Time Bungles Precision Medicine

Personalized pancreatic cancer therapies based on tumor genomics may take too long to prepare to be helpful, according to a small clinical trial.

Apr 20, 2015
Kerry Grens

Micrograph of pancreatic ductal adenocarcinomaWIKIMEDIA, KGHFor patients with pancreatic cancer, the wait for a personalized treatment based on genomic data may be too long to be practical, according to a study presented today (April 20) at the American Association for Cancer Research (AACR) meeting in Philadelphia, and published concurrently in AACR’s Clinical Cancer Research. The experience reflects the practical challenges of precision medicine studies, said study coauthor Andrew Biankin, director of the Wolfson Wohl Cancer Research Centre at the University of Glasgow.

“The science is the easy part,” Biankin told The Scientist. “The science will work out. But the clinical systems just aren’t there to do this properly.”

As part of a small clinical study, Biankin and his colleagues had set out to offer pancreatic cancer patients personalized therapies by sequencing their tumors in search of a variety of genetic alterations that would dictate an appropriate drug. Outlooks for these patients are particularly bleak: 95 percent die within five years of diagnosis.

Although Biankin’s team received the genetic results roughly three weeks after obtaining consent from patients, none of the consenting patients eligible for personalized treatments were available to complete the study. Among the 22 patients who had a positive hit on the genetic screen, six had died, others had medical complications arise, and a few withdrew from the study.

Biankin said that a number of cultural barriers challenge progress in clinical trials for genomics-based cancer treatments. For one, his team experienced logistical complications. If tissue biopsies were chaperoned through processing by one of the study members, samples were prepared for genetic testing within days; otherwise, it would take months to get samples processed—months that many pancreatic cancer patients don’t have.

Pathology labs should be aware when a sample needs priority handling, said Biankin. Research “should be integrated into pathology work flows. It’s really an oncological emergency, and some of the best options for those patients are to be part of a clinical trial,” he told The Scientist.

Another challenge Biankin’s team encountered was a reluctance by physicians to do biopsies. Considering that pancreatic cancer patients have few treatment options available, biopsies—the results of which could open up opportunities for clinical trial participation—should be a priority, he said.

“It is important for the public to know how hard it is to put into practice molecularly guided treatment within the constraints of our health service delivery,” Lorraine Chantrill, Biankin’s collaborator and an oncologist at Campbelltown Hospital in Australia, said in a press release. “We hope that our work will help others who are planning similar studies.”