Ovarian Yolk Sac Tumor: The Experience of a Regional Cancer Center

Wang, Xiaojing MD; Ma, Zebiao MD; Li, Yanfang MD

International Journal of Gynecological Cancer:
doi: 10.1097/IGC.0000000000000704
Ovarian Cancer

Objective: The aim of this study was to evaluate the clinicopathologic characteristics of patients with ovarian yolk sac tumor and the benefit of omentectomy in patients with clinical early-stage disease.

Methods: The medical records of 66 patients with ovarian yolk sac tumor were reviewed retrospectively.

Results: There were 37, 8, 14, and 7 patients with stages I, II, III, and IV disease, respectively. Sixty-five patients received surgery and adjuvant chemotherapy, and 1 had chemotherapy only. The median follow-up was 78 months. The overall 5-year survival rate was 86.0%. Univariate analysis revealed that stage (P = 0 .022), age (P = 0.001), residual tumor (P = 0.036), and satisfactory α-fetoprotein (AFP) decline (defined as normalization of AFP after the first or second cycles of postsurgery chemotherapy, P = 0.006) were significant prognostic factors. Multivariate analysis revealed that satisfactory AFP decline was an independent significant prognostic factor for overall survival (P = 0.028). The postoperative pathology showed that only 1 (2.7%) of 37 patients who received omentectomy without gross spread had omentum metastasis microscopically. The 5-year survival rates were 89.2% and 100.0% for stage I-II patients with (36 cases) or without (9 cases) omentectomy, respectively (P > 0.05). Three of the 7 patients with recurrence were successfully salvaged and lived 38.0, 102.6, and 45.2 months after initial diagnosis.

Conclusions: Postsurgery satisfactory AFP decline was an independent significant prognostic factor for patient survival. Omentectomy might not be of therapeutic significance for clinical stage I-II patients.

Author Information

*Department of Gynecologic Oncology, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou; and †Department of Gynecologic Oncology, the Affiliated Cancer Hospital of Zhengzhou University, Henan Province Cancer Hospital, Zhengzhou, Henan, People’s Republic of China.

X.W. and Z.M. contributed equally to this work.

Address correspondence and reprint requests to Yanfang Li, MD, Department of Gynecologic Oncology, Sun Yat-sen University Cancer Center, 651 Dongfeng Rd E, Guangzhou, People’s Republic of China 510060. E-mail: liyf@sysucc.org.cn.

The authors declare no conflicts of interest.

This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially.

Received October 31, 2015

Received in revised form February 3, 2016

Accepted February 5, 2016

Article Outline

Ovarian yolk sac tumor (YST), also known as ovarian endodermal sinus tumor, is the second most frequent malignant ovarian germ cell tumor just after dysgerminoma, with an incidence of approximately 1% of ovarian malignancies.1 Serum α-fetoprotein (AFP) is a useful tumor marker for diagnosis and monitoring recurrence. Because of its high progression, the 3-year survival rate was 13% prior to the availability of chemotherapy.2 Cisplatin-based chemotherapy has dramatically improved the prognosis of ovarian YST.3 Overall, the 5-year survival and disease-free survival rates were 94% and 90%, respectively.4 Because of its high effect on women of childbearing age, a therapy to preserve reproductive function is of utmost importance. In several retrospective reviews, fertility-sparing surgery has been demonstrated to be as effective as radical surgery. Unilateral salpingo-oophorectomy and postoperative combination chemotherapy, which have been proven to be feasible and safe while the contralateral ovary and uterus are not affected, are now widely recommended in patients of reproductive age.3–7 Despite the improvement of this treatment, there are still unsolved problems in the treatment of this disease such as the necessity of omentectomy in clinical stage I-II disease and the prognostic significance of performing retroperitoneal lymphadenectomy. Given the rarity of the tumor and the few randomized trials, we reviewed the clinicopathologic characteristics, treatment outcomes, and prognosis factors of 66 patients with ovarian YST treated in our cancer center to add knowledge to this rare disease.

