Thick mucus that can drown the lungs of a child has long been the hallmark of cystic fibrosis. The hereditary disease affects 30,000 Americans, and patients die unless they receive treatment to clear their lungs. But new research suggests that this pulmonary view of cystic fibrosis is only half of the picture: a suite of symptoms associated with cystic fibrosis can also occur in patients who do not have lung disease at all, indicating that cystic fibrosis is really two diseases. This second version, it appears, causes pancreatitis.

"Cystic fibrosis has been evaluated and managed by pulmonary doctors focusing on the lung, but other important problems are never seen by the pulmonologist and nobody's put the pieces together," says David Whitcomb of the University of Pittsburgh, who studies disorders of the pancreas.

Cystic fibrosis results from mutations in a gene that produces a tube-shaped protein known as CFTR, essential to the balance of electrolytes in the body. Specifically, this protein allows chloride ions to pass in and out of cells. When it malfunctions in classic cystic fibrosis, cells in the airway cannot produce normal mucus but instead make a thicker, stickier substance that clogs the lungs.

But CFTR leads a double life. Whitcomb's team screened a group of nearly 1,000 patients with pancreatitis and found nine abnormal but supposedly harmless versions of the CFTR gene. Their study suggests that the seemingly benign mutations break the switch that turns CFTR from a chloride portal to a channel for bicarbonate, a chemical that the pancreas produces to neutralize stomach acid. Patients with these mutations do not have the problems associated with the chloride channel, but the faulty bicarbonate channel means that they can suffer from painful pancreatitis, as well as sinusitis and, in men, infertility. Computer simulations confirmed that the mutations are all in places that would inhibit bicarbonate but not chloride from passing through.

Without the ability to secrete bicarbonate, Whitcomb says, patients cannot flush digestive enzymes out of their pancreas and the pancreas essentially dissolves itself, a horrifically painful condition. Other organs also depend on the bicarbonate channel: cells in the sinuses use it to produce the right consistency of mucus, and it is essential for pH-balancing semen. The CFTR mutations that only affect bicarbonate thus cause a recognizable syndrome that combines the symptoms of pancreatitis, sinusitis and male infertility. Meanwhile, since the chloride-control function of the channel is unaffected, these patients pass the sweat chloride test—the standard for diagnosing cystic fibrosis—with flying colors.

Drugs that help cystic fibrosis patients may also relieve this form of pancreatitis. If the number of patients with the syndrome turns out to be large—Whitcomb suspects it is more than the number with classic cystic fibrosis—then the market for those drugs will grow, and the drugs themselves could become more affordable. The cystic fibrosis drug Kalydeco, for example, currently costs more than $300,000 per year.

Julie Forman-Kay, a biochemist at the Hospital for Sick Children in Toronto, notes that the techniques the researchers used to figure out the details of how each mutation changes the protein are "extremely challenging" and "kind of an art form," and more work is needed to confirm that the mutations actually cause the changes that the computer model predicts. "The real potential impact of this paper," she says, "is that it kind of wakes up the research community a bit and pushes for expanding understanding of the role of CFTR in other diseases besides cystic fibrosis."

In a sense, cystic fibrosis is returning to its roots. It was originally named "cystic fibrosis of the pancreas," and children with the disease experienced pancreatic failure in infancy and died within their first few years of life. Once pills were available to replace the digestive enzymes they lost, those children were able to grow up—the average life expectancy is now 40 years—and the focus of treatment shifted to the lungs.

Whitcomb's eventual goal is to disentangle the distinct causes of what, until recently, appeared to be a single disease. "Chronic pancreatitis was considered to be a total enigma," Whitcomb says, with 42 percent of cases having no known cause, "but we're finding that it's five or six or more different diseases that all look the same on CAT scans. What we're able to do now is unravel that mystery on a case-by-case basis."

CFTR-caused pancreatitis is one of those variants. Another, which Whitcomb's group identified over a decade ago, is caused when a digestive enzyme, trypsin, is activated at the wrong time and digests the pancreas from within. "Modern medicine is built on the germ theory of disease, that one factor will cause a complex disorder," Whitcomb says, but personalized medicine is showing that many disorders have different causes in different patients. As the differences are untangled, Whitcomb says, "the percentage of patients that are mystery patients is getting smaller and smaller."