Article Text

other Versions

Download PDFPDF
Emerging biomarkers in ovarian granulosa cell tumors
  1. Anne M Mills1,
  2. Zachary Chinn1,
  3. Lisa A Rauh2,
  4. Anna C Dusenbery1,
  5. Rachel M Whitehair1,
  6. Erin Saks2 and
  7. Linda R Duska2
  1. 1Department of Pathology, University of Virginia, Charlottesville, Virginia, USA
  2. 2Department of Obstetrics and Gynecology, University of Virginia, Charlottesville, Virginia, USA
  1. Correspondence to Anne M Mills, Department of Pathology, University of Virginia, Charlottesville, VA 22904, USA; amm7r{at}


Objective Although the majority of ovarian granulosa cell tumors can be successfully managed with surgery, a subset require chemotherapy for residual and recurrent disease. The benefit of chemotherapy in this population, however, remains controversial. There is therefore interest in the development of more tolerable and effective treatment options for advanced ovarian granulosa cell tumors. We report the use of immunohistochemistry to investigate how biomarkers could inform clinical trials in granulosa cell tumors with an emphasis on emerging androgen antagonistic, immunotherapeutic, and anti-angiogenic approaches.

Methods Immunohistochemistry for androgen receptor, the immune markers programmed cell death ligand 1, indoleamine-2,3 dioxygenase, and cluster of differentiation 8, and the vascular marker cluster of differentiation 31 were evaluated on formalin-fixed paraffin-embedded whole tissue sections from 29 cases of adult-type granulosa cell tumors. Results were evaluated with clinicopathologic variables including recurrence.

Results 59% of granulosa cell tumors were androgen receptor-positive, suggesting a potential role for anti-androgen therapy in this tumor type. In contrast, the targetable immune modulatory molecules programmed cell death ligand 1 and indoleamine-2,3 dioxygenase were scarcely expressed, with no cases showing tumorous programmed cell death ligand 1 and a single case demonstrating very focal tumorous indoleamine-2,3 dioxygenase staining. A minority of cases expressed programmed cell death ligand 1 in occasional tumor-associated macrophages and indoleamine-2,3 dioxygenase in peritumoral vessels. Tumor-infiltrating cytotoxic T cells were also scarce in granulosa cell tumors, arguing against a significant role for immunotherapy in the absence of additional immunostimulation. Cluster of differentiation 31 immunostaining revealed a range of vascular densities across granulosa cell tumors, and future studies evaluating the role of vascular density as a predictor of response to angiogenesis inhibition are warranted. None of the biomarkers investigated were significantly correlated with recurrence, and the only clinicopathologic feature significantly correlated with outcome was stage at presentation.

Conclusions Biomarker data suggest that many ovarian granulosa cell tumors could be candidates for anti-androgen therapy, while the potential role for immunotherapy appears more limited. Vascular density could be useful for identifying optimal candidates for angiogenesis inhibition. Incorporation of these biomarkers into clinical trials could help optimize patient selection.

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.


  • Presented at This work was previously presented in part at the 2018 Annual Meeting of the College of American Pathologists.

  • Funding This work was funded by the Peyton Taylor Jr Women’s Oncology Research Fund.

  • Disclaimer The authors have no relevant disclosures to report. This manuscript is not under consideration for publication elsewhere.

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