for Proliferation cell nuclear antigen PCNA (586, PC10, 1:500), and CD206 (24595, \nE6T5JXP,1:200) diluted in Da Vinci Green diluent (BioCare Cat no. PD900). Detection was \nperformed using the MACH4 universal HRP-polymer (BioCare, cat no. MACH 4™) kit followed \nby the Betazoid DAB Chromogen Kit (BioCare, cat no BDB2004) as per manufacturer’s \ninstructions. Sections were counterstained with hematoxylin for 1 minute, dehydrated through \nethanol gradients (70%, 90%, and 100%), and cleared in Xylene for 1 minute, followed by \nmounting in Fisher Chemical TM Paramount TM mounting medium (SP15-100) for microscopic \nanalysis. A minimum  of five regions per tumor core from each mouse were analyzed using \nQuPath (v0.5.0) Bioimage analysis software [63] and presented as a percentage of total cells. \n \nImmunofluorescence: For Immunofluorescent detection of immune cells, Ovarian and omental   \nsections were processed similarly to IHC until antigen blocking, omitting peroxidase treatment. \nSections were incubated overnight at 4°C with primary antibodies, including Goat anti -mouse \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n19 \n \nCD45 (Cat no. AF114 1:100, Novus biologicals) for immune cells , anti -rabbit Arginine1(93668, \nD4E3MTM XP,1:100) and rat anti -mouse CD68 (MCA341, FA -11,1:100) for macrophage \ndetection and Foxp3 (D6O8R) 12653 (1:100) for regulatory T cells. Antibodies were  diluted in \nDa Vinci Green diluent (Biocare Cat no. PD900). Following primary antibody incubation, \nsections were washed twice in PBS with 1% tween -20 for 5 minutes and incubated with \nsecondary antibodies, Goat anti -Rabbit IgG Alexa Fluor ™ 594 (A11012), Goat anti -Rabbit IgG \n488 (A11008), and Donkey anti -rat IgG 488 (A48269TR), at 1:500 each for 2 hours at room \ntemperature. Nuclei were counterstained with DAPI for 5 minutes, and sections were mounted \nusing ProLong™ Gold Antifade Mountant (Cat no. P36930). Multiple Immunofluorescence \nimages  (3-5  per tumor core from each mouse) were analyzed using QuPath (v0.5.0) Bioimage \nanalysis software[28]. Double-positive cells were manually counted using Fiji (ImageJ) software \n[29] by an investigator blinded to the study. Results were presented as the percentage of total \npositive cells.  \n \nFlow cytometry:  A scites fluid was collected from the mice's peritoneal cavity, and cells were \nseparated from fluid after centrifugation at 1200 rpm at 4°C. Red blood cells were lysed using \nRBC lysis with incubation at room temperature for 10 minutes. For the Vismodegib inhibition \nexperiment ovaries and omentum were collected at endpoints (Day 42), after perfusing mice with \nPBS, digested using Collagenase II (Thermo Fisher Cat no.17101015, 1.0   mg/mL), and DNase I \n(Sigma Cat no. 10104159001, 25  μg/mL) for 30 minutes at 37°C, passed through 70µm nylon \nmesh and RBCs were lysed as described above. Freshly isolated cells were stained sequentially \nwith Live/Dead  (Zombie Yellow TM, Cat no, 423103, Biolegend) for 15 minutes at room \ntemperature followed by Fc -receptor blockade (rat ani -mouse CD16/CD32, Cat no 553141) for \n10 minutes on ice and fluorophore -labeled antibodies: PerCP/Cyanine5.5- labeled, anti -mouse \nCD45 Cat no. 103131 or FITC -labeled anti -mouse CD45 Cat no. 147709, PerCP /Cyanine5.5 \nlabeled anti-mouse F4/80 Cat no.123107, APC -labeled anti-mouse CD80 Cat no. 104714, APC -\nlabeled anti-mouse CD206 Cat no.141706, PerCP /Cyanine5.5- labeled anti-mouse CD3 Cat no.  \n100218, PE/Cy7-labeled anti-mouse CD4 Cat no. 100422, APC labeled anti -mouse CD8 Cat no. \n10071, APC -labeled anti -mouse CD25 Cat no. 102012, PE -labeled anti -mouse Foxp3 cat no. \n126403 for 30 minutes on ice in flow cytometry buffer (FACS) containing 2% fetal bovine \nserum in PBS at recommended concentrations. For Foxp3, intracellular staining, cells were \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n20 \n \npermeabilized using True -Nuclear™ Transcription Factor buffer kit (Cat no. 424401, \nBiolegend). Post -staining, cells were resuspended in FACS buffer and analy zed on BD \nFACSymphony flow cytometer, and data were analyzed using FlowJo TM v10 software (TreeStar \nInc., Ashland, OR USA). \n \nStatistical analysis : All data were compared using unpaired Student t- tests and presented as \nmean ± SEM. Data was analyzed using Graph Pad Prism 9.0 (La  Jolla, CA, USA), and statistical \nsignificance was considered at p < 0.05. \n \n \nAcknowledgements \n \nFunding for this work was provided  in part by NIHR01CA219495 and O’Neal Invests grant to \nMythreye Karthikeyan (KM). This study was supported by National Institutes of Health (NIH) \naward NIHR00AG068309 an AFAR Grant for Junior Faculty awardee  to Daniel Tyrrell (DJT).  \nThe funders had no role in study design, data collection and analysis, decision to publish, or \npreparation of the manuscript . We want to acknowledge the UAB flow cytometry core facility,  \n(AI027767),  O’Neal Cancer Center, P30CA013148, and shared instrument grant S100D032296. \nWe also acknowledge the Preclinical Imaging Shared Facility (P30CA013148, 1S10OD021697), \nHigh Resolution Imaging Facility  and  the Pathology Core Research Lab at UAB for assistance \nwith processing the histological specimens . Schematics were made using Biorender  (licensed \nagreement to KM and UAB) . We thank  Manan Nayyar and Emily O’Brian for technical \nassistance, Dr. Arend for sharing cell lines  and Dr. Darshan Shimoga Chandrashekar for helpful \ndiscussions. \n \n \nAuthor contributions \n \nConceptualization: KM, AK. Investigation: AK, CM. MHE, RR, KS, MM, LMQ, DJ T, LAS, \nCMM, RR, AK, LQM, SS, FM, MC. Analysis: KM, CM, AK, KS, Reagents: AK, KF, DJ T, \nLAS Resources/Supervision: KM, CM, DT, LAS . Writing : original draft: AK, KM  Writing: \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n21 \n \nreview & editing: all authors. Funding acquisition as described in acknowledgements and project \nadministration: KM \n \n \nData availability statement  \n \nGeoMx data are available at doi: 10.17632/j4f8r6z3b3.1. Additional data are available from the \ncorresponding author upon reasonable request. \n \nKeywords: Aging, Ovarian cancer, menopausal changes, vasculatures, Angiogenesis \n \nReferences \n \n1. Mancebo G, Sole-Sedeno JM, Fabregó B, Pinto G, Vizoso A, Alvarez M, Sabaté -Garcia \nRA and Miralpeix E. Influence of Age on Treatment and Prognosis in Ovarian Cancer Patients. \nCancers (Basel). 2025; 17(9). \n2. Withrow DR, Nicholson BD, Morris EJA, Wong ML and Pilleron S. Age -related \ndifferences in cancer relative survival in the United States: A SEER -18 analysis. Int J Cancer. \n2023; 152(11):2283-2291. \n3. Labidi- Galy SI, Papp E, Hallberg D, Niknafs N, Adleff V, Noe M, Bhattacharya R, \nNovak M, Jones S, Phallen J, Hruban CA, Hirsch MS, Lin DI, et al. High grade serous ovarian \ncarcinomas originate in the fallopian tube. Nature Communications. 2017; 8(1):1093. \n4. Köbel M and Kang EY. The Evolution of Ovarian Carcinoma Subclassification. Cancers \n(Basel). 2022; 14(2). \n5. Kurman RJ and Shih Ie M. The origin and pathogenesis of epithelial ovarian cancer: a \nproposed unifying theory. Am J Surg Pathol. 