IN 2007, LINDA GRIFFITH was putting together her first grant to study endometriosis. Her teenage niece had recently been diagnosed with the disorder. Griffith knew the symptoms from experience: severe, debilitating cramps, among other complications, which are brought on by displaced tissue from the lining of the uterus that grows outside the womb. Imagining what her niece would face during the coming decades, Griffith was appalled by the lack of significant progress on understanding or treatment since her own diagnosis. “It seemed like nothing had changed in 30 years,” says Griffith, a biological engineer and scientific director of the Center for Gynepathology Research at MIT.

As a scientist, Griffith had always viewed her own experience of the condition with critical interest. But her research role began in earnest after an emergency hysterectomy that was prompted in part by her endometriosis. The surgery was performed by Keith Isaacson, director of the Center for Minimally Invasive Gynecologic Surgery and Infertility at Newton- Wellesley Hospital. After the surgery, she flooded Isaacson  with questions about his work on endometriosis, and before long, they had begun  to collaborate.

Endometriosis occurs when cells from the uterine lining—the endometrium—grow outside the uterus. These cells may turn up on the ovaries, fallopian tubes, bladder or in other parts of the pelvic region. The condition strikes about one in 10 women and symptoms include severe menstrual pain, gastrointestinal problems, fatigue and infertility. Its signature symptom is “killer cramps”—debilitating menstrual cramps that don’t go away with a little ibuprofen. “The pain can be so intense that you can’t get out of bed or go to school or go about your normal activities,” says Jenny Hancher, education and outreach coordinator for the Endometriosis Foundation of America.

Current treatments include anti-inflammatory drugs, surgery to remove lesions, hormonal drugs and hysterectomy. And on the surgical front, at least, the past 15 years have brought important advances. For instance, being able to operate laparoscopically rather than through open surgery speeds recovery time and helps surgeons locate errant cells. Yet there hasn’t been a new Food and Drug Administration–approved therapy for endometriosis in more than two decades, surgery is rarely a cure, and treatments can cause life-altering side effects such as early menopause or suppressed ovulation. “We’re treating endometriosis with the same hammer that we were using 25 years ago, and we’ve made very little progress on the basic understanding and treatment,” Isaacson says. “As long as a patient is having recurring menstrual cycles, she’s at high risk of recurrence no matter what we do.”

And so countless patients remain in limbo. “I’ve had nine surgeries—you can’t live like that,” Griffith says. Endometriosis is often called a “benign” disease, but to the women whose lives are disrupted by the condition, she says, it feels anything but benign. She prefers to call endometriosis “non-malignant.” It may not be deadly, but it upends women’s lives. In her own research, and in her efforts to raise the profile of the disease in the medical community and the public eye, Griffith wants to change that.

DESPITE ITS PREVALENCE, endometriosis remains a scientific mystery. “We don’t know why some people have it and some people don’t, and why, for those who have the disease, the symptoms are so variable,” Isaacson says. It’s not clear how endometrial cells end up growing outside of the uterus. It’s also puzzling that the severity of symptoms doesn’t necessarily correlate with how extensive the disease is. The American Society of Reproductive Medicine has developed a staging system for endometriosis that is based on how much it has spread, what organs or other body structures are involved, the extent of pelvic adhesions and the degree to which the fallopian tubes are blocked. Yet these stages don’t seem to line up with symptoms. “Some people have stage one with a single spot [of endometrial tissue outside the uterus] and experience maximum pain, and others have stage four [the highest stage, signifying extensive lesions] and no pain,” Isaacson says.

The disease’s natural history also remains unclear. About 70% of those with stage one or stage two endometriosis will never advance to stage three or four, Isaacson says. “It doesn’t progress predictably like cancer does, so we treat it based on current symptoms, not because we’re afraid of what it will do five years from now.”

Griffith’s thinking on endometriosis took on a new perspective when she found a tumor in her breast in 2010. She learned that breast cancer tumors are analyzed for three molecular markers—estrogen receptor, progesterone receptor and HER2. Griffith was negative for estrogen and progesterone receptors, but borderline positive for HER2. Those results helped her doctors classify the aggressiveness of her cancer and aided decision-making about which therapies to try. Much of the recent progress in the battle against breast cancer comes from being able to employ targeted weapons that are more likely to be effective against a particular kind of tumor.

“Nothing like those classifications exists for endometriosis,” Griffith says. So while it’s plausible to think endometriosis, too, may exist in several molecularly distinct forms, not being able to identify those different types hamstrings efforts to find effective, nonsurgical treatments.

sp15_endo_spot1_630x420

BECAUSE OF THE WIDE variation in symptoms and how endometriosis may progress, Griffith and her colleagues hypothesize that there might be several distinct disease mechanisms at work—and it would be normal to expect those mechanisms to be distinguished by genetic differences. Yet while genes probably play a role in endometriosis, genetic research hasn’t found a primary culprit, and it’s likely that the condition, in all of its forms, can arise from multiple genetic factors.

So Griffith and her colleagues, in trying to parse the disease’s molecular structures, have taken a novel approach. Rather than starting with genes, they’re tackling the problem from the other end—proteins. In the body, the codes carried by genes in DNA are transcribed by RNA and then used to produce proteins, which are responsible for many biological functions. If the research team, looking at those end products—various proteins—could find similar proteins in patients who had the same symptoms, the scientists might begin to build a classification scheme that reflected the disease’s underlying mechanisms. That would be an improvement on the current system for staging endometriosis that is based primarily on what is seen during surgery.

