Clinical Challenges of Low-Grade Ovarian Cancer

Interview With David M. Gershenson, MD

Janis C. Kelly; David M. Gershenson, MD


January 26, 2012

Editor's Note:

About 10% of all serous ovarian carcinomas are low grade. Recent research indicates that these tumors differ in major ways from high-grade serous ovarian carcinomas. David M. Gershenson, MD, who holds the J. Taylor Wharton, MD, Distinguished Chair in Gynecologic Oncology at The University of Texas MD Anderson Cancer Center in Houston, Texas, discussed with Medscape current approaches to low-grade serous carcinoma of the ovary and peritoneum.

Medscape: What are the main challenges in diagnosing and treating low-grade serous carcinoma (LGSC) of the ovary?

Dr. Gershenson: There's little difference in presentation, diagnosis, or initial treatment of low-grade serous carcinoma and high-grade serous carcinoma (HGSC). Both frequently present with abdominal bloating or pain, distension, and vague gastrointestinal symptoms. Patients typically present with a pelvic mass, and primary treatment is usually surgery. Preoperatively, transvaginal sonography, serum CA-125, and CT scanning may suggest an ovarian or peritoneal malignancy and indicate referral to a gynecologic oncologist. In early-stage LGSC, comprehensive surgical staging includes removal of the ovarian mass, biopsies of peritoneal surfaces, omentectomy, and lymphadenectomy. Most LGSC presents with advanced-stage disease, which is treated with maximal cytoreductive surgery followed by paclitaxel/carboplatin chemotherapy.

Standard primary treatment currently is no different from that for the typical high-grade serous carcinoma, but LGSC appears not to be as chemosensitive as the typical HGSC.[1] We think that these tumors have a slower cell cycle than HGSC, which makes them less sensitive to conventional chemotherapy.[2] The longer survival in LGSC may be related to either tumor biology or treatment.

Medscape: What are the implications for patients with LGSC?

Dr. Gershenson: Some patients are overly influenced by reports that these tumors are relatively chemoresistant compared with high-grade serous carcinomas. We are not suggesting that they are never sensitive to chemotherapy. Many patients with LGSC do respond, but some decide that they shouldn't even try chemotherapy. For instance, in the recurrent setting, we have found that about 60% of LGSC patients will have stable disease while they are on chemotherapy.[1] Often, patients who are deciding about treatment don't understand that stable disease is a favorable outcome. Of course, everybody wants a complete remission, but a patient with an indolent disease like LGSC could live years with stable disease.

However, you have to balance the side effects of chemotherapy against the benefits. That led us to search for other treatments.

Medscape: What is the relationship between low-grade serous ovarian carcinoma and peritoneal carcinoma?

Dr. Gershenson: Either HGSC or LGSC can arise in the peritoneum as well as in the ovary. The criteria we use to distinguish those two entities in the peritoneum are the same as in the ovary, including tumor distribution.[3] We think that LGSC frequency is similar to that of HGSC -- about 10 ovarian cases for every 1 peritoneal case -- but there may be some differences in clinical course.

Medscape: What other therapies do you use in LGSC?

Dr. Gershenson: For several years we have used hormonal therapies in LGSC, and there's no question that they are effective. We've just completed a study showing that the response rate with hormonal therapy in recurrent LGSC was higher than with chemotherapy. With standard chemotherapy drugs, the response rate was less than 5%, while about 60% of patients had stable disease. With hormonal therapy, the rate of stable disease was also about 60%, but the objective response rate was higher. Of course, it is not possible to directly compare response rates across different studies.

The hormonal therapies that are used for breast cancer (tamoxifen, letrozole, anastrozole) all have some activity in LGSC.[4,5] Although there's never been a head-to-head comparison, we do not believe that they are as active in high-grade serous carcinomas.

We are also studying the targeted agents, as we did with AZD6244 in the GOG 0239 trial.[6] We're currently planning a large, randomized phase 2 international trial with colleagues in the United Kingdom to test another MAPK/ERK kinase (MEK) inhibitor. The trial design we are discussing most intensively would include about 130 patients with recurrent LGSC who will be randomly assigned to the MEK inhibitor or to standard therapy (to include a choice of 5 chemotherapy agents and hormonal agents). We hope to activate the trial in the summer of 2012.

