{"paper_id":"3188ad93-850b-4bdf-a0b1-5da213924d74","body_text":"10\nThe effect of antiangiogenic agent aflibercept on surgically \ninduced endometriosis in a rat model\nEl efecto del agente antiangiogénico aflibercept sobre la endometriosis inducida \nquirúrgicamente en un modelo de rata\nCaglayan Ates 1*, Berna Dilbaz 2, Seval Yılmaz Ergani 1, and Fuad Atabay 3\n1Department of Obstetrics and Gynecology, University of Health Sciences, Etlik Zubeyde Hanim Training and Research Hospital; 2Department of \nReproductive Endocrinology and Infertility, University of Health Sciences, Etlik Zubeyde Hanim Training and Research Hospital; 3Department of \nPathology, University of Health Sciences, Gulhane Training and Research Hospital. Ankara, Turkey\nAbstract\nObjective: The aim of the study is to show for the first time how aflibercept affects endometriosis lesions. \nMaterial and methods:  Surgically induced endometriosis in Wistar albino female rats. Rats with endometriosis were ran -\ndomly divided into three groups: control (Co), aflibercept (Af), and leuprolide acetate (Le). Then, Af, aflibercept, and Le re -\nceived leuprolide acetate. The control group was not treated. The weights and changes in intra-abdominal adhesions of the \nrats before and after treatment were recorded according to the Blauer adhesion score. Blood extracted for sacrifice was \nanalyzed. Endometriotic lesions were evaluated for size, volume, histology, and immunohistochemistry (vascular endothelial \ngrowth factor [VEGF] and CD31). Significance level was accepted as p < 0.05. Results: Aflibercept significantly reduced \nendometrial implant volume (p = 0.002). The explant epithelial histological score showed a significant difference between \naflibercept and leuprolide acetate (p = 0.006) and between aflibercept and control groups (p = 0.002). Aflibercept decreased \nVEGF-H and CD31 expression (p = 0.001) more than leuprolide acetate. Aflibercept improved adhesions (p = 0.006). Con‑\nclusion: Aflibercept is more successful than leuprolide acetate in the treatment of endometriosis.\nKeywords: Aflibercept. Leuprolide acetate. Angiogenesis. Endometriosis. Vascular endothelial growth factor.\nResumen\nObjetivo: Mostrar por primera vez cómo afecta aflibercept a las lesiones de endometriosis. Material y métodos: Endometriosis \ninducida quirúrgicamente en ratas hembras albinas Wistar. Las ratas con endometriosis se dividieron aleatoriamente en tres gru-\npos: control (Co), aflibercept (Af) y acetato de leuprolida (Le). Luego, Af, aflibercept y Le recibieron acetato de leuprolida. El grupo \nde control no fue tratado. Los pesos y cambios en las adherencias intraabdominales de las ratas antes y después del tratamien-\nto se registraron de acuerdo con la puntuación de adherencia de Blauer. La sangre extraída para el sacrificio fue analizada. Las \nlesiones endometriósicas se evaluaron en tamaño, volumen, histología e inmunohistoquímica (factor de crecimiento endotelial \nvascular [VEGF] y CD31). El nivel de significación se aceptó como p < 0.05. Resultados: Aflibercept redujo significativamente el \nvolumen del implante endometrial (p = 0.002). La puntuación histológica epitelial (EHS) del explante mostró una diferencia signi-\nficativa entre aflibercept y acetato de leuprolida (p = 0.006) y entre los grupos de aflibercept y control (p = 0.002). Aflibercept \ndisminuyó la expresión de VEGF-H y CD31 (p = 0.001) más que el acetato de leuprolida. Aflibercept mejoró las adherencias \n(p = 0.006). Conclusión: Aflibercept tiene más éxito que el acetato de leuprolide en el tratamiento de la endometriosis.\nPalabras clave: Aflibercept. Acetato de leuprolida. Angiogénesis. Endometriosis. Factor de crecimiento del endotelio vascular.\nORIGINAL ARTICLE\nCir Cir. 2024;92(1):10-19  \nContents available at PubMed  \nwww.cirugiaycirujanos.com\n*Correspondence \nCaglayan Ates \nE-mail: caglayanctf@hotmail.com\n0009-7411/© 2023 Academia Mexicana de Cirugía. Published by Permanyer. This is an open access article under the terms of the CC BY-NC-ND license  \n(http://creativecommons.org/licenses/by-nc-nd/4.0/ ).\nCIRUGIA Y CIRUJANOS\nDate of reception: 15-02-2023\nDate of acceptance: 13-05-2023\nDOI: 10.24875/CIRU.23000072\n\nC. Ates et al. Endometriosis and aflibercept\n11\nIntroduction\nEndometriosis is defined as the presence of a tissue \nsimilar to the endometrium at sites outside the uterine \ncavity1. It is difficult to determine the actual prevalence \nof endometriosis as it might be aymptomatic in some \nof the affected women and a biopsy is required for a \ndefinitive diagnosis. However, the prevalence among \nwomen of reproductive age is estimated to be 10% 2. \nEndometriosis frequently presents itself with symp -\ntoms such as chronic pelvic pain, infertility, dysmenor -\nrhea, dyspareunia, dysuria, dysgeusia, and fatigue \nand therefore might have a negative effect on physi -\ncal, mental, sexual, or social life and productivity. \nAlthough the pathophysiology of endometriosis has \nnot yet been fully understood, it is known to be closely \nrelated to inflammation and angiogenesis 3,4.