Possibility of women treated with fertility-sparing surgery for non-epithelial ovarian tumors to safely and successfully become pregnant---a Chinese retrospective cohort study among 148 cases

Abstract

This study was performed to evaluate the oncological and reproductive outcomes of childbearing-age women treated with fertility-sparing surgery (FSS) for non-epithelial ovarian tumors in China. One hundred and forty eight non-epithelial ovarian tumor women treated with FSS between January 1, 2000 and August 31, 2015 from two medical centers in China were identified. Progression-free survival (PFS) was 88.5%, whereas overall survival (OS) was 93.9%. Univariate analysis suggested that delivery after treatment is related to PFS (P=0.023), whereas histology significantly influenced OS. Cox regression analysis suggested that only histology was associated with PFS and OS (P<0.05). Among the 129 women who completed adjuvant chemotherapy (ACT), none developed amenorrhea. Among the 44 women who desired pregnancy, 35 (79.5%) successfully had 51 gestations including 35 live births without birth defects. Non-epithelial ovarian tumors can achieve fulfilling prognosis after FSS and chemotherapy. Histology might be the only independent prognostic factor for PFS and OS. FSS followed by ACT appeared to have little or no effect on fertility. Meanwhile, postoperative pregnancy did not increase the PFS or OS. Use of gonadotropin-releasing hormone agonist was not beneficial for fertility.

Keywords

malignant germ cell tumors / ovarian sex cord-stromal tumors / fertility-sparing surgery / prognosis / fertility

Introduction

Non-epithelial ovarian tumors, which mostly affect the health of adolescent, include malignant germ cell tumors (MOGCT), which is less than 5%, and ovarian sex cord-stromal tumors (OSCST) accounting for 5% to 8% [¹,²]. MOGCT usually involve the unilateral ovary, which is sensitive for chemotherapy [³,⁴]. In addition, OSCST generally grow slowly and occur at an early stage [⁵]. The 5-year overall survival (OS) of MOGCT and OSCST can reach 75%–100% and 97.2%, respectively [⁴,⁶–⁸]. The International Federation of Gynecology and Obstetrics (FIGO) and National Comprehensive Cancer Network (NCCN) suggest that fertility-sparing surgery (FSS) can be safely implemented at any stage of MOGCT or OSCST [⁵,⁹], which gradually shifts the focus of prognosis to fertility outcomes.

Many institutions and cooperative groups have provided attention to the treatment, management, and survival rate in MOGCT or OSCST, but very few researchers have reported the details of reproductive ability and outcomes [²,³,⁵,⁹]. Only few studies have suggested that the survivor fertility rate is almost comparable with those expected among age-matched women from the general population [¹,³,⁴]. In general, the ovarian function, pelvic adhesion, and tubal patency, as well as patients’ emotions and doctors’ suggestions, can affect the fertility rate. In addition, disease and surgery may affect marriage of patients. Therefore, we have conducted the present study to evaluate the prognosis of MOGCT and OSCST patients and the post-treatment fertility outcomes. At the same time, we analyzed the reasons why patients do not want to conceive despite their capability. In addition, the effect of pregnancy and the potential factors affecting the fertility outcomes have been studied among Chinese women with non-epithelial ovarian tumors after FSS.

Materials and methods

Patients

After obtaining Institutional Review Board approval of the Tongji Hospital, Huazhong University of Science and Technology in Hubei Province and Women’s Hospital, Zhejiang University in Zhejiang Province, we identified all gynecology and obstetrics patients from January 1, 2000 to August 31, 2015. In this study, the criteria were as follows: (1) women who had a final histological diagnosis of MOGCT or OSCST; (2) women who underwent FSS at these two medical centers and had a complete medical record; and (3) women≤40 years old of age who signed informed consent and availability to follow-up.

Clinical and demographic data were gathered from the patients’ medical records, including patient charts, operative and pathology reports, and chemotherapy records. Data collection included the age during diagnosis, disease stage, date and type of surgery, chemotherapy regimen, and cycle numbers. Follow-up data on the recurrence date, recurrence site, date of death, menstruation, menopause, fertility, and gestational outcomes were completed by telephone interview or multiple letters. The outpatient examinations were invoked as a reference. Informed consent was obtained from all contacted patients.

