Blood test predicts colon cancer patients’ response to chemotherapy pill

University of Southern California study shows testing for genetic variations can steer patients to drugs with the best chance for effectiveness

SAN FRANCISCO, May 13—Colorectal cancer patients’ own genetic blueprint can show whether a promising chemotherapy pill is likely to help them—before ever starting treatment, according to a study by USC/Norris Comprehensive Cancer Center researchers presented today at the American Society of Clinical Oncology annual meeting.

The researchers showed that patients with colorectal cancer who have a variation in a key gene respond best to the anti-cancer pill Xeloda (capecitabine), which was just approved by the Food and Drug Administration as a first-line therapy against colon cancer last month.

According to research oncologist Heinz-Josef Lenz, M.D., associate professor of medicine at the Keck School of Medicine of the University of Southern California, the findings are a pioneering first step toward customizing the most effective therapies for each patient diagnosed with colon cancer.

"Although further study is needed, these data indicate that identifying patients with specific genetic characteristics may dramatically improve treatment outcomes for some people with advanced colon cancer," Lenz says.

A simple blood test before treatment may right away steer the right patients toward Xeloda, Lenz says, instead of other therapies less likely to help. In a phase II clinical study of 22 patients with advanced colorectal cancer, he says, 80 percent of the patients with the specific genetic twist responded to Xeloda. Of those without the genetic variation, 10 to 14 percent responded to Xeloda.

The critical chain of events works like this:

Human cells have a gene that gives orders to produce a substance called thymidylate synthase, or TS. TS is a key enzyme needed to make DNA. Unfortunately, cancerous tumor cells can feed off the TS to reproduce their own DNA.

When a patient takes Xeloda, though, the tumors produce an enzyme that converts the Xeloda into fluororacil, a potent cancer-blasting substance. (Since colorectal tumors produce more of this enzyme than healthy tissue does, Xeloda does its work preferentially at the site of the tumor—making the drug unique.)

Fluororacil then breaks down in the body, and ideally, its product attaches itself to TS, making the TS useless to tumor cells. But in practice, the drug’s effectiveness varies from patient to patient and population to population.

The USC researchers, though, have found the genetic twist that makes the drug work better or worse. The TS gene regularly comes in two forms, or polymorphisms. In one form, a string of 28 base pairs (the building blocks of DNA) in the gene repeats itself two times. In the other form, the string repeats itself three times.

Each person has a set of two TS genes. So, the two genes in a set can come in an identical pair (either both double repeaters or triple repeaters) or a patient can have one of each type of gene (a double and a triple). In total, three combinations are possible.

The researchers found that patients who had a pair of genes that were both double repeaters responded best to Xeloda. Response means that tumors shrank by half or more.

The cells in these patients expressed fewer signals to produce TS, Lenz explains.

Using knowledge about each patient’s genetic makeup, physicians in the future may be able to recommend alternate medicines that work differently to patients who would not likely respond to Xeloda, Lenz says.

Results of another study by the USC/Norris team describing the relationship between genetic variations and the effectiveness of the drug 5-FU (fluororacil) against colon cancer are scheduled to be published in The Pharmacogenomics Journal in late May.

Colorectal cancer is the third most common cancer in the United States, and experts expect 130,000 new cases to be diagnosed this year. The five-year survival rate for patients with advanced or metastatic disease is less than 10 percent.