\nAngiogenesis is necessary for supplying nutrients \nrequired for implantation and invasion of endometriotic \nimplants. Vascularization of lesions is mainly regulated \nby vascular endothelial growth factor (VEGF) 5. Afliber-\ncept is a recombinant fusion protein containing the \nsecond Ig domain and the third Ig domain of the VEGF \nreceptors (VEGFR1 and VEGFR2, respectively), fused \nto the Fc portion of human immunoglobulin G1 (IgG1)6-8. \nIt also binds to VEGF-A, VEGF-B, placental growth \nfactor (PlGF)-1, and PlGF-2. Consequently, aflibercept \nacts as a trap for VEGFR by preventing the ligands \nfrom binding to their respective receptors, and it binds \nto both ends of VEGF very tightly 6,9. This binding in -\nhibits the biological action of VEGF and prevents the \nabnormal development of blood vessels. Aflibercept is \nalso referred to in the literature as “VEGF trap” 7.\nAflibercept monotherapy is used in conjunction with \nchemotherapy as it significantly inhibits tumor growth \nand improves survival in various orthotropic animal \nmodels. It has been experimentally shown that afliber -\ncept prevents and slows down the formation of cho -\nroidal neovascularization 10.\nIn this pioneering study, we aimed to evaluate the ef -\nfect of aflibercept, a recombinant fusion protein that has \nan antiangiogenic effect by inhibiting VEGF on endo -\nmetriotic foci, and compared the effect of aflibercept \nwith leuprolide acetate, a GnRH agonist currently used \nin the routine treatment and the no-treatment group.\nMaterials and methods\nThe effects of the antiangiogenic agent aflibercept \nand a gonadotropin analog, leuprolide acetate, on \nectopic endometrial lesions were compared to the \ncontrol group in a rat model.\nEthical approval\nThis experiment was conducted in accordance with \nthe standards of the Local Ethics Committee Directory \nof Turkish Ministry of Health, Health Sciences Univer -\nsity Gülhane animal experiments. The experimental \nanimals were obtained from Gülhane Experimental \nAnimals Production and Research Unit. As recom -\nmended by the Gülhane Animal Studies Ethics Com -\nmittee (03.03.21/ETİK-2021/07-21/09), a preliminary \nstudy was first conducted on two female rats for test -\ning the experimental endometriosis model and the \nexperiment was started after the success of the pro -\nposed model was confirmed and an approval from the \nGülhane Animal Studies Ethics Committee  was ob -\ntained (25.03.21/ETİK-2021/08-21/10).\nAnimals\nA total of 30 female Wistar albino rats were included \nin the study. The 8-week-old rats, weighing 250-\n300 g, were kept in temperature-controlled cages \nthroughout the study with standard rat chow and ad -\nequate water. During the day, each rat was kept in \nspecial standard cages at 21-24°C and 50% humidity. \nThe automatic 12-h light-dark cycle was maintained. \nRats were kept in the same cage for 20 days to ensure \nestrus.\nSurgical procedures\nAnesthesia was administered by intraperitoneal ad -\nministration of 90 mg/kg ketamine hydrochloride (Ke -\ntalar; Eczacıbaşı Warner-Lambert pharmaceutical \nindustry, Levent/İstanbul) and 10 mg/kg xylazine hy -\ndrochloride (Rompun-Bayer, Şişli/İstanbul) for the op -\nerations. In immobilized rats, the surgical area was \nshaved in the dorsal position and cleaned with povi -\ndone-iodine solution for antisepsis.\n1\nst\n operation\nA rat endometriosis model was developed by the \nsurgical endometriosis induction method defined by \nVernon and Wilson11. A 3-cm median skin incision was \nmade while protecting the integrity of the intra-abdom -\ninal organs, and then, the uterine horns were exposed. \n\nCirugía y Cirujanos. 2024;92(1)\n12\nThe right uterine horn was excised after ligating both \nthe uterotubal junction and the cervix. The endometrial \ntissue inside the excised horn section was excised to a \n5 × 5 mm piece and then implanted with 4-0 Vicryl su -\ntures into the relatively vascular area on the ipsilateral \ninner lateral wall of the abdomen, with the endometrium \nfacing the peritoneal surface. After the bleeding was \ncontrolled, 1 ml of saline was applied to the abdominal \ncavity, and the median incision was closed by continu -\nous suturing using 3-0 Vicryl and prolene sutures ac -\ncording to the anatomic plan. Subsequently, all rats \nwere transferred to the post-operative unit in separate \ncages. Paracetamol (Parol oral suspension; Atabay İlaç, \nKadıköy/Istanbul) at a dosage of 100 mg/kg was given \nto the rats in 500 ml of water for pain control during the \npost-operative period. Routine daily feeding of the ani -\nmals was continued during the following 3 weeks.