MOGCT and OSCST were classified according to the World Health Organization. The stage was determined according to FIGO (2010) staging system. They were further divided into two groups for statistical analysis as follows: early stage (FIGO I) and advanced stage (FIGO II–IV).

Treatment

FSS is defined as surgery sparing the uterus and some ovarian tissues to allow natural conception in the future. Comprehensive surgical staging is described as follows: peritoneal lavage fluid or ascite cytology, unilateral salpingo-oophorectomy, omentum removal, pelvic and/or abdominal paraaortic lymph node dissection, removal of the visible lesions, and biopsy of the suspicious sites. Adjuvant chemotherapy (ACT) was administered according to the department protocol, and the standard first line regimen was bleomycin [BEP (15 mg, days 1, 3, and 5), etoposide (100 mg/m², days 1–5), and cisplatin (20 mg/m², days 1–5)] every four weeks at the two centers. The same criteria were also applied in the two medical centers for FSS and adjuvant treatment according to NCCN guidelines for ovarian cancer. The number of cycle depended on the clinical circumstance, patient’s tolerance of chemotherapy, and objective response for any measurable index.

All patients were proposed to have follow-up examinations every three months for the first year and every six months for the next year. The follow-up cutoff date was March 2016. Recurrence was considered if new evidence of tumors (such as histological or sequential increase in serum CA-125 accompanied by imaging abnormalities at the same time) occurred after initial complete resection or complete regression of all measurable diseases with tumor markers normalized for at least one month.

Statistical analysis

Statistical Package for the Social Sciences (SPSS) 19.0 statistical software was used for statistical analysis. Progression-free survival (PFS) was calculated by the time interval in months from the date of primary surgery to the date of disease progression or recurrence. OS was calculated by the time interval in months from the date of primary surgery to the date of death or last contact. PFS and OS were estimated using Kaplan–Meier model, whereas Cox proportional hazard regression model was used for multivariate analysis. Variables with P<0.5 in univariate analysis were included in multivariate analysis. Binomial logistic regression analysis model was used to assess the independent effects of factors found to be significantly associated with fertility in univariate analysis. All reported P values were two tailed, and P<0.05 indicates statistical significance.

Results

From 2000 to 2015, 175 women diagnosed with MOGCT or OSCST were registered for FSS treatment at the two centers. In addition, 148 (84.6%) women had fulfilled the inclusion criteria. The characteristics of these 148 women with MOGCT or OSCST are shown in Table 1. Among these women, 70.3% were less than 25 years old. The median age was 22 years, ranging from eight to 37 years. The histological subtypes and clinical stages of these women were as follows: 122 (82.4%) MOGCT, 26 (17.6%) OSCST, 125 (84.5%) early stage (FIGO I), and 23 (15.5%) advanced stage (FIGO II–IV). Staging surgery was performed among 97 women (65.5%), whereas 120 (81.1%) underwent laparotomy. In addition, 129 women (87.2%) were treated with postoperative ACT.

Disease outcome

The mean follow-up time was 72 months (ranging from seven months to 179 months).The 5-year PFS was 88.5%, and 17 (11.5%) relapsed with a mean interval to recurrence of 37.8 months. From these numbers, six women died. Detailed information of these 17 women is shown in Supplementary Table S1. The 5-year OS was 93.9%, and nine women with tumor-related death were found during follow-up, with a mean time to death of 32.3 months, wherein three women abandoned treatment.

Of the possible clinical and pathological factor studies in women, the presence of different histological types (P = 0.008) is a significant independent prognostic factor related to the OS. In addition, being pregnant or delivering after treatment (P = 0.005) was correlated with the PFS (Fig. 1) in Kaplan–Meier analysis. Further analysis by Cox proportional hazard regression model found that the different histology types were significant with PFS and OS (P<0.05) (Table 2).

