Comparative reproductive pathology, a walk on the wild side

Endometrial hyperplasia (EH) is the excess proliferation of endometrial tissue, sometimes resulting in thickened endometrium and cystic glands (cystic EH) and leading to mineralization or fibrosis, followed by infertility or subfertility, depending on the severity of the lesions ( Fig. 1 ). EH is extremely common in canids and felids that have spent time without a pregnancy and is accelerated in those exposed to exogenous progestins ( 4 , 5 ). Among canids, African painted dogs ( Lycaon pictus ) are 8 times more likely to develop EH than other species ( 5 ); the reason for this is unknown, but unique reproductive characteristics include the largest litters of all canids and very stressful social dynamics in terms of social dominance and grouping. Among primates, the prevalence of EH in lemurs (52%; n = 98) and orangutans ( Pongo pygmaeus ) (52%; n = 25) is high relative to other primates (29%; n = 145) (RHSP, unpublished data). Lemur’s reproductive anatomy and physiology appear more similar to those of carnivores, suggesting similar predispositions may be at play, but the high prevalence in orangutans compared with other primates is unexplained. Figure 1 Photomicrograph of endometrial cystic hyperplasia in a non-domestic carnivore. HE = hematoxylin and eosin. Photomicrograph of endometrial cystic hyperplasia in a non-domestic carnivore. HE = hematoxylin and eosin. Endometriosis (defined as endometrial tissue outside of the uterus) is a painful condition that can result in infertility, yet the exact pathophysiology remains unclear. The overall prevalence in women varies among studies but has been documented to be as high as 18% in the general population in developed countries ( 6 ) and up to 80% among women with chronic pelvic pain ( 7 ). New hypotheses on its pathogenesis are being formulated as molecular tools improve. Among animals, spontaneous endometriosis has only been documented in NHPs ( Fig. 2 ), in particular, great apes and Afroeurasian monkeys ( 8 , 9 , 10 ), but reports on Neotropical monkeys are rare ( 11 , 12 , 13 ), and there are no published reports of endometriosis in lemurs. This progressively lower prevalence mirrors the amount of endometrial tissue that is shed at the end of the luteal phase: from Afroeurasian species with visible amounts of blood to Neotropical species with microscopic amounts and to prosimimans with no true menses. Some species of bats and the spiny mouse have been shown to also menstruate and have been proposed as potential models to study endometriosis, although spontaneous endometriosis has not yet been documented in either species ( 14 ). The data about primates support retrograde menstruation as a factor in the pathogenesis of endometriosis; thus, studying other species may shed light on additional factors or pathways to the development of endometriosis. It should be noted that in some older veterinary literature studies, “endometrial hyperplasia” has been labeled as “endometriosis,” but this terminology can be misleading and is no longer used. Figure 2 Endometriosis in a gorilla ( Gorilla gorilla ). Arrows indicate areas of endometrial tissue hemorrhage (chocolate cysts). Endometriosis in a gorilla ( Gorilla gorilla ). Arrows indicate areas of endometrial tissue hemorrhage (chocolate cysts). Adenomyosis is the migration of endometrial glands into the myometrium. Endometriosis and adenomyosis are characterized by the presence of ectopic endometrial tissue. The pathogenesis of both conditions is still under debate, but there is evidence to suggest that they might be different phenotypes of a single disease ( 15 ). Adenomyosis is commonly seen in several species, but extremely rare in other closely related species. African painted dogs are prone to severe adenomyosis, sometimes resulting in uterine rupture ( Fig. 3 and 4 ), although implantation of glandular material into the peritoneal cavity (endometriosis) has not been reported. Even though other canids also have a high prevalence of EH, none equal the African painted dog’s prevalence and severity of adenomyosis. Some bat species known to menstruate have been reported to develop adenomyosis, providing an interesting comparison to humans ( 16 ). In humans, prevalence is higher in parous women compared with nulliparous women, supporting the theory of endometrial-myometrial disruption where endometrial tissue can invaginate into the myometrium when the tissue is disrupted by peristalsis, pregnancy, or medical procedures such as curettage ( 17 ). Interestingly, the reports on bats pertain to bats housed in all-female groups, and therefore, there were no pregnancies during the time covered by the study. African painted dogs, with their extremely large litters and presumably marked stretching of the uterus, would support the endometrial-myometrial disruption theory of adenomyosis pathogenesis, although adenomyosis is seen in nulliparous and multiparous individuals. Among felids, there were noteworthy differences: tigers ( Panthera tigris ; 6.25%, n = 32) and snow leopards ( Panthera uncia ; 0%, n = 20) seem to be significantly less susceptible to it than leopards ( Panthera pardus ; 41.18%, n = 17) (RHSP, unpublished data). Figure 3 Rupture of the uterus in an African painted dog ( Lycaon pictus ) due to adenomyosis. Figure 4 Photomicrograph of the uterine wall of an African painted dog ( Lycaon pictus ) with adenomyosis (arrows), where glands migrate from the luminal