Objective
We sought to determine the prevalence of adenomyosis and assess its effect on lymph node status in endometrioid adenocarcinoma of the endometrium (EAC).
Study Design
Hysterectomy specimens from a single institution were reviewed for the presence of adenomyosis, lymphovascular space invasion (LVSI), tumor grade, histology, and lymph node status. Standard statistical analysis was used to compare variables.
Results
Adenomyosis was present in 42% of total and 66% of malignant hysterectomy specimens ( P = .009). Adenomyosis was most commonly associated with EAC histology ( P = .023). LVSI was found to be an independent predictor of lymph node metastasis in EAC patients without adenomyosis, but not in those with coexisting adenomyosis (odds ratio, 58.7; P = .03; and odds ratio, 4.98; P = .15; respectively).
Conclusion
Adenomyosis was associated with a lower risk of lymph node metastasis in EAC patients with LVSI. Further studies are needed to investigate the role of adenomyosis in lymphatic tumor infiltration.
Endometrial cancer is the most common gynecologic malignancy, with an annual incidence of approximately 46,470 cases in the United States. Factors such as histology, grade, depth of myometrial invasion, presence of lymphovascular space invasion (LVSI), and lymph node metastases are known prognostic factors. In contrast to endometrial carcinoma, adenomyosis is a common benign condition defined by the presence of endometrial glands and stroma within the myometrium. Adenomyosis typically presents as menorrhagia and dysmenorrhea in multiparous women during the fifth and sixth decade of life.
Several previous studies have documented coexistent adenomyosis and endometrial cancer ( Figure 1 ). The reported incidence of the presence of these 2 conditions in hysterectomy specimens is variable, ranging from 10–70%. The clinical relevance of endometrial adenocarcinoma involving adenomyosis remains unclear. Many of the early investigations on this subject found that endometrial adenocarcinomas with adenomyosis tend to have low histologic grades, superficial myometrial invasion, and excellent prognosis. However, more recent studies have been contradictory and suggest adenomyosis is associated with deep myometrial invasion. Previous investigators have suggested that adenomyosis can undergo malignant change and therefore may represent a precursor lesion to adenocarcinoma. However, there is no clear study demonstrating the natural transformation of adenomyosis to adenocarcinoma and the relationship between these 2 diseases remains speculative. In this study we assess the prevalence of adenomyosis in our population as well as its association with histology, LVSI, and likelihood of lymph node involvement of coexistent endometrial adenocarcinomas.
Materials and Methods
Approval for this study was obtained from the institutional review board. Using the institution’s pathology database we identified and reviewed retrospectively all patients who underwent total hysterectomy from 2000 through 2006. Patients undergoing hysterectomy via either laparoscopy or laparotomy at a single institution who were found to have endometrial adenocarcinoma were included. Exclusion criteria included >1 concurrent primary tumor, surgery performed at an outside institution, and tumors of pure mesenchymal origin. Patients with endometrial carcinosarcoma were included if the epithelial component of the tumor was predominantly endometrioid.
We reviewed patient demographic information and pathology reports. Pathologic variables abstracted included presence of adenomyosis, tumor grade, depth of myometrial invasion, LVSI, lymph node status, and presence of distant spread. Clinical information obtained from hospital and outpatient records included patient age, ethnicity, body mass index (BMI), history of comorbid conditions (hypertension and/or diabetes), postoperative course, survival status, and date of last follow-up.
The pathology department at this institution defines adenomyosis as the presence of endometrial glands and stroma in the myometrial wall at least 3-μm distance from the endometrium. The protocol for sectioning uteri depends on the underlying diagnosis. For a completely benign uterus, 6 sections are obtained: anterior and posterior cervix, anterior and posterior lower uterine segment, and anterior and posterior full-thickness endometrium to serosa.
For a malignant uterus the protocol is the same; however, the pathologist attempts to obtain a full-thickness section at the point of deepest invasion of the tumor, as well as 1 additional section of tumor per centimeter of tumor present. For a uterus with complex atypical hyperplasia or tumors with no apparent tumor the sectioning is the same as for benign cases; however, the entire endometrium is examined as well. The endometrial samples generally extend 3-5 mm into the myometrium.