Menstrual function

Our study found that not a single one had amenorrhea after completing chemotherapy among the 129 women who received ACT. Basic information of these patients’ menorrhea status when they had surgery is shown in Table 1. A total of 125 women (96.9%) resumed normal cycles in less than one year. Among the 129 women in the chemotherapy group, 38 received gonadotropin-releasing hormone agonist (GnRH-a), and 91 did not receive injection (Fig. 2). The mean time to resume menstruation of the GnRH-a group was 6.97 months instead of 4.62 months in the non-GnRH-a group (P<0.05). In Supplementary Table S2, patients in the GnRH-a and non-GnRH-a groups show no difference in demographics, pathological outcomes, menstrual status, and cycles of chemotherapy used. The use of GnRH-a significantly lengthened the interval of normal menstrual cycle recovery.

Fertility outcome

FSS was performed among 148 women including 129 women who received ACT (Fig. 3). Of the 44 women who desired pregnancy, 35 (79.5%) displayed 51 successful gestations, resulting in 35 live births without birth defects. Of the remaining nine (20.5%) women who failed to conceive, two were newly married, two completed the treatment for a few months and were trying to conceive, and five had infertility problems (all had undergone chemotherapy, two had yolk sac tumors, and two had immature teratomas). With respect to the remaining 76 women who did not attempt to conceive, 49 (61.8%) lacked a partner, among which, 29 were adolescent (≤18 years old) and three were≥27 years old but remained single. Meanwhile, among the contacted 19 women who did not receive ACT, only five women had tried to conceive and ultimately delivered five healthy children. None of these women were infertile, and no evidence of birth defects or other disabilities was observed in any of the offspring.

Binomial logistic regression analysis showed that no factor displayed a significant effect on the ovarian function in women desiring pregnancy (n = 49), including use (or not) of GnRH-a (Table 3).

Discussions

MOGCT and OSCST are rare neoplasms of the upper female genital tract, but they represent 70% of malignant ovarian tumors in young women (aged 20–25 years old) and adolescent girls (aged 11–19 years old) [¹,²]. MOGCT are characterized with rapid growth, large tumor size, and high malignancy, but almost all MOGCT are hypersensitive to cisplatin-based chemotherapy regimens [¹,³,⁴,⁷]. OSCST are less chemosensitive, but they may grow slowly [⁵]. The evolution of modern chemotherapy transformed the poor prognosis of MOGCT or OSCST into highly curable cancers [¹⁰,¹¹]. Traditionally, FSS with or without ACT arises at a historic moment and has become the standard treatment. A review of the literature [¹²] reveals an equivalent cure rate after FSS compared with a nonconservative procedure.

In accordance with several previous studies, the present study shows that FSS followed by ACT is an admissible option for women with MOGCT or OSCST. Univariate analysis of prognosis indicated that with or without delivery after treatment is related to PFS (P = 0.023), whereas histology significantly influenced OS (Table 4, Fig. 1). However, with the result shown in Supplementary Table S3, with or without delivery after treatment did not affect PFS because the age of the two groups did not match. In addition, we analyzed the possible risk factors of PFS and OS using a Cox regression analysis model, suggesting that only histology was a significant and independent prognostic factor for PFS and OS (P<0.05) (Table 3).

In Supplementary Table S1, 17 patients showed recurrence including three who had stage II–IV tumor and 14 who had stage I tumor. However, in our study, only 14 patients have stage III tumor, indicating 21.43% (3/14) stage II–IV tumor relapse. In addition, only 10.69% (14/131) showed stage I tumor relapse. The rate is even lower than that in Mangili et al.’s study [¹³], which reported recurrence of stage I disease occurring among 23 patients (16.0%). Furthermore, 64.7% (11/17) recurrence occurred in the contralateral ovary, which is very different with that of the previous reports. However, four patients showed bilateral involvement during diagnosis and chose to remove one side while the other side received ovarian cystectomy. The patients’ original records were verified by the researchers at both centers and found to be true.

Additionally, FIGO and NCCN suggest that FSS can be safely applied in any MOGCT stage. By contrast, after FSS, leaving only one ovary, pelvic adhesions and ACT are likely to reduce the fertility rate among these patients. Moreover, chemotherapy may cause premature menopause and infertility [¹⁴]. BEP can cause chromosome breakage, decreasing the general egg-formation capacity in treated females [¹¹]. Numerous follicular cysts and pronounced growth of the connective tissue were found in the ovaries three and six months after etoposide treatment [¹⁵]. Cisplatin triggers cell apoptosis in ovarian stromal cells [¹⁶]. Other investigators have reported that etoposide can increase the risk of premature menopause after treatment for Hodgkin’s lymphoma [¹⁷] and MOGCT [¹⁸].