\n2\nnd\n operation\nTwenty-one days after the initial surgery, rats under -\nwent exploratory laparotomy to evaluate endometriotic \nlesions. One rat died during this operation during an -\nesthesia was given and was excluded from the study. \nIn the remaining 29 rats, the development of endome -\ntriotic implants at the transplanted areas was con -\nfirmed. Then, the rats were randomly divided into three \ngroups: aflibercept group (Af) n = 12, leuprolide acetate \ngroup (Le) n = 12, and the control group (Co) n = 5. \nThe local ethics committee (25.03.21/ETIK-2021/08-\n21/10) recommended recruitment of a reduced number \nof rats to the control group. The rats were numbered \naccording to the intra-group tail staining method. The \nvolumes (0.52 × width [mm] × length [mm] × height \n[mm]) of all tissues transplanted in the abdominal wall \n3 weeks before and changed into endometriotic struc -\nture were then measured (Fig. 1A). During the examina-\ntion, intra-abdominal adhesions were scored according \nto the Blauer scoring system (0-4): 0 = no adhe -\nsion, 1 = weak adhesion, 2 = dense adhesion confined \nto a single area, 3 = dense adhesion over a large area, \nand 4 = strong adhesions including internal organs 12 \n(Fig. 1B and C). Surgery was completed with closure \nof the abdominal cavity with 3-0 Vicryl and prolene. \nBody weights of all rats were measured and 25 mg/kg \naflibercept (Eylea ®; Regeneron, NY, USA) was admin -\nistered intraperitoneally (i.p.) in the Af group, and 1 mg/\nkg leuprolide acetate (Lucrin Depot; Abbott, Cedex, \nIstanbul) was administered subcutaneously (s.c.), Tur -\nkey, in the Le group. The doses were based on the \nstudies from the literature 13,14. The co-group received \nno treatment. The rats received the same post-opera -\ntive pain control treatment and they were routinely fol -\nlowed up for 21 days until the third operation.\n3\nrd\n operation\nA third laparotomy was performed 21 days after the \nexploratory laparotomy. During this operation, the diam-\neters of endometriotic lesions were measured as done \nduring the previous operation. The adhesions were \nscored again according to the “Blauer scoring system” \nfor comparison with the pre-treatment scores 11. The \nbody weights of all rats were measured again. Subse -\nquently, all rats were sacrificed by exsanguination. \nBlood samples collected were sent to the laboratory in \ntubes containing ethylenediaminetetraacetic acid \n(EDTA) for analysis. Finally, all endometriotic lesions \nwere excised and sent to the pathology laboratory in \ncontainers containing 10% formaldehyde solution.\nHistopathological evaluation\nThe excised endometriotic tissues were stored in \ncontainers containing 10% formaldehyde, numbered, \nand sent to the pathology laboratory of Gülhane Train -\ning and Research Hospital, where they were examined \nby the same pathologist (F.A.) who was blinded to the \nstudy groups. Sections of 5 µm (microns) were taken \nby the microtome (Leica-M225-Thermo HM3555-Ther -\nmo scientific). Sections stained with hematoxylin and \neosin (H&E) were examined under a microscope \n(Nikon® ECLIPSE 80i, Japan) at ×100, ×200 and ×400 \nmagnifications (Supplementary Figure  1). The persis -\ntence of endometrial cells within the endometrial im -\nplants was assessed by semiquantitative explant \nepithelial histological scoring (EHS) (score 0-3): 3 = a \nwell-preserved epithelial layer, 2 = a moderately pre -\nserved epithelium with leukocyte infiltration, 1 = a \npoorly preserved epithelium (containing only occasion -\nal epithelial cells), and 0 = no epithelium 15. The soft -\nware “NIS-Elements D Ver 5.02.03 for 64-bit edition” \nwas used for photographing from the microscope.\nImmunohistochemical evaluation\nImmunohistochemical staining was performed automati-\ncally using the Ventana BenchMark XT System (Ventana \nMedical Systems, Roche, Basel, Switzerland). Ultraview uni-\nversal 3,3′-diaminobenzidine (DAB) detection kit (Ventana®) \n\nC. Ates et al. Endometriosis and aflibercept\n13\nwas used for automatic immunohistochemistry device. The \nprimary antibodies against VEGF (Flt-1/VEGFR1, 0.1 ml \nconcentrate 1:501:200 antibody, GenomeME, Richmond \nBC, Canada) and CD31 (JC70, 0.1 ml concentrate 1:251:100 \nantibody, Santa Kruz) were used. The slides were evalu-\nated under the microscope, with histological scoring (H-\nscore) for VEGF as previously described in the literature16-18: \nH-score = ∑Pi i = 0 (negatively stained cells) to i = 3 (highly \nstained cells), P = 1, 2, 3, 4, 5 values < 15%, 15-50%, 50-\n85%, > 85%, and 100% positively stained cells, respec-\ntively. For the CD31 antibody, the number of CD31-positive \nstained microvessels (endothelial cells/endothelial cell clus-\nters) was calculated per 1 mm2 area5.\nBlood parameters\nIntracardiac blood samples collected during the sac -\nrifice phase by exsanguination were transferred to \nEDTA tubes. Unfortunately, clotting abnormalities oc -\ncurred in 4 tubes (1 tube from the Af group, 2 tubes \nfrom the Le group, and 1 tube from the Co group) and \nthese tubes were excluded from the study. The remain -\ning 25 blood samples were sent to laboratory for analy -\nsis. An automatic analyzer was used to determine the \nhemoglobin level (HGB) (gr/dL), total white blood cell \ncount (WBC) (× 10^3 µL), and platelet count (PLT) \n(× 10^\\mm^3), which are among the laboratory com -\nplete blood count parameters (Mindray BC-6000).