The normality of the data was evaluated using the Kolmogorov–Smirnov test and normality plots and histograms for each of the variables, with evaluation of skewness and kurtosis. Patients with endometrioid adenocarcinoma of the endometrium (EAC) and adenomyosis were compared to patients with endometrioid EAC alone with regard to the abstracted pathologic variables. Normally distributed continuous variables were compared using the Student t test. Categorical variables were compared using the χ 2 test and Fisher exact test. Variables that were nonnormally distributed were analyzed with Mann-Whitney U test. Logistic regression analysis was performed to determine the independent predictors of lymph node metastasis. Survival was calculated using the method of Kaplan-Meier and compared using the log rank test. Data were analyzed using commercially available software (SPSS, version 16.0; IBM Corp, Armonk, NY). A P value of < .05 was considered significant for all tests.
Results
We identified 2346 hysterectomies performed at a single institution from 2000 through 2006. Among these, final pathology demonstrated EAC in 197 cases (8.4%). Adenomyosis was diagnosed in 42% of hysterectomy specimens and 66% of specimens with EAC ( P = .009). The prevalence of adenomyosis was significantly greater in patients with endometrioid EAC as compared to patients with EAC of other histologic subtypes (75% vs 48%, P = .023) ( Table 1 ).
| Adenomyosis diagnosed in: | |
| 42% | Hysterectomy specimens |
| 66% | EAC specimens |
| Prevalence of adenomyosis by EAC histology: | |
| 75% | Endometrioid histology |
| 48% | Other histology |
There was no significant difference among EAC patients with and without adenomyosis with regard to patient age (60.5 vs 63.4 years, P = .78) or BMI (28.4 vs 28.0, P = .13). There was also no significant difference in the presence of hypertension in patients with and without adenomyosis (38.6% vs 40.0%, P = .88). However, EAC patients without adenomyosis were more likely to have diabetes than EAC patients with adenomyosis (20.0% vs 7.9%, P = .04) ( Table 2 ).
| Demographic | Adenomyosis absent (n = 67) | Adenomyosis present (n = 130) | P value |
|---|---|---|---|
| Age, y | 63.4 | 60.5 | .780 |
| BMI | 28.0 | 28.4 | .125 |
| Hypertension | 40.0% | 38.6% | .876 |
| Diabetes | 20.0% | 7.9% | .044 |
Among patients with EAC, adenomyosis was associated with lower tumor grade, less myometrial invasion, negative LVSI, and negative lymph node pathology ( Table 3 ). LVSI was more common in patients without adenomyosis (33.3% of patient without adenomyosis, 8.6% of patients with adenomyosis). In endometrioid EAC patients without adenomyosis, 87.5% with LVSI had lymph node involvement (odds ratio [OR], 29.4; P = .001). In endometrioid EAC patients with adenomyosis, only 60% demonstrating LVSI had lymph node involvement (OR, 17.1; P = .008). When controlling for tumor grade and depth of myometrial invasion, LVSI was independently associated with lymph node metastasis in patients without adenomyosis (OR, 58.7; P = .03), but the relationship was not significant in patients with adenomyosis (OR, 4.98; P = .15) ( Table 4 ).
| Variable | Adenomyosis absent (n = 67) | Adenomyosis present (n = 130) | P value |
|---|---|---|---|
| Tumor grade | |||
| 1 | 36.8% | 73.5% | |
| 2 | 47.4% | 19.5% | |
| 3 | 15.8% | 7.1% | < .001 |
| Depth of myometrial invasion | |||
| 0 | 22.5% | 42.5% | |
| <50% | 50.0% | 46.0% | |
| >50% | 25.0% | 10.6% | .044 |
| Positive LVSI | 33.3% | 8.7% | .001 |
| Positive LN | 22.5% | 4.3% | .002 |
| Positive LVSI and positive LN | 87.5% | 60% | .008 |
| Variable | Adenomyosis absent (n = 67) | Adenomyosis present (n = 130) | ||
|---|---|---|---|---|
| OR | P value | OR | P value | |
| Grade | 1.35 | .85 | 1.99 | .35 |
| Depth of invasion | 0.23 | .35 | 4.32 | .1 |
| LVSI | 58.69 | .03 | 4.98 | .15 |
Figure 2 shows the survival curves of EAC patients with and without adenomyosis. A significantly improved survival was demonstrated in patients with adenomyosis compared with those lacking adenomyosis (77 vs 63 months, P = .02). Cox regression models were built to model overall survival adjusting for depth of myometrial invasion, tumor grade, and presence of adenomyosis. When controlling for tumor grade and the depth of myometrial invasion the presence of adenomyosis had no independent effect on survival ( Figure 3 ).