2010; 34(3):433-443. \n6. Ali AT, Al -Ani O and Al -Ani F. Epidemiology and risk factors for ovarian cancer. Prz \nMenopauzalny. 2023; 22(2):93-104. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n22 \n \n7. Briley SM, Jasti S, McCracken JM, Hornick JE, Fegley B, Pritchard MT and Duncan FE. \nReproductive age-associated fibrosis in the stroma of the mammalian ovary. Reproduction. 2016; \n152(3):245-260. \n8. Maruyama N, Fukunaga I, Kogo T, Endo T, Fujii W, Kanai -Azuma M, Naito K and \nSugiura K. Accumulation of senescent cells in the stroma of aged mouse ovary. J Reprod Dev. \n2023; 69(6):328-336. \n9. Zhang Z, Schlamp F, Huang L, Clark H and Brayboy L. Inflammaging is associated with \nshifted macrophage ontogeny and polarization in the aging mouse ovary. Reproduction. 2020; \n159(3):325-337. \n10. Ben Yaakov T, Wasserman T, Aknin E and Savir Y. Single -cell analysis of the aged \novarian immune system reveals a shift towards adaptive immunity and attenuated cell function. \nElife. 2023; 12. \n11. Loughran EA, Leonard AK, Hilliard TS, Phan RC, Yemc MG, Harper E, Sheedy E, \nKlymenko Y, Asem M, Liu Y, Yang J, Johnson J, Tarwater L, et al. Aging Increases \nSusceptibility to Ovarian Cancer Metastasis in Murine Allograft Models and Alters Immune \nComposition of Peritoneal Adipose Tissue. Neoplasia. 2018; 20(6):621-631. \n12. Hou X, Zhai Y, Hu K, Liu CJ, Udager A, Pearce CL, Fearon ER and Cho KR. Aging \naccelerates while multiparity delays tumorigenesis in mouse models of high -grade serous \ncarcinoma. Gynecol Oncol. 2022; 165(3):552-559. \n13. Flurkey K, Mcurrer J and Harrison D. (2007). Mouse Models in Aging Research. Mouse \nBiomed. Res. Elsevier Amsterdam). \n14. Bhattarai T, Datta S, Chaudhuri P, Bhattacharya K and Sengupta P. Effect of \nprogesterone supplementation on post -coital unilaterally ovariectomized superovulated mice in \nrelation to implantation and pregnancy. Asian Journal of Pharmaceutical and Clinical Research. \n2014:29-31. \n15. Wang S, Lai X, Deng Y and Song Y. Correlation between mouse age and human age in \nanti-tumor research: Significance and method establishment. Life Sciences. 2020; 242:117242. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n23 \n \n16. Hou X, Zhai Y, Hu K, Liu C -J, Udager A, Pearce CL, Fearon ER and Cho KR. Aging \naccelerates while multiparity delays tumorigenesis in mouse models of high -grade serous \ncarcinoma. Gynecologic Oncology. 2022; 165(3):552-559. \n17. Donovan MKR, D’Antonio- Chronowska A, D’Antonio M and Frazer KA. Cellular \ndeconvolution of GTEx tissues powers discovery of disease and cell -type associated regulatory \nvariants. Nature Communications. 2020; 11(1):955. \n18. Yuan X, Zhang J, Li D, Mao Y, Mo F, Du W and Ma X. Prognostic significance of \ntumor-associated macrophages in ovarian cancer: A meta- analysis. Gynecol Oncol. 2017; \n147(1):181-187. \n19. Hensler M, Kasikova L, Fiser K, Rakova J, Skapa P, Laco J, Lanickova T, Pecen L, \nTruxova I, Vosahlikova S, Moserova I, Praznovec I, Drochytek V, et al. M2- like macrophages \ndictate clinically relevant immunosuppression in metastatic ovarian cancer. J Immunother \nCancer. 2020; 8(2). \n20. Hinshaw DC, Hanna A, Lama -Sherpa T, Metge B, Kammerud SC, Benavides GA, \nKumar A, Alsheikh HA, Mota M, Chen D, Ballinger SW, Rathmell JC, Ponnazhagan S, et al. \nHedgehog Signaling Regulates Metabolism and Polarization of Mammary Tumor- Associated \nMacrophages. Cancer Res. 2021; 81(21):5425-5437. \n21. Roque DM, Buza N, Glasgow M, Bellone S, Bortolomai I, Gasparrini S, Cocco E, Ratner \nE, Silasi DA, Azodi M, Rutherford TJ, Schwartz PE and Santin AD. Class III β -tubulin \noverexpression within the tumor microenvironment is a prognostic biomarker for poor overall \nsurvival in ovarian cancer patients treated with neoadjuvant carboplatin/paclitaxel. Clin Exp \nMetastasis. 