ANIMAL MODELS HAVE NEVER BEEN A PERFECT STAND-IN FOR HUMANS, but their use took a giant step forward with CRISPR and other genetic manipulation technologies (“Made to Measure,” Fall 2016). The techniques made it much easier to replicate human disease mutations in laboratory animals.

That development has particularly benefited cancer research. Tumors show a wide range of genetic diversity—too wide to replicate easily through traditional breeding techniques—but the new tools have made it possible to engineer features of a single tumor into a mouse. A “co-clinical trial” then can try out a new medication in the mouse first; if it is successful, the same treatment can be given to the human patient.

Mice in that scenario might even be replaced by Drosophila melanogaster, the common fruit fly. A team led by Ross Cagan, a developmental and cancer biologist at Mount Sinai Hospital in New York City, used a fruit fly approach to discover an effective treatment for a patient with metastatic colorectal cancer. The results of the trial appeared in the May 2019 issue of Science Advances.

The team first sequenced the man’s primary tumor and identified mutations that might be fueling its growth. Then the researchers narrowed the field to nine mutations within the tumor, and engineered those nine mutations into a generation of fruit flies.

While the flies were still larvae, they had various treatments mixed in with their food. They received a nutrient mixture that was either plain or mixed with one of 121 drugs that have been shown to have some kind of antitumor effect.

On all of these regimens, fewer than 20% survived. But when researchers tried combining drugs, one pair—a cancer drug and a drug traditionally used for osteoporosis—saw survival rates for the flies triple in some cases. When this combination was administered to the patient, his tumors shrank by 45% and stayed that way for the next 11 months, although new resistant lesions eventually did emerge.

A fruit fly may seem a poor stand-in for a human patient, but roughly 75% of human disease-causing genes have a functional counterpart in the fly, according to Norbert Perrimon, a geneticist and developmental biologist at the Broad Institute of MIT and Harvard. Drosophila has been used since the early 1900s to study human diseases, including cancer. “The signaling pathways involved in cancer are conserved in fruit flies,” he says. “And in fact many components of these pathways were first discovered in the fly.”

A London company, My Personal Therapeutics, is trying to make this approach widely available. For each patient, the company engineers five to 20 mutations into about 400,000 flies (called “fly avatars”). Then it laces the flies’ food with each of about 2,000 FDA-approved drugs, alone and in combination, including both traditional chemotherapy agents and others with reported antitumor effects. When there is a match—a fly given a particular drug lives much longer than expected—the company replicates the drug screen eight times to make sure the observed effect holds up.

Flies have several advantages over mice in cancer research. “In mice or rats, we can generate only two or three mutations and it takes a year or more,” says Nahuel Villegas, chief scientific officer of My Personal Therapeutics. “In flies, we can create 20 mutations in a month and a half and can activate them all at the same time to recreate the patient’s tumor.” That greater number of mutations more accurately reflects the complexity of real-world tumors, he says, which are typically driven by a dozen or so key mutations. In addition, hundreds of thousands of flies can be handled in a small room and, compared to other models, they eat very little. Because they require only minute doses of drugs, they offer a significant cost benefit when expensive cutting-edge therapies are on the menu.

But the process of testing treatments in this way is still labor intensive, and even with the short reproductive cycles of the flies, agonizingly slow. “One challenge is to build the avatar and perform the screen fast enough to deliver a treatment,” Perrimon says. Currently the whole process can take four or five months, but the company aims to shrink the timeline to weeks.

My Personal Therapeutics is now offering its technology commercially to patients worldwide, with a focus on colorectal and GI cancers, and is conducting multi-center clinical studies. Though the company is banking on the success of avatars, its ultimate aim is to build a “flybrary” of data showing which treatments may be successful therapies for patients with particular mutations. The company hopes that future patients will be able to get treatment recommendations based on what worked for previous patients—rather than getting their own personal swarm.