Medscape: Does the surgical approach differ for LGSC and HGSC?

Dr. Gershenson: Primary surgery is usually no different than for any type of ovarian malignancy. We use secondary cytoreduction for patients with a long interval between the primary treatment and a recurrence of LGSC and for patients with borderline tumors that recur as LGSC. There are no randomized studies of secondary cytoreduction surgery, but we believe it can be beneficial for certain patients.

For both LGSC and HGSC patients who present with extensive disease, we may utilize the strategy of neoadjuvant chemotherapy for 3 cycles followed by interval debulking surgery and further chemotherapy.

Medscape: What is the next step for LGSC clinical research?

Dr. Gershenson: The major finding in the GOG 0239 study was a higher response with the MEK inhibitor than with conventional chemotherapy or hormonal therapy, so these drugs need to be studied further. They should also be tested in conjunction with drugs that block the PI3-kinase/Akt pathway. Dual pathway blockade may be required to produce a more robust response.

In the GOG 0239 trial, we saw a response rate of slightly over 15%, which is about 3-fold higher than with conventional chemotherapy and about 50% higher than with hormonal therapies in the recurrent setting. Unlike the situation with melanoma, there was no correlation with mutational status. That was something of a surprise and suggests that we don't understand what the impact of having a mutation is. I was certainly disappointed that there was not a correlation. We are looking at other markers in that pathway using data from that trial, and that analysis has not yet been completed.

Medscape: Where do things stand with regard to biomarkers in LGSC?

Dr. Gershenson: Initial reports suggested that the mitogen-activated protein (MAP) kinase pathway was activated in LGSC and that a high portion of patients had mutations of the BRAF gene or the KRAS gene, which are in this pathway.[7] Our data showed that KRAS and BRAF mutations were not as common as originally reported.[8] BRAF mutations were found in about 5% of cases. We know that the MAP kinase pathway somehow is activated, but we don't yet have a good biomarker.

Medscape: What else is on the horizon in LGSC?

Dr. Gershenson: We are completing our analysis of hormonal therapy in a retrospective cohort study, and we are conducting genomic studies, including deep sequencing, in which we look with more detail at a larger number of genes. We are trying to identify mutations that haven't yet been described and that may offer potential targets for therapy.

We now have a database of about 500 patients, including about 320 with LGSC. We're about to launch a study focused on biomarkers, such as estrogen receptor, progesterone receptor, BRAF mutations, KRAS mutations, and several other potential genes to see if any have prognostic value.

Medscape: What are the most important unanswered questions about LGSC?

Dr. Gershenson: We need biomarkers that can predict response so that we are better able to select patients for various treatments. We need better treatments. So far, our focus has been mainly on women with recurrent LGSC, but we also need to figure out how to treat patients with newly diagnosed advanced-stage LGSC. Should we give them standard chemotherapy, or should we be treating them with hormonal therapy or a combination of chemotherapy plus a targeted agent?

Other challenges are to understand the epidemiology of borderline tumors and LGSCs. What women are at risk for developing these? We know that the average age is about 20 years younger for LGSC than for ovarian cancer in general.

Another challenge is to try to understand the effects of the environment. One clue is that fertility drugs increase a woman's risk for developing a borderline tumor. Since we know that borderline tumors can lead to development of LGSC, there probably is a connection.

We also know that some of these tumors are responsive to antiestrogen hormonal therapies, just like in breast cancer.[4] This raises a clinical question: If a young woman with a new diagnosis of LGSC has a hysterectomy and a bilateral salpingo-oophorectomy (BSO), should she have hormone replacement therapy? I have come to believe that that is probably not a good idea. It is possible that the hormonal environment may influence development of these tumors. I advise my patients with hysterectomies and BSOs not to take hormone replacement therapy. Even though I don't have any hard evidence, there are enough clues that it could be related and might predispose to earlier recurrence.

Finally, the principal advance over the past decade is the recognition of LGSC as a distinct entity[9] -- very different from the typical high-grade ovarian cancers. Women with this subtype deserve separate clinical trials and treatment options.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.