Article Text
Abstract
Objective Fertility preservation is an option for selected patients with endometrial hyperplasia or cancer. This study aimed to evaluate whether duration of treatment impacts the oncologic and reproductive outcomes.
Methods We retrospectively reviewed patients diagnosed with endometrial cancer/atypical endometrial hyperplasia who underwent fertility-sparing treatment from January 2012 to December 2016. As the duration of treatment required by the patients was different, the patients who achieved a complete response were grouped according to the treatment duration as groups A (≤6 months), B (6–9 months), and C (>9 months).
Results With the prolongation of treatment duration from 6 months to 9 months to >9 months, the accumulative complete response rates for 67 patients were 58%, 76%, and 95.5%, respectively. Among groups A, B, and C there was no significant difference in the relapse rates (21.1%, 25%, and 36.4%, respectively, p=0.60) or the median time interval to relapse (14, 13, and 13.5 months, respectively, p=0.90). Maintenance treatment was an independent protective factor for recurrence (p=0.001), while the complication of diabetes was an independent risk factor for recurrence (p=0.03). Fertility rates (31%, 18.2%, and 62.5%, respectively, p=0.12) and the time interval to pregnancy (14, 13, and 8 months, respectively, p=0.67) were not significantly different among the three groups. Assisted reproductive technology was positively associated with a higher pregnancy rate (p=0.02) and a body mass index ≥25 kg/m2 was negatively associated with the pregnancy rate (p=0.047).
Conclusions Longer treatment duration was associated with higher rates of complete response. Longer treatment duration (>9 months) was not associated with a decrease in success rates of pregnancy.
- endometrial cancer
- atypical endometrial hyperplasia
- fertility-preserving
- progestin
- treatment duration
Statistics from Altmetric.com
HIGHLIGHTS
A longer treatment duration may increase rates of complete response in patients with endometrial hyperplasia or cancer
A longer treatment duration does not increase the risk of recurrence or decrease pregnancy rates
The treatment period should be prolonged to >9 months in the setting of conservative management of endometrial hyperplasia or cancer
Introduction
Endometrial cancer is one of the most common gynecological malignancies. Young patients with endometrial cancer are confronted with the issue of tumor treatment and sparing fertility. As many as 54% of pre-menopausal patients with endometrial cancer have not had children.1 Continuous oral progestin is the preferred therapy for fertility-sparing treatment, and a large number of studies have proved the efficacy of this approach.2–4 The median treatment duration is 5–9.5 months, but the duration can sometimes be even longer than 12 months.4–7 Although in recent years published studies have proved the effectiveness of the levonorgestrel intrauterine system and gonadotropin-releasing hormone agonist,8 9 oral progestin has been in use for the longest time and has been the most widely reported. However, at present, there is no consensus on the duration of conservative treatment and whether it is safe to prolong progestin-based therapy considering disease progression or relapse. On the other hand, whether long-term progestin use can compromise pregnancy outcomes is still unknown. This retrospective study aimed to determine whether a longer duration of treatment (1) influences the oncologic results by causing disease progression or recurrence; and (2) is beneficial to the pregnancy rate.
We retrospectively reviewed patients with endometrial cancer or atypical endometrial hyperplasia who underwent fertility-sparing treatment from January 2012 to December 2016 in 10 medical centers in China. The patients were followed up to February 1, 2018. The clinicopathological findings of the patients were retrieved by reviewing their medical records. Each center enrolled patients according to the unified protocol. The inclusion criteria were the following: (1) age no more than 40 years; (2) a strong desire to preserve fertility; (3) International Federation of Gynecology and Obstetrics 2009 IA grade 1 with the lesion confined in the endometrium; (4) positive pathology for the expression of progestin receptors and estrogen receptors; (5) no evidence of extrauterine or distant metastasis; and (6) no contraindication of conservative treatment. The protocol for this study was approved by the Independent Ethics Committee of Peking University People’s Hospital (approval number 2015BAI13B06-E). Informed consent was obtained from the patients.
The patients were treated with a progestin-based therapy, including medroxyprogesterone, megestrol acetate, gonadotropin-releasing hormone agonist, or the levonorgestrel intrauterine system. Doctors suggested a combined therapy with gonadotropin-releasing hormone agonist if patients had a poor response to oral progestin alone. Endometrial sampling was performed by hysteroscopy at the initial diagnosis and every 3 months during treatment. The histologies were reviewed according to the World Health Organization system by two pathologists specializing in gynecologic oncology.9 After a pathological complete response, the patients were monitored for recurrence every 3–6 months. If the patient desired fertility, she was recommended to become pregnant as soon as possible or to use assisted reproductive technology. If the patient had no current desire for fertility, she was recommended for maintenance treatment, including low-dose progestin, the levonorgestrel intrauterine system, or oral contraceptives. We defined the treatment duration as the time from the beginning of the treatment to the achievement of a complete response, not including the time of maintenance treatment.