However, many studies to date consistently show that the reproductive function is barely affected after chemotherapy for MOGCT or OSCST, and that healthy pregnancies can be anticipated [⁴]. Our results about menstrual status after treatment is consistent with several studies [⁴,¹⁹], reporting that 95%–100% return to normal menstrual status after completing BEP chemotherapy. Unfortunately, retrospective studies can only indirectly evaluate the ovarian function by observing the menstruation situation instead of specific direct clinical indicators, such as the estrogen level, follicle-stimulating hormone, and antimullerian hormone. Reduction in the ovarian reserve after chemotherapy and decline in fertility may be overlooked. This disadvantage exists in the present study, and prospective studies are needed to provide more pieces of evidence in the future.

In general, many reports are available regarding the fertility outcomes after FSS among women with MOGCT and OSCST. However, few studies have provided attention to the survival condition of women without fertility intentions. In our study, because of fear of relapse, doctors discourage them from trying pregnancy. In addition, those above 25 years old but still single were proved to be the reasons of the refusal to attempt pregnancy, wherein a high percentage of 6.7% is still unmarried. Additionally, 2.2% of women were forced to divorce because of possible future infertility. The above data demonstrate that the public’s awareness and understanding are limited, causing a group of people to avoid fertility and marriage because of group prejudice. This stigma significantly affects the patient’s marriage and fertility.

With respect to 44 women who desire pregnancy, seven women (15.9%) failed to conceive, which may be comparable with those expected among age-matched women from the general population, whose infertility rate can be up to 13.9% in North China [²⁰]. Many studies to date consistently show that current reproductive function is associated with a very low level after chemotherapy for MOGCT or OSCST, and that healthy pregnancies can be anticipated. Weinberg et al. [⁴] reported on 40 women with MOGCT, wherein 22 received FSS and 16 received ACT. Among the 10 who desired pregnancy, eight (80%) had 11 successful spontaneous pregnancies, resulting in 14 live births. Meanwhile, no relapse or death occurs in women who successfully conceived after treatment. This result may support that subsequent fertility did not increase the risk of relapse or death.

In summary, both MOGCT and OSCST carry a good prognosis, even when diagnosed in the advanced stage. FSS followed by ACT has been proved to exhibit minimal effect on fertility and the menstrual cycle. Meanwhile, pregnancy or even delivery after completing chemotherapy may not affect recurrence or death. Although preliminary, our study has provided insight into the importance of emotion and social factors when young women desire to become pregnant especially in developing countries, which is beneficial for women to overcome their fear, including providing psychological counseling.

References

[1] Ertas IE, Taskin S, Goklu R, Bilgin M, Goc G, Yildirim Y, Ortac F. Long-term oncological and reproductive outcomes of fertility-sparing cytoreductive surgery in females aged 25 years and younger with malignant ovarian germ cell tumors. J Obstet Gynaecol Res 2014; 40(3): 797–805

[2] Sieh W, Sundquist K, Sundquist J, Winkleby MA, Crump C. Intrauterine factors and risk of nonepithelial ovarian cancers. Gynecol Oncol 2014; 133(2): 293–297

[3] Nishio S, Ushijima K, Fukui A, Fujiyoshi N, Kawano K, Komai K, Ota S, Fujiyoshi K, Kamura T. Fertility-preserving treatment for patients with malignant germ cell tumors of the ovary. J Obstet Gynaecol Res 2006; 32(4): 416–421

[4] Weinberg LE, Lurain JR, Singh DK, Schink JC. Survival and reproductive outcomes in women treated for malignant ovarian germ cell tumors. Gynecol Oncol 2011; 121(2): 285–289

[5] Bryk S, Farkkila A, Butzow R, Leminen A, Heikinheimo M, Anttonen M, Riska A, Unkila-Kallio L. Clinical characteristics and survival of patients with an adult-type ovarian granulosa cell tumor: a 56-year single-center experience. Int J Gynecol Cancer 2015; 25(1): 33–41