\nStatistical analysis\nThe sample size of the study was calculated by \nG-power analysis, and the number of rats required for \nthe study was set at 30. Normality assumptions of the \ncontinuous variables were tested using the Shapiro–\nWilk test. The mean ± standard deviation of the nor -\nmally distributed variables and the median values \n(25th-75th percentiles) of the non-normally distributed \nvariables are indicated. The Wilcoxon signed-rank test \nwas used to compare the parameters of the rats be -\nfore and after treatment. One-way analysis of variance \n(one-way ANOVA) was used to compare the normally \ndistributed parameters between groups. In case of a \nsignificant difference, the post hoc  Tukey or Games-\nHowell test was used according to the result of Lev -\nene’s test for homogeneity of variance. The \nKruskal–Wallis test was used to compare the param -\neters that were not normally distributed between \ngroups. In the case of a significant difference, the \nMann–Whitney test with Bonferroni correction was \nused to determine from which groups the difference \noriginated. IBM SPSS 25 program was used in all \nanalyses, and p < 0.05 was accepted as the signifi -\ncance level.\nResults\nWeight of the rats and volume of the endometriotic \ntissue measured at the time of the 2 nd and 3 rd opera -\ntions (pre and post-treatment) were compared in the \nAf, Le, and the control groups. The post-treatment \nweight of both the Co group and the Af group was \nsignificantly lower than that of the Le group (p < 0.001) \n(Table 1). The weight of the rats increased significantly \nin all groups; 14.89% in the Co group, 8.3% in the Af \ngroup, and 27.7% in the Le group (Table 1). However, \nFigure 1. Several images captured throughout the experiment. A: appearance compatible with endometriosis on exploration. B: example of \nBlauer adhesion score 1. C: example of Blauer adhesion score 4.\nA B C\n\nCirugía y Cirujanos. 2024;92(1)\n14\nthe increase in weight was statistically significant in \nthe Af and Le groups. The endometriotic foci volume \nof the Co group was found to be increased by 333.3% \nduring the 3 rd operation. However, the volume of the \nendometriotic foci was found to be decreased by \n40.4% in the Le group and 79.5% in the Af group. Al -\nthough endometriotic foci volume regressed in both Af \nand Le groups, the reduction in the Af group was sta -\ntistically significantly higher (p =  0.002) (Table 1).\nIntraabdominal Blauer adhesion scores of the three \ngroups recorded during the 2 nd and 3 rd operations \nwere compared. No change in the Blauer adhesion \nscore was observed in the Le and Co groups, while \nthere was a statistically significant decrease in Blauer \nscore after treatment with Af (p = 0.006) (Table 1).\nHistology of the excised endometriotic tissue obtained \nduring the 3 rd operation and the immunohistochemical \nassessment of VEGF and CD31 expression in excised \ntissues were compared in the control group with the two \nintervention groups ( Table 2 ). Explant epithelial histo -\nlogical scoring (EHS) difference among groups was sig-\nnificant (p = 0.002) (Table 2). When the persistence of \nendometrial cells within the endometrial implants was \nassessed by semiquantitative explant EHS, none of the \nspecimens in the intervention and control groups had a \nscore of 0. In the control group, 40% had a score of 2 \nand 60% had a score of 3. In the Le group, the distribu -\ntion of the scores 1, 2, and 3 was 25.0%, 16.7%, and \n58.3%, respectively. In the Af group, none of the speci -\nmens had a score of 3, 75% received a score of 1 while \nthe remaining 25% had a score of 2 (Table 3). According \nto Chi-square analysis, when three groups were com -\npared, the p value was 0.07 (Table 3). When the median \nEHS scores were compared, there was a significant \nTable 1. Comparison of groups in terms of weight, endometriotic lesion width, length, height, volume, and Blauer adhesion score\nVariables Group Co\n(n = 5)\nGroup Af\n(n = 12)\nGroup Le\n(n = 12)\np value Difference \namong groups\nWeight (gr)\nBefore treatment\nPost‑treatment\n235.00 (215.00‑242.50)\n270.00 (245.00‑277.50) c\n242.50 (235.00‑253.75)\n262.50 (255.00‑273.75) a\n235.00 (235.00‑247.50)\n300.00 (290.00‑305.00) b\n0.282†\n < 0.001†\n‑\na < b; c < b\np value 0.042* 0.002* 0.002*\nEndometriotic lesion width (mm)\nBefore treatment\nPost‑treatment\n6.00 (4.00‑8.00)\n8.60 (6.15‑10.05)\n9.00 (5.40‑9.10)\n5.60 (2.95‑7.15)\n5.55 (3.80‑6.68)\n5.00 (3.58‑5.52)\n0.153†\n0.067†\n‑\n‑\np value 0.043* 0.002* 0.036*\nEndometriotic lesion length (mm)\nBefore treatment\nPost‑treatment\n4.00 (2.15‑5.05) c\n4.50 (3.05‑6.55)\n7.40 (5.55‑8.00) a\n4.65 (4.00‑6.30)\n4.45 (2.82‑5.23) b\n3.45 (2.58‑4.40)\n0.004†\n0.175†\nb < a; c < a\n‑\np value 0.042* 0.012* 0.006*\nEndometriotic lesion height (mm)\nBefore treatment\nPost‑treatment\n1.50 (1.25‑2.45) c\n2.50 (1.65‑5.10)\n3.65 (2.90‑4.57) a\n1.85 (1.55‑2.08)\n2.05 (1.45‑2.88) b\n1.50 (1.00‑2.50)\n0.002†\n0.230†\nb < a; c < a\n‑\np value 0.043* 0.002* 0.016*\nEndometriotic lesion volume (mm3)\nBefore treatment\nPost‑treatment\n18.00 (10.00‑35.50) c\n78.00 (24.50‑107.50)\n134.50 (59.75‑170.25) a\n27.50 (10.75‑43.00)\n21.00 (15.25‑46.75) b\n12.50 (7.00‑27.25)\n0.013†\n0.106†\nb < a\n‑\np value 0.043* 0.002* 0.084*\nBlauer adhesion score\nBefore treatment\nPost‑treatment\n2.00 (1.00‑2.00)\n2.00 (1.50‑3.50)\n2.00 (2.00‑3.00)\n1.00 (1.00‑2.00)\n2.00 (2.00‑3.00)\n2.00 (1.00‑4.00)\n0.145†\n0.091†\n‑\n‑\np value 0.102* 0.006* 1.000*\n*Wilcoxon signed‑rank test.