Article Text

Download PDFPDF

Frequency and prediction of deep uterine involvement in advanced high-grade epithelial ovarian cancer: is uterine preservation an option?
  1. Tamar Perri1,2,
  2. Gal Harel1,2,
  3. Tal Dadon2,
  4. Aya Mor-Sasson1,2,
  5. Itai Yagel1,2,
  6. Gilad Ben-Baruch1,2 and
  7. Jacob Korach1,2
  1. 1 Gynecologic Oncology, Sheba Medical Center, Tel Hashomer, Israel
  2. 2 Tel Aviv University Sackler Faculty of Medicine, Tel Aviv, Israel
  1. Correspondence to Dr Tamar Perri, Gynecologic Oncology, Sheba Medical Center, Tel Hashomer, Israel; tamarperri{at}gmail.com

Abstract

Introduction Hysterectomy is traditionally part of the surgical treatment for advanced high-grade epithelial ovarian carcinomas, although the incidence of uterine involvement has not been fully investigated. Some young patients with advanced high-grade epithelial ovarian carcinomas want uterine preservation. We aimed to determine the frequency of non-serosal (deep) uterine involvement in patients with high-grade epithelial ovarian carcinomas and to establish predictive factors for such involvement.

Methods A retrospective cohort study was performed of 366 consecutive patients with advanced high-grade epithelial ovarian carcinomas who had surgery between January 2012 and December 2019. Data collected included demographic and clinical details, and surgical and pathological reports to determine macroscopic and microscopic deep uterine involvement. The characteristics of the patients with and without deep uterine involvement were compared and univariate and multivariate Cox proportional hazard models were used to assess correlations and determine risk factors.

Results A total of 311 patients were included in the final analysis. The mean age was 62±11.6 years, with 32 (10.3%) being younger than 45. Most (92.3%) had serous carcinoma. Uterine involvement, excluding superficial (serosa-only), was present microscopically in 194 patients (62.4%) but was detected macroscopically at surgery in only 166 patients. Deep involvement was missed at surgery in 28 patients (14.4%), including parametrial involvement (n=18), parametria plus cervix (n=2), cervical involvement (n=3), endometrium (n=3), and myometrium (n=2). Multivariate analysis identified factors associated with deep uterine involvement including residual disease at surgery (HR 2.43, 95% CI 1.13 to 4.48; p=0.004) and CA125 >1000 U (HR 1.8, 95% CI 1.09 to 2.94; p=0.02).

Conclusions The incidence of deep uterine involvement in high-grade epithelial ovarian carcinomas is high. It can be diagnosed in most but not all cases on gross examination at surgery and is associated with residual disease and CA125 >1000 U. Patients who desire uterine preservation should be advised on an individual basis, given these factors and the operative findings.

  • cystadenocarcinoma
  • ascitic fluid
  • gynecologic surgical procedures
  • surgical procedures
  • operative
  • hysterectomy

Statistics from Altmetric.com

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.

HIGHLIGHTS

  • In advanced high-grade epithelial ovarian carcinoma, 62.4% had uterine involvement, 85.5% detected macroscopically at surgery.

  • Deep uterine involvement was associated with residual tumor at surgery and CA125 >1000 U.

  • Patients who desire uterine preservation should be advised based on these factors and the operative findings.

Introduction

Treatment of advanced high-grade epithelial ovarian carcinoma includes primary or interval debulking surgery with hysterectomy, bilateral salpingo-oophorectomy, and cytoreduction aiming toward minimal or no macroscopic disease along with platinum-based chemotherapy.1 2 While the age at which most patients with high-grade epithelial ovarian carcinoma are diagnosed is after the fifth decade, about 12% are diagnosed at age 20–44 years and 5.3% at 20–34 years.3 Some patients want fertility-preserving surgery, which might include preservation of the uterus. Uterine preservation, even after bilateral salpingo-oophorectomy, would theoretically enable further potential pregnancies for recovered patients either by transfer of pre-operatively frozen embryos or by oocyte donation.

While acceptable in early stage disease, in patients with advanced high-grade epithelial ovarian carcinoma, selection for uterine preservation necessitates careful consideration of the risks. Residual disease has proved to be one of the most significant prognostic factors for survival after high-grade epithelial ovarian carcinoma.4 If there is involvement of the tumor in the retained uterus, avoidance of hysterectomy might decrease the survival rate.