The continuous data of clinicopathological variables was expressed using the mean±SD or the median value (first quartile, fourth quartile). Intergroup differences were assessed using the t-test or one-way ANOVA for normally distributed variables and the Kruskal–Wallis test for the median time interval to relapse and pregnancy. Categorical variables were assessed using the χ2 test or Fisher’s exact χ2 test. Univariate and multivariate analyses were performed with logistic regression to evaluate the relationship between clinicopathological parameters and relapse or pregnancy. All statistical tests were two-sided at the 5% level of significance and were performed using IBM Statistical Package for the Social Science for Windows Version 19.0.
Results
A total of 68 patients were eligible and underwent analysis. Endometrial adenocarcinoma was diagnosed in 37 women and atypical endometrial hyperplasia was found in 31 women. The average age was 31±5 years. Sixty-four patients were treated with progestin-based therapy. Among these patients, 44 were treated with medroxyprogesterone, 13 with megestrol acetate, and 7 with medroxyprogesterone/megestrol acetate combined with gonadotropin-releasing hormone agonist. Three patients were treated with gonadotropin-releasing hormone agonist alone and one received the levonorgestrel intrauterine system alone. Patients were given medroxyprogesterone at 250–500 mg/day or megestrol acetate at 160–320 mg/day. The three cases treated with gonadotropin-releasing hormone agonist and the one case treated with the levonorgestrel intrauterine system are shown in online supplementary table 1.
Supplemental material
One patient underwent a hysterectomy and thus stopped treatment due to the degeneration of the leiomyoma; among the remaining 67 patients the time to complete response was 5 months (range 3–9). The complete response rate was 95.5% (64/67 patients). Apart from two patients for whom the exact treatment time was lost during the follow-up period, all the other 62 patients who achieved a complete response had available records. The median follow-up time for these 62 patients was 37.5 months (range 12–60). Details of the pathological responses are shown in online supplementary table 2.
A Kaplan–Meier analysis of the treatment duration is shown in Figure 1A. Among the patients who achieved a complete response, 56 (90%) reached complete response within 12 months but 6 (10%) required longer than 12 months. We further observed that, with the prolongation of treatment duration from 6 months to 9 months to >9 months, the accumulative complete response rates for the 67 patients were 58% (39 cases), 76% (51 cases), and 95.5% (64 cases), respectively. The distribution of the patients who achieved a complete response with different treatment durations is shown in Figure 1B.
We found that the treatment duration required by the 62 patients who achieved a complete response was different: 39 patients had treatment for <6 months, 12 patients had treatment lasting 6–9 months, and 11 patients were treated for >9 months. According to the treatment duration, we grouped these 62 patients into groups A (≤6 months), B (6–9 months), and C (>9 months). The baseline and prognostic characteristics are shown in Table 1.
There were no significant differences in the baseline information among the three groups regarding age, menstruation, history of pregnancy, body mass index, or complications of polycystic ovary syndrome, diabetes, or thyroid diseases. The follow-up times were also comparable (p=0.66). However, group C had more patients treated with combined therapy of gonadotropin-releasing hormone agonist than did groups A or B (45.5% vs 2.8% and 9.1%, p=0.003).
We analyzed the relapse results of groups A, B, and C (Table 1). The relapse rates were 21.1%, 25%, and 36.4%, respectively (p=0.60). Furthermore, no significant difference in the median time interval from complete response to relapse was found among the three groups (14, 13, and 13.5 months, respectively, p=0.90). The multivariate analysis of factors related to relapse among the 62 patients (Table 2) showed that maintenance therapy was an independent protective factor for relapse with an OR of 0.025 (p=0.001). Diabetes was an independent risk factor for relapse with an OR of 18.256 (p=0.03). There was no relationship between the body mass index, pathology, treatment duration, and relapse (Table 2).
Forty-eight patients had plans for future fertility after complete response (29 patients in group A, 11 patients in group B, and 8 patients in group C). A χ2 test of the pregnancy rates of the three groups showed that the patients in group C appeared to have a higher pregnancy rate (62.5%) than did the patients in either group A (31%) or group B (18.2%) (p=0.12) (Table 1). The time interval to pregnancy for groups A, B, and C was 14 months (range 11–26), 13 months (range 8–13), and 8 months (range 5–39.5), respectively. No significant difference in the time interval to pregnancy was found among the three groups (p=0.67) (Table 1). A logistic analysis of the factors related to pregnancy is shown in Table 3. A univariate analysis showed that treatment with assisted reproductive technology (p=0.03) combined with gonadotropin-releasing hormone agonist (p=0.19) appeared to be positively associated with pregnancy, while a body mass index ≥25 kg/m2 (p=0.11) and treatment time >9 months (p=0.07) appeared to have negative influences. Further multivariate analysis showed that only assisted reproductive technology was positively correlated with a higher pregnancy rate (OR 5.890; p=0.02), and only a body mass index ≥25 kg/m2 was negatively correlated with the pregnancy rate (OR 0.192; p=0.047). However, no relationship was found between the treatment duration or the combined use of gonadotropin-releasing hormone agonist and pregnancy.