2014; 31(1):101-110. \n22. Ma X. The omentum, a niche for premetastatic ovari an cancer. Journal of Experimental \nMedicine. 2020; 217(4). \n23. Munder M. Arginase: an emerging key player in the mammalian immune system. Br J \nPharmacol. 2009; 158(3):638-651. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n24 \n \n24. Menjivar RE, Nwosu ZC, Du W, Donahue KL, Hong HS, Espinoza C, Brown K, Velez -\nDelgado A, Yan W, Lima F, Bischoff A, Kadiyala P, Salas -Escabillas D, et al. Arginase 1 is a \nkey driver of immune suppression in pancreatic cancer. eLife. 2023; 12:e80721. \n25. Arlauckas SP, Garren SB, Garris CS, Kohler RH, Oh J, Pittet MJ and Weissleder R. Arg1 \nexpression defines immunosuppressive subsets of tumor -associated macrophages. Theranostics. \n2018; 8(21):5842-5854. \n26. Hibbs JB, Jr., Vavrin Z and Taintor RR. L -arginine is required for expression of the \nactivated macrophage effector mechanism causing selective metabolic inhibition in target cells. J \nImmunol. 1987; 138(2):550-565. \n27. Colvin EK. Tumor -associated macrophages contribute to tumor progression in ovarian \ncancer. Front Oncol. 2014; 4:137. \n28. Carroll MJ, Kapur A, Felder M, Patankar MS and Kreeger PK. M2 macrophages induce \novarian cancer cell proliferation via a heparin binding epidermal growth factor/matrix \nmetalloproteinase 9 intercellular feedback loop. Oncotarget. 2016; 7(52):86608-86620. \n29. Recognizing the importance of ovarian aging research. Nature Aging. 2022; 2(12):1071-\n1072. \n30. Hanna A, Metge BJ, Bailey SK, Chen D, Chandrashekar DS, Varambally S, Samant RS \nand Shevde LA. Inhibition of Hedgehog signaling reprograms the dysfunctional immune \nmicroenvironment in breast cancer. Oncoimmunology. 2019; 8(3):1548241. \n31. Hinshaw DC, Hanna A, Lama -Sherpa T, Metge B, Kammerud SC, Benavides GA, \nKumar A, Alsheikh HA, Mota M, Chen D, Ballinger SW, Rathmell JC, Ponnazhagan S, et al. \nHedgehog Signaling Regulates Metabolism and Polarization of Mammary Tumor -Associated \nMacrophages. Cancer Research. 2021; 81(21):5425-5437. \n32. Hinshaw DC, Benavides GA, Metge BJ, Swain CA, Kammerud SC, Alsheikh HA, \nElhamamsy A, Chen D, Darley-Usmar V, Rathmell JC, Welner RS, Samant RS and Shevde LA. \nHedgehog Signaling Regulates Treg to Th17 Conversion Through Metabolic Rewiring in Breast \nCancer. Cancer Immunol Res. 2023; 11(5):687-702. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n25 \n \n33. Sabatier R, Calderon B, Jr., Lambaudie E, Chereau E, Provansal M, Cappiello MA, Viens \nP and Rousseau F. Prognostic factors for ovarian epithelial cancer in the elderly: a case -control \nstudy. Int J Gynecol Cancer. 2015; 25(5):815-822. \n34. Melamed A, Bercow AS, Bunnell K, Rauh- Hain JA, Wright JD, Rice LW and del \nCarmen MG. Age -Associated Risk of 90- Day Postoperative Mortality After Cytoreductive \nSurgery for Advanced Ovarian Cancer. JAMA Surgery. 2019; 154(7):669-671. \n35. Fourcadier E, Trétarre B, Gras -Aygon C, Ecarnot F, Daurès J -P and Bessaoud F. Under -\ntreatment of elderly patients with ovarian cancer: a population based study. BMC cancer. 2015; \n15(1):937. \n36. Isola JVV, Hense JD, Osório CAP, Biswas S, Alberola -Ila J, Ocañas SR, Schneider A \nand Stout MB. Reproductive Ageing: Inflammation, immune cells, and cellular senescence in the \naging ovary. Reproduction. 