Diabetes Forecast

Better Care for PCOS

Studying women's genes to find out what works

By Andrew Curry , ,

Corrine Welt, MD

Corrine Welt, MD
Assistant Professor,
Harvard Medical School
Prediabetes/Insulin Resistance
ADA Research Funding
Clinical Translational Award

First identified in 1935, polycystic ovary syndrome, or PCOS, is more a collection of symptoms than a disease with just one cause. Doctors usually diagnose the condition based on symptoms such as overweight or obesity, elevated testosterone, excessive hair growth, acne, irregular menstrual cycles, and difficulty conceiving children. Many women are first diagnosed when they try—and fail—to get pregnant.

"I think it could be a disorder where a number of different things are happening, but it all shows up in the same way," says Corrine Welt, MD, a researcher at Harvard Medical School. Specifically connected to the reproductive system, PCOS starts at puberty; as many as 1 in 10 women of reproductive age have it. That's 7 million women in the United States alone.

The cause of PCOS is still unclear, but researchers know one thing for sure: There's a link between PCOS and diabetes. "It has been shown that these women are at increased risk for developing type 2 diabetes, not just compared to other women their age but also compared to women of the same weight," Welt says. Women with PCOS are often insulin resistant, a condition that's an important trigger for type 2 diabetes.

Drawing on the DNA of thousands of women from around the world, Welt is trying to unravel the relationship between PCOS and diabetes and figure out the best way to treat it. At its most basic, her research looks at genes, the human body's blueprints, to isolate the cause of PCOS.

Every human genome contains around 25,000 genes, each the instructions for one or more structures or processes in the body. (Researchers have isolated the genes responsible for everything from whether your earwax is wet or dry to speech and language disorders.) To sort through the thousands of genes and figure out which ones might have something to do with PCOS, Welt and her collaborators put the genes of thousands of women on easily scanned chips, about the size of a credit card.

Help support diabetes science: Join the Summit Circle, ADA's society of individuals who make a planned gift. Please call 1-888-700-7029 or go to diabetes.org/giving.

If genes are the body's blueprints, then imagine the chips as a way to lay one blueprint atop another. Wherever the chips don't line up, researchers take note. If all the women with one pattern have an illness, like PCOS, and all the women without the pattern don't, it's a good signal that the genetic variant may have something to do with the illness. "If enough of the cases have the variant that's changed, we call it associated with the disorder," Welt says. "Then we go on to see what the gene is and what it's doing."

With the help of collaborators around the world, Welt has recruited more than 2,000 women with PCOS from Iceland, the Netherlands, Greece, and several U.S. cities, including Boston and Chicago. Working with an Icelandic company called deCODE, Welt is comparing their genes to the genetic code of about 15,000 women without the disorder.

In an effort to help people with diabetes, Welt is taking the research a step further. The insulin resistance of PCOS is often treated with metformin, a drug also prescribed to people with type 2 diabetes.

For PCOS patients, "there's a big push to use metformin, which helps decrease insulin resistance," Welt says. "It looks like in type 2 diabetes [prevention] trials if you take metformin, you do have decreased risks." That, Welt says, is enough to convince doctors that metformin can protect PCOS patients against type 2 diabetes.

Yet metformin sometimes doesn't improve insulin resistance or other PCOS symptoms, such as irregular menstrual cycles and high androgen levels, which can sometimes lead to infertility and uterine cancer. That means there are a lot of people taking metformin for no good reason. Welt thinks there may be a genetic component to women's response to metformin that would enable doctors to design better treatments. "If you put someone on metformin and they don't respond, you're not really treating them," Welt says. "For those patients, you'd be better off using other things."

With the help of the American Diabetes Association, she's recruiting 200 people with PCOS for a study that will look at their response to the popular drug. "We're treating patients with metformin and looking at insulin changes, ovulation response, and testosterone levels before and after treatment," she says. Welt's team checks on patients periodically over the course of four months to see how they're responding to the drug.

Using DNA taken from blood samples, the genetic "blueprints" of the women who don't respond to metformin will be compared with those of the women who benefit from the drug. "We can go back and look at their genetic makeup as a function of their response to metformin," Welt says. In the future, perhaps doctors can study a patient's genetic code to tailor treatment to individuals, making sure everyone gets the best care possible.



Take the Type 2
Diabetes Risk Test