The researchers started by examining peritoneal fluid, found inside the abdominal cavity, where the lesions of endometriosis appear. Because inflammation is a hallmark of endometriosis, it stood to reason that proteins known to be associated with inflammation that appeared in this fluid might yield clues about what was causing the disease.

For a study published in the February 2014 issue of Science Translational Medicine (Isaacson is a co-author), Griffith’s team measured the concentration of 50 proteins in peritoneal fluid collected from 77 women who had various stages and symptoms of endometriosis.

Rather than dividing the samples according to the severity of the women’s symptoms, the researchers analyzed them without any preconceived classifications. “We proceeded with our mathematical analysis without looking at clinical data about the patients,” Griffith says. “We put all of the women in one big database and did the mathematical analysis that asked, ‘Could we put them into groups that have common features?’”

The answer was yes. The researchers looked for groups of proteins that seemed to act in concert. In endometriosis, as in other complex inflammatory disorders—asthma, rheumatoid arthritis, inflammatory bowel disease—the immune system seems to mount an inappropriate response, and the scientists’ analysis turned up a pattern of activity involving 13 cytokines—cell-signaling proteins that can trigger inflammation in endometriosis patients—that were associated with specific symptoms. “The way we did our math, these were all molecules that changed together,” Griffith says.

If particular proteins change in combination with one another, that suggests they may be part of a network that is related in some mechanistic way, according to Griffith. About one-third of the women in the study had the 13 proteins the team identified as being related. And sure enough, those women also shared similar symptoms.

Now that they have this collection of cytokines, they can try to trace their origins to understand what’s happening on a molecular level—“who’s driving this conversation,” Griffith says. She says that’s akin to trying to figure out who was at a cocktail party you didn’t attend by listening to snippets of audio recorded at the event. “You can identify people by their manner of speaking, and you might know that this person tends to speak in a particular way when he has had too much to drink.” Using the cytokine signatures they have identified, the researchers can begin to work through how the networks might function by using databases that show which cytokines are expressed by various types of immune cells. The ultimate goal is to create something like what exists for breast cancer—a classification system that can guide treatment decisions, and that could provide targets for new drugs aimed at curtailing the inflammatory mechanisms of endometriosis.

THIS STUDY BY GRIFFITH and her team was groundbreaking, but it involved a relatively small number of patients. To generalize more broadly they will have to extend their analysis to diverse groups of women. And to do that, they need to be certain that outside researchers are collecting and analyzing peritoneal fluid and other biological samples in a way that is consistent with their own methods. The World Endometriosis Research Foundation has recently developed a set of standardized tools that may assist researchers like Griffith. Led by Stacey Missmer, an associate professor of obstetrics, gynecology and reproductive biology at Harvard Medical School, the project gathered more than 50 endometriosis experts from 16 countries to create protocols that will help facilitate the kind of big data analysis that researchers such as Griffith are doing.

That effort represents a much larger movement to foster collaboration among basic scientists, clinicians, surgeons and patient advocates on endometriosis. When Isaacson was president of the American Association of Gynecologic Laparoscopists, he invited Griffith to speak to their annual meeting about analogies between endometriosis and cancer and molecular approaches to stratifying these diseases.

“They’ve invited her back every year since,” says Isaacson. “They loved the idea of a surgeon like me with limited basic science background collaborating with someone who has become one of the leaders in this field.” In another example of cross-disciplinary collaboration, Griffith, an expert in tissue engineering who had previously helped grow a human ear on a mouse, is now working with Isaacson to develop a three-dimensional endometrium culture with immune cells inside that could help them study the mechanisms of endometriosis.

In the meantime, public outreach and education is crucial. Celebrity chef Padma Lakshmi, co-founder of the Endometriosis Foundation of America, recently recruited Griffith’s help in developing an app. The foundation will use it to educate teenagers. Though the average age at diagnosis is 27, symptoms typically appear much earlier, and delays in diagnosis are common. Some women are misdiagnosed with other conditions such as irritable bowel syndrome or pelvic inflammatory disease.

“We hear all too often of this unnecessary delay in diagnosis,” says Lone Hummelshoj, secretary-general of the World Endometriosis Society. “Both women and physicians continue to normalize the pains associated with endometriosis.” Women whose mothers or close female relatives have endometriosis are seven times more likely to be diagnosed with it than those without a family history, and if a woman’s relatives experienced terrible menstrual pain, she may assume that what she’s enduring is normal. When Griffith’s own niece developed symptoms of endometriosis, her complaints were regarded with suspicion, and her niece’s doctor suggested that the young woman was overdramatizing her debilitating menstrual pain to get out of school.

The app Griffith and her MIT colleagues are developing with the EFA aims to help young women avoid such diagnostic delays by teaching them about endometriosis and helping them track symptoms. No woman should have to fight her doctor to have her pain taken seriously, Griffith says. And now, if the research she and others are pushing forward succeeds at removing some of the mystery long attached to the disease, women who are diagnosed with endometriosis may have better, more targeted treatment options to consider.