endometrial glands (L) to the serosa, sometimes leading to rupture. HE = hematoxylin and eosin. Rupture of the uterus in an African painted dog ( Lycaon pictus ) due to adenomyosis. Photomicrograph of the uterine wall of an African painted dog ( Lycaon pictus ) with adenomyosis (arrows), where glands migrate from the luminal endometrial glands (L) to the serosa, sometimes leading to rupture. HE = hematoxylin and eosin. Ovarian cysts can include multiple conditions, such as cystic follicles, anovulatory follicles, epithelial inclusion cysts, cystic rete ovarii, and Müllerian duct cysts, among others. Some conditions can have significant effects on the estrous cycle and others will have none, representing a clinical problem only if it invites unnecessary surgery or medical intervention. Among the callitrichids, golden lion tamarins ( Leontopithecus rosalia ) have a markedly higher prevalence of cystic ovarian disease (COD) (mostly rete cysts) than other callitrichids (23.3%, n = 30, 7.3%, n = 41, respectively) (RHSP, unpublished data), with cysts occasionally becoming large enough to need surgery because of their space-occupying effect (Moresco A, unpublished data). Among ungulates, COD is relatively common in some suids and is associated with exposure to progestin contraception ( 18 ). In contrast to the condition in women, COD with an active hormonal component is rarely diagnosed in zoo animals, except for granulosa cell tumors. Mammary gland adenocarcinoma (MG-AdCa) is relatively common in intact female felids and the histopathologic characteristics of feline MG-AdCa resemble recurrent breast cancer and BRCA gene mutation in women: they tend to be of high grade and are often hormone receptor negative ( 19 ). Exposure to exogenous progestins increases the odds ratio of developing MG-AdCa in felids ( 20 ). In women, the evidence for progestin exposure as a risk factor is less clear: the Million Women Study on hormone replacement therapy found an increased risk of breast cancer in women who took a combined (estrogen + progesterone) therapy compared with women who took the estrogen only therapy ( 21 ). However, more recent observational studies have failed to show the same effect ( 22 ). Interestingly, large felids ( Panthera spp ) are more likely to develop malignant neoplasia, including MG-AdCa, than small non-domestic felids ( Felis spp ) ( 23 ,  24 ). Systematic research is lacking in many non-felid species and thus the prevalence of mammary pathology is unknown, but other carnivores such as the black-footed ferrets ( Mustela nigripes ) and binturong ( Arctictis binturong ) also appear to have a relatively high risk of developing MG-AdCa ( 25 , 26 ). In contrast, mammary neoplasia rarely develops in ungulates; among domestic ungulates (cow, sheep, and goats), this lack of mammary cancer is thought to be associated with the fact that they are production animals, giving birth and lactating at an early age. That is, mammary gland epithelium terminally develops early in life, reducing the risk of MG-AdCa, similar to the decreased risk in women who had children early compared with nulliparous women or women who had their children later in life. Interestingly, malignant mammary neoplasia is uncommon among NHPs ( 9 ). Jaguars ( Panthera onca ) are an intriguing comparative example of reproductive pathology in multiple organs, because they are highly susceptible to ovarian, mammary, and endometrial carcinoma, reminiscent of similar susceptibilities in women with BRCA gene mutations ( Fig. 5 ). Genetic and molecular analyses demonstrated missense germline variants in the jaguar BRCA1 and BRCA2, but population analyses showed only the BRCA2 variant to be present in all the jaguars analyzed, which had ovarian carcinoma. Further studies using more robust molecular tools are required to better define this effect, compare it to non-affected wild jaguar samples as well as other species, and rule out other possible undiscovered genes with tumor suppression function in this species ( 27 ). Jaguars are also the only species among felids in which ovarian papillary cystadenocarcinomas have been diagnosed, potentially associated with the BRCA mutation observed in them ( 28 ). Figure 5 Photomicrograph of ovarian papillary cystadenocarcinoma in a jaguar ( Panthera onca ). HE = hematoxylin and eosin. Photomicrograph of ovarian papillary cystadenocarcinoma in a jaguar ( Panthera onca ). HE = hematoxylin and eosin. Endometrial adenocarcinoma (E-AdCa) in felids is associated with exposure to progestin contraception ( 29 ). However, the direct effect of exposure to progestin contraception as opposed to a lack of pregnancy should be investigated more. Among great apes, although the sample size is small, the prevalence of E-AdCa varies, with a higher prevalence in gorillas than in orangutans (16% [3/19] and 4% [1/24], respectively) ( 30 , 31 ). Although E-AdCa has been reported across many taxonomic groups, there are some species in which it seems to be more prevalent, like domestic rabbits, which are an interesting species for comparative research into the relationship between EH and E-AdCa ( 32 ). In addition to other reproductive and non-reproductive neoplasia, E-AdCa has been reported in free-ranging belugas (Delphinapterus leucas) and was thought to be associated with environmental toxins in the water ( 33 ). Uterine leiomyomas (smooth muscle tumors or fibroids) have been documented in a wide range of species and are a common cause of infertility