Article Text

Download PDFPDF

Fertility preservation in rare ovarian tumors
  1. Anca Chelariu-Raicu1,
  2. Lauren P Cobb2 and
  3. David M Gershenson2
  1. 1 Department of Obstetrics and Gynecology, University Hospital, LMU Munich, Munich, Germany
  2. 2 Department of Gynecologic Oncology and Reproductive Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
  1. Correspondence to Dr David M Gershenson, Gynecologic Oncology, MD Anderson Cancer Center, Houston, TX 77230, USA; dgershen{at}


Although gynecologic cancers usually affect older women, a significant proportion of patients with rare ovarian tumors are of reproductive age. In a young patient who presents with a pelvic mass, a primary consideration should be the probability of a malignancy. If there is any suspicion of a cancer diagnosis, the patient should be referred to a gynecologic oncologist. Key factors in clinical management include assessment of preoperative studies (physical examination, tumor markers, and imaging) to determine the likelihood of a malignancy, appropriate preoperative counseling (including discussion of fertility preservation), choice of surgical approach (minimally invasive vs open), frozen section examination by a gynecologic pathologist, and intraoperative decision making. Fortunately, the clinical features of several rare ovarian tumors are compatible with fertility preservation. These characteristics include a high proportion of stage I disease and unilateral ovarian involvement for most rare histotypes. Once a final diagnosis of a rare ovarian tumor is determined, further clinical management may include the need for further studies, possible referral to a fertility expert, consideration of further surgery (if the initial surgery was incomplete), and recommendations for postoperative therapy. This article reviews the literature on fertility preservation in the context of the treatment of several rare ovarian tumor subtypes, including malignant germ cell tumors, sex cord-stromal tumors, borderline tumors, low grade serous carcinoma, clear cell carcinoma, mucinous carcinoma, and small cell carcinoma of the hypercalcemic type.

  • ovarian cancer
  • surgery

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Our knowledge base on fertility preservation for patients with gynecologic malignancies has expanded significantly over the past two decades. Although gynecologic cancers usually affect older women, a significant proportion of patients with rare ovarian tumors are of reproductive age. There are several key factors in the approach to a young patient who presents with a pelvic mass. A primary consideration, based on diagnostic studies, is the probability of a malignancy. If there is any suspicion of a cancer diagnosis, the patient should be referred to a gynecologic oncologist. All too frequently, obstetricians and gynecologists operate on a child or young woman only to be surprised by a final pathologic diagnosis of cancer when, in retrospect, there were indicators predicting this outcome. Out of these missteps arise several potential adverse outcomes. For instance, an error in judgment may result in the patient requiring a second surgery to resect metastatic tumor or to adequately stage the patient to determine recommendations for postoperative management. In addition, with the increased practice of minimally invasive surgery, the operative procedure may result in rupture of an ovarian malignancy, upstaging the tumor, which could have been prevented had an open approach been selected based on the size of the ovarian mass or the results of preoperative diagnostic studies.

For any child or woman of childbearing age who presents with an ovarian mass, especially bilateral ovarian masses, preoperative counseling regarding the operative procedure, possible complications, and the possibility or lack thereof of fertility preservation is essential. Whenever possible, a physician must discuss standard of care, either preoperatively or postoperatively, for the patient’s condition. Any deviation from the standard of care with the objective of preserving fertility must be discussed in the context of associated risk. In some instances, preoperative referral to a reproductive endocrine infertility expert for fertility preservation counseling is appropriate. In others, postoperative referral should occur. In an ideal situation, fertility potential can be preserved without compromising cure. Common options for fertility preservation include controlled ovarian hyperstimulation, followed by either embryo cryopreservation or oocyte cryopreservation, and ovarian tissue cryopreservation.

In addition to the patient’s primary attending physician and the fertility preservation expert or team, the other important individual in optimizing the potential for fertility preservation is the gynecologic pathologist. Errors in frozen section diagnosis or final diagnosis may result in inappropriate treatment recommendations, intraoperatively or postoperatively. There are many cases where even the most expert gynecologic pathologist cannot render a precise frozen section diagnosis, distinguishing benign from malignant. In such instances, conservative management should generally be employed with the understanding that subsequent surgery may be necessary if the final diagnosis is cancer. In certain circumstances, a second opinion on the pathologic diagnosis should be sought. Whatever the circumstances that lead to a final diagnosis of an ovarian malignancy, the oncologist must have a complete understanding of the biologic behavior of each subtype in order to optimize recommendations for subsequent clinical management, including surveillance and, in consultation with a reproductive medicine expert, to determine how best to integrate fertility preservation strategies, when appropriate.