Back to Top | Article Outline


We performed a retrospective review of the clinical records of all patients with ovarian YST at Sun Yat-sen University Cancer Center from January 1, 1988, to June 30, 2014, to identify patients who received initial treatment at our cancer center. The patient list was obtained from the database, and patient hospital records were reviewed to obtain demographic data and treatment information. Only patients who met the following criteria were included in the study: (1) patients with tumors only of the yolk sac component and (2) patients whose initial treatment was administered at our cancer center. The pathological specimens of all patients were reviewed histologically. The study was approved by the ethics committee of Sun Yat-sen University Cancer Center.

The Kaplan-Meier method was used to calculate patient survival. Overall survival (OS) was defined from the date of the diagnosis to the time of death or last follow-up. Statistical significance was determined using the log-rank test. The prognostic effect of clinicopathologic variables in patients for OS was identified by univariate and multivariate Cox proportional hazards regression analysis. The result was considered statistically significant at P < 0.05.

Back to Top | Article Outline


Patient Characteristics

A total of 66 patients with exclusive ovarian YST were identified (Table 1). Seventeen patients (25.8%) were 14 years or younger, whereas 5 (7.6%) were 50 years or older, including 4 who were postmenopausal. The tumor site was noted in 65 of the patients. The tumors occurred on the right ovary in 64.6% of the patients, on the left ovary in 32.3%, and were bilateral in 3.1%. The tumor site in 1 stage IV patient who underwent chemotherapy only was not recorded. Nine (52.9%) of the 17 patients 14 years or younger had advanced (stage III or IV) disease, whereas 12 (27.3%) of the 44 patients of reproductive age did.

Back to Top | Article Outline
Treatment and Outcomes
Uterus-Sparing Versus Non–Uterus-Sparing Surgery and Survival

Of the 66 patients, 65 underwent surgery and postoperative adjuvant chemotherapy (Table 2). Uterus-sparing surgery and non–uterus-sparing surgery were defined as surgery in which the uterus was preserved or removed, respectively, regardless of the extent of the whole surgery. In 51 patients receiving uterus-sparing surgery, 30 had stage I disease, 6 stage II, 10 stage III, and 5 had stage IV disease. Of the 14 patients who had non–uterus-sparing surgery, 7 had stage I disease, 2 stage II, 4 stage III, and 1 had stage IV disease. The 5-year survival rates were 88.2% and 85.7% for patients with uterus-sparing surgery and non–uterus-sparing surgery, respectively (P > 0.05) (Table 3).

Back to Top | Article Outline
Lymphadenectomy and Survival

The stage distribution for the 22 patients with lymphadenectomy was as follows: 12 stage I, 4 stage II, 4 stage III, and 2 stage IV. Nine of them had a grossly enlarged lymph node. Lymph node metastasis was found in 2 patients (9.1%, 2/22). The lymph node metastasis rate for patients with clinical stage III or IV disease was 33.3% (2/6). None of the patients with clinical stage I or II disease (clinically no gross extrapelvic metastasis) in whom lymphadenectomy was performed had lymph node metastases. The 5-year survival rates for patients with and without node dissection were 86.4% and 86.0%, respectively (P > 0.05).

Back to Top | Article Outline
Omentectomy and Survival

The stage distribution for the 55 patients with omentectomy was as follows: 29 stage I, 7 stage II, 14 stage III, and 5 stage IV. Thirty-seven of them had no gross lesions on the omentum. The postoperative pathology showed that only 1 patient (2.7%, 1/37) had tumor metastasis microscopically. The 5-year survival rates were 88.9% and 100.0% for stage I or II patients with and without omentectomy, respectively (P > 0.05).

