{shortcode-2cea3959415a50feee9b67916b98d3d6e44d3e37}

A recent study published by Harvard scientists examined brain tumors in unprecedented detail and found that brain tumor cells shift between distinct states, making them more resilient to cancer treatment.

The study, published in Nature Genetics last month, looked at individual gliomas — the most common type of brain tumors — in unprecedented detail, offering new insights about their classification and genetic architecture.

Researchers from Harvard Medical School, Weill Cornell Medicine, Massachusetts General Hospital, the New York Genome Center, and the Broad Institute of MIT and Harvard co-authored the journal article.

Dana Silverbush, a postdoctoral fellow in the Suvà Laboratory at Harvard-affiliated MGH and co-first author of the study, said the researchers found how DNA methylation — cellular marking events that affect gene expression — causes brain tumor cells to change states.

“There are different cell states within the tumor and they probably respond differently to treatments, and we wanted to see whether we could look at the DNA methylation of the tumors — their heterogeneous DNA methylation states,” Silverbush said. “Because the DNA methylation are events that can dictate the cell state, they can dictate how the cells respond to treatment.”

According to Silverbush, the team found that specific “subpopulations” of the glioma cells had different DNA methylation “signatures,” which explains why different subpopulations within the heterogeneous tumors respond differently to treatment and why the whole tumor can survive despite treatment.

“Stem-like” tumor cells targeted by stem-cell-killing treatments, for example, might be regenerated from the tumor’s pool of more mature cells.

Silverbush added that this kind of analysis had been done “in bulk” before with whole brain tumors, but never before “in single-cell resolution.”

Mario L. Suvà, associate professor of Pathology at MGH and co-senior author of the study, wrote in an email that their study is the first to apply “multi-omics” to single brain tumor cells. “Multi-omics” is an analytic approach that draws from aspects of multiple “omes.” In this study, the group analyzed gene sequencing and transcription within the genome and DNA methylation within the epigenome.

The group sampled the individual cells from patients with IDH-wildtype glioblastoma — an aggressive type of brain tumor that is highly resistant to treatment — and patients with IDH-mutant glioma — a less aggressive type of brain tumor that progresses more slowly.

Silverbush also said that the DNA methylations the group studied can also be viewed as a molecular “clock” leaving “time marks,” which enables the researchers to track a tumor’s evolution.

Silverbush said that the evolutionary analysis revealed the glioblastomas have more “plasticity,” allowing them to “escape” treatments.

“Glioblastomas showed a lot more plasticity, so the cells can easily change their states, which makes them more aggressive, because if you treat the tumor with a certain treatment, it can escape this treatment and adapt to a different state,” Silverbush said.

Suvà said that a key implication of the study for cancer treatment is that the “molecular switches for cell states could serve as targets” for treatment.

Silverbush also said that future research should explore whether methylation events could serve as “markers” to find the “correct, optimal window of interference.”

“These could be markers for when it’s best to intervene because it’s possible that at some point an event happens that makes the tumor less likely to respond to a certain treatment,” Silverbush said. “So this would not be to develop a new treatment but more to know when to use existing treatments.”

—Staff writer Justin Lee can be reached at justin.lee@thecrimson.com.