As a cancer spreads throughout a person’s body—bits of a primary tumor lodging themselves in new organs and spawning new tumors—the cancerous cells are constantly changing and evolving. Exactly when, and how fast, cancers spread—or metastasize—and how the timing of such metastasis affects survival and treatment plans has been a long-standing question among cancer biologists. Now, a team of researchers has come up with a way to answer some of these questions, by building a genetic family tree of the relatedness of different tumors within single patients’ bodies. What they’ve found is described in a new PNAS Early Edition paper.
“The question I originally wanted to answer was whether metastasis occurs early or late in a cancer’s progression,” says Kamila Naxerova of Harvard Medical School, first author of the new paper. “This has implications for how similar metastases are to the primary tumor and to each other, and that has implications for treatment.”
Naxerova and her colleagues knew that if they could compare the genetics of tumors and metastases within a patient’s body, they could figure out when different populations of cells had split—just as evolutionary biologists use genetics to determine when in history two populations of birds, plants, or animals split apart. But sequencing the full set of genes from patients’ tumor samples requires explicit, up-front patient permission and would involve a large, long, prospective trial.
“We realized that we could be more efficient if we just focused on regions that are particularly enriched for mutations,” says Naxerova. This would allow them to use tumor samples already collected by pathologists.
They decided to look at regions of-so called “polyguanine repeats,” strings of many guanine nucleotides in a row. When cells copy themselves, mutations are often introduced into these areas of the genome, so they can vary even in cell populations that have recently diverged. Without sequencing, researchers can determine how varied these regions are by simply measuring their length.
The team focused on colon cancer for their first test run of the technique, since surgeons operating on colon cancers often remove not only primary tumors, but metastases throughout the abdominal cavity. With such samples from 22 patients, Naxerova and her colleagues built lineage trees showing when each metastasis—some in the liver, others in lymph nodes or ovaries—diverged from the primary colon cancer. The more different the polyguanine regions were between two samples, the longer ago these cell populations had separated.
“I went into this being a little bit biased toward thinking that metastases emerged early in a cancer’s course and would be very different from the primary tumor,” says Naxerova. “But it turns out that’s not the case. Most of the cases we look at, the primary tumor and metastases are pretty closely related.”
Now that the researchers have shown that analyzing only the polyguanine regions of tumor samples is sufficient to show their relatedness, Naxerova wants to use the approach to answer more basic questions about how cancer spreads. Are cells that metastasize to the liver different than cells that metastasize to the lymph nodes? Can the number of polyguanine mutations in a tumor predict a patient’s treatment outcome or odds of survival?
Clinically, the test could be used to determine where a wide-spread cancer originated. “Sometimes you encounter a weird case where just by looking at the cell morphology of biopsies, you can’t tell what’s going on,” says Naxerova. “Is this a colon cancer that metastasized to the ovary or an ovarian tumor that metastasized to the colon?” These two tumor types would be treated very differently, and sequencing the polyguanine region of the cancers could determine which tumor was older.