2024; 168(2). \n37. Cao LB, Leung CK, Law PW -N, Lv Y, Ng C -H, Liu HB, Lu G, Ma JL and Chan WY. \nSystemic changes in a mouse model of VCD -induced premature ovarian failure. Life Sciences. \n2020; 262:118543. \n38. Laviolette LA, Ethier JF, Senterman MK, Devine PJ and Vanderhyden BC. Induction of \na menopausal state alters the growth and histology of ovarian tumors in a mouse model of \novarian cancer. Menopause. 2011; 18(5):549-557. \n39. Marion SL, Watson J, Sen N, Brewer MA, Barton JK and Hoyer PB. 7,12-\ndimethylbenz[a]anthracene-induced malignancies in a mouse model of menopause. Comp Med. \n2013; 63(1):6-12. \n40. Terry S, Savagner P, Ortiz-Cuaran S, Mahjoubi L, Saintigny P, Thiery JP and Chouaib S. \nNew insights into the role of EMT in tumor immune escape. Mol Oncol. 2017; 11(7):824-846. \n41. Dongre A, Rashidian M, Reinhardt F, Bagnato A, Keckesova Z, Ploegh HL and \nWeinberg RA. Epithelial -to-Mesenchymal Transition Contributes to Immunosuppression in \nBreast Carcinomas. Cancer Res. 2017; 77(15):3982-3989. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n26 \n \n42. Dongre A and Weinberg RA. New insights into the mechanisms of epithelial -\nmesenchymal transition and implications for cancer. Nat Rev Mol Cell Biol. 2019; 20(2):69-84. \n43. Jin J, Li Y, Muluh TA, Zhi L and Zhao Q. Identification of CXCL10- Relevant Tumor \nMicroenvironment Characterization and Clinical Outcome in Ovarian Cancer. Front Genet. \n2021; 12:678747. \n44. Zhou B, Sun C, Li N, Shan W, Lu H, Guo L, Guo E, Xia M, Weng D, Meng L, Hu J, Ma \nD and Chen G. Cisplatin- induced CCL5 secretion from CAFs promotes cisplatin- resistance in \novarian cancer via regulation of the STAT3 and PI3K/Akt signaling pathways. Int J Oncol. 2016; \n48(5):2087-2097. \n45. You Y, Li Y, Li M, Lei M, Wu M, Qu Y, Yuan Y, Chen T and Jiang H. Ovarian cancer \nstem cells promote tumour immune privilege and invasion via CCL5 and regulatory T cells. Clin \nExp Immunol. 2018; 191(1):60-73. \n46. Wertel I, Surówka J, Polak G, Barczyński B, Bednarek W, Jakubowicz -Gil J, Bojarska -\nJunak A and Kotarski J. Macrophage -derived chemokine CCL22 and regulatory T cells in \novarian cancer patients. Tumour Biol. 2015; 36(6):4811-4817. \n47. Marshall LA, Marubayashi S, Jorapur A, Jacobson S, Zibinsky M, Robles O, Hu DX, \nJackson JJ, Pookot D, Sanchez J, Brovarney M, Wadsworth A, Chian D, et al. Tumors establish \nresistance to immunotherapy by regulating T(reg) recruitment via CCR4. J Immunother Cancer. \n2020; 8(2). \n48. Bellora F, Castriconi R, Dondero A, Pessino A, Nencioni A, Liggieri G, Moretta L, \nMantovani A, Moretta A and Bottino C. TLR activation of tumor -associated macrophages from \novarian cancer patients triggers cytolytic activity of NK cells. Eur J Immunol. 2014; 44(6):1814-\n1822. \n49. Dijkgraaf EM, Heusinkveld M, Tummers B, Vogelpoel LT, Goedemans R, Jha V, Nortier \nJW, Welters MJ, Kroep JR and van der Burg SH. Chemotherapy alters monocyte differentiation \nto favor generation of cancer -supporting M2 macrophages in the tumor microenvironment. \nCancer Res. 2013; 73(8):2480-2492. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n27 \n \n50. Curiel TJ, Coukos G, Zou L, Alvarez X, Cheng P, Mottram P, Evdemon- Hogan M, \nConejo-Garcia JR, Zhang L, Burow M, Zhu Y, Wei S, Kryczek I, et al. Specific recruitment of \nregulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. \nNature Medicine. 2004; 10(9):942-949. \n51. Jing J, Wu Z, Wang J, Luo G, Lin H, Fan Y and Zhou C. Hedgehog signaling in tissue \nhomeostasis, cancers, and targeted therapies. Signal Transduct Target Ther. 2023; 8(1):315. \n52. Gampala S and Yang JY. Hedgehog Pathway Inhibitors against Tumor \nMicroenvironment. Cells. 