in women and other species. In humans, leiomyomas have been associated with multiple factors, including exposure to exogenous hormones and nulliparity. Similar effects have been seen in some veterinary species but not in others. Leiomyomas are commonly reported in Asian elephants with strong evidence supporting an association with age and nulliparity ( 34 ). In large felids, leiomyomas were not associated with exogenous progestin contraception (melengestrol acetate [MGA]) when all felids were evaluated together; however, when felid species were considered separately, lions showed an increase in prevalence associated with MGA exposure (RHSP, unpublished data; Fig. 6 and 7 ). Large felids without an increase in reproductive leiomyoma prevalence associated with MGA exposure (leopards and tigers) typically have a solitary lifestyle, whereas lions are a highly social species. This suggests a need to investigate genetic, evolutionary, physiologic, or population management factors that may drive this difference in response to MGA exposure among closely related species. Among ungulates, the Tragelaphines (kudu and similar antelope) are more prone to uterine leiomyomas than other ungulate species, suggesting again that there may be genetic drivers linked to their natural history or breeding strategies; however, further research is required to tease out these potential mechanisms (RHSP, unpublished data). Figure 6 Multiple leiomyomas in the uterine lumen and wall of a lion ( Panthera leo ). Figure 7 Photomicrograph of a leiomyoma in a lion ( Panthera leo ). HE = hematoxylin and eosin. Multiple leiomyomas in the uterine lumen and wall of a lion ( Panthera leo ). Photomicrograph of a leiomyoma in a lion ( Panthera leo ). HE = hematoxylin and eosin. The full range of common testicular tumors has been reported in wild animals, including seminomas, sustentacular (Sertoli) cell tumors, and interstitial (Leydig) cell tumors ( 35 ). In humans, seminomas are typically highly malignant and occur in younger males; histologically, they are considered “classical seminomas”( 36 ), whereas in domestic and zoo species, classical seminomas are more rarely diagnosed and spermatocytic seminomas are more commonly identified. Spermatocytic seminomas are benign, indolent neoplasms commonly seen in older animals. The prevalence in the literature is likely skewed by the fact that classical seminomas are more likely to result in surgical intervention and/or death, whereas spermatocytic seminomas are incidental findings identified after routine castration or death as incidental findings and are apt to be considered unworthy of publication. Similarly, sustentacular and interstitial cell tumors are commonly seen but rarely reported in exotic species because they are often benign and identified incidentally at the time of sterilization or death. However, this perspective belies the potential impact these tumors may have on an individual’s fertility, and in an endangered species, on the impact to the population’s survival. Although many of the previous conditions may affect a subpopulation because of sharing the same environment, genetic background, or similar reproductive history, infectious diseases (bacterial, viral, and protozoal) can be transmitted between animals and can have a devastating effect in a naïve population because they can have a major impact on fertility in animals and in humans. Similarities in the pathogenesis across species can drive decisions on diagnostic and control methods. Herpesviruses are ubiquitous across the animal kingdom and can provide valuable information on vaccine strategies and other control methods. Our understanding of herpesviruses as causative agents of urogenital cancers in sea lions and toward their compounding role with environmental toxins provides valuable insight into herpesviral pathogenesis in humans ( 37 ). Less well-known examples of infectious reproductive disease similarities across species are the Treponema sp. associated with syphilis in humans and similar treponemal infections seen in rabbits, Gilbert’s potoroos ( Potorous gilbertii ), and chimpanzees ( 38 ,  39 ). Disease in these species can lead to significant infertility and potentially more severe systemic diseases, as in humans. One important aspect to keep in mind when carrying out comparative research is to make sure that one has the “normal ranges” for that species, because these can vary significantly between species. As an example, an investigation into infertility in an individual gorilla included measuring prolactin levels because hyperprolactinemia is a common cause of infertility in women. When compared with the human normal range from the laboratory, this gorilla had very high levels and was initially diagnosed with hyperprolactinemia ( 40 ). However, when additional gorillas were sampled to obtain normal ranges for the species, it was discovered that gorillas naturally have higher levels than humans at every phase of the cycle ( 41 ). In conclusion, it is clear there are many “natural animal models” of human disease that can be tapped as we seek a fuller understanding of reproductive diseases. We have provided just a few examples. In fact, as we try to save species endangered by their own infertility, humans may provide excellent “models of animal disease” from which veterinarians may extract useful information to enhance reproduction and reduce disease in their patients.

A.M. has nothing to disclose. D.W.A. has nothing to disclose.