For four years, MIT bioengineer Linda Griffith has been slowly unraveling the biology of endometriosis, a complicated and poorly understood disease that she has silently struggled with her entire life. The condition—in which tissue normally found in the uterus grows elsewhere in the body—is remarkably common. It causes severe pain and infertility and affects up to 10 percent of women, yet relatively little is known about what causes it, how to prevent it, or even how to effectively treat it.
In a study published Wednesday, Griffith and colleagues present new work that is a first step toward providing a more informed way of classifying endometriosis based on the underlying biological cause of the disease. Such tools are sorely needed, Griffith said. The professional guidelines for classifying the disease once lesions are surgically removed vary—and those classifications of tissue samples are not always related to the amount of pain or infertility a woman suffers.
The research, published in the journal Science Translational Medicine, is influenced by Griffith’s own experience as a breast cancer survivor. When she was diagnosed with breast cancer in 2010, Griffith’s experience with that disease was very different than with her lifelong struggle with endometriosis. Doctors quickly did molecular tests to discern the underlying molecular drivers of her cancer and to help select the treatment most likely to work. She wanted to find a way to bring that personalized medicine approach to endometriosis, which is usually treated with a combination of hormones and surgery.
“We’re trying to start a conversation with the community where we can say, like cancer, the surgical appearance is part of what you use to understand the patient situation. But there are molecular markers that are very informative about mechanism,’’ Griffith said.
In the study, Griffith and colleagues studied samples taken from 77 women with suspected endometriosis. The samples, removed during surgery, were analyzed for immune system molecules involved in inflammation. The researchers were searching for patterns that would give them new ways to subdivide the women, who had variable levels of infertility and pain and may even have different causes driving their disease. Instead of looking for just one immune system protein that was elevated, they looked for networks of molecules that seemed to be elevated in concert; that approach is key because the disease is complex and not likely to be due to a single errant protein.
In about a third of the women with confirmed endometriosis, they found a network of 13 immune system proteins that were elevated. They cross-referenced those to try and figure out what kind of immune cell might be releasing them, which led them back to a particular type of cell called a macrophage. Using that information, they selected a specific experimental drug that they thought would inhibit the process and administered it to the patient cells in a dish. They found that they could reduce the inflammatory factors secreted by that cell.
The work is early and Griffith said her hope would be that a pharmaceutical company might pick up on the research and move toward developing a targeted drug that could be tested in a subset of patients likely to respond. They also hope to look in a larger patient population, to see if they can find similar hallmarks that could point to new ways to diagnose and treat the disease.
They also hope to extend the approach to other complex diseases that involve inflammation, such as inflammatory bowel disease.
“We’re not claiming we found a mechanism—the mechanism for endometriosis,’’ Griffith said. “We have found a very convincing approach to understand an immune network.’’