Angiogenetic and hereditary factors in endometrial disease

Abstract

The endometrium is a unique tissue in the adult human body because it is the only site where physiological renewal of vessels, angiogenesis, occurs. The regulation of this process under normal conditions is not fully understood. Heavy Menstrual Bleeding of endometrial origin (HMB-E) and Endometrial Cancer (EC) are two diseases which can neither occur nor be sustained without the process of angiogenesis. Clinically, both diseases are also associated with a strong hereditary component where female relatives of affected patients demonstrate a high incidence of the same condition. The aim of this thesis is to further elucidate the angiogenic and hereditary components of these diseases. The first part of this project (studies I and II) explores HMB-E and its relationship to angiogenesis, while the second part (studies III and IV) examines the association between incidence of EC in relation to previously known and unknown hereditary factors.

Study I explored the relationship between expression of the pro-angiogenic factor SDF-1 and the number of circulating epithelial progenitor cells (EPCs) in peripheral blood in 10 women with a confirmed diagnosis of HMB-E, and compared these results with those of 10 healthy controls using flow cytometry and cell culture. The results showed a significant decrease of SDF-1 throughout the entire menstrual cycle, a 16% decrease overall, with the most substantial decline noted during the proliferative phase of the menstrual cycle in women with HMB-E, compared with controls. The number of EPCs in peripheral blood was also significantly reduced in HMB-E patients, showing a significant positive correlation between number of EPCs and SDF-1 levels. These findings are consistent with the literature, which has shown that SDF-1 is essential for recruitment of EPCs from bone marrow into the bloodstream, and further suggests that this signaling axis is important for physiological angiogenesis in the endometrium.

Study II addressed microvascular morphology within the endometrium among 17 women with HMB-E and compared the results with 10 controls, using immunohistochemistry and electron microscopy. We found a significant decrease in pericyte coverage of microvessels during the mid-proliferative phase, as well as an increase in vessel perimeter among women with HMB-E. We also found a negative correlation between vascular expression of the known pro-angiogenic factor VEGF-A and pericyte coverage. These findings indicate that HMB-E may be caused by altered microvessel maturation and that the overexpression of VEGF-A seen in HMB-E might be the underlying cause.

Study III examined the clinical challenge of reducing EC risk in women with Lynch Syndrome (LS), who collectively have a 42-54% lifetime risk of developing this disease. Sweden has no national guidelines regarding surveillance of these women for gynecological cancers. Therefore, we took a retrospective look at the modalities and screening intervals that are currently used in Sweden for known LS carriers. In all, 86 women had a known LS mutation and participated in various screening programs. Of the 41 women who decided to have prophylactic hysterectomy while under screening, EC/complex hyperplasia was found in the uterus of 4 of them postoperatively. The remaining 45 women opted for annual surveillance without prophylactic surgery. In this group, 9 women developed EC, 2 complex endometrial hyperplasia and 2 ovarian cancers, both of which were found at an early stage by ultrasound. Among these 9 EC cases, 5 were detected through endometrial biopsy during regular surveillance at an asymptomatic stage, as were the two hyperplasias. Ultrasonography failed to detect any of these cases. The remaining 4 EC cases were associated with occult bleeding between screening visits. No difference in tumor stage was noted between the ECs found because they were symptomatic and those found through routine surveillance. No mortality resulted from the gynecological cancers found in the study group as a whole. The results did not reveal any clear benefit from conducting annual gynecological screening for women with LS if the endpoint is to reduce mortality. The single most important factor may be knowledge and awareness among doctors and patients of LS and the associated increased risk of cancer, which should trigger prompt investigation if symptoms develop. However, when screening is undertaken, endometrial biopsy seems to be the diagnostic modality of choice to detect asymptomatic ECs or hyperplasia.

Study IV investigated additional genetic risk factors that help to explain the twofold increase in relative risk of developing EC among women with a first or second degree relative who have the disease, even when known single gene mutations are excluded. After genotype studies and quality control, a total of 332,906 SNPs among genotypes derived from 1116 EC cases and 5021 controls were compared. The results showed five haplotypes located on chromosomes no. 2, 10, 13, 15 and 20 that were significantly more common among EC patients than controls. The frequency of these haplotypes in the EC population ranged between 1.58-3.69% and the odds ratio for EC ranged between 1.58-3.05 for the five haplotypes. The five haplotypes were found in regions of the genome with no previously known link to EC development and without close proximity to any genes known to be involved with this disease. These findings add to our knowledge about the genetic risk factors associated with EC and may offer one explanation for why the incidence and clinical course of this disease differs among people of different ethnic backgrounds. Nevertheless, other risk factors must be taken into account when interpreting these results, including known environmental risk factors for EC and those that could potentially be inherited, such as obesity.

In conclusion, this thesis provides new information concerning the angiogenic and hereditary factors involved in development of both HMB-E and EC. The papers add to our knowledge about how the SDF-1 axis is involved in endometrial vascular regeneration during physiological endometrial angiogenesis and how vascular maturation appears to be inadequate in HMB-E. The thesis also confirms the knowledge about the role of known inherited cancer syndromes in development of EC, and finds that gynecological surveillance per se does not seem to reduce mortality from gynecological cancer; instead, the most important protective factor appears to be knowledge and awareness of the cancer syndrome among both carriers of the syndrome and the care providers. Additionally, this thesis found five haploptypes that are overrepresented in a Swedish cohort of EC cases, which may explain some of the risk of inheriting EC. The future challenge will be to examine the potential link between HMB-E and EC from the perspective of both clinical and basic research.