Back to Top | Article Outline
Overall Survival and Other Prognostic Factors

The overall 5-year survival rate was 86.0%. For stages I, II, III, and IV patients, the 5-year survival rates were 94.2%, 75.0%, 77.4%, and 71.4%, respectively (Table 3). Optimal surgery was defined as surgery with no macroscopic residual disease. Satisfactory postoperative AFP decline was defined as normalization of AFP after the first or second cycles of postsurgery chemotherapy. All 65 patients received postoperative adjuvant chemotherapy. EP (etoposide, cisplatin), VBP (vincristine, bleomycin, cisplatin), and EBP (etoposide, bleomycin, cisplatin) regimens were administered to 4, 27, and 31 patients, with a median of 4, 5, and 5 cycles, respectively. Univariate analysis revealed that stage (P = 0 .022, Fig. 1A), age (P = 0.001 Fig. 1B), residual tumor (P = 0.036, Fig. 1C), and satisfactory AFP decline (P = 0.006 Fig. 1D) were significant prognostic factors (Table 3).

Multivariate Cox proportional hazard regression analysis was performed to determine the independent prognostic factors. The results revealed that AFP decline was an independent significant prognostic factor for OS (P = 0.028), whereas International Federation of Gynecology and Obstetrics stage, age, and residual tumor were not.

Back to Top | Article Outline
Preoperative Neoadjuvant Chemotherapy and Treatment Outcomes

Among the 6 patients receiving neoadjuvant chemotherapy (NACT), partial response was observed in 5 patients who received a BEP or VBP regimen for 1 to 4 cycles; no response was seen in the one who received 1 cycle of 5-fluorouracil plus carboplatin. After NACT, the optimal cytoreduction rate was 83.3% (5/6) compared with that of 50.0% (7/14) in stages III and IV patients without NACT. The 5-year survival rates in patients with or without NACT were 66.7% and 85.7%, respectively (P = 0.293).

Back to Top | Article Outline
Uterus-Sparing Surgery and Fertility Outcomes

Regular menstruations returned after discontinuation of chemotherapy in 39 patients (95.1% [39/41]) who had regular menstrual cycles before treatment. The recovery time of regular menses was 2 to 12 months (median, 4 months) after the cessation of chemotherapy. Fourteen of the 17 patients who attempted conception gave birth to normal children, resulting in 20 live births. Twelve cases conceived naturally, and 2 required in vitro fertilization.

Back to Top | Article Outline
Recurrence and Salvage Therapy

The median follow-up time was 70.0 months (range, 7.0–236.0 months). The median recurrent time was 9.3 months (range, 6.9–10.7 months). The recurrent rates for stage I, II, III, and IV disease were 5.3% (2/38), 0% (0/8), 20.0% (3/15), and 25.0% (2/8), respectively. The recurrent rates for patients 14 years or younger, 14 to 50 years, and 50 years or older were 23.5% (4/17), 4.5% (2/44), and 20.0% (1/5), respectively (Table 4). Of the 7 patients with recurrence, 3 were successfully salvaged and alive at 38.0, 102.6, and 45.2 months after initial diagnosis, whereas the other 4 patients failed and eventually died of the disease.

Back to Top | Article Outline


Ovarian YSTs are a rare germ cell tumor. In this study, we provide a detailed description of the clinical features of 66 patients with ovarian YST. To our knowledge, we first evaluate the therapeutic significance of omentectomy in patients with clinically early-stage disease without grossly visible lesions in the omentum, which is frequently performed in clinical practice in the treatment of ovarian YST. This study is one of the largest series about ovarian YST in the literature.