\n†Kruskal–Wallis test.\ngr: grams; mm: millimeters; mm3: cubic millimeters.  \nBlauer adhesion score; (0‑4): 0 = no adhesion; 1 = weak adhesion; 2 = dense adhesion limited to a single area; 3 = dense adhesion over a large area; 4 = dense adhesion involving internal organs.\nParameters are expressed as mean±standard deviation or median (25th‑75th percentile) taking into account normality assumptions. Values in bold are statistically significant.\n\nC. Ates et al. Endometriosis and aflibercept\n15\ndifference between the groups, and Af group had the \nlowest median score when compared to Le and Co \ngroups. (p = 0.002). The Bonferroni-corrected Mann–\nWhitney test showed a significant difference between \nthe Af group and the Le group (p = 0.006) and between \nthe Af group and the Co group (p = 0.002) (Table 2).\nWhen glandular, stromal, and epithelial cells were \nstained for calculation of microvessel density for the \nVEGF-H immunohistochemical score and CD31 anti -\nbody evaluation, Co group was found to have a stron -\nger staining in comparison to the Af and Le groups. In \nthe Af group, staining for both VEGF and CD31 anti -\nbodies was much weaker than that observed in the Co \nand Le groups ( Fig. 2). Counting CD31 and VEGF-H \nscores revealed a significant difference between the \ngroups (p < 0.001 and p = 0.002, respectively) (Table 2 \nand Fig. 3). In the Bonferroni-corrected Mann–Whitney \ntest, both CD31 and VEGF H scores were significantly \nlower in the Af group than the Co group (p = 0.001) \nand the Le group (p < 0.001) (Table 2).\nOut of the 29 blood samples obtained during the 3rd op-\neration, four samples could not be processed due to the \nhemolysis. Analysis from the remaining samples (Co \ngroup: [n = 4], Af group: n = 11, Le group: n = 10) dem -\nonstrated no significant difference between the three \ngroups in terms of WBC counts and hemoglobin levels. \nThe platelet counts in Af, Le, and Co groups (median \n[25th-75th percentile]) were 999.50 (R: 886.75-1002.25) × \n10^3 µL, 978.00 (R: 936.00-993.75) × 10^3 µL, and \n870.00 (R: 805.00-890.50) × 10^3 µL, respectively. The \nKruskal–Wallis test revealed a significant difference in the \nplatelet counts among the groups (p = 0.032) (Table 4).\nDiscussion\nAlthough more than a decade passed since the \nuniversally accepted definition of endometriosis, \nthere is not a consensus about the pathogenesis and \na set protocol for the diagnosis and treatment of this \ndisease 19. Endometriosis has a negative impact on \neducation, employment, and social relations of the \nwomen of reproductive age due to its effect on physi -\ncal, sexual, and reproductive health and thus is \ncalled as a “social disease” 20,21 . The researchers \nworking in this field have proposed different theories \nfor the explanation of various forms of endometriosis. \nTable 3. Distribution of explant epithelial histological scores in Study Groups\nGroups Score 0 Score 1 Score 2 Score 3 p value\nNumber of EHS\nGroup Af\nGroup Le\nGroup Co\n‑\n‑\n‑\n9 (75.0%)\n3 (25.0%)\n‑\n3 (25.0%)\n2 (16.7%)\n2 (40.0%)\n‑\n7 (58.3%)\n3 (60.0%)\n0.007*\n*Chi‑square analysis.\nGroup Le < group Af at score 1 (p < 0.001); group Le < group Co at score 2 (p < 0.001).\nEHS: explant epithelial histological scoring “semi‑quantitative” (0‑3); 0: no epithelium, 1: poorly preserved (only occasionally) epithelium, 2: moderately preserved epithelium with \nleukocyte infiltrates, 3: well‑preserved epithelial layer.\nNumber of EHS: it refers to the number of rats in the score groups according to the EHS. Values in bold are statistically significant.\nTable 2. Comparison of groups in terms of explant epithelial histological score, VEGF H score, and CD31 expression\nVariables Group Co\n(n = 5)\nGroup Af\n(n = 12)\nGroup Le\n(n = 12)\np value Difference among \ngroups\nEHS 3.00 (2.00‑3.00) c 1.00 (1.00‑1.75) a 3.00 (1.25‑3.00) b 0.002* a < b (p = 0.006)†\na < c (p = 0.002)†\nCD31 154.00 (105.50‑286.50) c 75.50 (62.25‑85.00) a 144.00 (112.75‑198.25) b  < 0.001* a < b (p < 0.001)†\na < c (p = 0.001)†\nVEGF 6.00 (6.00‑6.00) c 2.00 (2.00‑3.00) a 4.00 (2.00‑6.00) b 0.002* a < c (p = 0.001)†\n*Kruskal–Wallis test.\n†Mann–Whitney test with Bonferroni correction.\nEHS: Explant epithelial histological scoring “semi‑quantitative” (0‑3); 0: no epithelium, 1: poorly preserved (only occasionally) epithelium, 2: moderately preserved epithelium with \nleukocyte infiltrates, 3: well‑preserved epithelial layer; CD31: microvessel density; The number of CD31 positive‑stained microvessels (endothelial cell/endothelial cell clump) per 1 mm2 \narea; VEGF: H score  =  ∑Pi; i = 0 (stained negatively) to 3 (stained heavily), P = 1, 2, 3, 4, 5 values < 15%, 15‑50%, 50‑85%, > 85%, and 100%, respectively, positively stained cells.\nParameters are expressed as mean±standard deviation or median (25th‑75th percentile) taking into account the assumptions of normality.\na: median value for the Af group (25th‑75th percentile), b: median value for the Le group (25th‑75th percentile), c: median value for the control group (25th‑75th percentile).\n\nCirugía y Cirujanos. 