This article will address the topic of rare ovarian tumors, with emphasis on the role of fertility sparing surgery for several specific subtypes. In addition, a brief discussion of the role of assisted reproductive technology options and the potential effects of adjuvant postoperative therapy, where applicable, will be included.

Malignant Ovarian Germ Cell Tumors

One of the greatest success stories in the chronicles of gynecologic oncology is that of malignant ovarian germ cell tumors, which comprise approximately 5% of all ovarian malignancies. From early case series providing details of operative findings, contemporary surgical principles evolved.1–10 From these reports, we learned that ovarian germ cell tumors occur primarily in girls and young women, tend to be quite large, and approximately 60% are confined to the ovary. The majority are unilateral, with bilateral ovarian involvement occurring in 10–15% of pure dysgerminomas, in some mixed germ cell tumors in which the dysgerminoma component predominates, and in occasional cases of advanced stage disease. In addition, mature cystic teratomas may occur in 5–10% of ovarian germ cell tumors, underscoring the fact that bilateral ovarian masses do not automatically indicate a malignant process.

These host factors and tumor characteristics converge to form an ideal opportunity for fertility sparing surgery. Evidence to support such an approach initially emerged not from randomized clinical trials but rather from observational studies. Kurman and Norris observed no worse prognosis associated with fertility sparing surgery compared with bilateral salpingo-oophorectomy in 182 patients whose tumor was grossly confined to the ovary.5 Thus except in rare circumstances, fertility sparing surgery with unilateral salpingo-oophorectomy has become the standard of care for patients with ovarian germ cell tumors, even in most patients with metastatic disease. The presence of bilateral ovarian masses usually indicates either a contralateral ovarian mature cystic teratoma or a bilateral dysgerminoma component. In the former case, ovarian cystectomy is recommended. In the case of bilateral dysgerminoma, even unilateral salpingo-oophorectomy with ovarian cystectomy has been reported with excellent outcomes. A notable exception to the opportunity for fertility sparing surgery is the presence of dysgenetic gonads, in which case bilateral adnexectomy is recommended.

Subsequent reports of ovarian germ cell tumor series demonstrated fertility sparing surgery rates in the range of 54–87%.11–14 As noted, fertility sparing surgery is frequently feasible even in patients with stage II–IV disease. Nasioudis and colleagues reported 526 patients with stage II–IV ovarian germ cell tumors from the National Cancer Database.15 The rate of uterine preservation in this cohort was 79.8%. Rates of uterine preservation were similar among histologic subtypes and across stages. Importantly, uterine preservation was not associated with decreased overall survival. In a recent report, Morrison and Nasioudis reviewed 47 studies in which 2189 patients underwent fertility sparing surgery for ovarian germ cell tumors.16 Data on the type of adnexal surgery performed was available for 1354 patients; 1219 had unilateral salpingo-oophorectomy or oophorectomy, 81 had unilateral cystectomy, 4 had bilateral cystectomy, and 50 had unilateral salpingo-oophorectomy or oophorectomy and contralateral cystectomy.

In the United States, standard postoperative therapy is chemotherapy with the combination of bleomycin, etoposide, and cisplatin for all patients with ovarian germ cell tumors, except for stage IA dysgerminoma and stage IA, grade 1 immature teratoma. In parts of Europe and for US children under the care of pediatric oncologists, observation has been recommended for all stage I ovarian germ cell tumors. In a trial of the Children’s Oncology Group, 25 patients with stage I yolk sac tumor or mixed germ cell tumor underwent observation following primary surgery.17 Twelve (48%) patients relapsed and were treated with chemotherapy, and all but one was salvaged. This experience served as background for the AGCT1531 trial, which is currently recruiting. In this trial, a low risk cohort consisting of patients with stage IA or IB ovarian germ cell tumors, with the exception of dysgerminoma, undergo observation and receive chemotherapy only if they relapse. In addition, a standard risk cohort (stages IC–III) are randomized to either four cycles of standard bleomycin, etoposide, and cisplatin, or four cycles of bleomycin, etoposide, and carboplatin.