Recently, the diagnostic (for staging purpose) and therapeutic significance of omentectomy in clinically apparent early-stage epithelial ovarian cancer has been doubted.8,9 The significance of omentectomy in clinical early-stage ovarian YST has not been demonstrated in the literature, either. Many authors performed fertility-preserving comprehensive staging surgery for patients with early-stage disease in which omentectomy was also included.1,3,6 Recently, the justification for comprehensive staging surgery in malignant germ cell tumors has been questioned; patients may not benefit more from comprehensive staging surgery than from unilateral adnexectomy.5 Omentum is an important abdominal organ that can limit peritoneal bacteria infection and prevent adhesions of intestine to anterior abdominal wall following surgery. Most patients with ovarian YST had early-stage disease and a normal omentum. Theoretically, removing a grossly normal omentum may not be of therapeutic benefit for patients with early-stage disease. One may argue that a macroscopically normal omentum may have occult metastasis, which can be detected only microscopically. However, even if occult metastasis is present in the unremoved omentum, postoperative chemotherapy is likely to control the disease because ovarian YST is highly sensitive to chemotherapy. In our study, the occult metastatic rate of omentum is 2.7%. The 5-year survival rate was not significantly different for patients with or without omentectomy (89.2% vs 100.0%), and the survival results even favored those without omentectomy. Our data suggest that occult metastasis in the omentum is not common and that removing the grossly normal omentum might not confer benefit to patients with clinically apparent stage I-II disease. Although a definite conclusion could not be obtained based on a small number of data, those were interesting and promising data that justified further investigation, especially randomized trial.

In ovarian malignant germ cell tumors, dysgerminoma and malignant teratoma tend to spread to the retroperitoneal lymph node.10 For ovarian YST, few data are available about lymph node metastasis rate and the importance of lymphadenectomy. Two studies have shown that lymph node metastasis may be uncommon in this disease and that lymphadenectomy may not affect prognosis. In 1 study, lumbar-aortic lymph node dissection was performed in 11 patients, and lymph node metastasis was found in only 2 stage IIIC patients with bulky peritoneal disease and was not present in all clinical stage I patients whose tumors were grossly confined to the ovary.3 In another study, lymph node dissection was performed in 3 patients with stage II-IV disease, but lymph node involvement was not found in all cases.7 In our study, the pathological lymph node metastatic rate was 9.1% (2/22) for all patients who underwent lymphadenectomy and 33.3% (2/6) for those with stage III or IV disease. These results may indicate that the lymph node metastatic rate is very low for clinical stage I or II patients and higher for patients with advanced-stage disease. Of the 7 patients with recurrence, 4 (57.1%) had lymph node metastasis, and 3 of them had initial advanced disease, which also suggested that patients with advanced disease more easily develop lymph node metastasis. Thus, it is especially important to carefully inspect the lymph node status during operation for patients with advanced disease and remove the enlarged lymph node if the intraperitoneal lesion has been optimally removed. Although the 5-year survival rates for patients with and without node dissection were not significantly different (86.4% vs 86.0%) based on the small number of cases, the survival benefit and necessity for patients receiving lymphadenectomy with different disease stages can be further analyzed.

In the literature, the reported prognostic factors of ovarian YST are tumor stage,1,7,11 the volume of ascites,3,7 high level of serum AFP12,13 or postoperative serum AFP declining rate,3 residual tumor,7,11,14 cisplatin-based chemotherapy in advanced disease,14 and chemotherapy with a BEP regimen.3,15 In our study, univariate analysis showed that age, tumor stage, AFP decline, and residual tumors affect the prognosis. Multivariate Cox analysis revealed that only satisfactory postoperative AFP decline was an independent significant prognostic factor for 5-year survival rate. This is similar to the result of a previous report. That study showed that the serum half-life of AFP was significantly predictive of patient OS.3 Patients with a serum half-life of AFP of more than 10 days had lower 5-year OS than patients did with half-life of 10 days or less (59% vs 92%).3 AFP decline rate may help clinicians distinguish patients with low or high risk of recurrence and determine the number of chemotherapy cycles. For patients with satisfactory AFP decline, 2 more cycles of chemotherapy after normalization of AFP may be enough no matter the stage of disease. This situation is similar to that in patients with stage IA disease after the tumor has been removed. de la Motte et al3 suggested that patients with stage IA disease may need only 2 cycles of BEP chemotherapy. This point needs to be evaluated in a well-organized perspective study. For patients with unsatisfactory AFP decline, intense treatment may be necessary, such as more cycles of chemotherapy, or new treatment regimen may need to be investigated.

