Elsevier

Gynecologic Oncology

Volume 116, Issue 1, January 2010, Pages 50-56
Gynecologic Oncology

Tumor type and substage predict survival in stage I and II ovarian carcinoma: Insights and implications

https://doi.org/10.1016/j.ygyno.2009.09.029Get rights and content

Abstract

Objective

An ability to predict survival is of crucial importance in determining the need for cancer therapy. Recent advances in tumor typing of ovarian carcinomas lead to a classification which is more reproducible and reflects underlying biology more accurately than grade. We tested whether updated tumor type predicts outcome for patients with low-stage ovarian carcinoma.

Methods

From a population-based cohort of 1326 women diagnosed with stage I–II ovarian carcinoma between 1984 and 2003, 652 cases were available for central pathological slide review using contemporary criteria. Six hundred thirty cases were confirmed as ovarian carcinoma. Twenty-five ovarian carcinomas of rare types were excluded leaving 605 cases for this study. Recursive partitioning analysis and univariate models were used to identify subsets with an excellent outcome, i.e., disease-specific survival at 10 years (DSS10y) ≥ 95%.

Results

Seventy-seven ovarian carcinomas of endometrioid and mucinous type, stage Ia or Ib, were associated with an excellent outcome [DSS10y = 95%]. No subset of the high-grade serous type with an excellent outcome could be identified. Clear cell carcinomas of stage Ia or Ib had a favorable outcome [DSS10y = 87%] compared to stage Ic–II [DSS10y = 66%].

Conclusions

A subset of ovarian carcinoma patients with an excellent outcome can be identified based on tumor type (endometrioid or mucinous) and stage (Ia or Ib). Type is more reproducibly assigned than grade and identifies a larger cohort of women with stage I/II ovarian carcinoma with favorable outcomes (12.2% vs. 6.5%), and therefore is superior to grade in estimating risk of death from ovarian carcinoma.

Introduction

Young et al. reported in 1990 that women diagnosed with well-differentiated (grade 1) or moderately-differentiated (grade 2) ovarian carcinomas confined to the ovaries (stage Ia and Ib) had a similar 5-year overall survival rate (94% versus 98%) [1]. A much larger study of 1545 patients with stage I ovarian carcinoma reported grade as the most powerful prognostic indicator of disease-specific survival (grade 2 versus grade 1 hazard ratio 3.13 [95% confidence interval 1.68–5.85]) [2]. This resulted in a widely adopted treatment recommendation that women diagnosed with ovarian carcinoma, stage Ia and grade 1, should not receive adjuvant therapy [3]. Many centers include stage Ib, grade 1 in this low-risk category; some authors have suggested that the diploid subgroup of grade 2 tumors also does not need adjuvant therapy when diagnosed at stage Ia [4], [5]. A significant problem with these recommendations, however, is that grade assignment lacks reproducibility (k = 0.27–0.40, depending on the grading system used) [6], [7], [8], suggesting that grade is not a sufficiently reliable basis for individual treatment recommendations. It would be desirable to base practice on more robust variables.

Driven by better understanding of the molecular characteristics of tumor types, there have been recent changes in histopathological tumor type assignment [9], [10], [11]. Using contemporary histopathological criteria, type diagnosis is much more reproducible (kappa (k) = 0.77) than it was previously [6]. Tumor type is now appreciated to reflect different diseases, with type-specific risk factors, precursor lesions, molecular events during oncogenesis, biomarker expression, clinical behavior and response to chemotherapy [12], [13]. Tumor type-specific management has been suggested for ovarian carcinomas [14], as has been introduced for breast carcinomas [15]. Five tumor types account for 98% of ovarian carcinomas: high-grade serous, endometrioid, clear cell, mucinous and low-grade serous; the latter is a molecularly and clinically distinct ovarian carcinoma type, and not part of a continuum of disease with the more common high-grade serous carcinoma [16].

We hypothesized that tumor type could be used to identify patients with a low risk of death due to recurrent ovarian carcinoma (defined for the purpose of this study as a disease-specific survival at 10 years (DSS10y) ≥ 95%).

