Deeper Look at Tumor’s Genetic Mutations Challenges Current Precision Medicine Approach
WASHINGTON (December 5, 2019) — The number of genetic mutations in cancerous tumors appears to be significantly higher than previously thought, according to a new study using an advanced technology that can detect genetic mutations present only in rare cells within tumors. The finding incorporates a new mathematical approach to calculating total unique mutations in a tumor from sequencing a limited number of DNA molecules.
The study finds that at diagnosis, each patient’s tumor (minimally about a billion cells) likely contains at least one cell with nearly every possible mutation, possibly explaining why advanced cancer is so difficult to treat, even with current precision medicine methods.
“This suggests that resistance to any single therapy always exists before treatment starts,” says the study’s senior author, Robert A. Beckman, MD, a professor in the departments of oncology, and biostatistics, bioinformatics, and biomathematics at Georgetown University Medical Center. Lawrence A. Loeb, MD, PhD, a professor of pathology and biochemistry at the University of Washington, is that paper’s first author.
Previously, pre-existing resistance was thought to be frequent, not universal, says Beckman. The finding, published online December 5 in Proceedings of the National Academy of Sciences, PNAS, suggests a renewed look at precision medicine treatment strategies based solely on the majority of cells in a diagnostic biopsy, rather than also looking at rare cells, he adds.
“Further, as tumors grow, the likelihood that a single cell can become simultaneously resistant to more than one treatment increases,” Beckman says. “Eliminating these rare cells before they evolve resistance to additional treatments may be more important than shrinking the tumor at times.”
Treatment can destroy much of a tumor’s mass, but cancer cells that are genetically resistant to one or more therapies survive and repopulate a tumor.
Beckman says the current precision treatment approach based on average measurements from a diagnostic biopsy can often be defeated by rare cells with different mutations and by continued tumor evolution. As the tumor grows, some rare cells acquire simultaneous resistance to all the therapies that make up a combination therapy, the study shows.
“We should start thinking about cancer treatment as an adaptive game of chess, thinking multiple steps ahead when we move our pieces,” Beckman says. “Precision medicine treatments should change frequently as a tumor continues to evolve, even when a patient is doing well based on tumor size.”
For this study, Beckman, a member of Georgetown Lombardi Comprehensive Cancer Center and Georgetown Innovation Center for Biomedical Informatics, used mathematical models he developed to understand the genetic mutations uncovered by the University of Washington team. Led by Loeb, the group developed a novel technique, duplex sequencing, which can detect genetic mutations that are present only in rare cells within tumors with high accuracy.
The researchers examined DNA from 20,000 cells within tumors of 15 colorectal patients, searching for mutations, called subclonal, present in only a minority of the cancer cells, and believed to be the source for the eventual emergence of therapeutic resistance. Duplex sequencing was able to see if one cell out of 20,000 was genetically different from all the other tumor cells, Beckman says. They found many more unique mutations than expected based on previous results with less sensitive techniques.
Each tumor would have contained at least a billion cells at diagnosis. Beckman developed mathematical methods to estimate what might have been seen if DNA from all the tumor cells had been examined. He found cancers acquire much more diversity as they grow than previously thought.
“No one has ever looked so deeply at rare cells within a mixture from a colorectal tumor,” he says. “In the past, we were looking only at the tip of the genetic diversity iceberg.”
The researchers sequenced the same tumors at different depths to see how the subclonal genetic diversity was progressively revealed by examination of rarer cells.
Beckman describes the genetic growth of a tumor as a tree. The trunk consists of the original cell that led to development of the tumor. The branches and leaves represent progressive cell divisions and evolving genetic diversity in rarer cells. “Because of the branching nature of tumor evolution, being able to detect rare mutations lets us see further forward in time from the birth of the first tumor cell,” he says.
“We now know each new cell has about 2,000 genetic changes compared to its parent, so a future approach to precision medicine would dynamically use all available information from the patient and others with more advanced cancer, to quantify the chance of current or future mutations and design a treatment strategy that considers both present and future,” Beckman says.
In addition to Loeb and Beckman, authors include Brendan F. Kohrn, Kaitlyn J. Loubet-Senear, Yasmin J. Dunn, and Eun Hyun Ahn of the University of Washington; Jacintha N. O’Sullivan of University College Dublin; Jesse J. Salk, of the University of Washington and TwinStrand Biosciences in Seattle; and Mary P. Bronner of the University of Utah.
Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health (NCI P01-CA077852, NCI R01-CA160674).
The following personal financial interests were disclosed: Loeb and the University of Washington have a license agreement with TwinStrand Biosciences for the use and development of duplex sequencing technology. Loeb is a cofounder of TwinStrand Biosciences. Loeb is a member of the Scientific Advisor Board of Stratos Genomics Inc. Beckman is a stockholder in Johnson and Johnson, is/was a consultant for EMD Serono, Zymeworks, CStone, Vertex and AstraZeneca, and is chief scientific officer for Onco-Mind.