As of the impact of AFP level on patient survival, the results of previous studies were inconsistent. In some literature, high level of serum AFP was reported to be a prognostic factor,12,13 but in other articles, the AFP level did not significantly affect the prognosis.7,14 In our study, patients with high pretreatment AFP level tend to have poor survival (P = 0.07). The inconsistency between studies may be due to the small number of cases. What is more, pretreatment AFP level is related to tumor burden, which depends on tumor size and extent of tumor spread (tumor stage). Patients with early-stage disease may have a large-size tumor and high AFP level but good outcome; patients with late-stage disease may have small-volume tumor and low AFP level but poor outcome.

Studies showed that residual tumor decreased patient OS.7,11,14,16 For example, in the study of Nawa et al,7 the 5-year survival rate was 78% for patients with residual tumor of 2 cm or less and 29.2% for those with residual tumor of larger than 2 cm, respectively. However, there were 2 obvious drawbacks in previous studies. First, stage I patients were included in the group with no or less residual tumor.7,11,14,16 Actually, the term residual tumor is supposed to be used in advanced-stage disease with spread tumor in peritoneal cavity that cannot be removed. Second, mixed germ cell tumor was included in some studies, which has different characteristics compared with pure YST.7,14,16 In our study, we included only stage II-IV patients to evaluate the impact of residual tumor on survival. Univariate analysis showed that residual tumor is a significant prognostic factor, but multivariate analysis did not. However, the 5-year OS tends to favor the group without residual tumor (92.9% vs 60.0%). These data suggest that more aggressive surgery may render patients more survival benefit. What is more, whether there is survival difference between patients without residual tumor and those with only small residual tumors less than 1 or 2 cm remains to be further evaluated. Because YST tumor is highly sensitive to BEP chemotherapy, small residual tumors less than 1 or 2 cm may be diminished by postoperative chemotherapy.

Although cisplatinum-based chemotherapy dramatically improved the prognosis of ovarian YST, a few patients may still relapse, and rare data are available regarding salvage chemotherapy in the literature. High-dose chemotherapy combined with autologous stem cell transplantation may provide an opportunity for complete tumor remission.3,17,18 In our study, 1 patient had a para-aortic lymph node recurrence at 6 months after discontinuation of the BEP regimen and 10.7 months after the initial diagnosis. She was successfully salvaged with another 4 cycles of BEP regimen. Another patient had lung recurrence after BEP regimen and achieved complete remission on chemotherapy with paclitaxel plus cisplatin. A tumor reappearing in the lung also had complete remission after chemotherapy with epirubicin plus nedaplatin, which suggests that the 2 regimens, paclitaxel plus cisplatin and epirubicin plus nedaplatin, may be another option for salvage chemotherapy. Four of the 7 patients lived 3 years or more, and 2 patients lived even longer than 5 years after recurrence. These data demonstrate that salvage treatment with a combined modality may render long-term survival possible in some patients.

In patients with recurrence, patient 3 had brain metastasis following her lung metastasis (Table 4). To our knowledge, this is the first case of brain metastasis reported in an ovarian YST patient. She received radiation therapy for brain metastasis but failed and died of the disease 87.7 months after the initial diagnosis.

One should be aware of the limitations of this study. This study is retrospective and is subject to the limitations of this type of study. The number of cases was small. The treatment information in 2 of the recurrent cases is not available. The results of the study must be interpreted with caution.

In conclusion, postsurgery satisfactory AFP decline was an independent significant prognostic factor for patient survival. Omentectomy may not be necessary for clinical stage I or II patients. Recurrent patients may obtain long-term survival, and proper salvage treatments need further study.

