Almost 95% of pancreatic cancer patients die within 5 years of diagnosis, and traditional chemotherapy does little to save their lives. Now, cancer researchers think they know why--and how they might get around the tumor's defenses. The work "has huge ramifications for how we approach therapy of this disease," says Margaret Tempero, an oncologist at the University of California, San Francisco.
Cancers almost always grow in the duct system of the pancreas, a spongy handful of tissue nestled against the stomach that produces digestive enzymes and hormones such as insulin. Although these tumors are deadly when they spread to other parts of the body, malignant cells are surprisingly sparse in the pancreas. Biopsies typically turn up a few scattered cancer cells amid masses of stroma, webs of tough, fibrous connective tissue that look like a "big piece of gristle," says oncologist David Tuveson of the Cambridge Research Institute in the United Kingdom.
That gristle, Tuveson and his colleagues found, shields pancreatic tumors from blood-transported treatments like chemotherapy. Using mice genetically engineered to develop pancreatic cancer, the team mapped the blood supply of tumors using chemical tracers to mark blood flow. They found very few blood vessels and just a third as much circulation as in normal tissue. Human pancreatic cancer samples told a similar story: lots of fibrous tissue, few blood vessels.
If his team could shrink the stroma, Tuveson wondered, could doctors target the cancer cells hiding inside? The group focused on the whimsically named hedgehog signaling pathway, which helps organize the structure of organisms during embryonic development and promotes the growth of stroma. The researchers treated mice with IPI-926, a drug that suppresses the hedgehog pathway. Other mice got a common chemotherapy drug, whereas yet another group got IPI-926 and the chemo drug.
As the researchers had hoped, IPI-926 dramatically shrank the stroma. And there was a bonus: The drug also increased the number of blood vessels inside the tumors by threefold to fourfold. Feeding a tumor with extra blood might normally promote cancer growth, but in the case of pancreatic cancer, it was good for drug delivery; mice given a combination of chemo and IPI-926 nearly doubled their survival, from a median of 11 days to 25 days, the researchers report online today in Science. It's not a cure, Tuveson cautions, but the results leave him optimistic. "If we were to double the survival of pancreas cancer patients that had advanced disease, instead of dying in 6 months, they'd die in 12 months," he says. Compared with previous medical interventions, "that would be a bigger impact than we've ever seen."
Although other factors besides blood flow might affect chemotherapy resistance, the study "demonstrates pretty nicely that the cancer cell is not the only thing we need to be worried about," says oncologist James Abbruzzese of the M. D. Anderson Cancer Center in Houston, Texas. "We really should be paying more attention to these cancer-stroma interactions." The next step, he and Tempero say, is to find out if IPI-926 has any potential as a drug for humans and if it is safe to use on tumors outside the pancreas.