Fax: (212) 305-3412
Fertility preservation in young women with epithelial ovarian cancer
Article first published online: 10 AUG 2009
Copyright © 2009 American Cancer Society
Volume 115, Issue 18, pages 4118–4126, 15 September 2009
How to Cite
Wright, J. D., Shah, M., Mathew, L., Burke, W. M., Culhane, J., Goldman, N., Schiff, P. B. and Herzog, T. J. (2009), Fertility preservation in young women with epithelial ovarian cancer. Cancer, 115: 4118–4126. doi: 10.1002/cncr.24461
- Issue published online: 4 SEP 2009
- Article first published online: 10 AUG 2009
- Manuscript Accepted: 18 DEC 2008
- Manuscript Revised: 14 DEC 2008
- Manuscript Received: 21 OCT 2008
- ovarian cancer;
- fertility-conserving surgery
Surgical management of ovarian cancer consists of hysterectomy with bilateral oophorectomy. In young women, this results in the loss of reproductive function and estrogen deprivation. In the current study, the authors examined the safety of fertility-conserving surgery in premenopausal women with epithelial ovarian cancers.
Women aged ≤50 years with stage IA or IC epithelial ovarian cancer who were registered in the Surveillance, Epidemiology, and End Results database were examined. Patients who underwent bilateral oophorectomy were compared with those who underwent ovarian conservation. A second analysis examined uterine conservation versus hysterectomy. Multivariate Poisson regression models were developed to describe predictors of fertility preservation. Survival was examined using Cox proportional hazards models and the Kaplan-Meier method.
In total, 1186 women, including 754 women (64%) who underwent bilateral oophorectomy and 432 women (36%) who underwent ovarian preservation, were identified. Younger age, later year of diagnosis, and residence in the eastern or western United States were associated with ovarian preservation (P < .05 for all). Women with endometrioid and clear cell histologies and stage IC disease were less likely to have ovarian conservation (P < .05). In a Cox model, ovarian preservation had no effect on survival (hazard ratio [HR], 0.69; 95% confidence interval [CI], 0.39-1.20). Young age, later year of diagnosis, residence in the eastern or western United States, single women, mucinous tumors, and patients with stage IA disease were more likely to have uterine preservation (P < .05 for all). In a multivariate model, uterine preservation had no effect on survival (HR, 0.87; 95% CI, 0.62-1.22).
Ovarian and uterine-conserving surgery were safe in young women who had stage IA and IC epithelial ovarian cancer. Cancer 2009. © 2009 American Cancer Society.
In 2008, it is estimated that 21,650 women in the United States will be diagnosed with ovarian cancer and that 15,520 will die from the disease.1 Despite the finding that the majority of women will be diagnosed with metastatic disease at the time of presentation, a number of patients will have stage I tumors confined to the ovaries. Compared with patients who have advanced stage disease, the prognosis for patients who have stage I tumors is excellent.2-4 In a study that examined the natural history of stage I ovarian tumors managed without chemotherapy, the 5-year survival rate was 94% for stage IA lesions, 92% for stage IB tumors, and 84% for stage IC neoplasms.5
Although it is primarily a disease of older women, it is estimated that 3% to 17% of ovarian tumors occur in women aged ≤40 years.6-11 The significance of age as a prognostic factor for ovarian cancer has been debated.6-11 Young women with ovarian cancer are more likely to present with early stage disease and lower grade tumors.6-11 Several reports have suggested that, even after accounting for stage and grade, young women have a more favorable prognosis.6-11 In a population-based study, young women had a significant survival advantage across all stages. The 5-year survival rate for women with stage III and IV tumors aged <30 years was 56% compared with only 22% for patients aged ≥60 years.11
The primary surgical management of epithelial ovarian cancer consists of hysterectomy in combination with bilateral oophorectomy. In young women, this not only results in the loss of reproductive function but also subjects them to the long-term consequences of estrogen deprivation. Given the increasing frequency of ovarian cancer in young women, the objective of our study was to examine the safety of fertility-conserving surgery in premenopausal women with epithelial ovarian cancer.
MATERIALS AND METHODS
The National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database was used. SEER is a population-based cancer registry that includes approximately 26% of the United States population.12 SEER is made up of a number of geographically distinct tumor registries. The demographic characteristics of the SEER registries have been described previously.13, 14 Data from SEER 17 registries was used. Exemption from Institutional Review Board review was obtained.