Back to Top | Article Outline


1. Dallenbach P, Bonnefoi H, Pelte MF, et al. Yolk sac tumours of the ovary: an update. Eur J Surg Oncol. 2006; 32: 1063–1075.
2. Kurman RJ, Norris HJ. Endodermal sinus tumor of the ovary: a clinical and pathologic analysis of 71 cases. Cancer. 1976; 38: 2404–2419.
3. de la Motte RT, Pautier P, Rey A, et al. Prognostic factors in women treated for ovarian yolk sac tumour: a retrospective analysis of 84 cases. Eur J Cancer. 2011; 47: 175–182.
4. de la Motte RT, Pautier P, Duvillard P, et al. Survival and reproductive function of 52 women treated with surgery and bleomycin, etoposide, cisplatin (BEP) chemotherapy for ovarian yolk sac tumor. Ann Oncol. 2008; 19: 1435–1441.
5. Liu Q, Ding X, Yang J, et al. The significance of comprehensive staging surgery in malignant ovarian germ cell tumors. Gynecol Oncol. 2013; 131: 551–554.
6. Cicin I, Saip P, Guney N, et al. Yolk sac tumours of the ovary: evaluation of clinicopathological features and prognostic factors. Eur J Obstet Gynecol Reprod Biol. 2009; 146: 210–214.
7. Nawa A, Obata N, Kikkawa F, et al. Prognostic factors of patients with yolk sac tumors of the ovary. Am J Obstet Gynecol. 2001; 184: 1182–1188.
8. Arie AB, McNally L, Kapp DS, et al. The omentum and omentectomy in epithelial ovarian cancer: a reappraisal: part II—the role of omentectomy in the staging and treatment of apparent early stage epithelial ovarian cancer. Gynecol Oncol. 2013; 131: 784–790.
9. Powless CA, Bakkum-Gamez JN, Aletti GD, et al. Random peritoneal biopsies have limited value in staging of apparent early stage epithelial ovarian cancer after thorough exploration. Gynecol Oncol. 2009; 115: 86–89.
10. Kumar S, Shah JP, Bryant CS, et al. The prevalence and prognostic impact of lymph node metastasis in malignant germ cell tumors of the ovary. Gynecol Oncol. 2008; 110: 125–132.
11. Tong X, You Q, Li L, et al. Prognostic factors of patients with ovarian yolk sac tumors: a study in Chinese patients. Onkologie. 2008; 31: 679–684.
12. Murugaesu N, Schmid P, Dancey G, et al. Malignant ovarian germ cell tumors: identification of novel prognostic markers and long-term outcome after multimodality treatment. J Clin Oncol. 2006; 24: 4862–4866.
13. Mitchell PL, Al-Nasiri N, A’Hern R, et al. Treatment of nondysgerminomatous ovarian germ cell tumors: an analysis of 69 cases. Cancer. 1999; 85: 2232–2244.
14. Kawai M, Kano T, Furuhashi Y, et al. Prognostic factors in yolk sac tumors of the ovary. A clinicopathologic analysis of 29 cases. Cancer. 1991; 67: 184–192.
15. Williams SD, Birch R, Einhorn LH, et al. Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide. N Engl J Med. 1987; 316: 1435–1440.
16. Kojimahara T, Nakahara K, Takano T, et al. Yolk sac tumor of the ovary: a retrospective multicenter study of 33 Japanese women by Tohoku Gynecologic Cancer Unit (TGCU). Tohoku J Exp Med. 2013; 230: 211–217.
17. Einhorn LH, Williams SD, Chamness A, et al. High-dose chemotherapy and stem-cell rescue for metastatic germ-cell tumors. N Engl J Med. 2007; 357: 340–348.
18. Voigt W, Kegel T, Maher G, et al. Bevacizumab plus high-dose ifosfamide, etoposide and carboplatin (HD-ICE) as third-line salvage chemotherapy induced an unexpected dramatic response in highly platinum refractory germ-cell cancer. Ann Oncol. 2006; 17: 531–533.

AFP decline; Chemotherapy; Omentectomy; Ovarian yolk sac tumor

© 2016 by the International Gynecologic Cancer Society and the European Society of Gynaecological Oncology.