For patients with ovarian germ cell tumors who receive adjuvant chemotherapy, several studies have reported reproductive outcomes and estimated rates of chemotherapy induced ovarian failure.18–22 The evidence suggests that based on resumption of menstrual function, chemotherapy effects on the ovary are no greater than 10–15%.20 However, we now understand the anti-Mullerian hormone seems to be the most sensitive measure of ovarian reserve in cancer survivors, at times combined with antral follicle counts, luteinizing hormone, or follicle stimulating hormone.21 In the report by Morrison and Nasioudis, 28 of 796 patients (3.5%) who underwent adjuvant chemotherapy had premature ovarian failure.16 Of 710 patients from 19 studies who received chemotherapy, 239 (33.7%) had achieved at least one pregnancy. Based on 594 patients from 13 studies, 152 (25.6%) patients had at least one live birth.

For those young patients with ovarian germ cell tumors who do undergo chemotherapy following fertility sparing surgery, the question arises as to whether the use of gonadotropin releasing hormone analogs can achieve ovarian protection from chemotherapeutic damage. In women undergoing chemotherapy for breast cancer, the findings have been conflicting. However, randomized trials that used quantitative measures of ovarian reserve, such as anti-Mullerian hormone with or without antral follicle counts, demonstrated no benefit from gonadotropin releasing hormone analogs.23 24

Ovarian Sex Cord-Stromal Tumors

Ovarian sex cord-stromal tumors account for less than 5% of all ovarian malignancies. Approximately 90% are stage I, and 95% are confined to one ovary. Thus fertility sparing surgery for ovarian sex cord-stromal tumors is also feasible in young patients. Granulosa cell tumors may occur at any age, although they are most commonly observed in postmenopausal women. In addition, approximately 25–30% are associated with endometrial pathology—hyperplasia or cancer.25 Sertoli–Leydig cell tumors also may occur at any age, but the majority are diagnosed in women in their twenties or thirties. Given the age range for these subtypes, particularly granulosa cell tumors, fertility sparing surgery is not nearly as common as for ovarian germ cell tumors.

In a French National Network study of 460 patients with granulosa cell tumors who had adequate data, the authors evaluated the surgical procedure in two periods: before 2012 and after 2012.26 Before 2012, the surgical approach was minimally invasive in 43.7% and open in 55.5%, whereas after 2012, the figures were 48.9% and 47.8%, respectively. Fertility sparing surgery was practiced in 37.3% before 2012 and in 26.9% after 2012. In other reports from China, Sweden, and Italy, the rate of fertility sparing surgery in stage I granulosa cell tumors was 54%, 76%, and 33%, respectively.14 27 28 Of 229 patients in the Italian study, 32.6% underwent fertility sparing surgery and 67.4% underwent radical surgery.28 Of those who underwent fertility sparing surgery, 62.8% underwent unilateral salpingo-oophorectomy, 16.7% underwent ovarian cystectomy, and 20.5% underwent cystectomy followed by unilateral salpingo-oophorectomy. The median disease free survival was significantly worse in the fertility sparing surgery group compared with the radical surgery group (50% vs 74%). However, there was no significant difference between the unilateral salpingo-oophorectomy versus the radical surgery group (70% vs 75%); there were significant differences in disease free survival between the unilateral salpingo-oophorectomy group compared with the cystectomy group, and the unilateral salpingo-oophorectomy versus cystectomy followed by unilateral salpingo-oophorectomy groups (70% vs 16% and 70% vs 41%, respectively). Of the 113 patients in the Chinese report, 54% underwent fertility sparing surgery and 46% underwent radical surgery.27 There was no difference in disease free survival between the two groups. In addition, of 22 patients who attempted to conceive following fertility sparing surgery, 29 (86.4%) succeeded in achieving 20 singleton pregnancies. Of 19 patients in the Swedish study, three (16%) had given birth at the time of the report.14