2021; 10(11). \n53. Kaye SB, Fehrenbacher L, Holloway R, Amit A, Karlan B, Slomovitz B, Sabbatini P, Fu \nL, Yauch RL, Chang I and Reddy JC. A phase II, randomized, placebo- controlled study of \nvismodegib as maintenance therapy in patients with ovarian cancer in second or third complete \nremission. Clin Cancer Res. 2012; 18(23):6509-6518. \n54. Iyer S, Zhang S, Yucel S, Horn H, Smith SG, Reinhardt F, Hoefsmit E, Assatova B, \nCasado J, Meinsohn MC, Barrasa MI, Bell GW, Pérez- Villatoro F, et al. Genetically Defined \nSyngeneic Mouse Models of Ovarian Cancer as Tools for the Discovery of Combination \nImmunotherapy. Cancer Discov. 2021; 11(2):384-407. \n55. Evans ET, Page EF, Choi AS, Shonibare Z, Kahn AG, Arend RC and Mythreye K. \nActivin levels correlate with lymphocytic infiltration in epithelial ovarian cancer. Cancer Med. \n2024; 13(17):e7368. \n56. Luo X, Fu Y, Loza AJ, Murali B, Leahy KM, Ruhland MK, Gang M, Su X, Zamani A, \nShi Y, Lavine KJ, Ornitz DM, Weilbaecher KN, et al. Stromal -Initiated Changes in the Bone \nPromote Metastatic Niche Development. Cell Rep. 2016; 14(1):82-92. \n57. Mayer LP, Devine PJ, Dyer CA and Hoyer PB. The follicle -deplete mouse ovary \nproduces androgen. Biol Reprod. 2004; 71(1):130-138. \n58. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich \nA, Pomeroy SL, Golub TR, Lander ES and Mesirov JP. Gene set enrichment analysis: A \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n28 \n \nknowledge-based approach for interpreting genome -wide expression profiles. Proceedings of the \nNational Academy of Sciences. 2005; 102(43):15545-15550. \n59. Koopmans F, van Nierop P, Andres -Alonso M, Byrnes A, Cijsouw T, Coba MP, \nCornelisse LN, Farrell RJ, Goldschmidt HL, Howrigan DP, Hussain NK, Imig C, de Jong APH, \net al. SynGO: An Evidence -Based, Expert -Curated Knowledge Base for the Synapse. Neuron. \n2019; 103(2):217-234.e214. \n60. Li T, Fu J, Zeng Z, Cohen D, Li J, Chen Q, Li B and Liu XS. TIMER2.0 for analysis of \ntumor-infiltrating immune cells. Nucleic Acids Res. 2020; 48(W1):W509-w514. \n61. Kernick K, Woudstra R, Berjanskii M, MacKay S and Wishart DS. Heatmapper2: web -\nenabled heat mapping made easy. Nucleic Acids Res. 2025; 53(W1):W316-w323. \n62. Tang D, Chen M, Huang X, Zhang G, Zeng L, Zhang G, Wu S and Wang Y. SRplot: A \nfree online platform for data visualization and graphing. PLoS One. 2023; 18(11):e0294236. \n63. Bankhead P, Loughrey MB, Fernández JA, Dombrowski Y, McArt DG, Dunne PD, \nMcQuaid S, Gray RT, Murray LJ, Coleman HG, James JA, Salto- Tellez M and Hamilton PW. \nQuPath: Open source software for digital pathology image analysis. Scientific Reports. 2017; \n7(1):16878.\n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \n \n \n \nFigure 1. ( A)(i)&(iv) Representative images of tumor burden in the ovaries and omentum of \nmice of the indicated age groups. (ii) Quantification of ovar y weights (iii) ascites fluid volume \n(v) omental weights (vi) total tumor burden, on day 55  post intrabursal implantation of \nID8Trp53−/−cells in young and aged mice ovaries (n=9). (B-C) Immunohistochemical analysis of \nproliferating nuclear antigen (PCNA) in (B)(i) ovarian and (C)(i) omental tumors  from young \nand aged mice in ID8Trp53−/−model, (ii)&(iii) Percentage of PCNA-positive cells in ovarian and \nomental tumors ( n=3). (D)(i)&(iv) Representative images of tumor burden in the ovaries and \nomentum of mice of the indicated age groups. ( ii) Quantification of ovary weights (iii) ascites \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \nfluid volume (v) omental weights (vi) tota l tumor burden on