Discussion
In this study the median time to a complete response was 5 months (range 3–9) and the complete response rate was 95.5%. The retrospective studies by Park et al,4 Baek et al,5 Chen et al,6 and Simpson et al7 concluded that the median time interval to a complete response for conservative treatment was 3–6 months (range 1–24), which coincides with our results. In the study by Park et al,4 77.7% (115/148 cases) of patients achieved a complete response within 13 months, which is in accordance with our data showing that 76% (51/67) of patients achieved a complete response within 12 months. However, as many as 95.5% of patients achieved a complete response after >9 months of treatment. Approximately 10% (6 cases) of patients may require >12 months of treatment to achieve a late complete response.
The optimal duration of treatment with progestin has not been determined to date. A number of reports have provided information on patients with late remission (>6–9 months); this information is summarized in Table 4.6 7 10–13 Erkanli’s10 meta-analysis showed that 47% of patients with endometrial cancer had a treatment duration of <6 months, while 17% had 7–9 months and 13% had >9 months of treatment. The retrospective study by Chen et al6 showed that five (13%) patients achieved a complete response after a treatment period of >12 months. In addition, we may consider longer therapy for obese patients who may be resistant to progestin therapy. The study by Penner et al11 found that patients with body mass index ≥35 kg/m2 showed an increase in the complete response rate from 33% with a treatment duration of 12 months to 66% with 18 months of treatment. However, some studies suggested that a longer treatment duration did not lead to a higher complete response rate.7 13 Simpson et al indicated that 92% of patients achieved a complete response within 12 months; otherwise, the adjustment of dosage or a hysterectomy should be considered.7 A meta-analysis by Koskas et al13 showed that the complete response probabilities after 3, 6, 12, 18, and 24 months of treatment were 30.4%, 72.4%, 78.0%, 80%, and 81.4%, respectively, and suggested that the complete response rate reached a plateau after 12 months.
Based on our study, the complete response rate improved with the extension of treatment. Moreover, some patients may require longer than 12 months of therapy. Our treatment extension was as long as 18 months; however, we should certainly communicate with patients to raise their awareness and carefully monitor the potential progression of diseases. Our results showed that there was no significant difference among groups A, B, and C in the recurrence rates (p=0.601) and the median time intervals to relapse (p=0.903). On logistic analysis, the patients with diabetes (OR 18.256; p=0.027) were more likely to experience recurrence and those undergoing maintenance therapy (OR 0.025; p=0.001) were less likely to relapse. However, no relationship was found between treatment duration and recurrence. This finding reminds us that longer progestin therapy may be due to complex patient factors such as obesity, diabetes, and insulin resistance. The relapse of disease might be related to the above risk factors and have no relationship to treatment duration.
Apart from our study, few papers have discussed the influence of conservative treatment duration on recurrence rate. The retrospective study by Park et al showed that the median treatment duration was 18 weeks and that treatment duration had no relationship with the recurrence rate (37.5% vs 21.6%, p=0.10).4 In addition, the study by Tamauchi et al12 showed no relationship between the median time to primary complete response (26 weeks vs 30 weeks) and relapse (OR 1.00; p=0.88). However, a small-sample retrospective study by Chen et al6 found that those with recurrence required a significantly longer time to achieve complete response (p<0.001) than those without recurrence. In our study we did not find a significant difference in the relapse rates among these three groups; however, the relatively small sample size and retrospective nature of our study are limitations.
Few studies have analyzed whether treatment duration influences the fertility outcome. Only the study by Chen et al reported that there was no difference in the pregnancy rate between groups with different treatment durations (60% vs 38%, p=0.226).6 The beneficial factor for the pregnancy rate was assisted reproductive technology, while a body mass index ≥25 kg/m2 predicted poor fertility results. Additionally, there was no difference in the time interval to pregnancy among the three groups (p=0.672).
In conclusion, a longer treatment duration (>9 months) might lead to a higher complete response rate and, at the same time, would not increase the risk of recurrence, decrease the pregnancy rate, or postpone the time interval to fertility for young patients with early-stage endometrial cancer or atypical endometrial hyperplasia. More evidence is still required to verify the efficiency and safety of longer treatment durations. Moreover, maintenance treatment predicts a lower risk of recurrence, while diabetes has a higher risk. Assisted reproductive technology is an independent protective factor for pregnancy, while a body mass index ≥25 kg/m2 is an independent risk factor.
Footnotes
YW and RZ are joint senior authors.
YW and RZ contributed equally.
Funding This study was supported by the National Key Technology R&D Program (No. 2015BAI13B06) and the National Natural Science Foundation of China (grant 81672571).
Competing interests None declared.
Provenance and peer review Not commissioned, externally peer reviewed.