Personalized Cancer Treatment
The Holy Grail of cancer treatment has always been a way to identify and understand the specific characteristics of an individual tumor. Such genetic information could, theoretically, make it possible to design treatments that were more successful and targeted--and might even avoid the current side effects associated with chemotherapy. Here is a report from Med Page on a recent study that moves us a little closer to that goal.
A measurement of circulating tumor DNA could lead to patient-specific tests for cancer, a preliminary study suggests.
Using advanced genomic sequencing techniques and polymerase chain reaction technology, researchers detected mutant DNA in patient plasma at levels <0.001%. Mutant DNA in plasma proved to be identical to genetic rearrangements found in tumor samples from the same patients.
Investigators believe the studies have laid the groundwork for development of personalized tumor biomarker assays. "We showed this approach was very sensitive and that we could detect even single rearranged tumor DNA molecules in the presence of hundreds of thousands of normal DNA molecules," Victor Velculescu, MD, PhD, of Johns Hopkins, said during a media briefing at the American Association for the Advancement of Science meeting in San Diego.
"These proof-in-principle experiments showed that the approach we created could provide a way for developing personalized biomarkers for virtually any cancer patient."
Chromosomal rearrangement has long been recognized as a universal feature of cancer. But until recently, scientists' ability to detect the changes had been limited by the technology of genomic sequencing. Now, scientific and technologic advances have dramatically increased the resolution of detection capabilities, Velculescu and colleagues wrote in an article for the Feb. 24 issue of Science Translational Medicine.
Tumor-specific chromosomal rearrangements offer the potential for development of highly sensitive biomarkers for tumor detection. Toward that end, Velculescu and colleagues have developed the personalized analysis of rearranged ends (PARE) approach to identify patient-specific rearrangements in tumor samples.
Conventional sequencing approaches have focused on single-letter DNA changes. The Hopkins group looked for rearrangements of entire sections of a tumor's genome. Velculescu compared the approach to identifying out-of-order chapters in a book as opposed to looking for an isolated typographical error. To test the strategy, investigators studied tumor specimens from four patients with colorectal cancer and two with breast cancer. They searched for regions of the tumor genome for areas that had too many or too few DNA copies and for sections of different chromosomes fused together.
The analysis identified an average of nine DNA rearrangements per specimen and a range of four to 17. "We found two general characteristics," said Velculescu. "Every cancer we analyzed had the rearrangements, and every rearrangement was unique and occurred in a different location in the genome." Velculescu and colleagues then evaluated blood samples from the two patients with colorectal cancer, looking for the same rearrangements found in the tumor specimens. Using DNA amplification techniques, they found rearrangements identical to those in the patients' tumor specimens.
Finally, the investigators applied the PARE approach to the evaluation of a patient at different times during the course of treatment.
The mutant DNA was isolated from all six blood samples obtained from the patient, but at varying concentrations. The concentration was highest before surgery, then decreased by more than 50% the day after surgery, and continued to decline after chemotherapy and resection of a liver metastasis. However, the mutant tumor DNA never disappeared entirely, consistent with residual liver metastases.
Velculescu cited two potential limitations of the PARE approach. Conceivably, some rearrangements could be lost during tumor progression, although identification of
several biomarkers in the same tumor would mitigate that concern. Cost is a second concern. Identification of patient-specific genomic alterations costs about $5,000 per patient. However, the cost of the type of sequencing required for the PARE approach has declined substantially in recent years and continues to decline, Velculescu said.