Women aged ≤50 years who had epithelial ovarian cancer diagnosed between 1988 and 2004 were analyzed. Only women who had tumors classified as serous carcinomas, endometrioid carcinomas, mucinous tumors, and clear cell carcinomas were included. Patients with stage IA and IC tumors were examined. Staging information was derived from the American Joint Cancer Committee (AJCC) classification system and was recorded according to extent of disease codes. Clinical and pathologic data, including age at diagnosis (ages <25 years, 26-35 years, or 35-50 years), race (white, black, or other), marital status, tumor grade, performance of lymphadenectomy, and receipt of radiation were collected. Year of diagnosis was stratified as either 1988 to 1997 or 1998 to 2004 for analysis. Patients were categorized based on the geographic area of residence at the time of diagnosis: central (Detroit, Iowa, Kentucky, Louisiana), eastern (Connecticut, New Jersey, Atlanta, rural Georgia) and western (Alaska, California, Hawaii, Los Angeles, New Mexico, San Francisco, San Jose, Seattle, Utah).
Two separate analyses were performed. First, patients were stratified based on whether they underwent oophorectomy (oophorectomy group) or whether 1, or a portion of 1, of their ovaries was retained (ovarian preservation). A second analysis examined the safety of uterine conservation. For this portion of the study, patients were stratified based on whether they underwent hysterectomy or the uterus was retained (uterine preservation).
The vital status of each patient was recorded. Survival was calculated as the number of months from cancer diagnosis to the date of death. Patients who were alive at last follow-up were censored. Both overall survival and cancer-specific survival were calculated.
Frequency distributions between categorical variables were compared using the chi-square test. The Fisher exact test was used if the expected frequency was <5. Because the prevalence of ovarian and uterine preservation was greater than 20%, Poisson regression models with robust estimation of standard error were developed to describe predictors of the receipt of ovarian and uterine preservation. In the Poisson regression analysis, we modeled all categories of each variable and controlled for all other variables in the model. Cox proportional hazards models were developed to examine survival. In Cox proportional hazards analyses, we modeled the overall mortality hazard ratios by comparing patients who underwent oophorectomy or hysterectomy with patients who did not and controlling for the other predictive variables. Kaplan-Meier curves were generated to examine overall survival and cancer-specific survival based on whether patients underwent oophorectomy or hysterectomy. All hypothesis tests were 2-sided, and the significance level was set at .05. Statistical analysis was conducted using Stata 10.0 (Stata Corporation, College Station, Tex).
In total, 1186 women were identified for the analysis of ovarian preservation. The cohort included 754 women (64%) who underwent bilateral oophorectomy and 432 women (36%) who had preservation of at least a portion of 1 ovary. The demographic characteristics of the population are displayed in Table 1. Ovarian preservation was performed more commonly in young women. Among women in the ovarian-preservation cohort, 15% were aged <25 years compared with only 0.5% of patients who underwent bilateral oophorectomy.
|Characteristic||Ovarian Preservation||Uterine Preservation|
|Oophorectomy, n=754||Ovarian Preservation, n=432||P||Hysterectomy, n=2242||Uterine Preservation, n=679||P|
|Year of diagnosis||.84||.23|
|Lymph node dissection||<.0001||<.0001|
A Poisson regression model of factors associated with ovarian preservation is shown in Table 2. Young age, later year of diagnosis, residence in the eastern or western United States, and early stage all were associated with ovarian preservation. In contrast, women who had endometrioid and clear cell tumors and those who underwent lymphadenectomy were less likely to have ovarian preservation. Compared with women in the central United States, those in the eastern United States were 52% more likely to have ovarian preservation (relative risk [RR], 1.52; 95% confidence interval [CI], 1.23-1.86), whereas those in the western United States had a 36% increased chance of having ovarian preservation (RR, 1.36; 95% CI, 1.12-1.64). Patients who had stage IC tumors were 33% less likely (RR, 0.77; 95% CI, 0.62-0.95) to have ovarian preservation than patients who had stage IA lesions.
|Variable||RR (95% CI)|
|Ovarian Preservation, n=1186||Uterine Preservation, n=2921|
|25-35||0.80 (0.71-0.89)||0.68 (0.59-0.76)|
|36-50||0.27 (0.23-0.32)||0.23 (0.19-0.27)|
|Black||1.02 (0.78-1.31)||1.09 (0.85-1.38)|
|Other||1.03 (0.86-1.22)||1.07 (0.91-1.25)|
|Year of diagnosis|
|1997-2004||1.17 (1.03-1.34)||1.15 (1.02-1.30)|
|Eastern||1.52 (1.23-1.86)||1.31 (1.09-1.57)|
|Western||1.36 (1.12-1.64)||1.23 (1.05-1.45)|
|Single/unknown||0.91 (0.76-1.04)||1.16 (1.02-1.31)|
|Serous||0.89 (0.75-1.06)||0.86 (0.73-1.00)|
|Endometrioid||0.74 (0.61-0.89)||0.81 (0.69-0.95)|
|Clear cell||0.63 (0.46-.0.86)||0.51 (0.38-0.68)|
|2||0.98 (0.82-1.17)||0.89 (0.76-1.03)|
|3||0.99 (0.74-1.32)||0.76 (0.58-0.97)|
|Unknown||1.15 (0.98-1.35)||1.17 (1.01-1.34)|
|Yes||0.53 (0.21-1.37)||0.10 (0.02-0.63)|
|Unknown||0.83 (0.39-1.77)||0.80 (0.38-1.67)|
|Not performed||1.19 (1.03-1.38)||1.58 (1.39-1.79)|
|IC||0.77 (0.62-0.95)||0.75 (0.63-0.88)|
A Kaplan-Meier analysis stratified by disease stage demonstrated no difference in survival based on performance of oophorectomy for either stage IA tumors (P = .47) or stage IC tumors (P = .15) (Fig. 1). In a Cox proportional hazards model that was adjusted for other prognostic variables, ovarian preservation was safe and had no effect on survival (HR, 0.69; 95% CI, 0.39-1.20) (Table 3). Only tumor grade and stage were predictive of survival.
|Variable||Ovarian Preservation, n=1148||Uterine Preservation, n=2921|
|Ovarian preservation||0.69 (0.39-1.20)||—|
|Uterine preservation||—||0.87 (0.62-1.22)|
|25-35||0.84 (0.29-2.36)||1.15 (0.53-2.49)|
|36-50||0.97 (0.34-2.79)||1.31 (0.61-2.81)|
|Black||1.91 (0.86-4.24)||1.42 (0.84-2.40)|
|Other||0.99 (0.54-1.80)||0.96 (0.65-1.43)|
|Year of diagnosis|
|1997-2004||0.61 (0.37-1.00)||0.66 (0.49-0.88)|
|Eastern||1.24 (0.62-2.44)||1.12 (0.78-1.63)|
|Western||1.42 (0.79-2.54)||1.04 (0.76-1.41)|
|Single/unknown||1.47 (0.96-2.24)||1.42 (1.11-1.82)|
|Serous||0.76 (0.42-1.37)||0.75 (0.53-1.07)|
|Endometrioid||0.66 (0.38-1.16)||0.67 (0.47-0.94)|
|Clear cell||1.13 (0.58-2.18)||1.02 (0.69-1.49)|
|2||0.77 (0.41-1.42)||1.24 (0.85-1.79)|
|3||2.39 (1.29-4.45)||3.04 (2.07-4.44)|
|Unknown||1.01 (0.57-1.77)||1.51 (1.05-2.18)|
|Yes||1.77 (0.69-4.57)||1.17 (0.62-2.24)|
|Unknown||3.53 (0.76-16.31)||0.89 (0.22-3.67)|
|Not performed||1.31 (0.82-2.07)||1.11 (0.86-1.43)|
|IC||2.43 (1.55-3.80)||1.75 (1.35-2.26)|
In total, 2911 women were identified for an analysis of the effect of uterine conservation. The group included 2242 women (77%) who underwent hysterectomy and 679 women (23%) who had uterine preservation (Table 1). Uterine preservation was performed more commonly in younger women, those who lived outside the central United States, and single women. In contrast, women with clear cell and endometrioid tumors and those with stage IC lesions were less likely to have uterine conservation (Table 2).
A Kaplan-Meier analysis revealed no difference in survival based on undergoing hysterectomy for either stage IA tumors (P = .55) or stage IC tumors (P = .26) (Fig. 2). In a multivariate Cox model, uterine preservation had no effect on survival (HR, 0.87; 95% CI, 0.62-1.22). Histology, marital status, tumor grade, and disease stage were predictors of survival.
Five-year survival also was examined (Table 4). There were no differences in survival based on undergoing either oophorectomy or hysterectomy in any of the groups that we examined. The 5-year survival rate among women who underwent oophorectomy was 91% (95% CI, 88%-93%) versus 94% (95% CI, 91%-96%) among women who underwent ovarian conservation. Similarly, the 5-year survival rate was 91% (95% CI, 90%-93%) among women who underwent hysterectomy compared with 92% (95% CI, 89%-94%) among women who retained their uterus.
|Oophorectomy||Ovarian Preservation||Hysterectomy||Uterine Preservation|
|5-Y Survival, % (No.)||91 (754)||94 (432)||91 (2242)||92 (679)|
|95% CI, %||88-93||91-96||90-93||89-94|
|5-Y Survival, % (No.)||94 (551)||95 (370)||93 (1506)||93 (548)|
|95% CI, %||91-96||91-97||92-95||90-95|
|5-Y Survival, % (No.)||83 (203)||88 (62)||87 (736)||88 (131)|
|95% CI, %||76-88||71-96||84-89||79-93|
The current findings suggest that, among young women with epithelial ovarian cancer, conservative, fertility-preserving surgery is safe. In our cohort, neither ovarian preservation nor uterine conservation had an adverse impact on survival. These findings were noted for patients who had stage IA and IC tumors.
Previous investigations of ovarian preservation for women with stage I epithelial ovarian cancer consisted primarily of small retrospective reports.15-19 In those studies, the recurrence rate after fertility-sparing surgery ranged from 4% to 28%.15-17 Most studies included stage IA and IC tumors, although fertility-sparing surgery was reported in a small number of women with more advanced stage disease.15-18 In a series of 52 patients with stage IA and IC disease, Schilder and colleagues reported an estimated 5-year survival rate of 98% and concluded that ovarian conservation was safe and was associated with an excellent outcome.17 Morice and coworkers noted a 5-year survival rate of 82% in their series and concluded that ovarian preservation should be approached cautiously and should be offered only to women with stage IA, grade 1 tumors.16 Relatively little data specifically have addressed uterine conservation. In a previous report of SEER data, Chan et al indicated that uterine conservation was safe in patients with stage I/II ovarian tumors.20 Our analysis differs, in that we focused on premenopausal women with stage IA and IC tumors and analyzed ovarian preservation in addition to uterine conservation.
A major concern after conservative therapy for early stage ovarian cancer centers on the ability to achieve pregnancy after treatment, particularly because many patients with stage I disease will receive adjuvant chemotherapy.2-4 Studies of young women with germ cell tumors of the ovary, breast cancer, and Hodgkin lymphoma have indicated that many women resume menstrual function after cytotoxic chemotherapy.21-23 Among women with early stage ovarian cancer who underwent fertility-conserving surgery, menstrual function was preserved in 94% of patients in 1 series.16 In a second series of young women who underwent fertility preservation, 71% of patients who attempted pregnancy conceived with no congenital anomalies in the offspring.17 Thus, it appears that the majority of women who undergo fertility-sparing surgery maintain reproductive function and that many are able to conceive, even after receipt of chemotherapy.16-18
Aside from reproductive concerns, surgical menopause in young women results in a number of long-term sequelae that can be avoided with ovarian preservation. Surgical menopause in young women increases the risk of coronary disease, osteoporosis and hip fracture, and cognitive dysfunction.24-27 A recent meta-analysis demonstrated that both early menopause and oophorectomy are associated with a significantly increased risk of cardiovascular disease.24 Of even greater concern, early oophorectomy has been correlated directly with increased mortality. In a modeling study, there was an 8.6% increase in mortality by age 80 years in women who underwent bilateral oophorectomy at age 55 years.28 This increased mortality was confirmed in a prospective cohort study in which prophylactic bilateral oophorectomy before age 45 years was associated with a 67% increase in mortality.29 The nonreproductive benefits of ovarian conservation should be considered in young women with early stage ovarian cancer.