Studies of the much less common Sertoli–Leydig cell tumors are fewer. Sertoli–Leydig cell tumors are classified as well differentiated, with intermediate differentiation, or poorly differentiated. In addition, tumors may include heterologous elements or a retiform pattern, which are associated with an increased risk of relapse. In general, patients with stage I well differentiated tumors or intermediate differentiation are treated with surgery alone. Patients with stage I poorly differentiated tumors or stage II–IV tumors are generally treated with adjuvant chemotherapy. The presence of heterologous elements or a retiform pattern may also prompt a recommendation for chemotherapy. Thus just as in ovarian germ cell tumors, adjuvant chemotherapy may compromise fertility in some patients. Johansen et al reported nine patients with Sertoli–Leydig cell tumors, seven of whom underwent fertility sparing surgery.14 Two patients gave birth following fertility sparing surgery. In a study from the Italian MITO group, Sigismondi et al reported on 21 patients with Sertoli–Leydig cell tumors treated in 11 MITO centers.29 Median age was 37 years (range 16–76). Eighty-one per cent had stage I tumors. Eleven (53%) women underwent fertility sparing surgery, and 47% had hysterectomy and bilateral salpingo-oophorectomy. Five patients received adjuvant chemotherapy.

Borderline Ovarian Tumors

Borderline ovarian tumors account for approximately 10–15% of epithelial tumors. Histological types include serous, mucinous, endometrioid, clear cell, and Brenner (or transitional cell). The most common type is serous followed by mucinous, the two types discussed here. The primary surgical approach may be either minimally invasive or open, depending on several factors, not the least of which is tumor size. Since the average age of patients with borderline ovarian tumors is younger than those with invasive ovarian cancers, fertility sparing surgery is an important consideration in this population.

Approximately 60–70% of serous borderline tumors are stage I, and 50% are bilateral. Approximately 90% of mucinous borderline tumors are stage I, and 80–90% are confined to one ovary at diagnosis. Options for fertility sparing surgery in the surgical management of borderline ovarian tumors, depending on surgical findings, include unilateral salpingo-oophorectomy, unilateral salpingo-oophorectomy with ovarian cystectomy, unilateral ovarian cystectomy, or bilateral ovarian cystectomy.

Surgical Approach: Laparoscopy Versus Laparotomy (Open Approach)

Over the past two decades, minimally invasive surgery has assumed a more prominent role in the surgical approach to women with adnexal masses. Table 1 shows details of studies of laparoscopy versus laparotomy, as well as comparison of outcomes in women with borderline ovarian tumors.30–39 Several studies have reported no difference in relapse rates, disease free survival, progression free survival, or overall survival between laparoscopic and open approaches.30–36 38 39 Vancraeynest et al reported a higher risk of recurrence after fertility sparing surgery compared with radical surgery; however, no effect on overall survival was observed.37 Two reports observed a higher rate of tumor rupture associated with laparoscopy compared with laparotomy.30 32

Table 1

Comparison of patients with borderline ovarian tumors who underwent laparoscopy versus laparotomy

Park et al reported 360 patients who underwent surgical management for serous or mucinous borderline tumors.33 Eighty per cent underwent laparotomy and 20% underwent laparoscopy. They found no differences in relapse rates or disease free survival based on the surgical approach. Du Bois and colleagues described a cohort study of 1236 patients with stage I–III serous or mucinous borderline tumors who underwent surgery.35 Laparotomy was performed in 62% and laparoscopy in 38%, and they found no difference in relapse rates or overall survival based on surgical approach. Song and associates reported on 687 patients from the Korean registry who underwent surgery for serous or mucinous borderline tumors.38 Approximately 55% underwent laparotomy and 45% underwent laparoscopy; there was no difference in relapse free survival between the two groups.

Fertility Sparing Surgery Versus Complete or Radical Surgery

There are also several reports of fertility sparing surgery compared with complete or radical surgery (terms used as a surrogate for bilateral salpingo-oophorectomy with or without other procedures) for borderline ovarian tumors.33 35 37 39–46 For serous borderline tumors, studies reveal fertility sparing surgery rates of 12–66% and radical surgery rates of 34–88%.33 37 40–44 For mucinous borderline tumors, reports indicate fertility sparing surgery rates of 26–60% and radical surgery rates of 40–74%.33 41 42 44–46 For studies that include both serous and mucinous borderline tumors, fertility sparing surgery rates are 26–35% and radical surgery rates 65–75%.35 39

For serous borderline tumors, risk factors for relapse include: (1) histologic factors—micropapillary pattern within the ovary (non-invasive low grade serous carcinoma), microinvasion in the ovary, and presence of peritoneal implants; and (2) clinical factors—age at diagnosis, baseline serum CA125, conservative surgery, International Federation of Gynecology and Obstetrics (FIGO) stage, and residual disease. In addition, several studies showed that the type of fertility sparing surgery influences the risk of relapse, with a higher rate after ovarian cystectomy than following unilateral salpingo-oophorectomy. The majority of recurrences of serous borderline tumors occur in the ipsilateral or contralateral ovary following fertility sparing surgery, but a small number of cases can be associated with peritoneal implants. Rarely, relapses may be in the form of invasive low grade serous carcinoma and, even more rarely, in the form of high grade serous carcinoma.

