Health & Medicine / Cancer

Despite intense investigation over the past decade, the relationship between diabetes and pancreatic cancer remains poorly understood. Many patients with pancreatic cancer develop high blood sugar, or hyperglycemia, which can progress to full-fledged diabetes. The reasons for this remain unknown. One hypothesis is that tumors of the pancreas destroy enough of the insulin-producing islet cells in the pancreas to cause diabetes. Another possibility is that pancreatic tumors somehow make patients become insensitive to insulin. The basis for this hypothesis is that patients often regain sensitivity to insulin following the removal of the tumor.

Some physicians have advocated studying patients with new-onset diabetes as a population at high risk for pancreatic cancer. They have proposed screening these patients with special blood tests, ultrasound exams, and CT scans. However, there are reasons to be cautious. More than 1 million new cases of diabetes are diagnosed in the United States each year, but pancreatic cancer is comparatively rare, with only about 30,000 new cases in the United States during the same period of time. Testing everyone who developed diabetes in hopes of finding a small number of patients with pancreatic cancer would be prohibitively costly.

Still, if the diabetes leads to the discovery of the cancer (and not the other way around), the cancer is more likely to be at a stage at which it can be successfully removed. Understanding this possibility is especially important because the overwhelming majority of patients have their pancreatic cancer diagnosed when the cancer has become incurable. Most patients die within six months of diagnosis. Usually the tumor has already spread to another organ, or the primary tumor has wrapped itself around one of the nearby major abdominal arteries, making surgical removal of the tumor too risky.

Later that day I got the results of the patient's CT scan. It showed fullness in her pancreas that could be a tumor. She was hospitalized for further testing.

The next morning, I performed an endoscopic ultrasound. After the patient was sedated, I inserted an endoscope with a tiny built-in ultrasound instrument through her mouth and into her stomach and intestines. The ultrasound let me look through the wall of her stomach and small intestine and see her pancreas in exquisite detail. What had appeared as fullness on the CT scan looked like a classic pancreatic cancer. To confirm the diagnosis, I inserted a needle through the endoscope to take a biopsy from the core of the mass. A pathologist standing by examined the tissue and verified that the cells from the mass were cancerous.

The tumor was surgically removed the following week.

My patient was lucky. If she had waited a few more months to get medical attention, she might have died from inoperable pancreatic cancer.

The relationship between diabetes and pancreatic cancer is still being explored, and doctors must rely on their clinical judgment and experience when deciding which patients with new-onset diabetes should be screened for pancreatic cancer. Yet, given how few tools exist for discovering patients with pancreatic cancer early in the course of the illness, any new way to identify potentially curable patients is a source of hope.

Douglas G. Adler is assistant professor of medicine at the University of Texas Medical School at Houston. The cases described in Vital Signs are true stories, but the authors have changed some details about the patients to protect their privacy.

Previous Vital Signs columns: