Statistician crunches numbers for better cancer treatment outcomes

January 29, 2014 by Kathleen Haughney

Florida State University Associate Professor Jinfeng Zhang and his graduate student Kaixian Yu aren't your typical cancer researchers.

For one, they're not doctors. They are statisticians. And rather than looking at cancer cells in a lab, they are analyzing numbers representing more than 1,000 breast , their genetic markers and their response rates to different cancer treatments.

Zhang and Yu's unique approach has led to the creation of a statistical program that personalizes chemotherapy treatments based on genetics, resulting in more favorable outcomes for each patient.

The program's initial results were so promising that Florida State University's Office of Commercialization awarded Zhang $50,000 through its annual GAP competition, a grant program that is designed to help bridge the gap between a faculty member's research and the commercial sphere for product development.

"I always wanted to do research that had a higher impact," said Zhang, an associate professor in the Department of Statistics. "All the time we hear of relatives or friends dying from cancer."

The initial statistical analysis showed that if the cancer treatments had been tailored to the patient, the response rate would have risen from 21 percent to 39 percent.

And it all came from crunching data.

"Every digit, every single one is a patient," Yu said. "It makes me feel really excited to be working on this. Science can be really exciting. It can help a human life."

According to the American Cancer Society, more than 1.6 million new cancer cases were projected for 2012—final numbers for the year are not yet available—and of those, 22 percent would receive chemotherapy.

But, the response rate of chemotherapy is less than 30 percent.

Cancer treatment is tricky. Chemotherapy, while having many benefits, is extremely toxic and comes with painful side effects. Sometimes, patients are given too much chemotherapy and cannot handle the toxicity.

Zhang said he had spoken with doctors in the area and other researchers at Florida State who acknowledged there was a need to take the large volumes of available data—so called big data—on cancer patients and analyze them to see if there was a way to improve the response rate.

But doctors, though trained to diagnose and treat, are not statisticians. That's where Zhang came in with his plan.

Admittedly, it was not smooth sailing at first.

Originally, Zhang, with the help of Yu, designed a program in May 2013 to examine whether the patients should have received chemotherapy at all. But the results were inconclusive and in August they went back to the drawing board and changed the approach to see if they could improve response rates by matching specific chemotherapy treatments to individual patients.

"You have to have faith that you will discover something," Zhang said.

After they had designed the statistical model, Yu set to work crunching the numbers.

This time, the numbers went their way.

"I pretty much did a dance every time we got a good result," Yu said.

The GAP award will help Zhang collect more data and do additional studies on the potential efficacy of his new system. The money will also help him get the system out into the marketplace so that patients can start benefiting from it.

Zhang said he hopes to do more studies and also look at different types of cancer. This first study focused solely on .

For now, Zhang and Yu are logging their results in a paper to be submitted for publication this spring and developing the statistical program so it could be used for other researchers in the future.

Zhang has filed a patent application for his design and also formed a company, Innomedicine LLC, with hopes of marketing the new method, called PERS (Personalized Regimen Selection), to patients around the country.

The research must undergo regulatory requirements before it can be sold to hospitals, so when it will be ready for commercial use is uncertain.

Explore further: Surgery beats chemotherapy for tongue cancer, study finds

Related Stories

Surgery beats chemotherapy for tongue cancer, study finds

December 26, 2013

Patients with tongue cancer who started their treatment with a course of chemotherapy fared significantly worse than patients who received surgery first, according to a new study from researchers at the University of Michigan ...

Classification system for bladder cancer prognosis

January 29, 2014

Researchers at Lund University have developed a classification system to determine the prognosis for bladder cancer. It is hoped that this will prove useful for future bladder cancer research and drug development.

Blacks have less access to cancer specialists, treatment

November 19, 2013

Researchers at the University of California, San Diego School of Medicine say metastatic colorectal cancer patients of African-American descent are less likely to be seen by cancer specialists or receive cancer treatments. ...

Recommended for you

Ancient stress response provides clues to cancer resistance

April 25, 2017

Cancer is often able to craftily outwit the best techniques modern medicine has developed to treat it. In an attempt to understand and combat cancer's vaunted prowess, an unusual collaboration between physicists and a leading ...


Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.