To date, only a few studies have reported on the incidence of uterine involvement in high-grade epithelial ovarian carcinoma, with highly inconsistent results ranging from 4.5% to 52%.5–7 Risk factors for uterine involvement are not well established.7 The aim of the present study was to determine the incidence of macroscopic and microscopic deep uterine involvement (infiltration beyond the serosa) in advanced high-grade epithelial ovarian carcinoma at a single medical center, and to define factors that could help in advising those patients who have advanced high-grade epithelial ovarian carcinoma and who desire to retain their uterus about the likelihood of deep uterine involvement.

Methods

The study was approved by the Ethics Committee of the Sheba Medical Center.

We reviewed the charts, surgical reports, and pathology reports of all 366 consecutive patients listed in our prospectively created computerized database who had been diagnosed with non-mucinous high-grade epithelial ovarian carcinoma and had undergone surgery at our tertiary center between January 2012 and December 2019. For the purposes of this study, we included only those patients who, at the time of diagnosis, had advanced disease. Exclusion criteria were low-grade carcinomas, no hysterectomy at time of study, pregnancy at diagnosis, and early stage disease.

All included patients had undergone laparotomy and surgery that included hysterectomy and aimed at optimal debulking. All patients had also completed 6–8 courses of platinum-based chemotherapy (post-operatively or as a neoadjuvant, usually with three courses before and three after surgery). Patients' demographics and clinical data, cancer characteristics, treatment, and outcomes had been recorded. Stage and grade had been established according to International Federation of Gynecology and Obstetrics criteria 2018.8 The CA125 levels at diagnosis, pre-operative imaging, and surgical records as well as pathology reports and imaging studies were reviewed in detail. Macroscopic and/or microscopic involvement was recorded including that of the uterine serosa, myometrium and endometrium, cervix and parametria, as well as the Douglas pouch adnexa, omentum, bladder plica, liver and spleen surfaces or parenchyma, mediastinum, brain, and other distant metastases.

Deep uterine involvement was categorized as involvement of one or more of the following: endometrium, myometrium, cervix, and parametrium. When serosal involvement had been only superficial so that the surgeon had apparently been able to peel the involved serosa from the uterus if attempted leaving a functional uterus, the finding was classified as ‘uterus not involved’.

Statistical analysis

Categorical variables were reported as frequencies and percentages. Visual evaluations of a histogram and Q–Q plot of each continuous variable were performed in order to evaluate approximation to the normal distribution and were reported as mean and SD or median and IQR. Between the two groups (‘uterus involved’ vs ‘uterus not involved’), categorical variables were compared using the χ2 or Fisher’s exact test and continuous variables were compared using the independent samples t-test or the Mann–Whitney test. Cox regression analysis was used to assess the association between deep uterine involvement and each of the following variables: age at diagnosis, menopausal status, body mass index, BRCA mutation status, gravidity and parity, CA125 levels at diagnosis, histology (serous, endometrial or other), stage (IIb−IIIb, IIIc or IV), primary ovarian/tubal carcinoma as opposed to primary peritoneal carcinoma, upfront surgery versus neoadjuvant chemotherapy, ascites at surgery, metastases in the mediastinum, intra-parenchymatic liver and/or spleen, diaphragmatic metastases, and residual disease at surgery. Variables significantly associated with deep uterine involvement were included in a multivariate analysis. The area under the receiver operating characteristic curve was used to evaluate the discriminatory ability of the CA125 level and of multivariate logistic regression findings. Nagelkerke’s R-squared was used as a measurement for the pseudo R-squared of the logistic regression.

All statistical tests were two-sided and p<0.05 was considered as statistically significant. IBM SPSS statistics software Version 23 (IBM Corp, Armonk, New York, USA) was used for all statistical analyses.

Results

Of the 366 consecutive patients with high-grade epithelial ovarian carcinoma who had undergone surgery at our medical center within the study period, 55 were excluded from this study for the following reasons: 11 had undergone hysterectomy for benign reasons prior to ovarian cancer diagnosis, 4 had opted for uterine-preserving surgery, 2 had been operated on during pregnancy, and 38 had been diagnosed at an early stage (I−IIa). The study group thus comprised 311 women with advanced (stage IIb−IV) high-grade epithelial ovarian carcinoma who had undergone hysterectomy as part of their cancer treatment. Their mean (SD) age was 62 (11.6) years and 32 patients (10.3%) were <45 years old at diagnosis. Their clinicopathological characteristics are shown in Table 1. Serous carcinoma was diagnosed in 287 patients (92.3%) and 68 patients in the study group (21.9%) were known to be BRCA mutation carriers.