A number of clinical considerations arise when contemplating fertility-sparing surgery for women with ovarian cancer. Lymphadenectomy should be performed routinely when stage I ovarian cancer is suspected. Occult lymph node disease will be identified in 10% to 25% of women with apparent stage I disease.30, 31 A second consideration is evaluation of the contralateral ovary. It appears that the risk of occult metastases in macroscopically normal appearing ovaries is small. No metastases were identified in random biopsies of normal appearing ovaries in 2 series of conservative surgical treatment.15, 16 In a report of 118 women with stage I ovarian cancer, occult disease in the contralateral ovary was noted in only 3% of patients.32 Finally, the question of whether completion oophorectomy/hysterectomy should be undertaken in women after the conclusion of childbearing frequently arises. Completion surgery is appealing, because many of the recurrences in patients who undergo conservative surgery occur in the contralateral adnexa.15-17 Schilder and colleagues noted that 3 of 5 recurrence (60%) in their series occurred in the contralateral ovary, whereas Morice et al noted 5 of 7 recurrences (71%) in the retained adnexa.16, 17 Whether completion surgery alters long-term outcome is unknown; thus, the decision to undergo surgery should be individualized.
In our series, fertility-conserving surgery, both ovarian and uterine preservation, was more performed commonly in younger women and in those with stage IA tumors. The rate of fertility preservation appears to be increasing, because women who were treated in the later years of the study were more likely to undergo conservative surgery. We noted that both ovarian and uterine preservation were performed more commonly in patients in the eastern and western United States as opposed to patients residing in the central United States. Finally, conservative surgery was undertaken less commonly in young women who had tumors with endometrioid and clear cell histologies. Given the aggressive behavior of clear cell ovarian tumors, conservative surgical therapy should be approached with caution in these patients.20
Although our study benefits from the inclusion of a large number of women, several limitations must be acknowledged. A priori, we assumed that premenopausal women had not previously undergone oophorectomy or hysterectomy. Although a small fraction of patients likely had undergone previous gynecologic surgery, the number of women who would have undergone bilateral ovarian removal probably was quite small. Not all patients in the series underwent lymphadenectomy; thus, the possibility of occult stage III disease cannot be excluded in all women. SEER lacks data on chemotherapy as well as patterns of recurrence. Although the cohort was large, the number of patients with high-risk features, such as stage IC disease, grade 3 tumors, and clear cell histology, was small. For example, to detect a 20% difference in survival for patients with stage IC tumors, a cohort of 1282 patients with 52 deaths would be required. Thus, given the limited sample size in these high-risk populations, fertility conservation should be undertaken with caution in these subgroups. Finally, individual patient and physician preferences that influenced ovarian and uterine preservation cannot be measured and certainly may have led to biases in surgical treatment.
In conclusion, the current findings suggest that ovarian-conserving and uterine-conserving surgery are safe in young women with stage IA and IC epithelial ovarian cancer. Five-year survival for patients in our series who underwent conservative surgery compared favorably with that of women who had stage I ovarian tumors reported in the literature.2-4 Given the potential reproductive and nonreproductive benefits of ovarian and uterine preservation, the benefits of conservative surgical management should be considered in young women with ovarian cancer.
Conflict of Interest Disclosures
The authors made no disclosures.
- 2International Collaborative Ovarian Neoplasm Trial 1 and Adjuvant Chemotherapy in Ovarian Neoplasm trial: 2 parallel randomized phase III trials of adjuvant chemotherapy in patients with early stage ovarian carcinoma. J Natl Cancer Inst. 2003; 95: 105-112., , , et al.
- 12National Cancer Institute, Surveillance, Epidemiology and End Results (SEER) Program. SEER*Stat Database: Incidence-SEER 9 registries limited-use, November 2006 submission (1973-2004), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch. Released April 2007, based on the November 2006 submission. Available at: http://seer.cancer.gov. Accessed October 1, 2007.
- 13National Cancer Institute, Surveillance, Epidemiology and End Results (SEER) Program. The SEER Program, Data Quality. Bethesda, Md: National Cancer Institute; 2007. Available at: http://seer.cancer.gov/about/. Accessed October 1, 2007.