day 42 following intrabursal \nimplantation of murine fallopian tube -derived PPNM cells in young and aged mice  ovaries \n(n=8). PCNA immunohistochemistry of (E)(i) ovarian and (F)(i) omental tumors from young and \naged mice in the PPNM model, showing (ii)  & (iii) percentage and intensity of PCNA -positive \ncells in ovarian and omental tumors (n= 3). Color codes and symbols in the graphs represent the \nnumber of images analyzed per mouse. All data are presented as mean±SEM, *p<0.05,**p<0.01, \n***p<0.001, ****p<0.0001, unpaired t test.  \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \nFigure 2. (A-B) Representative multiplex immunofluorescence images of an ovarian tumor from \nyoung or aged hosts. Tumor cells are identified as CK8⁺(yellow), immune cells as CD45⁺ \n(white), endothelial cells as CD31⁺(red) , and nuclei (blue). Magnified panels highlight \nrepresentative regions of interest (ROIs) selected from CD45⁺depleted or CD45⁺enriched areas \nfor Spatial GeoMx analysis  (n=3). (C) Intra-tumoral Gene Set Enrichment Analysis (GSEA) \ncomparing CD45⁺depleted or CD45⁺-enriched regions from (i) young and (ii) aged hosts tumors \nusing normalized enrichment scores (NES) against MSigDB Hallmark gene set s (FDR<0.05). \n(D) Heatmaps depicting unique inflammatory genes contributing to the ‘Inflammatory Hallmark’ \nenrichment in ovarian tumors from  (i ) young (i i) aged hosts . ( E) CD45\n+ cells deconvolution \nusing CIBERSORTX  from young and aged host tumors . (F)(i-iv) GSEA of CD45⁺ -enriched \nregions compared between young and aged hosts using Hallmark gene sets ( FDR<0.05). (G) (i-\niii) The top three upregulated genes in aged host ovarian tumors are associated with higher B \ncells and prolonged survival of high- grade ovarian cancer patients , and (i v-vi) The top three \nupregulated genes in aged host ovarian tumors are associated with higher M2 macrophages and \nlower survival of high-grade serous ovarian cancer patients. \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \n \nFigure 3\n. (A-B) Representative i mmunofluorescence images from  ovarian tumors (A)(i)  \nID8Trp53−/− (B)(i) PPNM cells in ovarian tumors showing CD45+(green), Arg1+(red), and DAPI \n(blue). Adjacent graphs ( A)(ii) and (B)(ii) showing quantification of percentage CD45+Arg1+ \ncells in ovar ian tumors from both models . (C-D) Representative immunofluorescen ce images \nfrom (C)(i)  ID8Trp53−/− (D)(i)   PPNM   cells in ovarian tumors showing CD68+(green), \nArg1+(red), and DAPI ( blue) cells. Quantification of percentage C D68+Arg1+ cells (C)(ii) and \n(D)(ii) showing ovarian tumors from both models . (E-F) Immunohistochemistry of ovarian \ntumors showing CD206 + cells (E)(i) ID8 Trp53−/− (F)(i)  PPNM cells in ovarian tumors.  \nQuantification of percentage CD206 +cells. (E)(ii) and  (F)(ii) in both models . (G-H) \nRepresentative immunofluorescence images of Foxp3 +(red) and DAPI (bl ue) \nimmunostaining(G)(i) ID8Trp53−/− (H)(i) PPNM  in ovarian tumor. Graphs (G)(ii) and (H)(ii) \nshowing quantification of percentage Foxp3+ cells in both models. (I-J)  Representative \nimmunofluorescence images from (I) (i) ID8Trp53−/− (J)(i) PPNM  cells in omental tumors \nshowing CD45+(green), Arg1+(red), and DAPI(blue). Adjacent graphs (I)(ii) and (J)(ii) showing \nquantification of percentage CD45+Arg1+ cells in omental tumors from both models.  (K-L) \nRepresentative i mmunofluorescence images from (K) (i)  ID8Trp53−/− (L)(i)   PPNM cells in \nomental tumors showing CD68 +(green), Arg1+(red), and DAPI ( blue) cells. (K) (ii) and (L)(ii) \nQuantification of percentage CD68 +Arg1+ cells showing omental tumors from both models in \n(p=0.169 and p<0.05, respectively).  (M-N) Immunohistochemistry images of omental tumors \nshowing CD206+ cells (M)(i) ID8Trp53−/− (N)(i)  PPNM  cells in o mental tumors.  (M)(ii) and \n(N)(ii) Quantification of percentage CD206 + cells in both models. (O-P) Representative \nimmunofluorescence images of Foxp3 +(red) and DAPI ( blue) immunostaining ( O)(i) \nID8Trp53−/−(P)(i) PPNM in omental tumor. Adjacent graphs  (O)(ii)  and (P)(ii)  showing \nquantification of percentage Foxp3 + cells in both models.  Color codes and symbols indicate the \nnumber of images analyzed per  mouse in each group. All data are m eant SEM (n=3), *p<0.05, \n**p<0.01, *** p<0.001, ****p<0.0001, unpaired t test. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \n \n \nFigure 4. (A) Schematic of Hedgehog pathway  (B) Treatment regimen of Vismodegib post \nID8Trp53−/−cell implantation in the left ovaries of aged mice . (C) Probability plot based on \nabdominal girth changes showing probability of remaining below control derived threshold \n(n=9). (D) Ascites fluid collected from both groups at the endpoint. (i) vehicle (ii) vismodegib, \nshowing percentage of mice with or without ascites ( E) Representative images of diaphragmatic \nmets (i)  v ehicle (ii)  vismodegib, (iii)&(iv) showing percentage of mice with or without \ndiaphragmatic Mets in each group . (F) Representative images of  m ets on peritoneal wall (i)  \nVehicle (ii) Vismodegib, (iii)&(iv) showing percentage of mice with or without peritoneal wall \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \nmets in each group  (G) Representative images of (i) ovaries and (ii) ovary weights at end points \n(p=0.179) (iii) Representative images of PCNA Immunohistochemistry in ovarian sections (iv) & \n(v) %PCNA, positive cells (n=3). (H) Representative images of (i) omentum (ii) omental weights \nat endpoint (iii) Representative images of PCNA Immunohistochemistry in omental sections \n(iv)&(v) %PCNA, positive cells. All data are meant SEM(n=3), *p<0.05, **p<0.01, ***p<0.001, \n***p<0.0001, unpaired t test \n \n \n \n \nFigure 5. (A-B) Representative immunofluorescence images of CD45 +(green), Arg1+(red), and \nDAPI (blue) in (A)(i) ovarian  (B)(i) omental tumors from vismodegib  and vehicle treated mice. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint \n\n \n \n \n \nAdjacent graphs showing percentage of CD45\n+Arg1+cells in  (A)(ii)  ovarian tumors (p=0.062)  \nand (B)(ii) in omentum  (p=0.372). (C-D)  Representative immunofluorescence images of \nCD68+(green), Arg1+(red), and DAPI (Blue) in  (C)(i) ovarian or (D)(i) omental  tumors from  \nvismodegib  and vehicle treated mice. Adjacent graphs showing percentage of CD 68+Arg1+cells  \nin (C)(ii) ovaries (p=0.502), (D)(ii) in omentum (p=0.317). (E-F) Immunohistochemistry of \nCD206+ cells in  (E)(i) ovarian (F)(i) oment al tumors. Adjacent graph showing percentage  \nCD206+ cells (E)(ii) in ovaries  and (F)(ii) in omentum  (p=0.624). (G-H) Representative \nimmunofluorescence images of  Foxp3 + (red) and DAPI (Blue)  cells (G)(i) ovarian (H)(i) \nomental tumors. Adjacent graph showing percentage Foxp3 + cells in (G)(ii) ovarian and (H) (ii) \nomental tumors.  Color codes and symbols indicate the number of images analyzed per mouse in \neach group. All data are presented as mean±SEM  (n=3), *p<0.05, **p<0.01, ***p<0.001, \n***p<0.0001, unpaired t test.   \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted December 27, 2025. ; https://doi.org/10.64898/2025.12.23.695206doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}