Table 1

Clinico-pathological characteristics of the study group (n=311)

Other than those with superficial serosal involvement, deep uterine involvement was detected macroscopically in 166 patients (53.3%) at surgery. According to the pathology reports, however, actual deep uterine involvement had been detected microscopically in 194 patients (62.4%), correctly reported in all 166 patients in whom it was found macroscopically but missed at surgery in 28 patients (14.4%) in whom it had not been diagnosed macroscopically. Parametrial involvement had been missed macroscopically in 18 cases and parametria plus cervix in two cases. Also missed were cervical involvement (three cases), uterine involvement limited to the endometrium (three cases), and involvement limited to the myometrium (two cases). In patients who had neoadjuvant chemotherapy (n=197), the uterus was microscopically involved in 117 (59.3%), six were missed at surgery.

The distribution of different clinicopathological characteristics between patients with and without deep uterine involvement is shown in Table 2. Univariate analysis showed that the following variables were significant factors for deep uterine involvement: ovarian involvement (as opposed to primary peritoneal carcinoma), ascites, diaphragmatic disease, higher CA125 levels at diagnosis, and residual macroscopic disease at surgery. As shown in Table 3, multivariate logistic regression analysis indicated that only residual disease at surgery and CA125 >1000 U were significant predictors of deep uterine involvement (HR 2.43, 95% CI 1.13 to 4.48 and HR 1.8, 95% CI 1.09 to 2.94 for residual disease at surgery and CA125 >1000 U, p=0.004 and p=0.02, respectively).

Table 2

Distribution of different characteristics between patients with and without microscopic deep uterine involvement in advanced high-grade epithelial ovarian carcinoma and univariate analysis

Table 3

Multivariate analysis of risk factors for deep uterine involvement in advanced high-grade epithelial ovarian carcinoma

Discussion

Our data suggest that deep uterine involvement occurs in a large proportion of patients with advanced high-grade epithelial ovarian carcinoma (62.4% in our study), and in most cases may be detected macroscopically at surgery. The risk of microscopic deep uterine involvement is lower in patients with advanced high-grade epithelial ovarian carcinoma without ascites or diaphragmatic disease, with primary peritoneal disease without ovarian involvement, with CA125 <1000 U at diagnosis, and without residual disease at surgery.

Although hysterectomy is considered part of staging surgery in patients with advanced high-grade epithelial ovarian carcinoma, the incidence of uterine involvement in high-grade epithelial ovarian carcinoma has so far been addressed in only a few studies, with conflicting results. Chitrathara et al5 and Behtash et al6 reported involvement of only 13–17% whereas Menczer et al7 reported 52.5% involvement, closer to our findings. The difference might be attributed to our inclusion of parametrial and cervical involvement whereas others included only endometrial and/or myometrial involvement.

If the uterus is grossly involved at surgery, the surgeon would probably decide not to preserve it. However, in one study7 the frequency of macroscopic uterine involvement was found to be low, whereas microscopic-only involvement was as high as 38%, perhaps suggesting that the decision on hysterectomy should not be guided by intra-operative macroscopic findings alone. The discrepancy between this study and ours might be explained by the multicenter retrospective study design. As a result, data were based on non-uniform operative notes and pathology reports, as well as the different numbers of sections taken for pathology according to the policies of different institutions. All of these factors when taken together might have introduced a significant bias towards detection of uterine metastasis. In contrast, the groups of both Chitrathara et al5 and Behtash et al6 rarely detected microscopic uterine involvement in the absence of discernible gross involvement. In the present study, however, we found that detection of uterine involvement in 14.4% of patients would have been missed had the decision on whether or not to remove the uterus been based solely on intra-operative findings. We therefore suggest that, while in the majority of cases the surgeon can make decisions during surgery with regard to the safety of uterine preservation, the risk of missing deep uterine involvement should be discussed pre-operatively with the patient.

A major consideration is the impact of retained microscopic disease on prognosis in macroscopically complete cytoreduction in patients with high-grade epithelial ovarian cancer. Theoretically, retained uterus with microscopic disease might affect prognosis. This issue is under-explored for uterine microscopic disease, but it was previously shown that retained microscopic disease did not impact survival. In a study by Du-Bois et al,9 systematic pelvic and para-aortic lymphadenectomy of clinically uninvolved lymph nodes was not associated with better outcomes than no lymphadenectomy in patients with completely resected advanced high-grade epithelial ovarian cancer, despite the 55.7% of the nodes that were microscopically involved. One can postulate that perhaps retaining clinically uninvolved uterus might also be unrelated to prognosis even if microscopically involved.