[6] Yesilyurt H, Tokmak A, Guzel AI, Simsek HS, Terzioglu SG, Erkaya S, Gungor T. Parameters for predicting granulosa cell tumor of the ovary: a single center retrospective comparative study. Asian Pac J Cancer Prev 2014; 15(19): 8447–8450

[7] Pectasides D, Pectasides E, Kassanos D. Germ cell tumors of the ovary. Cancer Treat Rev 2008; 34(5): 427–441

[8] Tomao F, Peccatori F, Del Pup L, Franchi D, Zanagnolo V, Panici PB, Colombo N. Special issues in fertility preservation for gynecologic malignancies. Crit Rev Oncol Hematol 2016; 97: 206–219

[9] Yang ZJ, Liu ZC, Wei RJ, Li L. An analysis of prognostic factors in patients with ovarian malignant germ cell tumors who are treated with fertility-preserving surgery. Gynecol Obstet Invest 2016; 81(1): 1–9

[10] .Slayton RE, Park RC, Silverberg SG, Shingleton H, Creasman WT, Blessing JA. Vincristine, dactinomycin, and cyclophosphamide in the treatment ofmalignant germ cell tumors of the ovary. A Gynecologic Oncology Group Study (a final report). Cancer 1985; 56(2):243–248

[11] Williams SD, Birch R, Einhorn LH, Irwin L, Greco FA, Loehrer PJ. Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide. N Engl J Med 1987; 316(23):1435–1440

[12] Low JJ, Ilancheran A, Ng JS. Malignant ovarian germ-cell tumours. Best Pract Res Clin Obstet Gynaecol 2012; 26(3): 347–355

[13] Mangili G, Sigismondi C, Lorusso D, Cormio G, Candiani M, Scarfone G, Mascilini F, Gadducci A, Mosconi AM, Scollo P, Cassani C, Pignata S, Ferrandina G. The role of staging and adjuvant chemotherapy in stage I Malignant ovarian germ cell tumors (MOGTs): the MITO-9 study. Ann Oncol 2017; 28(2):333–338

[14] Pedersen-Bjergaard J, Daugaard G, Hansen SW, Philip P, Larsen SO, Rørth M. Increased risk of myelodysplasia and leukaemia after etoposide, cisplatin, and bleomycin for germ-cell tumours. Lancet 1991; 338(8763):359–363

[15] Borovskaya TG, Timina EA, Perova AV, Pakhomova AV, Shchemerova YA, Gol'dberg VE. Pharmacological correction of etoposide ovariotoxicity. Bull Exp Biol Med 2007; 143(5):614–616

[16] Fabbri R, Macciocca M, Vicenti R, Paradisi R, Klinger FG, Pasquinelli G, Spisni E, Seracchioli R, Papi A. Doxorubicin and cisplatin induce apoptosis in ovarian stromal cells obtained from cryopreserved human ovarian tissue. Future Oncol 2016; 12(14):1699–1711

[17] Swerdlow AJ, Cooke R, Bates A, Cunningham D, Falk SJ, Gilson D, Hancock BW, Harris SJ, Horwich A, Hoskin PJ, Linch DC, Lister A, Lucraft HH, Radford J, Stevens AM, Syndikus I, Williams MV; England and Wales Hodgkin Lymphoma Follow-up Group. Risk of premature menopause after treatment for Hodgkin’s lymphoma. J Natl Cancer Inst 2014; 106(9): dju207

[18] Yoo SC, Kim WY, Yoon JH, Chang SJ, Chang KH, Ryu HS. Young girls with malignant ovarian germ cell tumors can undergo normal menarche and menstruation after fertility-preserving surgery and adjuvant chemotherapy. Acta Obstet Gynecol Scand 2010; 89(1): 126–130

[19] Gadducci A, Lanfredini N, Tana R. Menstrual function and childbearing potential after fertility-sparing surgery and platinum-based chemotherapy for malignant ovarian germ cell tumours. Gynecol Endocrinol 2014; 30(7): 467–471

[20] Cong J, Li P, Zheng L, Tan J. Prevalence and Risk Factors of Infertility at a Rural Site of Northern China. PLoS One 2016; 11(5): e0155563

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany