Intravascular Adenomyosis in a Cynomolgus Macaque (Macaca Fascicularis)

An adult female cynomolgus macaque ( Macaca Fascicularis ; 8y 6 m old) weighing 6.9 kg reached study endpoint, was humanely euthanized, and underwent experimental necropsy. The animal was originally purchased from Alpha Genesis and had an uneventful clinical history other than being overweight due to her experimental diet, a diet designed with high fat content to mirror a typical “Western Diet”. There were no reproductive manipulations in her experimental history. On gross exam the uterus weighed 3.17 g (0.05% BW) and was in range with the rest of the experimental cohort ( n = 7, 2.91–8.04 g). The uterus was 24.88 mm long, 24.88 mm wide, and 10.63 mm thick (anterioposterior) and again was in range with the rest of the experimental cohort ( n = 7, 20.97–33.44 mm L x 13.48–24.1 mm W x 10.5–16.95 mm T). On sectioned surface the myometrium was pale pink and the endometrium was hemorrhagic. The uterus was collected as five equally thick cross sections spanning from the cervical junction to the apex. On histologic examination, the myometrium had increased clear space due to diffuse dilation of thin-walled vessels interpreted as veins ( Figure 1B ). Within multiple regions of the myometrium in individual cross sections and extending throughout all five collected cross-sections were multiple profiles of veins that contained intimately associated islands or stalks of endometrial tissue containing both glandular and stromal elements ( Figure 1C ). Invasion depth was grade 3 and glandular involvement was marked. Cellular detail of the endometrial tissue within these areas was consistent with the intact endometrial lining (menstrual phase; Figure 1D ). There was no evidence of increased mitotic figures or disorganization in the vessel-associated endometrial tissue. Immunohistochemistry was performed on tissue sections using horse-radish peroxidase conjugated secondary antibodies and 3,3′-diaminobenzidine (DAB) detection chromogen. The antibody panel consisted of cytokeratin (Abcam, ab9377, 1:500) and CD31 (Biocare Medical, CM 131 A, C, 1:200). The endometrial glands demonstrated immunopositivity for cytokeratin ( Figure 2A , C ). The vascular associated islands and stalks of endometrium were surrounded by a coat of CD31 immunopositive endothelium ( Figure 2B , D ). Findings were consistent with a large case series of human intravascular adenomyosis [ 7 ]. A comprehensive assessment was conducted of the heart, lungs, spleen, left kidney, left adrenal gland, pelvic lymph nodes, vagina, cervix, and left ovary to identify any abnormalities. A few minor incidental findings were noted, none of which were related to the intravascular adenomyosis. In the lungs, small, scattered clusters of lymphocytes were observed that were not associated with the bronchi. In the kidneys, Bowman's space frequently appeared distended, containing pale pink protein-aceous fluid.

The theory that metastatic or embolic endometriosis originates from endometrial tissue disseminated through venous circulation can be traced back to John A. Sampson in 1927 [ 5 ]. Sampson described surgical observations in women where he filled menstrual phase uterine lumen with saline and the fluid dilated and then escaped through the uterine and ovarian veins. Upon this discovery he began to map the uterine blood supply in detail discovering that numerous veins in the uterus are merely endothelially lined spaces with poorly defined walls, often mistaken for lymphatics, that communicate directly with endometrial capillaries. These veins do not contain valves, and he postulated that at menstruation capillaries of the endometrium rupture and menstrual tissue becomes entrapped due to venous pressure. He then documented instances of intravascular endometrial stroma and glands. This documentation of intravascular endometrial stroma and glands was repeated by Sahin et al. in 1989 where they found intravascular fragments of endometrial tissue in 17.9% of all cases of adenomyosis with no limitation to menstrual phase [ 6 ]. In 2010, Meenakshi et al. again found similar cases of adenomyosis with vascular involvement at an incidence of 12.4% [ 7 ]. Meenakshi followed up with immunostaining for CD31 and CD34 in five of their 54 identified cases and documented an intact layer of CD31 immunopositive endothelial cells surrounding cases of intravascular adenomyosis. We report the first documented case of intravascular adenomyosis in a cynomolgus macaque. The presentation closely mirrors descriptions in human literature, notably featuring an intact surrounding layer of CD31-immunopositive endothelium. The occurrence of intravascular adenomyosis in multiple species that naturally develop both adenomyosis and endometriosis lends additional support to the embolic endometriosis theory originally proposed by John A. Sampson in 1927.

Diagnosis of adenomyosis, presence of endometrial tissue within the myometrium, is relatively straightforward on histologic examination of uterine tissue. Case reports exist for adenomyosis in chimpanzees ( Pan troglodytes ), baboons ( Papio sp.), rhesus ( Macaca mulatta ), pigtails ( Macaca nemestrina ), and cynomolgus macaques ( Macaca fascicularis ), but overall rates of incidence in nonhuman primates are limited [ 1 ]. In our cynomolgus macaque cohort at Wake Forest, we have necropsied and evaluated uteri in 89 adult reproductive age (> 4years old) females of various origins (Cambodia, Chinese, Indonesian, Malaysian, Mauritius, and Vietnamese) between 2007–2025. Twenty-five (28.1%) had a diagnosis of adenomyosis (10/29 Cambodia, 1/1 Chinese, 10/40 Indonesian, 1/1 Malaysian, 2/17 Mauritius, and 1/1 Vietnamese). This is comparable to the mean prevalence reported in human literature of 20%–25% [ 2 ]. In 1972, Bird et al. developed a grading system to denote the depth of invasion into the myometrium: grade 1-adenomyosis is limited to within one low-power field of the basal endometrium, grade 2-adenomyosis penetrates to mid-myometrium, and grade 3-adenomyosis penetrates to beyond mid-myometrium [ 3 ]. Bird et al. also developed a scoring system for glandular involvement. A score of ‘slight’ glandular involvement denotes few (1–3) glands per low-power field, a score of ‘moderate’ denotes several (4–9) glands per low-power field, and a score of ‘marked’ denotes many (10+) glands per low power field [ 3 ]. These scoring paradigms are clinically beneficial as depth of adenomyosis penetration correlates with severity of associated dysmenorrhea and pelvic pain [ 3 ]. However, these scoring systems have had little adoption and there is an urgent need for this to be remedied [ 4 ]. Aside from the depth of penetration and granularity, variable presentations of adenomyosis have been underreported in the literature. Herein we describe a unique case of vascular associated adenomyosis in a cynomolgus macaque which shows similarity to reports in human literature of intravascular adenomyosis [ 5 - 7 ]. These presentations of intravascular adenomyosis support a vascular dissemination pathogenesis for adenomyosis and endometriosis in primates.