Section snippets

Study population

The British Columbia Cancer Agency (BCCA) provides cancer care for British Columbia's population, now approximately 4.1 million. Between the years 1984 and 2003 a total of 2947 women were referred to the BCCA for ovarian cancer management of which 1326 were stage I–II carcinomas. All cases were requested and 652 cases (49%) were available for full slide review. Two expert gynecological pathologists (C.B.G. and M.K.) reviewed all slides blinded to patient outcome data. Tumor type and grade (

Results

The clinicopathological features of the cases studied are presented in Table 1. Mean patient age was 57.2 years (range 26.4–89.6 years). Follow-up time ranged from 26 months to 23.6 years, with a mean follow-up time for censored patients of 7.2 years. Two hundred sixteen patients died during follow-up and 139 of the 216 deaths were due to ovarian carcinoma. The Kaplan–Meier survival analyses showed a relapse-free survival (RFS) at 5 and 10 years of 72.4% and 66.6%, respectively, and DSS5y and

Discussion

We defined the study endpoint of excellent outcome as ≥ 95% DSS10y. Establishing risk thresholds above which no meaningful benefit from adjuvant therapy can be expected is controversial. In the early 1990s a risk of 20% seemed acceptable for avoiding adjuvant therapy [1], [20]. With the development of more tolerable therapy this threshold is now often set between 5 and 10%. In addition to results based upon our strict definition of excellent outcome [DSS10y ≥ 95%], we present results in a way

Disclosure statement

The authors declare no potential conflicts of interest.

Acknowledgments

M.K. has received fellowship support from Eli Lilly Canada. C.B.G. was supported by the National Cancer Institute of Canada (#017051) and an unrestricted educational grant from sanofi aventis. This work was supported by a unit grant to OvCaRe from the Michael Smith Foundation for Health Research (#INRUA005045). We thank Nhu Le, Anna Tinker and Paul Hoskins for critical review of the manuscript.

References (43)

  • J.A. Irving et al.

    Synchronous endometrioid carcinomas of the uterine corpus and ovary: alterations in the beta-catenin (CTNNB1) pathway are associated with independent primary tumors and favorable prognosis

    Hum. Pathol.

    (2005)
  • R.C. Young et al.

    Adjuvant therapy in stage I and stage II epithelial ovarian cancer. Results of two prospective randomized trials

    N. Engl. J. Med.

    (1990)
  • K.D. Swenerton

    Prognostic indices in ovarian cancer. Their significance in treatment planning

    Acta. Obstet. Gynecol. Scand. Suppl.

    (1992)
  • C. Trope et al.

    Adjuvant chemotherapy for early-stage ovarian cancer: review of the literature

    J. Clin. Oncol.

    (2007)
  • N. Singh et al.

    Grading of serous ovarian carcinoma: further evidence of a lack of agreement between conventional grading systems

    Histopathology

    (2008)
  • S. Kommoss et al.

    Histological grading in a large series of advanced stage ovarian carcinomas by three widely used grading systems: consistent lack of prognostic significance. A translational research subprotocol of a prospective randomized phase III study (AGO-OVAR 3 protocol)

    Virchows Arch.

    (2009)
  • M.M. Leitao et al.

    Clinicopathologic analysis of early-stage sporadic ovarian carcinoma

    Am. J. Surg. Pathol.

    (2004)
  • R.J. Kurman et al.

    Pathogenesis of ovarian cancer: lessons from morphology and molecular biology and their clinical implications

    Int. J. Gynecol. Pathol.

    (2008)
  • R.A. Soslow

    Histologic subtypes of ovarian carcinoma: an overview

    Int. J. Gynecol. Pathol.

    (2008)
  • C.N. Landen et al.

    Early events in the pathogenesis of epithelial ovarian cancer

    J. Clin. Oncol.

    (2008)
  • M. Kobel et al.

    Ovarian carcinoma subtypes are different diseases: implications for biomarker studies

    PLoS Med.

    (2008)
  • Cited by (106)

    View all citing articles on Scopus
    1

    Present address: Department of Pathology and Laboratory Services, University of Calgary, Calgary, Alberta, Canada.

    View full text