One of the promises of “personalized medicine” is using new scientific discoveries about the human genome to pinpoint causes of disease and tailor treatments to a patient’s individual genetic makeup. Unfortunately, sequencing a patient’s genome and searching for every mutation known to cause a disease is impractical and expensive.
The costs of genetic testing are going down, but for now doctors need to know when it makes the most sense to screen for diseases with known genetic causes. In some cases, such as testing for the BRCA1 mutation that drastically increases a woman’s risk for breast cancer, the decision is simple. For others, where finding a genetic mutation offers opportunities for improvements to existing treatments or quality of life, it’s a little more complicated.
Rochelle Naylor, MD, a pediatric endocrinologist at the University of Chicago, studies a disease that makes for one of those complicated decisions: maturity-onset diabetes of the young, or MODY. It’s the most prevalent form of monogenic diabetes, which is caused by mutations in any one of about 20 different genes.
When correctly diagnosed, patients with certain types of monogenic diabetes can often take relatively inexpensive pills called sulfonylureas to manage blood glucose levels instead of relying on an insulin pump or injections. That can lead to a better quality of life and fewer diabetes-related complications.
But all forms of monogenic diabetes account for only 1-2 percent of diabetes cases, and it can be difficult to distinguish MODY from the far more common type 1 or type 2. So when does it make sense to screen for MODY?“The question is, does it save enough money over time to make up the initial cost of that genetic testing?” Naylor said. “If you test every diabetes patient when there’s less than 2 percent prevalence of MODY, then you’re taking the cost of testing all those people to pick up a few. Can you recover that cost?”
To find out, she and her colleagues built a statistical model to simulate the cost-effectiveness of screening for MODY in patients with type 2 diabetes. The model, published in the American Diabetes Association’s journal Diabetes Care, included hypothetical patients with true type 2 diabetes and patients with three genetic forms of MODY. They then calculated the cost savings of using cheaper sulfonylureas instead of insulin, along with the improvements to quality of life assuming better blood glucose control and fewer complications, such as renal failure and cardiovascular problems.
They found that in a population with a 6 percent prevalence of MODY, a policy of testing all type 2 patients for MODY would produce enough quality of life improvements for enough patients to make up for the screening costs. If MODY prevalence were greater than 30 percent, widespread testing would actually save costs.
But don’t all forms of monogenic diabetes account for less than 2 percent of diabetes cases? How do you find a group of patients big enough to hit that 6 percent threshold?
Naylor said you could start with patients who fit the classic phenotype for MODY: type 1 patients who still produce insulin long after initial diagnosis, young type 2 patients who are not significantly overweight or don’t display signs of insulin resistance, and those with certain family patterns of the disease.
This phenotype accounts for 50 percent of MODY patients, so testing those patients makes a lot of sense. But that also misses another 50 percent of MODY patients who don’t fit the classic picture.
Naylor said the decision to screen for MODY in the remainder of these less-obvious cases often comes down to the up-front cost of genetic testing. She co-manages the Kovler Diabetes Center’s MODY Registry, a database to track patients on a long-term basis to learn more about symptoms and optimal treatments. As she and her colleagues learn more, she said the decision to screen for MODY in the right situations would become easier.
“What our long-term follow up says so far is that patients do much better on sulfonylureas, so it actually does matter because their control and rates of complications are better,” she said. “But I think the education for providers is the big second piece. People have to know that the right treatment matters, and in the long term it makes financial sense.”
Naylor R.N., John P.M., Winn A.N., Carmody D., Greeley S.A.W., Philipson L.H., Bell G.I. & Huang E.S. (2013). The Cost-Effectiveness of Maturity-Onset Diabetes of the Young Genetic Testing – Translating Genomic Advances into Practical Health Applications., Diabetes care, PMID: 24026547