Barnhill et al reported 146 patients with stage I serous borderline tumors, of whom 23 (16%) underwent fertility sparing surgery; no patients had recurrence of disease.40 On the other hand, Uzan et al described 119 patients with stage I serous borderline tumors who underwent fertility sparing surgery, and 38 (32%) relapsed.36 All but one recurrence involved an ovary, and in 10 cases there were associated non-invasive peritoneal implants. In a study of 132 patients with serous borderline tumors, Vancraeynest et al reported that 42 (32%) patients underwent fertility sparing surgery and 90 (68%) underwent radical surgery.37 Although the risk of relapse was higher in the former group, no difference in overall survival was observed.

For mucinous borderline tumors, risk factors for relapse include: (1) histologic factors—microinvasion and intraepithelial carcinoma; and (2) clinical factors—age at diagnosis, conservative surgery, and FIGO stage (although most experts believe all mucinous borderline tumors are stage I and only seromucinous tumors may be stage II or III). Practically speaking, essentially all recurrences of mucinous borderline tumors occur in the ipsilateral or contralateral ovary following fertility sparing surgery.

Khunamornpong et al reported 171 patients with mucinous borderline tumors, 60 (35%) of whom underwent fertility sparing surgery and 111 (65%) had bilateral salpingo-oophorectomy.45 Six of 144 patients (4%) relapsed, with no difference between the fertility sparing surgery and radical surgery groups. In the study by De Decker et al, 2/74 (3%) patients with mucinous borderline tumors relapsed.46

Relapse rates of 5–16% have occurred in studies comparing fertility sparing surgery with radical surgery in cohorts of both serous and mucinous borderline tumors.33 35 40 41 44 Two studies have reported a significantly higher risk of relapse for fertility sparing surgery versus radical surgery,35 44 while two studies did not show a difference in relapse rates based on type of surgery.33 41 Details of studies comparing fertility sparing surgery and radical surgery are shown in Table 2. 33 35 37 39–46

Table 2

Comparison of patients with borderline ovarian tumors who underwent fertility sparing surgery versus radical surgery (complete resection)

Reproductive Outcomes Following Fertility Sparing Surgery for Borderline Tumors

Reproductive outcomes have been reported in several cohorts of women following fertility sparing surgery for borderline tumors.32–34 36 39 42 47–50 In one of the earliest reports, Morris et al described 43 patients who underwent fertility sparing surgery for borderline tumors.47 Twenty-four women attempted pregnancy, and 12 conceived 25 pregnancies (16 live births, 4 spontaneous abortions, 3 ectopic pregnancies, and 2 ongoing pregnancies). In a cohort of 184 patients who underwent fertility sparing surgery, 130 were contacted for a telephone survey.33 Thirty-one patients attempted to conceive and 27 patients became pregnant. Three of these patients had received platinum based chemotherapy and all three had term pregnancies. Song et al reported 155 patients who underwent fertility sparing surgery, of whom 116 were contacted for a telephone interview.42 Fifty-one attempted to conceive and 45 patients had 48 singleton pregnancies, 3 twin pregnancies, 3 miscarriages, and 4 ongoing pregnancies, for a total of 54 full term infants. Following fertility sparing surgery in 119 patients with borderline tumors, 50 pregnancies were observed in 33 patients.36 Seven infertile patients achieved pregnancy following ovulation stimulation or an in vitro fertilization procedure.