Uterine conservation is usually reserved for patients with early stage non-epithelial ovarian cancers10 11 and low-grade epithelial tumors,12 and more recently, even for patients with early stage (but not advanced stage) high-grade epithelial ovarian carcinoma.13–16 However, new advances in the understanding of tumor biology and new strategies for the treatment of patients with high-grade epithelial ovarian carcinoma have led to better survival rates,17 18 and there is growing interest in uterine-preserving surgeries in young patients with advanced high-grade epithelial ovarian carcinoma.12 There are a few published case reports of conservative surgery for advanced high-grade epithelial ovarian carcinoma in selected patients with uterine preservation who conceived after receiving successful treatment of their disease,19 20 but there is yet only sparse available information about the long-term oncologic outcome.7 21

Another potential downside of retaining the uterus is the possible co-existence of ovarian and endometrial cancers, especially in younger patients. That incidence was recently reported to be about 7.4% in a large cohort of 133 481 cases of ovarian malignancies.22 It was suggested that, if uterine preservation is planned, endometrial sampling should be part of the pre-operative evaluation.5 It was also argued that BRCA mutation might expose affected patients to endometrial in addition to ovarian cancer.23 24 Our data, however, do not show that BRCA mutation carriers with high-grade epithelial ovarian carcinoma were at higher risk of deep uterine involvement.

One of the main concerns is that the anatomic proximity of the ovaries and the uterus and their common blood vessel coverage are likely to expose the uterus to tumor cells. Thus, the uterus might be a site of metastasis in women with high-grade epithelial ovarian carcinoma, and if not removed, would affect the prognosis because of non-optimal cytoreduction. Such proximity might also explain the higher incidence of uterine involvement in women with ovarian involvement (as opposed to primary peritoneal disease) found in our study.

No existing factors have yet been identified as predictors of uterine involvement. The results of our study suggest that, in most patients, uterine involvement can be detected during surgery and that, in those without gross involvement, uterine microscopic involvement is less common in patients with lower levels of CA125 and/or in those with primary peritoneal high-grade epithelial ovarian carcinoma, without ascites and with no diaphragmatic involvement, as well as in those who had surgery without residual disease. Except for primary peritoneal disease (discussed above), all of these parameters might be signs of a less aggressive disease.

Our study has certain limitations. The number of cases, although the largest so far reported from a single center, is not high. Because of the retrospective design, it is inevitable that operative reports might not be completely accurate. For instance, macroscopic parametrial involvement might not be reported as uterine involvement. However, our system required uniform surgical reports and in most cases the necessary information was available. Parametrial involvement was related to uterine involvement for the purpose of the study. The single-center design assured uniform handling of pathology specimens, with uniform numbers of sections taken from the uterus, parametria, and cervix. At the same time, a single-center study such as ours has the disadvantage that our experience may not always be applicable to other practices. The study included patients of all ages, pre- and post-menopausal, thus allowing us to determine the incidence of deep uterine involvement in all patients with advanced high-grade epithelial ovarian carcinoma. Since no difference was found in the incidence of deep uterine involvement when stratified by age and neither age nor menopausal status were risk factors for deep uterine involvement, our results reflect the status in all age groups including the younger patients who might want uterine preservation.

Notwithstanding the limitations, our study indicates that, despite the common prevalence of deep uterine involvement in women with high-grade epithelial ovarian carcinoma, uterine preservation might be considered under certain conditions if the patient desires. Young patients seeking fertility preservation require special interdisciplinary consultation, including pre-treatment evaluation with gynecological oncologists, medical oncologists, and fertility specialists. Furthermore, uterine conservation is not only relevant for fertility sparing but also to potentially reduce complications and risk for other women with high-grade epithelial ovarian cancer undergoing cytoreductive surgery. Our observation that deep uterine involvement is similar in all age groups suggests that uterine preservation might be an option also in morbid patients, when the length and complexity of the surgical procedure is a limitation for completing optimal treatment, if CA125 <1000 and complete cytoreduction with no residual macroscopic disease is achievable.

The pros and cons of uterine preservation should be thoroughly discussed and a carefully considered decision agreed on by the patient, her family, and physicians. Larger-scale studies are needed in this connection and, before more definitive decisions can be adopted, uterine preservation should meanwhile be considered on an individual basis.

References

Footnotes

  • TP and GH are joint first authors.

  • Contributors Conception or design of the work: TP, JK, GB-B. Acquisition, analysis or interpretation of data: TP, JK, GH, TD, AM-S, IY, GB-B. Drafting the work: TP. Revising it critically for important intellectual content: JK, GH, TD, AM-S, GB-B, IY. Final approval of the version published: TP, JK, GH, TD, AM-S, IY, GB-B. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

  • 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.

  • Data availability statement Data and statistics are available upon reasonable request from tamarperri@sheba.health.gov.il.

  • Author note This work was performed in partial fulfillment of the M.D. thesis requirements of the Sackler Faculty of Medicine, Tel Aviv University.