NORWALK — As a clinical researcher who spends most of his time at his oncology practice, Dr. Richard Frank has few conversations more agonizing than those with his pancreatic cancer patients.

They ask how long they have. Frank tells them there’s no reason to despair, because remarkable things happen when people have hope.

But the numbers are against them. The typical prognosis for late-stage pancreatic cancer is six to 12 months, although some patients live two to three years. One person in 10 might survive for five years.

The patient invariably asks whether the cancer could have been caught sooner.

While the answer for many other types of cancer is increasingly yes, the answer for pancreatic cancer continues to be no.

The difficulty of detecting the disease early is one reason it kills 43,000 Americans each year — more than breast cancer.

“I am a medical oncologist, and I watch people with pancreatic cancer generally die,” says Frank, the director of clinical cancer research for the health network that operates the hospitals in Danbury, Norwalk and New Milford. “It’s unacceptable, right?”

Frank’s frustration that medical science has not developed a way to screen for pancreatic cancer is behind an ambitious $2.7 million clinical trial he launched earlier this month to develop an early detection test that can save lives.

The three-year trial, which seeks to enroll 800 people, will study the relationship between pancreatic cancer and a far more prevalent disease of the pancreas — diabetes.

“Fifty percent of individuals who get pancreatic cancer develop diabetes in the two to three years before they are diagnosed with pancreatic cancer,” Frank said during an interview last week at the Western Connecticut Health Network’s Biomedical Research Institute in Danbury.

“The cancer is doing something in the pancreas to cause the diabetes.”

Frank believes that a specific kind of diabetes known as Type 2 is a symptom of pancreatic cancer.

To test his hypothesis, he needs many hundreds of blood samples and MRIs from patients with Type 2 diabetes or with a family history of pancreatic cancer.

In the end, Frank hopes to come up with a blood test much like the PSA test for prostate cancer, a disease that kills 26,700 men annually in America.

The stakes for Frank’s trial are high: Pancreatic cancer is on track to become the second-leading cause of cancer death, after lung cancer, by 2020.

Although the trial has launched with 30 participants, its success depends on Frank’s ability to enroll hundreds more over the next year. Success will also depend on his ability to win grants and contributions from philanthropists.

A website with details about participation requirements and donation information is at http://bit.ly/2zbiGr5.

Already Frank’s prevention study has raised $2 million in private money, thanks to a boost in May from a Weston family that raised $1.2 million in honor of Pamela Naughton, who died of prostate cancer.

Her husband, Tony Ward-winning actor James Naughton, said in a video on the prevention study’s website that he will never forget the day he got the call from his wife.

“She said, “They found a mass in my pancreas,’ and I will never forget that call or that feeling,” Naughton said. “It was like taking an arrow right to your heart, because I knew that was a really bad thing to get.”

Early detection

Some physicians use MRIs or ultrasound to look for pancreatic cancer, but there is no standard test, and screenings are far from routine, Frank says.

But more advances are made in cancer treatment than in any other medical field, says Frank, who did research at Memorial Sloan Kettering Cancer Center in Manhattan before joining Norwalk Hospital 17 years ago.

So why isn’t pancreatic cancer treatment advancing?

“The discoveries in genomics and the immunotherapy of cancer are not impacting the treatment of pancreatic cancer,” Frank said.

“It spreads very early, and it is resistant to the treatments that we have ... so we are not going to save lives by finding a better chemotherapy — even though we need to do that.”

The future of pancreatic cancer treatment, Frank says, is prevention through early detection.

Frank, a father of two who grew up on Long Island, knows that it is unusual for an oncologist to be running a prevention study.

His boss has full confidence in him.

“He is both a physician and a scientist — and not many physicians are capable of doing both well,” said John Murphy, the president and CEO of Western Connecticut Health Network, in a recorded statement. “He is a deeply compassionate man.”

Frank plans to expand the scope of his trial to as many at 5,000 patients across the tri-state area by forming research partnerships with teaching universities.

“This is the only study of its kind in the world, and it has lofty goals,” Frank said. “It could save the lives of a significant percentage of people who were destined to get pancreatic cancer.”