The use of assisted reproductive technology has also been reported in patients with borderline tumors. Fain-Kahn et al reported 17 patients with borderline tumors on frozen section examination were planned for ovarian tissue cryopreservation.51 Fifteen of these patients had previously been treated conservatively for borderline tumors. At the time of surgery, ovarian tissue cryopreservation was aborted for various reasons, and nine patients underwent successful cryopreservation. Fortin et al described 30 patients treated for borderline tumors of whom ovarian induction was performed in 25 for infertility after conservative surgery for borderline tumors.52 Five patients with a single ovary underwent ovarian induction before surgery for recurrent disease. Three of the 30 patients were treated with clomiphene and 27 patients underwent an in vitro fertilization procedure with hyperstimulation. Thirteen (43%) patients became pregnant. The mean interval between the assisted reproductive technology and pregnancy was 28 months. Three of the five patients treated with ovarian induction before surgery for recurrent disease had three normal pregnancies. Among 10 pregnancies in patients treated for infertility, outcomes included a single miscarriage, one premature delivery at 24 weeks, and eight normal pregnancies.

Low Grade Serous Carcinoma

Low grade serous carcinoma accounts for approximately 10% of all serous ovarian carcinomas. It is characterized by younger age at diagnosis, relative insensitivity to chemotherapy, responsiveness to endocrine therapy, and prolonged overall survival compared with high grade serous ovarian carcinoma.53–57 Similar to high grade serous carcinoma, most patients present with advanced stage disease, and over 70% relapse. In addition, low grade serous carcinoma has a high frequency of estrogen and progesterone receptor expression, and the mitogen activated protein kinase (MAPK) signaling pathway plays a prominent role in its pathogenesis, with frequent KRAS mutations as well as BRAF and NRAS mutations.55 58–65

The frequency of bilaterality is 74–85%.53 57 66–70 In addition, only 2–27% of cases are stage I.57 67–70 Thus the opportunity for fertility sparing surgery in low grade serous carcinoma is somewhat limited. In addition, for those patients who undergo fertility sparing surgery but require adjuvant therapy (generally stages IC–IV), treatment with an aromatase inhibitor, either as primary or maintenance therapy, will require the addition of a gonadotropin releasing hormone analog to suppress ovarian function.

Although we understand that fertility sparing surgery is feasible in a small proportion of women with low grade serous carcinoma, mostly stage I but also in rare patients with stages II–IV, there is almost no information in the literature regarding reproductive outcomes. Yin and colleagues reported one patient with stage IC low grade serous carcinoma who underwent primary surgery following six cycles of paclitaxel/carboplatin.71 She delivered by cesarean section at 37 weeks.

Clear Cell Carcinoma

Ovarian clear cell carcinoma is regarded as an aggressive histological subtype of cancer; women who develop such tumors are considered high risk patients. In accordance with this classification, clear cell carcinoma proved to have the poorest prognosis among epithelial ovarian cancer histological subtypes.72 Furthermore, the response rate to platinum based chemotherapy in patients with ovarian cancer was significantly lower in clear cell carcinoma compared with serous carcinoma.73 Currently, both the National Comprehensive Cancer Network (NCCN) and the European Society of Gynecologic Oncology (ESGO) guidelines do not recommend fertility sparing surgery as an option for the management of clear cell carcinoma patients with childbearing potential.74 75 Therefore, the available data on clear cell carcinoma are limited due to a small proportion of patients who were included in retrospective studies along with mucinous, endometrioid, and serous subtypes76 (Table 3).77–83

Table 3

Studies evaluating fertility sparing surgery in early stage (International Federation of Gynecology and Obstetrics stages IA–IC) mucinous and clear cell carcinomas

Kajiyama et al were among the first to retrospectively compare the prognosis between patients with or without stage IA and IC clear cell carcinoma who underwent fertility sparing surgery.77 Considering the limitations of the analysis, including small number of cases (n=16) and no statistical significance, the study failed to demonstrate that fertility sparing surgery is safe for stage I clear cell carcinoma. However, given the lower rate of recurrence in the stage IA clear cell carcinoma group (4.3%), the authors concluded that patients with stage IA may benefit from fertility sparing surgery.

A more recent retrospective study of Nasioudis et al included a cohort of 741 premenopausal women (age <50 years) from 18 cancer registries, diagnosed between 1988 and 2013 with a primary ovarian tumor, clear cell histology, stage IA and IC.84 In this study, the rate of uterine preservation was 14.5% and the rate of ovarian preservation was 28.1%. After controlling for disease substage (IA vs IC), neither uterine nor ovarian preservation was associated with a worse cancer specific or overall mortality. Among the uterine preservation group, the authors reported a hazard ratio (HR) of 0.75 (95% confidence interval (CI) 0.36 to 1.56) for overall mortality and a HR of 0.84 (95% CI 0.44 to 1.62) for cancer specific mortality. In addition, the results for the ovarian preservation group were HR 0.36 (95% CI 0.13 to 1.02) for cancer specific mortality and HR 0.44 (95% CI 0.17 to 1.13) for overall mortality.

Lastly, Park et al designed a case-control study and compared the oncologic outcomes of 22 premenopausal women (mean age 36.5 years) with clear cell carcinoma, stage IA and IC, who underwent fertility sparing surgery, with a group of 25 patients who had radical surgery (mean age 40.9 years).85 In all of the above mentioned studies, no difference was noted in disease free survival or overall survival between the two groups after a median follow-up of 72 months, even when stratification by disease substage was applied.

The majority of studies reporting obstetric outcome after fertility sparing surgery in patients with stage I epithelial ovarian cancer are limited due the low number of subjects who benefit from fertility sparing surgery.86 One recent study reported one pregnancy in a patient with stage IA clear cell carcinoma who had a successful vaginal delivery at 39 weeks.71 The overall consensus is that planning a pregnancy in clear cell carcinoma patients should be carefully considered, given the risk of recurrences in extraovarian locations within the first 2 years following initial surgical staging. This is especially important for patients with stage IC, where recurrences were much higher when compared with stage IA.87

Taken together, fertility sparing surgery management should be carefully discussed with clear cell carcinoma patients, based on the limited data on clinical and obstetric outcome. However, given the low recurrence rate reported by several retrospective studies in patients with stage IA, fertility sparing surgery could still be an opportunity to consider.

Mucinous Carcinoma

In 2004,Hess et al reported a case-control study in which they suggested reduced platinum sensitivity for mucinous carcinomas compared with other histological subtypes of epithelial ovarian cancer.88 In addition, 83% of mucinous carcinomas are stage I and approximately 80% are unilateral.89 The recurrence rate for stage I mucinous carcinoma is only 5.8%, and 5 year survival is 91%.90 Thus NCCN guidelines recommend observation for stages IA-B and either observation or consideration of adjuvant chemotherapy for stage IC mucinous carcinoma.74

Therefore, fertility sparing surgery represents a reasonable option for patients with stage I mucinous carcinoma. Given the good prognosis of these patients, several studies evaluating the safety of fertility sparing surgery in epithelial ovarian cancer enrolled mostly patients with this histological subtype (Table 3).77–83 In view of these observations, both NCCN and ESGO guidelines included fertility sparing surgery management for stage IA–B mucinous carcinoma and considered it an option for high risk patients with stage IC.74 75 In addition, a more recent study reporting more granularity regarding mucinous carcinoma subtypes compared fertility sparing surgery results for expansile with infiltrative subtypes.90 Even though infiltrative type tumors are considered to be more aggressive than expansile, the authors of the study reported no difference in oncologic outcomes between the two subtypes and also recommended fertility sparing surgery for early stage infiltrative type tumors.

Data on obstetric outcome were reported for five patients who underwent fertility sparing surgery for mucinous carcinoma.71 Among the patients evaluable for obstetric outcome, one patient with stage IA and four patients with stage IC received at least three cycles of an adjuvant platinum based chemotherapy. Three patients delivered vaginally at 33, 36, and 38 weeks. The other two patients had fetal anomalies at 23 and 24 weeks, which resulted in pregnancy termination. In addition, other studies, including a large cohort of patients with mucinous histology, reported good obstetric outcome, reaching 80% pregnancy rates.86 91 However, data from these studies are reported for all patients with epithelial ovarian cancer (serous, mucinous, endometrioid, and clear cell) and not specifically for patients with mucinous carcinoma.

In summary, stage I mucinous carcinoma has the best clinical outcome within epithelial ovarian cancer, and fertility sparing surgery is a safe option for young women with stages IA–B and for selected patients with stage IC with childbearing potential. Consequently, fertility sparing surgery is currently included in the international guidelines for the treatment of ovarian cancer.74 75 Women with stage IA–B mucinous carcinoma could consider initiating pregnancy planning shortly after staging surgery, given the fact that adjuvant therapy is not recommended. However, for stage IC mucinous carcinoma patients, adjuvant chemotherapy may be recommended, in which case pregnancy planning should be delayed.

Small Cell Carcinoma of the Ovary, Hypercalcemic Type

Small cell carcinoma of the ovary, hypercalcemic type, is a rare aggressive malignancy characterized by an average age of diagnosis at 24 years.92 Prognosis is poor with median overall survival of 14.9 months.93 Given the rarity of this disease, with fewer than 300 cases reported in the literature, there is no consensus on the optimal treatment strategy or consideration of fertility preservation.

In the largest clinicopathologic analysis to date, Young and colleagues reviewed the features of 150 patients with small cell carcinoma of the ovary, hypercalcemic type.92 Almost all cases documented unilateral ovarian involvement. Even so, they suggested that bilateral salpingo-oophorectomy might lead to better survival outcomes. In a subset analysis of 50 patients with stage 1A disease, 14 (33%) were alive without disease. They reported that 8/14 (57%) of stage 1A patients who underwent bilateral salpingo-oophorectomy were disease free, whereas only 5/21 (23%) who underwent unilateral salpingo-oophorectomy were disease free.

In contrast, in a Gynecologic Cancer InterGroup study of 17 patients with small cell carcinoma of the ovary, hypercalcemic type, Harrison and colleagues concluded that given the young average age at diagnosis and unilaterality in 99% of cases, unilateral salpingo-oophorectomy is reasonable.94 However, they did not specifically address the outcomes of their patients undergoing unilateral salpingo-oophorectomy. Of the 10 patients with stage I disease, six initially underwent either unilateral oophorectomy or unilateral salpingo-oophorectomy. Three of these six patients were alive without disease (although two of three of these patients received radiation as part of their treatment). One of six who did not receive radiation died of disease, and two of six were alive with disease. These findings may speak more to the role of radiation in the treatment rather than to the feasibility of fertility preservation.

In a retrospective study by Callegaro-Filho et al, the clinical characteristics, treatment modalities, and outcomes of 47 patients with small cell carcinoma of the ovary, hypercalcemic type, were reviewed.93 Consistent with other case series, the tumor was unilateral in the majority of cases (96.3%). All patients underwent cytoreductive surgery, with 26 patients (55%) undergoing unilateral salpingo-oophorectomy and 21 patients (44.7%) undergoing a total abdominal hysterectomy with bilateral salpingo-oophorectomy as part of their cytoreductive effort. Most patients received chemotherapy, with only four patients receiving chemotherapy and radiation. Thirty-five patients (74.5%) developed recurrent disease, with the most common sites of recurrence including the pelvis, retroperitoneal lymph nodes, and contralateral ovary. In a subset analysis of patients with stage I disease (n=16), eight (50%) underwent fertility sparing unilateral salpingo-oophorectomy. Of those eight patients, only two (25%) were alive without evidence of disease (median follow-up of 13.2 months) with no reported pregnancies.

There are additional small case series and case reports that have reported good outcomes in patients who received fertility preserving treatment.95–98 In a review of four patients (two stage 1A, one stage 2B, and one stage IIIC) who underwent cytoreductive or staging surgery (including unilateral salpingo-oophorectomy in three patients) followed by chemotherapy, only the patient with stage IIIC disease relapsed and died.96 The authors concluded that stage appears to play a significant role in outcome, and that fertility preservation may be considered in early stages. In a case report from our group, a 26-year-old woman with stage IA disease underwent unilateral salpingo-oophorectomy followed by chemotherapy with vinblastine, cisplatin, cyclophosphamide, bleomycin, doxorubicin, and etoposide; 6 years later she conceived naturally.98 This is the first reported case of pregnancy after treatment for small cell carcinoma of the ovary, hypercalcemic type. Because fertility preservation in this highly aggressive malignancy remains controversial, further experience is necessary to determine if this is a feasible approach.


In summary, fertility sparing surgery is feasible in several rare ovarian tumors, including all stages of malignant ovarian germ cell tumors, all stages of borderline ovarian tumors, stage I ovarian sex cord-stromal tumors, and stage I invasive epithelial ovarian cancer. Table 4 presents the international guidelines regarding fertility sparing surgery for rare ovarian tumors by the NCCN and ESMO-ESGO.74 99

Table 4

International guidelines for fertility sparing surgery in patients with rare ovarian tumors



  • Contributors All authors contributed to writing and review of final version.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.