Eight years ago, developmental biologist Rita Fior learned that her mother, who needed cancer treatment at the time, would receive different drugs depending on nothing more than which hospital she chose. Fior was taken aback. “You don’t know if it’s better to take drug A or B,” she says. “This is a big problem.” Now she is addressing the problem—with a fish.
This week, Fior, who is at the Champalimaud Centre for the Unknown in Lisbon, and her colleagues reported growing implanted human tumor cells in zebrafish larvae. Each fish became a minuscule model of a patient’s cancer—and a testbed for treatments. Similar cancer “avatars” have been created with mice, but the piscine approach may be faster and cheaper, making it accessible for more patients. “Zebrafish could have a unique niche [in cancer treatment],” says Leonard Zon of Harvard Medical School in Boston, who has used the fish for more than a decade to study how cancer develops.
To create mouse avatars, researchers implant some of a patient’s cancer cells into rodents lacking a normal immune system and measure whether various drugs destroy the tumors that sprout in the animals. But the mice are expensive to create and typically require between 2 and 6 months to deliver a verdict.
Zebrafish, which have become widely studied lab organisms, are cheaper and faster to raise, but they are evolutionarily more distant from humans than the rodents. So Fior and her colleagues first determined whether human tumors behave the same in zebrafish. When the scientists inserted human colorectal cancer cells into zebrafish embryos and allowed them to grow for 4 days, the resulting tumors showed three hallmarks of human solid tumors: rapid cell division, formation of blood vessels to supply nutrients, and the ability to spread to other locations in the body.
The team then added either of two chemotherapy cocktails commonly used for colorectal cancer to the fishes’ water and found that some of the tumors shrank and others didn’t. That result suggests that the fish avatars can discriminate between effective and ineffective drugs. Next, the researchers injected cells from five patients’ colorectal tumors into different zebrafish embryos and dosed the fish with the same chemotherapy drug combination the patients had received. In four out of five cases, the response of the tumors in the zebrafish avatars correctly predicted whether the patients’ tumors would rebound within 3 months to 6 months after surgery, the team reports in the Proceedings of the National Academy of Sciences.
Because of the time required to produce mouse cancer avatars, Fior says, doctors typically use them to choose a follow-up treatment if the first one fails. But all the steps for producing fish avatars, from implanting the tumors to analyzing the results, would take just 2 weeks to 3 weeks, so they could be useful for choosing a patient’s initial treatment, she says.
She and her colleagues aren’t the first to try to create zebrafish cancer avatars, but their work is the most persuasive so far, Zon says. Oncologist Richard White of the Memorial Sloan Kettering Cancer Center in New York City predicts that the fish have a future in personalizing cancer treatment. “The advantage is simple—numbers,” he says. To create mouse avatars for 1000 patients wouldn’t be feasible, he says, “but to do that in fish is easy.”
Still, Zon cautions that not all human drugs work in zebrafish, so “we need to study a lot more patients to see, in a broad view, how this approach performs.” Fior and colleagues plan to do that with a larger group of colorectal cancer patients and with patients who have other types of cancer, such as breast cancer. “We will ask if we give the fish the same drug as the patients and see if we get the same results,” Fiori says.