Risk factors for the development of endometrial lesions in breast cancer patients using tamoxyphen: a retrospective cohort study

ABSTRACT Introduction: breast cancer is the cancer with the highest incidence in women in Brazil, representing 29.7% of all cancers. More than two thirds of women with breast cancer show expression for hormone receptors, and in these cases, hormone therapy with tamoxifen is indicated, which may represent a risk factor for the development of endometrial cancer (four-fold greater relative risk). Objective: this study aimed to evaluate the association of tamoxifen and the development of endometrial disturbances and to assess possible other associated risk factors. Patients and method: a total of 364 breast cancer patients were evaluated, 286 who used tamoxifen and 78 who did not use this hormone therapy. Results: patients who used tamoxifen had a mean follow-up time of 51.42 months similar to those without hormone therapy (p=0.081). A total of 21 (7.3%) women who used tamofixen and no cases among women without hormone therapy presented endometrial changes during follow-up (p=0.01). Despite information regarding obesity was available for only 270 women, obesity was also significantly associated with the development of endometrial changes (p=0.008). Conclusion: furthermore, the association between tamofixen and endometrial changes remained significant (p=0.039) after adjusting for obesity.

women with breast cancer exposed to tamoxifen and assess which factors could be associated with it.

PATIENTS AND METHODS
This study has a retrospective cohort design.
We evaluated women with breast cancer treated at the Oncology Service of a general hospital in the interior of Paraná, between 2010 and 2020, diagnosed with breast carcinoma and with a minimum follow-up of six months.We excluded patients with other types of breast tumors, as well as those whose records did not contain information on the start date of hormone therapy in the group that received tamoxifen or on the date of surgery in the group without hormone therapy.
We collected data retrospectively, through access to the electronic medical records, after approval by the Ethics and Ethics in Research Committees and registration on the Brazil Platform.
• Group 2 (G2): patients with negative immunohistochemical reaction for hormone receptors (ER-/PR-) or who have not received hormone therapy with tamoxifen.
In both groups, we evaluated the development of endometrial disorders, such as endometrial thickening on ultrasound (thickness ≥10mm), endometrial hyperplasia, endometrial polyp, or endometrial cancer.
In addition, we collected data on the associated factors obesity, arterial hypertension, and diabetes mellitus.
We compared the frequency of associated factors in the two groups using the Chi-square test with Pearson's correction, and the follow-up time for each group using the Student's t test for independent samples.
We also used the chi-square to analyze the association between exposure to tamoxifen, as well as the associated factors, and the onset of endometrial disorders.As the time of exposure to tamoxifen was not homogeneous, we assessed the risk of endometrial disease with survival curves (Kaplan-Meier) in both groups.We applied this same analysis to compare the occurrence of the outcome between obese and non-obese women.We performed the statistical analyzes with the SPSS ® version 21 software.

RESULTS
We of the patients at diagnosis is shown in Table 1.2).In the 78 patients who did not receive hormone therapy, there were no cases of endometrial disease.
The use of tamoxifen was significantly associated with the development of endometrial changes, as shown in Table 5 and Figure1 (p=0.014).Of the 286 women treated with tamoxifen, 21 (7.3%) had some type of endometrial alteration, and eight had endometrial thickening (endovaginal pelvic ultrasound showing endometrial echo >10mm) and were not submitted to any further investigative procedure.The other 13 patients underwent uterine curettage, hysteroscopy, or hysterectomy, whose anatomopathological studies showed the presence of endometrial hyperplasia in eight cases, endometrial polyp in two, endometrial adenocarcinoma in one case, atrophic endometrium in one, and submucous leiomyoma in one case (Table 4).The assessment of the association of tamoxifen use adjusted for obesity is shown in Table 7 (Mantel-Haenzel chi-square, p=0.088).

DISCUSSION
In the present study, the incidence of endometrial disorders occurred in 7.3% of the patients who received this medication for more than six months and in none of those without hormone therapy.
Although it was not possible to assess the risk (odds ratio) of hormone therapy in the development of endometrial changes, as in the present sample patients who did not receive hormone therapy with tamoxifen did not develop endometrial changes, the difference in the incidence of such changes between groups was statistically significant (p =0.014).

The use of tamoxifen as an adjuvant therapy
for breast cancer is an effective and widely used treatment in these patients, as already demonstrated in several studies in the literature 3,4,14,18 .Although its effect on the endometrium is controversial 6 , its administration for long periods is related to the appearance of endometrial alterations, possibly due to an "estrogenic action" on the endometrium, although it is an antagonist of the receptor of this hormone [7][8][9] .
The incidence of endometrial diseases in the group of women using tamoxifen for the treatment of breast cancer is quite variable in the literature, between 29% In several studies [8][9][10] , obesity was a risk factor significantly associated with the development of endometrial diseases, especially endometrial cancer, in women with breast cancer using tamoxifen.In this study, obesity was present in 16.3% of the women who used tamoxifen and in 21.3% of those who did not receive it.In the group that used tamoxifen, obesity increased the relative risk of developing endometrial disease by 5.17 when compared with non-obese women who took tamoxifen, this increase being significant (p=0.001).The Regarding to arterial hypertension, among the women who used tamoxifen, 78 were hypertensive (27%); of those, three (3.8%)developed endometrial changes, versus 13 (7.8%) of the 166 patients without arterial hypertension (Table 6, p=0.241).Diabetes mellitus was also not associated with the outcome, with 3.8% and 6.9% in the groups exposed and not exposed to tamoxifen, respectively (p=0.548).

Vizzotto Jr
Risk factors for the development of endometrial lesions in breast cancer patients using tamoxyphen: a retrospective cohort study non-obese women (Table 7) (Mantel-Haenzel chi-square, p=0.088).
Arterial hypertension, present in 27% of women who used tamoxifen and in 29% of women who did not use it, was not associated with endometrial changes during follow-up (p=0.241).

Table 6
shows the univariate association of the assessed risk factors with the development of endometrial changes during follow-up.Only obesity showed a significant association with the development of endometrial changes during follow-up p=0.001.

Figure 2 .
Figure 2. Outcome evolution (endometrial changes) in patients with and without tamoxifen use (time in months).

Figure 2
Figure 2 shows the association of obesity with the development of endometrial changes during followup according to their time of onset.

Figure 1 .
Figure 1.Evolution of the outcome (endometrial changes) in patients with and without obesity (time in months).

(
Fisher et al., 2005) 4 and 61% (Exacoustós et al., 1995) 5 .Bernstein et al. (1999) 3 showed a risk 1.52 times greater for endometrial cancer in patients who used tamoxifen, reaching 4.06 times greater when this use lasted for more than five years.Our findings corroborate the data presented in the literature.

Table 1 -
Distribution of clinical staging.

Table 2
ER: Estrogen Receptor; PR: progesterone receptor; HER2: presence of the her-2 gene expression product receptor.

Table 3 -
Frequency of risk factors in groups of women who used and did not use tamoxifen.

Table 3
lists the frequency of risk factors in the two groups of women with and without tamoxifen.

Table 5 -
Hormone therapy and endometrial changes.

Table 6 -
Association between risk factors and development of endometrial changes (outcome) in the group of women exposed to tamoxifen.