2024;92(1)\n16\nBesides lymphangiogenesis and neurogenesis, an -\ngiogenesis plays an important role in the pathophysi -\nology of endometriosis 22. The nutrients and oxygen \nrequired for the development of endometriotic le -\nsions are provided by angiogenesis and thus neo -\nvascularization 23. The presented study is a pioneering \nstudy that aims to investigate aflibercept, an antian -\ngiogenic agent in the treatment of endometriosis in \na surgically induced rat model. In the present study, \naflibercept was more efficient in the regression of \nendometriotic lesions and treating adhesions than \nthe control group and the leuprolide acetate group.\nChanges in implant volume have been reported most-\nly as a marker for treatment efficacy of induced endo -\nmetriotic foci in animal studies 24,25. Bakacak et al. \nreported a significant reduction in the endometrial im -\nplant volume after treatment with the antiangiogenic \nagent thalidomide (p = 0.001) 26. Aflibercept group had \nFigure 2. Immunohistochemical examination of endometriotic lesions. Image for Group Co CD31 is at ×100 magnification. All other images are \nat ×200 magnification. VEGF: vascular endotelial growth factor.\n\nC. Ates et al. Endometriosis and aflibercept\n17\na significant reduction in the volume of the endometriotic \nimplants (p = 0.002) (Table 1). The significant increase \nin the total body weight of the rats in all groups after \ntreatment could be speculated as the rats being in their \ngrowth period during the experimental study. However, \nthe weight gain in the group receiving leuprolide acetate \nwas more significant than the other groups (Table 1).\nOzer et al. analyzed and compared the effect of two \nantiangiogenic agents – bevacizumab and sorafenib \non the volume of the endometriotic foci 5 and the \nchanges in VEGF and CD31.The reported results of \nthis study were similar to our results. In contrast, Ozer \net al. observed that sorafenib cleared some endome -\ntriotic lesions entirely. For aflibercept, we did not meet \nsuch a circumstance in our research. They compared \ntwo anti-angiogenic drugs while in the present study, \nthe efficacy of aflibercept; an antiangiogenic drug was \ncompared with a GnRH agonist leuprolide acetate in \nthe treatment of endometriosis. Because, in addition \nto its hypoestrogenic impact, leuprolide acetate has a \ndemonstrated anti-VEGF activity 27. In our study, with \nleuprolide acetate, VEGF expression decreased when \ncompared with the control group but this decrease \nwas not significant. In our study, aflibercept statisti -\ncally significantly reduced VEGF expression. This in -\ndicates that aflibercept has a stronger antiangiogenic \neffect than leuprolide acetate.\nAdhesion formation was decreased with two anti-\nVEGF agents bevacizumab28 and sunitinib29 in two ani -\nmal studies presented by Moraloglu and Pala. According \nto the Blauer scoring system, we also found that afliber-\ncept reduced adhesion after treatment. Consistent with \nthe literature our study demonstrated that anti-VEGF \nagents have a reducing effect on adhesions.\nSiracusa et al observed to have a significant reduc -\ntion in both markers of angiogenesis – VEGF and \nCD34 expression on the endometriotic surfaces with \nanother anti-VEGF agent, rapamycin 30. In the present \nstudy, besides VEGF expression levels, another en -\ndothelial marker, CD31 was also evaluated in the en -\ndometriosis rat model and the findings with aflibercept \nwere similar to those obtained with rapamycin.\nFigure 3. Results of CD31 and vascular endothelial growth factor H score by groups. CD31: microvessel density: number of CD31-positive \nstained microvessels (endothelial cell/endothelial cell clump) per 1 mm2; VEGF: H score  = ∑Pi; i =  0 (stained negatively) to 3 (stained heavily), \np = 1, 2, 3, 4, 5 values < 15%, 15-50%, 50-85%, > 85%, and 100, respectively, % positively stained cells.   \nTable 4. Comparison of intra‑group and inter‑group blood parameters\nParameters Group Co (n = 4) Group Af (n = 11) Group Le (n = 10) p value Difference among groups\nWBC (× 10^3 µL) 5196.00±930.61 6870.00±1552.26 7870.00±2531.40 0.061* ‑\nHGB (gr/dL) 13.38±0.36 13.94±0.27 14.06±0.79 0.097* ‑\nPLT (× 10^\\mm^3) 870.00 (805.00‑890.50) c 999.50 (886.75‑1002.25) a 978.00 (936.00‑993.75) b 0.032† c < b\nc < a\n*One‑way ANOVA analysis. \n†Kruskal–Wallis test. \nWBC: white blood cell count; HGB: amount of hemoglobin; PLT: platelet count.\nParameters are expressed as mean±standard deviation or median (25th‑75th percentile) taking into account normality assumptions. Values in bold are statistically significant.\n\nCirugía y Cirujanos. 2024;92(1)\n18\nZhang et al studied rosiglitazone, an antiangiogenic \nPPARγ (peroxisome proliferator-activated receptor γ) \nagonist that acts by inhibiting macrophage activation in \nendometriotic lesions in a surgically induced rat model4. \nThe expression of VEGF and caspase-3 immunohisto -\nchemically in endometriotic tissue was evaluated, and \nexpression of VEGF and caspase-3 was significantly \nreduced by rosiglitazone (p < 0.05, p < 0.05) 4. Zhang \net al. compared rosiglitazone with the control (no treat -\nment) and saline groups without comparing it with any \nother agent currently being used in the treatment of \nendometriosis. In the present study, the efficacy of \naflibercept was compared with leuprolide acetate in a \nrat model, to demonstrate its potential in comparison to \nan agent that has been currently being used.