(ChemotherapyAdvisor) - The appearance of new genetic mutations during cancer therapy has long been correlated with drug resistance, treatment failure, and ultimately, relapse. But using single-cell genome sequencing, researchers at the Ontario Cancer Institute and Princess Margaret Cancer Center in Toronto, Canada have shown that gene mutations alone cannot explain drug-resistant cancer.
Tracking individual human colorectal tumor subclone cells that had been xenografted into mice, and sequencing their exomes (the gene-encoding regions of subclone genomes), the team discovered dramatic functional heterogeneity even among genetically identical clones; these included different tumor propagation patterns and different susceptibilities to oxaliplatin, which was reported by the the researchers in Science.
The findings represent “a major conceptual advance in understanding tumor growth and treatment response,” said senior author John Dick, PhD, a pioneer in cancer stem cell research. “The data show that gene sequencing of tumors to find the spectrum of their mutations is definitely not the whole story when it comes to determining which therapies will be most effective.”
While some cells of a subclone contributed to tumor growth, others quickly became dormant, even though they harbored the same mutations as the more active cells, Dr. Dick's team found—and these dormant cells survived oxaliplatin therapy.
“This is a paradigm shift that shows research also needs to focus on the biological properties of cells,” Dr. Dick said. Treatments that force dormant cells back into growth cycles could make them more sensitive to chemotherapies, he theorizes.
“Targeting the biology and growth properties of cancer cells could expand the repertoire of usable therapeutic agents and provide better outcomes for patients,” he added.
The study is “avante garde in its documentation that types of subcloning can happen against a stability of genetic changes—a clone within a given set of genetic changes can evolve into subpopulations within that clone without changing their genetic background, their mutations,” said Stephen Baylin, MD, Professor of Oncology and Deputy Director of the Cancer Center at the Johns Hopkins University School of Medicine.
Epigenetic factors, such as different DNA-methylation patterns that can silence gene expression, might help explain behavioral heterogeneity among genetically identical subclones, Dr. Baylin postulates.
“If you have got such genetic stability, then it's likely that the other facets of the subclones that emerged could have an epigenetic basis—long-term changes in gene expression,” he told ChemotherapyAdvisor. “Things like epigenetic abnormalities could be contributing to the emergence of new subclones with distinct properties.”
The new study “emphasizes those possibilities,” he said. When dormant tumor cells “come out and replenish the tumor,” other studies have shown that they do so with a “different epigenetic state” that appears to contribute to their drug resistance, he noted.
The traditional paradigm, with new mutations causing some tumor cells' drug resistance, is not completely wrong, Dr. Baylin is quick to point out. “That can happen,” he said of mutation-driven resistance. However, Baylin believes,the new findings reported by Dr. Dick and his colleagues strongly suggest epigenetics is another “big player” in drug resistance.
Epigenetics-targeting drugs already exist, he notes. For example, azacitidine (5-azacytidine) and decitabine (5-aza-2′deoxycytidine) inhibit DNA methylation and are approved by the FDA for myelodysplastic syndrome. Histone deacetylase inhibitors might also target epigenetic pathways in tumors.
At low doses, ongoing studies in Dr. Baylin's lab suggest that azacitadine “sensitizes patients to subsequent chemotherapies or a new form of immunotherapy,” he said.
The findings reported by Dr. Dick's team indeed suggest that nongenetic targets for personalized anticancer agents are waiting to be identified, agrees Charis Eng, MD, PhD, FACP. Candidate targets include both epigenetic alterations and tumor microenvironments (healthy cells adjacent to tumors), she said.
Dr. Eng is the Hardis and American Cancer Society Professor and founding Chair of the Genomic Medicine Institute and directs the Institute's clinical component, the Center for Personalized Genetic Healthcare at the Cleveland Clinic. She believes that even though tumors' subclone gene mutations were identical from cell to cell, their functional genomics—“the ways they interact with each other, the ways they signal and make transcripts”—might still be quite variable.
Gene mutations, in other words, are just one part of a larger puzzle. Epigenetics, proteomics, even microbiomes, or the genomes of bacteria living on epithelial tissues in which tumors emerge, may all help explain why some subclone cells go dormant and evade chemotherapeutic attacks, while others succumb to treatment, she believes.
“Genomic changes are like the skeleton. The genome is the skeleton and everything else—the methylation, microenvironment, the microbiome—will be the flesh, the meat, the muscle and the skin,” Dr. Eng explained to ChemotherapyAdvisor. “So a look at everything, a snapshot profile of all the –omes, or what I call ‘integrated –omics,' which is the strength of my lab, integrates all the –omic platforms to see whether we can come up with an integrated view of what a cancer looks like.”
Even the Cancer Genome Atlas Project has added RNA and epigenetic assays to its profiling of tumor genomes, she notes.
The “sum total of the integration of all the ‘-omes' from all the cancer cells,” rather than any one component, may dictate chemotherapy responses in many cancers, she suspects. If that's the case, gene mutation-targeting drugs could one day represent just one part of clinical oncology's personalized-therapy arsenal.
“I have a funny feeling that even the three-dimensional positioning of the cells (within tumors) and how they talk to each other—whether by message or protein or exchanging genes—also matters,” Dr. Eng said.
But she is quick to point out that the Science study involved xenografting human tumors into mice. Even though that model worked elegantly to show that subclone heterogeneity is not attributable to genetic mutations alone, it may not be the best way to find out exactly what else is responsible, she cautioned—especially if tumor microenvironments are involved.
“When the microenvironment is not represented well, alterations in the microenvironment might be missed,” she explained. “We have to ask: what is the interaction of each of these subclones with the mouse environment?”
Taking human tumors out of their human microenvironmental context might itself “have effects on different expressions of genes in the cancer,” Dr. Eng noted.
“We'd been assuming the enemy was simple-minded,” she said. “The enemy is complex. We need a multidisciplinary approach to look at the DNA and to understand the microenvironment, how it turns genes off and on in different contexts and even in different (cell) positions within tumors.”
Dr. Baylin agrees that the new findings will open doors to new avenues of research.
“We need to work on extending these observations to other tumors and to really keep studying the mechanisms that account for how these subclones emerge,” he said. “And then we need to correlate those changes really carefully with drug resistance so we can understand the molecular underpinnings in resistance patterns, and learn how to tailor therapeutic approaches to those molecular mechanisms.”
Reference: Kreso A, O'Brien CA, van Galen P, et al. Variable clonal repopulation dynamics influence chemotherapy responses in colorectal cancer. Science.
http://www.sciencemag.org/content/ea...b-46531083d53e