Hormone therapy for endometrial cancer targets connective tissue, not tumor cells

The female hormone progesterone has been used for several decades as a therapy for endometrial cancer, which starts in the lining of the uterus. Yet scientists didn't understand the mechanisms behind the therapy or its site of action.

Now, a new a study from the G.O. Discovery Lab(Link is external) (Link opens in new window) team at UCLA and UCLA collaborators shows that progesterone, rather than directly targeting tumors, eliminates tumor cells indirectly by binding to progesterone receptors in connective tissue cells (stroma) in the tumors' microenvironment.

Like breast and prostate tumors, endometrial cancer is regulated by hormones. But unlike therapies for those cancers, in which drugs block hormone signaling, progesterone therapy actually stimulates its hormone receptor, the researchers found.

While doctors know that a certain subsets of patients will benefit from progesterone treatment, prescribing the therapy is like shooting in the dark because it's unclear in advance which women will respond and which may have resistant tumors, according to the study's senior author, Dr. Sanaz Memarzadeh, an assistant professor of obstetrics and gynecology and director of the G.O. Discovery Lab at UCLA. Therefore, while progesterone can be effective as a therapy in endometrial cancer, its use is not widely embraced in clinical practice, she said.

"When viewing tumors under the microscope, clinicians often focus on the cancer cells and neglect the supporting stroma in the microenvironment," said Memarzadeh, who also is a researcher at the Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research at UCLA(Link is external) (Link opens in new window) and UCLA's Jonsson Comprehensive Cancer Center(Link is external) (Link opens in new window). "In this study, we found that all of the progesterone anti-tumor effects are in fact mediated through the stroma, even though it makes up a minor fraction of the tumor. I believe these exciting findings are going to surprise the clinical community and change the way people look at patterns of hormone-receptor expression in endometrial tumors."

The results of the three-year study, done using a specially developed laboratory model created by Memarzadeh's team that closely mimics human endometrial cancer, appear in the early online edition of Cancer Research, a peer-reviewed journal of the American Association for Cancer Research.

Memarzadeh and her team showed that when progesterone receptors are deleted from stromal cells in the tumor microenvironment, progesterone therapy won't work. However, in a model of hormone-resistant endometrial cancer, they found that tumor cells became sensitive to hormone therapy when the progesterone receptors were returned to adjacent cells in the microenvironment.

"We were really surprised to find that when we added back the progesterone receptor to the microenvironment, tumors that before did not respond to the treatment simply melted away," said the study's first author, Deanna Janzen, a senior research associate with the G.O. Discovery Lab. "Making one genetic change in the tumor microenvironment — deleting or adding back the progesterone receptor — completely changed the biology of the tumor. That was a striking finding."

Going forward, Memarzadeh and her team will translate this work into studies of human samples of endometrial cancer to see if their findings apply to patients. They hope to discover biomarkers that indicate response or resistance to hormone therapy. They also plan to find and test drugs that can reverse progesterone resistance, making cells sensitive to hormone therapy. This approach will provide a potential combination therapy that could prove effective for women with disseminated endometrial cancer.

Currently, the most common treatment for early-stage endometrial cancer is a hysterectomy, followed by radiation and/or chemotherapy. Doctors may prescribe progesterone to endometrial cancer patients who are seeking to preserve their fertility, although they don't know whether it will be effective. Therefore, finding biomarkers that indicate a response to therapy would provide clinicians with a valuable tool, Memarzadeh said.

Endometrial cancer is the most common gynecologic cancer in the United States. About 49,000 new cases will be diagnosed this year alone, and about 8,000 women in the U.S. will die from their cancers. The chance of a woman being diagnosed with this cancer in her lifetime is about one in 38, according to the American Cancer Society.

"This finding may have critical clinical implications, as it demonstrates that modulation of the tumor microenvironment can reverse hormone resistance in endometrial tumors," the study states. "In future work, we will test if stromal-specific delivery of DNA methyltransferase inhibitors may be an effective way to re-sensitize hormone refractory endometrial cancers to progesterone therapy."

Ultimately, the G.O. Discovery Lab team hopes to develop a simple test so that, after biopsy and analysis of an endometrial tumor, physicians will be able to figure out whether or not their patient is a good candidate for hormonal therapy.

The study was funded by the Veterans Administration Greater Los Angeles Healthcare System, the Scholars in Translational Medicine Program, a Mary Kay Award, a Charles Drew University/UCLA National Institutes of Health grant (U54-CA-143931), a Sidney Kimmel Foundation award, the Concern Foundation, and a UCLA Cancer Research Coordinating Committee grant.