\nYıldız et al. studied imatinib, a tyrosine kinase recep -\ntor inhibitor, and obtained a significant improvement (p \n< 0.05) compared with the control group using VEGF-H \nscore system 31. However, in our study, aflibercept re -\nduced VEGF expression more prominently, therefore \nit can be speculated that the antiangiogenic effect of \naflibercept might be stronger that of imatinib (p = 0.01).\nIn the study performed by Ozdemir et al on ranibi -\nzumab which has an antiangiogenic effect, the percent -\nage of specimens with an epithelial histological score of \n0 was 33.3%. In our study, there were no specimens in \nany of the groups with a score of 0; however, the rates \nof the other scores were similar. Ozdemir et al reported \nscore 1 in 66.7% of the ranibizumab-treated group 32. In \nour study, 75% of the aflibercept group received a score \nof 1. This suggests that aflibercept does not completely \neradicate endometriotic lesions but can destroy the pro-\ntective epithelium of the lesion similar to ranibizumab.\nIt is well known that antiangiogenic agents may cause \nthromboembolic complications30. In the present study, \nthe blood samples taken at the final phase of the study \nwere analyzed and complete blood count results showed \nthat the platelet count was higher in the aflibercept group \nthan in the other groups. This result was significant when \ncompared with the control group but not significant com-\npared with the leuprolide acetate group. These findings \nsuggest that aflibercept may cause thrombocytosis.\nStudies with experimental animals cannot be directly \napplied to humans. Therefore, acceptance of aflibercept \nas a treatment option for endometriosis depends on \nmore comprehensive animal studies and clinical stud -\nies. Any results we obtained should be carefully evalu -\nated because the number of rats was restricted due to \nthe regulations of the local ethics committee for animal \nstudies and the parameters studied to determine the \nside effect profile were limited.\nConclusion\nIt is necessary to suppress the process of angio -\ngenesis to achieve effective results in the treatment \nof endometriosis. The anti-angiogenic agent afliber -\ncept was used for the first time in the present study \nfor the treatment of endometriosis. Aflibercept proved \nmore successful than the control group and leuprolide \nacetate in both regression of endometriotic lesions \nand treatment of adhesions.\nFunding\nThe University of Health Sciences Scientific Research \nProjects Unit funded this study as part of a medical \nspecialty thesis (project number: 2021/122). Author B.D. \nhas received research support from The University of \nHealth Sciences Scientific Research Projects Unit.\nConflicts of interest\nThe authors declare that they have no conflicts of \ninterest.\nEthical disclosures\nProtection of human and animal subjects. The \nauthors declare that the procedures followed were in \naccordance with the regulations of the relevant clinical \nresearch ethics committee and with those of the Code \nof Ethics of the World Medical Association (Declaration \nof Helsinki).\nConfidentiality of data. The authors declare that \nno patient data appear in this article.\nRight to privacy and informed consent. The au -\nthors declare that no patient data appear in this article.\nSupplementary data\nSupplementary data are available at DOI: 10.24875/\nCIRU.23000072. These data are provided by the cor -\nresponding author and published online for the benefit \nof the reader. The contents of supplementary data are \nthe sole responsibility of the authors.\nReferences\n 1. Samartzis EP, Labidi-Galy SI, Moschetta M, Uccello M, Kalaitzopoulos \nDR, Perez-Fidalgo JA, et al. Endometriosis-associated ovarian carcino -\nmas: insights into pathogenesis, diagnostics, and therapeutic targets-a \nnarrative review. Ann Transl Med. 2020;8:1712.\n 2. Zondervan KT, Becker CM, Missmer SA. Endometriosis. N Engl J Med. \n2020;382:1244-56.\n\nC. Ates et al. Endometriosis and aflibercept\n19\n 3. Alborzi S, Azimirad A, Azimirad M. Experimental endometriosis: review \nof the literature through a century and the Iranian experience. Arch Iran \nMed. 2018;21:536-43.\n 4. Zhang S, Zhuang L, Liu Q, Yu X, Min Q, Chen M, et al. Rosiglitazone \naffects the progression of surgicallyinduced endometriosis in a rat model. \nMol Med Rep.  2021;23:35.\n 5. Ozer H, Boztosun A, Açmaz G, Atilgan R, Akkar OB, Kosar MI. The \nefficacy of bevacizumab, sorafenib, and retinoic acid on rat endometrio -\nsis model. Reprod Sci. 2013;20:26-32.\n 6. Ciombor KK, Berlin J, Chan E. Aflibercept. Clin Cancer Res. \n2013;19:1920-5.\n 7. Sarwar S, Clearfield E, Soliman MK, Sadiq MA, Baldwin AJ, Hanout M, \net al. Aflibercept for neovascular age-related macular degeneration. Co -\nchrane Database Syst Rev. 2016;2:CD011346.\n 8. Avery RL, Castellarin AA, Steinle NC, Dhoot DS, Pieramici DJ, See R, \net al. Systemic pharmacokinetics and pharmacodynamics of intravi -\ntreal aflibercept, bevacızumab, and ranibizumab. Retina. 