Cancer genetics gets personal

Researchers have developed a novel technique for identifying patient-specific biomarkers in tumor DNA which they say can reliably monitor the progression of individual patients' cancers. Their findings are presented this week at the American Association for the Advancement of Science meeting in San Diego and will be published next week in Science Translational Medicine.

Image: Courtesy of Life Technologies and Digizyme, Inc.

"This study pushes the limits of what we can do and what we might be able to do in the future with regards to cancer treatment," said linkurl:Arul Chinnaiyan,;http://www.pathology.med.umich.edu/faculty/Chinnaiyan/ a University of Michigan cancer geneticist who was not involved in the study. For years researchers have known that human cancer cells have scrambled genomes, but finding these genetic rearrangements amongst the hundreds of thousands of normal DNA molecules in a patient's blood or other tissue had proved difficult. Previous attempts looked for mutations, or single base changes, in the tumor genome, but finding such mutations was as tricky as finding a single typo in an entire book. In this study, researchers from Johns Hopkins Kimmel Cancer Center and Life Technologies instead looked for rearrangements of large chunks of DNA in the entire genome of four colorectal cancer patients and two breast cancer patients. The rearrangements they found, approximately 1 to 1.5 kb in size, only occurred in tumor samples and to their surprise, each individual's tumor had its own unique rearrangement. "We were looking for a translocation that would define breast cancer or colon cancer," said linkurl:Luis Diaz,;http://www.hopkinskimmelcancercenter.org/index.cfm/cID/1686/mpage/expertdata.cfm/expID/359 an oncologist at the Johns Hopkins Kimmel Cancer Center and a coauthor on the study. "But what we found was that each cancer tumor had a unique rearrangement that could be used as a biomarker. So basically, each individual's tumor is as unique as they are." "I would expect that's the nature of the cancer beast, that each one is different" said linkurl:Michael Snyder,;http://snyderlab.stanford.edu/ a geneticist at Stanford University who was not involved in the study. "I am a big believer that these high sequencing methods are going to be the way for understanding a lot of cancers...they really let you look at a level of detail you couldn't see before." Once the researchers found and isolated these rearranged pieces, they created a PCR-based blood test that could detect the biomarker DNA fragments. The test was sensitive enough to detect one tumor biomarker amongst a sea of normal DNA in a patient's blood, plasma or other tissue. Using this new approach, coined Personalized Analysis of Rearranged Ends, or PARE, the researchers tracked cancer progression of their six patients as they underwent treatment. When patients had surgery to remove the tumor or underwent chemotherapy, researchers found fewer rearranged fragments, and when the tumor grew back, the fragments' number increased. "The ebb and flow of the biomarker, which we tracked by searching for the DNA rearrangement in the blood, correlated perfectly with the clinical scenario," Diaz said. Currently more than half of cancer patients go through potentially curative surgical treatments where the tumor is excised. But did the surgeon removed the entire tumor, curing the patient, or inadvertently leave some stray tumor cells behind? "[T]here isn't any methodology to help answer this question," Diaz said. "So we are blind at that point as oncologists." To reduce the chances of recurrence, all patients receive additional treatment such as chemotherapy and radiation. "We would like to introduce our test at that point to help discriminate between the cured and non-cured," he said. While the ability to individualize treatment would be ideal, the feasibility is challenged by the exuberant cost. Doing paired end analysis using next generation sequencing costs roughly $5000, and takes two to three weeks. "It's almost like a mini research project that you would have to carry out for each patient," Chinnaiyan said. "Trying to scale that up for hundreds to thousands of patients could be an issue." But as the technology improves and demand increases, costs will continue to drop, Diaz said. He predicts PARE monitoring will be come more affordable in the future if they can increase demand by showing makes a difference for patients. "If we can tell patients after their surgery 'It looks like you're cured,' or 'You've got something left, we better treat it with intensive therapy,' that makes a big difference," he said.
**__Related stories:__***linkurl:A revolutionary approach to biomarker discovery;http://www.the-scientist.com/article/display/25357/
[11th January 2006]*linkurl:How to build a cancer sensor system;http://www.the-scientist.com/article/display/15734/
[26th September 2005]*linkurl:Reducing risks, maximizing impacts with cancer biomarkers;http://www.the-scientist.com/article/display/14535/
[15th March 2004]