2017;37:  \n1847-58.\n 9. Azanza Perea JR, Sádaba B. Aflibercept. An approach to pharmacology. \nArch Soc Esp Oftalmol. 2015;90 Suppl 1:6-10.\n 10. Stewart W. Aflibercept (VEGF-TRAP): the next anti-VEGF drug. Inflamm \nAllergy Drug Targets. 2011;10:497-508.\n 11. Vernon MW, Wilson EA. Studies on the surgical induction of endome -\ntriosis in the rat. Fertil Steril. 1985;44:684-94.\n 12. De Iaco PA, Stefanetti M, Pressato D, Piana S, Doná M, Pavesio A, et  al. \nA novel hyaluronan-based gel in laparoscopic adhesion prevention: pre -\nclinical evaluation in an animal model. Fertil Steril. 1998;69:318-23.\n 13. Kiykac Altinbas S, Tapisiz OL, Cavkaytar S, Simsek G, Oguztuzun S, \nGoktolga U. Is montelukast effective in regression of endometrial im -\nplants in an experimentally induced endometriosis model in rats? Eur J \nObstet Gynecol Reprod Biol. 2015;184:7-12.\n 14. Forsgård RA, Marrachelli VG, Lindén J, Frias R, Collado MC, Korpela R, \net al. Two-week aflibercept or erlotinib administration does not ınduce \nchanges in intestinal morphology in male sprague-dawley rats but afli -\nbercept affects serum and urine metabolic profiles. Transl Oncol. \n2019;12:1122-30.\n 15. Keenan JA, Williams-Boyce PK, Massey PJ, Chen TT, Caudle MR, \nBukovsky A. Regression of endometrial explants in a rat model of endo -\nmetriosis treated with the immune modulators loxoribine and levamisole. \nFertil Steril. 1999;72:135-41.\n 16. Donnez J, Smoes P, Gillerot S, Casanas-Roux F, Nisolle M. Vascular \nendothelial growth factor (VEGF) in endometriosis. Hum Reprod. \n1998;13:1686-90.\n 17. Hagedorn A, Germann PG, Junker-Walker U, Tomovic A, Seewald W, \nPolkinghorne A, et al. Immunohistochemical study about the Flt-1/VEG -\nFR1 expression in the gastrointestinal tract of mouse, rat, dog, swine \nand monkey. Exp Toxicol Pathol. 2005;57:149-59.\n 18. Miettinen M, Lindenmayer AE, Chaubal A. Endothelial cell markers \nCD31, CD34, and BNH9 antibody to H-and Y-antigens--evaluation of \ntheir specificity and sensitivity in the diagnosis of vascular tumors and \ncomparison with von Willebrand factor. Mod Pathol. 1994;7:82-90.\n 19. Burns KA, Pearson AM, Slack JL, Por ED, Scribner AN, Eti NA, et al. \nEndometriosis in the mouse: challenges and progress toward a ‘Best Fit’ \nmurine model. Front Physiol. 2021;12:806574.\n 20. La Rosa VL, De Franciscis P, Barra F, Schiattarella A, Török P, Shah M, \net al. Quality of life in women with endometriosis: a narrative overview. \nMinerva Med. 2020;111:68-78.\n 21. Simoens S, Dunselman G, Dirksen C, Hummelshoj L, Bokor A,  Brandes  I, \net al. The burden of endometriosis: costs and quality of life of women with \nendometriosis and treated in referral centres. Hum Reprod. 2012;27:1292-9.\n 22. Hey-Cunningham AJ, Peters KM, Zevallos HB, Berbic M, Markham R, \nFraser IS. Angiogenesis, lymphangiogenesis and neurogenesis in endo -\nmetriosis. Front Biosci (Elite Ed). 2013;5:1033-56.\n 23. Hung SW, Zhang R, Tan Z, Chung JP, Zhang T, Wang CC. Pharmaceu -\nticals targeting signaling pathways of endometriosis as potential new \nmedical treatment: a review. Med Res Rev. 2021;41:2489-564.\n 24. Soysal D, Kızıldağ S, Saatlı B, Posacı C, Soysal S, Koyuncuoğlu M, et al. \nA novel angiogenesis inhibitor bevacizumab induces apoptosis in the rat \nendometriosis model. Balkan J Med Genet. 2014;17:73-80.\n 25. Karslıoglu T, Karasu AF, Yildiz P. The effects of micronized progestero -\nne and cabergoline on a rat autotransplantation endometriosis model: \nA placebo controlled randomized trial. J Invest Surg. 2021;34:897-901.\n 26. Bakacak M, Ercan Ö, Köstü B, Bostancı MS, İnanç F, Yaylalı A, et al. \nThe effects of thalidomide in a rat model of surgically-induced endome -\ntriosis. Turk J Obstet Gynecol. 2015;12:125-31.\n 27. Meresman GF, Bilotas MA, Lombardi E, Tesone M, Sueldo C,  Barañao RI. \nEffect of GnRH analogues on apoptosis and release of interleukin-1beta \nand vascular endothelial growth factor in endometrial cell cultures from \npatients with endometriosis. Hum Reprod. 2003;18:1767-71.\n 28. Moraloglu O, Işik H, Kiliç S, Sahin U, Caydere M, Ustün H, et al. Effect \nof bevacizumab on post-operative adhesion formation in a rat uterine \nhorn adhesion model and the correlation with vascular endothelial grow -\nth factor and Ki-67 immunopositivity. Fertil Steril. 2011;95:2638-41.\n 29. Pala HG, Erbas O, Pala EE, Artunc Ulkumen B, Akman L, Akman T, et al. \nThe effects of sunitinib on endometriosis. J Obstet Gynaecol. 2015;35:183-7.\n 30. Siracusa R, D’Amico R, Impellizzeri D, Cordaro M, Peritore AF, Guglian -\ndolo E, et al. Autophagy and mitophagy promotion in a rat model of \nendometriosis. Int J Mol Sci. 2021;22:5074.\n 31. Yildiz C, Kacan T, Akkar OB, Karakus S, Seker M, Kacan SB, et al. Effect \nof imatinib on growth of experimental endometriosis in rats. Eur J Obstet \nGynecol Reprod Biol. 2016;197:159-63.\n 32. Ozdemir EU, Adali E, Taskin MI, Yavasoglu A, Aktug H, Oltulu F, et al. \nEffects of ranibizumab and zoledronic acid on endometriosis in a rat \nmodel. Arch Gynecol Obstet. 2022;305:267-74.","source_license":"CC0","license_restricted":false}